The specific identification of Staphylococcus aureus with new fluorescence in situ hybridization...

The specific identification of

Staphylococcus aureus with new

fluorescence in situ hybridization

(FISH) methods

By

Thomas Sutherland Lawson

A thesis submitted to Macquarie University

for the degree of

Doctor of Philosophy

Faculty of Science

January 2012

Examiner’s Copy

mailto:[email protected]

http://www.mq.edu.au/

ii

c© Thomas Sutherland Lawson, 2012.

Typeset in LATEX 2ε.


iii

Unless otherwise indicated, the material in this thesis-

by-publication is original and the work of the candidate.

The findings of the thesis were based on investigations

at Macquarie University, Sydney, Australia. Tests were

performed with the approval of the University Biosafety

Committee (09/14/LAB and 5201000927) and were lim-

ited to pure cultures of patient isolates at a non-clinical

location. Its Chapters contain sections that were pub-

lished in peer-reviewed Journals and are included as such.

To comply with Journal requirements, sections may dif-

fer in their format or contain material that overlaps. In

some instances the page size and byline of the original

publication was modified so that it could be integrated

into the thesis. The thesis did not italicize Latin expres-

sions that have common English usage such as in situ, in

vivo or in vitro. It followed the American convention for

spelling and the citation style of the Journal of Clinical

Laboratory Analysis (Online ISSN: 1098-2825).

Thomas Sutherland Lawson


Acknowledgements

I would like to express my gratitude to my supervisors Dr. Russell Connally, Dr. Jon

Iredell, Associate Professor Subramanyam Vemulpad and Professor Jim Piper.

The staff and students at the Faculty of Science, Macquarie University and at the

Centre for Infectious Diseases and Microbiology, Westmead Hospital are thanked for

their help. I would like to thank the anonymous reviewers of the manuscripts submitted

to journals for their helpful feedback. Finally, thanks to my family and friends for their

support.

I would like to acknowledge the Australian Research Councils Linkage Projects

(LP0775196) for funding this research and the Australian Proteome Analysis Facility

for providing laboratory facilities.

v

vi Acknowledgements

List of publications and awards

Published manuscripts:

1. Lawson TS, Connally RE, Vemulpad S, Piper JA. In silico evaluation and testing

of fluorescence in situ hybridization 16S rRNA probes for Staphylococcus aureus.

Lab Med 2011;42:587-591 (Chapter 3) (1).

2. Lawson TS, Connally RE, Vemulpad S, Piper JA. Optimization of a two-step

permeabilization fluorescence in situ hybridization (FISH) assay for the detection

of Staphylococcus aureus. J Clin Lab Anal 2011;25:359-365. (Chapter 3) (2).

3. Lawson TS, Connally RE, Vemulpad S, Piper JA. Express fluorescence in situ hy-

bridization methods for the detection of Staphylococcus aureus. Clin lab 2011;57:789-

794 (Chapter 3) (3).

4. Lawson TS, Connally RE, Iredell JR, Vemulpad S, Piper JA. Detection of Staphy-

lococcus aureus with a fluorescence in situ hybridization that does not require

lysostaphin. J Clin Lab Anal 2011;25:142-147. (Chapter 4) (4).

5. Lawson T, Connally R, Vemulpad S, Piper JA. Dimethyl formamide-free, urea-

NaCl fluorescence in situ hybridization (FISH) assay for Staphylococcus aureus.

Lett Appl Microbiol 2012;10.1111/j.1472-765X.2011.03197.x:(in press). (Chap-

ter 4). (5).

6. Lawson TS, Connally RE, Vemulpad S, Piper JA. In reference to targeted imag-

ing modality selection for bacterial biofilms in chronic rhinosinusitis and different

biofilms, different disease? a clinical outcomes study. Laryngoscope 2011;121:2043-

2044. A) (6).

7. Lawson TS, Connally RE, Iredell JR, Piper JA. The simultaneous detection and

differentiation of staphylococcus species in blood cultures using fluorescence in

vii

viii List of publications and awards

situ hybridization: A comment. Med Princ Pract 2011;20:390-391. (Appendix

A) (7).

Candidate contribution to the above manuscripts: concept (75%), experimental (100%),

analysis of results (85%) and writing (80%).

Conference proceeding:

1. Hamey LG, Connally RE, Yen SW, Lawson TS, Piper JA, Iredell JR, Lumines-

cent microspheres resolved from strong background on an automated Time-Gated

luminescence microscopy workstation. DICTA 2009 2009;1:223-228. (8).

The candidate contributed 5% to this manuscript’s concept, analysis and writing.

Awards:

1. Automated pathogen detection using time-gated luminescence microscopy, ICS

APAI PhD Scholarship, Macquarie University, 2008 to 2011.

2. Commercialization training scheme (CTS) Scholarship and Postgraduate Certifi-

cate in Entrepreneurship, Macquarie University, 2008 to 2009.

3. FABLS Support Scheme for Emerging Research Projects, Macquarie University,

2008.

Abstract

Staphylococcus aureus (SA) is a common bacterium associated with potentially serious

infections affecting both humans and other mammals. It is of particular concern that

SA can rapidly develop resistance to a range of antibiotics. Consequently, SA can cause

death and severe disability as a result of treatment failure. Antibiotic resistant SA is

especially prevalent in modern hospitals. For these reasons the capacity to rapidly

identify SA in patients is a crucial endeavor. The rapid identification of SA plus

information about its sensitivity to specific antibiotics can be lifesaving.

A range of laboratory based techniques are available for the identification of pathogens

such as SA. However, these current techniques have important limitations. These lim-

itations include (i) initial inadequate specificity and (ii) a delay by days of the precise

identification of the pathogen. As a result initial treatment of possible infections with

SA is usually based on broad clinical judgments and not precise information. These

clinical judgments usually lead to the use of broad spectrum antibiotics which may

have modest or no impact.

These are the reasons for this current project whose aim is to develop techniques

which offer (i) specificity concerning the identity of the pathogen infecting a patient

and (ii) rapid results.

In order to achieve these aims we have sought to further develop and optimize

an established laboratory technique - fluorescent in situ hybridization (FISH). This

technique is most commonly used following the outcomes of a blood culture to identify

suspect pathogens such as SA. The use of FISH in this context is to confirm the accuracy

of the diagnosis based on the blood culture and Gram-stain. The positive attributes of

FISH include its applicability to a range of specimen types plus its accuracy, robustness,

short turnaround time and its ability to offer in situ (cellular location of the pathogen

in the specimen) information.

In technical terms FISH binds oligonucleotides to its complementary nucleotide

sequence targets, usually 16S rRNA. The oligonucleotides are then usually visualized

ix

x Abstract

by fluorochrome labels and an epifluorescent microscope.

There are limitations to analyses based on FISH. There are delays in the use of FISH

because of the need to first complete a blood culture. This requirement is to make sure

the number or load of microbes is sufficiently high to allow accurate detection. The

assay is usually not automated and requires handling by a technician. In addition, the

establishment of FISH techniques can be complex and the costs can be high if dedicated

equipment is purchased. Finally, routine FISH techniques may lead to the inhibition

and obscuring of the signals generated by FISH. It is because of these limitations that

FISH is rarely used in clinical diagnostics.

Accordingly, the aim of this project was to overcome these limitations and to develop

FISH as a viable diagnostic tool. The specific aim was to investigate the accurate and

rapid identification of SA with FISH techniques.

SA was chosen as the subject and target of this investigation firstly because of

its clinical importance, and secondly because it is potentially a difficult pathogen to

detect with DNA based FISH techniques. Here, it should be noted that SA is frequently

misidentified with coagulase-negative staphylococci (CoNS). In addition, if refined and

redeveloped FISH techniques can identify SA, it can be reasonably assumed that the

same or similar techniques would be effective with most other common and important

pathogens.

There are three outcomes of this project.

(i) The existing FISH method for detecting SA was improved (Chapter 3). New

probe sequences for FISH that were specific to SA were identified. These probes had

binding and formamide requirements that were more useful than the existing sequences

that are commonly reported and used. High-yield fluorophores were found to label SA

with a high and consistent signal intensity and were also more resistant to photo-

bleaching.

New techniques for the preparation of FISH were developed which facilitate the ap-

plication of FISH. These techniques eliminate tedious and time-consuming preparation

for FISH assays. This was achieved by the development of premixed materials. The

adhesion of the specimen containing SA to glass slides was improved. The need for

xi

adequate cell adhesion to glass slides is a technical issue with FISH that is not widely

recognized. This technical issue of cell adhesion can be directly linked to the accuracy

of the FISH assay. Resolution of this technical problem has been shown in this project

to allow detection of SA with much less cell loss and with the considerable benefit that

SA at lower numbers could be detected.

A two-step permeabilization treatment using lysozyme and lysostaphin was devel-

oped which was useful if high molecular weight probes were used. This approach

shortened the time needed for hybridization incubation when smaller probes were used.

These techniques have previously been reported. However, past approaches have been

substantially improved and optimized so that SA integrity could be maintained and

the FISH assay could be completed in one hour instead of several hours. This is a

considerable achievement with potential advantage to patients with serious infections.

Tests were run to determine if the time taken to conduct conventional FISH assays

could be substantially reduced. A range of techniques were developed all aimed at

reducing the time taken to conduct FISH assays. These techniques included the com-

bination of existing permeabilization steps. These developments were successful. It

was possible to detect by FISH techniques the presence of SA in 24 minutes (in place

of the current 45 to 127 minutes) and to complete a Gram-stain and follow up FISH

test within one hour of a positive blood-culture.

(ii) New approaches to FISH were tested (Chapter 4). Improving the existing tech-

nique is useful, but does not extend the potential of the assay. A FISH technique that

can detect SA with DNA probes in the absence of permeabilization with lysostaphin

was developed. Lysostaphin is a significant burden to the routine use of FISH to de-

tect SA. It can be costly and its handling, storage and application are difficult. When

lysostaphin was omitted, the permeabilization step was simplified. Usually when FISH

was run, two permeabilization treatment arms were needed, one for SA and another

for other Gram-positive bacteria. With DNA probes and this new technique, only one

was required for the detection of Gram-positive bacteria.

A FISH method was developed that detected SA in the absence of formamide with

urea-based reagents. Urea is non-toxic and so its handling and disposal is simpler and

xii Abstract

it can be used in both the hybridization and washing reagents. Previously, the washing

step had to rely solely on NaCl to remove partly bound oligonucleotides. When urea

was used, the FISH signal was more intense and non-specific binding was minimized.

Possibly urea is partially permeabilizing the SA and more effectively removing unbound

probe. Because of the attributes of urea, FISH could be run entirely on a hot-plate with

a precise temperature control. This removed the need for the conventional dedicated

incubator and water-bath.

(iii) A FISH assay was developed that could detect SA in a complex autofluorescent

blood specimen using a europium chelate and time-gated luminescence microscopy

(TGLM) (Chapter 5). Specimens were prepared for testing with TGLM by spiking

fresh whole-blood with SA and incubating. The SA was then separated from most of

the blood and detected with conventional oligonucleotide probes and FISH and with

a europium (Eu3+) probe and time-gated luminescence microscopy (TGLM). Eu3+

probes and TGLM provided higher clarity than the conventional probe as most of the

background signal or autofluorescence from the specimen was suppressed.

The technique developed for the separation of SA from the spiked blood sample was

simple, rapid and accurate, collecting nearly all the intra and inter-cellular SA. The

separated SA remained viable and could be cultured in nutrient broth. Cultures of

the separated SA became turbid more rapidly than cultures of the unseparated spiked

blood sample.

The central aim of this research project, namely, the enhancement of the use of

FISH for the rapid detection of SA, was achieved. The existing FISH methodology

and techniques were greatly enhanced and new methods including a TGLM technique

for the use of FISH in highly autofluorescent specimens were successfully developed.

Increased permeabilization of SA for FISH and DNA probes was achieved. As other

bacteria need less or no permeabilization, the findings are likely to be applicable to

other pathogens. Accordingly, extension of these investigations and additional testing

of patient specimens in clinical settings is the next important step.

Contents

Acknowledgements v

List of publications and awards vii

Abstract ix

List of Figures xvii

List of Tables xix

1 Introduction 1

1.1 Rationale for the present project . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Septicemia and S. aureus . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.3 Diagnosis and treatment of S. aureus septicemia . . . . . . . . . . . . . 4

1.3.1 Standard diagnostic pathway for septicemia and its limitations . 4

1.3.2 Possible improvements to S. aureus septicemia diagnostics . . . 6

1.3.3 The role of fluorescence microscopy in diagnosing S. aureus sep-

ticemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.4 Improving the FISH for identification of S. aureus directly in blood

cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.4.1 Recent developments in FISH . . . . . . . . . . . . . . . . . . . 11

1.4.2 Limitations of FISH as applied to S. aureus . . . . . . . . . . . 14

1.5 Improvements required for the application of FISH in the detection of

S. aureus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.5.1 Re-engineering FISH for the detection of S. aureus . . . . . . . 18

1.5.2 Issues concerning the use of FISH for the detection of S. aureus

in complex samples . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.6 Outline of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

xiii

xiv Contents

2 Methodology: FISH with rRNA-targeted oligonucleotide probes 27

2.1 Preparation of reagents, probes and S. aureus samples . . . . . . . . . 31

2.1.1 Hybridization and post-hybridization washing buffer preparation 31

2.1.2 In Silico Evaluation and Testing of FISH 16S rRNA Probes for

S. aureus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.1.3 In situ probing of S. aureus with specific 16S rRNA targeted

oligonucleotides . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

2.1.4 Bacterial isolates . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.1.5 Separation of S. aureus from an in vitro model of bacteraemia . 41

2.2 Step 1: Method for adhering specimens to slides . . . . . . . . . . . . . 42

2.3 Step 2: S. aureus fixation . . . . . . . . . . . . . . . . . . . . . . . . . 43

2.4 Step 3: S. aureus permeabilization . . . . . . . . . . . . . . . . . . . . 43

2.5 Step 4: In situ hybridization with rRNA-targeted, fluorescently labeled

oligonucleotides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

2.6 Step 5: Specimen washing with buffer . . . . . . . . . . . . . . . . . . . 47

2.7 Direct visualization of microorganisms . . . . . . . . . . . . . . . . . . 47

2.7.1 FISH image and statistical analysis . . . . . . . . . . . . . . . . 49

3 Improvements to the existing FISH method 51

3.1 In silico evaluation and testing of fluorescence in situ hybridization 16S

rRNA probes for Staphylococcus aureus . . . . . . . . . . . . . . . . . . 53

3.2 Optimization of a two-step permeabilization fluorescence in situ hy-

bridization assay for the detection of Staphylococcus aureus . . . . . . . 59

3.3 Express fluorescence in situ hybridization methods for the detection of

Staphylococcus aureus . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

4 Development of new FISH methods 75

4.1 Detection of Staphylococcus aureus with a fluorescence in situ hybridiza-

tion that does not require lysostaphin . . . . . . . . . . . . . . . . . . . 76

4.2 Dimethyl formamide-free, urea-NaCl fluorescence in situ hybridization

(FISH) assay for Staphylococcus aureus . . . . . . . . . . . . . . . . . . 82

Contents xv

5 Time-gated fluorescence imaging of a europium chelate label 87

5.1 Time-gating of a europium probe rapidly labeled with luminescence in

situ hybridization for the detection of Staphylococcus aureus . . . . . . 87

5.1.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

5.1.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

5.1.3 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

5.1.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

5.1.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

6 Conclusion 103

A Appendix A: Other publications that emerged from the thesis 109

A.1 In reference to targeted imaging modality selection for bacterial biofilms

in CRS and different biofilms, different disease? . . . . . . . . . . . . . 110

A.2 The simultaneous detection and differentiation of staphylococcus species

in blood cultures using fluorescence in situ hybridization . . . . . . . . 114

B Appendix B: Analysis of common oligonucleotides used in the detec-

tion of S. aureus with FISH 117

List of abbreviations 123

References 125

xvi Contents

List of Figures

1.1 Time-gated Giardia lamblia . . . . . . . . . . . . . . . . . . . . . . . . 20

1.2 Schema of time-gated luminescence microscopy (TGLM) . . . . . . . . 23

2.1 Probe melting temperature and efficiency by formamide concentration . 33

2.2 Binding affinity of 18, 19, 22, 24 and 25 base oligonucleotides . . . . . . 39

2.3 Melting temperature of the Staaur probe . . . . . . . . . . . . . . . . . 46

5.1 SA cultures labeled with BHTEGS and visualized with TGLM . . . . . 96

5.2 SA and SE incubated and labeled with the BHTEGS chelate in blood . 97

5.3 SA labeled with BHTEGS and Alexa with Figure 5.4 plot locations . . 98

5.4 Plots of the TGLM and conventional FISH signal . . . . . . . . . . . . 99

xvii

xviii List of Figures

List of Tables

1.1 Advantages and limitations of fluorescence in situ hybridization (FISH) 10

1.2 FISH studies that have identified SA . . . . . . . . . . . . . . . . . . . 13

2.1 FISH method typically used for the detection of bacteria . . . . . . . . 30

2.2 Washing buffer NaCl (M) by hybridization buffer formamide (%) . . . . 35

2.3 Guide to judging the performance of a probe sequence . . . . . . . . . . 38

5.1 S/N calculations of a BHTEGS and a conventional probe . . . . . . . . 100

B.1 Binding affinity of the EUB338 probe . . . . . . . . . . . . . . . . . . . 118

B.2 Binding affinity of the KT18-16S68 probe . . . . . . . . . . . . . . . . . 119

B.3 Binding affinity of the Staaur probe . . . . . . . . . . . . . . . . . . . . 120

B.4 Binding affinity of the Staphy probe . . . . . . . . . . . . . . . . . . . . 121

xix

xx List of Tables

1Introduction

1.1 Rationale for the present project

Staphylococcus aureus (SA) is a Gram-positive bacterium ubiquitous in the environ-

ment and in humans and is linked to infection with high rates of morbidity and mor-

tality (9, 10, 11). It is persistent in the upper respiratory tract, is easily transmitted

in hospitals (12) and is often resistant to many antibiotics (9, 13). SA is the most

common cause of septicemia, that is, an infection of the blood (referred to as Staphylo-

coccus aureus bacteremia or SAB) (14), and is also associated with invasive procedures

(15).

Identification of SA as the cause of septicemia is difficult (16). The symptoms of

a SA infection can be inconsistent (17). The foci of infection may not be found and

other coagulase negative staphylococci (CONS) such as S. epidermidis (SE), a common

contaminant of blood cultures (18), can mimic its features (19). Rapid blood tests can

confirm infection, but not identity of the pathogen (20). Tests that can identify SA are

usually much slower (21) as they need culturing first (22). Septicemia if severe, needs

immediate treatment (16). The initial treatment is, therefore, based on presumptive

diagnosis (23) and maybe incorrect (24) and the infection clearance delayed (25). A

solution is to use diagnostic tests that are both rapid and accurate (26, 27, 28).

1

2 Introduction

Fluorescence in situ hybridization (FISH) is an accurate and rapid test for the

identification of intact SA in specimens (29). The technique often involves hybridizing

slide-based fluorescent labeled DNA probes to in situ rRNA (30) sequences of SA in

blood cultures (Table 1.2). Labeled SA is then visualized with fluorescence microscopy

(31). It can be applied either to cultures (32) or directly to specimens that are not

cultured (33). Automation is simpler if peptide nucleic acid (PNA)-based probes are

used and the SA detected with a flow-cytometer (34, 35).

There are certain limitations to the use of FISH as a test for SA which may explain

why it is not reported more often in routine diagnostics (23, 36, 31). Limitations

include: the preparation of its reagents can be exacting and time-consuming (37, 38).

The sensitivity of the assay is low (39, 40) and usually necessitates its application to

blood cultures which take two days to complete (41). Material other than SA in the

specimen (42, 43) can hinder access of the probes to SA and can obscure its signal

(44, 45). The natural emission of the specimen (referred to as autofluorescence) can

interfere with the signal from the probe (46, 47). There are other tests for SA detection

which, although not as accurate or quick as FISH and do not visualize SA, nonetheless

are simple and inexpensive to perform (48, 49, 50, 51).

The project investigates the use of FISH for the detection of SA and its differen-

tiation from the coagulase negative staphylococci (CONS) Staphylococcus epidermidis

(SE) (19). There were three aims to this research. The first was to improve the cur-

rent FISH protocols for the identification of SA in blood cultures (37, 52, 32). Blood

cultures was chosen as it is the most tested specimen (Table 1.2) (41, 18).

The second aim of the project was to re-engineer the FISH procedure. Improvements

to a conventional technique are useful, but are usually incremental. Radical alternatives

to the conventional FISH reagents, equipment and procedure were tested in order to

achieve significant gains to its performance.

The third aim was to reduce interference of the FISH signal from non-target material

in the specimen (43, 47). The tests performed in the project and the changes made to

the FISH assay were aimed at making its use in routine microbiology for the detection

of SA more practical.

1.2 Septicemia and S. aureus 3

1.2 Septicemia and S. aureus

Septicemia is a common infection and a type of sepsis where pathogens have invaded

the blood-stream (53, 54). It is a serious condition which can be fatal and SA is its

most common cause (55). The frequency and role of SA in septicemia is related to its

potential virulence (13). SA can remain viable both within the body and on surfaces

for many months (12). Its highly cross-linked peptidoglycan and capsule layer makes

it hardy and resistant to antibacterial reagents (12). SA is easily transmitted from

surfaces to people and between people and can evade and inhibit the immune system

(56). It can quickly develop antibiotic resistance and share this resistance with other

stains of SA (27, 13). It has been suggested that because of its ubiquity, SA could

serve as a hygiene indicator in laboratories that wish to maintain sterility (12).

The diagnosis and treatment of SA related septicemia has improved with time (25),

but its incidence remains high (53, 54). Its control in hospitals has proven difficult to

achieve (10). A third of surfaces (12) and a third of otherwise healthy people tested

were found to be positive for SA (57). The number of invasive procedures carried

out in hospitals has risen (54) and patients with strains of SA resistant to first-line

antibiotics have become more common (55). This increase in resistance is related to

the acquisition of genes that make SA less susceptible to antimicrobials (13).

Strains of SA can be categorized as community acquired methicillin resistant SA

(CA-MRSA) and hospital-acquired MRSA (HA-MRSA) (58). CA-MRSA is mostly

acquired by the young in the community, transmitted via people and shared items,

resistant to β-lactams, but sensitive to other antibiotics, presents with skin and soft

tissue infection and possesses the SCCmec gene (type IV or V) (13). HA-MRSA

is mostly acquired by the elderly in health facilities, transmitted via patients and

staff, resistant to most antibiotics, presents with bacteraemia, wound infections and

infections of the respiratory and urinary tracts and possesses the SCCmec gene (type

I, II or III) (13). The frequency, virulence and ability to resist treatment suggest that

for now, SA related infections will remain a major health issue (59, 25).

4 Introduction

1.3 Diagnosis and treatment of S. aureus septicemia

High risk patients with suspected septic shock and septicemia need immediate treat-

ment (54). Each hour of delay in its treatment substantially increases the rate of mortal-

ity (10, 60). The choice of initial antibiotics is occasionally ’hit or miss’ (23) with major

adverse consequences if the treatment is incorrect (24, 25). Cefazolin, flucloxacillin,

nafcillin and oxacillin are the antibiotics commonly given if methicillin-sensitive SA

(MSSA) is suspected and vancomycin if MRSA is suspected (16). Treatment is con-

tinued for one to two weeks and is not stopped until the patient is symptom-free (25).

If the response is poor, the antibiotic treatment is changed or rotated (16).

At the start of treatment the identity the etiologic pathogen might be suspected,

but is not usually known (20). The initial examination and empiric data is used to

inform treatment (23), but this is not specific to the patient and cannot confirm the

identity and susceptibility of the pathogen (26, 24, 17). Even with incomplete diagnosis,

most antibiotic treatment is adequate against SA and other prevalent pathogens (23,

9). Misdiagnoses and incorrect treatment (23), however remain the most common

causes of avoidable death due to sepsis (24, 17). Poor diagnosis can also contribute to

increasing the resistance of SA to antibiotics (25). This suggests that early diagnosis

of septicemia that can also identify the etiologic pathogen is important for effective

treatment (26, 10).

1.3.1 Standard diagnostic pathway for septicemia and its lim-

itations

If the patient is febrile and septicemia is suspected, blood is collected and antibiotic

treatment started (26, 16). The blood can then be tested to confirm infection and

the identity of the pathogen (21). Initial tests applied directly to whole-blood such as

coagulation screening, C-reactive protein, full blood count, liver function and urea and

electrolytes are rapid and simple to perform (27, 16). They can indicate an infection

(16), but they cannot establish the identity of the etiologic pathogen.

1.3 Diagnosis and treatment of S. aureus septicemia 5

The role of blood culture in diagnosing septicemia

Further microscopy, culturing and a variety of tests help to identify the cause of sep-

ticemia and its susceptibility to antibiotic treatment (22, 61). Blood cultures can be

run with an automated continuous-monitoring blood culture system (20, 41) to con-

firm an infection and increase its pathogen numbers (60). Other more accurate tests

can then be applied to identify the pathogen (27, 22). Blood cultures are necessary

for these confirmatory tests, but delay their start by two days and can be inaccurate

(62, 28, 60, 21, 18). Blood cultures may not detect an infection if the pathogen in the

blood is slow-growing and produces a false-negative result (63, 24) or they may detect

a contaminant such as SE from skin flora inoculated at the time of blood collection

(18) and produce a false-positive result (64, 65). Thus, the tests that follow a blood

culture, although highly accurate and rapid themselves (22) such as FISH (32), are

compromised somewhat by their necessity for blood cultures (60, 21, 18).

The role of Gram-stain for the detection of septicemia

Once it has become positive, blood culturing is followed by a Gram-stain (16). The

Gram-stain uses reagents to stain the microbes so that they can be observed with a

light microscope fitted with a 40× or higher objective. (66). Gram-staining is simple

and rapid (10 to 20 minutes) to complete and can be reasonably accurate, indicating

the presence of bacteria (48). Gram-negative bacteria such as Escherichia coli (EC)

stain red-pink and SA and other Gram-positive bacteria stain blue-violet (48). SA can

be identified as it is observed as a 1 µm diameter spheres and that regularly form tetrad

clusters (48). However, because SE and other coagulase-negative staphylococci (CoNS)

contaminants are similar in appearance and also stain blue-violet (66), they are often

misidentified as SA (19). In spite of this drawback, if the patient is not responding

well, antibiotic treatment for septicemia is modified on the basis of the Gram-stain

report (23, 66, 61).

6 Introduction

The role of confirmatory and antibacterial susceptibility tests

To confirm the identity of SA, tests are run after the Gram-stain and completed the

same day (65, 67). Tube coagulase (51) and DNase (51) biochemical tests are com-

monly performed as they are rapid and inexpensive (50); both tests can be completed

with some accuracy (22) in two to four hours (67). These tests, however do not directly

visualize the SA (49) and can be relatively inaccurate (22) unless performed in conjunc-

tion with other tests (67). Tests to determine the resistance of SA to antibiotics such

as chromogenic media, Mueller-Hinton agar or the disc diffusion test (22), are started

after Gram-staining, but are not completed until the next day (27). The confirmatory

and susceptibility tests that follow a blood culture rarely inform the current antibiotic

treatment (23) or change its outcome (68). Rather the value of these tests is to confirm

diagnosis and inform the treatment of future infections (26).

1.3.2 Possible improvements to S. aureus septicemia diagnos-

tics

Blood cultures and Gram-stains are adequate for most septicemia diagnoses (18), but

they do not identify SA or its susceptibility to antibiotics (66). Reliance on these two

procedures may become less acceptable as the number of resistant strains of SA in-

creases (55, 54). The tests that can confirm the identity of SA and its susceptibility

cannot be completed fast enough after a blood culture to play a role in the patient’s

treatment (67). Attempts have been made to address these shortcomings (69). Solu-

tions are focused on developing tests for SA that are applied directly to whole-blood

(70) or improving the confirmatory tests applied to blood cultures (65). Possibly, iden-

tification of etiologic pathogens in whole-blood has the most potential (21), but the

problems are still technically complex and difficult to solve (70).

1.3 Diagnosis and treatment of S. aureus septicemia 7

1.3.3 The role of fluorescence microscopy in diagnosing S. au-

reus septicemia

Although recently not the focus of diagnostics (71), because of the accuracy and sim-

plicity of its procedures (28), fluorescence microscopy might be a promising avenue for

investigation (72). Fluorescent microscopes are more expensive than their equivalent

bright-field microscopes, but this cost is decreasing (73) and they are now common in

microbiology laboratories (74). Non-specific fluorescent stains, such as acridine-orange

(AO) (Sigma, A6014) (1 µg/ml of AO in 1 µM of acetic acid and sodium acetate)

(75), are simpler and faster (5 minutes) to apply than Gram-stains (10 to 20 minutes)

as they can be completed in a single incubation step and can be more accurate (71).

Nonetheless, fluorescent staining has the same constraint as Gram-staining, namely

these stains indicate, but do not distinguish SA from CoNS (71).

Assessment of the value of immunofluorescence microscopy in the diagnosis

Another fluorescent technique, immunofluorescence, usually takes longer than Gram-

stain to complete (15 to 30 minutes), but is as simple to apply as a general fluorescent

stain (76). Like the general fluorescent stain, its procedure can include a single incuba-

tion step at room temperature followed by a quick rinse to remove unbound probe (77).

Unlike Gram-staining and general fluorescent staining, immunofluorescence can posi-

tively identify SA. The technique binds fluorochrome conjugated antibodies to antigens

specific to SA on its cell wall (78). If the SA is in a suspension (referred to as plank-

tonic) such as a blood culture (18), immunofluorescence can be completed as rapidly

as a Gram-stain (Virostat, 6883) (78).

Despite its ability to directly visualize and identity SA, it may not be possible to

apply immunofluorescence consistently to SA and this might explain why it is reported

infrequently. Virulence factors in SA disrupt the formation of antibodies to SA and the

binding of antibodies to SA (56). SA antigen expression is inconsistent between strains

and, with changes to its micro-environment or phase of growth, can differ within a single

strain (56). SA inhibits formation and binding of antibodies by forming a capsular or

8 Introduction

slime layer (79), congregating in clusters (48) or biofilms (80) and expressing Protein

A (81).

Initial work carried out during the project (data not shown) applied commercial

monoclonal and polyclonal antibodies against SA to cultures of clinical isolates of SA.

A single product, a rabbit polyclonal antibody conjugated to FITC (Virostat, 6883)

(78) identified the SA tested and did not react with isolates of SE, a finding that was

repeated elsewhere (82, 77). It was not known if this was a true antibody to antigen

binding or a rabbit immunoglobulin G to Protein A binding (83, 77). Furthermore,

these results were inconsistent; when the same product in a biotinylated (Virostat,

6887) and in an unconjugated (Virostat, 6881) form were tested, they failed to bind.

For these reasons, the use of immunofluorescence to identify SA in routine diagnostics

might be impractical (56).

1.4 Improving the FISH for identification of S. au-

reus directly in blood cultures

At 45 minutes (32), fluorescence in situ hybridization (FISH), takes longer to complete

than immunofluorescence, but its identification of SA can be more robust (76). Both

assay types have similar characteristics for the determination of SA. They can confirm

with certainty the presence of SA, detect SA in situ (with reference to the specimen)

(19) and, although often performed, do not need culturing (33). Both methods have

limitations, of which those for immunofluorescence were already touched on. In the

case of FISH, its reagent preparation and optimization can be complex and, if con-

ventionally applied, cannot distinguish between sub-strains (31, 38). There is also a

fundamental difference between the two assays. Unlike immunofluorescence, which la-

bels cell-wall antigens exterior to the pathogen, FISH labels nucleic sequences interior

to the pathogen (44). This simplifies its interaction with SA and allows its labeling of

SA to be consistent (76). The advantages and limitations of the identification of SA

with FISH are detailed in Table 1.1.

1.4 Improving the FISH for identification of S. aureus directly in bloodcultures 9

The FISH assay creates the conditions necessary for the hybridization of a probe

sequence (referred to as an oligonucleotide) to its in situ complementary sequence (30).

Flurophore conjugated DNA is hybridized to 16S rRNA to identify pathogens such as

SA from the species-level through to the domain-level (29). FISH can detect unknown

pathogens in situ, whose culturability is also unknown, without disturbing the matrix

of the specimen (44, 84). The assay can thus provide in situ data of the relationship

between SA, its host and other pathogens (85). Even if first cultured, it is possible to

identify SA by its growth patterns in the culture media as they usually form tetrad

clusters demonstrable in Gram stained smears (48, 79).

The slide and DNA-based FISH procedure commonly used to identify SA (32) is

completed in five steps: (i) specimen preparation (42, 86, 43), where slides are pre-

pared, spotted with the collected (and possibly cultured) specimen; (ii) fixation of the

specimen, usually with an alcohol (32); (iii) permeabilization, where SA is permeabi-

lized with lysozyme and lysostaphin enzymes (87); (iv) hybridization of the probe to

SA (88); (v) and washing, where unbound probe is removed (87). If the final slide

rinse, cover-slip mounting and microscopy is also included, the assay is completed in

six, not five steps (44). Thus, the FISH assay itself is relatively simple to perform.

What can complicate its routine use is its setup and quality control and the access of

probes to targets in complex specimens (Table 1.1).

10 Introduction

Table 1.1: Advantages of using fluorescence in situ hybridization (FISH) to detect SAand its limitations (36).

Advantages Limitations

A range of frequently encountered mi-crobes can be targeted with FISH 16SrRNA probes (64).

Microbe targets need to be predeter-mined. Probes for SE are limited intheir availability and accuracy (39, 40).

SA can be differentiated at the specieslevel (87).

Resistant strains of SA cannot be dif-ferentiated from non-resistant strains(32).

SA can be tested for directly in whole-blood (ISH) (62), stool (89), sputum(90) and urine (33) specimens withouta culturing step.

Sensitivity limit is approximately 103

to 104 cfu/ml (39, 91, 33, 40).

Assay is relatively reliable and robust(52).

Known SA and SE controls are neededwith each test batch (37).

Multiple probes can be applied at thesame time (45, 64).

Probes need similar formamide concen-trations to be applied together (92).

Assay can detect SA from blood cul-tures in 45 minutes (32).

Blood cultures on average delay FISHfor 2 days (60). Simple tube coagulase(22) and DNase (49) tests can be com-pleted in 2 to 4 hours (67).

Assay simple to perform if reagents arepremixed and stored before start.

Complex manual handling required forreagent preparation and optimization(93) and, if slide-based, is difficult tofully automate (94).

Expense per test is relatively low ifDNA probes are used (50).

Tube coagulase (22) and DNase (49)tests are less expensive (50).

Small quantities of probes and lytic en-zymes are used per slide.

Probes and lytic enzymes are expensiveto first source (95).

Assay does not need a relatively largeamount of bench-space.

An (epi)fluorescent microscope is re-quired to visualize the cells.


1.4.1 Recent developments in FISH

The disadvantages of the FISH assay listed in Table 1.1 were addressed to some extent

by recent advances to its procedure and use in blood cultures for e.g. (18).

An online resource for the optimization of probe sequences for SA (and other

pathogens) as well as the conditions of their incubation was made available by Yil-

maz et al. (92) (mathfish.cee.wisc.edu). This resource makes it possible to rapidly test

probes in silico (performed on computer) (84, 96) against the 16S rRNA sequences of

SA (97). Locations on the SA 16S rRNA sequence with the highest affinity for bind-

ing can be identified (98) and then tested against complementary sequences of various

lengths. Once the most efficient of these sequences is found, the ideal concentration

of formamide and NaCl and temperature for their hybridization can also be calculated

(99). With this tool, the characteristics of established and new probe sequences can be

rapidly and accurately compared and the time spent optimizing their use with FISH

in the laboratory can be reduced.

If FISH can be applied to a specimen in the absence of culturing, the time to result

after specimen collection is dramatically shortened and in situ data can be collected

(85). SA in urine was recently detected directly with a FISH assay by Wu et al. (33).

Conventional detection of pathogens in urine takes at least one day to be cultured and

then detected. FISH applied directly to urine can detect and identify the pathogens

present in two hours, thus informing their treatment the same day as specimen col-

lection. Other specimens tested successfully with FISH for SA without first culturing

have included cerebrospinal fluid (39), sputum (90), stool (89), and whole-blood with

in situ hybridization (ISH) (100, 63, 62). Unfortunately, FISH cannot be applied di-

rectly to whole-blood without either a series of complex purification and blocking steps

(100, 63, 62) or more commonly after completing a two day blood culture (32).

The turnaround time for a diagnostic test is important as it determines its usefulness

(101). A 45 minute FISH assay that successfully detected SA was reported by Poppert

et al. (32). This was a significant improvement as it was faster than the commonly used

two hour assay (64). A confirmatory test that takes longer than one hour is usually

12 Introduction

of a lesser value to the clinician (101). Most septicemia treatment decisions are made

at blood collection or after the report of a Gram-stain (66). At 45 minutes, the FISH

assay is more useful after a blood culture (32), but it may still be too lengthy to inform

treatment (23) because Gram-staining can be completed in 10 to 20 minutes (23).

The introduction and use of PNA based probes for FISH instead of DNA probes

has simplified and improved the assay (102). Multiple probes can be combined more

easily if PNA is used as it is not as sensitive as DNA probes to the stringency of the

buffer (102). No permeabilization is required for SA as PNA probes do not carry a

charge (103). For the same reason, the hybridization step is more efficient and can be

shortened (104) and the use of a flow-cytometer is less hampered by material in the

specimen (105, 34). With the use of PNA probes and the omission of permeabilization

and improvement to hybridization, it is possible to run a FISH assay in one step instead

of five (106).

Shrestha et al. (35) reported distinguishing MSSA from MRSA strains in approx-

imately three hours using PNA probes and a FISH assay visualized with a flow-

cytometer. Since FISH probes for SA cannot distinguish sub-stains (87), the deter-

mination of antibiotic susceptibility of SA with FISH was previously implied (107),

but not thought practical until this report (35). The FISH procedure used was indirect

(35); blood cultures were re-cultured in growth medium with or without antibiotics.

Differences in cell-counts and signal between the stains was then detected with a PNA

based FISH assay and a flow-cytometer (35). Unlike other tests for susceptibility which

take a day to complete (22), this study could determine susceptibility with FISH the

same day as the Gram-stain result was available (35).

The current high cost of PNA probes (Advandx, AC005) (50) and flow-cytometry

may deter its routine use. This could change with the lapse of the original patent

(108) and the development of inexpensive and easy-to-use desktop flow-cytometers

(BD Accuri, C6).


Table 1.2: FISH studies that have identified SA.

Specimensource/type

Specimen form‡ Culturing be-fore FISH

Reference

Blood culture Planktonic Yes (88, 37, 109, 103,104, 105, 110,111, 112, 113,50, 52, 114, 32,40, 68, 35)

Brain abscess Non-planktonic No (115)

Bone Non-planktonic No (116)

Cerebrospinal fluid Planktonic Yes (39)

Ear Non-planktonic No (117, 118, 119,120)

Heart valve Non-planktonic No (28)

Laboratory strains Planktonic Yes (107, 102)

Menses Planktonic No (121)

Milk Planktonic No (79, 80)

Nose Non-planktonic No (122, 123, 124,125, 126, 127,128, 129, 130)

Sputum Planktonic No (90)

Stool Planktonic No (89)

Throat Non-planktonic No (131, 132)

Tampon Non-planktonic No (121)

Urine Planktonic No (33)

Whole-blood Planktonic No ISH assay (100,63, 62)

Wound Non-planktonic No (19, 87, 133, 85,134, 135)

‡ Planktonic specimens contain free-floating pathogens in dilution. Non-planktonicspecimens contain pathogens adhered to its matrix or tissue.

