Methodology for the measurement of HIF-1α in bovine ...
Transcript of Methodology for the measurement of HIF-1α in bovine ...
Methodology for the measurement of HIF-1α in bovine leukocytes and assessment of the molecule as a biomarker for bovine respiratory disease outcome
Sarah Gestier BVSc
A thesis submitted for the degree of Master of Philosophy at
The University of Queensland in 2014
School of Veterinary Science
ii
Abstract
Bovine respiratory disease (BRD) is a multi-factorial inflammatory respiratory disease complex and
is a significant problem for the beef industry. While the infectious agents involved are known and
vaccines exist for some, the host factors determining outcome of infections are still only partially
understood. The aim of this project was to explore the usefulness of an inflammation regulatory
molecule as a biomarker for susceptibility/resistance to the development of BRD in beef cattle.
The transcription factor HIF-1α is recognised for its importance in the development and
coordination of an immune response to hypoxia and inflammation caused by infectious agents. For
this reason, HIF-1α was identified as an attractive biomarker candidate for BRD susceptibility.
The initial objective of this study was to develop methods by which the level of HIF-1α stabilisation
in response to hypoxia could be measured in key immune cells. Once developed, the second
objective of the study was to use these methodologies to determine whether there was significant
positive or negative correlation between the regulation of HIF-1α and the development of clinical
BRD. The experimental hypothesis was that there is a significant difference in HIF-1α regulation
between animals with clinical BRD compared with other clinically healthy animals, and thus in
response to hypoxia there would be a difference in the lymphocyte and monocyte HIF-1α
expression between these two groups of cattle.
This project developed a methodology for hypoxic culture of bovine lymphocytes and monocytes to
induce HIF-1α stabilisation. Tube cultures for lymphocyte analysis were cultured with 1.25 x106
cells in 250 µl serum-free medium suspension treated with 200 µM cobalt chloride and 5 µg/ml
LPS; with a non-stimulated control tube containing only the cells suspended in serum-free medium.
Chamber slides for monocyte analysis were cultured containing 2.5 x 106 cells in 500µl serum-free
medium with or without stimulants. Tubes and chamber slides were cultured for 18 hours in 5%
CO2 at 37°C. Tube cultured cells underwent post-culture fluorescent staining using a fluorochrome-
conjugated HIF-1α antibody for measurement of HIF-1α fluorescence in B lymphocytes and T
lymphocytes using flow cytometry. T lymphocytes were labelled using a fluorochrome-conjugated
anti-CD3 antibody and B lymphocytes labelled using a fluorochrome-conjugated anti-CD20
antibody. An additional protocol was developed to measure monocyte HIF-1α expression using
immunocytochemical staining of chamber slide cultures. This alternative approach was necessitated
by the finding of poor monocyte survival during hypoxic culture and the the pre-flow cytometry
staining procedures. These methodologies for the stabilisation and measurement of HIF-1α in
bovine lymphocytes and monocytes were then applied to a sample population of 88 feedlot cattle,
34 of which had been treated in the feedlot for clinical BRD (cases), while the remainder was
untreated, clinically healthy controls. The groups were compared using two-sample t-tests for both
iii
flow cytometry and immunocytochemistry. No statistical significance difference was detected in the
HIF-1α expression of leukocytes when cases were compared to control animals. However, due to
time constraints only a relatively small number of animals were tested and the possibility remains
that with a larger cohort the HIF-1α response may still prove useful alone or in combination with
other biomarkers in the setting of BRD. Further work is needed to improve these methodologies, or
develop alternative methods. If this can be done successfully a higher powered study with a larger
cohort of animals could be undertaken. However the extent of the within-group variation of cases
and controls suggests that any difference between the groups would need to be very substantial for
such a difference to be statistically significant, even with a much larger sample size.
Based on this study, the expression of HIF-1α and its value as a biomarker for the development of
clinical BRD as a disease outcome could still be significant and remains to be established. The
expression of HIF-1α could still be shown as significant in the context of the bovine acute immune
responses to BRD viral pathogens if further investigations incorporated relevant and appropriate
viral challenges. The lack of a detected difference in HIF-1α could also mean that HIF-1α does not
have a significant role in the pathogenesis of clinical BRD, and further studies involving viral
challenge could also help to establish or refute this possibility.
iv
Declaration by author
(All candidates to reproduce this section in their thesis verbatim)
This thesis is composed of my original work, and contains no material previously published or
written by another person except where due reference has been made in the text. I have clearly
stated the contribution by others to jointly-authored works that I have included in my thesis.
I have clearly stated the contribution of others to my thesis as a whole, including statistical
assistance, survey design, data analysis, significant technical procedures, professional editorial
advice, and any other original research work used or reported in my thesis. The content of my thesis
is the result of work I have carried out since the commencement of my research higher degree
candidature and does not include a substantial part of work that has been submitted to qualify for
the award of any other degree or diploma in any university or other tertiary institution. I have
clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.
I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,
subject to the General Award Rules of The University of Queensland, immediately made available
for research and study in accordance with the Copyright Act 1968.
I acknowledge that copyright of all material contained in my thesis resides with the copyright
holder(s) of that material. Where appropriate I have obtained copyright permission from the
copyright holder to reproduce material in this thesis.
v
Publications during candidature
No publications.
Publications included in this thesis
No publications included.
vi
Contributions by others to the thesis
Dr Helle Bielefeldt-Ohmann contributed to the conception and design of the project, performed
reading of ICC slides as well as critical revision of written work. Dr Tamsin Barnes contributed to
the analysis and interpretation of experimental data and revision of written work. Dr Helen Owens
contributed to the revision of written work. Dr Nana Satake contributed flow cytometry technical
and interpretative expertise.
Statement of parts of the thesis submitted to qualify for the award of another degree
None.
vii
Acknowledgements
Acknowledgement and gratitude to Dr Helle Bielefeldt-Ohmann, Dr Tamsin Barnes and Dr Helen
Owens for their supportive roles as supervisors, to the Meat and Livestock Association for financial
support for a scholarship and research funds and to the technical staff at the School of Veterinary
Science, University of Queensland, Gatton.
viii
Keywords
hypoxia-inducible factor-1α, biomarker, bovine respiratory disease
Australian and New Zealand Standard Research Classifications (ANZSRC)
ANZSRC code: 070705, Veterinary Immunology, 100%
Fields of Research (FoR) Classification
FoR code: 0707, Veterinary Sciences, 100%
ix
Table of Contents
Abstract ii Declaration by author iv Publications during candidature v Publications included in this thesis v Contributions by others to the thesis vi Statement of parts of the thesis submitted to qualify for the award of another degree vi Acknowledgements vii Keywords viii Australian and New Zealand Standard Research Classifications (ANZSRC) viii Fields of Research (FoR) Classification viii Table of Contents ix List of Figures xii List of Tables xiv Abbreviations xv 1 Introduction, Literature Review, Conclusions and Research Proposal 1 1.1 Introduction 1 1.2 Literature Review 1 1.2.1 HIF-1α regulation during normoxia, development and homeostasis 3 1.2.2 HIF-1α during hypoxic conditions 5 1.2.3 The importance of HIF-1α and HIF-1 in mammalian immunity 6 1.2.4 HIF-1α and BRD 12 1.3 Conclusions and Research Proposal 16 1.3.1 Experimental hypothesis 16 1.3.2 Null Hypothesis 16 1.3.3 Research Objectives 16 2 Thesis overview 18 3 Optimization of hypoxic cell culture and harvest 19 3.1 The production of HIF-1α in hypoxic conditions in vivo 19 3.1.1 Hypoxic cell culture 19 3.1.2 Bovine leukocytes chosen for hypoxic culture 19 3.2 Materials, Methods and Results 20 3.2.1 Bovine leukocytes used in development of hypoxic-mimetic cell culture methods 21 3.2.2 Procedure for mononuclear cell isolation 21 3.2.3 Preparation of frozen test samples for culture 21 3.2.4 Hypoxic mimetic culture, stimulant and culture medium 22
x
3.2.5 Culture duration testing 24 3.2.6 Culture vessel testing, post-culture viability and HIF-1α degradation 24 3.3 Discussion 29 3.4 Conclusion – final cell culture and post-culture harvest protocol 33
4 Optimization of flow cytometry protocol for measuring HIF 1α expression in lymphocyte subtypes 34
4.1 Introduction 34 4.2 Materials, methods and results 34 4.2.1 Flow cytometer maintenance and flow cytometry controls 34 4.2.2 Cellular staining method for flow cytometry 35 4.2.3 Antibody titration 36 4.2.4 T lymphocyte CD3 antibody selection and titration 38 4.2.5 B lymphocyte CD20 antibody selection and titration 41 4.2.6 Monocyte CD14 antibody selection and titration 41 4.2.7 HIF-1α antibody selection 41 4.2.8 Triple staining of cultured cells for flow cytometry 45 4.3 Discussion 48 4.4 Conclusion - Final post-culture staining and flow protocol for 3-colour flow
cytometry of T lymphocytes, B lymphocytes and HIF-1α 49
5 Development of a protocol to detect HIF-1α expression in monocytes 50 5.1 Introduction 50 5.2 Materials, methods and results 51 5.2.1 HIF-1α antibody selection for ICC 51 5.2.2 ICC protocol for HIF-1α expression in monocytes 51 5.3 Discussion 53 5.4 Conclusion 54
6 Methodology application to assess HIF-1α expression as a biomarker of bovine respiratory disease 55
6.1 Introduction 55 6.2 Materials and methods 56 6.2.1 Study population and animal selection 56 6.2.2 Baseline data 57 6.2.3 Blood sample collection 57 6.2.4 Mononuclear cell isolation from experimental samples 58 6.2.5 Cell culture and post-culture harvest of experimental samples 58 6.2.6 Flow cytometry protocol 58 6.2.7 ICC for monocyte expression of HIF-1α 60 6.3 Results 61 6.3.1 Mononuclear cell isolation 61
xi
6.3.2 Comparison of baseline data 61 6.3.3 Flow cytometry results 64 6.3.4 ICC results 65 6.4 Discussion 69 6.5 Conclusion 71 7 General discussion and future directions 72 7.1 The importance of finding a biomarker for BRD 72 7.2 Methodology development issues 72 7.2.1 Hypoxic cell culture and harvest methods 72 7.2.2 Flow cytometry methods 73 7.2.3 ICC methods 75 7.3 Future directions for investigation of biomarkers for BRD 75 8 References 77 9 Appendices 90 9.1 Initial T lymphocyte CD3 antibody 90 9.2 CD20 antibody titration on flow cytometry 90 9.3 Initial CD14 antibody 91
xii
List of Figures