14 Introduction

1.4.2 Limitations of FISH as applied to S. aureus

In spite of the new developments to the FISH assay, its use is not often reported in

routine microbiology. There could be a number of reasons for this. Most of the decisions

about the treatment of septicemia are made at the time of blood collection or after its

culturing (23). The FISH assay reported in blood culture studies cannot be applied

to whole-blood and so cannot be used at that time to indicate SA. The sensitivity of

the assay is limited to 103 to 104 cfu/ml (colony forming units) or more (39, 40) and

the blood from septic patients contains SA at no more than 10 cfu/ml and often only

1 cfu/ml or less (41). An ISH assay can be applied to whole-blood (100, 63, 62), but

its use is not widely reported as it is laborious and complex. The application of FISH

to positive blood cultures is also problematic. Blood cultures delay the start of FISH

by two days (60, 21) and remove most of the collectible in situ data (20). FISH can

identify SA in 45 minutes, but this may be too long after Gram-staining, which can be

completed in 10 to 20 minutes (23), for it to inform treatment (68). The delay to the

start of FISH and the time taken for its completion (21) are disadvantages that other

tests, performed after a blood culture to identify SA, share with FISH (22, 60).

Preparation of specimens and reagents

Material in the specimen can interfere with the FISH procedure and its signal. The

probe can bind non-specifically to the specimen or be unable to access SA (42, 43). The

signal from SA can be concealed by the specimen or be overwhelmed by its autofluo-

rescence (136, 47). Signal interference from debris in the blood cultures is usually not

an issue (18), but can be if FISH is applied to blood cultures that are accelerated (137)

or to specimens that are not cultured (121). There are several possible approaches to

avoid interference. Separation and purification of the specimen can increase its ratio of

SA to non-target material (42, 138, 43). The specimen can be pretreated with reagents

that block its non-specific binding to the probe (100, 76). Selective lysis of the spec-

imen can make SA more accessible to probes (19). The specimen can be illuminated

at longer wavelengths to reduce its autofluorescence (40).


These additional treatments, however complicate the FISH procedure, disrupt the

specimen matrix and, if overdone, can weaken and reduce the resolution of the signal

from SA (100)

Aspects of permeabilization of S. aureus

The use of a FISH assay with DNA-based probes applied to slides has drawbacks (139).

If it is not done correctly, preparation of its reagents and the procedure at each step

can lead to a poor signal from the probe (38). The formulation, storage of its reagents

and their correct application is not simple (44). Determining the correct concentration

involves the titration of reagents against reference strains of SA and testing with FISH

(32). The concentration of the reagent is optimal when SA produces the highest signal

and SE generates a weak or non-existent signal (140).

A poor signal from the FISH assay is often a result of inadequate fixation or per-

meabilization of SA. Insufficient permeablization of SA is possibly the most common

cause of a weak or non-detectable FISH signal and a false-negative result (103, 93).

The fixation step is important because its failure can cause the permeabilization step

that follows it to also fail. Over-fixed SA can be resistant to permeabilization, which

then reduces the access of the probe to SA, its binding and the signal of labeled SA. In

contrast, under-fixed SA can lyse when permeabilized and its cells and signal are lost

(39). Even if the fixation of SA is correct, the permeabilization of SA that follows can

still fail. As described, poorly permeabilized SA can result in no signal or loss of cells.

To avoid an incorrect result, reference stains of SA can be tested with FISH and the

assay adjusted if the signal is incorrect. This pretesting with FISH, however complicates

and delays its implementation (32). Once it is optimized, the fixation procedure is easily

repeated. This is not the case with the SA permeabilization reagents lysozyme (141)

and lysostaphin (95). When first prepared, these enzymes require titration and testing

with FISH against reference strains of SA to determine their correct concentration for

permeabilization. If these lytic enzymes are applied again after their long-term storage,

this testing is repeated.

16 Introduction

Duplex binding of DNA to RNA

The correct fixation and permeabilization of SA is a prerequisite for the successful de-

tection of SA with DNA-based FISH. Nonetheless, if the conditions for its hybridization

and washing are not ideal, the probe signal can also be weak or non-specific (142). If

the stringency of the hybridization and washing buffer is incorrect, probes can either

bind indiscriminately or not at all (140). Even if hybridization of SA is correct, the

washing step may be incorrect and fail (143). A further complication is that probe

sequences for SA differ in their capacity to differentiate SA from SE (Appendix B)

(52) and incorrect labeling of SE can be worsened by the conventional washing buffer

which relies on NaCl alone to adjust its stringency (142). Formamide is effective at

denaturing nucleic acids (144), but it is absent from the washing buffer as it is toxic,

difficult to dispose of and thus cannot be used in the larger volumes of washing buffer

used to remove the unbound probe (142).

Steps can be taken to improve the signal differentiation of SA from SE. The signal

can be amplified by tyramide signal amplification and multiple labeling with probes,

but this complicates the assay (145, 45). In silico calculations can be run to predict the

optimal formamide and NaCl concentrations for the probes and incubation conditions

(99). These calculations can reduce the time spent in the laboratory optimizing the

hybridization and washing reagents and the likelihood of their incorrect application

and a poor result (92). In silico calculations can also predict those probe sequences

with the greatest capacity to differentiate SA from SE (52) and can then optimize

the assay to this chosen probe sequence (92). As well, an extra non-toxic denaturing

reagent can be added to the washing buffer to improve its stringency control (144).

In summary, DNA based FISH is simple and rapid to carry out, but the preparation

of its reagents and their quality control complicates its implementation in routine

diagnostics (37, 40).

1.5 Improvements required for the application of FISH in thedetection of S. aureus 17

1.5 Improvements required for the application of

FISH in the detection of S. aureus

This project addressed some of the deficiencies of the FISH assay as a test for sep-

ticemia (146), by choosing SA as a target for its investigation. Apart from its clinical

importance in septicemia and other infections (55), SA is easily misidentified with

CoNS such as SE (19). The blood culture and Gram-stain tests indicate, but do not

identify SA with certainty (18) and so a test is needed to confirm its identity (44).

Furthermore, SA is peculiar in DNA-based FISH as it, unlike other Gram-positive bac-

teria, is resistant to the permeabilizing effect of lysozyme, but sensitive to the action

of lysostaphin (147). Thus, a DNA-based FISH assay developed for SA, with its per-

meabilization simplified or omitted, could be applied to other pathogens (32). There

were also practical reasons for selecting SA. The safe handling of clinical isolates of

SA at a non-clinical location was not onerous. It could be stored long-term (86) and,

when it was needed, quickly cultured for testing with FISH.

Firstly, the project investigated the conventional slide and DNA-based FISH assay

used to detect SA (Chapter 3). Established probe sequences for SA were tested. It

was not known if these probes for SA were optimal or if new, more efficient probes

could be identified (1). The conventional method for formulation of reagents was

reassessed. Preparation of these reagents for FISH was lengthy and it was hoped

that it could be shortened (2). It was not known if the conditions typically used

were optimal or merely followed convention (2). The standard incubation conditions

for permeabilization, hybridization and washing were optimized. These incremental

improvements were then used to further shorten the turnaround time of the FISH

assay. It was not known if shortening the FISH assay to less than 45 minutes (32)

would also compromise its signal and accuracy (3). Since the investigation focused

on basic aspects of FISH procedure, PNA and flow-cytometry were not tested in this

project.

18 Introduction

1.5.1 Re-engineering FISH for the detection of S. aureus

Next, the project investigated new approaches to the identification of SA with DNA-

based FISH (Chapter 4). Alternatives to the use of lysostaphin were tested (4). For

a DNA-based FISH assay that detects SA as well as other pathogens, three perme-

abilization treatments are normally applied (52): (i) lysostaphin and lysozyme for SA

(32), (ii) lysozyme for other Gram-positive bacteria and (iii) no treatment, apart from

fixation, for other pathogen types (37). If an alternative to lysostaphin were found,

the FISH procedure would be simpler to carry out and its costs possibly halved. An

alternative could also be to simplify the preparation for FISH as the initial preparation

of lysostaphin is exacting and once diluted, its activity needs to be monitored as it

declines with time (95).

Alternatives to the formamide-based reagents, incubators and water-baths com-

monly used with FISH were tested. Functionally the hybridization and washing buffers

are similar (142) as both are used to denature nucleic acid. The use of formamide (142),

however is restricted to the hybridization buffer as it is toxic (144) and large volumes

are used in washing (40). As a result, the washing buffer may not be as efficient

at removing unbound probe. The denaturing efficiency of FISH might be increased

and assay preparation simplified if a non-toxic alternative to formamide (142) were

found which could be used in all incubations (144). With more robust FISH buffers,

it might become practical to carry out the assay without an incubator or water-bath.

This equipment is usually purchased specifically for FISH and can take up bench-space

which might otherwise be better used.

1.5.2 Issues concerning the use of FISH for the detection of

S. aureus in complex samples

Lastly, the project investigated solutions to the management of interference of the

FISH procedure and signal from material in the specimen. Blood cultures do not

usually suffer this interference as the proportion of debris is low (32). Specimens that

are directly tested (116) or rapidly cultured (137), however have a higher proportion of


background material or debris. This can then block access of the FISH probe to SA, can

hide the SA and can autofluoresce and overwhelm the FISH signal (44, 121, 39, 125).

Two techniques for reducing interference were investigated: the purification of the

specimen (138) and the time-gating of its autofluorescence (46, 47). An example of

time-gating is illustrated in Figure 1.1. Giardia lamblia cells were labeled with a

europium long-lifetime probe (136) using an immunofluorescence technique and then

visualized by time-gating the signal (148).

20 Introduction

(a) Ungated fluorescence image

(b) Time-gated luminescence image

Figure 1.1: (a) Micrographs of pond water containing flocculation. Giardia lambliacysts were immunofluorescently labeled with a europium chelate and then inoculatedin the water and illuminated at 365 nm; they fluoresce a bright red. (b) The sameview of the specimen with the UV emission time-resolved with a time-gated auto-synchronous luminescence detector (GALD). Micrographs included with permissionfrom Russell Connally (148).


Sample purification and preparation technique

The aim of reducing specimen interference deserves more comment. As a first approach,

and before testing with an ISH assay, the project investigated purifying whole-blood

spiked with SA (138). Purification increases the ratio of SA to non-target material

(138). By removing non-target material, probe access is improved and autofluores-

cence reduced (42, 138, 43). Purification has its own disadvantages; it disrupts the

specimen, alters in situ data and can lengthen the time taken for the preparation of

the FISH procedure (138). Of possibly greater concern is the potential purification has

for removing SA with other material (42, 43). Separation and removal of non-target

material from SA is only of use if it can be done accurately, rapidly and simply.

Reduction of autofluorescence in fluorescence microscopy

As a second approach, the autofluorescence signal was blocked or time-gated (referred

to as time-resolved) so that it did not interfere with observation of the FISH signal

(47). Time-gating the signal lessens the need to apply other treatments for autofluo-

rescence such as purifying the specimen or illuminating it at longer-wavelengths (136).

These treatments can remove SA or reduce its resolution when viewed with a micro-

scope. Time gating illuminates the specimen with short (800 µs) pulses, blocks its

first emission, but captures emissions that follow after a predefined gating period (Fig-

ure 1.2) (148). This time gated luminescence microscopy (TGLM) technique relies on

probes with long lifetime emission. After the excitation pulse ends the initial short life-

time autofluorescence (and conventional fluorophore), emission (tens of nanoseconds)

is blocked until its decays (47). The emission from the luminophore probe that lasts

longer (10 to 100s of microseconds), is observed free of autofluorescence (47).

The detection of pathogens by time-gating cells labeled with luminophore probes

using an in situ hybridization (ISH) assay is not often reported. The equipment needed

is specialized (136) and its signal can be weak (149, 150). Its occasional use can be

attributed to the probes (referred to as chelates) used, which have relatively poor

stability and solubility (151). To overcome the limitations of the chelates, changes were

22 Introduction

made to the ISH assay that complicated and lengthened it. These included blocking

steps, overnight incubations and signal amplification with streptavidin conjugates (149)

or tyramides (150). This rendered the time-gated ISH assay unsuitable for most routine

diagnostics (45). Immunofluorescence might be well suited as its labeling is exterior to

the cell (136), but as discussed, its use with SA is limited (83).

As a possible remedy, a new europium chelate (BHTEGS), which had properties that

were more desirable than those of earlier chelates (151), was developed by a research

group at Macquarie University (personal communication with Russell Connally). This

chelate was stable and soluble enough that it could be conjugated directly to DNA

(KT18 5’- GCAAGCTTCTCGTCCGTT -3’) so that SA could be labeled with a rapid

assay based on FISH. The labeled SA was then time-gated with a newly developed

GALD device (148). This is the first report on the application of the chelate, the LISH

assay and GALD to the detection of SA.


Figure 1.2: The time-resolved technique uses the difference in the emission lifetime ofa europium chelate (BHTEGS) and the autofluorescence of the specimen. The schemaof TGLM shows time on the X-axis and the intensity of the luminescence from thepathogen is on the Y-axis. The specimen is illuminated with pulses, but the captureof its emission is delayed (referred to as gated) until the short-lived autofluorescencehas decayed. The luminescence from the europium chelate is then collected withoutthis background signal. Illustration is included with permission from Russell Connally(148).

24 Introduction

1.6 Outline of the thesis

The Chapters in the thesis address, in order, the aims of the project. Chapter 2

describes the broad aspects of the methodology used in this project, structured around

the five steps of the whole-cell prokaryote FISH assay (30). Chapter 3 reports on

improvements to the FISH assay commonly used to detect SA in blood cultures (32):

1. Established and new probe sequences for the identification of SA with FISH were

assessed. This was published in a peer reviewed journal and included as such:

Lawson TS, Connally RE, Vemulpad S, Piper JA. In silico evaluation and testing


Lab Med 2011;42:587-591 (1).

2. The preparation and storage of reagents for FISH was shortened. A FISH tech-

nique that used probes of large molecular-weight was optimized for the detection

of SA. This was published in a peer reviewed journal and included as such: Law-

son TS, Connally RE, Vemulpad S, Piper JA. Optimization of a two-step per-

meabilization fluorescence in situ hybridization (FISH) assay for the detection of

Staphylococcus aureus. J Clin Lab Anal 2011;25:359-365 (2)

3. The FISH procedure for the identification of SA was shortened so that it could

be completed in half an hour. This was published in a peer reviewed journal

and included as such: Lawson TS, Connally RE, Vemulpad S, Piper JA. Express

fluorescence in situ hybridization methods for the detection of Staphylococcus

aureus. Clin lab 2011;57:789-794 (3).

Chapter 4 reports on the re-engineering of the FISH assay for the identification of SA:

1. A novel FISH technique free of the SA permeabilizing reagent lysostaphin was

developed for the detection of SA. This was published in a peer reviewed journal

and included as such: Lawson TS, Connally RE, Iredell JR, Vemulpad S, Piper

JA. Detection of Staphylococcus aureus with a fluorescence in situ hybridization

that does not require lysostaphin. J Clin Lab Anal 2011;25:142-147 (4).

1.6 Outline of the thesis 25

2. A novel FISH technique free of formamide, an incubator and a water-bath was

developed for the detection of SA. This was published in a peer reviewed jour-

nal and included as such: Dimethyl formamide-free, urea-NaCl fluorescence in

situ hybridization (FISH) assay for Staphylococcus aureus. Lett Appl Microbiol

2012;10.1111/j.1472-765X.2011.03197.x:(in press) (5).

Chapter 5 reports on an investigation into the reduction of specimen autofluorescence

which can overwhelm a FISH signal (47). Techniques were tested for removal of non-

target material from the specimen. A new europium chelate BHTEGS was trialled.

The probe made it possible to apply a rapid in situ hybridization assay similar to

FISH and which could rapidly detect SA in complex blood specimens which exhibited

autofluorescence.

Chapter 6 summaries the findings of the project and draws conclusions from its results.

Appendix A includes the following two publications that also emerged from the thesis

(6, 7):

1. Lawson TS, Connally RE, Vemulpad S, Piper JA. In reference to targeted imag-

ing modality selection for bacterial biofilms in chronic rhinosinusitis and different

biofilms, different disease? a clinical outcomes study. Laryngoscope 2011;121:2043-

2044 (6).

2. Lawson TS, Connally RE, Iredell JR, Piper JA. The simultaneous detection and


situ hybridization: A comment. Med Princ Pract 2011;20:390-391 (7).

These letters provide commentaries on contemporary SA FISH studies, in light of the

findings of this thesis.

Appendix B provides technical details on the binding affinity of established and new

oligonucleotides specific for SA (98).

26 Introduction

2Methodology: FISH with rRNA-targeted

oligonucleotide probes

The fluorescence in situ hybridization (FISH) method, used by this project to identify

S. aureus (SA), is described in this Chapter. This method labels 16S rRNA in whole

prokaryote cells such as SA with DNA probes conjugated to fluorophores (38). DNA

probes are less efficient at crossing the cell-wall and accessing their targets than peptide

nucleic acid (PNA) based probes (102), but are better suited to routine work as they

are far less expensive. The FISH method labels RNA; as this type of nucleic acid

exists in high numbers within prokaryote cells and it can thus produce a signal of

high intensity (140). A sub-type of rRNA, 16S rRNA is usually the target of most

established FISH probes for SA (44) since knowledge of its prokaryote sequences is

relatively comprehensive (88, 64) and accurate (84).

The identification and differentiation of SA from Staphylococcus epidermidis (SE)

with FISH is the focus of this project. The labeling of SA with FISH is not as straight-

forward as the labeling of other pathogens such as Escherichia coli (EC). SA requires

permeabilization with lysozyme and lysostaphin for DNA probes to cross the cell-wall

(147, 52). The 16S rRNA sequence for SA is almost identical to that for SE (1). SE

is a frequent contaminant of blood-cultures and its appearance in Gram-stains can

27

28 Methodology: FISH with rRNA-targeted oligonucleotide probes

be indistinguishable from SA (48). In contrast, E. coli requires simple fixation with

alcohol and no permeabilization before it can be labeled with FISH and its 16S rRNA

sequence is dissimilar to other common pathogens (52).

SA was chosen as a target in part because its labeling with DNA probes and FISH is

more complex than other pathogens (52). SE was chosen to act as a negative control as

its appearance is similar to SA and its 16S rRNA is almost identical. If simplified, the

methods developed for SA could be applied to other less demanding pathogens. The

reverse scenario is less likely to be true. A single pathogen and not multiple pathogens

was also chosen as it permitted a greater focus on the methodology of FISH. As the

extra complexity associated with targeting more than one type of pathogen was avoided

(32), a greater number of experimental iterations could be tested each day.

A recent report by Poppert et al. (32), which developed a new FISH assay for SA

(referred to as accelerated), was taken as the starting point for this project. This accel-

erated method was chosen as it had a 45 minute turnaround time, the fastest reported

for the detection of SA and used DNA probes to test blood-cultures, the most common

specimen type tested for SA (32). Apart from its turnaround time, other aspects of

this FISH method were conventional. This is illustrated in Table 2.1 which lists the

accelerated method (32) and another method (referred to as comprehensive) recently

used by Gescher et al. (64). Although the accelerated method (32) is over two times

shorter than the comprehensive method (64), both methods share the same compo-

nents of the FISH assay. They both have five steps (or six if microscopy is included):

(i) specimen preparation, (ii) fixation, (iii) permeabilization, (iv) hybridization and (v)

washing. Both used slides for testing blood-cultures with DNA probes targeting 16S

rRNA and were visualized with an epifluorescent microscope to find SA.

The accelerated FISH method (32) was not only used as a starting-point for the

project, but also as a control throughout the project. If a new and enhanced FISH

method was developed, it was not considered a practical improvement unless it demon-

strated the same accuracy and signal intensity as the accelerated method (32).

The FISH methods developed and tested in this project applied, like the two meth-

ods listed in Table 2.1, DNA probes to SA affixed to slides and were visualized with

29

an epifluorescent microscope. This differs from other FISH studies that have used con-

focal microscopes (36) or flow-cytometer (105, 34) for the visualization of the probes.

Confocal microscopy (36) has a high sensitivity and can be useful in tissues that have

thickness. Flow-cytometer also has a high sensitivity, can collect quantitative data and

be automated (105). They both can be costly and complex to carry out, however and

were not thought necessary for an investigation of core aspects of the FISH procedure.

It should be noted that the prokaryote FISH method tested by the project differs

from FISH reported elsewhere that targets eukaryote chromosomes and their abnor-

malities (152).


Table 2.1: A comparison of the FISH methods described by Poppert et al. (32) andGescher et al. (64) for the detection of SA.

Accelerated FISH (32) Comprehensive FISH (64)

Preparation: Blood culture iso-lates were diluted with PBS (1 min),spotted to slides (1 min), air-dried (5min), fixed with methanol (10 min)and air-dried (1 min).

Preparation: Blood culture iso-lates were fixed with ethanol (1min), spotted onto slides (1 min)and air-dried (5 min).

Permeabilization: Slides werespotted (1 min) with lysis reagent(2 mg/ml lysozyme, 100 µg/mllysostaphin (147, 52), 10 mM Tris-HCl at pH 8.0) and incubated at 46◦C (5 min), washed with methanol(3 min) and air-dried (1 min).

Permeabilization: Slides werespotted with 1 mg/ml lysozyme (1min) and incubated at 30 ◦C (10min). Lysozyme was removed (1min), slides were spotted with 1mg/ml lysostaphin and incubated at30 ◦C (5 min) (32). Slides werewashed with filtered (Milli-Q, MQ)water (1 min) and air-dried (5 min).

Hybridization: Slides were spot-ted with hybridization buffer (30%formamide, 0.9 M NaCl, 10 mMTris-HCl at pH 8.0, 0.01% SDS, 25ng/ml probe and MQ water) (1 min)and incubated at 46 ◦C (10 min).

Hybridization: Hybridizationbuffer (40% formamide, 0.9 MNaCl, 20 mM Tris-HCl at pH 7.3,0.01% SDS, 10 pM of probe andMQ water) was spotted to slides (1min) and incubated at 49 ◦C (90min).

Washing/Mounting: Slides wereincubated with washing buffer(0.112 M NaCl, 10 mM Tris-HCl pH8.0, 0.01% SDS, 5 mM EDTA andMQ water) at 48 ◦C (5 min) andair-dried (1 min)

Washing/Mounting: Slides werewashed with water (1 min) andmounting media added with DAPI(1 min).

Total time: 45 minutes. Total time: 127 minutes.

2.1 Preparation of reagents, probes and S. aureus samples 31

2.1 Preparation of reagents, probes and S. aureus

samples

To save time, the hybridization buffer and washing buffer were prepared in advance

(2). Hybridization buffer (0.9 M NaCl (153), 20 mM Tris-HCl, 0.01% (w/v) SDS,

and 1 µg/ml DAPI) with no formamide or with 60% (v/v) deionized formamide were

prepared and stored for up to a year at -20 ◦C in 5 ml sterile plastic screw-top tubes

(2). The hybridization buffer contained DAPI. Many FISH studies counter-stain the

cells with a general DNA intercalating fluorescent dye such as DAPI (94, 36, 34) or

Hoechst (85) as a control for visualizing target and non-target pathogens and for cell

counting (85). When needed, the buffers were thawed and mixed to the desired target

formamide concentration (2).

2.1.1 Hybridization and post-hybridization washing buffer prepa-

ration

NaCl (Sigma, S619) was prepared at a 5 M concentration as stock solution in Milli-

Q (MQ) water (Millipore), sterilized with a 2 µm syringe-filter and stored at room

temperature. NaCl was diluted to 0.9 M (153) and used in the FISH buffers to increase

the stability of nucleic acid duplexes (142). Hydrochloric acid (HCl) (Sigma, H1758)

and Sodium hydroxide (NaOH) (Sigma, S8045) were used to adjust the pH of the

buffers and other reagents such as Tris-HCl. Most buffering with Tris-HCl was in the

physiological range of pH 7.0 to 8.0 and at concentrations of 10 to 100 mM (87). Tris-

HCl was prepared with Trizma hydrochloride (Sigma, T3253), MQ water and Trizma

base (Sigma, T1503) and HCl was used to adjust the pH. It was sterilized with a 2 µm

syringe filter before use.

Sodium dodecyl sulfate (SDS) (Sigma, L4390) was used at low concentrations (0.01

to 0.02% v/v) in hybridization and washing buffers as a surfactant and as a mild

permeabilizing agent (87). SDS was added to MQ water and the solution was mixed

and heated to 68 ◦C until the SDS dissolved. SDS was not autoclaved, but instead was


sterilized with a 2 µm syringe filter.

Formamide and in situ hybridization

Formamide was used to destabilize nucleic acid duplexes in the hybridization buffer

(87). The buffer’s stringency was adjusted with the formamide against a fixed 0.9 M

concentration of NaCl (153, 142). No preparation of the formamide was necessary if

it was already deionized (Applichem, A2156) and fresh (colorless). Fresh deionized

formamide was aliquoted and stored at -20 ◦C for up to one year before use.

If de-ionization of the formamide was needed, wet mixed bed ion exchange resin 5%

(w/v) (Sigma, Amberlite c© MB-1 hydrogen and hydroxide form, 501999) was added,

removed with a coffee filter and the formamide supernatant stored as described above.

Urea (153) was also used as an alternative denaturing agent for formamide (142) be-

cause formamide is toxic and difficult to dispose of (144). Stock solution of urea (Sigma,

U6504) was prepared at 80 M and filtered before use.

The ideal formamide concentration to use with a probe was predicted with in silico

calculation and confirmed by testing with FISH (92). The melting temperature of an

oligonucleotide probe binding to SA was calculated with the formula 81.5 + 16.6(log

M [Na+]) + 0.41(%G+C) - 0.72(% formamide) or with a corresponding and easier to

use online algorithm such as mathFISH (mathfish.cee.wisc.edu) (92) (Figure 2.1). G

and C are the number of Guanine and Cytosine bases in the sequence. The melting

temperature of a probe is when 50% of its sequence is annealed to its target (84).

If the incubation temperature and NaCl concentration is kept constant, the melting

temperature of a buffer can be adjusted with formamide. As the formamide is increased

so that the melting temperature of the nucleic acid is lower, the stringency of the buffer

also increases and the likelihood of mismatched DNA:RNA binding decreases.

To confirm the ideal formamide concentration for a probe, the hybridization buffer

was prepared with formamide at 15% (v/v) increments from zero to 60%. A FISH

assay was then applied, that used these different concentrations, to reference strains

of SA affixed to slides. The concentration with the highest signal intensity from the

FISH probe was chosen. Most probes can maintain a signal over two increments of


formamide (140). If two of these formamide concentrations were optimal, the higher

concentration was chosen as it is less likely to produce a non-specific signal.

For most of the probes tested in the project, formamide at 30% (v/v) produced

an adequate signal and was prepared by mixing the 60% prepared buffer at 1:8 with

the buffer that contained no formamide. This is about 5% higher than the lowest

formamide concentration for the probes calculated with mathFISH (92) in Figure 2.1.

The calculations assumed 47 ◦C incubation, 0.9 M NaCl (153) and 1 µM of probe in

the buffer. The oligonucleotide could be added to this buffer mix and stored at 4 ◦C

in 1.5 ml sterile plastic aliquots for a week before the FISH assay was run.

Figure 2.1: The hybridization efficiency of EUB338, KT18-16S68, Staaur or Staphyprobes to SA by the amount of formamide in the buffer. [FA]m is the melting formamideconcentration for the DNA:RNA duplex (92).

Buffer type in the washing media

In an approach similar to the preparation for the hybridization buffer, washing buffer

without salt (20 mM Tris-HCl, 5 mM EDTA and 0.01% (w/v) SDS) and with 1.8 M

NaCl was prepared and stored for up to a year at 4 ◦C in one liter bottles (Schott,


GL45) (2). When needed, these buffers were mixed to their target NaCl concentration

in a 50 ml tube and preheated in a water-bath to 47 ◦C.

The concentration of NaCl used in the washing buffer was a function of the concen-

tration of formamide used in the hybridization buffer (Table 2.2) (143). The formamide

concentration was determined, as noted before, by the probe and target pathogen se-

quence used (Figure 2.1) (92). The NaCl concentration was confirmed by testing with

FISH in the laboratory at 0.014, 0.04, 0.112, 0.318 and 0.9 M NaCl concentrations

(143). The NaCl concentration was optimal when the ratio of SA signal to SE signal

was at its highest (140).

Low concentrations of ions can affect the stringency of the buffer. To counter-act

this, ethylenediaminetetraacetic acid (EDTA) (Sigma, EDS) was used as a chelate in

the washing buffers when the NaCl concentration was lower than 0.225 M (102). It

was also used as a chelate in washing buffers that contained sodium citrate buffer

(SSC) (116). In addition, it was added to the TE buffer (TE is 10 mM Tris-HCl

and 1 mM EDTA) to protect stock solutions of fluorescent probes and to blood as an

anticoagulant. Stock solution of EDTA was prepared at 0.5 M, sterilized with a 0.2

µm syringe filter and stored at room-temperature before use.


Table 2.2: NaCl (M) in the washing buffer as a function of formamide (%) in thehybridization buffer (142, 143).

Formamide(%)‡ NaCl (M)

0 0.900

5 0.636

10 0.450

15 0.318

20 0.2250

25 0.159

30 0.112

35 0.080

40 0.056

45 0.040

50 0.028

55 0.020

60 0.014

65 -

70 -

‡ The percentage of formamide is depended upon the particular probe(s) sequence tobe hybridized as indicated in Figure 2.1.


2.1.2 In Silico Evaluation and Testing of FISH 16S rRNA

Probes for S. aureus

FISH oligonucleotide probes that are unique to SA were identified and then assessed.

Only 16S rRNA targets was considered. Its sequences for most pathogens are well docu-

mented, but there were no established probes that targeted the 18S or 23S rRNA of SA

(84). To identify unique SA sequences, SA and other non-target pathogen 16S rRNA se-

quences were collected from the NCBI-Nucleotide database (ncbi.nlm.nih.gov/nuccore)

(97).

Pathogens that had a similar 16S sequence to SA were identified (rdp.cme.msu.edu)

(154). When aligned with the online tool NCBI-Blast (blast.ncbi.nlm.nih.gov), SE had

an almost identical 16S rRNA sequence to SA (97). SE is a benign microbe often

found in blood-cultures as a contaminant and often misidentified as SA. The near

perfect alignment of SA to SE meant that any mismatches identified between the two

would probably be unique to SA and that their number would be small.

For this project, a typical sequence for SA (GenBank: CP000253.1) and one for SE

(GenBank: AF397060.1) were aligned and mismatches were identified at 69 to 89, 183

to 198, 452 to 477 and 999 to 1024 positions relative to E. coli 16S rRNA (97, 155).

The uniqueness of the identified mismatched sequences to SA was then confirmed by

reapplying it to other pathogen sequences (microbial-ecology.net/probecheck). The se-

quence 999 to 1024 was not unique to SA and so was not considered further. SA shared

this sequence with Staphylococcus haemolyticus which is also occasionally detected in

blood-cultures (65).

Quantitative assessment of oligonucleotides for S. aureus

To assess the usefulness of an oligonucleotide, the ∆Gooverall (DeltaGo) and the hy-

bridization efficiency of its sequence were calculated (92). DeltaGo indicates the prob-

ability of a probe to target binding (DNA:RNA) given the competing interactions of

probe (DNA:DNA) and target (RNA:RNA) self-binding (156). The higher the negative

number, the higher the binding potential of the probe to their targets. Hybridization


efficiency indicates the predicted ratio of target molecules bound with probe to all the

target molecules. A hybridization efficiency of one is equal to saturation binding and

zero to no binding.

Several criteria can be used to judge the affinity of a particular oligonucleotide to SA

(Table 2.3). A DeltaGo between -17 and -13 kcal/mol for SA, a difference in DeltaGo

between SA and SE greater than 3 kcal/mol, a difference in formamide concentration

of 20% (v/v) and hybridization efficiency greater than 0.8 indicates a highly sensitive

sequence. A DeltaGo greater than -13 kcal/mol and less than -10 kcal/mol to SE and

a hybridization efficiency greater than 0.8 indicates a highly specific sequence (92).


Table 2.3: A guide to judging the performance of a probe sequence to SA and SE.

Thermodynamic descriptor†Sensitivity∗ Specitivity£

Low≺ High≺ Low≺ High≺

∆G target§ (kcal/mol) < -17` -17 to -13 > -10 NA

∆G non-target§ (kcal/mol) < -13 NA NA > -10

∆G difference∼ (kcal/mol) < 3 > 3 NA NA

FA difference‡ (%) < 20 > 20 NA NA

HE target∝ (ratio) NA > 0.9 < 0.9 NA

HE non-target∝ (ratio) > 0.1 NA NA < 0.1

HE difference¶ (ratio) < 0.9 > 0.9 < 0.9 > 0.9

† Thermodynamic calculations assume a single DNA probe binding to a target 16SrRNA sequence (92).∗ Sensitivity = TP/(TP+FP), where TP= true positive and FP = false positive.£ Specificity = TN/(TN+FN), where TN = true negative and FN = false negative.≺ Low and high sensitivity and specificity cutoffs were based on Yilmaz et al. (92)§ Overall Gibbs binding potential of probe (DeltaGo kcal/mol) (156).` The DeltaGo indicates the standard state overall Gibbs free energy of the probe-target hybrid: the probability of probe to target binding (DNA:RNA) given thecompeting interactions of probe (DNA:DNA) and target (RNA:RNA) self-binding.The higher the negative number, the greater the probe-target binding affinity (156).^ NA: not applicable.∼ ∆G difference = DeltaGo target - DeltaGo non-target (96).‡ Melting formamide concentration (FA %) for the probe-target duplex (99).∝ Hybridization efficiency (HE) indicated the predicted ratio of target moleculesbound with probe to all target molecules. A hybridization efficiency of 1 indicatedsaturation and 0 no hybridization (98).¶ HE difference = HE target - HE non-target (92).


The sequences 183 to 193 and 452 to 477 were judged to have a low hybridization

efficiency (98). The 69 to 89 was analyzed in more detail with probes of 18, 19, 22, and

25 bases long (Figure 2.2). A pattern emerged where the 5’ end of a potential probe

was most efficient at the 65 to 67 positions. Probes from 15 to 30 bases were tested at

this location using the mismatch feature of mathFISH (96), and a number of potential

probe candidates were realized including all the established probes already reported

for SA. These calculations assumed 47 ◦C incubation, 0.9 M NaCl (153) and 1 µM of

probe in the buffer.

Figure 2.2: The 5’ end binding affinity (∆Gooverall) of 18, 19, 22, 24 and 25 bases long

oligonucleotides to the 54 to 73 SA 16S rRNA sequence.


2.1.3 In situ probing of S. aureus with specific 16S rRNA

targeted oligonucleotides

The sequences that were identified as highly specific to SA were then tested as fluores-

cent probes with FISH. This included two probes KT18 (16S68: 5’- GCAAGCTTCTCGTC-

CGTT -3’) (1) and STAAUR (16S69: 5’- GAAGCAAGCTTCTCGTCCG -3’) (87)

specific for SA and STAPHY (16S697 5’-TCCTCCATATCTCTGCGC-3’) (87) specific

for Staphylococcus. EUB338 (16S337: 5’- GCTGCCTCCCGTAGGAGT -3’) (157)

specific for eubacteria was used as a positive control. More detailed information on

the alignment and binding affinity of these sequences to SA and SE is provided in

Appendix B.

The oligonucleotides were either directly conjugated to fluorophores (157) or bi-

otinylated (100) at the 5’ end. The fluorophores that were used were Dylight R© 488

(Jackson), Alexa Fluor R© 488 or 555 (Invitrogen) and FITC or Cy3 (Genworks) (2).

For the time-gated luminescence microscopy (TGLM) visualization of the europium

(Eu3+) BHTEGS chelate (developed by a Macquarie University research group), a

member of this research team (Russell Connally) conjugated the new Eu3+ chelate

BHTEGS to the sequence KT18 (1).

Oligonucleotide resuspension, and storage

DNA oligonucleotides were supplied (Invitrogen or Geneworks) and stored dry for up

to a year at 4 ◦C. To use, these probes (as well as salmon sperm DNA) were diluted

in TE buffer at 100 µM, aliquoted out at 100 µl each to reduce freeze-thaw cycles and

stored for up to a year at -20 ◦C. Before use, an aliquot of the probe stock was thawed

and stored at 4 ◦C. To use, the probes were diluted 1:100 in the final hybridization

buffer mix to make 1 µM and stored at 4 ◦C and applied within a week.

Oligonucleotide probes used in FISH to detect SA can be synthesized from ribonu-

cleic acid (RNA), deoxyribonucleic acid (DNA) or peptide nucleic acid (PNA) (102).

RNA is rarely used as it can be easily degraded by the ubiquitous RNase. The use

of PNA (Advandx, AC005) based probes is often reported (109, 105) as they do not


possess a charge and so do not need to SA to be permeabilized for hybridization and

can bind rapidly to their targets with a high affinity (102). As they are low in cost, the

use of DNA based probes is also often reported (87, 32) and were used in this project,

but do need SA to be permeabilized for hybridization before they can be applied.

2.1.4 Bacterial isolates

So that isolates could be tested at a non-clinical location, clinical patient isolates of

SA, SE and E. coli were collected at a major hospital (Westmead Hospital, Sydney)

on agar plates. To control for potential differences between strains, 10 isolates of each

type of bacteria were randomly collected. Initial testing observed no difference in

the FISH signal between methicillin-susceptible SA (MSSA) and methicillin-resistant

SA (MRSA). To lower risk, only non antibiotic resistant strains were tested further.

Isolate identity was confirmed with polymerase chain reaction (PCR) (158) and then

de-identified for testing.

To compare FISH procedures, collected isolates were re-cultured in 50 ml tubes until

turbid, aliquoted, centrifuged for three minutes at 3000 rcf, supernatant removed,

frozen and stored long-term (86). Before its use, the nutrient broth was sterilized

with a 2 µm syringe filter. Before testing with FISH, isolates were thawed and re-

cultured, usually for 70 minutes, in the nutrient broth until turbid (0.5 McFarland).

No difference in the signals were observed if the isolates were tested by FISH directly

from the collected agar plates.

2.1.5 Separation of S. aureus from an in vitro model of bac-

teraemia

For the TGLM detection of SA in whole-blood, cultures of SA were washed and diluted

in saline to an optical density of 1.0 at 600 nm (159). NaCl at 0.9 % (v/v) was used

to dilute SA so that when the diluted SA was added, the tonicity of the blood would

be maintained. Venous blood was collected from a healthy volunteer in EDTA tubes

(Becton Dickinson, 367863). A simple in vitro bacteraemia model was created by


spiking fresh whole-blood with SA and incubating (160). For 1 ml of blood, 10 µl of

the SA in saline was added (1.0 optical density at 600 nm) and the blood incubated

with gentle agitation at 37 ◦C for one hour.

To simplify the procedure, the incubated blood was lysed with alkaline water which

released intra-cellular SA (138). The SA could then be detected with FISH. This differs

from the approach taken elsewhere that separated blood components with Dextran

500 and labeled intra-cellular SA in the leukocytes with an in situ hybridization (ISH)

assay (100). Lysing the blood allowed the ratio of SA to blood cells to be increased,

simplified and shortened the assay and reduced non-specific labeling of the FISH probe

to leukocytes and other blood debris (data not shown). The alkaline water was prepared

by adding 4 mM NaOH to Milli-Q (MQ) water at pH 10.0 (138). Blood and alkaline

water were then mixed at a ratio of 1:10 (to make a pH of 8.5) by vortexing and then

centrifuged at 3,000 rcf. The supernatant was removed and the treatment repeated

before spotting and fixing the pellet to slides for the FISH procedure.