Figure 1.1: Diagram of (a) degradation of HIF-1α during normoxia and (b) HIF -1α stabilization during hypoxia (right) (From Nizet 2009) .................................................................................................................... 4 Figure 2.1: Thesis overview showing the interconnected relationships between various aspects of the methodologies developed in chapters 3, 4 and 5. Flow cytometry results influenced the cell culture methodology and lead to the development of ICC. All developed methodology is applied to experimental samples in chapter 6, followed by a discussion of developed methodology and future directions in chapter 7. ......................................................................................................................................................................... 18 Figure 3.1: A: SS/Pacific Blue dot plot obtained from Kaluza of all events within the Cells gate. Colour gradients denote areas of lower (blue) to higher event density (red), percentages are of all events within the Cells gate. An uncultured aliquot of cells stained with pacific blue-labelled CD14 antibody shows a distinct CD14 positive population to the right of a larger group of negative cells. B: An aliquot of cells cultured without stimulants shows a similar but depleted CD14 positive population, and generalised increased SS of all cells. C: An aliquot of cells cultured with CoCl2 and LPS shows generalised increased SS of all cells but the CD14 positive population cannot be distinctly visualised. ........................................................................ 27 Figure 3.2: A: ICC of uncultured mononuclear cells with anti-activated caspase-3. There is moderate staining intensity within both cells with abundant and scant cytoplasm (putative monocytes and lymphocytes, respectively). B: ICC of mononuclear cells cultured with stimulants before undergoing staining procedures for flow cytometry. There is marked staining intensity of cells with abundant cytoplasm (putative monocytes). ...................................................................................................................................................... 29 Figure 4.1: A bivariate scatter plot of the forward scatter (FS) and side scatter (SS) of the uncultured unstained test cells as visualised on a linear scale. Debris usually has low FS and SS. Lymphocytes have moderate FS but relatively low amounts of SS in contrast to monocytes. Events with high FS and SS are postulated to be aggregates of cells. ................................................................................................................ 36 Figure 4.2: A: Peripheral blood mononuclear cells - histogram of CD3 fluorescence of events with the cells gate. A peak of cells positive for CD3 can be visualised to the right of the larger negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD3 fluoresence versus SS, log scale. A population positive for CD3 has been gated by eye (black border). This positive population is distinct from and far to the right of the negative population. ................................................................................................ 39 Figure 4.3 A: Peripheral blood mononuclear cells - histogram of CD20 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD20 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD20 fluorescence versus SS, log scale. A population of cells positive for CD20 is visualised and gated by eye (black border). 41 Figure 4.4: Graph of the mean fluorescence of HIF-1α antibody at 1µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. ......................................................................... 43 Figure 4.5: Graph of the mean fluorescence of HIF-1α antibody at 2µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. ......................................................................... 44 Figure 4.6: A. Scatter plot of HIF-1α fluorescence versus SS, log scale. Unstimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. A single population with one central red (i.e. high density) focus is visible. B. Scatter plot of HIF-1α fluorescence versus SS, log scale. Stimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. The mean HIF-1α fluorescence was consistently lower in stimulated cells however stimulation caused a repeatable change in population shape with respect to HIF-1α and SS. A larger population with greater SS is now visible (upper large red foci) situated above a smaller population with lower SS (lower red foci). ...................................................................................................... 45 Figure 4.7: Scatter plot of CD3 fluorescence and CD20 fluorescence, triple-stained stimulated cells, log scale. B and T cell populations are gated (solid black borders) with percentages of gated cells that are CD3 positive, CD20 positive or positive for both phenotypic markers. .................................................................. 47
xiii
Figure 5.1: Post-culture in situ ICC of adherent monocytes. Secondary controls Wells 1 and 2 contained unstimulated and stimulated cells, respectively, and did not receive primary antibody. The lower wells 3 and 4 contained duplicate cell populations to 1 and 2 and were treated with primary HIF-1α antibody. .............. 52 Figure 6.1 A: An example of the forward scatter/side scatter plot obtained from Kaluza. Events included in cell population lie within the gate “Cells” comprising a solid black line. This gate excludes debris and aggregate cells. B: An example of a T lymphocyte fluorescent marker/B lymphocyte fluorescent marker scatter plot obtained from Kaluza, based on the events included in the “Cells” gate. “T cells” and “B cells” populations are both gated (solid black lines). C: An example of a histogram of HIF-1α fluorescence detected within the T cell gate on flow cytometry with Kaluza-generated summary statistics. .................................... 59 Figure 6.2: ICC of stimulated cells showing several monocytes with scattered intracytoplasmic HIF-1α expression (arrow) and a central, activated monocyte with strong intracytoplasmic and intranuclear HIF-1α signal. ............................................................................................................................................................... 66 Figure 6.3. A: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. B: Percentage difference in ICC HIF-αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. C: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. D: Percentage difference in ICC HIF-1αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. ........................................... 68
xiv
List of Tables
Table 4.1: Detectors used for detection of cell characteristics and the cytometer parameters for each detector output as set using single and then triple-stained cells. ................................................................................... 46 Table 6.1: Baseline data for cases and controls for both subsets ..................................................................... 62 Table 6.2: Descriptive statistics for the absolute difference in mean, median and standard deviation of HIF-1α in B cells and T cells between unstimulated and stimulated cultures for case group and control groups. . 65 Table 6.3: ICC results for case and control groups from both subsets. The mean, median and standard deviation of the percentage of cells displaying cytoplasmic or combined cytoplasmic and nuclear HIF-1α positive ICC staining in stimulated and unstimulated cell cultures, as well as for the percentage difference of these two types of HIF-1α positive staining between stimulated and unstimulated cultures. ......................... 66
xv
Abbreviations
APC allophycocyanin
ARD1 arrest defective-1 acetyltransferase
ATP adenosine triphosphate
BHV-1 bovine herpesvirus-1
BRD bovine respiratory disease
BRSV bovine respiratory syncytial virus
BVDV bovine viral diarrhoea virus
CO2 carbon dioxide
CoCl2 cobalt chloride
DFO desferrioxamine
ELISA enzyme-linked immunosorbent assay
EPO erythropoietin
FIH factor-inhibiting HIF
FITC fluorescein isothiocyanate
FS forward scatter
HHV-8 human herpesvirus-8
HIF-1α human immunodeficiency virus-1
HIV-1 hypoxia-inducible factor-1α
HRE hypoxic response element
ICC immunocytochemistry
IFN interferon
IHC immunohistochemistry
IL interleukin
iNOS inducible nitric oxide synthase
LANA latency associated nuclear antigen
LMD list mode data file
LPS lipopolysaccharide
mRNA messenger ribonucleic acid
NaN3 sodium azide
NF-κB nuclear factor [kappa] B
NO nitric oxide
ODDD oxygen-dependent degradation domain
PBS phosphate-buffered saline
xvi
PE phycoerytherin
PHD prolyl hydroxylase domain-containing protein
PI-3 parainfluenza-3 virus
pVHL von Hippel-Lindau protein
ROS reactive oxygen species
RSV respiratory syncytial virus
SS side scatter
TLR toll-like receptor
TNF-α tumour necrosis factor-α
UQ The University of Queensland
VEGF vascular endothelial growth factor
VSV vesicular stomatitis virus
1
1 Introduction, Literature Review, Conclusions and Research Proposal
1.1 Introduction
Previously an experimental model of bovine respiratory disease (BRD) has shown that animals are
most susceptible to secondary bacterial infections at the height of the virus-induced inflammatory
response and will develop severe, even fatal, fibrinous bronchopneumonia (Bielefeldt-Ohmann et
al., 2008). The significance of hypoxia-inducible factor-1α (HIF-1α) in the development and
coordination of an immune response to hypoxic microenvironments caused by infectious agents
makes HIF-1α an attractive biomarker for BRD susceptibility. There is evidence in the current
literature to suggest that HIF-1α is an important regulator of innate immunity, that it is induced
during immune stress responses independent of hypoxia, and that both too vigorous and too little of
a HIF-1α response could be detrimental in the probability of overcoming a sophisticated infectious
challenge such as that inherent in the BRD complex.
To investigate the possibility of the cytokine/reactive oxygen species (ROS)- induced positive
feedback loop of HIF-1α induction contributing to the development of clinical BRD, this project
aimed to develop methods for the stabilization and measurement of HIF-1α in bovine immune cells.
Once developed, a further aim was to apply these methods experimentally to quantify leukocyte
HIF-1α responsiveness of a group of feedlot cattle with a history of clinical BRD, and compare with
leukocyte HIF-1α responsiveness of feedlot cattle without clinical respiratory disease.
The discovery of a relationship between HIF-1α expression and the development of clinical BRD
may help to better understand why certain animals develop clinical BRD while other animals do
not, and may establish HIF-1α as a potential biomarker to be used in the development of breeding
programs aimed at reducing BRD susceptibility in feedlot cattle.
1.2 Literature Review
Oxygen is crucial in the survival of most eukaryotic cells due to the oxygen requirements of
important cellular processes, most notably the oxidation of nutrients to produce adenosine
triphosphate (ATP) by oxidative phosphorylation (Lipmann, 1941, Kalckar, 1941). Tissue hypoxia
is defined as low oxygen within a tissue, organ or cell compared to the oxygen level normally
demanded within that tissue at sea level ([Anon], 1973). The rapid cellular and systemic response
mechanisms to adapt to hypoxic conditions are coordinated by Hypoxia-inducible factor-1 (HIF-1),
a heterodimeric helix-loop-helix transcription factor that is considered an important regulator of
cellular homeostasis, cellular adaptation to stress caused by hypoxia, and innate immunity (Nizet
and Johnson, 2009).
Hypoxia-inducible factor-1 was first identified when the molecule was found to be associated with
marked induction of erythropoietin (EPO) gene transcription that occurs under hypoxic conditions
2
(Goldberg et al., 1988, Semenza et al., 1991, Semenza and Wang, 1992). Subsequent study of HIF-
1 recognized its existence in all mammalian cells, not just those involved in EPO production (Wang
and Semenza, 1993c, Maxwell et al., 1993).
HIF-1 is composed of 2 subunit proteins, a hypoxia-inducible factor-1α (HIF-1α) subunit (Wang
and Semenza, 1993a, Wang and Semenza, 1995) which is oxygen-sensitive (Wang et al., 1995,
Jiang et al., 1996) and a constitutively expressed hypoxia-inducible factor-1β (HIF-ß) subunit
(Wang and Semenza, 1995).
When stabilised by hypoxia or other factors, HIF-1α translocates into the nucleus where it binds to
HIF-1ß (Wang et al., 1995). The resultant heterodimer HIF-1 binds to 50-base pair cis-acting
sequence termed a hypoxic response element (HRE) of target gene enhancer and promotor
sequences, and activates the expression of these target genes (Semenza et al., 1991, Semenza and
Wang, 1992, Wang et al., 1995). HIF-1 is known to regulate the expression of more than 100 target
genes (Nizet and Johnson, 2009) involved in systemic and cellular adaptive responses to hypoxia.
HIF-1 regulates genes important for erythropoiesis and iron metabolism (Semenza et al., 1991),
angiogenesis (Hertig, 1935), cell proliferation and survival, as well as responses supporting innate
immunity within the characteristic hypoxic microenvironment of inflamed tissues (Nizet and
Johnson, 2009). Many pro-inflammatory cytokines and ROS can stabilize HIF-1α even under
normoxic conditions, with subsequent HIF-1 induction of proteins that promote inflammation in a
seemingly positive feedback loop of inflammation regulation.
More recently, two other closely related forms of the HIF-α subunit have been identified and
characterized as HIF-2α (Tian et al., 1997) and HIF-3α (Gu et al., 1998). Both HIF-2α and HIF-3α
can also complex with HIF-1ß (Tian et al., 1997, Gu et al., 1998). HIF-1α and HIF-2α have a
similar domain structure, are inducible by hypoxia and bind HRE of target genes (Tian et al., 1997,
Wiesener et al., 1998, O'Rourke et al., 1999). HIF-2α expression has been found predominantly in
vascular structures, lung, endothelium and carotid body (Ema et al., 1997, Tian et al., 1997, Tian et
al., 1998), in contrast to the ubiquitous HIF-1α. HIF-3α is expressed in a variety of tissues (Gu et
al., 1998) but lacks the C-terminal activation domain required for co-activator binding, so cannot
recruit co-transcriptional regulators and transcriptional machinery to gene targets (Gu et al., 1998).
There is evidence to suggest that HIF-3α splice variants function as inhibitors of HIF-1 induced
gene expression, possibly to protect against hypoxic damage (Makino et al., 2001, Forristal et al.,
2010, Augstein et al., 2011).
3
1.2.1 HIF-1α regulation during normoxia, development and homeostasis
An appreciation for the metabolism of HIF-1α during normoxia is necessary to better understand
the currently known mechanisms of HIF-1α in the hypoxic microenvironment.
1.2.1.1 HIF-1α during normoxic conditions
Normoxia is currently defined for tissue culture as being the equivalent of atmospheric oxygen
pressure at sea level or as physiological oxygenation for well-vascularised and perfused tissue
(Nizet and Johnson, 2009). Reported oxygen pressure values for healthy tissue vary from 10-30
mm Hg (Intaglietta et al., 1996, Braun et al., 2001). During normoxic conditions HIF-1α is a short-
lived protein with a half-life of less than five minutes (Huang et al., 1996). In the absence of
stabilization, it is removed via ubiquitination followed by proteasomal degradation (Salceda and
Caro, 1997, Huang et al., 1998, Kallio et al., 1999).
The ubiquitination of HIF-1α is performed by an E3 ubiquitin ligase complex that also contains the
von Hippel-Lindau (pVHL) tumor suppressor protein, which acts as the recognition point for HIF-
1α during binding for ubiquitination (Maxwell et al., 1999). To promote VHL interaction with HIF-
1α, there must be hydroxylation of prolyl residues(Ivan et al., 2001, Jaakkola et al., 2001, Yu et al.,
2001) within identified oxygen-dependent degradation domains (ODDDs)(Huang et al., 1998) of
HIF-1α (Huang et al., 1998). Hydroxylation of the prolyl residues is carried out by three prolyl
hydroxylase domain-containing protein (PHD) enzymes, and all PHD enzymes require oxygen and
iron to complete this activity (Epstein et al., 2001, Jaakkola et al., 2001, Bruick and McKnight,
2001). This implicates PHD enzymes as important oxygen sensors within the HIF-1α pathway
(Schofield and Ratcliffe, 2004, Walshe and D'Amore, 2008). HIF-1α is subsequently tagged with
polyubiquitin and undergoes proteosomal proteolysis (Salceda and Caro, 1997, Kallio et al., 1999,
Masson et al., 2001) (see Figure 1.1).
4
Figure 1.1: Diagram of (a) degradation of HIF-1α during normoxia and (b) HIF -1α stabilization
during hypoxia (right) (From Nizet 2009) In addition to PHD enzyme activity, there are further oxygen-sensing mechanisms in the HIF-1α
pathway. HIF-1α contains a carboxy-terminal activation domain (Jiang et al., 1997) in which there
is hydroxylation of an asparagine (Lando et al., 2002b) by an asparaginyl hydroxylase called factor-
inhibiting HIF (FIH) (Lando et al., 2002a) during normoxia. This hydroxylation prevents binding of
the p300/cAMP- response element-binding protein, a co-activator which, during hypoxia, binds to
HIF-1α and enhances its transcriptional activity (Lando et al., 2002b)(see Figure 1.1). FIH also
interacts with VHL to form a ternary complex with HIF-1α in which pHVL inhibits HIF-1α
transactivation function (Mahon et al., 2001). FIH also has an absolute requirement for oxygen
(Koivunen et al., 2004, Ehrismann et al., 2007) and so is considered a second oxygen sensing
mechanism in the inhibitory pathway (Walshe and D'Amore, 2008).
A study by Jeong et al using HEK293 cells showed HIF-1α is also further negatively regulated by
acetylation performed by arrest defective 1acetyltransferase (ARD1) which binds in the ODDD and
enhances pVHL binding to HIF-1α (Jeong et al., 2002). This study concluded ARD1 functions as a
negative regulator of HIF-1α during normoxia, as there were decreased levels of ARD1 mRNA and
5
decreased acetylated HIF-1α during hypoxia (Jeong et al., 2002). However, another study used
several cell types to show that neither the mRNA nor the protein levels of ARD1 are regulated by
short term or long term hypoxia (Bilton et al., 2005), and that overexpression or silencing of the
ARD1 gene had no effect on HIF-1α stability in normoxic or hypoxic conditions (Bilton et al.,
2005). The relationship between normoxia, hypoxia and HIF-1α acetylation by ARD1 remains
unclear.
1.2.2 HIF-1α during hypoxic conditions
Several important pathological processes involve or induce hypoxic conditions in tissue including
inflammation and neoplasia (Najafipour and Ferrell, 1995, Helmlinger et al., 1997). The hypoxia
within acutely inflamed tissue is due to decreased perfusion from microvascular injury and
thrombosis, clogging of vessels with recruited inflammatory cells, and the metabolic activities of
recruited inflammatory cells and any infectious pathogen (Simmen et al., 1994, Saadi et al., 2002,
Sitkovsky and Lukashev, 2005, Kempf et al., 2005, Nizet and Johnson, 2009). Within the context of
the hypoxic inflammatory microenvironment and the immune response, HIF-1α is established as a
molecule of interest to this project about BRD susceptibility.
Under hypoxic conditions HIF-1α ubiquitination and degradation is decreased (Kallio et al., 1999)
due to reduced activity of oxygen-dependent PHD enzymes (Epstein et al., 2001, Jaakkola et al.,
2001). PHD are also actively degraded by the E3 ubiquitin ligase (Nakayama et al., 2004).
HIF-1α accumulates in the cytoplasm and translocates to the nucleus (Kallio et al., 1999) where it
binds to HIF-1β (Wang et al., 1995) and forms activated HIF-1 for transcription (Huang et al.,
1996, Kallio et al., 1997)}. Experiments in vitro and in vivo have also shown increased
transcription of HIF-1α mRNA in response to hypoxia (Wiener et al., 1996, Yu et al., 1998,
Semenza, 2000, Belaiba et al., 2007) and have replaced an original concept of solely
posttranslational hypoxic HIF-1α regulation.
1.2.2.1 Transcription of HIF-1 target genes in response to hypoxia
The HIF-1 transcription factor regulates the expression of more than 100 target genes (Nizet and
Johnson, 2009). In response to hypoxia, HIF-1 activates genes important for erythropoiesis and iron
metabolism, the products of which enhance tissue oxygenation. Gene products include
erythropoietin (EPO) for new erythrocyte formation (Semenza et al., 1991), transferrin that
transports ferric iron into cells (Rolfs et al., 1997) and ceruloplasmin, which oxidises ferrous iron to
ferric iron (Mukhopadhyay et al., 2000).
6
1.2.2.1.1 Angiogenesis
HIF-1 also promotes angiogenesis, defined as the formation of new vessels from existing
vasculature (Hertig, 1935) to increase vascularisation in response to hypoxia. The important
endothelial mitogen vascular endothelial growth factor (VEGF) (Leung et al., 1989, Keck et al.,
1989) is produced following HIF-1 gene transcription regulation (Forsythe et al., 1996). Vascular
endothelial growth factor causes recruitment to and proliferation of vascular endothelial cells within
hypoxic or avascular tissues, including during the development of tumours, in diseases with an
angiogenic pathogenesis and wound healing (Shweiki et al., 1992, Folkman and Shing, 1992,
Brown et al., 1992, Adamis et al., 1994).
There is evidence of inhibition of HIF-1 angiogenesis in specific tissues. The cornea is subject to
continuous hypoxia during sleep but vessel proliferation does not occur due to a splice variant of
the HIF-3α called inhibitory Per/Arnt/Sim protein, which inhibits HIF-1 controlled gene expression
(Makino et al., 2001). Other examples are vascular endothelium and smooth muscle (Augstein et
al., 2011), as well as the Purkinje cell layer of the cerebellum (Makino et al., 2001).
1.2.2.1.2 Cell proliferation and survival
HIF-1 can control cell proliferation and survival. Hypoxia and HIF-1 induce growth factors such as
insulin-like growth factor-2 and transforming growth factor-α that activate signal transduction
pathways leading to cell-specific proliferation and survival, as well as stimulating HIF-1α
expression. These events implicate HIF-1α in an autocrine-signalling pathway that can be crucial
for neoplastic disease (Semenza, 2003). Hypoxia and HIF-1 has also been shown to induce
apoptosis via several mechanisms (Greijer and van der Wall, 2004), as well as inhibit neutrophil
apoptosis (Walmsley et al., 2005).
1.2.3 The importance of HIF-1α and HIF-1 in mammalian immunity
The transcription of certain other target genes by the HIF-1 transcription factor has an impact on
immune function. This impact can come from the hypoxic stabilisation of HIF-1α, but also through
the stabilisation of HIF-1α by non-hypoxic pathways such as those initiated by pathogens and
proinflammatory stimuli, which can regulate HIF-1 at transcriptional, translational and
posttranslational levels.
1.2.3.1 HIF-1α and cellular energy production in immune cells
The HIF-1 transcription factor is involved in transcribing genes for glycolytic energy generation
inimmune cells. Hypoxia is a characteristic of inflamed tissue (Sitkovsky and Lukashev, 2005) due
7
to the rapid consumption of oxygen while blood vessels are clogged with phagocytes (Sitkovsky
and Lukashev, 2005). During hypoxic conditions, cellular utilisation of oxygen-dependent cellular
oxidative phosphorylation is decreased in favour of anaerobic glycolysis. This adaption was first
noted by Pasteur in the 19th century and termed the Pasteur effect. With an available glucose source,
glycolysis allows optimized cellular energy production, function and survival in low oxygen
environments, and the HIF-1 transcription factor has been shown as a necessary mediator in
performing this metabolic switch in mammalian cells (Seagroves et al., 2001). HIF-1 up-regulates
the expression of many of the glycolytic enzymes and glucose transporters allowing increased
glycolytic pathway activity to maintain ATP levels (Semenza et al., 1994, O'Rourke et al., 1996,
Iyer et al., 1998, Chen et al., 2001).
Myeloid cells such as neutrophils and macrophages are amongst the first responders to changes in
tissue integrity or entry of pathogenic organisms (Nizet and Johnson, 2009), constituting important
components of the innate immune inflammatory response to invading microbial organisms and
foreign bodies. These cells release many different anitimicrobial molecules and pro-inflammatory
mediators, localise and phagocytise pathogens and tissue debris; fight local infection and prevent
systemic spread (Nizet and Johnson, 2009). It has been shown that when HIF-1α is absent, the ATP
pool in macrophages and neutrophils is drastically reduced to 15-40% of wild-type levels with
concurrent impairment of myeloid immune function, even under normoxic culture conditions
(Cramer et al., 2003). A number of other studies have linked ATP levels in myeloid cells with the
capacity for inflammation (Kittlick, 1986). Such ATP depletion in the absence of HIF-1α suggests
that myeloid cells may rely on glycolysis as a main mechanism of energy production.
1.2.3.2 HIF-1α and immune cell function
The importance of HIF-1α in myeloid cell function has been thoroughly demonstrated in the
aforementioned, key study by Cramer et al (2003). This study employed conditional gene targeting
of myeloid cells in knock-out mice (silencing HIF-1α, its negative regulator pVHL and a known
downstream target, VEGF). These knock-out mice demonstrated no abnormalities under normal
conditions, however showed marked aberrations in experimental models of acute inflammation. The
knock-out mice did not develop severe joint swelling and cartilage destruction in a collagen induced
arthritis and showed no cutaneous redness or oedema, cardinal signs of inflammation, following
application of a chemical irritant to the skin, suggesting an impaired acute inflammatory response.