2.2 Step 1: Method for adhering specimens to slides

SA adhesion to slides can be poor if it is air-dried (40). The adhesion of SA to the

slides can be increased by heat fixing the specimen to slides pretreated with agarose

and then, after the SA is spotted and dried, fixing with alcohol (139). The number

of SA that remain adhered to the slide as well as the effectiveness of permeabilization

can then be quickly determined with fluorescent DAPI (Sigma, D9564) stain excited

with UV light.

To prepare the slide, an agarose (Bio-Rad, 162-0102) bed was applied to diagnostic

glass slides (Menzel-Glaser, X1XER308B). The bed was prepared by adding 0.02%

(w/v) agarose with 0.01% (w/v) sodium azide (Sigma, S2002) to Milli-Q water R© (MQ)

(Millipore) and dissolving it in water by heating in a microwave oven without boiling

(139). This diluted agarose was spotted (10 µl) to each slide well and dried on a 60 ◦C

hotplate.

If cell loss persisted, the broth culture of the isolates could also be diluted 1:1 in

2.3 Step 2: S. aureus fixation 43

prewarmed 0.4% (w/v) agarose (139). The agarose-isolate dilute was then spotted

(10 µl) to slides not treated with agarose and fixed with a 60 ◦C hotplate until dry.

For specimens that were heat-fixed to plain glass slides, rinsing these slides in 1 M

urea was more effective than agarose at reducing cell loss (data not shown). As a

further improvement to the use of slides with FISH, the specimen or reagent run-off

was contained by marking the slides with a wax-pencil (Staedtler R©, Chinagraph).

2.3 Step 2: S. aureus fixation

Fixation was necessary to inactivate the pathogens, avoid cell lysis in Gram-negative

bacteria such as EC and to improve the consistency of permeabilization and hybridiza-

tion of the FISH probes. To fix as well as partly permeabilize pathogens, slides with

were washed in 50 ml sterile tubes with either absolute methanol or ethanol for three

minutes (32). Slides were removed and dried on a 60 ◦C hot-plate.

For more rapid fixation, slides of SA were spotted with alcohol, left on the bench

for one minute and then dried on the hot-plate. Fixation with methanol produced a

more consistent FISH signal, but was toxic to use. Fixation with ethanol was less toxic

and produced a higher, but also a less consistent FISH signal.

2.4 Step 3: S. aureus permeabilization

Gram-negative bacteria were fixed with alcohol and did not need permeabilization.

Permeabilization with enzymes was necessary for DNA probes to reach in situ targets

in Gram-positive pathogens such as SA. Most Gram-positive bacteria lysed rapidly

with lysozyme. SA permeabilizes slowly with lysozyme, but quickly with lysostaphin

(147). To cut preparation time, stock solutions of 30 µg/ml lysozyme (Sigma, L6876)

and 2 µg/ml lysostaphin (Sigma, L4402) (32) were prepared and stored in 1.5 ml sterile

aliquots. These solutions were frozen and stored long-term at -20 ◦C. Unless frozen,

the enzymes gradually lost their permeabilizing activity.

For use, the solutions were thawed, diluted 1:1 with MQ water and 40 µM Tris-HCl


for buffering and used within a week. Lysozyme was most active at pH 7.0 and at 37

◦C in the absence of NaCl (141). Lysostaphin was most active at pH 8.0 and at 47 ◦C

in the absence of formamide (95). To permeabilize in a single step, 2 mg/ml lysozyme

and 0.1 mg/ml lysostaphin (147) at pH 7.0 was spotted to the slides and incubated at

47 ◦C for five minutes in 50 ml tubes (Greiner, 210-261) (2) before rinsing the slides in

absolute methanol (32).

To permeabilize SA in two steps, 10 µl of lysozyme at 15 mg/ml in MQ water

(147, 4), was spotted onto the slide wells and incubated in 50 ml tubes (Greiner, 210-

261) for six minutes at 38 ◦C (141). The lysozyme was rinsed off with PBS (Sigma,

P4417) and the slides were rapidly dried with pressurized air (32) or by centrifuging in

50 ml tubes for one minute at 100 rcf. The order of their application mattered; lysozyme

followed by lysostaphin permeabilization was more effective than in the reverse order

(52). Permeabilization was stopped by rinsing the slides again in absolute methanol.

Permeabilization that does not require lysostaphin

To permeabilize without lysostaphin, 10 µl of freshly prepared 15 mg/ml lysozyme in

unbuffered MQ water (147) was spotted to the slides and incubated in 50 ml tubes at

47◦C for 30 minutes before rinsing the slides in absolute methanol (4). If permeabi-

lization was performed on a 47 ◦C hot-plate, the treatment was the same except that

the slides were covered with a clear plastic lid and the incubation was extended to 40

minutes (5). The hot-plate was developed by one of the authors (Russell Connally)

and had an accuracy of ± 0.5 ◦C at 47 ◦C. If the permeabilization was applied to SA

separated from blood, incubation was extended to one hour.

After its permeabilization and before its hybridization, the SA rRNA can be de-

graded by endogenous RNase. To reduce loss from RNase, the equipment and most

of the reagents can be treated with RNase-Zap (Ambion, AM9780), 0.1% diethylpy-

rocarbonate (DEPC) (Aldrich, 159220) or autoclaved. However, in this project, no

difference was observed in the FISH signal with or without this treatment if clean

laboratory standards were maintained, pre-sterilized polypropylene plastic disposables

and MQ water were used and gloves were regularly changed.

2.5 Step 4: In situ hybridization with rRNA-targeted, fluorescentlylabeled oligonucleotides 45

2.5 Step 4: In situ hybridization with rRNA-targeted,

fluorescently labeled oligonucleotides

Incubation in hybridization buffer binds oligonucleotides to their complementary se-

quences. Incubation in the washing buffer that follows, washes away probe that is not

fully hybridized (84). The hybridization buffer uses formamide at a NaCl concentration

of 0.9 M (153) to adjust its stringency (Figure 2.3) (142). In contrast, the washing

buffer uses varying amounts of NaCl to adjust its stringency as formamide is toxic. The

formamide concentration in the hybridization buffer is dependent on its oligonucleotide

sequence (Figure 2.1) (92). In turn, the NaCl concentration in the washing buffer is

dependent on the formamide concentration used in the hybridization buffer (Table 2.2)

(143).

The FISH assay reported by Poppert et al. (32) was applied with changes. So that

a single incubator or water-bath could be used, all steps in the assay were set to 47 ◦C

(2). To reduce the reaction time and the drying out of reagents, for slide incubations,

preheated 50 ml centrifuge tubes with screw-caps (Greiner, 210-261) were used (3). To

simplify reagent preparation, the hybridization buffers in the project mostly used 30%

formamide and a washing buffer set to 0.225 M NaCl (1).

For hybridization, 10 µl of buffer [30 % formamide (v/v), 0.9 mol/L NaCl (153), 20

mM Tris-HCl pH 8.0, 0.02 % (v/v) SDS, 0.5 µg/ml DAPI, and Milli-Q water] with 1

µM of oligonucleotide probe was spotted to the slides, the slides were fitted in 50 ml

tubes and placed in a 47 ◦C incubator for 20 minutes (32). If urea was substituted for

formamide (153), 30 µl of urea-NaCl [1 mol l−1 urea (Sigma, U6504), 0.9 mol l−1 NaCl,

20 µmol l−1 Tris-HCl (pH 7.0) in MQ water] with 1 µmol l−1 of probe was spotted to

each well (5) and the slides were incubated as before.

If impure or uncultured specimens were tested and non-specific binding was high,

a FISH assay that was more complex than the common assay for SA was used (116).

The hybridization buffer [35 mM Tris-HCl pH 7.5, 2.5× standard sodium citrate buffer


(SSC), 5 mM EDTA, 0.05% SDS, 0.05% Na-Pyrophosphate, 0.45 M NaCl, 22.5% deion-

ized formamide] also contained blocking agents [2.5×Denhardt’s and 50 µg/ml herring-

sperm-DNA]. The hybridization buffer was incubated twice with the specimen. The

first time without the oligonucleotide probe to block non-specific binding, and the sec-

ond time for hybridization with the probe. The washing buffer that followed contained

2×SSC.

Figure 2.3: The melting temperature (Tm) in ◦C of 1 µM of the Staaur probe to SAby formamide (Fa %) or NaCl (M) concentration (92).

2.6 Step 5: Specimen washing with buffer 47

2.6 Step 5: Specimen washing with buffer

After hybridization, slides were immediately fitted into 50 ml tubes of prewarmed

washing buffer [5 mM EDTA (Sigma, EDS), 0.64 M NaCl, 20 mM Tris-HCl and 0.02%

(w/v) SDS in MQ water] (142). Tubes were then placed in a 47 ◦C water bath for

three minutes and agitated (4). This washing action was stopped by briefly rinsing the

slides at room temperature in a 50 ml tube of MQ water (32).

Buffers other than the conventional NaCl-based washing buffer were also tested.

Preheated PBS was used to remove unbound probes since PBS has a surfactant quality

and its ionicity is approximately 0.15 M (2). The results were not as specific as a full

washing buffer, but were simple to apply and adequate for rapid testing (3). Preheated

urea with NaCl was also tested to remove unbound probe [8 mol l−1 urea, 0.9 mol l−1

NaCl (153), MQ water and 20 µmol l−1 Tris-HCl (pH 7.0)] (5). After washing, the

slides were mounted while wet for viewing with a cover-slip.

If biotinylated oligonucleotides were used (2), after the washing step, the slides were

dried with pressurized air and then spotted with 10 µL of streptavidin conjugated to

Alexa Fluor R© 488 (Invitrogen, S-32354), DyLight R© 488 (Thermo Fisher, 21832) or

Alexa Fluor R© 555 (Invitrogen, S-32355) at 10 µg/ml in PBS (145, 45). Slides were

incubated at 47 ◦C for 10 minutes, rinsed with PBS and mounted as before for viewing.

For time-resolved europium chelates labeled with FISH, the hybridization buffer

was rinsed off with MQ water, the slides air-dried and 10 µl of fluorescence enhancing

buffer (FEB) buffer (148) containing 0.4 mM Eu3+ was spotted to each well. The

slides were mounted while wet with a cover-slip and left at room temperature for 20

minutes before viewing.

2.7 Direct visualization of microorganisms

SA on the slides were observed with an epifluorescence microscope (Olympus, BX51)

fitted with a 40 or 60× dry objective (Olympus, UPLFLN) and FITC/DAPI filters

(Olympus, U-MWU2, U-MWIB2). Images were acquired at a resolution of 1360×1024


with a color camera (Olympus, DP72) and software (Olympus, DP2-BSW v2.2) set to

a gain of 200 ISO and an exposure of 0.5 to 2 seconds.

Fluorophores that become excited at a particular wavelength, will always emit at

a wavelength that is longer. A blue light excited fluorophore will usually emit green,

a green excited fluorophore will emit red and so on (72). The signal intensity of the

FISH probe is dependent on the type of fluorophore and oligonucleotide used and the

accessibility and abundance of the in situ rRNA (84). Alexa Fluor R© (Invitrogen) and

Dylight R© (Jackson) fluorochromes have a higher-yield than dyes such as Cy3 and Cy5

cyanine. PNA has a higher binding affinity than DNA (102). SA in exponential growth

phase has a higher number of rRNA than SA in stasis and the target rRNA sequence

for EUB338 is more accessible than the target for Staaur (data not shown).

SA could be identified by its specific oligonucleotide signal and by its arrangement

on the slide even after first culturing. Cultures of SA tended to cluster in tetrad

arrangements and cultures of SE in staphylococci arrangements (48, 79). SA and SE

differed also in their reaction to permeabilization treatments (32). SA is more sensitive

to lysostaphin and SE to lysozyme (147). If only lysostaphin was applied, SA labeled

brightly and SE did not. If only lysozyme was applied, SE labeled brightly and SA did

not.

Time-gated bio-imaging of a europium chelate label

For the TGLM visualization of SA, slides were viewed with an epifluorescence micro-

scope (BX51, Olympus) and a 40 and 60× objective (UPLFLN, Olympus) fitted with

a time-gated auto-synchronous luminescence detector (GALD) held in its DIC prism

slot (148). The short-lived background signal of the specimen was removed and the

long-lifetime probe emission was detected by gating the emission signal. The GALD

device was excited with 355 nm UV from a 100 mW YAG laser source. It used a

rotating element that simultaneously pulsed the specimen, suppressed its short-lived

autofluorescence and allowed the passage of long-lived probe emission (148).

2.7 Direct visualization of microorganisms 49

2.7.1 FISH image and statistical analysis

A representative image of each FISH experiment, that had a SA count of at least

100, was selected for analysis. These images were analyzed using standard algorithms

with ImageJ (NIH, v1.43u). Counts, morphology and permeabilization of the SA were

assessed against a 50 µm haemocytometer grid, the FISH signal and DAPI staining

(2). The SA cells were then masked with automatic thresholding so that the mean

FISH signal in 8-bit Grey-scale, the size of the cells and the ratio of cells with signal

to those without, could be calculated with FISH and DAPI (3).

For statistical analysis, parametric assumptions were tested with a histogram of

the FISH signal and a P value of < 0.05 was considered significant. The mean signal

intensity in 8-bit gray-scale, standard deviation and its 95% confidence interval were

calculated. The summary statistics of a new FISH treatment were compared to a con-

trol (32) with either an unpaired two-tail t-test or with a one-way analysis of variance

(ANOVA) to test for a significant difference (3).

3Improvements to the existing FISH method

The project investigated the use of FISH for the detection of S. aureus (SA) (19, 32).

This Chapter reports on improvements that were made to the FISH assay during its

investigation. It includes an assessment of established and newly identified probes

(often called oligonucleotides) specific for SA (1). The shortening of the preparation

and storage of reagents for the FISH assay is described (2). It recounts an attempt

to optimize the permeabilization of SA so that probes of large-molecular weight can

access SA 16S rRNA (2) without lengthening the FISH assay beyond one hour (101).

Finally, this Chapter reports on the identification of SA with a FISH assay that can

be completed in half the time (3) of the previous fastest reported FISH assay (32).

The Chapter comprises of three sections. Each of these sections was published in a

peer reviewed journal and included as such. In the first paper, an in silico evaluation

and testing of FISH probes that target 16S rRNA SA is described: Lawson TS, Con-

nally RE, Vemulpad S, Piper JA. In silico evaluation and testing of fluorescence in situ

hybridization 16S rRNA probes for Staphylococcus aureus. Lab Med 2011;42:587-591

(1). In the second paper, optimization of a two-step permeabilization technique for SA

is described: Lawson TS, Connally RE, Vemulpad S, Piper JA. Optimization of a two-

step permeabilization fluorescence in situ hybridization (FISH) assay for the detection

of Staphylococcus aureus. J Clin Lab Anal 2011;25:359-365 (2). In the third paper, the

51

52 Improvements to the existing FISH method

procedure for substantially shortening the typical FISH assay for SA is described: Law-

son TS, Connally RE, Vemulpad S, Piper JA. Express fluorescence in situ hybridization

methods for the detection of Staphylococcus aureus. Clin lab 2011;57:789-794 (3).

labmedicine.com December 2011 ■ Volume 42 Number 12 ■ LABMEDICINE 1

Science

Staphylococcus aureus is a clinically important pathogen.1-3 Whole-cell slide based fluorescence in situ hybridization (FISH) is a molecular assay that can reliably detect and dif-ferentiate S. aureus from S. epidermidis.4-6 Fluorescence in situ hybridization detection involves hybridizing small subunit ribosomal ribonucleic acid (16S rRNA) with DNA probes.7 The accuracy of FISH is dependent upon the hybridization efficiency (HE) of its probes to S. aureus and to non-targets such as S. epidermidis.5,6,8,9 The number of probes reported to be specific to S. aureus is small.3,6,8,10,11 Until recently, probe design calculations were limited to online software provided for polymerase chain reaction (PCR).12 With new online software tools available,13,14 and mathFISH specific for FISH probe design,15 it may be possible to predict and compare

the accuracy of these existing probes as well as identify better probes that can target S. aureus.

Staphylococcus aureus also provides a unique opportunity to test the accuracy of these tools. Because of the similarity between S. aureus and S. epidermidis 16S rRNA, there are few possible misaligned sequences that can be targeted by probes. This is evident in the overlap of all reported probes for S. aureus about the 16S69 5'- AAGCTTCTCGTCCG -3' se-quence as illustrated in Figure 1. Even so, the number of pos-sible probes per misaligned sequence can still be large; about 450 for the 16S69 sequence. Online tools such as NCBI-Nucleotide,13 Ribosomal Database Project,16 NCBI-Blast,13 Reverse-Complement,12 Probecheck,14 and mathFISH15 are useful as they can rapidly characterize these sequences. The accuracy of the tools can be assessed thoroughly by comparing predicted results to those in the laboratory from this and pre-vious studies for the 16S69 sequence. Where probe names are listed, the original author designation is chosen first, otherwise the common name or the author’s initials with the number of bases is given.

Materials and Methods

Identifying 16S rRNA S. aureus Probes In Silico

The S. aureus and non-target 16S rRNA sequences were acquired from NCBI-Nucleotide (ncbi.nlm.nih.gov/nuc-core),17 the Ribosomal Database Project identified closely

In Silico Evaluation and Testing of Fluorescence In Situ Hybridization 16S rRNA Probes for Staphylococcus aureusThomas S. Lawson, MSc, Russell E. Connally, PhD, Subramanyam Vemulpad, PhD, James A. Piper, PhD

(Faculty of Science, Macquarie University, New South Wales, Australia)

DOI: 10.1309/LMI4L6CF6HGFBGYA

Abstract

Background: Staphylococcus aureus is a clinically important pathogen. A small number of whole-cell fluorescence in situ hybridization (FISH) probes have been reported to detect S. aureus. New online computational tools for in silico design and testing make it possible to assess candidate FISH probes for S. aureus. Materials and Methods: Six online tools, NCBI-Nucleotide, Ribosomal Database Project, NCBI-Blast, Reverse-Complement, Probecheck, and mathFISH, were employed in a workflow

to evaluate FISH probes for S. aureus. A previously reported probe Staaur-16S69 was compared to a new probe KT18-16S68 predicted by mathFISH to have the same performance. Results: A number of new probes for S. aureus were predicted to perform as well or better in silico as those previously reported. When tested in a FISH assay, Staaur and a new probe KT18 were found to have the same performance.

Conclusion: Existing and new FISH probes for S. aureus were found to be accurately identified and characterized with online computational tools. In silico evaluation of probes has the potential to reduce the time spent evaluating probes in the laboratory. Keywords: fluorescence in situ hybridization, FISH, hybridization efficiency, mathFISH, probes, Staphylococcus aureus

Submitted 2.25.2011 | Revision Received 4.25.2011 | Accepted 5.13.2011

Corresponding Author

Thomas S. Lawson, MSc

[email protected], [email protected]

Abbreviations

FISH, fluorescence in situ hybridization; HE, hybridization effi-

ciency; PCR, polymerase chain reaction; PNA, peptide nucleic acid;

CoNS, staphylococci; FITC, fluorescein isothiocyanate

3.1 In silico evaluation and testing of fluorescence in situhybridization 16S rRNA probes for Staphylococcus aureus 53

3.1 In silico evaluation and testing of fluorescence in situ hybridization 16S

rRNA probes for Staphylococcus aureus

Science

2 LABMEDICINE ■ Volume 42 Number 12 ■ December 2011 labmedicine.com

related microbes such as S. epidermidis (rdp.cme.msu.edu),16 NCBI-Blast aligned S. aureus and S. epidermidis sequences NCBI-Blast (blast.ncbi.nlm.nih.gov),13 and the cross-reactiv-ity to other microbes of identified mismatch sequences was confirmed with Probecheck (microbial-ecology.net/probe-check) against the SILVA sequence collection.14 Staphylococcus

aureus (GenBank: CP000253.1) and S. epidermidis (GenBank: AF397060.1) 16S ribosomal RNA alignment was compared, and mismatches were identified at the 69 to 89, 183 to 198, 452 to 477, and 999 to 1024 positions relative to E. coli 16S rRNA.13,14 The sequence of 999 to 1024 was not consid-ered further as it was found to hybridize with a number of Staphylococcus species including S. haemolyticus, which is fre-quently detected in blood cultures.2,3

Next, the mathFISH tool analyzed the HE of each mismatch 16S rRNA sequence (mathfish.cee.wisc.edu).15 The 183 to 193 and 452 to 477 sequences were assessed to have poor HE by the oligonucleotide walk-though feature of mathFISH set at 18 bases.18 The first mismatch 16S rRNA sequence at 69 to 89 was analyzed in more detail. The oli-gonucleotide walk-though feature of mathFISH was applied again at 18, 19, 22, and 25 bases lengths, which matched the lengths of previously reported S. aureus probes for this sequence.3,6,8,11 A pattern emerged where the 5' end of a potential probe was most efficient at the 65 to 67 positions. Probes from 15 to 30 bases were tested at this location using the mismatch feature of mathFISH,19 and a number of poten-tial probe candidates were realized including probes already reported (Table 1 and Table 2). For the calculations, 0.9 NaCl in the hybridization buffer, 47°C hybridization incubation, and 1 µM of probe was assumed.20

Figure 1_Alignment of S. aureus probes with S. aureus and S. epi-dermidis sequence rRNA. Mismatches to S. epidermidis at 16S72, 16S79, and 16S88 are highlighted. As genome 16S rRNA was aligned, thymine (T), instead of RNA’s uracil (U) is shown.

Sau66-16S66

KT18-SA68

Staaur-16S69

Staur-16S69

WQ25-16S67

S. aureus

S. epidermidis

5’ end 3’ endCommon sequence

3’ end 5’ end

Table 1_The Predicted Hybridization Efficiency of Previously Reported DNA Probe Sequences to S. aureus 16S rRNA and Not to S. epidermidis

Name* DNA probe (5'– 3')† FA %‡ SA ∆G kcal/mol§ SE ∆G kcal/mol# HE¶

Sau66-16S663 AAGCTTCTCGTCCGTTCG 29.9 –14.1 –4.7 1.00

WQ25-16S676 AGAGAAGCAAGCTTCTCGTCCGTTC 42.3 –14.8 –5.4 1.00

JG24-16S6810 AGAGAAGCAAGCTTCTCGTCCGTT 41.9 –15.0 –5.3 1.00

Staaur-16S6911 GAAGCAAGCTTCTCGTCCG 23.7 –12.1 –5.8 0.99

Staur-16S698 AGAGAAGCAAGCTTCTCGTCCG 30.6 –12.2 –4.5 0.99

*Probe designation (name) includes number of bases and 5 ' end location relative to the 16S rRNA sequence of E. coli.†Thermodynamic calculations assume a single DNA probe binding to a target 16S rRNA sequence.‡Formamide concentration (FA %) assume 1 µM probe dissociation at 47°C in 0.9 NaCl buffer. §Overall Gibbs binding potential to S. aureus (SA DeltaGo kcal/mol). #Overall Gibbs binding potential to S. epidermidis (SE DeltaGo kcal/mol). ¶Difference in hybridization efficiency (HE) was calculated by subtracting HE of the probe to S. epidermidis from the probe to S. aureus.

Table 2_The Predicted Hybridization Efficiency of DNA Probe Sequences Identified in This Study to S. aureus 16S rRNA and Not to S. epidermidis

Name* DNA Probe (5'–3')† FA %‡ SA ∆G kcal/mol§ SE ∆G kcal/mol# HE¶

KT16-16S65 CTTCTCGTCCGTTCGC 33.7 –14.5 –4.9 1.00

KT15-16S66 CTTCTCGTCCGTTCG 20.2 –12.0 –2.4 0.99

KT20-16S66 GCAAGCTTCTCGTCCGTTCG 38.3 –15.0 –5.5 1.00

KT25-16S66 GAGAAGCAAGCTTCTCGTCCGTTCG 43.3 –15.0 –5.1 0.99

KT26-16S66 AGAGAAGCAAGCTTCTCGTCCGTTCG 44.5 –15.3 –4.9 1.00

KT18-16S68 GCAAGCTTCTCGTCCGTT 35.3 –14.8 –5.9 0.99

KT30-16S69 CTAACATCAGAGAAGCAAGCTTCTCGTCCG 47.0 –17.3 –4.3 1.00

*Probe designation (name) includes number of bases and 5 ' end location relative to the 16S rRNA sequence of E. coli.†Thermodynamic calculations assume a single DNA probe binding to a target 16S rRNA sequence.‡Formamide concentration (FA %) assume 1 µM probe dissociation at 47°C in 0.9 NaCl buffer. §Overall Gibbs binding potential to S. aureus (SA DeltaGo kcal/mol). #Overall Gibbs binding potential to S. epidermidis (SE DeltaGo kcal/mol). ¶Difference in hybridization efficiency (HE) was calculated by subtracting HE of the probe to S. epidermidis from the probe to S. aureus.


Science


Assessing 16S rRNA S. aureus Probes In Silico

Table 1 and Table 2 list the DeltaGo for S. aureus and the DeltaGo for S. epidermidis and the difference in HE for probes between S. aureus and S. epidermidis. The DeltaGo indicates the standard state overall Gibbs free energy of the probe-target hybrid: the probability of probe to target binding (DNA:RNA) given the competing interactions of probe (DNA:DNA) and target (RNA:RNA) self-binding.21 The higher the negative number, the stronger the potential probe-target binding. Hybridization efficiency indicated the predicted ratio of target molecules bound with probe to all target molecules. An HE of 1 indicated saturation and 0 no hybridization.

A probe’s usefulness was usually defined by its predicted (and tested) sensitivity and specificity to its target and non-target microbes.2,6,22 Sensitivity refers to how good a probe was at correctly identifying the target microbe S. aureus.23 Specificity, on the other hand, indicated how good the probe was at not binding to non-target microbes.23 Generally the predicted DeltaGo and HE indicates both sensitivity and specificity whereas the formamide dissociation indicates only sensitivity.15

A probe was predicted to have a high sensitivity if it has a DeltaGo between -17 and -13 kcal/mol to S. aureus, a differ-ence in DeltaGo between S. aureus and a non-target such as S. epidermidis greater than 3 kcal/mol, a formamide dissociation concentration difference between S. aureus and S. epidermidis greater than 20% (v/v), an HE greater than 0.9, and a differ-ence in HE between S. aureus and S. epidermidis greater than 0.8.15

A probe should have at least -10 kcal/mol DeltaGo to S. aureus to be sensitive, but it does not need to meet all of the other criteria. A S. aureus probe such as Staphy11 with no dif-ference in DeltaGo and HE to non-targets, but a formamide difference greater than 20%, can still be useful. A probe was predicted to have a high specificity if the DeltaGo to S. aureus was greater than -13 kcal/mol, the DeltaGo to non-targets was less than -10 kcal/mol, and the difference in HE between S. aureus and S. epidermidis was greater than 0.8.15

Testing S. aureus 16S rRNA Probes With FISH

The accuracy of the online tools was tested by compar-ing predicted results to results in the laboratory with FISH. Two probes were compared, Staaur (Invitrogen, Carlsbad, CA, Staaur-16S69: 5'- GAAGCAAGCTTCTCGTCCG -3')11 and a new probe KT18 (Invitrogen, KT18-16S68: 5'- GCAAGCTTCTCGTCCGTT -3') (present study). Staaur is frequently cited in FISH studies of S. aureus.2,4,22 KT18 was chosen as it was predicted to match the HE of Staaur (Table 2).

Specimens were collected from blood agar plates at a major hospital and stored so that 10 isolates of S. aureus and S. epidermidis could be rapidly tested with FISH at a non-clinical location. Isolate identity was confirmed with PCR24 and then de-identified for FISH. After culturing in nutrient broth (Oxoid, Hampshire, U.K., CM0001), isolates were aliquoted and pelleted for up to 3 months storage at -20°C as described by Baldrich and colleagues.25 Before testing with FISH, isolates were thawed and re-cultured for 70 minutes in nutrient broth until turbid (0.5 McFarland).25 No difference was observed in the results when isolates were tested directly from agar plates.

A FISH assay reported by Poppert and colleagues22 was applied with some modifications. To minimize reac-tion time, preheated 50 mL centrifuge tubes with screw-caps (Greiner, 210-261) were used for reagents and slides. As it simplified the assay without a reduction in HE, the hybrid-ization and washing steps were set at 47°C. The hybridiza-tion buffer contained 0.9 NaCl and probes at 1 µM. The 2 S. aureus DNA probes tested and a control eubacteria probe EUB338 probe (Invitrogen, EUB338-16S337: 5'- GCTGCCTCCCGTAGGAGT -3') were conjugated at the 5' end to Alexa Fluor 488 (Invitrogen). As the mathFISH calculations were based upon DNA probes, peptide nucleic acid (PNA) probes were not tested.15 With each experiment, the 2 slides with 5 wells each were tested (Menzel-Gläser, Braunschweig, Germany, X1XER308B).

To compare the 2 S. aureus probes, cells were observed with an epifluorescence microscope (Olympus, Tokyo, Japan, BX51) fitted with a 60× dry objective (Olympus, UPLFLN) and FITC/DAPI filters (Olympus, U-MWU2, U-MWIB2); images were acquired at a resolution of 1360 × 1024 with a color camera (Olympus, DP72) and software (Olympus, DP2-BSW v2.2) set to a gain of 200 ISO and an exposure of 400 ms; and analyzed with ImageJ using standard algorithms (NIH, v1.43u). Cell counts were estimated with a 50 µm haemacytometer grid and a DAPI (Sigma, St. Louis, MO, D9564) counter-stain. Signal intensity was measured by seg-menting images with the same threshold level. For statistical analysis, parametric assumptions were tested with a histogram of the signal and a P value of less than 0.05 was considered significant. The mean signal intensity 8-bit grayscale, stan-dard deviation, and a 95% confidence interval for each probe image was calculated. Summary statistics were compared with an unpaired 2-tail t test. The ratio of cells with FISH and DAPI signal to those with just DAPI signal was also measured for both probes.

Results

Assessing 16S rRNA S. aureus Probes In Silico

A number of new probes for S. aureus were predicted to perform as well or better in silico than those previously cited. Table 1 lists probe sequences previously tested and found to be sensitive and specific to S. aureus. Table 2 lists probe sequences predicted in this study as sensitive and specific to S. aureus. Because of the single viable region on S. aureus for probe targeting (Figure 1), all the potential probes and the reported probes overlapped and, in some cases, were almost identical. It is apparent from Table 1 and Table 2 that KT15-16S66 is identical to KT16-16S65, except that the latter was missing the last base at the 3' end; and WQ25-16S67 is identical to KT26-16S66, except that the latter was missing the last base at the 3' end.

A majority of probes in Table 1 and Table 2 had a pre-dicted high sensitivity to S. aureus; DeltaGo to S. aureus was between -17 and -13 kcal/mol, the DeltaGo to S. epidermidis was less than -10 kcal/mol, a formamide difference greater than 20% (not shown) and an HE difference greater than 0.8. A majority of probes also had a predicted high specificity to S. aureus; DeltaGo to S. aureus was greater than -13 kcal/mol, the DeltaGo to S. epidermidis was less than -10 kcal/mol, and the difference in HE was greater than 0.8. The reported


Science


probes Staaur-16S69, Staur-16S69, and the newly identified probe KT15-16S66 had DeltaGo lower than -13 kcal/mol, which suggests they might be marginally prone to false nega-tives as compared to the other probes.

Table 1 and Table 2 also list the predicted formamide melting concentration (v/v) as a percentage for each probe. This was the stringency at which half the probe has annealed (or dissociated) from the target. The melting point formamide concentrations ranged from 20.2% (v/v) for the shortest probe KT15-16S66 to 47.0% for the longest probe KT30-16S69. It should be noted that the predicted formamide concentration is not the actual concentration used by the hy-bridization buffer and FISH for that probe. The hybridization buffer should have a lower stringency, and so the formamide concentration used is typically 5% to 10% higher than what was predicted.2,6,22 The washing buffer stringency (set with NaCl and not formamide) is higher than the hybridization buffer.2,6,22

Testing S. aureus 16S rRNA Probes With FISH

Except for Staur-16S69, the reported sensitivity and spec-ificity of S. aureus probes used in other studies were similar to the sensitivity and specificity predicted in this study. The studies by Jansen and colleagues,10 Tavares and colleagues,3 and Poppert and colleagues22 were completed in a clinical setting with blood cultures that contained S. aureus, coagu-lase-negative Staphylococci (CoNS), Micrococcus spp., and Rotia spp. Probe sensitivity and specificity to S. aureus were as follows: Sau66-16S66, 100% and 99%;3 JG24-16S68, no data available; Staaur-16S69, 100% and 100%;22 and Staur-16S69, 68% and 100%.10 The poor Staur-16S69 findings in Jansen and colleagues10 were confirmed elsewhere26 and were attributed to either differences in the permeability5 of the S. aureus cell wall or steric hindrance.26 A study by Wu and col-leagues6 was also completed in a clinical setting, but the study tested urinary tract infection samples containing S. aureus, Escherichia coli, and Enterococcus faecalis directly without first

culturing. The sensitivity and specificity of the WQ25-16S67 probe used to detect S. aureus was 100% and 98% respec-tively.6

When tested with a FISH assay in this study, Staaur-16S69 and a new probe KT18-16S68 were found to have a similar signal intensity to S. aureus and similar specificity to S. epidermidis. The FISH experiment was performed twice. First, to ascertain the optimal formamide concentration, each probe was tested at 0%, 15%, 30%, 45%, and 60% formamide (v/v). The optimal formamide concentration was found to be 30% for both Staaur and KT18. Next, the signal intensity of the probes was compared at 30% formamide. No difference in signal was observed in Image 1. Lastly, S. aureus and S. epidermidis were mixed and analyzed in the same sample to determine interference, sensitivity, and the limit of detection. Staphylococcus aureus was clearly identified with either the Staaur or the KT18 probes at 10(3) density amongst S. epi-dermidis at 10(8). To ensure that S. epidermidis was not also labeled with these probes, a polyclonal antibody conjugated to fluorescein isothiocyanate (FITC) and specific to S. aureus (ViroStat, Portland, ME) was applied to the slides after FISH. The immunofluorescence signal labeled the outside of only those cells that had a FISH signal.

The signal intensity was measured with ImageJ and found to be 28.201 ± 0.59 for Staaur and 28.614 ± 0.52 for KT18 (8-bit grayscale). An unpaired t test of the signal intensity for each probe was not significantly different (cell count >200; P=0.39). The ratio of cells with FISH signal to those without was also compared. No difference was measured between the 2 probes. For both FISH experiments, no cross-reactivity was observed; S. aureus was positive for the Staaur and KT18 probes and S. epidermidis was not.

Discussion

We set out to test the efficacy of a number of online tools to evaluate FISH probes for S. aureus. Since S. aureus has a single sequence that can be targeted for FISH, it was considered

Image 1_S. aureus labeled with (A) Staaur probe and (B) KT18 probe were compared. Both probes were conjugated with Alexa Fluor 488. No difference was observed. Bar is 10 µm.

A B


Science


unlikely that any new probes determined in silico would be as efficient as those previously reported (Table 1). It was therefore surprising that a number of new probes for S. aureus were predicted to perform as well or better in silico (Table 2). In particular, KT26-16S66 and KT30-16S69 were found to have a high predicted difference in DeltaGo and HE.18 When tested with a FISH assay, Staaur and a new probe KT18 were found to have similar HE18 and melting formamide point concentrations27 as predicted. In addition, a new online tool, mathFISH, was found to offer a number of advantages in the development of FISH probes.15 First it calculated the interac-tions observed in FISH between the probe and the target, self-folding within the probe itself and within the rRNA target. Next, it predicted the most efficient 5' end location for a DNA probe within a mismatch sequence,19 the HE for probes of different lengths at that location,18 and the melting formamide point concentration of selected probes.27

A general probe for Staphylococcus was also tested in silico. The results were more difficult to assess than with S. aureus as the number of probe possibilities and poten-tial non-target microbes were larger and were complicated by mismatches spanning across more than 1 base. The speed advantage of mathFISH to walk an oligonucleotide of a set length across a sequence range became more ap-parent when more than 1 target sequence was analyzed.21 Potential target sequences were sought by aligning and analyzing mismatches between S. aureus and Streptococcus agalactiae (GenBank, HQ658089.1). Candidate probe sequences were also compared with Enterococcus faecalis (GenBank, FJ749378.1) and Micrococcus luteus (GenBank, HQ323416.1) as these gram-positive aerobic cocci are fre-quently encountered in blood culture and can be confused with S. aureus.1-3 A probe for Staphylococcus RB17-16S696 5'- CTCCATATCTCTGCGCA -3', was found to be at least as efficient as the reported Staphy probe (Staphy-16S697 5'- TCCTCCATATCTCTGCGC -3').11 But further testing in the laboratory with FISH is needed to confirm its specificity.

This study had a number of limitations. The large sub-unit rRNA was not assessed in silico,7 nor was a reported 23S rRNA probe Saur72 analyzed.9 A representative GenBank sequence was chosen for S. aureus and S. epidermidis. The in silico calculation assumptions were experimentally dem-onstrated on E. coli after an extended hybridization step by Yilmaz and Noguera,21 but the FISH assay in the present study verified the in silico results on S. aureus with a 10 min-ute incubation step.22 Hybridization efficiency calculations assumed no added formamide,18 but in practice formamide is rarely excluded.2,4,22 When confirming the computed predic-tions with a FISH assay, only 1 probe candidate KT18 was tested with FISH against a known probe Staaur. The FISH assay was run on pure cultures of patient isolates and not on reference strains or as is usually the case, with clinical FISH, directly from blood cultures.2,4,22 Since there was some varia-tion in the signal observed between slides and between FISH experiments, representative images were chosen from slides treated the same day with FISH and from the same location on each slide. Image analysis was complicated by the inclusion of cells at a lower signal outside the focal plane and by DAPI bleeding.

In conclusion, the characterization of existing and new probes for S. aureus was greatly enhanced by in silico testing. We were able to assess the HE of KT18-SA68 to S. aureus in silico and confirm these calculations with minimal FISH

testing. To determine their applicability and reliability, the probe sequences predicted in this study as specific for S. au-reus, warrant further testing with FISH in a routine clinical setting against positive blood cultures containing a variety of additional Staphylococcus spp. LM

Acknowledgements: The study was supported by the Australian Research Council’s Linkage Projects (LP0775196). Our thanks to the Australian Proteome Analysis Facility for laboratory access.

1. Kempf VA, Trebesius K, Autenrieth IB. Fluorescent in situ hybridization allows rapid identification of microorganisms in blood cultures. J Clin Microbiol. 2000;38:830-838.

2. Wang P. Simultaneous detection and differentiation of Staphylococcus species in blood cultures using fluorescence in situ hybridization. Med Princ Pract. 2010;19:218-221.

3. Tavares A, Inácio J, Melo-Cristino J, et al. Use of fluorescence in situ hybridization for rapid identification of staphylococci in blood culture samples collected in a Portuguese hospital. J Clin Microbiol. 2008;46:3097-3100.

4. Gescher DM, Kovacevic D, Schmiedel D, et al. Fluorescence in situ hybridisation (FISH) accelerates identification of Gram-positive cocci in positive blood cultures. Int J Antimicrob Agents. 2008;32(suppl 1):S51-S59.

5. Jansen GJ, Mooibroek M, Idema J, et al. Rapid identification of bacteria in blood cultures by using fluorescently labeled oligonucleotide probes. J Clin Microbiol. 2000;38:814-817.