(Cramer et al., 2003). This study also demonstrated that loss of HIF-1α impaired key macrophage
cell activities such as aggregation, invasion and motility, and that HIF-1α is essential for myeloid
cell infiltration and activation in vivo through a mechanism independent of VEGF (Cramer et al.,
8
2003). There have been studies with conflicting results however. A 2005 study by Peyssonnaux
found that HIF-1α -null murine neutrophils were recruited to sites of microbial infection as
efficiently as wild type neutrophils. It is possible the presence of a bacterial stimulus in the latter
study and the use of different myeloid cell types may factor into the difference in myeloid cell
recruitment after loss of HIF-1α.
HIF-1α exerts its mechanism of action in acute inflammation through induction of ß2 integrin
adhesion receptors whose expression coordinates the recruitment of myeloid cells to sites of
inflammation (Kong et al., 2004), and HIF-1α activity delays neutrophil apoptosis to sustain a local
immune response (Walmsley et al., 2005). Overexpression of HIF-1α in macrophages has shown
surprising increases in phagocytosis, with macrophages better able to kill bacteria under hypoxic
conditions than normoxic conditions (Anand et al., 2007).
In contrast, HIF-1α appears to dampen more chronic inflammatory processes through negative
regulation of lymphocyte adaptive immune functions. Chimeric mice with HIF-1α deficient T and
B cells showed greatly increased autoimmune tissue damage (Kojima et al., 2002) indicating a
possible tissue-protecting role for HIF-1α. Increased production of HIF-1α in T cells induces a shift
from type 1 helper T-cell to type 2 helper T-cell regulated immune responses, which then inhibit
Th1-mediated microbicidal functions (Eltzschig and Carmeliet, 2011). HIF-1α also stimulates
proliferation of regulatory T-cells which are inhibitory T-cells (Ben-Shoshan et al., 2008). HIF-1α
transcriptional regulation negatively modulates proinflammatory cytokine production by CD4+ and
CD8+ T cells, as activated T cells with deletion of the Hif1a gene released greater levels of tumour
necrosis factor-α (TNF-α) and interferon- γ (IFN-γ) than wild type T cells (Lukashev et al., 2006).
Another study supported these findings by using a murine model of sepsis to show that activated
Hif1a gene-deficient T cells underwent increased cell proliferation and enhanced secretion of
interleukin-2 (IL-2), IFN-γ, TNFα and interleukin-6 (IL-6), blocking bacterial proliferation and
improving survival of mice (Thiel et al., 2007).
These immunosuppressive actions are thought to be the result of hypoxia-induced accumulation of
extracellular adenosine due to hypoxic regulation of enzymes involved in adenosine metabolism
(Synnestvedt et al., 2002). Adenosine accumulation is a rapid response to hypoxia, and subsequent
signaling via adenosine receptors results in accumulation of immunosuppressive intracellular cylic
AMP (cAMP) and thus inhibition of T-cell receptor activated T-cells (Sitkovsky, 2009).
Strict regulation of HIF-1 function in T cells is important to prevent cell death and suggests T cells
play a role in hypoxia-mediated and HIF-1 dependent cell death in hypoxic tissue (Biju et al.,
2004).
9
1.2.3.3 HIF-1α induction of pro-inflammatory molecules and further regulation of the HIF-1
pathway due to inflammation and pathogens
1.2.3.3.1 HIF-1α and inflammatory mediators
During the innate immune response, HIF-1α promotes the myeloid expression of antimicrobial
peptides, granule proteases, VEGF and pro-inflammatory cytokines such as TNF-α, interleukin-1α,
interleukin-1ß (IL-1ß) and interleukin-12. In addition, it upregulates toll-like receptor (TLR)
expression and activates production of nitric oxide (NO) by inducible nitric oxide synthase (iNOS)
(Peyssonnaux et al., 2005, Peyssonnaux et al., 2007). Increased VEGF and the released
proinflammatory cytokines recruit and activate more immune effector cells (Zinkernagel et al.,
2007).
TNF-α and IL-1ß are proinflammatory cytokines that have been shown to increase the activity of
HIF-1α via increased mRNA translation, as seen in tumour cells (Hellwig-Burgel et al., 1999), renal
proximal tubular cells (Zhou et al., 2003), epithelial cells(Sandau et al., 2001b) and vascular smooth
muscle cells (Richard et al., 2000). Many of these cell types also express pro-inflammatory
cytokines like IL-1β and TNF-α, thus creating an autocrine feedback loop (Frede et al., 2007). TNF-
α also increases HIF-1α protein expression in neutrophils and macrophages participating in wound
healing (Albina et al., 2001).
TNF- α induces HIF-1α via NO and ROS pathways (Haddad and Land, 2001, Brune and Zhou,
2007) and can also post-translationally induce HIF-α activity in a nuclear factor [kappa] B (NF-
κB)-dependent manner (Jung et al., 2003, van Uden et al., 2008). In addition, platelet-derived
growth factor (PDGF) is increasingly produced in response to hypoxia and can itself stimulate HIF-
1 dependent gene expression in an autocrine-paracrine way (Zhang et al., 2003).
NO is a microbicidal molecule produced by activated macrophages and granulocytes during
inflammation via the inducible NO synthetase (iNOS) (Thomas et al., 2008). It has been long
known that HIF-1 activates the expression of iNOS to produce NO (Melillo et al., 1997, Jung et al.,
2000, Hu et al., 2002) however more recent work indicates a contrast between classic pro-
inflammatory macrophages (M1macrophages) in which Th1 cytokines induce HIF-1α, ultimately
upregulating NO, to that of alternatively activated macrophages (M2 macrophages) in which Th2
cytokines induce HIF-2α and ultimately suppress NO (Takeda et al., 2010). In the context of
bacterial infection, iNOS expression in wild-type macrophages is principally regulated by HIF-1α
(Peyssonnaux et al., 2005). It has also been demonstrated that in the same circumstances, inhibition
of iNOS blocks the expected generation of increased levels of HIF-1α, indicating an amplification
10
loop where NO plays an important role in further stabilization of HIF-1α in macrophages (Sandau
et al., 2001a, Peyssonnaux et al., 2005).
NO has been shown to stabilize HIF-1α by redistributing cellular oxygen, inhibiting PHD activity
and preventing HIF-1α ubiquination during normoxia (Sandau et al., 2001a, Hagen et al., 2003,
Peyssonnaux et al., 2005, Brune and Zhou, 2007). However, under hypoxic conditions it has been
shown that NO can also have an inhibitory affect on HIF-1α stabilization and activity by
reactivation of PHD (Callapina et al., 2005, Zhou and Brune, 2006).
During acute inflammation, myeloid cells produce a burst of ROS that is critical for the bactericidal
action of phagocytes, but may result in damage to host tissues if ROS formation proceeds
uncontrolled (Dehne and Brune, 2009). This oxidative burst occurs equally well in wild-type and
HIF-1α deficient macrophages (Peyssonnaux et al., 2005). During hypoxia, subsequent ROS
generated within mitochondria stabilise HIF-1α through the same mechanism of inhibition of PHD
enzyme function (Chandel et al., 2000, Guzy et al., 2005).
Ascorbate is an antioxidant that keeps PHD enzymes in a reduced form and therefore functioning to
tag HIF-1α for degradation. An ascorbate deficiency has been shown to limit PHD function in
human tumours, resulting in increased HIF-1α levels (Knowles et al., 2003). Other work involving
deletion of a transcriptional regulator responsible for increasing antioxidant levels resulted in
increased HIF-1α levels due to increased free ROS, which in turn cause direct oxidation of Fe (II) at
the catalytic site of PHD, which ultimately limits PHD enzyme activity (Gerald et al., 2004).
In addition, a non-hypoxic pathway mediating the effect of TNF-α in regulating the stabilization,
translocation and activation of HIF-1α in a ROS-sensitive mechanism has also been reported
(Haddad, 2002). Taken together, these findings indicate that an increase in ROS during acute
inflammation and/or hypoxia may result in further HIF-1α stabilization.
One may speculate that HIF-1α may be situated at the centre of an amplification loop mechanism
for innate immune activation: stimulation of HIF-1α by hypoxia and bacteria induces the production
of NO and TNF-α, functioning not only to generate inflammation and control bacterial infection but
also to further stabilize HIF-1α in recruited myeloid cells (Peyssonnaux et al., 2005). To complete
the loop, increases in ROS/nitrogen species in the inflammatory process have been shown to
stabilize HIF-1α (Pouyssegur and Mechta-Grigoriou, 2006). Transgenic mice expressing
constitutively active HIF-1α in the epidermis showed increased VEGF and capillary formation, but
no evidence of vascular leakage or inflammation (Elson et al., 2001), indicating that the effects of
increased constitutive HIF-1α (instead as part of a positive feedback loop) within non-inflamed
11
epidermis may be controlled by the physiochemical microenvironment. The effects of increased
HIF-1α within an inflammatory process as induced by hypoxia and inflammatory mediators may be
different.
1.2.3.3.2 HIF-1α and the immune response to viral pathogens
The relationship between oxygen tension and viral replication is more complex than previously
thought, with recent studies collectively highlighting the major but variable role that HIF-1α plays
in modulating viral infections in both normoxic and hypoxic conditions (Morinet et al., 2013).
For example hepatitis C virus replication is enhanced under hypoxic conditions, although the
mechanism appears to be HIF-1α –independent (Vassilaki et al., 2013). The nuclear antigens of
Epstein-Barr virus bind to PHD, inhibiting HIF-1α degredation (Darekar et al., 2012). The
respiratory tract pathogen respiratory syncytial virus (RSV) induces HIF-1α in human bronchial
cells via a NO-dependent pathway, and the resultant increased VEGF also enhances monolayer
permeability, a pathway which may contribute to airway oedema of acute RSV infection (Kilani et
al., 2004). Hypoxia-independent stabilization of HIF-1α in vitro and in vivo has also been observed
in another study of RSV (Haeberle et al., 2008), further supporting HIF-1α stabilisation occurring
as a result of a viral infection.
Human immunodeficiency virus-1 (HIV-1) viral protein vpr induces ROS, which contribute to HIF-
1α expression, with the subsequent interesting mechanism of HIV-1 gene transcription brought
about by HIF-1α forming a heterodimer with vpr (Deshmane et al., 2009).
Hypoxia induces the lytic replication cycle of human herpesvirus-8 (HHV-8, also known as
Kaposi’s sarcoma-associated herpesvirus) in vitro via a functional HRE (Davis et al., 2001, Haque
et al., 2003). HHV-8 also contains a latency associated nuclear antigen (LANA) that enhances the
transcriptional level of HIF-1α mRNA but also induces the lytic replication cycle during hypoxia
(Dalton-Griffin et al., 2009). LANA also stabilises HIF-1α during normoxia by targeting pVHL
(Cai et al., 2007). However, a more complex relationship between HHV-8 and HIF-1α has emerged
with evidence that the HHV-8 viral protein ORF34, previously transcribed via HIF-1, was then able
to induce HIF-1α degradation via the proteasome (Haque and Kousoulas, 2013).
Reduction of viral replication and protein expression of an adenovirus has also been reported
(Pipiya et al., 2005). The replication of vesicular stomatitis virus (VSV) is inhibited in vitro by HIF-
1α (Hwang et al., 2006). Renal carcinoma cells with constitutive HIF-1 activity during nomoxia
demonstrate enhanced cellular resistance to the cytoltic effect of VSV when compared to wild-type
with the expected low HIF-1α activity (Hwang et al., 2006).
12
1.2.3.3.3 HIF-1α and the immune response to bacterial pathogens
Bacteria lipopolysaccharide (LPS) is a component of the cell wall of gram-negative bacteria (Quinn
et al., 2011) and has been shown to induce stabilisation of HIF-1α mRNA in resident lung
macrophages under normoxic conditions (Blouin et al., 2004, Eitam et al., 2010).
HIF-1α expression increased four-fold in wild type macrophages at normoxia following exposure to
gram-positive or gram-negative bacteria with resulting HIF-1α transcriptional gene activity levels
similar to or greater than levels achieved with hypoxia (Peyssonnaux et al., 2005). In this study by
Peyssonnaux et al, mice lacking HIF-1α in their myeloid lineage showed decreased myeloid
bactericidal activity, with 2-fold decrease in intracellular killing under normoxia and 3-fold
decrease intracellular killing of a Group A Streptococcus under hypoxia, with macrophage killing
of the gram negative P. aeruginosa likewise impaired. Conversely, activation of the HIF-1α
pathway by deletion of the VHL protein or by pharmacologic inducers improved myeloid cell
bactericidal activity (Peyssonnaux et al., 2005).
1.2.4 HIF-1α and BRD
1.2.4.1 HIF-1α and bovine pulmonary immune responses
Pulmonary diseases associated with intra-alveolar exudates, edematous septal thickening, infection
and inflammation can often result in decreased mucosal oxygenation and ultimately blood and
tissue hypoxia (Schaible et al., 2010). As with other tissues, hypoxia and immune signaling
pathways in the lung are closely connected and so subsequent activation of various hypoxia
sensitive pathways may contribute to the development of infectious and inflammatory lung disease
(Schaible et al., 2010) hypoxic induction of HIF-1 in ferrets has been shown to occur in many
pulmonary cell types including pulmonary arterial endothelium, smooth muscle, bronchial
epithelium, alveolar epithelium alveolar macrophages and microvascular endothelium (Yu et al.,
1998).
In the bovine respiratory tract there are similar innate immune responses as identified in other
mammalian species, and given the association of hypoxia with lung disease it is reasonable to
expect that HIF-1α plays a similar role in regulating these responses. Activation of HIF-1α in
macrophages and neutrophils in the bovine lung likely enhances the innate immunological functions
of these cells via increased bacterial killing, phagocytosis, cell survival and cytokine production
(Cramer et al., 2003, Schaible et al., 2010). Alveolar epithelial cells, resident alveolar macrophages
and dendritic cells recognize various microbial components via pattern recognition receptors (PRR)
with resultant production of proinflamamtory cytokines via the NF-kB pathway, directly affecting
expression of genes involved in innate immunity as well as indirectly causing amplification of the
HIF pathway (Schaible et al., 2010).
13
The anti-inflammatory actions of HIF, brought on by concurrent local hypoxia and accumulation of
adenosine that negatively regulate T-cells (see section 1.2.3.2) also participate in negative
regulation of the pulmonary innate immune response in vivo, as demonstrated by a study of
neutrophil-mediated lung inflammation in mice (Thiel et al., 2005). In addition, HIF-1α is involved
in the regulation of several barrier-protective genes that are thought to facilitate a protective role of
HIF-1α activation in a model of mucosal inflammation using murine colitis (Robinson et al., 2008)
a role which, given similar structural features and a common embryonic origin of the foregut
(Gilbert, 2010), may also apply to respiratory mucosal inflammation.
1.2.4.2 HIF-1α as a potential biomarker for development of clinical BRD
Of diseases in beef cattle, respiratory disease has the greatest economic impact. Economic losses
occur due to morbidity, mortality, treatment and prevention costs and loss of production (Griffin,
1997). Due to the high incidence of pneumonia in cattle, the ubiquity of the respiratory pathogens,
and the various anatomical factors predisposing cattle to respiratory infection, investigations are
focused on ways to enhance effective and non-injurious immune responses to these pathogens
(Ackermann et al., 2010b) in addition to combating environmental and managerial issues that may
predispose cattle to respiratory disease.
BRD as a specific entity is a multi-factorial, inflammatory respiratory disease complex which poses
a significant problem for the beef industry. The disease complex involves both viral and bacterial
pathogens and invokes both innate and adaptive immune responses as well as the development of
clinical pneumonia in certain feedlot cattle. The main viruses of this disease complex are bovine
herpesvirus-1 (BHV-1), bovine respiratory syncytial virus (BRSV), bovine viral diarrhoea virus
(BVDV) and parainfluenza-3 virus (PI-3)(Srikumaran et al., 2007) and the prominent bacterial
pathogens are Mannheimia haemolytica, Histophilus somni and Mycoplasma bovis (Srikumaran et
al., 2007). BRD involves complex interactions amongst these viral and bacterial pathogens with
viral infection predisposing cattle to secondary bacterial pneumonia, resulting in intense pulmonary
inflammation (Hodgson et al., 2005). An experimental model of this viral-bacterial synergy has
been described which utilizes primary BHV-1 respiratory challenge followed by aerosol challenge
with M. haemolytica (Bielefeldt Ohmann et al., 1991).
BHV-1, PI-3, BRSV and BVDV can infect lung epithelial cells, and can signal through TLR 3, 7, 8
and 9 (Ackermann et al., 2010b). Following BHV-1 infection, innate immune responses include the
antiviral action of and the recruitment of lymphoid cells, macrophages, neutrophils and natural
killer cells to the site of infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Bovine
14
bronchial cells that are infected with BHV-1 will release cytokines, which subsequently promote
neutrophil recruitment to the lung (Rivera-Rivas et al., 2009). In vitro, TNF-α, IL-1β and IFN-γ are
secreted in response to BHV-1 infection, causing the increased expression of the β2 integrin on
bovine alveolar macrophages and neutrophils (Czuprynski, 2009). The β2 integrin is the receptor
for the leukotoxin (LKT) of M. haemolytica, and LKT binding causes neutrophils and macrophages
to release cytokines and reactive oxygen intermediates that might further exacerbate the
inflammatory response in the respiratory tract (Czuprynski, 2009).
Although BHV-1 is a cytopathic virus, it is not this effect, but rather the virus-induced release of
cytokines and their subsequent effects, which constitute the basic mechanisms in the inflammatory
process and lung pathology (Bielefeldt Ohmann et al., 1991). Although vigorous inflammatory
response initiated by the virus infection can be beneficial in host defense, it also establishes
conditions that predispose to poorly regulated inflammation in the lung. Activation of the HIF-1a
pathway by microenvironmental hypoxia in myeloid cells will enhance the innate immune response
to microbial pathogens via cytokine production, microbial killing, increased phagocytosis, NO
production and TLR expression as already reviewed (Schaible et al., 2010). However this enhanced
inflammatory response may also contribute to tissue damage and further lung disease.
The increased production of pro-inflammatory cytokines that occurs during a primary BHV-1
infection is of significance to this project. It is in the context of this vigorous inflammatory response
that is characteristic of BRD viral-bacterial synergy, that the possibility of HIF-1α as a biomarker is
raised. The previously described involvement of HIF-1α in a veritable amplification loop of
cytokine-ROS innate inflammation (see section 1.2.3.3) may make HIF-1α an interesting molecule
to study in the context of the cytokine-involved inflammatory process culminating in BRD.
Work on H5N1 human influenza virus in cynomolgus macaques has found a pronounced
upregulation of HIF-1α expression in peripheral blood leukocytes and tissue macrophages in the
context of severe pneumonia (Tolnay et al., 2010), along with profound and sustained upregulation
of mRNA for IFNs, IL-1, IL-6 and TNF-α proteins (Baskin et al., 2009). Alternatively a lack of
HIF-1α response could also increase susceptibility to BRD. A sluggish HIF-1α response could
decrease the hypoxia-triggered HIF-1α adenosinergic suppression of proinflammatory cytokines in
CD4+ and CD8+ lymphocytes, an important immunosuppressive mechanism which downregulates
the immune response to protect local tissue and maximize the anti-pathogen cellular response
during pathogen encounter (Sitkovsky and Lukashev, 2005, Zinkernagel et al., 2007, Sitkovsky,
2009). However, the variable interactions of HIF-1α and viruses during the establishment of viral
15
infections (see section 1.2.3.3.2) suggests that many other mechanisms or indeed complex
combination of mechanisms may be at play.
1.2.4.3 Previous BRD biomarker studies
There have been previous studies investigating possible biomarkers for BRD outcome (Aich et al.,
2009, Eitam et al.). Aich et al 2009 established the value of using combinatorial “omics”
approaches to identify candidate biomarkers and used multimodal analysis of these biomarkers to
predict disease outcome. Twenty calves seronegative on enzyme-linked immunosorbent assay
(ELISA) for BHV-1 and leukotoxin of M. haemolytica were challenged on day 0 with BHV-1, with
serum samples collected prior to this infection. All calves were then challenged with M.
haemolytica on day 4 post infection when susceptibility to fatal secondary bacterial infection is
greatest, with serum samples collected days 1-4. Eight infected calves survived and twelve died.
Comparing samples from control animals on day 0 with samples from BHV-1 infected calves on
days 1-4 identified differentially expressed and statistically significant biomarkers. Multivariate
analysis was used to look for significant differences when animals were grouped by BRD outcome
of survival versus mortality. Proteomic studies identified a group of acute phase proteins,
haptoglobin and apolipoprotein, which could be linked to the primary viral respiratory infection, but
there was no significant association observed with fatal BRD. In contrast, metabolomic and
elemental analyses identified biomarkers for a primary viral infection (glucose, low-density
lipoprotein, valine, phosphorous, and iron) and also the disease outcome (lactate, glucose, iron).
While multivariate analysis of proteomic and metabolite profiles did not discriminate between
animals that survived or died post-synergic viral–bacterial infection by analyzing preinfection
serum samples, analysis of pre-infection serum elemental profiles, did allow some discrimination
(Aich et al., 2009). This study employed high-throughput methodologies but only challenged
twenty calves, and the number of control animals used is unclear. This study was designed to
discriminate between the complex biological responses induced by the primary viral infection and
the secondary bacterial infection resulting in mortality. While death due to BRD is a significant
disease outcome, the production loss in animals with less severe disease is also important. Using
this animal model achieving 60% mortality, there was little scope to assess any relationship of these
biomarkers to the varying severity of non-fatal BRD. It remains to be determined whether identified
biomarker profiles as predictors for mortality are only associated with the maximal viral-bacterial
synergy on day 4, or whether these biomarkers could be predictive of mortality with earlier or later
bacterial challenges.
Eitam et al 2010 aimed to detect individual variations in the stress responses of 13 newly received
16
young calves through measurement of their leukocyte heat shock protein response, selected
neutrophil-related gene expression and oxidative stress, and relate these variations to pulmonary
adhesions at slaughter as an indicative sign of clinical and subclinical episodes of BRD at an early
age. This study developed a discriminant analysis model, based on levels of linoleoyl tyrosine
oxidative products and β-glycan, according to which vulnerable individuals may be predicted at
near 100% probability after seven days arrival at the feedlot. The design and results of this study are
intriguing, but do address the problem of early detection of subclinical signs. However, as the
authors note, a larger scale experiment (i.e. n > 13) should be carried out to confirm this model
(Eitam et al.).
1.3 Conclusions and Research Proposal
To investigate the possibility of the cytokine/ROS- induced positive feedback loop of HIF-1α
induction contributing to the development of clinical BRD, or a possible lack of HIF-1α
stabilisation contributing to the development of clinical BRD, the following experiment, null
hypothesis and alternative hypotheses have been established.
1.3.1 Experimental hypothesis
The experimental hypothesis is that there is a significant difference in HIF-1α regulation between
animals with clinical BRD compared with other clinically healthy animals.
1.3.2 Null Hypothesis
The null hypothesis is that there is no significant difference in HIF-1α regulation in bovine
leukocytes of animals with clinical BRD compared with other clinically healthy animals.