6. Wu Q, Li Y, Wang M, et al. Fluorescence in situ hybridization rapidly detects three different pathogenic bacteria in urinary tract infection samples. J Microbiol Methods. 2010;83:175-178.

7. Amann R, Fuchs BM. Single-cell identification in microbial communities by improved fluorescence in situ hybridization techniques. Nat Rev Microbiol. 2008;6:339-348.

8. Bentley RW, Harland NM, Leigh JA, et al. A Staphylococcus aureus-specific oligonucleotide probe derived from 16S rRNA gene sequences. Lett Appl Microbiol. 1993;16:203-206.

9. Veeh RH, Shirtliff ME, Petik JR, et al. Detection of Staphylococcus aureus biofilm on tampons and menses components. J Infect Dis. 2003;188:519-530.

10. Jansen G, Degener J, Welling G. Method for the rapid determination of bacteria. Eur pat. 1999. WO Patent WO/1999/054,502.

11. Trebesius K, Leitritz L, Adler K, et al. Culture independent and rapid identification of bacterial pathogens in necrotising fasciitis and streptococcal toxic shock syndrome by fluorescence in situ hybridisation. Med Microbiol Immunol. 2000;188:169-175.

12. Stothard P. The sequence manipulation suite: JavaScript programs for analyzing and formatting protein and DNA sequences. Biotechniques. 2000;28:1102-1104.

13. Johnson M, Zaretskaya I, Raytselis Y, et al. NCBI BLAST: A better Web interface. Nucleic Acids Res. 2008;36:W5-W9.

14. Loy A, Arnold R, Tischler P, et al. probeCheck—a central resource for evaluating oligonucleotide probe coverage and specificity. Environ Microbiol. 2008;10:2894-2898.

15. Yilmaz LS, Parnerkar S, Noguera DR. mathFISH, a Web tool that uses thermodynamics-based mathematical models for in silico evaluation of oligonucleotide probes for fluorescence in situ hybridization. Appl Environ Microbiol. 2011;77:1118-1122.

16. Cole JR, Wang Q, Cardenas E, et al. The Ribosomal Database Project: Improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009;37:141-145.

17. Madden T. The NCBI Handbook. Bethesda, MD: National Center for Biotechnology Information; 2003;16:1-17.

18. Yilmaz LS, Okten HE, Noguera DR. Making all parts of the 16S rRNA of Escherichia coli accessible in situ to single DNA oligonucleotides. Appl Environ Microbiol. 2006;72:733-744.

19. Yilmaz LS, Bergsven LI, Noguera DR. Systematic evaluation of single mismatch stability predictors for fluorescence in situ hybridization. Environmental Microbiology. 2008;10:2872-2885.

20. Szweda P, Kotlowski R, Kur J. New effective sources of the Staphylococcus simulans lysostaphin. Journal of Biotechnology. 2005;117:203-213.

21. Yilmaz LS, Noguera DR. Mechanistic approach to the problem of hybridization efficiency in fluorescent in situ hybridization. Appl Environ Microbiol. 2004;70:7126-7139.


Science


22. Poppert S, Riecker M, Wellinghausen N, et al. Accelerated identification of Staphylococcus aureus from blood cultures by a modified fluorescence in situ hybridization procedure. J Med Microbiol. 2010;59:65-68.

23. Loong T. Understanding sensitivity and specificity with the right side of the brain. BMJ. 2003;327:716-719.

24. Thomas LC, Gidding HF, Ginn AN, et al. Development of a real-time Staphylococcus aureus and MRSA (SAM-) PCR for routine blood culture. J Microbiol Methods. 2007;68:296-302.

25. Baldrich E, Vigués N, Mas J, et al. Sensing bacteria but treating them well: Determination of optimal incubation and storage conditions. Anal Biochem. 2008;383:68-75.

26. Ikeda M, Yamaguchi N, Tani K, et al. Development of phylogenetic oligonucleotide probes for screening foodborne bacteria. J Health Sci. 2005;51:469-476.

27. Yilmaz LS, Noguera DR. Development of thermodynamic models for simulating probe dissociation profiles in fluorescence in situ hybridization. Biotechnol Bioeng. 2007;96:349-363.


Journal of Clinical Laboratory Analysis 25 : 359–365 (2011)

Optimization of a Two-Step Permeabilization FluorescenceIn Situ Hybridization (FISH) Assay for the Detection

of Staphylococcus aureusThomas S. Lawson,� Russell E. Connally, Subramanyam Vemulpad,

and James A. PiperMacquarie University, Faculty of Science, Sydney, New South Wales, Australia

Background: Aspects of the fluorescence insitu hybridization (FISH) method for thedetection of clinically important bacteria,such as Staphylococcus aureus, Staphylo-coccus epidermidis, and Escherichiacoli, were investigated for optimization.Methods: Various approaches to optimizingthe FISH procedure were taken and differ-ent methods were compared. To save time,hybridization and washing buffers wereprepared beforehand and stored at �201Cand mixed to their final formamide andNaCl concentrations just before use.The use of 50-ml tubes for hybridization

incubation reduced drying out, reagentwastage, and reaction times. Results:A two-step permeabilization FISH assaywas developed that used phosphate-bufferedsaline as a buffer for lysostaphin. It coulddetect bacteria with DNA probes conjugatedto fluorophores with a higher signal intensityand the less expensive biotinylatedDNA probes with minimal cell lysis in 1 hr.Conclusions: The two-step assay might beused when the FISH signal is weak, bacterialnumbers are low or if there is a need to useother reporter molecules. J. Clin. Lab. Anal.25:359–365, 2011. r 2011 Wiley-Liss, Inc.

Key words: fluorescence in situ hybridization; FISH; Gram-positive bacteria; moleculardiagnostic techniques; Staphylococcus aureus; Staphylococcus epidermidis;Staphylococci

INTRODUCTION

Following a positive blood-culture and Gram-stain,fluorescence in situ hybridization (FISH) can be used toidentify the bacteria present such as the clinically importantStaphylococcus aureus (1–4). The FISH procedure typicallyuses a single permeabilization step (hereafter referred to asthe one-step FISH assay) and DNA probes (also calledoligonucleotides or oligos) conjugated to fluorophores(1,5–8). To permeabilize S. aureus, the one-step FISHassay applies a lytic enzyme mixture of lysozyme andlysostaphin. As it can be more robust, FISH can also use atwo-step permeabilization (two-step FISH assay) to detectS. aureus (2,4,9–14). To permeabilize S. aureus, two-stepFISH assay applies a lysozyme step, and a quick waterrinse followed by a lysostaphin step.The DNA probes conjugated to fluorophores (here-

after, oligo-f) are relatively small in molecular weightand so can gain rapid access to in situ rRNA targets.The detection of S. aureus with FISH and biotinylatedprobes (hereafter, oligo-b) is rarely reported (15–17).

Greater permeabilization is needed for streptavidin togain in situ access to acteria as it has a high molecularweight. This can lengthen the assay time (15–17) andlead to over-permeabilization or cell lysis. A rapid oligo-bFISH assay, however could offer cost savings.As far as we are aware, there are no reports of the

detection of S. aureus with a two-step FISH assay andoligo-b in 1hr or less. S. aureus was chosen for testing witholigo-b and FISH as it is an important pathogen and itspermeabilization for DNA probes is more involved. SinceS. epidermidis is phylogenetically similar to S. aureus andits rRNA nearly identical (7), it was included as a negative

Published online in Wiley Online Library (wileyonlinelibrary.com).

DOI 10.1002/jcla.20486

Received 29 April 2011; Accepted 21 July 2011

Grant sponsor: Australian Research Council’s Linkage Projects; Grant

number: LP0775196.

�Correspondence to: Thomas S. Lawson, Macquarie University,

Faculty of Science, Sydney, New South Wales, Australia.

E-mail: [email protected]

�c 2011 Wiley-Liss, Inc.

3.2 Optimization of a two-step permeabilization fluorescence in situhybridization assay for the detection of Staphylococcus aureus 59

3.2 Optimization of a two-step permeabilization fluorescence in situ hybridiza-

tion assay for the detection of Staphylococcus aureus

control. E. coli was also tested to ensure that the FISHassays developed could also detect Gram-negativebacteria (11). DNA rather than peptide nucleic acid(PNA) were used as they are less expensive.

MATERIALS AND METHODS

Two FISH assays were developed and tested. Theseare listed and compared in their optimized form inTable 1. The one-step permeablization FISH assay is amodified version of the FISH assay described byPoppert et al. (1). The two-step permeablization FISHassay extends the one-step assay by the addition of anextra permeabilization step and a streptavidin incuba-tion step (15). Details for the optimized one-step andtwo-step FISH assays are not included in Table 1.

Specimen Preparation

So that isolates could be tested at a nonclinicallocation, clinical patient isolates of S. aureus, Staphylo-coccus epidermidis and Escherichia coli were collected onagar plates from a major hospital. Only penicillin

binding protein (PBP2)-negative S. aureus were selected.Isolate identity was confirmed with polymerase chainreaction (18) and was then de-identified for experimen-tation. To control for potential differences betweenstrains, ten isolates of each type of bacteria werecollected. To enable the isolates to be compared over anumber of FISH experiments, the collected isolates werecultured in nutrient broth (CM0001; Oxoid, Hampshire,UK), centrifuged at 4,000 rcf, aliquoted and frozen asdescribed by Baldrich et al. (19). An aliquot of S. aureus,S. epidermidis, and E. coli was thawed and recultured innutrient broth for testing with FISH. As a control,isolates were tested directly from the plates and nodifference in signal was observed. To test with FISH,cultures of clinical isolates in nutrient broth werespotted (10 ml) onto the slides (X1XER308B; MenzelGlaser, Braunschweig, DE), dried at 801C for 3min, andthen fixed with absolute (m)ethanol for 1min.

Cell Permeabilization

To reduce preparation time, stock solution of lysozyme(L6876; Sigma-Aldrich, St. Louis, MO) and lysostaphin(L4402; Sigma) were prepared up to a week in advanceand stored as 1.5-ml aliquots in sterile plastic tubes. Forthe one-step permeabilization FISH assay, 15mg/mllysozyme and 0.1mg/ml lysostaphin were applied inone-step as described by Poppert et al. (1). For the two-step permeabilization assay, 10ml of lysozyme at 15mg/mlin Milli-Qs (MQ) water (Millipore, Billerica, MA) (20,21)was spotted onto the slide wells and incubated in 50-mltubes (210–261; Greiner, Frickenhausen, Germany) for6min at 381C (22). The lysozyme was rinsed off with, PBS(P4417; Sigma), and the slides were rapidly dried withpressurized air (1) or by centrifuging in 50-ml tubes for1min at 100 rcf. Lysostaphin at 0.1mg/ml in PBS wasspotted (10ml) onto the slides and incubated in 50-mltubes at 471C for 6min. Lysostaphin was removed byrinsing the slides in absolute (m)ethanol for 1min andthen drying at 801C for 1min. When E. coli isolates weretested, permeabilization was omitted.

Hybridization

To save time, the hybridization buffer and washingbuffer were prepared in advance. Hybridization buffer(0.9M NaCl, 20mM Tris–HCl, 0.01% (w/v) SDS, and1mg/ml DAPI) with no formamide or with 60% (v/v)deionized formamide were prepared and stored for up to ayear at �201C in 5-ml sterile plastic screw-top tubes.When needed, the buffers were thawed and mixed to thedesired target formamide concentration. The concentra-tion of formamide used in this study was 30% (v/v),about 5% higher than the lowest probe formamidedissociation concentration estimated with mathFISH

TABLE 1. Comparison of the Optimized One-Step or Two-Step Permeabilization FISH Assays

One-step (1) Two-step (This study)

Preparation: Cultures of clinical isolates were diluted with PBS,

spotted, fixed to slides at 801C, fixed in methanol, and air-dried

Permeabilization: Slides were

spotted with lysis reagent

(15mg/ml lysozyme,

0.1mg/ml lysostaphin,

20mM Tris–HCl at pH

7.0 inMQwater) and incu-

bated at 401C, rinsed with

methanol, and air-dried


spotted with 15mg/ml

lysozyme with 20mM

Tris–HCl at pH 7.0 in

MQ water and incubated

at 381C, rinsed with PBS,

and dried with

pressurized air


spotted with 0.1mg/ml

lysostaphin, 20mM

Tris–HCl diluted in PBS

and incubated at 471C,

rinsed with methanol,

and air-dried

Hybridization: Slides were spotted with hybridization buffer (30%

formamide, 0.9M NaCl, 20mM Tris–HCl at pH 8.0, 0.01% SDS, a

1-mM probe, and MQ water), and incubated at 471C

Washing: Slides were incubated with washing buffer (0.3–0.9M NaCl,

20mM Tris–HCl pH 8.0, 0.01% SDS, 10mM EDTA, and MQ

water) at 471C, rinsed with PBS

Mounting: Slides were

mounted with a cover-slip

while wet

Streptavidin/Mounting:

Slides were spotted with

streptavidin-f and

incubated at 381C, rinsed

with PBS, and mounted

with a cover-slip while

wet

360 Lawson et al.

J. Clin. Lab. Anal.


(mathfish.cee.wisc.edu) (471C hybridization incubation,0.9M NaCl and 1mM of probe) (23). The oligo (1mM)could be added to the buffer mix and stored at 41C in 1.5-mlsterile plastic aliquots for a week before performing FISH.For hybridization, 10ml of buffer with oligo (1mM) was

spotted onto the slides and incubated at 471C for 10min.Oligos tested were STAAUR specific for S. aureus(STAAUR-16S69: 5- GAAGCAAGCTTCTCGTCCG -3)(12) (Invitrogen, Carlsbad, CA), STAPHY specific forStaphylococcus (STAPHY-16S697 5-TCCTCCA-TATCTCTGCGC-3) (12) (Invitrogen), and EUB338specific for eubacteria (EUB338 16S337: 5- GCTGC-CTCCCGTAGGAGT -3) (Sigma) (24) (Invitrogen).These oligos were biotinylated (oligo-b) or directlyconjugated to Alexa Fluors 488 (Invitrogen), AlexaFluors 555, FITC or Cy3 (Genworks, Adelaide,Australia) labeled on the 50 end (oligo-f).

Specimen Washing

In a similar fashion to the hybridization buffer, thewashing buffer without salt (20mM Tris-HCl, 5mMEDTA and 0.01% (w/v) SDS) or with 1.8M NaCl wereprepared and stored for up to a year at 41C in 1 l bottles(GL45; Schott, Mainz, Germany). When needed, thebuffers were mixed to the target NaCl concentration in a50-ml tube, typically 1:4, to make approximately 0.3MNaCl. So that the washing buffer could be reusedmultiple times, before immersing the slide in the 50-mltubes of washing buffer for 3min at 471C, thehybridization buffer was rinsed off with washing buffer.

Streptavidin Conjugation

The slides were removed from the washing buffer andrinsed in PBS. For the one-step FISH assay, the slides weremounted wet with PBS and cover-slips for microscopy. Forthe two-step FISH assay, the slides were dried withpressurized air and spotted with 10ml of streptavidinconjugated to Alexa Fluors 488 (S-32354; Invitrogen)(hereafter, streptavidin-f), DyLights 488 (21832; ThermoFisher, Waltham, MA) or Alexa Fluors 555 (S-32355;Invitrogen) at 10mg/ml in PBS. Slides were incubated at471C for 10min, rinsed with PBS, and mounted as before,for microscopy.

Microscopy and Statistical Analysis

The FISH signal was observed with an epifluorescencemicroscope (BX51; Olympus, Tokyo, Japan) fitted witha 60� dry objective (UPLFLN; Olympus) and FITC/DAPI filters (U-MWU2, U-MWIB2; Olympus). Imageswere acquired at a resolution of 1,360� 1,024 with anOlympus DP72 camera and software (DP2-BSW v2.2;Olympus) set to a gain of 200 ISO and an exposure of

400msec. A representative image with a cell count of atleast 100 was selected for each treatment from threeexperiment runs. Images were analyzed with ImageJusing its standard segmentation algorithms (v1.43u;NIH, Bethesda, MD). Cell numbers, morphology, andpermeabilization were assessed with the FISH signal and40,6-diamidino-2-phenylindole (DAPI) (D9564; Sigma).The mean signal intensity (8-bit gray-scale) and stan-dard deviation for each FISH method were calculated.Summary statistics were compared with one-way analy-sis of variance (ANOVA) and a P value of o0.05 wasconsidered significant.

RESULTS

Effect of Different FISH Assays and Probe Typeson Signal

Two FISH assays were tested: the assay by Poppertet al. (1), which included a one-step permeabilizationtreatment and a modified version of that assay thatincluded a two-step permeabilization treatment (Table 1).Each FISH assay was tested with two types of probes:Oligo-f probes that had DNA sequences conjugated tofluorophores and oligo-b probes that had DNA sequencesconjugated to biotin and visualized with streptavidin-f.Each probe type was tested with three probe sequences:the STAAUR probe that was specific for S. aureus and theSTAPHY probe specific for Staphylococcus (12), and theEUB338 probe specific for eubacteria (24).As reported by Poppert et al. (1), the one-step

permeabilization FISH assay successfully detectedS. aureus and differentiated it from S. epidermidis witholigo-f probes in 45min (Fig. 1A). The one-step assay (1),however produced a poor signal for S. aureus with theSTAAUR oligo-b probe (Table 2) and a weak signal forS. epidermidis with the STAPHY oligo-b probe andstreptavidin-f (data not shown). In contrast, the two-steppermeabilization FISH assay successfully detectedS. aureus with oligo-f, oligo-b probes, and streptavidin-fin 1hr (Fig. 1B and D). Furthermore, the two-step assayproduced a higher FISH signal with oligo-f probes than theone-step FISH assay (Table 2). The difference in oligo-fsignal intensity was found to be significant (One-wayANOVA; Po0.05).

Effect of Different Slide and Fixation Preparation

Other aspects of the FISH method werealso investigated for optimization (data not shown).Heat fixing the bacteria on the slides at 801C rather thanair-drying shortened drying time from 10 to 3min. Cellloss from the slides was observed after processingwith FISH. Rinsing the slides with molten 0.2%(w/v) agarose (162-0102; Bio-Rad Laboratories, CA)

361Optimization of FISH for Staphylococcus aureus

J. Clin. Lab. Anal.


minimized this loss (25,26). A number of specimenfixation techniques were tested. Alcohol fixation ofisolates during dilution or after drying onto a slide wasfound to be necessary for a consistent FISH signal or ifthe isolates were stored for later testing. No difference in

the signal was observed between isolates fixed for 3 and10min. If fixed for 1min, the FISH signal was less thanwhen fixed for 3min, but was used to shorten the assay.Ethanol fixation was observed to produce a highersignal intensity, whereas methanol was observed toproduce a more consistent signal. Diluting isolates inabsolute (m)ethanol 1:1 and heating to 801C for 10mindid not improve the signal (27). For S. aureus,paraformaldehyde at 1% produced a weak signal (7,11).

Effect of Lysozyme and LysostaphinPermeabilization Treatments

As reported by Poppert et al. (1), when 2mg/ml lysozymeand 0.1mg/ml lysostaphin in 10mM Tris/HCl (pH 8) werecombined and applied in a one 5-min step at 461C, theFISH assay was simple and rapid for oligo-f probes. Thistreatment was further shortened to 3min by combining15mg/ml lysozyme with 0.1mg/ml lysostaphin in 10mMTris/HCl (pH 7) and incubating for 3min at 401C. Othervariations of lysozyme and lysostaphin permeabilizationwere tested with oligo-f probes. If lysozyme was appliedwithout lysostaphin, a 30-min incubation at 381C wasnecessary for STAAUR signal (21). If lysostaphinwas applied without lysozyme, a 10-min incubation at

TABLE 2. Comparison of the Staphylococcus aureus SI and CIfor One- and Two-Step Permeabilization FISH Methods Using

a STAAUR Oligo

Treatment SIa CIb

One-step FISH assay with oligo-fc 23.44 0.55

One-step FISH assay with oligo-f

and Tween 20s d24.85 0.71

One-step FISH assay with oligo-bc 17.19 0.19

Two-step FISH assay with oligo-bc 23.27 0.48

Two-step FISH assay with oligo-fe 27.31 0.45

SI, signal intensity; CI, confidence interval; FISH, fluorescence in situ

hybridization. Exposure time was the same for each image acquisition.

The results were consistent across the ten isolates tested.aMean signal intensity in 8-bit Gray-scale.bConfidence interval was calculated at 95%.cAs described in Table 1.dThe one-step FISH assay with a 5-min, 1% Tween 20s step at room

temperature before hybridization.eThe two-step FISH assay using a oligo-f, without the streptavidin

incubation step.

Fig. 1. Staphylococcus aureus treated with (A) the one-step permeabilization FISH assay or (B) the two-step permeabilization FISH assay and

then labeled with 1 mM/ml STAAUR oligo directly conjugated to Alexa Fluors 488. S. aureus treated with the two-step permeabilization FISH

assay and (C) not washed with PBS before lysostaphin, and (D) washed with PBS before lysostaphin and then labeled with 1 mM/ml oligo-b

STAAUR and 10 mg/ml streptavidin Alexa Fluors 488. (A) S. aureus treated with the one-step FISH assay. (B) S. aureus treated with the two-

step FISH assay. (C) S. aureus not washed with PBS before lysostaphin. (D) S. aureus washed with PBS before lysostaphin. PBS, phosphate-

buffered saline; FISH, fluorescence in situ hybridization.

362 Lawson et al.

J. Clin. Lab. Anal.


471C was sufficient for S. aureus, but the S. epidermidisEUB338 signal was poor. If oligo-b probes and streptavi-din-f was used, the one-step permeabilization treatmentproduced a weak signal (Table 2).To produce a satisfactory signal for oligo-b probes

and streptavidin-f, lysozyme and lysostaphin wereapplied separately. The sequence of the lysozyme andlysostaphin steps was found to be important. Asreported by Tavares et al. (2), lysozyme treatmentbefore lysostaphin produced a higher and more con-sistent signal than vice versa. Buffering of the enzymeswas also important. No difference was observed inSTAAUR signal if lysozyme was buffered at pH 7.0, 8.0,or left unbuffered. Lysostaphin when unbuffered,however, led to over-permeabilization and cell lysis.When lysostaphin buffered in Tris-HCl pH 7.0 or 8.0was applied, cell lysis was reduced but not completelyabrogated (Fig. 1C). Cell lysis, however was minimizedif a 1-min PBS wash step was added between thelysozyme and lysostaphin steps (Fig. 1D). The PBStreatment was further optimized by omitting the PBSwash step and instead diluting lysostaphin in PBS. Sincethe Tris–HCl buffering was not needed in the lysozymestep and lysostaphin was buffered with PBS, the assaypreparation was simplified. For the different permeabi-lization treatments tested, no differences were observedamong the ten isolates of each of the three bacteria(S. aureus, S. epidermidis, and E. coli).

Optimizing Hybridization

A surfactant step before hybridization was not necessary,however, a 1% (v/v) Tween 20s or 0.1% (v/v) TritonX-100swith 1% (w/v) bovine serum albumin inMQwaterspotted (10ml) onto the slides and incubated for 5min atroom temperature increased the signal intensity by 6%(Table 2). A number of different oligo and stain treatmentswere tested with the hybridization buffer. No differencewas observed in the signal if the oligos were tested atconcentrations of 0.25–3mM. Since it was easier to prepare,1mMwas chosen for further FISH testing. If Alexa Fluors

488 and Alexa Fluors 555 rather than FITC and Cy3fluorophores were used, photo-stability increased from10 sec to 1min with a 100-W Olympus U-RFL-T burner.The use of 50-ml tubes for hybridization incubationreduced drying out, reagent wastage, and reaction times.By using 30% (v/v) formamide for all the oligos tested inthis study, preparation was simpler and multiple oligoscould be applied at the same time.

Optimizing Specimen Washing and StreptavidinConjugation

The washing step could be substantially shortened ifthe NaCl concentration was increased from 0.3 to 0.9M.

A 1-min PBS wash step at room temperature was alsotested. This rapid and simple wash produced a highsignal intensity, but also some nonspecific staining. If thenonspecific signal was unacceptable with a PBS wash,the wash could be repeated with regular washing bufferuntil the nonspecific staining was removed withoutadversely affecting the FISH signal. When oligo-bprobes were used, an additional streptavidin-f incubationstep was needed. Its signal was highest when streptavi-din-f diluted in PBS was incubated at 381C for 10min.A number of measures were taken to counteract thenonspecific background signal associated with streptavi-din-f. NaCl concentration in the previous washing bufferwas increased from 0.3 to 0.9M and the washing timelengthened from 3 to 10min. Before applying, strepta-vidin-f was centrifuged at 10,000 rcf for 1min. Finally,the streptavidin concentration was minimized withoutlosing signal by diluting to a range of 1–10 mg/ml.

DISCUSSION

The study set out to optimize FISH for the detectionof S. aureus and differentiation from S. epidermidis.E. coli was also tested to ensure that the FISH assaysdeveloped would work for Gram-negative bacteria aswell. A merit of the study is that it addresses someaspects of the FISH procedure concerning fixation,permeablization, buffers and fluroescent dyes that werepreviously unquestioned. Although some of the studyexperiments did not result in major improvements to theassay, they may still be valuable for further studies,especially those that aim to optimize FISH.

Preparing Buffers

To test a range of FISH treatments, hybridization andwashing buffers were prepared beforehand and storedlong-term at �201C and mixed to their final formamideand NaCl concentrations just before use. It was relativelystraightforward to prepare large volumes of buffer andstore. As far as the authors are aware, this approach topreparing FISH buffers has not been reported elsewhereand is useful and applicable to routine diagnostics wherelabor costs are important. Since it was not necessary toprepare the reagents for each batch of FISH experiments,time-savings were made. Preparing and storing buffers forup to an year did not affect their application in the FISHassay or its signal. Hybridization buffer prepared witholigos could also be used for up to a week without signalloss or nonspecific binding (data not shown).

One-Step vs. Two-Step Permeabilization

The biotin–streptavidin system is rarely used inclinical FISH studies. The oligo-b assays are more


J. Clin. Lab. Anal.

fluorescent


involved and take longer than those that use oligo-f.Multiple oligo-b probes applied to the same specimenfor different microbes cannot be distinguished bystreptavidin-f as it binds to all of them. This can be amajor disadvantage since it is necessary in diagnosticFISH to combine the use of a species-specific probe witha eubacterial probe as an internal control. In addition,nonspecific background staining is higher when strepta-vidin-f is used. The study found, however that it ispossible to detect bacteria quickly with a relativelysimple oligo-b FISH assay. It was not possible to applyand distinguish between multiple oligo-b probes simul-taneously with streptavidin-f. As a simple work-around,the same specimen was spotted to more than one slidewell and a different oligo-b probe was applied to eachwell. The flip-side of this biotin–streptavidin systemlimitation is that only one streptavidin-f is required forvisualization and the cost of an oligo-b probe is about aquarter of its oligo-f counterpart. Nonspecific back-ground staining was controlled by more stringentwashes, and minimizing the amount of streptavidin-f.An unexpected outcome of the study was that the 1-hr

two-step FISH assay produced a higher signal intensitythan the one-step assay when oligo-f probes were used.This suggests that permeabilization is a key factor inhybridization. The time-to-result is 15min longer, but if theFISH signal is low or the background nonspecific signal ishigh, the two-step FISH assay might be a more practicalchoice.

Study Limitations

Cultures of clinical samples rather than clinical isolateswere tested. Reference strains and other frequentlyencountered microbes were not tested. Since it was difficultto control for the variation when comparing the FISHsignal intensity between treatments, representative imageswere used. Each treatment was repeated three times, animage was taken and if the variation between the threeimages was not significant, one image was chosen asrepresentative. The two-step FISH assay that was devel-oped could not be shortened to less than 1-hr without lossof S. epidermidis EUB338 signal. The optimized protocolslisted in Table 1 were a compromise to capture the variedresponses of the microbes tested. S. aureus produced ahigher and more consistent signal when treated withmethanol and lysostaphin, whereas S. epidermidis produceda higher and more consistent signal with ethanol andlysozyme.

CONCLUSION

The study found that a FISH assay that used alysozyme step followed by a PBS–lysostaphin step had ahigher STAAUR signal and could be applied almost as

rapidly as the FISH assay that combined lysozyme andlysostaphin into one-step. The two permeabilizationsteps lengthen the assay, but provided optimal condi-tions for lysozyme and lysostaphin enzyme activity,better control over the process of permeabilization aswell as a higher level of permeabilization. The two-stepassay might be used when the FISH signal is weak,bacterial numbers are low or if there is a need to useother reporter molecules such as catalyzed reporterdeposition (CARD)-FISH (26). Further testing of thefindings is warranted in a clinical scenario.

ACKNOWLEDGMENTS

The authors declare that no conflict of interest exists.Our thanks to the Australian Proteome Analysis Facilityfor laboratory facilities.

REFERENCES

1. Poppert S, Riecker M, Wellinghausen N, et al. Accelerated

identification of Staphylococcus aureus from blood cultures by a

modified fluorescence in situ hybridization procedure. J Med

Microbiol 2010;59:65–68.

2. Tavares A, Inacio J, MeloCristino J, et al. Use of fluorescence in

situ hybridization for rapid identification of staphylococci in blood

culture samples collected in a Portuguese hospital. J Clin

Microbiol 2008;46:3097–3100.

3. Oliveira K, Brecher SM, Durbin A, et al. Direct identification of

Staphylococcus aureus from positive blood culture bottles. J Clin

Microbiol 2003;41:889–891.

4. Kempf VA, Trebesius K, Autenrieth IB. Fluorescent in situ

hybridization allows rapid identification of microorganisms in

blood cultures. J Clin Microbiol 2000;38:830–838.

5. Peters RPH, Savelkoul PHM, SimoonsSmit AM, et al. Faster

identification of pathogens in positive blood cultures by fluores-

cence in situ hybridization in routine practice. J Clin Microbiol

2006;44:119–123.

6. Peters R, Agtmael MAV, SimoonsSmit AM, et al. Rapid

identification of pathogens in blood cultures with a modified

fluorescence in situ hybridization assay. J Clin Microbiol

2006;44:4186–4188.

7. Krimmer V, Merkert H, Eiff CV, et al. Detection of Staphylo-

coccus aureus and Staphylococcus epidermidis in clinical samples by

16S rRNAdirected in situ hybridization. J Clin Microbiol

1999;37:2667–2673.

8. Horvath A, Kristof K, KonkolyThege M, et al. Rapid identifica-

tion of pathogens in blood culture with fluorescent in situ

hybridization (FISH). Acta Microbiol Immunol Hung

2010;57:225–234.

9. Wang P. Simultaneous detection and differentiation of staphylo-

coccus species in blood cultures using fluorescence in situ

hybridization. Med Princ Pract 2010;19:218–221.

10. Gescher DM, Kovacevic D, Schmiedel D, et al. Fluorescence in

situ hybridisation (FISH) accelerates identification of gram-

positive cocci in positive blood cultures. Int J Antimicrob Agents

2008;32:51–59.

11. Jansen GJ, Mooibroek M, Idema J, et al. Rapid identification of

bacteria in blood cultures by using fluorescently labeled oligonu-

cleotide probes. J Clin Microbiol 2000;38:814–817.

12. Trebesius K, Leitritz L, Adler K, et al. Culture independent and

rapid identification of bacterial pathogens in necrotising fasciitis

364 Lawson et al.

J. Clin. Lab. Anal.


and streptococcal toxic shock syndrome by fluorescence in situ

hybridisation. Med Microbiol Immunol 2000;188:

169–175.

13. Hogardt M, Trebesius K, Geiger AM, et al. Specific and rapid

detection by fluorescent in situ hybridization of bacteria in clinical

samples obtained from cystic fibrosis patients. J Clin Microbiol

2000;38:818–825.

14. Wu Q, Li Y, Wang M, et al. Fluorescence in situ hybridization

rapidly detects three different pathogenic bacteria in urinary tract

infection samples. J Microbiol Methods 2010;83:175–178.

15. Matsuhisa A, Saito Y, Ueyama H, et al. Detection of

staphylococci in mouse phagocytic cells by in situ hybridization

using biotinylated DNA probes. Biotech Histochem 1994;69:

31–37.

16. Shimada J, Hayashi I, Inamatsu T, et al. Clinical trial of insitu

hybridization method for the rapid diagnosis of sepsis. J Infect

Chemother 1999;5:21–31.

17. Kudo M, Matsuo Y, Nakasendo A, et al. Potential clinical benefit

of the in situ hybridization method for the diagnosis of sepsis.

J Infect Chemother 2009;15:23–26.

18. Thomas LC, Gidding HF, Ginn AN, et al. Development

of a realtime Staphylococcus aureus and MRSA (SAM�)

PCR for routine blood culture. J Microbiol Methods 2007;68:

296–302.

19. Baldrich E, Vigues N, Mas J, et al. Sensing bacteria but treating

them well: Determination of optimal incubation and storage

conditions. Anal Biochem 2008;383:68–75.

20. Cisani G, Varaldo PE, Grazi G, et al. High-level potentiation of

lysostaphin anti-staphylococcal activity by lysozyme. Antimicrob

Agents Chemother 1982;21:531–535.

21. Lawson TS, Connally RE, Iredell JR, et al. Detection of

Staphylococcus aureus with a fluorescence in situ hybridization that

does not require lysostaphin. J Clin Lab Anal 2011;25:142–147.

22. Yan C, Ding B, Lan X, et al. The toxicity study on marine

lowtemperature lysozyme. Food Chem Toxicol 2008;46:604–609.

23. Yilmaz LS, Parnerkar S, Noguera DR. MathFISH, a web tool for

the in silico evaluation of oligonucleotide probes for fluorescence

in situ hybridization based on thermodynamic models. Appl

Environ Microbiol 2010;77:1118–1122.

24. Amann RI, Binder BJ, Olson RJ, et al. Combination of 16S

rRNAtargeted oligonucleotide probes with flow cytometry for

analyzing mixed microbial populations. Appl Environ Microbiol

1990;56:1919–1925.

25. Daims H, Ramsing NB, Schleifer KH, et al. Cultivation-

independent, semiautomatic determination of absolute bacterial

cell numbers in environmental samples by fluorescence in situ

hybridization. Appl Environ Microbiol 2001;67:5810–5818.

26. Pernthaler A, Pernthaler J, Amann R. Fluorescence in situ

hybridization and catalyzed reporter deposition for the identification

of marine bacteria. Appl Environ Microbiol 2002;68:

3094–3101.

27. Deere D, Vesey G, Milner M, et al. Rapid method for fluorescent

in situ ribosomal RNA labelling of Cryptosporidium parvum.

J Appl Microbiol 1998;85:807–818.


J. Clin. Lab. Anal.


Clin. Lab. 9+10/2011 789

Clin. Lab. 2011;57:789-794 ©Copyright

SHORT COMMUNICATION

Express Fluorescence in Situ Hybridization Methods for the Detection of Staphylococcus Aureus

THOMAS S. LAWSON, RUSSELL E. CONNALLY, SUBRAMANYAM VEMULPAD, JAMES A. PIPER

Faculty of Science, Macquarie University, Sydney, Australia

SUMMARY Background: As a proof-of-concept, the feasibility of detecting Staphylococcus aureus faster than previous whole-cell fluorescent in situ hybridization (FISH) methods was tested. Methods: Isolates of Staphylococcus were treated with three rapid slide-based FISH protocols and DNA probes. Protocols were shortened by optimizing, combining or omitting steps. Results: All FISH protocols detected S. aureus and not the phenotypically similar Staphylococcus epidermidis. The express FISH assay was completed in 24 minutes. The one-step FISH assay with NaCl and the one-step with phos-phate buffered saline (PBS) assay took 19 minutes, but yielded a weaker signal. Conclusions: The exploratory study identified S. aureus two to three times faster than previous methods. Addi-tional testing in a clinical laboratory scenario (for example with positive blood-culture bottles) is warranted. (Clin. Lab. 2011;57:789-794)

INTRODUCTION

Faster detection of S. aureus in clinical diagnostics is desirable (1-3). FISH is a reliable method for detecting S. aureus and its time-to-result has been shortened from 127 minues (4,5) and 581,3 minutes down to 45 minutes (2). Peptide nucleic acid (PNA) probes can further re-duce this time, but the cost is significant (Panagene) (6). FISH might be better utilized if its results were avail-able in a similar time-frame as Gram-staining (7). We believed that the conventional FISH assay could be fur-ther shortened if it was optimized. FISH typically uses lysostaphin to permeabilize S. aureus for DNA probe access (2,4,5). Other diagnostic assays have also used lysostaphin (8,9) and so we reasoned that lysostaphin might allow certain FISH steps to be combined or omit-ted.


Three FISH methods were tested and assessed against a method reported by Poppert et al. (2). Details not listed in Table 1 are as follows. To rectify a report of cell loss (5), 0.02 % (w/v) agarose (Bio-Rad, 162-0102) was

spotted to diagnostic glass slides and heat fixed at 80 oC (10,11). Ten clinical isolates of S. aureus and S. epidermidis were randomly collected from blood agar plates at a major hospital. Identity of the isolates was confirmed with polymerase chain reaction (PCR) (12) and then deidentified for FISH. To allow up to 4 X dai-ly FISH experiments at a non-clinical location, isolates were cultured in nutrient broth (Oxoid, CM0001), ali-quoted, and then pelleted and stored for up to three months as described by Baldrich et al. (13). Before test-ing each FISH method three times, isolates were thawed and re-cultured in nutrient broth until turbid (70 min-utes, 0.5 McFarland) (13). As a control, isolates were tested directly from agar plates with no difference in re-sults. To minimize reaction time, reagents and slides were held in preheated 50 mL centrifuge tubes with screw-caps (Greiner, 210-261). A DNA probe specific for S. aureus (Invitrogen, Staaur 16S69: 5'- GAAG-CAAGCTTCTCGTCCG -3’) and a eubacteria probe EUB338 both conjugated at the 5' end to Alexa Fluor® 488 (Invitrogen) were applied. A representational image (24-bit-RGB TIFF) of each FISH method was acquired (Olympus, BX51) and ana-_____________________________________________ Short Communication accepted April 29, 2011


3.3 Express fluorescence in situ hybridization methods for the detection of

Staphylococcus aureus

T. S. LAWSON et al.

Clin. Lab. 9+10/2011

790

Table 1. Express, one-step and one-step with PBS FISH methods.

Express FISH One-step One-step with PBS

Preparation: Isolates were cultured in nutrient broth until turbid, spotted (1 minute), fixed to slides at 80 oC (3 min-utes), methanol fixed (1 minute), and air-dried (1 minute)

Preparation: Same as the express FISH assay

Preparation: Same as the express FISH assay

Permeabilization: Slides were spotted (1 minute) with lysis reagent (15 mg/mL lysozyme, 100 µg/mL lysostaphin, 20 mM Tris-HCl pH 7.0) and incubated at 40 oC (4 minutes), rinsed with methanol, and air-dried (1 minute)

No permeabilization step No permeabilization step

Hybridization: Slides were spotted with buffer [20 % formamide (v/v), 0.9 mol/L NaCl, 20 mM Tris-HCl pH 8.0, 0.02 % (v/v) SDS, 2 µM probe, 0.5 µg/mL DAPI, and Milli-Q water] (1 minute), and in-cubated at 47 oC (8 minutes)

Hybridization: Slides were spotted with buffer [100 µg/mL lysostaphin, 1.2 mol/L NaCl, 0.02 % SDS, 20 mM Tris-HCl pH 7.0, 2 µM probe, 0.5 µg/mL DAPI, Milli-Q water], and incubated at 47 oC (10 minutes)

Hybridization: Slides were spotted with buffer [100 µg/mL lysostaphin, 8XPBS (Sigma, P4417) buffer, 1 % (v/v) Tween 20, 2 µM probe, 0.5 µg/mL DAPI, and Milli-Q water] (1 minute), and incubated at 47 oC (10 minutes)

Washing: Slides were incubated with washing buffer [0.318 mol/L NaCl, 20 mM Tris-HCl pH 8.0, 0.01 % SDS, 10 mM EDTA, and Milli-Q water] at 47 oC (1 minute), and rinsed with PBS (1 min-ute)

Washing: Slides were rinsed with PBS (1 minute)

Washing: Slides were rinsed with PBS (1 minute)

Total time: 24 minutes Total time: 19 minutes Total time: 19 minutes lyzed with ImageJ (NIH, v1.43u). Three image attri-butes were compared: the signal intensity, the size of the cell with FISH staining and the ratio of cells with FISH signal. To measure signal intensity, an image mask was created to delineate the cells for analysis. Threshold levels were set automatically for the control FISH image and kept constant for the other images. To measure the size of individual cells stained with FISH, the diameter of representative cells was measured and the area calculated. To measure the ratio of cells with FISH signal to those without, the FISH mask was inverted, formatted to a red channel (24-bit, RGB), and merged with a blue DAPI (Sigma, D9564) channel image, and the differences measured. Parametric assumptions were tested with a histogram of the signal and a p value of <0.05 was con-sidered significant. The mean signal intensity gray-scale, standard deviation and a 95 % confidence interval for each FISH method image were calculated (Table 2). Summary statistics were compared with one-way analy-sis of variance (ANOVA).

RESULTS

The control (2) and express FISH methods had similar signal intensity, PBS FISH was weaker, and the one-step signal weaker again (Table 2). Image analysis was confirmed by observation (Figure 1). A one-way ANO-VA of the mean signal intensity between methods was found to be significantly different (p <0.000). If agarose was not added to the slides, signal variance increased but its intensity did not change. A further control where molten 0.4 % (w/v) agarose and the broth culture of the isolates was diluted 1:1 and applied to slides (data not shown) (11) confirmed this result. No difference was observed in the diameter of cells with FISH signal. For the ratio of cells labeled with FISH, there were minor discrepancies, but these were attributed to the threshold setting filtering out cells outside the focal-plane with FISH signal and not those with DAPI signal. For all methods, no cross-reactivity was observed; S. aureus was positive for the Staaur probe and S. epidermidis was not. With the one-step and PBS FISH methods, however the S. epidermidis EUB338 signal was insuffi-cient.

3.3 Express fluorescence in situ hybridization methods for thedetection of Staphylococcus aureus 67

EXPRESS DETECTION OF S. AUREUS WITH FISH

Clin. Lab. 9+10/2011

791

Table 2. Comparison of the signal intensity for the three FISH protocols and a control FISH assay2. FISH methods are listed in Table 1.

FISH assay Signal Intensity

Meana ± CIb

Control FISH2 43.9 0.9

Express FISHc 44.2 0.7

Express FISH without agarosed 44.8 3.7

One-step FISHe 37.7 1.0

One-step with PBS FISHf 41.4 0.7 ‘a’ Mean signal intensity was measured in 8-bit Gray-scale. ‘b’ Confidence interval was calculated at 95 % ‘c’ A 2 minute permeabilization step, 2 mg/mL lysozyme and 0.02 mg/mL lysostaphin treatment was sufficient for S. aureus and Staaur, but not for S. epidermidis and the EUB338 probe ‘d’ As a control no agarose was added to the slide before FISH ‘e’ Buffer with 0.9 mol/L NaCl was optimal for EUB338 but not for the Staaur probe ‘f’ Over-permeabilized if not treated with Tween 20 or a methanol bath

A)


T. S. LAWSON et al.

Clin. Lab. 9+10/2011

792

B)

C)

Figure 1. S. aureus labeled with the Staaur probe conjugated to Alexa Fluor® 488 after express (a), one-step (b) and one-step PBS FISH (c). Exposure time was kept constant and there was no processing after acquisition. Bar is 10 µM.


EXPRESS DETECTION OF S. AUREUS WITH FISH

Clin. Lab. 9+10/2011

793

DISCUSSION

The proof-of-concept study set out to test the feasibility of shortening the conventional FISH assay (1-3). The differences observed in FISH staining between methods was found to be attributed to signal intensity and not to the size of the cells or the ratio of cells stained with FISH (Figure 1). The results suggest that a 24 minute express FISH assay can identify S. aureus without com-promising the signal. By permeabilizing with only lyso-staphin, the specificity of the one-step and PBS FISH methods increased and the washing step became unnec-essary, but the signal was reduced. The weaker signal might be due to the limitation of using one instead of two lytic enzymes. Possibly PBS performed better than one-step FISH as its conditions are optimal for lyso-staphin activity (14). To develop the three FISH methods, Poppert et al. (2) was taken as a starting point. We considered that opti-mizing the conditions for each FISH step might reduce its overall length. The express FISH assay is listed in Table 1. For the preparation step, air-drying time was shortened to 3 minutes by placing slides on an 80 oC heat-block and methanol fixation was reduced without signal loss to one minute. For the permeabilization step, S. aureus took 2 minutes, but S. epidermidis required 4 minutes with 15 mg/mL (15) or 600 U/G lysozyme (Sigma, L6876) and 0.1 mg/mL (5,15) or 0.3 U/G lyso-staphin (Sigma, L9043) at pH 7.0 (14,16,17) and 40 oC (14,17). For the hybridization step, incubation was shortened to 8 minutes without signal loss and DAPI was added to avoid a separate counter-staining step. For the washing step, we found that washing could be short-ened to one minute by increasing NaCl stringency until no signal was observed with S. epidermidis. As a result, the express FISH assay turnaround-time was reduced from 45 (2) to 24 minutes. We then tested if combining and omitting FISH steps to reduce the time further was feasible. The one-step FISH assay is listed in Table 1. The preparation step was the same as express FISH. For the combined permeabilization and hybridization buffer, we were unable to add lysozyme at a NaCl concentra-tion above 0.1 mol/L or lysostaphin at a formamide (Applichem, A2156) concentration above 15 % (16). So only lysostaphin was added to the formamide-free buf-fer and tested at increments of 0.15 mol/L NaCl from 0.0 to 1.8 mol/L and found it to be optimal at 1.2 mol/L for the Staaur probe. Total turnaround-time was cut from 24 to 19 minutes but the signal was weaker. We recognized that PBS (Sigma, P4417) might provide ideal conditions for lysostaphin (14) and thought that using PBS buffer could overcome the weaker signal observed with one-step FISH. The one-step PBS FISH assay is listed in Table 1. For the buffer, an equivalent ionic concentration of PBS replaced NaCl, Tris-HCl was unnecessary and Tween 20® replaced SDS as it reduced over-permeabilization. As before, the washing step was omitted. The resulting signal was an improve-

ment over the one-step FISH (Table 2). Assay time for the one-step PBS FISH remained at 19 minutes. The results have a number of limitations. Blood culture bottles, S. aureus reference strains, and other microbes were not tested (3,5). Pretreating slides with agarose was helpful, but not necessary. For comparison lyso-zyme and lysostaphin was set at 15 mg/mL and 0.1 mg/ mL, respectively, but was not optimal for every method (15). As both FISH probes were conjugated to the same fluorophore, they could not be applied simultaneously (5). For image analysis, some cells were measured out-side the focal plane, a single representative image rather than multiple images was analyzed, and minor DAPI bleeding was observed. To conclude, shorter assay times are desirable, but longer hybridization times may still be necessary for reliable detection when patients are pretreated with antibiotics and pathogen rRNA is low (4). The findings warrant testing the applicability of these methods in a clinical laboratory scenario. Acknowledgment: The study was funded by the Australian Research Council’s Linkage Projects (LP0775196). Declaration of Interest: There are no conflicts of interest for the authors. References: 1. Peters RP, Agtmael MAV, Simoons-Smit AM, Danner SA, Van-

denbroucke-Grauls CM, Savelkoul PH. Rapid identification of pathogens in blood cultures with a modified fluorescence in situ hybridization assay. Journal of Clinical Microbiology 2006;44: 4186-8.

2. Poppert S, Riecker M, Wellinghausen N, Frickmann H, Essig A.

Accelerated identification of Staphylococcus aureus from blood cultures by a modified fluorescence in situ hybridization proce-dure. Journal of Medical Microbiology 2010;59:65-8.

3. Horvath A, Kristof K, Konkoly-Thege M, Nagy K. Rapid identi-

fication of pathogens in blood culture with fluorescent in situ hy-bridization (FISH). Acta Microbiologica et Immunologica Hun-garica 2010;57:225-34.

4. Gescher DM, Kovacevic D, Schmiedel D, et al. Fluorescence in

situ hybridisation (FISH) accelerates identification of gram-posi-tive cocci in positive blood cultures. International Journal of Antimicrobial Agents 2008;32 Suppl 1:51-9.


coccus species in blood cultures using fluorescence in situ hybrid-ization. Medical Principles and Practice 2010;19:218-21.

6. Hermsen ED, Shull SS, Klepser DG, et al. Pharmacoeconomic

analysis of microbiologic techniques for differentiating staphylo-cocci directly from blood culture bottles. Journal of Clinical Mi-crobiology 2008;46:2924-9.

7. Munson EL, Diekema DJ, Beekmann SE, Chapin KC, Doern GV.

Detection and treatment of bloodstream infection: laboratory re-porting and antimicrobial management. Journal of Clinical Mi-crobiology 2003;41:495-7.


T. S. LAWSON et al.

Clin. Lab. 9+10/2011

794

8. Geary C, Stevens M. Rapid lysostaphin test to differentiate staphylococcus and micrococcus species. Journal of Clinical Mi-crobiology 1986;23:1044-5.

9. Tuncan EU, Martin SE. Lysostaphin lysis procedure for detection

of Staphylococcus aureus by the firefly bioluminescent ATP method. Applied and Environmental Micro-biology 1987;53:88-91.

10. Daims H, Ramsing NB, Schleifer KH, Wagner M. Cultivation-

independent, semiautomatic determination of absolute bacterial cell numbers in environmental samples by fluorescence in situ hybridization. Applied and environmental microbiology 2001;67: 5810-8.

11. Pernthaler A, Pernthaler J, Amann R. Fluorescence in situ hy-

bridization and catalyzed reporter deposition for the identification of marine bacteria. Applied and Environmental Microbiology 2002;68:3094-101.

12. Thomas LC, Gidding HF, Ginn AN, Olma T, Iredell J. Develop-

ment of a real-time Staphylococcus aureus and MRSA (SAM-) PCR for routine blood culture. Journal of microbiological methods 2007;68:296-302.

13. Baldrich E, Vigues N, Mas J, Munoz FX. Sensing bacteria but

treating them well: Determination of optimal incubation and stor-age conditions. Analytical Biochemistry 2008; 383:68-75.

14. Schindler CA, Schuhardt VT. Purification and properties of lyso-staphin – A lytic agent for Staphylococcus aureus. Biochimica et Biophysica Acta 1965;97:242-50.

15. Cisani G, Varaldo PE, Grazi G, Soro O. High-level potentiation

of lysostaphin anti-staphylococcal activity by lysozyme. Antimi-crobial Agents and Chemotherapy 1982;21: 531-5.

16. Davies RC, Neuberger A, Wilson BM. The dependence of lyso-

zyme activity on pH and ionic strength. Biochimica et Biophysica Acta 1969;178:294-305.

17. Szweda P, Kotlowski R, Kur J. New effective sources of the

staphylococcus simulans lysostaphin. Journal of Biotechnology 2005;117:203-13.

Correspondence: Thomas S. Lawson Faculty of Science, Macquarie University Sydney, NSW, 2109, Australia Tel.: +61 2 9850-8938 Fax: +61 2 9850 8115. E-mail: [email protected]



Improvements to the existing FISH method: a summary

The findings reported in this Chapter are as follows:

1. The number of FISH probes that could potentially identify SA was doubled (1).

2. The newly identified probes were found to have characteristics that were superior

to the established probes (1).

3. Reagent preparation for FISH was improved (2). Reagents were formulated be-

forehand and stored long-term so that, just before commencing the procedure for

FISH, they could then be mixed rapidly to their final concentration(2).

4. Permeabilization of SA for the access of high-molecular weight probes was opti-

mized (2).

5. SA was sufficiently permeabilized so that FISH could be performed in one hour,

but without the loss of its cell integrity (2).

6. Turnaround time for the FISH assay was shortened from the previous fastest time

of 45 minutes (32) to 24 minutes (3). This suggests that it might be possible to

detect SA within half an hour of a blood culture becoming positive (3).

There were limitations to these findings. The FISH methods tested (52, 32, 40)

relied on lysostaphin to permeabilize SA. Lysostaphin is an enzyme that is effective

at permeabilizing SA (147), but can be costly and can complicate the preparation

and performance of the FISH assay (95). The FISH methods that were tested also

used formamide, a standard denaturing reagent for the hybridization of FISH probes

(45). Formamide, however is toxic to use and disposal can be difficult (144). The

FISH methods also relied on an incubator and a water-bath to offer the conditions

necessary for the hybridization and washing of the probes which, because they are

usually dedicated to the FISH assay, are an extra cost and take up bench-space when

not used.

The concerns raised here are addressed in Chapter 4. It reports on the feasibility

of permeabilizing SA for DNA probes with lysozyme alone (4), instead of with both


lysozyme and lysostaphin (52). Lysozyme is an inexpensive enzyme (141) commonly

used to permeabilize most Gram-positive bacteria in FISH (64). The replacement

of formamide with non-toxic urea is also described (5). Finally, the replacement of

an incubator and water-bath with a hot-plate whose temperature can be precisely

controlled is reported.

4Development of new FISH methods

This Chapter recounts an investigation into the re-engineering of the FISH assay for

the identification of S. aureus (SA). The feasibility of detecting SA with FISH that

was not first permeabilized with lysostaphin is reported (4). The Chapter then reports

on the feasibility of performing a FISH assay without formamide, an incubator or a

water-bath (5).

The Chapter comprises of two sections. Each of these sections were published in

a peer reviewed journal and included as such. In the first, SA is detected with a

DNA-based FISH assay that permeabilizes with lysozyme alone: Lawson TS, Connally

RE, Iredell JR, Vemulpad S, Piper JA. Detection of Staphylococcus aureus with a

fluorescence in situ hybridization that does not require lysostaphin. J Clin Lab Anal

2011;25:142-147 (4). In the second, SA is detected with an urea-NaCl based FISH

assay that uses a hot-plate with a precise temperature control for the incubation steps:

Lawson T, Connally R, Vemulpad S, Piper JA. Dimethyl formamide-free, urea-NaCl

fluorescence in situ hybridization (FISH) assay for Staphylococcus aureus. Lett Appl

Microbiol 2012;10.1111/j.1472-765X.2011.03197.x:(in press) (5).

75

Journal of Clinical Laboratory Analysis 25 : 142–147 (2011)

Detection of Staphylococcus aureus With a Fluorescence In SituHybridization That Does Not Require Lysostaphin

Thomas S. Lawson,1,2� Russell E. Connally,1 Jonathan R. Iredell,2 Subramanyam Vemulpad,1

and James A. Piper3

1Faculty of Science, Macquarie University, NSW, Australia2Centre for Infectious Diseases and Microbiology, Westmead Hospital, Sydney, NSW, Australia

3Department of Physics, Macquarie University, NSW, Australia

To detect with whole-cell fluorescence insitu hybridization (FISH), Staphylococcusaureus is typically permeabilized with lyo-zyme and lysostaphin. We tested whether itwas feasible to detect S. aureus anddifferentiate it from Staphylococcus epider-midis with lysozyme-only permeabilization.We compared lysozyme permeabilization

to S. aureus permeabilized with lysozymein combination with lysostaphin. It wasdetermined that S. aureus treated withagarose, methanol, and lysozyme couldbe detected with FISH. The 1 hr protocolis a useful alternative to conventionalFISH. J. Clin. Lab. Anal. 25:142–147,2011. r 2011 Wiley-Liss, Inc.

Key words: early diagnosis; fluorescent in situ hybridization; gram-positive bacteria;molecular diagnostic; Staphylococcus aureus; lysostaphin; lysozyme;techniques

INTRODUCTION

Slide-based fluorescence in situ hybridization (FISH)is a reliable method for detecting pathogenic Staphylo-coccus aureus and distinguishing it from the relativelybenign Staphylococcus epidermidis (1–3). If DNA ratherthan the costly Peptide Nucleic Acid probes (Panagene)are applied, permeabilization is necessary to ensure accessof probes to in situ ribosomal RNA (rRNA) (4,5).Usually, permeabilization is conducted with the enzymeslysozyme (Sigma, L6876; Sigma-Aldrich, St. Louis, MO)and lysostaphin (Sigma, L4402), either mixed together(6,7) or in two steps (2,8). Other permeabilizationtreatments, such as hydrochloric acid (9), nisin (10),proteinase K (9), staphylolysin (11) or Triton X–100 (12)are only sometimes adopted (2,3,5,6,13).Permeabilization can complicate the application of

FISH in routine laboratory diagnostics, as it has to beconducted precisely (2). Underpermeabilization canresult in a low FISH signal and overpermeabilizationin lysis and cell loss (4). A simplification of this stepleading to more consistent outcomes is desirable.Lysozyme applied on its own for the detection ofS. aureus was previously reported, but the assaysdescribed took a number of hours (14,15). We report

here the efficacy of applying a single enzyme (lysozyme)instead of two, to rapidly detect S. aureus with FISH.


Preparation

To reduce cell loss (16), an agarose (Bio–Rad, 162–0102; Bio-Rad Laboratories, CA) bed was applied todiagnostic glass slides (Menzel–Glaser, X1XER308B;Menzel Glaser, Braunschweig, DE). The bed wasprepared by adding 0.02% (w/v) agarose and 0.01%(w/v) sodium azide (Sigma, S2002) to Milli-Q waters

(MQ) (Millipore, Billerica, MA) and dissolving it byheating without boiling in a microwave oven. The agarosedilute was spotted 10ml to each slide well and dried on an801C hotplate. Blood agar plates of clinical isolates

Published online in Wiley Online Library (wileyonlinelibrary.com).

DOI 10.1002/jcla.20448

Received 16 November 2010; Accepted 14 January 2011

Grant sponsor: Australian Research Council’s Linkage Projects; Grant

number: LP0775196.

�Correspondence to: Thomas S. Lawson, Department of Physics,

Macquarie University, NSW 2109, Australia.


�c 2011 Wiley-Liss, Inc.

76 Development of new FISH methods

4.1 Detection of Staphylococcus aureus with a fluorescence in situ hybridization

that does not require lysostaphin

positive for S. aureus and S. epidermidis were randomlycollected from a major hospital. Isolate identity wasconfirmed via polymerase chain reaction (17). For safehandling, the first ten isolates of S. aureus negative forthe mecA gene and the first ten isolates identified asS. epidermidis were selected for further testing. Isolateswere deidentified at collection and labeled numerically toensure their identity was blinded when assessed. Isolateswere cultured in 50ml sterilized tubes of nutrient broth(Oxoid, CM0001; Oxoid, Hampshire, UK) and incubatedat 371C with a gentle rotation until turbid. Broth dilutionsfor blood cultures were used, as it allowed shortenedincubation times of 1–2hr from frozen isolates as opposedto day or overnight incubations. To enhance probe signaland further reduce cell loss, prewarmed 0.4% (w/v)agarose and the broth culture of the isolates was diluted1:1 (16). The agarose–isolate dilute was then spotted 10mlto each slide well and fixed with an 801C hotplate until dry.

Permeabilization

To further fix and partially permeabilize the isolates,the slides were washed in 50ml sterile tubes of absolutemethanol for 3min (6). Slides were removed and driedon a hotplate. The slides were cooled and 10 ml of freshlyprepared 15mg/ml lysozyme (18) in unbuffered MQwater (12,19,20) was pipetted to each well. Typically,lysozyme is buffered with Tris–HCl at pH 8.0(2,3,5,6,13), but for simplicity and to attain a moreintense signal (21), we followed reports where bufferingwas omitted (12,19,20). Slides were fitted in 50ml tubesto prevent evaporation and then placed in a 371C (21)incubator for 30min (12). The lysozyme action wasstopped by immersion in absolute methanol for 1minand then dried on a hotplate (6).The isolates were also permeabilized with a lysozyme–

lysostaphin mixture. The protocol was identical to thelysozyme-only treatment, but with the following modi-fications. Lysostaphin (Sigma, L4402) at 100 mg/ml(3) was added to the lysozyme in MQ water (19) and

incubated at 401C (21,22) for 3 (6) instead of 30min.As a control, lysozyme–lysostaphin was kept unbuf-fered, but we observed cell morphology was betterpreserved if it was buffered at pH 8.0 (2,3,5,6,13).As before, slides were immersed for 1min in absolutemethanol (6).Additional tests were performed to compare the

quality and applicability of other reagents. Fixativesand permeabilizers were selected on the basis of previousreports and the signal intensity, cells stained with FISH,cell loss after FISH, time taken for the assay, and costswere compared (Table 1), for different permeabilizationtreatments. The lysozyme and lysostaphin and lysozyme-only treatments are already described. The treatmentwith lysostaphin excluded lysozyme (23). The treatmentwithout agarose excluded agarose spotting to the slides oragarose in dilution with the isolates (16). The treatmentwith lysozyme after ethanol replaced the methanolfixation step with absolute ethanol (2). The proteinaseK treatment replaced the 30min lysozyme step with10min incubation in 1mg/ml proteinase K (P4850,Sigma) at 401C, a methanol rinse for inactivation, and10min incubation with 1mg/ml lysozyme at 401C. Thelysozyme after HCl acid treatment was the same as theproteinase K treatment, except proteinase K was replacedwith 1M HCL at 371C (24). The treatment with Tween20 (P7949, Sigma) after lysozyme added a 5min incuba-tion step at room temperature with Tween followed bya water rinse. The treatment with Triton X–100 (T8787,Sigma) after lysozyme (12) was the same as Tween, expectwith Triton. The no permeabilization treatment omittedthe personalization step. If not listed, other FISH stepswere the same as the lysozyme-only treatment.

FISH

A hybridization buffer was prepared with 0.9M NaCl(Sigma, S6191), 20mM Tris–HCl (Sigma, T1503,T3253), and 0.02% (w/v) SDS (Sigma, L4390) in MQwater (25). Either 15% (v/v) deionized formamide

TABLE 1. Comparison of Different S. aureus Permeabilization Treatments Concerning Quality and Robustness

Permeabilization treatment Signal intensity Cells stained Cell adhesion Time (min) Cost ($)

Lysozyme and lysostaphin 1111 1111 111 7 10

Lysostaphin 1111 111 111 7 9

Lysozyme 1111 1111 111 34 5

Lysozyme without agarose 11 11 11 33 5

Lysozyme after ethanol 11111 111 111 34 5

Lysozyme without alcohol fixation 111 11 111 31 5

Lysozyme after proteinase K 1111 1111 1 21 6

Lysozyme after HCl acid – – 11 21 5

Tween 20 after lysozyme 11111 1111 111 39 5

Triton X-100 after lysozyme 1111 1111 111 39 5

No permeabilization 1 1 111 4 4

143S. aureus Detection With Lysozyme FISH

J. Clin. Lab. Anal.

4.1 Detection of Staphylococcus aureus with a fluorescence in situhybridization that does not require lysostaphin 77

(Applichem, A2156; Applichem, Darmstadt, DE) and2 mM of Sau probe (Sau 16S69: 50-GAAGCAAGC-TTCTCGTCCG-30) specific for S. aureus or 30% (v/v)formamide and 2mM of EUB338 probe (EUB338 16S337:50-GCTGCCTCCCGTAGGAGT-30) specific for bacteriawas added (Invitrogen, Carlsbad, CA). Both oligonucleo-tide (DNA) probes were conjugated to the flurophoreAlexa Fluors 488 (Invitrogen). The buffer was spotted10ml to each well and the slides were fitted in 50ml tubesand placed in a 471C incubator for 20min.After hybridization, slides were immediately fitted in

50ml tubes of prewarmed washing buffer containing5mM EDTA (Sigma, EDS), 0.64M NaCl, 20mMTris–HCl, and 0.02% (w/v) SDS in MQ water (25).Tubes were then placed in a 471C water bath for3min (6). Washing action was stopped by rinsing ina 50ml tube of phosphate buffered saline (PBS) (Sigma,P4417) at room temperature and followed by dryingwith pressurized air (6). If required, isolates werecounterstained with 15 ml of 1 mg/ml DAPI for 1minand then rinsed with PBS (16). Cells were visualized witha fluorescence microscope (Olympus, BX51; Olympus,Tokyo, Japan) equipped with a fluorescein filter.Different permeabilization treatments are listed in

Table 1. The ratios are indicated by ‘‘1111’’ for all,‘‘111’’ for three quarters, ‘‘11’’ for half, and ‘‘1’’ fora quarter or less. A negative result is indicated by ‘‘–’’.The signal intensity (1) was measured relative to thelysozyme and lysostaphin FISH treatment. Cells stained(1) with FISH was measured from the ratio of cells withFISH to DAPI (Sigma, D9564) signal. Cell adhesion (1)was measured from the ratio of cells remaining afterFISH to cells observed with DAPI before FISH. Time(Min) taken for each treatment included the sum of theagarose, fixation, and permeabilization steps. The cost($) was rounded to the nearest dollar for a daily run offour FISH experiments, each with two slides (Sigma,USD). All treatments were adjusted so that cell lysis was

minimal. The treatment was repeated in its final formthree times. For each experimental variable, two wellswere tested and three fields of view with an objective ofX60 were assessed. Two independent, blinded observersanalyzed the images. Slight variation was observedbetween slide wells, but not between experimental runs.

RESULTS

Table 1 summarizes the results of different treatmentsin terms of quality and robustness. Both lysozyme-onlyand lysozyme–lysostaphin permeabilization detectedS. aureus and differentiated it from S. epidermidis withthe Sau probe. For the initial tests, all enzymes were leftunbuffered. Lysozyme–lysostaphin had a brighter signalthan lysozyme-only treated S. aureus. However, thelysozyme–lysostaphin left cells overpermeabilized andlysed. Once a buffer at pH 8.0 was added, the lysis wascontrolled, and S. aureus treated for 3min withlysozyme–lysostaphin, which had a result equivalent tothat of a 30min lysozyme-only treatment. Figure 1illustrates the ability to detect S. aureus with the Sauprobe for both treatments. In addition, no cross-reactivity was noted for the Sau probe; it was positivefor S. aureus and negative for the S. epidermidis isolates.Likewise, both treatments detected S. aureus andS. epidermidis with the universal EUB338 probe.We could not obtain a signal rapidly with lysozyme

alone unless the S. aureus isolates were diluted inagarose. This lengthened the assay, but it was only aslight encumbrance as the step was performed in 1min.Agarose doubled the signal intensity, the ratio of cellswith signal, and increased cell adhesion. Without theagarose dilution, similar signal intensity was realizedif the cells were hybridized for 70 instead of 20min.We tested cell loss of isolates in agarose spotted to slidesprepared and unprepared with an agarose bed. Weobserved that agarose spotted slides further reduced cell

Fig. 1. S. aureus permeabilized with lysozyme (A), and S. aureus permeabilized with lysozyme–lysostaphin (B). S. aureus were labeled with the

Sau probe conjugated to the fluorophore Alexa Fluors 488. Bar is 10mm. (A) S. aureus permeabilized with lysozyme. (B) S. aureus permeabilized

with lysozyme and lysostaphin.

144 Lawson et al.

J. Clin. Lab. Anal.

fluorophorefluorophorefluorophorefluorophore


loss. This might be an advantage if the number of targetcells is low. As the bed can be applied before a FISHprocedure, its preparation does not complicate orlengthen FISH.A number of different fixation procedures were tested

(Table 1). Applying 100% methanol to the slides wasobserved to be the most effective and rapid fixation (6).Ethanol fixation of slides produced a brighter signal, butwas less consistent than with methanol. Omitting thealcohol fixation step reduced both signal intensity andconsistency. Tween 20 enhanced the signal, but itinvolved an additional assay step. Yet, when we testedlysozyme diluted with Tween, the signal did not differfrom lysozyme-only permeabilization. In contradictionto previous reports (12), we saw no improvement withTriton X–100. Permeabilization with hydrochloric acidproduced poor results; it seemed to inhibit the action ofthe conjugated probe itself.As a control, FISH was performed with the permea-

bilization step omitted. Less than one quarter of theS. aureus cells had sufficient signal. As a further control,FISH was performed with only lysostaphin (19). Highsignal strength was observed, but the signal was lessconsistent than that of S. aureus treated with lysozyme–lysostaphin or the proposed lysozyme-only method.Isolates were tested directly from blood agar plates withlysozyme-only and in combination with lysostaphin.The results were consistent with tests of S. aureuscultured in nutrient broth. Poor results were obtainedwith lysozyme-only permeabilization if agarose wasomitted. Preliminary testing (data not shown) witha healthcare-associated meticillin-resistant S. aureus,(HA)–MRSA isolate and a community-associated(CA)–MRSA isolate, was comparable to the mecA-negative isolates (26).Lysoyme and lysostaphin are commonly applied at a

pH of 8.0 (2,3,5,6,13). Buffering at pH 8.0 was found toreduce the loss of cell morphology with lysostaphin.However, we observed that lysozyme-only assay pro-duced poor results unless the pH was reduced to 7.0. Thelysozyme-only assay was tested and found to be effectivewithout buffering, and so for simplicity, Tris-HCl bufferwas omitted from the final tests. We experienced somedifficulty applying Proteinase K. The precise concen-tration, incubation temperature, and time necessaryfor permeabilization but not overlysis was difficult tomanage. Washing with 100% methanol reduced over-permeabilization, but an agarose bed and dilution inagarose did not stop the loss of up to half the cells.To minimize thickness and visual aberration, we

tested the lowest concentration of agarose necessary tomaintain cell adhesion and signal intensity (27). Wefound that an agarose concentration of 0.02% (w/v) wassufficient for the slide bed and 0.2% sufficient for the

isolate dilution. For simplicity, we diluted 0.4% (w/v)agarose 1:1 with the isolates. This may, however, have anegative effect on the assay’s sensitivity if microbenumbers are low. To reduce overdilution of cells, wetrialled one part agarose at 0.8% to three parts ofnutrient broth with isolates, without signal loss. Anadditional benefit of agarose was that the probe concen-tration could be reduced by a factor of five without lossof signal. Initially, experiments were performed at 5 mMprobe concentrations, but after the addition of agarose,this was reduced to 1 mM. As a safety margin, the finalexperiments were performed at 2 mM.

DISCUSSION

We set out to validate whether lysostaphin wasnecessary for detecting S. aureus with FISH. Wedemonstrated that S. aureus can be successfully permea-bilized rapidly without lysostaphin. The ability oflysozyme-only to permeabilize S. aureus is likely owingto how the isolates were prepared after culturing andhow they were fixed and permeabilized. Isolates werediluted in agarose to enhance signal intensity (16,27);fixed in mid-log phase when rRNA numbers were high(1,13); permeabilized by heat, methanol (6) and lyso-zyme (12); treated with a relatively high concentration ofunbuffered lysozyme (12,18–20); and finally incubatedfor an extended period of time (12) at an optimaltemperature for lytic activity (21).There were some drawbacks to using an agarose bed

and an agarose isolate dilution. For agarose stockdilution to mix properly with isolates in nutrient buffer,it needed to be prewarmed. When viewed with afluorescence microscope, the agarose did create visualaberrations and thickening of the specimen. To see allthe cells in focus, it was necessary to adjust the micro-scope stage Z-axis up and down while viewing. Figure 1illustrates FISH-labeled S. aureus inside and outside thefocal plane. However, these encumbrances were offsetby the doubling in signal intensity and cell adhesion.Rapid and effective FISH with only lysozyme waspossible with this signal enhancement. When usingbacteria from pure culture, cell loss was not a problem.However, it was felt that this study would have a widerutility if this parameter was optimized as well.Handling of lysostaphin was not straightforward.

Minute amounts were involved (28) and upon weighing,the lyophilized powder (Sigma, L4402) readily absorbedmoisture from the atmosphere, making exact measure-ment difficult. When diluted in water, its decline inactivity was noticeable after 1 week. We saw somevariation in S. aureus strain response to lysostaphin.These variables made the titration of lysostaphinnecessary before each experiment to ensure that isolates


J. Clin. Lab. Anal.


were permeabilized optimally. Furthermore, lysostaphinwas approximately 40 times more expensive byvolume spotted than lysozyme (Sigma, L6876, L4402).In contrast, if only lysozyme was applied, the permeabi-lization step was more robust, less sensitive to variationin bacteria strains, less likely to overpermeabilize, andtitration was unnecessary. Dilutions can be stored at 41Cfor 2 months before activity loss was noticeable. Theweighing was relatively simple and did not requirea microbalance scale housed in a draft-free enclosure.If preparation mistakes are made, the enzyme wasreformulated quickly and without significant expense.A limitation of the lysozyme-only FISH protocol was

its turnaround time. At 1 hr, it was twice as long asthe fastest reported lysozyme–lysostaphin protocol (6).However, this was still half the time of other presump-tive tests for S. aureus (29,30). In conclusion, this studydetected and differentiated S. aureus from S. epidermidiswith a 1 hr FISH method that did not require lyso-staphin. The procedure worked with Staphylococcitaken directly from agar plates (data not shown), butfurther testing is required to assess the sensitivity andspecificity of this practical method on blood cultures.

ACKNOWLEDGMENTS

The authors thank the Australian Proteome AnalysisFacility and to Associate Professor Robert Willows atMacquarie University.

REFERENCES

1. Krimmer V, Merkert H, Eiff CV, et al. Detection of Staphylo-

coccus aureus and Staphylococcus epidermidis in clinical samples

by 16S rRNA-directed in situ hybridization. J Clin Microbiol

1999;37:2667–2673.

2. Tavares A, Inacio J, Melo-Cristino J, Couto I. Use of fluorescence

in situ hybridization for rapid identification of staphylococci in

blood culture samples collected in a Portuguese hospital. J Clin

Microbiol 2008;46:3097–3100.


coccus species in blood cultures using fluorescence in situ

hybridization. Med Princ Pract 2010;19:218–221.

4. Nistico L, Gieseke A, Stoodley P, Hall-Stoodley L, Kerschner JE,

Ehrlich GD. Fluorescence in-situ hybridization for the detection

of biofilm in the middle ear and upper respiratory tract mucosa.

J Methods Mol Biol 2009;493:191–213.

5. Mallmann C, Siemoneit S, Schmiedel D, et al. Fluorescence in situ

hybridization to improve the diagnosis of endocarditis: A pilot

study. Clin Microbiol Infect 2010;16:767–773.

6. Poppert S, Riecker M, Wellinghausen N, Frickmann H, Essig A.

Accelerated identification of Staphylococcus aureus from blood

cultures by a modified fluorescence in situ hybridization procedure.

J Med Microbiol 2010;59:65–68.

7. Zautner AE, Krause M, Stropahl G, et al. Intracellular persisting

Staphylococcus aureus is the major pathogen in recurrent

tonsillitis. PLoS ONE 2010;5:9452.

8. Wu Q, Li Y, Wang M, Pan XP, Tang YF. Fluorescence in situ

hybridization rapidly detects three different pathogenic bacteria in

urinary tract infection samples. J Microbiol Methods 2010;83:

175–178.

9. Moter A, Gobel UB. Fluorescence in situ hybridization (FISH) for

direct visualization of microorganisms. J Microbiol Methods

2000;41:85–112.

10. Peters RP, Agtmael MAV, Simoons-Smit AM, Danner SA,

Vandenbroucke-Grauls CM, Savelkoul PH. Rapid identi-

fication of pathogens in blood cultures with a modified fluores-

cence in situ hybridization assay. J Clin Microbiol 2006;44:

4186–4188.

11. Zwirglmaier K. Detection of prokaryotic cells with fluorescence in

situ hybridization. J Methods Mol Biol 2010;659:349.

12. Cimino M, Alamo L, Salazar L. Permeabilization of the

mycobacterial envelope for protein cytolocalization studies by

immunofluorescence microscopy. BMC Microbiol 2006;6:35.

13. Kempf VA, Trebesius K, Autenrieth IB. Fluorescent in situ

hybridization allows rapid identification of microorganisms in

blood cultures. J Clin Microbiol 2000;38:830–838.

14. Kipp F, Ziebuhr W, Becker K, et al. Detection of Staphylococcus

aureus by 16S rRNA directed in situ hybridisation in a patient with

a brain abscess caused by small colony variants. Br Med J

2003;56:746.

15. Hoa M, Tomovic S, Nistico L, et al. Identification of adenoid

biofilms with middle ear pathogens in otitis-prone children

utilizing SEM and FISH. Int J Pediatr Otorhinolaryngol

2009;73:1242–1248.

16. Pernthaler A, Pernthaler J, Amann R. Fluorescence in situ

hybridization and catalyzed reporter deposition for the identi-

fication of marine bacteria. Appl Environ Microbiol 2002;68:

3094–3101.

17. Thomas LC, Gidding HF, Ginn AN, Olma T, Iredell J.

Development of a real-time Staphylococcus aureus and MRSA

(SAM-) PCR for routine blood culture. J Microbiol Methods

2007;68:296–302.

18. Cisani G, Varaldo PE, Grazi G, Soro O. High-level potentiation of

lysostaphin anti-staphylococcal activity by lysozyme. Antimicrob

Agents Chemother 1982;21:531–535.

19. Oliveira M, Bexiga R, Nunes SF, et al. Biofilm-forming ability

profiling of Staphylococcus aureus and Staphylococcus epidermidis

mastitis isolates. Vet Microbiol 2006;118:133–140.

20. Gescher DM, Kovacevic D, Schmiedel D, et al. Fluorescence

in situ hybridisation (FISH) accelerates identification of gram-

positive cocci in positive blood cultures. Int J Antimicrob Agents

2008;32:51–59.

21. Yan C, Ding B, Lan X, Guo S, Xie Y, Wang C. The toxicity study

on marine low-temperature lysozyme. Food Chem Toxicol

2008;46:604–609.

22. Szweda P, Kotlowski R, Kur J. New effective sources of

the Staphylococcus simulans lysostaphin. J Biotechnol 2005;117:

203–213.

23. Veeh RH, Shirtliff ME, Petik JR, et al. Detection of Staphylo-

coccus aureus biofilm on tampons and menses components.

J Infect Dis 2003;188:519–530.

24. Macnaughton SJ, O’Donnell AG, Embley TM. Permeabilization

of mycolic-acid-containing actinomycetes for in situ hybridization

with fluorescently labelled oligonucleotide probes. Microbiology

1994;140:2859.

25. Manz W, Amann R, Ludwig W, Wagner M, Schleifer KH.

Phylogenetic oligodeoxynucleotide probes for the major subclasses

of proteobacteria: Problems and solutions. Syst Appl Microbiol

1992;15:593–600.

26. Millar BC, Loughrey A, Elborn JS, Moore JE. Proposed

definitions of community-associated meticillin-resistant Staphylo-

coccus aureus (CA-MRSA). J Hosp Infect 2007;67:109–113.

146 Lawson et al.

J. Clin. Lab. Anal.


27. Gijlswijk RPV, Wiegant J, Raap AK, Tanke HJ. Improved

localization of fluorescent tyramides for fluorescence in situ

hybridization using dextran sulfate and polyvinyl alcohol.

J Histochem Cytochem 1996;44:389–392.

28. Trebesius K, Leitritz L, Adler K, Schubert S, Autenrieth IB,

Heesemann J. Culture independent and rapid identification of

bacterial pathogens in necrotising fasciitis and streptococcal toxic

shock syndrome by fluorescence in situ hybridisation. Med

Microbiol Immunol 2000;188:169–175.