1.3.3 Research Objectives
1. Isolate bovine peripheral blood leukocytes from previously observable clinically ill and
healthy animals in large enough concentrations for cell culture.
2. Develop a methodology for short-term culture of isolated leukocytes with an agent that will
stimulate HIF-1α expression; successful culture also entails achieving good post-culture cell
percentage viability and adequate total numbers of viable cells.
3. Develop a methodology for successful molecular tagging of cells with a fluoresecent marker
for the HIF-1α molecule and other phenotypic markers to identify leukocyte population sub-
types using flow cytometry or an alternative technique.
4. Investigate the value of measuring HIF-1α expression as a predictor for the outcome of
BRD by implementing developed methodologies experimentally to compare the HIF-1α
17
expression in a group of clinically healthy beef feedlot cattle with the HIF-1α expression in
feedlot cattle that developed clinical respiratory disease while in the feedlot.
5. Perform statistical analysis of acquired flow data and immunohistochemical data to
determine if there is a statistically significant difference of acute HIF-1α expression in
stimulated leukocytes from animals with BRD compared to clinically healthy animals.
18
2 Thesis overview
The next five chapters cover the development of methodologies of i) a hypoxic cell culture, ii) a
flow cytometry protocol for detection of HIF-1α in lymphocytes and iii) an immunocytochemistry
(ICC) protocol for detection of HIF-1α in monocytes, followed by the application of developed
methodologies and a general discussion. The development of these methods were interconnected
and often conducted concurrently. Each method involved iterative cycles of trial and error to
optimise different aspects (see figure 2.1).
Figure 2.1: Thesis overview showing the interconnected relationships between various aspects of the methodologies developed in chapters 3, 4 and 5. Flow cytometry results influenced the cell culture methodology and lead to the development of ICC. All developed methodology is applied to experimental samples in chapter 6, followed by a discussion of developed methodology and future directions in chapter 7.
19
3 Optimization of hypoxic cell culture and harvest
3.1 The production of HIF-1α in hypoxic conditions in vivo
When a cell encounters a low oxygen environment, intracellular HIF-1α protein is stabilized by a
decrease in HIF-1α ubiquitination and degradation (Kallio et al., 1999) due to reduced activity of
oxygen-dependent PHD enzymes (Epstein et al., 2001, Jaakkola et al., 2001)}. PHDs are also
actively degraded by the E3 ubiquitin ligase (Nakayama et al., 2004). Activating this pathway by
exposing the cell to true or simulated hypoxia will thus potentially allow measurement of cellular
HIF-1α.
3.1.1 Hypoxic cell culture
The hypoxic cellular environment required for HIF-1α stabilization is created experimentally by
one of two cell culture methods. Firstly, a hypoxic cell culture can be achieved by use of a modular
incubator chamber (Wu and Yotnda, 2011) to create a hypoxic atmosphere in which cell cultures
are conducted. In tissue culture, hypoxia is routinely defined as levels that are equivalent to between
0.5% and 3% oxygen by volume of air that perfuses the growth medium (Linder et al., 2003, Nizet
and Johnson, 2009). The second common method is to use a hypoxic-mimetic agent to simulate
hypoxia in cell culture. Common hypoxic-mimetic agents that induce HIF-1α are the iron chelator
desferrioxamine (DFO) (Wang and Semenza, 1993b) and cobalt chloride (CoCl2) (Wang and
Semenza, 1993c).
DFO is thought to induce HIF- 1α by substitution or removal of the Fe2+ ion (Wang and Semenza,
1993b, Schofield and Ratcliffe, 2004), inhibiting PHD enzyme activity. Studies have shown CoCl2
may inhibit HIF-1α degradation though a variety of other mechanisms including binding directly to
HIF-1α, thus hindering recognition by the pVHL protein, and may also be dependent on ROS
formation (Yuan et al., 2003, Chachami et al., 2004, Salnikow et al., 2004).
Hypoxic-mimetic agents are commonly used in hypoxic cell culture with human cell-lines to
stabilize and measure HIF-1α (Wu and Yotnda, 2011). Based on the prevalence and convenience of
hypoxic-mimetic cell culture utilisation, a hypoxic mimetic was chosen to stimulate stabilisation of
HIF-1α (hereafter referred to as stimulation) within selected bovine leukocytes. The aim was to
develop a protocol for hypoxic cell culture of selected bovine leukocytes that provided an optimal
yield of the target cells under conditions that promoted stabilisation of HIF-1α.
3.1.2 Bovine leukocytes chosen for hypoxic culture
Following BHV-1 infection, innate immune responses include the antiviral action of IFN and the
recruitment of lymphoid cells, macrophages, neutrophils and natural killer cells to the site of
infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Although neutrophils are phagocytes
20
involved in these responses, the neutrophil populations from the test animal or experimental
animals were not included in this study because granulocytes are typically poorly preserved by
freezing (Boonlayangoor et al., 1980, Nishimura et al., 2001) and use of fresh cells for culture was
not practical. The importance of HIF-1α in myeloid cell function has been thoroughly demonstrated
(Cramer et al., 2003) and HIF-1α has also been shown to be critically involved in the regulation of
the antimicrobial functions of lymphocytes (Thiel et al., 2007). Based on the involvement of these
cells in BHV-1 infection and the functional importance of cellular utilisation of HIF-1α, bovine
monocytes and lymphocytes were chosen for development of method for a hypoxic cell culture.
Monocytes and lymphocytes also survive frozen storage and were a practical choice for this project.
HIF-1α protein in leukocytes has traditionally been measured using immunoblot assays and
immunohistochemistry (IHC)(Yu et al., 1998, Semenza, 2005, Peyssonnaux et al., 2005) or ICC
(Jantsch et al., 2008). Most studies involving measurement of HIF-1α protein have been performed
in human or mouse cells, and the transferability of the findings to bovine leukocytes was uncertain.
Western blot analysis of HIF-1α has been undertaken in cultured bovine luteal cells (Nishimura
and Okuda, 2010) and corneal stromal cells (Xing and Bonanno, 2009).
This hypoxic cell culture aimed to (i) induce intracellular HIF-1α stabilisation, with cell harvest
focused on minimizing degradation of stabilized HIF-1α and (ii) facilitate subsequent measurement
of HIF-1α in these cells. This project explored the use of flow cytometry to measure the stabilised
HIF-1α protein as a relatively novel technique with the hopes that development of a successful
methodology would allow enhanced sensitivity in detecting differences in HIF-1α expression
between individuals. In the wake of methodology development issues, ICC was also implemented
as a more established method of HIF-1α measurement.
3.2 Materials, Methods and Results
To maximize potential for successful HIF-1α stabilization via hypoxic-mimetic culture with
minimal post-culture HIF-1α degradation, many aspects of the hypoxic-mimetic cell culture and
harvest methods required concurrent optimization. These aspects included the choice of hypoxic
mimetic agent and other stimulants, choice of culture duration, choice of culture vessels and cell
harvest procedures. These methods and results for optimization of these aspects are described in
sections 3.2.4, 3.2.5 and 3.2.6 below. Pre-culture sample preparation methods did not require
further optimization and are described in sections 3.2.1, 3.2.2 and 3.2.3 below.
21
3.2.1 Bovine leukocytes used in development of hypoxic-mimetic cell culture methods
Bovine blood, used for methodology testing and development was collected in 450 ml blood bags
from clinically normal female, Holstein Fresian teaching cows residing at the University of
Queensland (UQ) Gatton Campus dairy facility.
3.2.2 Procedure for mononuclear cell isolation
Mononuclear leukocytes were isolated from 300 ml of whole blood. Using an Eppendorf Centrifuge
5810R (Eppendorf South Pacific, Australia), 50 ml aliquots of peripheral blood were centrifuged at
2000 g, 4 °C for 25 minutes to separate out the erythrocytes from the leukocytes (“buffy coat”
layer). The buffy coat layer was collected from each aliquot and transferred into a fresh 50 ml tube,
and phosphate-buffered saline solution (PBS, pH 7.2) added to achieve a 20 ml total volume. The
resulting cellular suspension was layered onto 25 ml of a solution containing polysucrose and
sodium diatrizoate, adjusted to a density of 1.077 g/ml (Histopaque-1077; Sigma Aldrich, St Louis,
MO) and centrifuged at 2000 g, 20 °C for 40 minutes. The mononuclear cell layer was then
collected and transferred into a fresh 50 ml tube, filled to 50 ml with PBS and centrifuged at 1000 g,
4 °C for 15 minutes. The supernatant was poured off, and the cell pellet resuspended in 50 ml of
fresh PBS. This last step was repeated twice, centrifuging for 10 and 5 minutes, respectively.
Ten µl of cell suspension was mixed with 90 µl trypan blue 0.4% (Life Technologies, Victoria,
Australia) and a manual cell count of viable and non-viable cells was performed for each animal
using a haemocytometer (Neubauer, Germany). Cell viability was assessed on exclusion of trypan
blue (Strober, 2001). Percentage viability was calculated as the number of viable cells divided by
the number of total cells counted x 100. The number of cells/ml of sample was calculated using the
following method: #Viable cells counted within the 1mm x 1mm haemocytometer counting grid x
105 = cells/ml; cells/ml x sample volume in ml = total cell number in sample. These methods were
used for all cell viability assessments and counts conducted during the project. Cells in PBS were
spun down and resuspended at a concentration of 1 x 107 cells/ml and 0.9 ml aliquots were
transferred into 1.8ml sterile round-bottom cryotubes (Nunc, Roskilde, Denmark) mixed at a 1:1
ratio with freezing solution comprising 40 % Roswell Park Memorial Institute medium (Sigma
Aldrich, St Louis, MO), 40 % horse serum (Life Technologies, New Zealand) and 20 % dimethyl
sulphoxide (Sigma Aldrich, Missouri, USA).
3.2.3 Preparation of frozen test samples for culture
All cell cultures described hereafter took place in a Sanyo CO2 Incubator MCO-19AIC (Sanyo,
Sakata, Japan) at 5% CO2 and 37°C. The pre-culture setup procedures took place in a Biological
22
Safety Cabinet Class II (Email Westinghouse Pty Ltd, New South Wales, Australia) after 15
minutes of ultraviolet light sterilization and treatment of all items and surfaces with 70% ethanol.
Each frozen 1.8ml cryotube theoretically contained approximately 9 x 106 viable cells. Eight
cryotubes frozen at -81°C were retrieved for each animal and thawed at 37°C in a Grant W14 water
bath (Grant, Cambridge, UK). The contents of the cryotubes were pooled and immediately washed
three times in 15 ml PBS, centrifuged at 1000 RPM for 7 minutes at 4°C during each wash. The
cells were recounted and post-freezing cellular viability ranged from 65% to 86%, with total cell
yields (out of a theoretically possible 7.2 x 107 cells from 8 cryotubes) ranging from 1 x107 to 6 x
107 cells.
3.2.4 Hypoxic mimetic culture, stimulant and culture medium
3.2.4.1 Culture medium
The culture medium used in all cultures was X-Vivo 15 serum-free medium with gentamicin and
phenol red (Lonza, Maryland, USA). Gentamicin was considered essential for this project because
cells isolated from non-sterile blood collections were being cultured.
3.2.4.2 Hypoxic-mimetic culture and comparing hypoxic mimetics
For cell culture, DFO (Sigma Aldrich, St Louis, MO, USA) and CoCl2 (Sigma Aldrich, St Louis,
MO, USA) were tested for suitability as a hypoxic-mimetic agent to stimulate bovine leukocytes.
The criteria for suitability were adequate cellular viability and adequate total cell numbers
harvestable post-culture.