29. Lagace-Wiens PR, Alfa MJ, Manickam K, Karlowsky JA.

Thermostable DNase is superior to tube coagulase for the direct

detection of Staphylococcus aureus in positive blood cultures.

J Clin Microbiol 2007;45:3478–3479.

30. Sturm PDJ, Kwa D, Vos FJ, Bartels CJM, Schulin T. Performance

of two tube coagulase methods for rapid identification of

Staphylococcus aureus from blood cultures and their impact

on antimicrobial management. Clin Microbiol Infect 2008;14:

510–513.


J. Clin. Lab. Anal.


NOTE TO THE EDITOR

Dimethyl formamide-free, urea-NaCl fluorescence in situhybridization assay for Staphylococcus aureusT.S. Lawson, R.E. Connally, S. Vemulpad and J.A. Piper

Faculty of Science, Macquarie University, NSW, Australia

Slide-based fluorescence in situ hybridization (FISH) is a

robust assay for characterizing intact bacteria in clinical

specimens. It is usually run with an incubator and a

water-bath and with dimethyl formamide (referred to as

formamide) and NaCl buffers (Wang 2010). Occasionally

a specialized hot-plate is used for hybridization with DNA

(Poppert et al. 2010) or PNA probes (AC005; AdvanDx,

Wobum, MA) followed by washing with a thermo-mixer

(Poppert et al. 2010) or water-bath (AC006; AdvanDx). A

regular hot-plate has not been used, however, for the

entire FISH assay with DNA probes.

This report identified Staphylococcus aureus with a

novel FISH assay performed on a single hot-plate using

DNA probes. Staphylococcus aureus is a clinically signifi-

cant Gram-positive bacteria (Wang 2010) and if DNA

probes are used, its permeabilization can be more com-

plex than other bacteria (Poppert et al. 2010). The new

assay tested is modified from an assay reported by the

authors (Lawson et al. 2011). Improvements include

marking the slides with a wax pencil, replacing the form-

amide-NaCl buffers with urea-NaCl reagents and using a

hot-plate with a plastic cover instead of 50 ml tubes, an

incubator and a water-bath (Lawson et al. 2011). To

compare the signal strength of the new assay, the previous

assay (Lawson et al. 2011) was run in parallel as a

control.

If a hot-plate is used for incubations, reagents can dry

out and the temperature can fluctuate, both of which can

result in a weaker hybridization signal. Possible solutions

to enhance the signal were tested including using urea as

an alternative denaturing reagent to formamide (Soe et al.

2011). Unlike formamide, urea is non-toxic (Simard et al.

2001), can inhibit RNase degradation (Simard et al. 2001)

and can act as an additional permeabilizer with the result

that the FISH signal could be increased (Huang et al.

2011).

In preparation for FISH, 10 clinical isolates of PBP2-

negative S. aureus and 10 of Staphylococcus epidermidis

previously identified with PCR (Thomas et al. 2007) were

randomly collected at a hospital and cultured in nutrient

broth to an optical density of 1Æ0 at 600 nm (CM0001;

Oxoid, Basingstoke, UK). Cultures were diluted 1 : 10 in

broth, and 10 ll aliquots were spotted to slide-wells

(X1XER308B; Menzel-Glaser, Braunschweig, Germany),

dried at 47�C on a hot-plate and fixed with absolute

methanol for 1 min. Wells were marked with a wax pen-

cil to restrain reagents to the wells.

Two FISH assays were tested, an incubator-bath assay

described in Lawson et al. (2011) and a hot-plate FISH

assay (this report). Two sets of hybridization and washing

reagents were tested with each assay. The first used con-

ventional hybridization (formamide-NaCl) and NaCl

Keywords

detection, identification, infection, rapid

methods, staphylococci.

Correspondence

Tom Lawson, Faculty of Science, Macquarie

University, NSW 2109, Australia.


2011 ⁄ 1642: received 27 September 2011,

revised 1 December 2011 and accepted 2

December 2011

doi:10.1111/j.1472-765X.2011.03197.x

Abstract

Aims: To test the feasibility of identifying Staphylococcus aureus with a fluores-

cence in situ hybridization (FISH) assay that uses a single hot-plate and urea-

NaCl reagents.

Methods and Results: Slides spotted with S. aureus and treated with methanol

and lysozyme were incubated with urea-NaCl reagents on a hot-plate with a

precise temperature control and identified with specific DNA probes.

Conclusions: Staphylococcus aureus was detected and differentiated from Staph-

ylococcus epidermidis in 1 h with a novel FISH method that used a single

hot-plate and in the absence of dimethyl formamide.

Significance and Impact of Study: A rapid hot-plate FISH assay with urea-

NaCl and without toxic dimethyl formamide might be useful if FISH is run

infrequently or where resources are limited.

Letters in Applied Microbiology ISSN 0266-8254

ª 2011 The Authors

Letters in Applied Microbiology ª 2011 The Society for Applied Microbiology 1


4.2 Dimethyl formamide-free, urea-NaCl fluorescence in situ hybridization (FISH)

assay for Staphylococcus aureus

washing buffers (Poppert et al. 2010; Lawson et al. 2011).

The second used urea with NaCl (urea-NaCl) for both

hybridization and washing (this report). With all treat-

ments, two DNA probes Staaur (16S69: 5¢-GAAGCAAG-

CTTCTCGTCCG -3¢) and EUB338 probe (16S337: 5¢-GCTGCCTCCCGTAGGAGT -3¢) conjugated at the 5¢ end

to Alexa Fluor 488 identified S. aureus and eubacteria,

respectively.

The first FISH assay was tested with an incubator-bath

and formamide-NaCl reagents (Poppert et al. 2010; Lawson

et al. 2011). To permeabilize the bacteria, wells were spot-

ted with 30 ll of 15 mg ml)1 lysozyme (pH 7Æ0). The slides

were fitted into 50 ml tubes (210–261; Greiner Bio-One,

Frickenhansen, Germany), incubated at 47�C for 30 min

and then rinsed with methanol and dried. For hybridiza-

tion, wells were spotted with 20 ll of formamide-NaCl buf-

fer [30% (v ⁄ v) formamide (A2156; Applichem, Darmstadt,

Germany), 0Æ9 mol l)1 NaCl (S6191; Sigma-Aldrich, St

Louis, MO), 20 lmol l)1 Tris–HCl (pH 7Æ0), 0Æ01% (w ⁄ v)

SDS (4390; Sigma, L4390) in Milli-Q (MQ) water; Milli-

pore, Bedford, MA] containing 1 lmol l)1 of probe.

The slides were fitted into 50 ml tubes and placed in a

47�C incubator for 20 min. After hybridization, slides

were rinsed with prewarmed NaCl washing buffer

[0Æ64 mol l)1 NaCl, 20 lmol l)1 Tris–HCl (pH 7Æ0) and

0Æ01% SDS in MQ water] and fitted into 50 ml tubes

containing 25 ml of prewarmed washing buffer and

placed in a gently agitated 47�C water-bath for 5 min.

Slides were removed and rinsed with MQ water and

mounted wet with a cover-slip for viewing.

The second FISH assay used a hot-plate and was opti-

mized with urea-NaCl (this report). The permeabilization

step was the same as before except that it was performed on

a 47�C hot-plate, and the slides were covered with a clear

plastic lid (78 · 78 mm, 123160; Decor, Melbourne, Aus-

tralia) to minimize temperature change and reagent drying.

The hot-plate was developed by one of the authors (Russell

Connally) and had an accuracy of ±0Æ5�C at 47�C. A

platinum resistance probe was used with a microcomputer

display for accurate temperature control. Slides were rinsed

with methanol and were dried before 20 ll of urea-NaCl

[1 mol l)1 urea (U6504; Sigma), 0Æ9 mol l)1 NaCl,

20 lmol l)1 Tris-HCl (pH 7Æ0) in MQ water] with 1 lmol

l)1 of probe was spotted to each well.

Slides were incubated on the 47�C hot-plate for 20 min

before rinsing twice with prewarmed 250 ll of urea-NaCl

[8 mol l)1 urea, 0Æ9 mol l)1 NaCl, MQ water and

20 lmol l)1 Tris-HCl (pH 7Æ0)]. Slides were placed on

the hot-plate and 30 ll of prewarmed urea-NaCl was

spotted immediately to each well. Slides were incubated

with the plastic cover for 5 min, rinsed twice with urea-

NaCl again before a final rinse with MQ water and

mounting as before.

Slides were visualized with an epifluorescence

microscope (BX51; Olympus, Tokyo, Japan) fitted with a

60 objective (UPLFLN; Olympus) and FITC ⁄ DAPI filters

(U-MWU2, U-MWIB2; Olympus). Images were acquired

at a resolution of 1360 · 1024 with an Olympus DP72 cam-

era and software (DP2-BSW v2Æ2; Olympus) set to a gain of

200 ISO and an exposure of 2 s. Images from three experi-

mental runs were analysed with ImageJ (v1Æ43u; NIH,

Bethesda, MD). Summary statistics were compared with

one-way analysis of variance (anova) and a P value of 0Æ05.

A summary of the results for the two FISH assays and

their reagents is listed in Table 1 and shown in Fig. 1(a–

d). Staphylococcus epidermidis was negative for the Staaur

probe (Fig. 1f), and S. aureus and Staph. epidermidis were

both positive for the EUB338 probe. The incubator-bath

assay produced a higher signal than the hot-plate. The

incubation temperature and humidity were observed to

vary more on a hot-plate as it was not sealed. The urea-

NaCl produced a higher signal than the formamide-NaCl.

Urea might be acting as an additional premeabilizer for

S. aureus (Huang et al. 2011). No difference in signal was

observed between the 10 isolates of each bacteria tested.

The one-way anova found a significant difference

(P < 0Æ000) for the four treatments listed in Table 1.

The hot-plate assay was initially developed with the con-

ventional formamide-NaCl hybridization and washing buf-

fers. The signal was observed to be weaker than the same

assay run with an incubator and a water-bath (Table 1). To

increase the signal, urea was tested as a substitute for form-

amide in the hybridization buffer at 0Æ5, 1, 2, 4 and 8 mol l)1

(Kourilsky et al. 1970; Simard et al. 2001; Soe et al. 2011)

with NaCl at 0Æ9 mol l)1 (Poppert et al. 2010; Lawson et al.

2011). The signal was highest at 1 and 2 mol l)1 urea, and

1 mol l)1 was chosen for further testing.

A number of washing treatments were tested with the 1

mol l)1 urea and 0Æ9 mol l)1 NaCl hybridization reagent.

Table 1 A comparison of the Staaur probe signal intensity for each

of the FISH treatments tested with Staphylococcus aureus.

FISH treatment

Signal intensity

Mean* CI�

Incubator-bath FISH assay�

Formamide-NaCl and NaCl buffers 36Æ09 0Æ62

Urea-NaCl reagents 49Æ5 1Æ23

Hot-plate FISH assay§

Formamide-NaCl and NaCl buffers 32Æ7 0Æ52

Urea-NaCl reagents 45Æ2 1Æ13

*Mean signal intensity was measured in 8-bit Gray-scale.

�Confidence interval was calculated at 95%.

�Incubator-bath FISH protocol (Lawson et al. 2011) as described in

this report.

§Hot-plate FISH protocol as described in this report.

FISH on a hot-plate with urea Lawson et al.

2 Letters in Applied Microbiology ª 2011 The Society for Applied Microbiology

ª 2011 The Authors

4.2 Dimethyl formamide-free, urea-NaCl fluorescence in situhybridization (FISH) assay for Staphylococcus aureus 83

A conventional wash containing 0Æ9 mol l)1 NaCl was

applied for 1 min as described previously (Lawson et al.

2011), but the signal was inconsistent. Urea was reported

by Hutton (1977) to reduce melting temperature by

approx. 2Æ25�C with each 1 mol l)1 increase in its concen-

tration whereas melting temperature was reduced by

approx. 0Æ6�C with each 1% (v ⁄ v) increase of formamide.

To improve the signal and to simplify the assay, the

washing buffer was replaced with the urea-NaCl reagents

used in the hybridization step.

The stringency of this new washing regime was

adjusted with urea rather than with the conventional

NaCl and was set higher than that used for hybridization

so that duplexes with mismatches would be removed. The

(a) (b)

(c) (d)

(e) (f)

(g)

Figure 1 Staphylococcus aureus visualized with Alexa Fluor� 488 after performing an incubator-bath FISH assay with (a) formamide-NaCl (Law-

son et al. 2011) or (b) urea-NaCl reagents (this report). Staphylococcus aureus after performing the hot-plate FISH assay (this report) with (c) form-

amide-NaCl or (d) urea-NaCl reagents. The same S. aureus visualized with (e) DAPI. Staphylococcus epidermidis after performing the hot-plate

assay with urea-NaCl and visualized with (f) Alexa or (g) DAPI. Bar in lower right corner is 5 lm.

Lawson et al. FISH on a hot-plate with urea

ª 2011 The Authors

Letters in Applied Microbiology ª 2011 The Society for Applied Microbiology 3


urea wash was tested at 0, 1, 2, 4 and 8 mol l)1 with

NaCl at 0Æ9 mol l)1 (Kourilsky et al. 1970; Simard et al.

2001; Soe et al. 2011). The non-specific signal for

Staph. epidermidis was the weakest at 4 and 8 mol l)1

urea. To minimize background signal, the 8 mol l)1 con-

centration was chosen (Kourilsky et al. 1970). Increasing

urea from 1 mol l)1 in the hybridization step to 8 mol l)1

in the washing step decreased the melting temperature by

about 26�C (Hutton 1977).

The findings suggest that if urea-NaCl reagents are

used, it is feasible to control hybridization conditions and

produce a sufficient signal with the hot-plate FISH

method. There are, however, some limitations to the find-

ing. The relationship between urea and formamide con-

centrations and urea’s action as a permeabilizer was not

established. Although it was not observed after the

20 min hybridization step, urea can have a slower rate of

denaturation than formamide (Hutton 1977). Other bac-

teria were not tested, and the specificity and sensitivity of

the assay were not assessed.

In conclusion, we described a novel FISH assay that does

not require an incubator, water-bath, formamide, lysosta-

phin or PNA probes (the latter two are expensive). Urea-

NaCl reagents were simple to prepare and unlike formam-

ide, non-toxic. The exclusion of formamide may open up

new applications, such as simplified FISH analysis using cell

sorters or FISH procedures using beacon probes without an

additional washing step. The findings warrant further spec-

ificity and sensitivity testing in a clinical scenario.

Acknowledgements

The authors acknowledge the Australian Research Coun-

cil’s Linkage Projects (LP0775196) for funding this

research and the Australian Proteome Analysis Facility

(APAF) for providing laboratory facilities.

References

Huang, E., Talukder, S., Hughes, T.R., Curk, T., Zupan, B.,

Shaulsky, G. and Katoh-Kurasawa, M. (2011) BzpF is a

CREB-like transcription factor that regulates spore

maturation and stability in dictyostelium. Dev Biol 358,

137–146.

Hutton, J.R. (1977) Renaturation kinetics and thermal stability

of DNA in aqueous solutions of formamide and urea.

Nucleic Acids Res 4, 3537–3555.

Kourilsky, P., Manteuil, S., Zamansky, M.H. and Gros, F.

(1970) DNA-RNA hybridization at low temperature in the

presence of urea. Biochem Biophys Res Commun 41, 1080–

1087.

Lawson, T.S., Connally, R.E., Iredell, J.R., Vemulpad, S. and

Piper, J.A. (2011) Detection of Staphylococcus aureus with

a fluorescence in situ hybridization that does not require

lysostaphin. J Clin Lab Anal 25, 142–147.

Poppert, S., Riecker, M., Wellinghausen, N., Frickmann, H.

and Essig, A. (2010) Accelerated identification of Staphylo-

coccus aureus from blood cultures by a modified fluores-

cence in situ hybridization procedure. J Med Microbiol 59,

65–68.

Simard, C., Lemieux, R. and Cote, S. (2001) Urea substitutes

toxic formamide as destabilizing agent in nucleic acid

hybridizations with RNA probes. Electrophoresis 22, 2679–

2683.

Soe, M.J., Moller, T., Dufva, M. and Holmstrom, K. (2011) A

sensitive alternative for MicroRNA in situ hybridizations

using probes of 2¢-O-Methyl RNA+ LNA. J Histochem Cy-

tochem 59, 661–672.

Thomas, L.C., Gidding, H.F., Ginn, A.N., Olma, T. and Iredell,

J. (2007) Development of a real-time Staphylococcus aureus

and MRSA (SAM-) PCR for routine blood culture. J

Microbiol Methods 68, 296–302.

Wang, P. (2010) Simultaneous detection and differentiation of

staphylococcus species in blood cultures using fluorescence

in situ hybridization. Med Princ Pract 19, 218–221.

FISH on a hot-plate with urea Lawson et al.

4 Letters in Applied Microbiology ª 2011 The Society for Applied Microbiology

ª 2011 The Authors

4.2 Dimethyl formamide-free, urea-NaCl fluorescence in situhybridization (FISH) assay for Staphylococcus aureus 85


Development of new FISH methods: a summary

A summary of the findings reported in this Chapter follows. SA was sufficiently per-

meabilized after methanol fixation and incubation with lysozyme so that it could be

detected with DNA probes and FISH in one hour (4). The lysozyme-only FISH assay

was repeated, but with changes. Formamide typically used in the hybridization buffer

was replaced with non-toxic urea and the incubator and water bath with a precision

temperature controlled hot-plate. SA was detected as rapidly as by the conventional

assay, but with a higher signal intensity when urea was used (5).

There were limitations to these findings. The earlier FISH assays tested cultures of

isolates of SA. If the same FISH procedure was applied to blood samples that contained

a high proportion of non-target material, these materials interfered with the access of

the FISH probe and its signal. When the urea-NaCl assay was repeated in purified

blood samples that contained SA, it was not possible to detect SA without interference

from the autofluorescence in the blood-debris that remained.

The following Chapter reports on methodological improvements for reducing signal

interference from non-target material in the specimen. Two approaches were investi-

gated, the non-target material was removed and the emission signal was time-resolved.

A simple blood bacteremia model was tested by spiking whole-blood with SA and

incubating the blood. To remove most of the non-target material, the separation of

intra and inter-cellular SA from the blood was attempted with a lysis and purifica-

tion procedure. A number of in situ hybridization (ISH) techniques were tested to

label the separated SA in the remaining blood-debris. This included a luminescence

in situ hybridization (LISH) assay, based on the lysozyme-only permeabilization FISH

assay developed in the project (4), to label SA with a long-lifetime luminescent probe.

Autofluorescence from the specimen could then be suppressed by visualizing the probe

with time-gated luminescence microscopy (TGLM).

5Time-gated fluorescence imaging of a

europium chelate label

This Chapter reports on the time-resolved visualization of S. aureus (SA) labeled with

europium probes using a luminescence in situ hybridization (LISH). A description is

offered of a novel technique that was developed for the separation and purification of

SA that was spiked and incubated in whole-blood (138). A report is then made on the

detection of SA labeled with a europium chelate probe. The signal of the probe was

time-resolved to suppress the autofluorescence of the remaining blood debris.

5.1 Time-gating of a europium probe rapidly la-

beled with luminescence in situ hybridization

for the detection of Staphylococcus aureus

5.1.1 Abstract

Aim: To identify SA in whole-blood with a europium (Eu3+) chelate, luminescence

in situ hybridization (LISH) assay and time-gated luminescence microscopy (TGLM).

87

88 Time-gated fluorescence imaging of a europium chelate label

Methods and Results: Whole-blood was spiked with SA and incubated for one hour.

SA was separated from the blood by lysis and centrifugation. SA was detected with a

Eu3+ chelate (BHTEGS) conjugated to a DNA (KT18) using a LISH assay. The SA

signal to noise ratio improved by a factor of at least 5 relative to SA detected with a

conventional non time-resolved fluorophore.

Conclusions: It was possible to rapidly identify SA with a Eu3+ chelate, a LISH

assay and TGLM in two-hours.

Significance and impact of study: Bacteria can be more readily identified with

TGLM, in specimens that are highly auto-fluorescent.

Keywords: autofluorescence, chelate, europium, in situ hybridization (ISH), lan-

thanide, luminescence in situ hybridization (LISH), Staphylococcus aureus, time-gated

luminescence microscopy (TGLM), time-resolved, whole-blood

5.1.2 Introduction

Fluorescent techniques such as fluorescence in situ hybridization (FISH) can accurately

and rapidly detect microbes in blood-cultures. The results from FISH can be poor if

it is applied to specimens with complex matrices such as whole-blood. The natural

background fluorescence of the specimen can overwhelm the signal from a FISH probe

(44, 121, 39, 125). One solution is to time-resolve the signal using time-gated lumines-

cence microscopy (TGLM) (46).

By gating the emission signal, the short-lived background signal of the specimen is

removed and the probe emission remains for detection (149, 161). The technique uses

an excitation pulse to create an emission signal from the specimen. While the detector

is in the off position, the pulse is terminated abruptly. After the short-lived specimen

fluorescence has decayed, the detector is turned on and a target signal acquired, that

is free of background noise (46).

Although it is useful, TGLM is not often used with in situ hybridization (ISH) as-

says (149). It requires probes and microscopy equipment that are specialized and can

be complex to use (162, 149). Lanthanide trivalent ions (Eu3+, Dy3+, Sm3+ and

5.1 Time-gating of a europium probe rapidly labeled with luminescencein situ hybridization for the detection of Staphylococcus aureus 89

Tb3+) are commonly used as the luminophores in TGLM. Ln3+ ions produce narrow-

spectrum emissions with long-lifetimes in the visible to infra-red spectrum (163). Be-

cause Ln3+ have low absorption coefficients they require chelating to produce a high

signal emission (164). The chelates absorb most of the light and transfer it to the Ln3+

for emission.

The chelates are complex and thus difficult to synthesize and are reactive and prone

to insolubility (151). Water from the specimen can also access the chelate and quench

the Ln3+ signal (47). For a high emission signal, the chelates require UV excitation

at wavelengths around 330 nm prolonged exposure of which can damage the specimen

(165). Even with chelating and short wavelength excitation, the signal from in situ

assays can still be too weak for routine diagnostics.

Attempts have been made to improve these shortcomings (47). Ln3+ ions such as

Eu3+ chelates can produce enough signal after excitation at the longer wavelengths

of 350 to 365 nm (47). New chelates have been developed such as europium (Eu3+)

BHHST, a derivative of BHHCT, tetradentate β-diketone (151). To increase its solubil-

ity and stability, it uses a hydrophilic molecular tether attached to a BHHCT molecule.

This allows BHHST to be conjugated to biomolecules without inducing precipitation

from solution (136).

BHHST has improved solubility and stability (161), but it is only a partial solu-

tion. It is still not stable or soluble enough to be applied as a streptavidin conjugate

to in situ hybridization assays (data not shown) (145, 45). To remedy this, another

BHHCT derivative Eu3+ chelate BHTEGS was developed by members of the candi-

date’s research group (Russell Connally and Nima Sayyadi, manuscript in preparation).

Initial tests suggest that it is more soluble and less reactive than BHHST (151) and

that it can be directly conjugated to DNA sequences and it does not interfere with its

hybridization (data not shown).

Specialized equipment and a modified microscope is typically required for the time-

resolution of the chelates (162). The pulsed excitation and time-resolution of the

emission signal has to be precisely synchronized. In this study, however, a simple-to-

use time-gated auto-synchronous luminescence detector (GALD) device (developed by


Russell Connally) was used (148). This device may make it simpler to carry out TGLM

in routine diagnostics. The pocket-size device fitted into the differential interference

and contrast (DIC) slot of the microscope and had its own UV light source. The

device simultaneously blocked the short-term fluorescence and transmitted the delayed

luminescence (148).

To compensate for the reactivity and low emission signal of early chelates (165),

early TGLM ISH studies were complicated and lengthy (149, 150). Usually the signal

was not visible through the eye-piece and could only be detected with a sensitive camera

(149). To amplify the weak signal (45), the ISH assays used biotinylated DNA probes

(150) or ran incubations over-night for probe hybridization (149). The weak chelate

signal could then amplified with a streptavidin (149) or tyramide conjugate (150).

These solutions, however complicated and lengthened the ISH assay making its

use in diagnostics limited (149, 150). If directly labeled chelates such as BHTEGS

could be applied to ISH with sufficient signal, a simpler, more rapid assay could be

used. Direct conjugation of chelates with DNA is possible. Ln3+ chelates have already

been conjugated directly to oligonucleotides and these bio-conjugates used in Forster

resonance energy transfer (FRET) assays where reactivity is less of a problem (165).

The aim of this study is to test the feasibility of rapid SA detection in a blood

sample with an ISH assay that is visualized with TGLM. SA is a frequent cause of

bacteraemia and severe sepsis (55) and blood can be highly autofluorescent (100). The

study proposed to spike blood with SA and then incubate it so that the SA undergoes

phagocytosis. The blood will be lysed and centrifuged to separate and collect the SA.

A slide-based LISH assay will be performed with the novel Eu3+ BHTEGS DNA probe

and an Alexa Fluor R© 488 (Invitrogen) (Alexa) probe as a control. To the candidate’s

knowledge, this is the first report of the time-gated luminescence detection of bacteria

hybridized to a luminescent probe.

5.1.3 Method

SA isolates identified with PCR (158) and sourced from clinical specimens were cultured

in nutrient broth (Oxoid, CM0001) Figure 5.1 (3). Cultures were washed and diluted


in saline to an optical density of 1.0 at 600 nm (159). Venous blood from a healthy

volunteer was collected in EDTA tubes (Becton Dickinson, 367863). A simple in vitro

bacteraemia model was created by spiking the whole-blood while fresh, with SA and

incubating (160). For 1 ml of blood, 10 µl of SA prepared in saline was added (1.0

optical density at 600 nm). The blood was then incubated with gentle agitation at 37

◦C for one hour.

Rather than labeling SA in intact blood with LISH or FISH as reported elsewhere

(100, 63, 62), the SA was first separated from the blood with alkaline water (138). The

alkaline water was prepared by adding 4 mM NaOH to Milli-Q (MQ) water (pH 10.0).

Blood and alkaline water at a ratio of 1:10 were mixed by vortexing (to make a final

pH of 8.5) and then centrifuged for five minutes at 3,000 rcf. The supernatant was

removed and this alkaline water lysis treatment repeated.

The vortexed pellet was spotted to slide-wells (Menzel-Glaser, X1XER308B) and

dried with a 60 ◦C hot-plate. Wells were marked with a wax-pencil (Staedtler R©,

Chinagraph) to reduced run-off and specimen contamination. SA samples were fixed

with absolute methanol for one minute and the slides dried again on the 60 ◦C hot-

plate (2). One of the Candidate’s research team (Russell Connally) conjugated the

novel Eu3+ BHTEGS chelate to a DNA sequence KT18 (Geneworks, 16S68: 5’-

GCAAGCTTCTCGTCCGTT -3’) (1) specific for SA 16S rRNA.

An in situ hybridization assay was applied (4). SA was permeabilized with lysozyme

at 37 ◦C for one hour, rinsed with absolute methanol and incubated at 47 ◦C with

hybridization buffer containing the probe for 30 minutes. The buffer was then rinsed

off with MQ water, the slides air-dried and 10 µl of fluorescence enhancing buffer (FEB)

(148) containing 0.4 mM Eu3+ was spotted to each well. The slides were mounted while

still wet, with a cover-slip and left at room temperature for 20 minutes before viewing.

Slides were viewed with an epifluorescence microscope (BX51, Olympus) fitted with

a 60× objective (UPLFLN, Olympus) and a time-gated auto-synchronous luminescence

detector (GALD) held in its DIC prism slot (148). The GALD device used a 355 nm UV

from a 100 mW YAG laser as the excitation source. Time-resolved images were acquired

after a two second exposure with an Olympus DP72 camera and software (Olympus,


DP2-BSW v2.2) set to 200 ISO and a resolution of 1,360×1,024. When Alexa was

visualized without TGLM, the GALD was locked into the open position (148). The

mean signal intensity (8-bit gray-scale) for BHTEGS and Alexa were compared and

analyzed with ImageJ using standard algorithms (NIH, v1.43u) (Figure 5.4).

5.1.4 Results

In this study, the feasibility of detecting SA with a europium chelate conjugated to

DNA, luminescent in situ hybridization (LISH) assay and time-gating of the lumines-

cent signal (TGLM) was tested. SA was separated from blood along with other debris

in the blood. A LISH assay with BHTEGS conjugated to K18 DNA (1) was com-

pleted in two hours. SA was identified with minimal background signal (Figure 5.3)

and S. epidermidis (SE) (a negative control) was not detected (Figure 5.2). SA was

also identified with BHTEGS in cultures containing no observed debris (Figure 5.1).

As a control the assay was tested with Alexa Fluor R© 488 (Alexa) also conjugated

to the K18 oligonucleotide (Figure 5.1 and 5.3). The SA cells had an Alexa signal

and the SE cells had a partial signal. A comparison of the signal to noise ratio (S/N)

for BHTEGS and Alexa is shown in Figure 5.4 and in Table 5.1. The S/N ratio for

BHTEGS was over five times higher than the same oligonucleotide conjugated to Alexa.

Although it is meant to be viewed under pulsed and not constant UV illumination,

BHTEGS was more photo-sensitive than Alexa. Its signal faded within 30 seconds of

UV excitation, whereas the Alexa signal was still visible after one minute.

BHTEGS was initially tested with a FISH assay reported earlier by the candidate

(4). The buffer contained reagents typically used in FISH washes (0.2818 M NaCl, 20

mM Tris-HCl pH 8.0, 0.01% SDS, 10 mM EDTA, and MQ water). The Eu3+ signal

was weak, so the EDTA was removed and the LISH assay was repeated. The signal

remained low until this washing buffer was replaced with a gentle MQ water rinse. To

compare the S/N of BHTEGS, Alexa was also tested (Figure 5.3). Although a signal

was not observed with BHTEGS and the negative control SE after a MQ water rinse,

weak Alexa signal was observed with SE. No improvement in the signal was observed

when the hybridization buffer with formamide, but without the probe, was applied as


a five minute wash.

The MQ water rinse was used in the final LISH assay. The BHTEGS signal was

further improved by extending incubation times from 30 minutes (4) to one hour for

permeabilization and from 20 minutes (4) to 30 minutes for the hybridization.

The accuracy of the blood model was also confirmed. Blood was separated with Dex-

tran 500 (Pharmacosmos) (166) and intra-cellular SA was observed when the leukocytes

were lysed with water. A comparison of inter-cellular and intra-cellular SA with 0.285

µg/ml DAPI (Sigma, D9542) staining suggested that a majority of the SA underwent

phagocytosis, but remained intact (19, 132) and, when separated from leukocytes and

placed in nutrient broth, was still viable and could be cultured rapidly. When the blood

was lysed by diluting in alkali (138) instead of with MQ water, debris was minimized

and the detection of SA with ISH was unhampered.

5.1.5 Discussion

We set out to test the efficacy of a novel TGLM Eu3+ probe, the chelate BHTEGS, and

a novel LISH assay. The assay was modified from a FISH assay (4) and was applied to

SA separated from whole-blood. The study found that BHTEGS and LISH suppressed

a majority of the autofluorescence of blood-debris. The S/N ratio of SA labeled with

BHTEGS was over five times greater than for SA labeled with Alexa Fluor R© 488

(Alexa) conjugated to the same DNA probe (Figure 5.3) (Table 5.1).

An earlier report applied an in situ hybridization method to label and detect viruses

with a time-resolved chelate BHHCT (149). In that study, europium was conjugated to

the streptavidin molecule and the oligonucleotide was biotinylated. The hybridization

buffer contained 50% formamide with 2×SSC (0.28 M NaCl and 0.28 M sodium citrate,

pH 7.2), 10% dextran sulfate and 0.4 mg/ml salmon sperm DNA. The DNA-biotin

conjugate was hybridized to its mRNA or DNA targets overnight at 42 ◦C and then

bound the next day to the europium conjugate streptavidin by incubating at room

temperature for 30 minutes.

This study also used an in situ hybridization method, but differed significantly in

its approach. SA was permeabilized with lysozyme and not proteinase K, an enzyme


whose use can lead to loss of specimen and cells from the slide. The in situ hybridization

method used a fluorochrome or europium chelate (BHTEGS) conjugated directly to

DNA. The hybridization buffer used also contained formamide and NaCl, but did not

contain sodium citrate, dextran sulfate or salmon sperm DNA.

Hybridization time for the assay was much shorter, 30 minutes instead of overnight.

No biotin-streptavidin incubation step was required (145, 45). The washing was a

simple MQ rinse rather than a series of 2 x SSC washes for one hour. Specimens were

available for viewing within two hours.

Unlike Alexa, BHTEGS was more likely to produce a weak signal if the ISH assay

was not optimized. Blood-debris possibly obstructed permeabilization, hybridization

and visualization of SA. The use of lysozyme simplified the assay, but did produce a

lower level of permeabilization than if lysostaphin was used. To maximize this signal,

blood debris was more completely removed by diluting in alkaline water (138) twice

and the lysozyme incubation was lengthened from half to one hour (4).

Initially a conventional washing buffer with NaCl and EDTA buffer was applied (2).

NaCl and EDTA buffer was observed to weaken the Eu3+ signal and was replaced with

a simple MQ water rinse. The MQ water wash was sufficient for the BHTEGS probe,

but was less effective at removing partially bound Alexa probe. Without NaCl and

EDTA there was less control over the stringency of the wash which resulted occasionally

in an inconsistent signal.

Other problems were encountered with the BHTEGS and the ISH assay. Unlike an

earlier study that produced an immediate signal when Eu3+ was applied to chelates

exterior to the cells (136), in this study LISH required slides to be left for 20 minutes

before an Eu3+ signal could be seen. The five fold increase in the S/N ratio with

TGLM was not as high as in a previous report that tested pond water (136). The

smaller improvement in the S/N ratio might be due to a difference in the type of

specimens tested.

There was a noticeable reduction in signal if the GALD TGLM was used instead of a

regular excitation source and filter (Olympus, U-RFL-T, U-MWIB2) to illuminate the


BHTEGS. Connally et al. (148) calculated that GALD applied with a 260 µm gate-

delay reduced the signal by at least a third. To detect this lower signal, hybridization

of the BHTEGS chelate to SA needed to be optimal. Variation in the blood specimens,

however made this difficult to meet, with every experimental run.

The lysis protocol used to separate SA from whole-blood (138) was necessary for

the removal of the bulk of the non-target material in the blood so that SA could

be labeled with LISH. There maybe, however, other applications for this purification

technique. The viability of the SA that was separated was confirmed by culturing the

pellet in nutrient broth with all specimens becoming turbid within two hours. A recent

study determined the antibiotic susceptibility of SA with FISH, PNA probes and flow-

cytometry (35). It may be possible to combine these two techniques: the separation

of SA from blood (this study) and the detection of antibiotic susceptibility with FISH

(35), so that SA bacteraemia in whole-blood and its resistance to antibiotics can be

determined without first having to perform a two-day blood-culture.

There were limitations to the findings. Clinical isolates, reference strains and other

commonly encountered bacteria were not tested. Because tests were performed at a

non-clinical location, sensitivity and specificity tests were not performed. An in vitro

bacteremia model was used to simulate an auto-fluorescent specimen. It would not

be expected that SA counts in the blood of sepsis patients would be high enough for

detection with LISH or FISH (20) without first culturing the bacteria.

In conclusion, it was possible to separate SA from whole-blood for identification

with TGLM using a Eu3+ chelate. The LISH assay used for hybridizing the probe

was similar to the method used in other clinical microbiology FISH studies (32, 40).

Further investigations of BHTEGS, LISH and TGLM against a range of bacteria and

highly auto-fluorescent specimens might be warranted.


(a) BHTEGS

(b) Alexa Fluor

Figure 5.1: SA in pure cultures of nutrient broth labeled with (a) BHTEGS and (b)Alexa Fluor R© 488. The BHTEGS signal is time-resolved and the Alexa Fluor signalis not. Bar is 5 µm.


(a) SA bright-field (b) SA TGLM

(c) S. epidermidis bright-field (d) S. epidermidis TGLM

Figure 5.2: Staphylococci separated from whole-blood and SA labeled with KT68 andBHTEGS and visualized with (a) bright-field and (b) LISH and TGLM. S. epidermidislabeled with KT68 and visualized with (c) bright-field and (d) LISH and TGLM. Baris 5 µm.


(a) BHTEGS

(b) Alexa Fluor

Figure 5.3: SA separated from whole-blood labeled with (a) BHTEGS and (b) AlexaFluor R© 488. The BHTEGS signal is time-resolved and the Alexa Fluor signal is not.P indicates plot locations in Figure 5.4. Bar is 5 µm.


(a) BHTEGS TGLM

(b) Alexa Fluor R©

Figure 5.4: Plots of the 8-bit Grey scale signal of SA labeled with (a) BHTEGS and(b) Alexa Fluor 488. Plots correspond to the line sample that transects at P in Figure5.3. Bar is 5 µm.


Table 5.1: Signal-to-noise ratio (S/N) of SA labeled∗ with BHTEGS or Alexa Fluor R©488 conjugated to KT68. The BHTEGS signal is time-resolved and the Alexa Fluorsignal is not.

ReporterSA† Background†

S/N

‡

∆ S/N

§

Mean¶ CI¶ Mean CI

BHTEGS 44.5 2.7 5.4 0.03 8.35.5

AlexaFluor R©488

43.8 0.8 28.6 0.03 1.5

∗ LISH method and visualization is described in the Methods section.† Figure 5.3 was used for sampling.¶ Mean signal (8-bit Grey scale) and 95% confidence interval.‡ S/N = Mean (SA)/Mean (Background).§ ∆ S/N = S/N (BHTEGS)/S/N (Alexa).

Acknowledgements

This study was supported by the Australian Research Council’s Linkage Projects

(LP0775196), the Australian Proteome Analysis Facility (APAF), Douglass Hanly Moir

Pathology (Macquarie University Hospital), Hunters Hill Medical Practice and Mac-

quarie University Medical Centre.


Time-resolved fluorescence imaging: a summary

The findings outlined in this Chapter were as follows:

1. SA was incubated in whole blood and was shown to undergo phagocytosis.

2. SA was efficiently separated from most of the blood by lysis with alkali water

and centrifuging.

3. The separated SA could then be either detected with FISH and a conventional

DNA-based fluorophore or rapidly cultured to increase its numbers for detection

again with FISH.

4. To overcome the background signal from the remaining blood-debris, a lumines-

cence in situ hybridization (LISH) assay that was similar to FISH that used a

long-lifetime probe was applied and the probe was visualized with time-gated

luminescence microscopy (TGLM).

5. The signal to noise ratio of the time-resolved long-lifetime probe was higher than

the conventional DNA-based fluorophore and minimal background signal was

observed.

There were limitations to these findings. The synthesis and purification of the

europium chelate probe was not robust and its conjugation to DNA was not optimized

(151). The conditions best suited to the hybridization and washing of the probes were

not fully determined (5). In the technique that was used, a simple wash with MQ

water was applied and unbound probe was sufficiently, but not fully removed. There

were also shortcomings to the TGLM (46). The signal intensity of the excitation source

used with the gated auto-synchronous luminescence detection (GALD) device was low

and because of this, the TGLM signal was weak unless the probe was applied at a high

concentration and the permeabilization was optimal (148).

Further investigation into the ideal use of this novel europium probe and improve-

ments to the excitation source of the GALD were beyond the scope of the current

project (151, 148).

6Conclusion

The central aim of this project was the development of laboratory techniques which

offer (i) specific identification of an important pathogen and (ii) rapid results. These

aims were achieved.

The well-known and serious pathogen Staphylococcus aureus (SA) was chosen as the

model for a series of experiments relevant to these aims. The reason for choosing SA

was because it is an ubiquitous pathogen which commonly infects humans, it has lethal

potential and it has an unrivaled capacity to develop resistance to most antibiotics

(55, 54).

There were three main outcomes of this project. Firstly, current methods for the

conduct of FISH analyses were refined and improved. These advances were achieved

by a range of technical changes to existing methods plus the development of a series of

novel innovations. These changes and innovations included (i) the development of new

probes (specific DNA sequences) for use with FISH techniques (1), (ii) the use of new

high-yield fluorophores (1), (iii) the development and use of pre-mixed materials and

reagents for FISH techniques (2), (iv) use of sealable 50 ml centrifuge tubes to hold

slides in order to reduce the time needed for the permeabilization, hybridization and

washing incubation steps in FISH procedures (2) and (v) the development of a rapid

one-step (in place of a multi-step) permeabilization treatment to significantly reduce

103

104 Conclusion

the time required for FISH procedures (3).

The details are as follows:

1. Development of new probes - new FISH probes specific for SA were developed

(1). The number of potential probes for the identification of SA has doubled. The

new probes still targeted the same region of the SA 16S rRNA as the existing

probes, but had the advantage of greatly increasing the accuracy and specificity

of the FISH techniques.

2. High-yield fluorophores were found to label SA with a higher, more consistent

and photo-stable signal (140) than the fluorophores that are currently used (1).

In particular, Alexa Fluor R© (Invitrogen) and Dylight R© (Jackson) probes were

found to be superior to fluorescein, cyanine and other conventional fluorescent

dyes (2). Although these new high yield fluorophores were more costly than

existing reagents, because of their higher signal intensity they could be used at

lower concentrations and hence a reduced cost.

3. Use of pre-mixed materials and reagents for FISH techniques (2). New prepara-

tion techniques were developed to make it more straightforward to apply FISH

routinely in the microbiology laboratory. The approach of preparing and stor-

ing stock solutions and then mixing them just before use is standard practice in

laboratories. However, the use of pre-mixed materials for use in FISH has not

been previously reported. This procedure has three advantages, (i) reduction in

the time required for completion of FISH analyses, (ii) a significant improvement

in the consistency of FISH based analytical outcomes and (iii) repeated control

FISH tests only need to be run (using reference strains of pathogens) with each

new batch of the basic reagents.

4. Use of sealable centrifuge tubes (2). The use of sealable 50 ml centrifuge tubes

(Greiner, 210-261) to hold slides substantially reduced the time needed for the

permeabilization, hybridization and washing steps of incubation. Permeabiliza-

tion or hybridization activity of reagents was higher, the drying out of reagents

minimized and the results were more consistent.

105

5. The adhesion of the specimens to the slides was improved (2). Additional im-

provements to standard techniques for FISH included the use of agarose which

when applied and dried to the slides before spotting SA reduced their loss and

the use of urea, either diluted in the specimen or applied to specimen already

dried on the slide, also reduced cell loss (5).

6. A two-step permeabilization treatment was developed (2). This is useful when

high molecular weight probes are used to identify SA by FISH techniques. This

two-step permeabilization approach had additional advantages including very fast

outcomes and maintenance of the biological integrity of SA. Finally, because the

level of permeabilization was high, the time needed for high molecular weight

oligonucleotide hybridization was shortened.

7. A series of changes to the current FISH techniques has allowed a reduction in

the time to detect SA, to 24 minutes (3). This compares to an earlier time of

45 minutes to achieve this identification (32). In some clinical circumstances this

reduction in time taken to confirm the identity of SA can be life saving as the

use of particular antibiotics can be dependent on an accurate identification of the

pathogen. These advances in techniques will make it possible to complete both

Gram-stain and FISH analyses within an hour of a positive blood-culture.

Secondly, novel FISH procedures were investigated for the detection of SA. This

was achieved by developing FISH techniques that do not require lysostaphin per-

meabilization (4). Usually, lysostaphin permeabilization is a requirement for DNA

oligonucleotide probes to gain access to the SA bacteria (52). The use of lysostaphin

can complicate FISH analyses (32). A FISH assay that detects SA without requiring

lysostaphin makes its use more practical in routine diagnostic laboratories.

A novel FISH technique free of formamide that used a hot-plate with a precise

temperature control was developed (5). Urea was used to denature and hybridize

an oligonucleotide to SA (153). The assay did not need a dedicated incubator or

water-bath (thus saving expense and bench-space) because it could be performed on

a hot-plate. Urea was less toxic (144) than formamide (142) and produced a more

106 Conclusion

intense signal (153). The same hybridization reagent mix could be used in both the

hybridization and washing steps. This simplified the assay and meant that the wash-

ing step was less likely to produce a non-specific signal. Collectively these technical

developments have greatly simplified FISH analyses.

Thirdly, SA was identified with a luminescence in situ hybridization (LISH) assay

in complex blood-specimens (Chapter 5). A simple procedure was developed for the

creation of an in vitro bacteraemia model with SA (160) and then separating the SA

from the model for testing with FISH and LISH. SA were separated from the sample

by lysing the blood with alkaline water (138), centrifuging to remove the supernatant

and repeating the procedure. Most of the SA in the whole-blood was separated and

collected. The separation technique was accurate, rapid and simple to perform. It was

found to be not only useful for the immediate detection of SA, but also for its rapid

culturing (137). Its application might be useful in other assays such as polymerase

chain reaction (PCR) for the detection of SA (39, 158, 69).

Separated SA from the blood was detected with a FISH and a LISH assay (Chapter

5). The LISH assay visualized the SA with TGLM which, unlike the FISH assay,

suppressed most of the background autofluorescence signal from the blood debris that

remained. A previous study detected viruses labeled with europium (Eu3+) chelates

using an in situ assay and visualized with TGLM, but this ran over-night (149). Unlike

the assay used in this project, it did not use directly conjugated oligonucleotide probes

to detect the clinically important bacteria SA in two hours.

It was not possible to apply all the newly developed FISH techniques at the same

time. The express FISH assay could not be completed in 24 minutes without the en-

zyme lysostaphin (4). Similarly, the two-step permeabilization FISH assay (2) required

the enzyme lysostaphin (4) and could not be completed in 24 minutes (3). Labeling

of europium probes with a LISH assay was not possible with the urea-NaCl based

hybridization reagents (5) and required formamide.

107

Future research directions

The research described in this thesis is ongoing and could take several directions.

1. There is a need to test the newly developed FISH techniques for SA in clinical

settings (40, 33). This is entirely feasible. As a first step these new techniques

could be used in parallel with the existing conventional FISH techniques. This

was shown to be practical as conventional FISH techniques were used as controls

(32) in this current project. It is not expected that the FISH assays would be

less accurate than the conventional FISH techniques used in other studies as they

performed as well as a control (32) with SA collected from whole-blood (Chapter

5).

2. There is also a need to test the new FISH techniques for SA with other pathogens

often detected in clinical microbiology (64). Since the detection of SA with

FISH is more complex, it may be expected that the new methods would be

compatible with these other types of pathogens (88, 52). If tested, a reduction

in the permeabilization treatment would be needed as these other pathogens are

usually more sensitive than SA to such treatment (32). Some of the new FISH

methods might also be applicable and useful with flow-cytometric visualization

(157, 105, 34, 35).

3. The new FISH techniques for SA were confined to the detection of SA with a

single probe since all SA probes target the same overlapping sequence of the

SA 16S rRNA (1). A probe sequence (KT26-1002: 5- AAGGCTCTATCTCTA-

GAGTTGTC -3) found to have a high binding affinity to SA, but not overlapping

with other probes, could be used simultaneously with an established probe for

SA (40). Although the probe also binds to Staphylococcus haemolyticus, it would

be more specific than the Staphy probe which binds to most Staphylococci (40).

4. Probes that target 18S or 23S rRNA could be tested for specificity to SA. Al-

though not as well documented as 16S rRNA, 23S rRNA may prove to have

108 Conclusion

sequences that have a high accuracy and affinity to SA or to S. epidermidis

(167).

5. It is also possible to apply the newly enhanced FISH techniques directly to short-

turnaround blood cultures to determine the identity and antibiotic susceptibility

of pathogens (107, 35). The SA separation from whole-blood was accurate and

did not inactivate the SA. Cultures of this SA could be performed with and

without antibiotics and FISH used to determine the susceptibility of the strain

(107, 35).

6. Specimens other than blood remain to be tested directly with the enhanced FISH

techniques. For example, intravenous and intra-arterial catheters are a significant

focus of infection and are a common cause of septicemia (168). The density of

the SA at these foci is high enough for direct testing with FISH, because they can

act as reservoirs for SA (19). Possibly identification of SA and other microbes of

catheter specimens could potentially be completed within an hour of collection

using the new FISH methods developed in this project.

Since peptide nucleic acid (PNA) probes are expensive (Advandx, AC005), they

were not investigated in this project. However, with the recent expiry of the original

patent (108), it would be expected that PNA probes would become more affordable.

A significant advantage of PNA probes is that SA can be detected without first having

to be permeabilized (102). With this step omitted, it becomes possible to combine

the remaining steps in the FISH procedure into one step, thus greatly simplifying and

shortening the assay (106).

A one step FISH assay could be applied to the detection of SA separated from the

blood (this study) of sepsis patients and grown in rapid blood cultures (137), some

of which contain antibiotics (107, 35). Strains of SA could be identified and their

antibiotic susceptibility ascertained the same day as blood-collection with a one-step

PNA FISH assay and an automated flow cytometer (105, 35) that could also be time-

resolved (169, 170).

AAppendix A: Other publications that

emerged from the thesis

Appendix A is composed of two sections. Each of these sections was published in a

peer reviewed journal and included as such. In the first section, the FISH methodology

of two chronic rhinosinusitis studies by Foreman et al. (129, 128) were assessed in

light of the new FISH methodology developed in this project: Lawson TS, Connally

RE, Vemulpad S, Piper JA. In reference to targeted imaging modality selection for

bacterial biofilms in chronic rhinosinusitis and different biofilms, different disease? a

clinical outcomes study. Laryngoscope 2011;121:2043-2044 (6). In the next section,

the FISH method of the blood-culture study by Wang (40) was assessed, also in light

of the findings of the project: Lawson TS, Connally RE, Iredell JR, Piper JA. The

simultaneous detection and differentiation of staphylococcus species in blood cultures

using fluorescence in situ hybridization: A comment. Med Princ Pract 2011;20:390-391

(7). Foreman et al. (171) and Wang (7) have given permission for their response to be

included in this thesis.

109

The LaryngoscopeVC 2011 The American Laryngological,Rhinological and Otological Society, Inc.

Letter to the Editor

In Reference to Targeted Imaging Modality Selection for BacterialBiofilms in Chronic Rhinosinusitis and Different Biofilms,Different Disease? A Clinical Outcomes Study

Dear Editor:

We read with interest two recent reports on biofilmsin chronic rhinosinusitis (CRS) by Foreman et al.1 andForeman and Wormald.2 We agree with the authors thatfuture studies of species-specific biofilms such as Staphy-lococcus aureus would be an effective approach toidentifying CRS patients who may progress poorly aftersurgery.2 Thus, we will limit our comments to the meth-odologies used to detect biofilms with S aureus.

In Foreman et al.,1 the authors astutely adoptedtwo assays, LIVE/DEAD BacLight (Invitrogen, Carlsbad,CA) and fluorescence in situ hybridization (FISH) (PNAFISH; AdvanDx, Woburn, MA), in parallel with confocallaser scanning microscopy (CLSM) to detect CRS bio-films. The emphasis by the authors on testing specimenswith complementary assays is welcome, but we questionwhy other techniques were not considered. Hematoxylin-eosin staining has reliably detected biofilm in CRSpatients in parallel with FISH.3 Gram staining has eval-uated CRS biofilms with FISH in conjunction withculturing to determine viability.4 The authors mentionedthe necessity of CLSM for biofilm analysis.1 We remarkthat if detection rather than analysis is the aim of thestudy, an epifluorescent microscope with an adjustablestage would be sufficient and less costly.5

For both studies,1,2 a FISH kit (AdvanDx) was runto identify microbes in the CRS biofilms. A kit is conven-ient but can be limiting. Because it is commercial, theFISH method was relatively short on detail; the fixationalcohol, probe sequences, fluorophores, and number ofprobes applied simultaneously were not stated andtherefore were difficult to evaluate. For a larger cohort ofpatients, the cost of a kit can be prohibitive (KT005;AdvanDx,). We remark that the efficacy of AdvanDx FISHwas confirmed with positive blood cultures6 of S aureus,but not with biofilm, intramucosal, and intracellularS aureus.7 The authors commented that they anticipatedwhat microbes were present before testing and used a lim-ited number of probes.1,2 We note that a study, similar toSwidsinski et al.,8 has yet to be done with CRS biofilms tocomprehensively determine their flora.

Differentiating between S aureus in biofilm or inplanktonic form may not be as straightforward as theauthors implied. A more rigorous criterion for detection9

may be required as planktonic S aureus regularly

adheres and clumps.10 We found this was further exas-perated by the alcohol-fixation step in FISH. The authors’reliance on a less intense ‘‘blush’’ of autofluorescence sur-rounding the microbes to represent the matrix of thebiofilm1 was suggestive, but may not confirm the pres-ence of biofilm. Other assays could be used in tandemwith BacLight and FISH to confirm S aureus bio-films.4,10,11 We concede that some of these confirmatoryassays are problematic, as CRS patients commonly yieldnegative sinus cultures.12 We also note that the sensitiv-ity and specificity of the BacLight and FISH were nottested in the studies against positive controls of biofilmproducing S aureus strains.1,2

In Foreman et al.,1 the authors recognized theinability to process a single sample with both BacLightand FISH assays. We agree it would be beneficial if bio-films could be visualized and species identified withinthe same specimen. The potential surface area for eachspecimen would be increased13 and the disparitybetween different assays reduced.1 To address this, wewere surprised that the nucleic acid stains DAPI(D9542; Sigma, St. Louis, MO) or Hoechst (94403;Sigma,) were not used with the BacLight or FISHassay.8 We recognize that a limitation of multiple stain-ing is the potential to remove a fraction of any biofilmpresent.14 Nevertheless, we are not aware of a CRS bio-film study that has taken advantage of combining DAPI,FISH probes at 488 nm excitation, and wheat germagglutinin conjugated to Alexa Fluor at 555 nm(W32464; Invitrogen).4 Such an assay cannot test for vi-ability, but could visualize the majority of microbespresent, identify the species of microbe, and positivelydelineate the biofilm4 within a single field of view.

Typically, CRS biofilm studies harvest sinonasalmucosa surgically from the ethmoid cavity.1,2 The confir-mation in Foreman and Wormald2 that culturescollected at the same time as surgery predicted postin-fection, suggest a role for nonsurgical tests of S aureusbiofilm.15,16 We observed that Keen et al.17 reported col-lecting specimens from nasal bottles and identifyingS aureus biofilms, and that Foreman et al.1 and Fore-man and Wormald2 prepared tissue for biofilm testingby washing to remove planktonic bacteria. It may be in-formative to test for S aureus biofilms suspensions insuch a wash directly11 or indirectly, by testing washcontaminants for strains of S aureus that produce

Laryngoscope 121: September 2011

2043

110 Appendix A: Other publications that emerged from the thesis

A.1 In reference to targeted imaging modality selection for bacterial biofilms

in CRS and different biofilms, different disease?

biofilms.4,10,11 We congratulate the authors on increasingour understanding of the role of biofilm with S aureusin CRS.

THOMAS LAWSON

RUSSELL CONNALLY

SUBRAMANYAM VEMULPAD

JAMES PIPER

Macquarie UniversitySydney, Australia

BIBLIOGRAPHY1. Foreman A, Singhal D, Psaltis AJ, et al. Targeted imaging modality selec-

tion for bacterial biofilms in chronic rhinosinusitis. Laryngoscope 2010;120:427–431.

2. Foreman A, Wormald PJ. Different biofilms, different disease? A clinicaloutcomes study. Laryngoscope 2010;120:1701–1706.

3. Hochstim CJ, Masood R, Rice DH. Biofilm and persistent inflammationin endoscopic sinus surgery. Otolaryngol Head Neck Surg 2010;143:697–698.

4. Kania RE, Lamers GE, Vonk MJ, et al. Characterization of mucosal bio-films on human adenoid tissues. Laryngoscope 2008;118:128–134.

5. Corriveau MN, Zhang N, Holtappels G, et al. Detection of Staphylococcusaureus in nasal tissue with peptide nucleic acid-fluorescence in situhybridization. Am J Rhinol Allergy 2009;23:461–465.

6. Gonzalez V, Padilla E, Gimenez M, et al. Rapid diagnosis of Staphylococ-cus aureus bacteremia using S. aureus PNA FISH. Eur J Clin MicrobiolInfect Dis 2004;23:396–398.

7. Clement S, Vaudaux P, Francois P, et al. Evidence of an intracellular res-ervoir in the nasal mucosa of patients with recurrent Staphylococcusaureus rhinosinusitis. J Infect Dis 2005;192:1023–1028.

8. Swidsinski A, Goktas O, Bessler C, et al. Spatial organisation of micro-biota in quiescent adenoiditis and tonsillitis. J Clin Pathol 2007;60:253–260.

9. Parsek MR, Singh PK. Bacterial biofilms: an emerging link to diseasepathogenesis. Annu Rev Microbiol 2003;57:677–701.

10. Kouidhi B, Zmantar T, Hentati H, et al. Cell surface hydrophobicity, bio-film formation, adhesives properties and molecular detection of adhesinsgenes in Staphylococcus aureus associated to dental caries. MicrobPathog 2010;49:14–22.

11. Oliveira M, Bexiga R, Nunes SF, et al. Biofilm-forming ability profiling ofStaphylococcus aureus and Staphylococcus epidermidis mastitis isolates.Vet Microbiol 2006;118:133–140.

12. Sanderson AR, Leid JG, Hunsaker D. Bacterial biofilms on the sinus mu-cosa of human subjects with chronic rhinosinusitis. Laryngoscope 2006;116:1121–1126.

13. Hoa M, Tomovic S, Nistico L, et al. Identification of adenoid biofilms withmiddle ear pathogens in otitis-prone children utilizing SEM and FISH.Int J Pediatr Otorhinolaryngol 2009;73:1242–1248.

14. Hall-Stoodley L, Hu FZ, Gieseke A, et al. Direct detection of bacterial bio-films on the middle-ear mucosa of children with chronic otitis media.JAMA 2006;296:202–211.

15. Veeh RH, Shirtliff ME, Petik JR, et al. Detection of Staphylococcus aur-eus biofilm on tampons and menses components. J Infect Dis 2003;188:519–530.

16. Homoe P, Bjarnsholt T, Wessman M, et al. Morphological evidence of bio-film formation in greenlanders with chronic suppurative otitis media.Eur Arch Otorhinolaryngol 2009;266:1533–1538.

17. Keen M, Foreman A, Wormald PJ. The clinical significance of nasal irriga-tion bottle contamination. Laryngoscope 2010;120:2110–2114.

Laryngoscope 121: September 2011

2044

A.1 In reference to targeted imaging modality selection forbacterial biofilms in CRS and different biofilms, different disease?111

The LaryngoscopeVC 2011 The American Laryngological,Rhinological and Otological Society, Inc.


In Response to Targeted Imaging Modality Selection for BacterialBiofilms in Chronic Rhinosinusitis and Different Biofilms, DifferentDisease? A Clinical Outcomes Study

Dear Editor,

We acknowledge Lawson et al.’s synthesis of ourrecent work investigating the role of biofilms in chronicrhinosinusitis (CRS), which in essence highlights therelative infancy of this field of research. This is partic-ularly true when compared with the current depth ofknowledge surrounding other biofilm-associated dis-eases such as otitis media with effusion. Clearly, thereis much work still to be done before the role of biofilmsin CRS is completely understood. Staphylococcus aureusdemonstrates an extraordinary repertoire of virulencefactors that enable it to both attack and evade thehuman host, of which biofilm formation is but one. Thevarious roles of planktonic, intracellular, small colonyvariant, and biofilm forms of S. aureus are still to beclearly defined, and our current understanding of thisdisease does not allow us to differentiate the pathologicimportance of each. Further study in this area coupledwith nucleic acid stains, as suggested by Lawson et al.and previously performed by Corriveau et al.,1 to local-ize the bacterial communities relative to the sinonasalepithelium may shed light on the multiplicity of exis-tence many bacteria demonstrate in human disease.

We are firm believers in the biofilm definitions pro-posed by Costerton et al.,2 and use this to guide us indelineating planktonic and biofilm S. aureus, which webelieve is possible by demonstrating adherent bacteria(the washing step removes planktonic clones) that con-gregate in a three-dimensional structure (in our opinion,replacement of the confocal microscope with an epifluor-escent microscope does not allow appreciation of thecharacteristic three-dimensional structure) and surroundthemselves with an exopolysaccharide matrix (the lessintense ‘‘blush’’). This has been successfully applied inboth human and animal work from our department.3–5

Furthermore, although not expressly stated in our arti-cle, S. aureus biofilms do conform to at least the firstfive out of six ‘‘rigorous’’ criteria set out by Parsek andSingh6 for determining a biofilm infection (we are yet toinvestigate the final criteria, colocalizing bacterial cellclusters with host inflammatory cells), highlighting theutility of current diagnostic techniques.

Clearly, there are multiple methods for detectingbiofilms in surgical specimens, of which only two, FISHand BacLight, were used in the studies in question.7,8 It

is important to recognize that all microscopic techniquesrely on the identification of a morphologic appearancecharacteristic of biofilm formation, which in itself is apotentially flawed diagnostic approach. H&E stainingand Gram staining may well complement our currentlyused techniques, but neither could overcome this defi-ciency. The only truly reliable method of detectingbiofilm bacteria will be to demonstrate the genotypicchanges that herald a transition to the biofilm pheno-type. This is not currently possible.

We accept that the requirement for preselection ofFISH probes is a limitation of our species-specific biofilmdiagnostic work to date. Molecular diagnostics have sincebeen employed in CRS patients, both by our group(unpublished data) and others,9 confirming the relativeabundance of S. aureus in this group of patients. The roleof anaerobic bacteria has been raised by molecular stud-ies, although not specifically investigated in CRS as yet.Finally, we would like to draw the attention of the read-ership to another of our articles investigating biofilms inCRS,10 which answers a number of the other methodo-logic queries raised by Lawson et al. Furthermore, it doesseem somewhat contradictory that on one hand the use ofconfocal microscopy would be criticized as being superflu-ous and adding to the complexity of the protocol, whereasthe use of a ‘‘methodology-simplifier’’ such as a commer-cially available FISH kit is also questioned.

In summary, the parallel development of biofilmdiagnostic modalities and our understanding of theirrole in CRS represent an exciting time for Rhinologyresearchers. Open discussion of and analysis of the cur-rent literature will undoubtedly enhance the quality andvalidity of future endeavors in this area.

DR. ANDREW FOREMAN, BMBS (Hons)PETER-JOHN WORMALD, MD

Department of Surgery–Otorhinolaryngology, Head and Neck SurgeryUniversity of Adelaide and Flinders University

Adelaide, Australia

BIBLIOGRAPY1. Corriveau MN, Zhang N, Holtappels G, Van Roy N, Bachert C. Detection

of Staphylococcus aureus in nasal tissue with peptide nucleic acid-fluo-rescence in situ hybridization. Am J Rhinol Allergy 2009;23:461–465.

2. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a commoncause of persistent infections. Science 1999;284:1318–1322.

Laryngoscope 121: September 2011 Foreman: Letter to the Editor

2045


3. Ha KR, Psaltis AJ, Tan L, Wormald PJ. A sheep model for thestudy of biofilms in rhinosinusitis. Am J Rhinol 2007;21:339–345.

4. Psaltis AJ, Ha KR, Beule AG, Tan LW, Wormald PJ. Confocal scanninglaser microscopy evidence of biofilms in patients with chronic rhinosinu-sitis. Laryngoscope 2007;117:1302–1306.

5. Singhal D, Psaltis AJ, Foreman A, Wormald PJ. The impact of biofilms onoutcomes after endoscopic sinus surgery. Am J Rhinol Allergy 2010;24:169–174.

6. Parsek MR, Singh PK. Bacterial biofilms: an emerging link to diseasepathogenesis. Annu Rev Microbiol 2003;57:677–701.

7. Foreman A, Singhal D, Psaltis AJ, Wormald PJ. Targeted imaging modal-ity selection for bacterial biofilms in chronic rhinosinusitis. Laryngo-scope 2010;120:427–431.

8. Foreman A, Wormald P. Different biofilms, different disease? A clinicaloutcomes study. Laryngoscope 2010;120:1701–1706.

9. Stephenson MF, Mfuna L, Dowd SE, et al. Molecular characterization ofthe polymicrobial flora in chronic rhinosinusitis. J Otolaryngol HeadNeck Surg 2010;39:182–187.

10. Foreman A, Psaltis AJ, Tan LW, Wormald PJ. Characterization of bacterialand fungal biofilms in chronic rhinosinusitis. Am J Rhinol Allergy 2009;23:556–561.

Laryngoscope 121: September 2011 Foreman: Letter to the Editor

2046

A.1 In reference to targeted imaging modality selection forbacterial biofilms in CRS and different biofilms, different disease?113

Fax +41 61 306 12 34E-Mail [email protected]


The author modified the commonly accepted FISH protocol [2–4] . It might have been simpler, however, to only adjust the hy-bridization buffer’s formamide concentration and the washing buffer’s NaCl concentrations rather than increasing the incuba-tion temperature to 50 ° C as well [6] . Using 46 ° C with a 0.9 M NaCl hybridization buffer has the advantage of allowing stringency to be optimized to formamide concentrations between 0 and 60% for most probes [2–4, 6] . Similarly, a 48 ° C washing buffer step allows stringency to be optimized to NaCl concentrations between 0.014 and 0.9 M for most probes [3, 4, 6] . FISH protocol turnaround is more likely to be shortened by effective specimen preparation and permeabilization rather than stringency adjustments [4] .

We found that the fluorescein isothiocyanate fluorophore,and to a lesser extent Cy3, suffers from poor signal, bleaching and spectral overlap. In our experience, Alexa Fluor � (Invitrogen) or DyLight Fluor � (Thermo Fisher) probes are superior in signal strength, photostability and spectral flexibility. For instance,fluorescence output could be doubled and potential photobleach-ing and bleed-through halved if fluorescein isothiocyanate and Cy3 were replaced with Alexa Fluor 488 and Alexa Fluor 555 flu-orophores (Invitrogen).

We consider the dual-probe FISH technique described by Wang [1] a valuable addition to the detection of S. aureus which can be further enhanced.

Wang [1] reported a useful slide-based fluorescence in situ hy-bridization (FISH) protocol that involved applying two oligonu-cleotides with different fluorophores simultaneously. Compared to applying the Staphylococcus aureus probe [2] alone, the author demonstrated that using it in conjunction with the Staphylococcus spp. probe [3] increased the sensitivity and specificity of S. aureus detection and differentiation from coagulase-negative staphylo-cocci (CoNS) [1] . We note that the multicolor dual-probe FISH method took at least 2 h to finish. However, it is possible to detect S. aureus with FISH in less than 1 h [4] .

The author described difficulties in cell adhesion to the slides. We found that slides spotted with agarose [5] reduced cell loss and that S. aureus permeabilization responds better to methanol than ethanol [4] . Spotting specimens onto a 10-well diagnostic glass slide (Menzel-Gläser, X1XER308B) rather than smearing was also more effective. A greater concentration of cells was heat-fixed to a smaller slide area, less reagents were needed throughout the FISH protocol and more accurate comparisons could be made be-tween specimens and FISH treatments on the same slide.

The lysis enzyme conditions stated may have resulted in the permeabilization time of at least 20 min. We observe that theoptimal permeabilization temperature is not 30 ° C but approxi-mately 35 ° C for lysozyme (Sigma, L6876) and 45 ° C for lysostaph-in (Sigma, L4402). If specimens were fixed with methanol and then permeabilized at 46 ° C with a lysozyme and lysostaphin mix-ture, the permeabilization time can be shortened to 5 min [4] .

We agree that using a Staphylococcus spp. and an S. aureus probe together may be the most robust probe option as the S. au-reus and CoNS 16S rRNA are conserved, but we observe that an untested CoNS probe is available (CoNS 16S1442: 5 � -CGACG-GCTAGCTCCAAATGGTTACT-3 � ). The EUB338 probe is a nec-essary control as it confirms if the protocol is effective, but we believe that the non-EUB338 probe is not necessary; we are not aware of any instances where it has failed as a negative control in the detection of staphylococci with FISH [2, 3, 5] .

Received: November 1, 2010

Accepted: November 29, 2010

© 2011 S. Karger AG, Basel1011–7571/11/0204–0390$38.00/0

Accessible online at:www.karger.com/mpp

Med Princ Pract 2011;20:390–391

DOI: 10.1159/000324875

The Simultaneous Detection and Differentiation

of Staphylococcus Species in Blood Cultures Using

Fluorescence in situ Hybridization: A Comment

Thomas S. Lawson a , Russell E. Connally a , Jon R. Iredell b ,

James A. Piper a

a Department of Physics, Macquarie University, and b Centre

for Infectious Diseases and Microbiology, Westmead Hospital,

Sydney, N.S.W., Australia

References

1 Wang P: Simultaneous detection and differentiation of Staphylococcus species in blood cultures using fluorescence in situ hybridization. Med Princ Pract 2010; 19: 218–221.

2 Kempf VA, Trebesius K, Autenrieth IB: Fluorescent in situ hybridiza-tion allows rapid identification of microorganisms in blood cultures. J Clin Microbiol 2000; 38: 830–838.

3 Trebesius K, Leitritz L, Adler K, Schubert S, Autenrieth IB, Heesemann J: Culture independent and rapid identification of bacterial pathogens in necrotising fasciitis and streptococcal toxic shock syndrome byfluorescence in situ hybridisation. Med Microbiol Immunol 2000; 188: 169–175.

4 Poppert S, Riecker M, Wellinghausen N, Frickmann H, Essig A: Accel-erated identification of Staphylococcus aureus from blood cultures by a modified fluorescence in situ hybridization procedure. J Med Micro-biol 2010; 59: 65–68.

5 Pernthaler A, Pernthaler J, Amann R: Fluorescence in situ hybridiza-tion and catalyzed reporter deposition for the identification of marine bacteria. Appl Environ Microbiol 2002; 68: 3094–3101.

6 Manz W, Amann R, Ludwig W, Wagner M, Schleifer KH: Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacte-ria: problems and solutions. Syst Appl Microbiol 1992; 15: 593–600.

Thomas S. Lawson Department of Physics , Macquarie University Sydney, NSW 2109 (Australia) Tel. +61 2 9850 8938, E-Mail tomxlawson @ gmail.com


A.2 The simultaneous detection and differentiation of staphylococcus species in

blood cultures using fluorescence in situ hybridization

Letter to the Editor Med Princ Pract 2011;20:390–391 391

I would like to thank Lawson et al. for providing an excellent opinion regarding the article ‘Simultaneous detection and differ-entiation of Staphylococcus species in blood cultures using fluo-rescence in situ hybridization’ [1]. I agree that heat fixation, and pretreatment at 46 ° C with a lysozyme and lysostaphin mixture is helpful for permeabilization.

As for the probe, it is difficult to design a single probe to cover all CoNS species. An alternative strategy could be to use two spe-cific probes that target different conserved sequences (data not published). Non-EUB338 is necessary to detect nonspecific bind-ing of oligonucleotides [2], although it seldom occurs to clinical specimens. The FISH protocol was a modification of that estab-lished by Jansen et al. [3], in which the cell wall of Staphylococcus aureus was rigid, such that an intensive hybridization signal could be taken into account as a priority rather than a turnaround time.

No doubt, Alexa Fluor � (Invitrogen) or DyLight Fluor � (Ther-mo Fisher) probes are superior to fluorescein isothiocyanate and

Cy3 in terms of signal strength; however, the expensive cost hin-ders its wide application in routine clinical laboratory.

There is room for enhancement in the permeabilization and hybridization protocol. I appreciate the effort made by the authors to share their experiences concerning the FISH protocol.

References 1 Wang P: Simultaneous detection and differentiation of Staphylococcus

species in blood cultures using fluorescence in situ hybridization. Med Princ Pract 2010; 19: 218–221.

2 Kempf VA, Trebesius K, Autenrieth IB: Fluorescent in situ hybridiza-tion allows rapid identification of microorganisms in blood cultures. J Clin Microbiol 2000; 38: 830–838.

3 Jansen GJ, Mooibroek M, Idema J, Harmsen HJ, Welling GW, Degener JE: Rapid identification of bacteria in blood cultures by using fluores-cently labeled oligonucleotide probes. J Clin Microbiol 2000; 38: 814–817.

Pei WangDepartment of Laboratory MedicineThe First People’s Hospital of JingmenJingmen 448000 (China)Tel. +86 724 230 5781, E-Mail peiwwien @ yahoo.com

Reply

Pei Wang

Division of Clinical Microbiology, Department of Laboratory

Medicine, The First People’s Hospital of Jingmen, Jingmen,

China

A.2 The simultaneous detection and differentiation of staphylococcusspecies in blood cultures using fluorescence in situ hybridization 115

BAppendix B: Analysis of common

oligonucleotides used in the detection of S.

aureus with FISH

Oligonucleotides were analyzed using mathFISH (mathfish.cee.wisc.edu) (92).

EUB338 16S337: 5’- GCTGCCTCCCGTAGGAGT -3’ (157)

Listing B.1: EUB338 alignment with S. aureus and S. epidermidis 16S rRNA

EUB338 alignment with S. aureus:

....................................

EUB338: TGAGGATGCCCTCCGTCG5 ’

..................||||||||||||||||||

S. aureus: ACTCCTACGGGAGGCAGC3 ’

....................................

EUB338 alignment with S. epidermidis:

....................................

EUB338: 3’TGAGGATGCCCTCCGTCG5 ’

..................||||||||||||||||||

S. epidermidis: 5’ACTCCTACGGGAGGCAGC3 ’

....................................

117

118Appendix B: Analysis of common oligonucleotides used in the

detection of S. aureus with FISH

Table B.1: The calculated binding affinity of the probe EUB338 to S. aureus and S.epidermidis.

∆G (kcal/mol) S. aureus S. epidermidis ∆value∗

∆Go1† -24.5 -24.5 0.00

∆Go2‡ -1.5 -1.5 NA

∆Go3§ -5.3 -5.3 0.00

∆Gooverall

‖ -17.7 -17.7 0.00

[FA]m⊕ (%) 58.0 58.0 0.00

Hybridization Efficiencyd 1.0000 1.0000 0.00

∗ Values for S. epidermidis−Values for S. aureus.† ∆Go

1 indicates the binding affinity of the DNA probe to its RNA target.‡ ∆Go

2 indicates the binding affinity of the DNA probe to itself.§ ∆Go

3. indicates the binding affinity of the RNA target to itself.‖ ∆Go

overall indicates the binding affinity of the DNA probe to its RNA target giventhe competing ∆Go

1 and ∆Go2 reactions.

⊕ The formamide concentration required to anneal or dissociation of 50% of the DNAprobe from its RNA target given 1 µM probe, 0.9 M NaCl in the buffer and 47 ◦Cincubation.d At 0% formamide.

119

KT18 16S68: 5’- GCAAGCTTCTCGTCCGTT -3’ (1)

Listing B.2: KT18-16S68 alignment with S. aureus and S. epidermidis 16S rRNA

KT18 -16S68 alignment with S. aureus:

....................................

....KT18 -16 S68: 3’TTGCCTGCTCTTCGAACG5 ’

..................||||||||||||||||||

S. epidermidis: 5’AACGGACGAGAAGCTTGC3 ’

....................................

KT18 -16S68 alignment with S. epidermidis:

.....................C......T.......

KT18 -16S68: 3’TTG.CTGCTC.TCGAACG5 ’

..................|||.||||||.|||||||

S. epidermidis: 5’AAC.GACGAG.AGCTTGC3 ’

.....................A......G.......

Table B.2: The calculated binding affinity of the probe KT18-16S68 to S. aureus andS. epidermidis.

∆G (kcal/mol) S. aureus S. epidermidis ∆value

∆Go1 -20.8 -14.2 6.60

∆Go2 1.7 1.7 NA

∆Go3 -5.9 -8.2 -2.30

∆Gooverall -14.8 -5.9 8.90

[FA]m (%) 35.3 -27.8 -63.10

Hybridization Efficiency 0.9999 0.0110 -0.99



STAAUR 16S69: 5’- GAAGCAAGCTTCTCGTCCG -3’ (87)

Listing B.3: Staaur alignment with S. aureus and S. epidermidis 16S rRNA

Staaur alignment with S. aureus:

.....................................

Staaur: 3’GCCTGCTCTTCGAACGAAG5 ’

..................|||||||||||||||||||

S. aureus: 5’CGGACGAGAAGCTTGCTTC3 ’

.....................................

Staaur alignment with S. epidermidis:

...................C......T........A.

Staaur: 3’G.CTGCTC.TCGAACGA.G5’

..................|.||||||.||||||||.|

S. epidermidis: 5’C.GACGAG.AGCTTGCT.C3’

...................A......G........C.

Table B.3: The calculated binding affinity of the FISH probe Staaur to S. aureus andS. epidermidis.

∆G (kcal/mol) S. aureus S. epidermidis ∆value

∆Go1 -19.8 -14.7 5.10

∆Go2 -0.4 -0.4 NA

∆Go3 -7.0 -8.2 -1.20

∆Gooverall -12.1 -5.8 6.30

[FA]m (%) 23.7 -50.7 -74.40


121

STAPHY 16S697 5’-TCCTCCATATCTCTGCGC-3’ (87)

Listing B.4: Staphy alignment with S. aureus and S. epidermidis 16S rRNA

Staphy alignment with S. aureus:

...............................

Staphy: 3’CGCGTCTCTATACCTCCT5 ’

.............||||||||||||||||||

S. aureus: 5’GCGCAGAGATATGGAGGA3 ’

...............................

Staphy alignment with E. coli:

................G......T.......

Staphy: 3’CGC.TCTCTA.ACCTCCT5 ’

.............|||.||||||.|||||||

E. coli: 5’GCG.AGAGAT.TGGAGGA3 ’

................T......C.......

Table B.4: The calculated binding affinity of the FISH probe Staphy to S. aureus andS. epidermidis.