DFO was the first available hypoxic mimetic to be tested. Fresh mononuclear cells were washed
three times in PBS, and then re-suspended in serum-free medium at an approximate concentration
of 1 x 107 cells/ml. Five cultures were seeded with 250 µl cell suspension (estimated 2.5 x 106 cells)
and 250 µl of 200 µM DFO, achieving a final concentration 100 µM DFO. Five negative control
cultures were seeded with 250 µl of cell suspension and 250 µl serum-free medium without DFO.
All culture plates were incubated in 5% CO2 at 37°C. After 42 hours, culture fluid was collected
and plates rinsed twice with PBS to collect lymphocytes. The percentage viability of lymphocytes
in DFO-treated cultures ranged from 55 to 83%, and harvestable cell counts ranged from 6 x 105 to
1.2 x 106 cells. Control cultures had 63 to 91% viability and cell yields ranging from 7 x 105 to 1.6 x
106 cells.
23
Further cell cultures were set up to compare the effects of DFO to CoCl2 on cell viability and total
cell yields post-culture. Duplicates of approximately 5 x 106 cells were treated with 100 µM DFO,
100 µM CoCl2 or 200 µM CoCl2 and cultured along with untreated control cultures. One duplicate
well was harvested after 18 hours of culture and the other at 42 hours. After 18 hours, only a minor
difference in total cell yield for the two CoCl2 concentrations was recorded (3.9 x106 cells for 100
µM and 3.2 x 106 cells for 200 µM), with cellular viability of 90% for 100µM and 94% for 200µM
CoCl2. In contrast, DFO culture had only 1.2 x106 cells and 63% viability and the 18-hour control
total cell count was 1.1 x 107 cells with 95% viability. Cultures harvested at 42-hours had total cell
yields of 1.7 x106 cells for both 100µM and 200µM CoCl2 with less than 50% cellular viability.
The DFO culture cell yield at 42-hours was 1.6 x 106 cells with 55% viability whereas the control
culture contained 5.5 x 106 cells with 91% viability. Thus, at the tested concentrations CoCl2 and
DFO gave similar cell viability and yields at 42 hours, while at 18 hours both concentrations of
CoCl2 yielded greater cell viability and total cell numbers in comparison to the DFO culture. The
18-hour higher cell viability of CoCl2 compared to culture of the same duration using half the
concentration of DFO indicated that CoCl2 could be used at higher concentrations to maximize
potential hypoxia while minimizing cytotoxicity. CoCl2 was therefore the hypoxic mimetic of
choice for further hypoxic cell culture development.
3.2.4.3 Lipopolysaccaride (LPS) as an additional culture stimulant
CoCl2 was titrated further with or without the addition of lipopolysaccharide (LPS) from
Escherichia coli (Sigma Aldrich, Missouri, USA). The inclusion of LPS as an additional stimulant
in these hypoxic cell cultures to stabilize the HIF-1α molecule was based on the previously verified
ability of bacteria or LPS to stabilize and contribute to accumulation of HIF-1α under both hypoxic
and normoxic conditions (Rius et al., 2008). LPS raises HIF-1α levels through pathways involving
MAPK and NF-kB (Peyssonnaux et al., 2007). Maximal induction of HIF-1α in alveolar
macrophages has been observed using an LPS concentration of 1µg/ml in 6-hour cultures of murine
alveolar macrophages (Blouin et al., 2004) but LPS concentrations in this project were increased to
test the upper limits of using this stimulant. CoCl2 was therefore further tested in six-hour cultures.
Each culture contained 2.5 x 106 cells exposed to either 200 µM, 600 µM or 1.8 mM CoCl2 in
permutations with relatively high LPS concentrations of 5 µg/ml, 15 µg/ml and 45 µg/ml. Results
showed that cellular viability decreased with increased CoCl2 dose; culture with 600 µM and 1.8
mM CoCl2 resulted in less than 50% cellular viability after 6 hours, with total cell counts of 1.8 x
24
106 and 9.7 x 105 cells for 600 µM and 1.8 mM CoCl2 respectively where CoCl2 was used without
LPS. Cultures with 600 µM and 1.8 mM CoCl2 in combination with all concentrations of LPS also
resulted in less than 50% cellular viability. The addition of LPS at 5 µg/ml to 200µM CoCl2 gave
the best cellular viability at 73% and the highest total cell count at 2.15 x 106 cells. LPS was
subsequently used at 5 µg/ml as this concentration gave the best cellular viability while being
closest to the previously published 1µg/ml.
3.2.5 Culture duration testing
Stabilisation and measurement of HIF-1α has been recorded with a culture time as short as three
hours (Pisani and Dechesne, 2005) or four hours (Bove et al., 2008). Forty-two-hour cultures were
initially tested; these cultures often resulted in extremely low post-culture viability and experiment
discontinuation due to lack of cells. Shorter culture times (6 to 8 hours) resulted in acceptable post-
culture viability of cells (refer to section 3.2.4.2 and 3.2.4.3).
3.2.6 Culture vessel testing, post-culture viability and HIF-1α degradation
3.2.6.1 Plate cultures
The selection of culture vessel had various far-reaching effects on the outcome of the culture.
Initially, individual, 6-well or 24-well plates (Nunc, Roskilde, Denmark) were trialed. The larger
numbers of cells and culture medium volume needed were issues in addition to the important issue
of adherent monocytes. Single, 6-well or 24-well culture plates provided large surface areas for
activated monocytes to adhere to during culture. The outcome to this adherence was that a great
proportion of activated monocytes would then remain on the plate surface after repeated washing,
remaining unavailable for harvest into suspension for subsequent analysis by flow cytometry.
Lidocaine may be used to detach adherent monocytes from plate surfaces (Ohmann et al., 1983,
Nielsen, 1987). To test the feasibility of using lidocaine to retrieve adherent monocytes from plate
cultures, four trials using round culture plates were performed. After a four hour culture, the cell
culture plates were placed on ice after washing with PBS to retrieve non-adherent cells, and each
plate surface was treated with 2ml of 3.5mg/ml Ilium Lidocaine 20 (Troy Laboratories,
Glendenning, Australia). After 15-20 minutes the lidocaine reduced monocyte adherence to the
plate surface and subsequent gentle physical removal of the monocyte population by light abrasion
of the culture surface with a 20µl pipette tip was possible. Total cell yields from this harvesting
technique were on average 2 x 106 cells from an original total of 1 x 107 cells spread across four
culture plates. Unfortunately the original percentage of monocytes in the pre-culture population was
not determined at the time, but was estimated off later aliquots using manual differential counts to
25
be approximately 39%, meaning that approximately half of cultured monocytes could be retrieved
when using lidocaine during harvest.
3.2.6.2 Culturing in tubes
Five ml polypropylene sterile ‘snap cap’ tubes (Beckton Dickinson, NJ, USA) were initially chosen
for culture to discourage the adhesion of activated monocytes to the internal tube surface, and
similar total cell yields were achieved when comparing this technique to culturing and treating
plates post-culture with lidocaine. Stimulated tube cultures contained 1.25 x106 cells in 250µl
serum-free medium suspension treated with 200 µM CoCl2 and 5µg/ml LPS. A tube containing
non-stimulated cells suspended in serum-free medium was used as a control. The use of 5ml tubes
in an 18-hour culture in 5% CO2 at 37 °C gave an average 20 % cellular survival rate.
3.2.6.3 Preservation of HIF-1α by placing post-culture cells on ice, with cellular fixation in
presence of CoCl2
Tube cultures were immediately placed on crushed ice (temperature approximately 4°C) with the
intention to slow cellular processes including potential degradation of the HIF-1α molecule by the
ubiquitin proteasome (Walshe and D'Amore, 2008). The chilled cells then remained in the presence
of CoCl2 until undergoing formalin fixation. Manufacturer’s recommendations stated that the
fixative solution Medium A (Life Technologies, Victoria, Australia) should be used at room
temperature, however, it was successfully used at 4°C after consultation with the manufacturer.
Post-culture cell counts and cell washings in PBS were omitted to keep cells exposed to CoCl2 up
until the point of cellular fixation to minimise potential HIF-1α degradation, Due to reduced
binding of the B cell surface marker anti-CD20 antibody to fixed B cells, a 20-minute incubation of
cells with this antibody took place while on ice prior to fixation. However there was no further
washing step needed to remove this antibody before use of Medium A to fix the cells.
3.2.6.4 Tube cultures: monocyte viability during harvest and cellular staining
After culture, cells were harvested, fixed and stained with antibodies for surface markers of B cells,
T cells, monocytes and HIF-1α for analysis by flow cytometry (refer to chapter 4 for further detail
of staining and flow cytometry). It was observed that the cell populations harvested from all tube
cultures showed a repeatable and problematic drop in the proportions of CD14 positive cells
(monocytes) detectable by flow cytometry by detection of a pacific blue fluorochrome attached to
the CD14 antibody. To assess the relationship between cell culture and the apparent absence of the
monocyte population, aliquots of cell suspension isolated from the same individual animal were
thawed, washed and left on ice without undergoing culture, thawed and cultured as negative
26
controls, or thawed and cultured with CoCl2 and LPS. After 18 hours, a post-culture 100µl of cell
suspension was removed for cytospin slide preparation and Wright’s Giemsa staining to facilitate a
manual differential cell count. All remaining cells were routinely stained with CD14 antibody
labelled with a pacific blue fluorochrome (refer to chapter 4 for further detail of staining). A post-
staining 100µl sample of cell suspension was taken for cytospin slide preparation. A post-thawing
cytospin slide was also prepared from the uncultured cells.
Analysis software (Kaluza, Beckman Coulter) was used to gate cell populations to exclude debris
and aggregated cells (refer to Chapter 4 for further detail). Side scatter (SS) is a measure of the
internal complexity of the cell and can be useful in determining populations of monocytes (Givan,
2001). For each sample, a dot plot of cells was created by plotting SS against the signal of the
pacific blue fluorochrome of the CD14 antibody (figure 3.1). The thawed sample had a well-
demarcated CD14 positive population, considered distinct due to the relatively narrow range of SS
of this population as shown in figure 3.1A. This population had distinct positive staining for pacific
blue as shown by the population’s location as furthest to the right on the increasing log scale of
pacific blue fluorescent signal (figure 3.1A). Based on the lower end of this positive signal, a
quadrant gate entitled ‘CD14 positive cells’ was applied to set the positivity cut-off point for pacific
blue fluorescence, and the percentage of CD14 positive cells in the uncultured sample was
calculated by Kaluza to be 22.0%. These gates were replicated on the graphs for the cultured
samples. The sample cultured without stimulants had a CD14 positive population of 14.4%, as well
as a general increase in SS in both negative and positive cells (figure 3.1B). The sample cultured
with CoCl2 and LPS had a CD14 positive population of only 1.6%, as well as increased SS, and the
distinct population of CD14 positive cells was no longer visible (figure 3.1C).
Manual cell differential counts of monocytes and lymphocytes were performed on the post-
thawing/post-culture, and post-staining cytospin slides. Counts were performed by the author and
an experienced clinical pathologist. Post-thaw uncultured cells comprised 61% lymphocytes and
39% monocytes, cultured cells without stimulants comprised 70% lymphocytes and 30%
monocytes, while cells cultured with CoCl2 and LPS comprised 96% lymphocytes and only 4%
monocytes. The post-staining cytospins displayed poor cellular preservation and manual cell
differential was deemed too unreliable, however it was concluded that manual cell differential
counts showed decreased proportions of monocytes present after exposure to culture and/or
stimulants, in agreement with flow cytometry results. These findings were likely to be due to
cellular fragility and monocyte apoptosis. Poor cellular preservation on post-staining cytospin slides
27
and the overall lower percentages on monocytes detected during flow cytometry suggested the
possibility that staining procedures also contributed to monocyte depletion.