∆G (kcal/mol) S. aureus E. coli ∆value

∆Go1 -21.1 -16.2 4.90

∆Go2 2.1 2.1 NA

∆Go3 -3.7 -2.9 0.80

∆Gooverall -17.4 -13.3 4.10

[FA]m (%) 43.7 21.8 -21.90


List of abbreviations

ANOVA Analysis of variance

C Cytosine

CA -MRSA Community -acquired MRSA

CARD -FISH Catalyzed reporter deposition FISH

CoNS Coagulase -negative staphylococci

CRS Chronic rhinosinusitis

Cy3 Cyanine dye excited at 550 nm

Cy5 Cyanine dye excited at 650 nm

DAPI 4’,6-diamidino -2- phenylindole dye excited

at 358 nm

Delta G Gibbs binding potential or free energy values

DEPC Diethylpyrocarbonate

DIC Differential interference and contrast

DNA Deoxyribonucleic acid

EC E. coli

EDTA Ethylenediaminetetraacetic acid

Eu3+ Europium ion excited at 350 nm

FA Formamide

FEB Fluorescence enhancing buffer

FISH Fluorescent in situ hybridisation

FISH Fluorescence in situ hybridization

FISH Fluorescent in situ hybridization

FITC Fluorescein isothiocyanate dye excited

at 495 nm

FN False negative

FP False positive

G Guanine

GALD Gated auto -synchronous luminescence detector

HA -MRSA Hospital -acquired MRSA

HE Hybridization efficiency

IgG Immunoglobulin G antibody

ISH In situ hybridization

LISH Luminescence in situ hybridization

Ln3+ Lanthanide trivalent ions Eu, Dy, Sm and Tb

LNA Locked nucleic acid

M Moles

MC&S Microscopy , culturing and susceptibility

Min Minutes

MQ water Milli -Q water

MRSA Methicillin -resistant Staphylococcus aureus

MSSA Methicillin -sensitive Staphylococcus aureus

123

124 List of abbreviations

NaCl Sodium chloride

NaOH Sodium hydroxide

Oligo Oligonucleotide

PBP2 Penicillin -Binding Protein 2 Gene

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PNA Peptide nucleic acid

RNA Ribonucleic acid

rRNA Ribosomal RNA

S/N Signal to noise ratio

SA S. aureus

SAB Staphylococcus aureus bacteremia

S. aureus Staphylococcus aureus

SCCmecA Staphylococcal cassette chromosome

SDS Sodium dodecyl sulfate

SE S. epidermidis

S. epidermidis Staphylococcus epidermidis

SSC Sodium citrate buffer

TE buffer Tris -HCl and EDTA buffer

TGLM Time -gated luminescence microscopy

TIFF Tagged Image File Format

Tm Melting temperature

TN True negative

TP True positive

Tris -HCl Tris(hydroxymethyl)aminomethane hydrochloric acid

v/v Volume by volume

w/v Weight by volume

References

1. Lawson TS, Connally RE, Vemulpad S, et al. In silico evaluation and testing


Lab Med 2011;42:587–591. vii, 17, 24, 27, 40, 45, 51, 72, 91, 92, 103, 104, 107,

119

2. Lawson TS, Connally RE, Vemulpad S, et al. Optimization of a two-step per-

meabilization fluorescence in situ hybridization (FISH) assay for the detection of

Staphylococcus aureus. J Clin Lab Anal 2011;25:359–365. vii, 17, 24, 31, 34, 40,

44, 45, 47, 49, 51, 72, 91, 94, 103, 104, 105, 106

3. Lawson TS, Connally RE, Vemulpad S, et al. Express fluorescence in situ hy-

bridization methods for the detection of Staphylococcus aureus. Clin lab 2011;

57:789–794. vii, 17, 24, 45, 47, 49, 51, 52, 72, 90, 104, 105, 106

4. Lawson TS, Connally RE, Iredell JR, et al. Detection of Staphylococcus aureus

with a fluorescence in situ hybridization that does not require lysostaphin. J Clin

Lab Anal 2011;25:142–147. vii, 18, 24, 44, 47, 72, 75, 86, 91, 92, 93, 94, 105, 106

5. Lawson TS, Connally RE, Vemulpad S, et al. Dimethyl formamide-free, urea-

NaCl fluorescence in situ hybridization (FISH) assay for Staphylococcus aureus.

Lett Appl Microbiol 2012;10.1111/j.1472-765X.2011.03197.x:(in press). vii,

25, 44, 45, 47, 73, 75, 86, 101, 105, 106

6. Lawson TS, Connally RE, Vemulpad S, et al. In reference to targeted imag-

ing modality selection for bacterial biofilms in chronic rhinosinusitis and differ-

ent biofilms, different disease? a clinical outcomes study. Laryngoscope 2011;

121:2043–2044. vii, 25, 109

7. Lawson TS, Connally RE, Iredell JR, et al. The simultaneous detection and


situ hybridization: A comment. Med Princ Pract 2011;20:390–391. viii, 25, 109

125

126 References

8. Hamey LGC, Connally RE, Yen SW, et al. Luminescent microspheres resolved

from strong background on an automated Time-Gated luminescence microscopy

workstation. DICTA 2009 2009;1:223–228. viii

9. Wang L, Barrett J. Control and prevention of MRSA infections. Methods Mol

Biol 2007;391:209–225. 1, 4

10. Aboud Z, Peters T. Early recognition and management of sepsis at west middlesex

university hospital. Crit Care 2009;13:32. 1, 3, 4

11. Edmond K, Zaidi A. New approaches to preventing, diagnosing, and treating

neonatal sepsis. PLoS Med 2010;7:e1000213. 1

12. Schmidlin M, Alt M, Vogel G, et al. Contaminations of laboratory surfaces with

Staphylococcus aureus are affected by the carrier status of laboratory staff. J Appl

Microbiol 2010;109:1284–1293. 1, 3

13. Palavecino E, et al. Clinical, epidemiological, and laboratory aspects of

methicillin-resistant Staphylococcus aureus (MRSA) infections. Methods Mol Biol

2007;391:1–19. 1, 3

14. Gosbell IB, Barbagiannakos T, Neville SA, et al. Non-multiresistant methicillin-

resistant Staphylococcus aureus bacteraemia in sydney, australia: emergence of

EMRSA-15, oceania, queensland and western australian MRSA strains. Pathology

2006;38:239–244. 1

15. Schulz M, Nonnenmacher C, Mutters R. Cost-effectiveness of rapid MRSA screen-

ing in surgical patients. Eur J Clin Microbiol Infect Dis 2009;28:1291–1296. 1

16. Haji-Michael P, Dodgson K, Kaczmarski E. Guidelines for the management of

sepsis (including neutropenic sepsis). christie.nhs.uk 2010. 1, 4, 5

17. LaRosa S. Prevalence and incidence. clevelandclinicmeded.com 2010. 1, 4

References 127

18. Venkatesh M, Flores A, Luna RA, et al. Molecular microbiological methods in

the diagnosis of neonatal sepsis. Expert Rev Anti Infect Ther 2010;8:1037–1048.

1, 2, 5, 6, 7, 11, 14, 17

19. Krimmer V, Merkert H, Eiff CV, et al. Detection of Staphylococcus aureus and

Staphylococcus epidermidis in clinical samples by 16S rRNA-directed in situ hy-

bridization. J Clin Microbiol 1999;37:2667–2673. 1, 2, 5, 8, 13, 14, 17, 51, 93,

108

20. Llewelyn M, Cohen J. Diagnosis of infection in sepsis. Intensive Care Med 2001;

27:10–32. 1, 4, 5, 14, 95

21. Nimmo GR. Rapid detection of bacteraemia. Clin Microbiol 2010;101:131. 1, 4,

5, 6, 14

22. Shah A, Gordon NC, Pheel L, et al. Comparison of microscopic, phenotypic

and molecular techniques for the rapid identification and susceptibility testing of

staphylococci from positive blood culture bottles. Crit Care 2009;13:44. 1, 5, 6,

10, 12, 14

23. Munson EL, Diekema DJ, Beekmann SE, et al. Detection and treatment of blood-

stream infection: laboratory reporting and antimicrobial management. J Clin

Microbiol 2003;41:495–497. 1, 2, 4, 5, 6, 12, 14

24. Lever A, Mackenzie I. Sepsis: definition, epidemiology, and diagnosis. Br Med J

2007;335:879–883. 1, 4, 5

25. Dryden M, Andrasevic AT, Bassetti M, et al. A European survey of antibiotic

management of methicillin resistant Staphylococcus aureus infection: current clin-

ical opinion and practice. Clin Microbiol Infect 2010;16:3–30. 1, 3, 4

26. Cohen J, Brun-Buisson C, Torres A, et al. Diagnosis of infection in sepsis: an

evidence-based review. Crit Care Med 2004;32:S466. 1, 4, 6

128 References

27. Brown DFJ, Edwards DI, Hawkey PM, et al. Guidelines for the laboratory di-

agnosis and susceptibility testing of methicillin-resistant Staphylococcus aureus

(MRSA). J Antimicrob Chemother 2005;56:1000–1018. 1, 3, 4, 5, 6

28. Mallmann C, Siemoneit S, Schmiedel D, et al. Fluorescence in situ hybridization

to improve the diagnosis of endocarditis: a pilot study. Clin Microbiol Infect 2010;

16:767–773. 1, 5, 7, 13

29. Bottari B, Ercolini D, Gatti M, et al. Application of FISH technology for micro-

biological analysis: current state and prospects. Appl Microbiol Biotechnol 2006;

73:485–494. 2, 9

30. Amann R, Fuchs BM. Single-cell identification in microbial communities by im-

proved fluorescence in situ hybridization techniques. Nat Rev Microbiol 2008;

6:339–348. 2, 9, 24

31. Schweickert B, Moter A, Lefmann M, et al. Let them fly or light them up: matrix-

assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrom-

etry and fluorescence in situ hybridization (FISH). APMIS 2004;112:856–885. 2,

8

32. Poppert S, Riecker M, Wellinghausen N, et al. Accelerated identification of

Staphylococcus aureus from blood cultures by a modified fluorescence in situ hy-

bridization procedure. J Med Microbiol 2010;59:65–68. 2, 5, 8, 9, 10, 11, 12, 13,

15, 17, 18, 24, 28, 30, 41, 43, 44, 45, 47, 48, 49, 51, 72, 95, 105, 107

33. Wu Q, Li Y, Wang M, et al. Fluorescence in situ hybridization rapidly detects

three different pathogenic bacteria in urinary tract infection samples. J Microbiol

Methods 2010;83:175–178. 2, 8, 10, 11, 13, 107

34. Kempf VAJ, Mandle T, Schumacher U, et al. Rapid detection and identifica-

tion of pathogens in blood cultures by fluorescence in situ hybridization and flow

cytometry. Int J Med Mol 2005;295:47–55. 2, 12, 29, 31, 107

References 129

35. Shrestha N, Scalera N, Wilson D, et al. Rapid identification of Staphylococcus

aureus and methicillin resistance by flow cytometry using a peptide nucleic acid

probe. J Clin Microbiol 2011;49:3383–3385. 2, 12, 13, 95, 107, 108

36. Wagner M, Horn M, Daims H. Fluorescence in situ hybridisation for the identifi-

cation and characterisation of prokaryotes. Curr Opin Microbiol 2003;6:302–309.

2, 10, 29, 31

37. Kempf VAJ, Trebesius K, Autenrieth IB. Fluorescent in situ hybridization allows

rapid identification of microorganisms in blood cultures. J Clin Microbiol 2000;

38:830–838. 2, 10, 13, 16, 18

38. Zwirglmaier K. Detection of prokaryotic cells with fluorescence in situ hybridiza-

tion. Methods Mol Biol 2010;659:349–362. 2, 8, 15, 27

39. Poppert S, Essig A, Stoehr B, et al. Rapid diagnosis of bacterial meningitis by

Real-Time PCR and fluorescence in situ hybridization. J Clin Microbiol 2005;

43:3390–3397. 2, 10, 11, 13, 14, 15, 19, 88, 106

40. Wang P. Simultaneous detection and differentiation of staphylococcus species in

blood cultures using fluorescence in situ hybridization. Med Princ Pract 2010;

19:218–221. 2, 10, 13, 14, 16, 18, 42, 72, 95, 107, 109

41. Riedel S, Carroll KC. Blood cultures: key elements for best practices and future

directions. J Infect Chemother 2010;16:301–316. 2, 5, 14

42. Stevens K, Jaykus L. Bacterial separation and concentration from complex sample

matrices: a review. Crit Rev Microbiol 2004;30:7–24. 2, 9, 14, 21

43. Brehm-Stecher B, Young C, Jaykus L, et al. Sample preparation: the forgotten

beginning. J Food Prot 2009;72:1774–1789. 2, 9, 14, 21

44. Moter A, Gobel UB. Fluorescence in situ hybridization (FISH) for direct visual-

ization of microorganisms. J Microbiol Methods 2000;41:85–112. 2, 8, 9, 15, 17,

19, 27, 88

130 References

45. Zwirglmaier K. Fluorescence in situ hybridisation (FISH) - the next generation.

FEMS Microbiol Lett 2005;246:151–158. 2, 10, 16, 22, 47, 72, 89, 90, 94

46. Connally RE, Piper JA. Time-Gated luminescence microscopy. Ann N Y Acad

Sci 2008;1130:106–116. 2, 19, 88, 101

47. Hagan A, Zuchner T. Lanthanide-based time-resolved luminescence immunoas-

says. Anal Bioanal Chem 2011;400:1–18. 2, 14, 19, 21, 25, 89

48. Murdoch DR, Greenlees RL. Rapid identification of Staphylococcus aureus from

BacT/ALERT blood culture bottles by direct gram stain characteristics. Br Med

J 2004;57:199–201. 2, 5, 8, 9, 28, 48

49. Lagace-Wiens PRS, Alfa MJ, Manickam K, et al. Thermostable DNase is superior

to tube coagulase for direct detection of Staphylococcus aureus in positive blood

cultures. J Clin Microbiol 2007;45:3478–3479. 2, 6, 10

50. Hermsen ED, Shull SS, Klepser DG, et al. Pharmacoeconomic analysis of micro-

biologic techniques for differentiating staphylococci directly from blood culture

bottles. J Clin Microbiol 2008;46:2924–2929. 2, 6, 10, 12, 13

51. Sturm PDJ, Kwa D, Vos FJ, et al. Performance of two tube coagulase methods

for rapid identification of Staphylococcus aureus from blood cultures and their

impact on antimicrobial management. Clin Microbiol Infect 2008;14:510–513. 2,

6

52. Tavares A, Inacio J, Melo-Cristino J, et al. Use of fluorescence in situ hybridization

for rapid identification of staphylococci in blood culture samples collected in a

portuguese hospital. J Clin Microbiol 2008;46:3097–3100. 2, 10, 13, 16, 18, 27,

28, 30, 44, 72, 73, 105, 107

53. Hotchkiss R, Karl I. The pathophysiology and treatment of sepsis. N Engl J Med

2003;348:138–150. 3

References 131

54. Parthasarathy P. Outcomes of Staphylococcus aureus bacteremia. Ph.D. thesis,

NUS 2009. 3, 4, 6, 103

55. Collignon P, Nimmo G, Gottlieb T, et al. Staphylococcus aureus bacteremia,

Australia. Emerg Infect Dis 2005;11:554–561. 3, 6, 17, 90, 103

56. Garcia-Lara J, Foster SJ. Anti-Staphylococcus aureus immunotherapy: current

status and prospects. Curr Opin Pharmacol 2009;9:552–557. 3, 7, 8

57. Halablab M, Hijazi S, Fawzi M, et al. Staphylococcus aureus nasal carriage rate

and associated risk factors in individuals in the community. Epidemiol Infect

2010;138:702–706. 3

58. Millar BC, Loughrey A, Elborn JS, et al. Proposed definitions of community-

associated meticillin-resistant Staphylococcus aureus (CA-MRSA). J Hosp Infect

2007;67:109–113. 3

59. Gould I. Costs of hospital-acquired methicillin-resistant Staphylococcus aureus

(MRSA) and its control. Int J Antimicrob Agents 2006;28:379–384. 3

60. Mancini N, Carletti S, Ghidoli N, et al. The era of molecular and other Non-

Culture-Based methods in diagnosis of sepsis. Clin Microbiol Rev 2010;23:235–

251. 4, 5, 10, 14

61. Wasserman E. What can you expect from a microbiology diagnostic service?

CME 2008;26:528–530. 5

62. Kudo M, Matsuo Y, Nakasendo A, et al. Potential clinical benefit of the in situ

hybridization method for the diagnosis of sepsis. J Infect Chemother 2009;15:23–

26. 5, 10, 11, 13, 14, 91

63. Shimada J, Hayashi I, Inamatsu T, et al. Clinical trial of in-situ hybridization

method for the rapid diagnosis of sepsis. J Infect Chemother 1999;5:21–31. 5, 11,

13, 14, 91

132 References

64. Gescher DM, Kovacevic D, Schmiedel D, et al. Fluorescence in situ hybridisation

(FISH) accelerates identification of gram-positive cocci in positive blood cultures.

Int J Antimicrob Agents 2008;32:51–59. 5, 10, 11, 27, 28, 30, 73, 107

65. Jukes L, Mikhail J, Bome-Mannathoko N, et al. Rapid differentiation of Staphy-

lococcus aureus, Staphylococcus epidermidis and other coagulase-negative staphy-

lococci and methicillin susceptibility testing directly from growth positive blood

cultures by multiplex real-time PCR. J Med Microbiol 2010;59:1456–1461. 5, 6,

36

66. Sogaard M, Norgaard M, Schonheyder HC. First notification of positive blood

cultures and the high accuracy of the gram stain report. J Clin Microbiol 2007;

45:1113–1117. 5, 6, 12

67. Kateete DP, Kimani CN, Katabazi FA, et al. Identification of Staphylococcus

aureus : DNase and mannitol salt agar improve the efficiency of the tube coagulase

test. Ann Clin Microbiol Antimicrob 2010;9:23. 6, 10

68. Holtzman C, Whitney D, Barlam T, et al. Assessment of impact of peptide

nucleic acid fluorescence in situ hybridization for rapid identification of Coagulase-

Negative staphylococci in the absence of antimicrobial stewardship intervention.

J Clin Microbiol 2011;49:1581–1582. 6, 13, 14

69. French GL. Methods for screening for methicillin-resistant Staphylococcus aureus

carriage. Clin Microbiol Infect 2009;15:10–16. 6, 106

70. Wellinghausen N, Siegel D, Gebert S, et al. Rapid detection of Staphylococcus

aureus bacteremia and methicillin resistance by real-time PCR in whole blood

samples. Eur J Clin Microbiol Infect Dis 2009;28:1001–1005. 6

71. Mirrett S, Lauer BA, Miller GA, et al. Comparison of acridine orange, methylene

blue, and gram stains for blood cultures. J Clin Microbiol 1982;15:562–566. 7

72. Gitai Z. New fluorescence microscopy methods for microbiology: sharper, faster,

and quantitative. Curr Opin Microbiol 2009;12:341–346. 7, 48

References 133

73. Albeanu D, Soucy E, Sato T, et al. LED arrays as cost effective and efficient light

sources for widefield microscopy. PLoS ONE 2008;3:2146. 7

74. Nishimura M, Shimakita T, Matsuzaki T, et al. Automatic counting of FISH-

labeled microbes by an LED illuminated detecting apparatus. Fish Sci 2008;

74:405–410. 7

75. Fazii P, Ciancaglini E, Riario Sforza G. Differential fluorescent staining method

for detection of bacteria in blood cultures, cerebrospinal fluid and other clinical

specimens. Eur J Clin Microbiol Infect Dis 2002;21:373–378. 7

76. Gmur R, Luthi-Schaller H. A combined immunofluorescence and fluorescent in

situ hybridization assay for single cell analyses of dental plaque microorganisms.

J Microbiol Methods 2007;69:402–405. 7, 8, 14

77. Clausen CR. Detection of bacterial pathogens in purulent clinical specimens by

immunofluorescence techniques. J Clin Microbiol 1981;13:1119. 7, 8

78. Schellenberg J, Smoragiewicz W, Karska-Wysocki B. A rapid method combining

immunofluorescence and flow cytometry for improved understanding of compet-

itive interactions between lactic acid bacteria (LAB) and methicillin-resistant S.

aureus (MRSA) in mixed culture. J Microbiol Methods 2006;65:1–9. 7, 8

79. Oliveira M, Bexiga R, Nunes SF, et al. Biofilm-forming ability profiling of Staphy-

lococcus aureus and Staphylococcus epidermidis mastitis isolates. Vet Microbiol

2006;118:133–140. 8, 9, 13, 48

80. Oliveira M, Nunes SF, Carneiro C, et al. Time course of biofilm formation by

Staphylococcus aureus and Staphylococcus epidermidis mastitis isolates. Vet Mi-

crobiol 2007;124:187–191. 8, 13

81. Forsgren A, Forsum U. Agglutination of Staphylococcus aureus by rabbit sera.

Infect Immun 1972;5:524–530. 8

134 References

82. Smith P, McCoy E, Wilson J. Identification of staphylococci in nonfat dry milk

by the fluorescent antibody technique [1] and [2]. J Dairy Sci 1962;45:729–734. 8

83. Forsgren A, Sjoquist J. Protein A from S. aureus. I. pseudo-immune reaction with

human gamma-globulin. J Immunol 1966;97:822–827. 8, 22

84. Hugenholtz P, Tyson GW, Blackall LL. Design and evaluation of 16S rRNA-

targeted oligonucleotide probes for fluorescence in situ hybridization. Methods

Mol Biol 2002;179:29–42. 9, 11, 27, 32, 36, 45, 48

85. Malic S, Hill KE, Hayes A, et al. Detection and identification of specific bacteria in

wound biofilms using peptide nucleic acid fluorescent in situ hybridization (PNA

FISH). Microbiology 2009;155:2603–2611. 9, 11, 13, 31

86. Baldrich E, Vigues N, Mas J, et al. Sensing bacteria but treating them well:

Determination of optimal incubation and storage conditions. Anal Biochem 2008;

383:68–75. 9, 17, 41

87. Trebesius K, Leitritz L, Adler K, et al. Culture independent and rapid identifica-

tion of bacterial pathogens in necrotising fasciitis and streptococcal toxic shock

syndrome by fluorescence in situ hybridisation. Med Microbiol Immunol 2000;

188:169–175. 9, 10, 12, 13, 31, 32, 40, 41, 120, 121

88. Jansen GJ, Mooibroek M, Idema J, et al. Rapid identification of bacteria in blood

cultures by using fluorescently labeled oligonucleotide probes. J Clin Microbiol

2000;38:814–817. 9, 13, 27, 107

89. Kalliomaki M, Collado MC, Salminen S, et al. Early differences in fecal microbiota

composition in children may predict overweight. Am J Clin Nutr 2008;87:534–538.

10, 11, 13

90. Hogardt M, Trebesius K, Geiger AM, et al. Specific and rapid detection by

fluorescent in situ hybridization of bacteria in clinical samples obtained from

cystic fibrosis patients. J Clin Microbiol 2000;38:818–825. 10, 11, 13

References 135

91. Hoshino T, Yilmaz LS, Noguera DR, et al. Quantification of target molecules

needed to detect microorganisms by fluorescence in situ hybridization (FISH) and

catalyzed reporter deposition-FISH. Appl Environ Microbiol 2008;74:5068–5077.

10

92. Yilmaz LS, Parnerkar S, Noguera DR. mathFISH, a web tool that uses

thermodynamics-based mathematical models for in silico evaluation of oligonu-

cleotide probes for fluorescence in situ hybridization. Appl Environ Microbiol

2011;77:1118–22. 10, 11, 16, 32, 33, 34, 36, 37, 38, 45, 46, 117

93. Chapin K, Musgnug M. Evaluation of three rapid methods for the direct identi-

fication of Staphylococcus aureus from positive blood cultures. J Clin Microbiol

2003;41:4324–4327. 10, 15

94. Pernthaler J, Pernthaler A, Amann R. Automated enumeration of groups of ma-

rine picoplankton after fluorescence in situ hybridization. Appl Environ Microbiol

2003;69:2631–2637. 10, 31

95. Szweda P, Kotlowski R, Kur J. New effective sources of the Staphylococcus sim-

ulans lysostaphin. J Biotechnol 2005;117:203–213. 10, 15, 18, 44, 72

96. Yilmaz LS, Bergsven LI, Noguera DR. Systematic evaluation of single mismatch

stability predictors for fluorescence in situ hybridization. Environ Microbiol 2008;

10:2872–2885. 11, 38, 39

97. Johnson M, Zaretskaya I, Raytselis Y, et al. NCBI BLAST: a better web interface.

Nucleic Acids Res 2008;36:5–9. 11, 36

98. Yilmaz LS, Okten HE, Noguera DR. Making all parts of the 16S rRNA of Es-

cherichia coli accessible in situ to single DNA oligonucleotides. Appl Environ

Microbiol 2006;72:733–744. 11, 25, 38, 39

99. Yilmaz LS, Noguera DR. Development of thermodynamic models for simulating

probe dissociation profiles in fluorescence in situ hybridization. Biotechnol Bioeng

2007;96:349–363. 11, 16, 38

136 References

100. Matsuhisa A, Saito Y, Ueyama H, et al. Detection of staphylococci in mouse

phagocytic cells by in situ hybridization using biotinylated DNA probes. Biotech

Histochem 1994;69:31–37. 11, 13, 14, 15, 40, 42, 90, 91

101. Hawkins RC. Laboratory turnaround time. Clin Biochem Rev 2007;28:179–194.

11, 12, 51

102. Perry-O’Keefe H, Rigby S, Oliveira K, et al. Identification of indicator microor-

ganisms using a standardized PNA FISH method. J Microbiol Methods 2001;

47:281–292. 12, 13, 27, 34, 40, 41, 48, 108

103. Oliveira K, Brecher SM, Durbin A, et al. Direct identification of Staphylococcus

aureus from positive blood culture bottles. J Clin Microbiol 2003;41:889–891. 12,

13, 15

104. Gonzalez V, Padilla E, Gimenez M, et al. Rapid diagnosis of Staphylococcus

aureus bacteremia using S. aureus PNA FISH. Eur J Clin Microbiol Infect Dis

2004;23:396–398. 12, 13

105. Hartmann H, Stender H, Schafer A, et al. Rapid identification of Staphylococcus

aureus in blood cultures by a combination of fluorescence in situ hybridization

using peptide nucleic acid probes and flow cytometry. J Clin Microbiol 2005;

43:4855–4857. 12, 13, 29, 40, 107, 108

106. Fiandaca M, Oliveira K, Stender H. Patent US20070128646 - Hybridization of

PNA probes in alcohol solutions 2007. 12, 108

107. Regnault B, Martin-Delautre S, Grimont PAD. Problems associated with the

direct viable count procedure applied to gram-positive bacteria. Int J Food Mi-

crobiol 2000;55:281–284. 12, 13, 108

108. Nielsen PE, Egholm M, Berg RH, et al. Sequence-selective recognition of DNA

by strand displacement with a thymine-substituted polyamide. Science (80) 1991;

254:1497–1500. 12, 108

References 137

109. Oliveira K, Procop GW, Wilson D, et al. Rapid identification of Staphylococcus

aureus directly from blood cultures by fluorescence in situ hybridization with

peptide nucleic acid probes. J Clin Microbiol 2002;40:247–251. 13, 40

110. Sogaard M, Stender H, Schonheyder HC. Direct identification of major blood

culture pathogens, including Pseudomonas aeruginosa and Escherichia coli, by a

panel of fluorescence in situ hybridization assays using peptide nucleic acid probes.

J Clin Microbiol 2005;43:1947–1949. 13

111. Forrest GN, Mehta S, Weekes E, et al. Impact of rapid in situ hybridization

testing on coagulase-negative staphylococci positive blood cultures. J Antimicrob

Chemother 2006;58:154–158. 13

112. Peters RP, Savelkoul PH, Simoons-Smit AM, et al. Faster identification of

pathogens in positive blood cultures by fluorescence in situ hybridization in rou-

tine practice. J Clin Microbiol 2006;44:119–123. 13

113. Peters RPH, Agtmael MAV, Simoons-Smit AM, et al. Rapid identification of

pathogens in blood cultures with a modified fluorescence in situ hybridization

assay. J Clin Microbiol 2006;44:4186–4188. 13

114. Horvath A, Kristof K, Konkoly-Thege M, et al. Rapid identification of pathogens

in blood culture with fluorescent in situ hybridization (FISH). Acta Microbiol

Immunol Hung 2010;57:225–234. 13

115. Kipp F, Ziebuhr W, Becker K, et al. Detection of Staphylococcus aureus by 16S

rRNA directed in situ hybridisation in a patient with a brain abscess caused by

small colony variants. J Neurol Neurosurg Psychiatry 2003;74:1000–1002. 13

116. Alt V, Lips KS, Henkenbehrens C, et al. A new animal model for implant-

related infected non-unions after intramedullary fixation of the tibia in rats with

fluorescent in situ hybridization of bacteria in bone infection. Bone 2011;48:1146–

1153. 13, 18, 34, 45

138 References

117. Hall-Stoodley L, Hu FZ, Gieseke A, et al. Direct detection of bacterial biofilms

on the Middle-Ear mucosa of children with chronic otitis media. JAMA 2006;

296:202–211. 13

118. Hoa M, Tomovic S, Nistico L, et al. Identification of adenoid biofilms with middle

ear pathogens in otitis-prone children utilizing SEM and FISH. Int J Pediatr

Otorhinolaryngol 2009;73:1242–1248. 13

119. Homoe P, Bjarnsholt T, Wessman M, et al. Morphological evidence of biofilm for-

mation in Greenlanders with chronic suppurative otitis media. Eur Arch Otorhi-

nolaryngol 2009;266:1533–1538. 13

120. Nistico L, Gieseke A, Stoodley P, et al. Fluorescence ’in situ’ hybridization for

the detection of biofilm in the middle ear and upper respiratory tract mucosa.

Methods Mol Biol 2009;493:191–213. 13

121. Veeh RH, Shirtliff ME, Petik JR, et al. Detection of Staphylococcus aureus biofilm

on tampons and menses components. J Infect Dis 2003;188:519–530. 13, 14, 19,

88

122. Sanderson AR, Leid JG, Hunsaker D. Bacterial biofilms on the sinus mucosa of

human subjects with chronic rhinosinusitis. Laryngoscope 2006;116:1121–1126.

13

123. Huvenne W, Zhang N, Tijsma E, et al. Pilot study using doxycycline-releasing

stents to ameliorate postoperative healing quality after sinus surgery. Wound

Repair Regen 2008;16:757–767. 13

124. Healy DY, Leid JG, Sanderson AR, et al. Biofilms with fungi in chronic rhinosi-

nusitis. Otolaryngol Head Neck Surg 2008;138:641–647. 13

125. Niederfuhr A, Kirsche H, Deutschle T, et al. Staphylococcus aureus in nasal lavage

and biopsy of patients with chronic rhinosinusitis. Allergy 2008;63:1359–1367. 13,

19, 88

References 139

126. Corriveau MN, Zhang N, Holtappels G, et al. Detection of Staphylococcus aureus

in nasal tissue with peptide nucleic acid-fluorescence in situ hybridization. Am J

Rhinol Allergy 2009;23:461–465. 13

127. Foreman A, Psaltis AJ, Tan LW, et al. Characterization of bacterial and fungal

biofilms in chronic rhinosinusitis. Am J Rhinol Allergy 2009;23:556–561. 13

128. Foreman A, Wormald PJ. Different biofilms, different disease? a clinical outcomes

study. Laryngoscope 2010;120:1701–1706. 13, 109

129. Foreman A, Singhal D, Psaltis AJ, et al. Targeted imaging modality selection for

bacterial biofilms in chronic rhinosinusitis. Laryngoscope 2010;120:427–431. 13,

109

130. Sachse F, Becker K, von Eiff C, et al. Staphylococcus aureus invades the epithelium

in nasal polyposis and induces IL-6 in nasal epithelial cells in vitro. Allergy 2010;

65:1430–1437. 13

131. Swidsinski A, Goktas O, Bessler C, et al. Spatial organisation of microbiota in

quiescent adenoiditis and tonsillitis. J Clin Pathol 2007;60:253–260. 13

132. Zautner AE, Krause M, Stropahl G, et al. Intracellular persisting Staphylococcus

aureus is the major pathogen in recurrent tonsillitis. PLoS One 2010;5:e9452. 13,

93

133. Kirketerp-Moller K, Jensen PO, Fazli M, et al. Distribution, organization, and

ecology of bacteria in chronic wounds. J Clin Microbiol 2008;46:2717–2722. 13

134. Thomsen T, Aasholm M, Rudkjobing V, et al. The bacteriology of chronic venous

leg ulcer examined by culture-independent molecular methods. Wound Repair

Regen 2010;18:38–49. 13

135. Stoodley P, Conti S, DeMeo P, et al. Characterization of a mixed MRSA/MRSE

biofilm in an explanted total ankle arthroplasty. FEMS Immunol Med Mic 2011;

62:66–74. 13

140 References

136. Connally R, Veal D, Piper J. Time-resolved fluorescence microscopy using an

improved europium chelate BHHST for the in situ detection of cryptosporidium

and giardia. Microsc Res Tech 2004;64:312–322. 14, 19, 21, 22, 89, 94

137. Bernhardt M, Pennell D, Almer L, et al. Detection of bacteria in blood by cen-

trifugation and filtration. J Clin Microbiol 1991;29:422–425. 14, 18, 106, 108

138. Decleva E, Menegazzi R, Busetto S, et al. Common methodology is inadequate

for studies on the microbicidal activity of neutrophils. J Leukocyte Biol 2006;

79:87–94. 14, 19, 21, 42, 87, 91, 93, 94, 95, 106

139. Pernthaler A, Pernthaler J, Amann R. Fluorescence in situ hybridization and cat-

alyzed reporter deposition for the identification of marine bacteria. Appl Environ

Microbiol 2002;68:3094–3101. 15, 42, 43

140. Bouvier T, Del Giorgio PA. Factors influencing the detection of bacterial cells

using fluorescence in situ hybridization (FISH): a quantitative review of published

reports. FEMS Microbiol Ecol 2003;44:3–15. 15, 16, 27, 33, 34, 104

141. Yan C, Ding B, Lan X, et al. The toxicity study on marine low-temperature

lysozyme. Food Chem Toxicol 2008;46:604–609. 15, 44, 73

142. Manz W, Amann R, Ludwig W, et al. Phylogenetic oligodeoxynucleotide probes

for the major subclasses of proteobacteria: problems and solutions. Syst Appl

Microbiol 1992;15:593–600. 16, 18, 31, 32, 35, 45, 47, 105

143. Manz W, Eisenbrecher M, Neu TR, et al. Abundance and spatial organization

of gram-negative sulfate-reducing bacteria in activated sludge investigated by in

situ probing with specific 16S rRNA targeted oligonucleotides. FEMS Microbiol

Ecol 1998;25:43–61. 16, 34, 35, 45

144. Simard C, Lemieux R, Cote S. Urea substitutes toxic formamide as destabilizing

agent in nucleic acid hybridizations with RNA probes. Electrophoresis 2001;

22:2679–2683. 16, 18, 32, 72, 105

References 141

145. Knoll J, Lichter P, Bakdounes K, et al. In situ hybridization and detection using

nonisotopic probes. Curr Protoc Mol Biol 2007;14:Unit 14.7. 16, 47, 89, 94

146. Bossuyt PM, Irwig L, Craig J, et al. Comparative accuracy: assessing new tests

against existing diagnostic pathways. BMJ 2006;332:1089. 17

147. Cisani G, Varaldo PE, Grazi G, et al. High-level potentiation of lysostaphin

anti-staphylococcal activity by lysozyme. Antimicrob Agents Chemother 1982;

21:531–535. 17, 27, 30, 43, 44, 48, 72

148. Connally R. A device for Gated Auto-Synchronous Luminescence Detection

(GALD). Anal Chem 2011;83:4782–4787. 19, 20, 21, 22, 23, 47, 48, 90, 91,

92, 95, 101

149. Seveus L, Vaisala M, Syrjanen S, et al. Time-resolved fluorescence imaging of

europium chelate label in immunohistochemistry and in situ hybridization. Cy-

tometry 1992;13:329–338. 21, 22, 88, 90, 93, 106

150. de Haas R, Verwoerd N, Van der Corput M, et al. The use of peroxidase-

mediated deposition of biotin-tyramide in combination with time-resolved flu-

orescence imaging of europium chelate label in immunohistochemistry and in situ

hybridization. J Histochem Cytochem 1996;44:1091–1099. 21, 22, 90

151. Connally R, Jin D, Piper J. BHHST: an improved lanthanide chelate for time-

resolved fluorescence applications. In Proc SPIE, vol. 5704. 93. 21, 22, 89, 101

152. Weise A, Mrasek K, Ewers E, et al. Diagnostic applications of fluorescence in situ

hybridization. Expert Opin Med Diagn 2009;3:453–460. 29

153. Hutton JR. Renaturation kinetics and thermal stability of DNA in aqueous so-

lutions of formamide and urea. Nucleic Acids Res 1977;4:3537–3555. 31, 32, 33,

39, 45, 47, 105, 106

154. Cole JR, Wang Q, Cardenas E, et al. The ribosomal database project: improved

alignments and new tools for rRNA analysis. Nucleic Acids Res 2008;37:141–145.

36

142 References

155. Loy A, Arnold R, Tischler P, et al. probeCheck-a central resource for evaluating

oligonucleotide probe coverage and specificity. Environ Microbiol 2008;10:2894–

2898. 36

156. Yilmaz LS, Noguera DR. Mechanistic approach to the problem of hybridiza-

tion efficiency in fluorescent in situ hybridization. Appl Environ Microbiol 2004;

70:7126–7139. 36, 38

157. Amann RI, Binder BJ, Olson RJ, et al. Combination of 16S rRNA-targeted

oligonucleotide probes with flow cytometry for analyzing mixed microbial popu-

lations. Appl Environ Microbiol 1990;56:1919–1925. 40, 107, 117

158. Thomas LC, Gidding HF, Ginn AN, et al. Development of a real-time Staphy-

lococcus aureus and MRSA (SAM-) PCR for routine blood culture. J Microbiol

Methods 2007;68:296–302. 41, 90, 106

159. Hinds A, Peterson G. Method for standardizing staphylococcal suspensions: Re-

lationship of optical density to viable cell count. J Bacteriol 1963;86:168. 41,

91

160. Hockey L, Fujita N, Gibson T, et al. Detection of fungemia obscured by concomi-

tant bacteremia: in vitro and in vivo studies. J Clin Microbiol 1982;16:1080–1085.

42, 91, 106

161. Krasnoperov L, Marras S, Kozlov M, et al. Luminescent probes for ultrasensitive

detection of nucleic acids. Bioconjugate Chem 2010;21:319–327. 88, 89

162. Soini E, Lovgren T, Reimer C. Time-resolved fluorescence of lanthanide probes

and applications in biotechnology. Crit Rev Anal Chem 1987;18:105–154. 88, 89

163. Selvin P. Principles and biophysical applications of lanthanide-based probes.

Annu Rev Biophys Biomol Struct 2002;31:275–302. 89

164. Bunzli J, Piguet C. Taking advantage of luminescent lanthanide ions. Chem Soc

Rev 2005;34:1048–1077. 89

References 143

165. Bunzli J. Lanthanide luminescence for biomedical analyses and imaging. Chem

Rev 2010;110:2729–2755. 89, 90

166. Nauseef W. Isolation of human neutrophils from venous blood. Methods Mol Biol

2007;412:15–20. 93

167. Roller C, Wagner M, Amann R, et al. In situ probing of gram-positive bacte-

ria with high DNA G + C content using 23S rRNA-targeted oligonucleotides.

Microbiology 1994;140:2849–2858. 108

168. Fowler JVG, Justice A, Moore C, et al. Risk factors for hematogenous compli-

cations of intravascular Catheter-Associated Staphylococcus aureus bacteremia.

Clin Infect Dis 2005;40:695–703. 108

169. Jin D, Connally R, Piper J. Practical time-gated luminescence flow cytometry. I:

concepts. Cytom Part A 2007;71:783–796. 108

170. Jin D, Connally R, Piper J. Practical time-gated luminescence flow cytometry. II:

experimental evaluation using UV LED excitation. Cytom Part A 2007;71:797–

808. 108

171. Foreman A, Wormald P. In response to targeted imaging modality selection for

bacterial biofilms in chronic rhinosinusitis and different biofilms, different disease?

a clinical outcomes study. Laryngoscope 2011;121:2045–2046. 109

The specific identification of Staphylococcus aureus with new fluorescence in situ hybridization...

Documents

Transcript of The specific identification of Staphylococcus aureus with new fluorescence in situ hybridization...