Figure 3.1: A: SS/Pacific Blue dot plot obtained from Kaluza of all events within the Cells gate. Colour gradients denote areas of lower (blue) to higher event density (red), percentages are of all events within the Cells gate. An uncultured aliquot of cells stained with pacific blue-labelled CD14 antibody shows a distinct CD14 positive population to the right of a larger group of negative cells. B: An aliquot of cells cultured without stimulants shows a similar but depleted CD14 positive population, and generalised increased SS of all cells. C: An aliquot of cells cultured with CoCl2 and LPS shows generalised increased SS of all cells but the CD14 positive population cannot be distinctly visualised.
28
3.2.6.4.1 Immunocytochemical (ICC) staining for activated caspase-3 as measure for apoptosis
Eighteen-hour cell cultures were repeated to collect post-culture and post-staining samples to
prepare further cytospin slides for ICC staining for activated caspase-3, a molecule which is
involved in programmed cellular death and can be used as a measure of apoptosis. Treatments were
i) cells that were thawed but put back on ice for 18 hours without culture, ii) cells that were cultured
without CoCl2 and LPS, iii) cells that were cultured with CoCl2 only, and iv) cells that were
cultured with CoCl2 and LPS. After culture and again after staining with CD14 antibody, 150µl of
cell suspension was taken to create three cytospin slides for each time-point. Cytospin slides were
heated at 95 °C in Target Retrieval Citrate Solution pH 6.0 (DAKO, Carpentaria, USA) for 25
minutes, followed by a 20 minute cooling period. For the detection of activated caspase-3, a rabbit
anti-activated caspase-3 polyclonal antibody (Abcam) was used in combination with the DAKO
Envision kit for detection of rabbit antibody binding. The chromogen was AEC. ICC staining of
mononuclear cells for activated caspase-3 confirmed the majority of mononuclear cells were
expressing this intracytoplasmic molecule and were likely undergoing apoptosis. Uncultured cells
exhibited moderate staining intensity for activated caspase-3 (Figure 3.2A) with a slight increased
intensity noted in uncultured cells in the post-staining slide. The most intense staining was evident
in cultured samples in both the post-culture and post-staining slides (Figure 3.2B). It was concluded
that cells that had undergone cell culture had increased levels of intracellular activated caspase-3
compared to similar uncultured populations that had been thawed and kept chilled before staining. It
was concluded the cell culture conditions were therefore likely causing increased cellular apoptosis,
while permeabilisation during staining (refer to chapter 4) might further deplete cell numbers by
causing the rupture of already fragile cells. Taken together, these findings indicated that the
detection of HIF-1α in cultured monocytes using flow cytometry was currently unachievable, and
that an alternative plan was necessary.
29
Figure 3.2: A: ICC of uncultured mononuclear cells with anti-activated caspase-3. There is moderate staining intensity within both cells with abundant and scant cytoplasm (putative monocytes and lymphocytes, respectively). B: ICC of mononuclear cells cultured with stimulants before undergoing staining procedures for flow cytometry. There is marked staining intensity of cells with abundant cytoplasm (putative monocytes).
3.2.6.5 Chamber slide cultures for monocyte HIF-1α
Nunc Lab-Tek chamber slides (Nunc, Roskilde, Denmark) were selected as an alternative to
plates/tubes to study the expression of HIF-1α in bovine monocytes. With four chambers on each
slide, this layout allowed for two treated (i.e. stimulated with 200 µM CoCl2 and 5µg/ml LPS) and
two untreated negative control chambers (refer to Chapter 5, Figure 5.1).
Each chamber was seeded with 2.5 x 106 cells in 500 µl serum-free medium (with or without
stimulants) and cultured for 18 hours in 5% CO2 at 37°C. Immediately post-culture the culture fluid
containing lymphocytes was discarded and the chamber slides immersed in 10% buffered formalin
for 5 minutes to fix the remaining adherent monocytes and minimize the potential for HIF-1α
degradation within the adherent cells. The chamber slides had detachable chamber walls that were
removed after monocyte fixation, leaving a short (3mm) plastic gasket designed to neatly facilitate
subsequent ICC staining of the monocyte population. Slides were air-dried and stored in clean slide
containers that were sealed in zip-lock plastic bags and then stored at 25°C away from light.
3.3 Discussion
The work up of factors such as variable numbers of cells cultured, choice of hypoxic mimetic,
culture duration, culture vessels and post-culture harvesting processes led to a number of key
decisions regarding the cell culture process and the subsequent methods used to detect the HIF-1α
protein. It was determined that a total of 2 x 107 cells were required for the monocyte cell cultures
for ICC and 2.5 x 106 cells were required for lymphocyte cultures to be used for flow cytometry.
30
Eight cryotubes were required from each animal. This was expected to provide sufficient cells for
ICC and flow cytometry considering the range of cellular viability observed.
CoCl2 was used in subsequent test cultures at no less than 200 µM, to attempt to maximize
stimulation with the view that cytotoxicity was going to be a consistent feature with this cell
culture. The results indicate likely increased cellular apoptosis in cultures with DFO when
compared to CoCl2, but that both hypoxic mimetics result in apoptosis. These results were
consistent with a previous study that found that DFO and CoCl2 treatment of cell preparations for
HIF-1α stabilisation resulted in a dose dependent reduction in cell viability via apoptosis in
leukemic cells with DFO being the more cytotoxic (Guo et al., 2006). The observation of a dose-
dependent reduction in cell viability is consistent with another study in which human pancreatic
cancer cells were cultured with different doses of CoCl2 for 72 hours, and HIF-1α protein
expression measured using western blot. Apoptotic cells were identified by ultrastructural changes
using electron microscopy at 48-hours and by measurement of Annexin V and propidium iodide
using flow cytometry after 72-hours (Dai et al., 2012). In addition to a dose-dependent increase in
HIF-1α protein expression, this study also reported a dose-dependent increase in apoptosis evident
from flow cytometry, with the highest dose of CoCl2 at 200 µM resulting in 52.3% of cells
identified as apoptotic after a 72-hour culture.
LPS as an additional stimulant appeared to improve post-culture cellular viability and was added to
the final protocol. It is possible this effect may be due to LPS-induced stimulation of monocytes
with IL-6 production, which in turn stimulates B lymphocytes (Mangan and Wahl, 1991) although
this seems unlikely considering the relatively short culture time of 6 hours used during LPS
titration.
The poor post-culture viability that required this optimisation has been shown to be a significant
issue with frozen-thawed cells. Regrettably, due to the ongoing trouble-shooting and work-up of
protocols and the restricted nature of sample collections, the opportunity to culture fresh cells
immediately after collection did not arise. Using fresh cells remains an avenue for future
investigation into hypoxic culture of bovine leukocytes for stabilisation of HIF-1α.
An 18-hour culture was used to maximise the culture time and thus potential for HIF-1α
stabilization by the cells of the bovine immune system while still retaining adequate numbers of
viable cells for post-culture analysis. A study subjecting human glioblastoma cells to in vitro
hypoxia for 1 hour, 6 hours or 18 hours with oxygen concentrations between 2% and <0.02%,
demonstrated, using western blotting and flow cytometry, that there was a clear pattern of
31
increasing HIF-1α protein expression in the nuclear extracts of these cells (Vordermark and Brown,
2003). Another study recommended a culture duration of 24 hours when culturing mammalian cells
in 100 µM CoCl2 in 5% CO2 and 37 °C (Wu and Yotnda, 2011).This project demonstrated that
drug-related cytotoxicity in bovine leukocytes with use of hypoxic mimetics was compounded by
length of culture, however as these other studies have shown that HIF-1α stabilisation increases
with culture time, a balance needs to be achieved. There does not appear to be any published data
on previous attempts of hypoxic culture for HIF-1α in bovine leukocytes in which to refer.
The use of lidocaine to retrieve adherent monocytes was not pursued once it was realised that the
cultured monocytes were not durable enough for flow cytomtery. The use of lidocaine to harvest
monocytes from plates resulted in percentages of harvested cells comparable to those using tubes
for culture. However, the prolonged time required for the lidocaine to take effect also precluded
further use as during this time the unfixed monocytes were re-exposed to normoxia after rinsing
away the CoCl2, resulting in degradation of HIF-1α to basal levels in all cultures due to a 5-10
minute half-life (Wang et al., 1995).
Post-culture monocyte viability investigations of tube cultures indicated that monocyte apoptosis
was likely being induced by cell culture stimulation to the degree that these cells would not tolerate
the staining process for flow cytometry.
The percentage of cells identified as monocytes in the thawed uncultured cell population and the
unstimulated cultured cell population was higher than expected. The possibility that some of these
cells were large activated lymphocytes cannot be discounted, thus meaning the drop in the
proportion of monocytes after culture may have been less than was apparent from the manual
differential counts. However, the practical reality remained that hypoxic culture of this cell
population resulted in the loss of an appreciable CD14 positive population on flow cytometry,
whereas within an uncultured cell population a distinct CD14 positive subset was clearly visible,
indicating a loss of monocytes or a loss of CD14 labelling by the CD14 antibody. As CD14 is a
receptor for LPS(Kielian and Blecha, 1995), there is a possibility that LPS binding could prevent
the CD14 labelled antibody from binding to CD14 via several possible mechanisms. LPS has been
shown to bind to CD14 without the presence of LPS-binding protein (LPB) in serum-free medium,
although LPB accelerates this process(Hailman et al., 1994). A short term 1-hour culture with LPS
followed by CD14 labelling could have investigated this idea further.
The flow cytometry results and discussion of this issue are included in this chapter as these
consequences for the flow experiment fundamentally influenced the final protocol for culture; tube
cultures were therefore used exclusively for lymphocyte culture and analysis and not for monocyte
32
analysis. The 20% cellular survival rate of cells in tube cultures suggested that the reduced
medium/CO2 interface (compared to traditional plates) did not adversely affect cellular viability.
The approximately 20% post-culture survival rate was consistent, and so post-culture manual cell
counts were discontinued as another way to further minimise harvest-to-fixation interval and
preserve the HIF-1α protein.
The use of chamber slide culture followed by immediate slide fixation was chosen as the best
alternative way to assess the monocytes. Chamber slide culture and immediate fixation would
address both the post-culture monocyte apoptosis issue by temporally minimizing the opportunity
for viable monocytes to undergo apoptosis in the post-culture period, as well as minimize HIF-1α
degradation in the adherent monocytes. The monocytes would be assessed using ICC to take
advantage of the monocyte adherence to slides, the ability to quickly fix the cells and due to time
constraints of the project.
Due to the short 5-10 minute half-life and consequent rapid degradation of HIF-1α under aerobic
conditions (Wang et al., 1995), it was deemed vital to perform the initial steps of tube culture
harvest on ice and to employ use of ice-cold reagents; similar techniques were employed
successfully in another published flow cytometry protocol used to assess HIF-1α as an intrinsic
marker of tumour hypoxia (Vordermark and Brown, 2003).
The rapid degradation of HIF-1α under normoxic conditions is mediated by the ubiquitin-
proteasome system (Walshe and D'Amore, 2008) and a number of proteasome inhibitors are
commercially available to halt protein degredation by these proteasomes within a cell. A study that
tested specific proteasome inhibitors, including Lactacystin, MG-132, calpain inhibitor I and
calpain inhibitor II. Using gel shift assay measurement of nuclear extracts it found that 100 µg/ml
calpain inhibitor I, when added during the last 15 min of a 4-hour, 0.5% O2 hypoxic culture of Hep
3B cells, protected the degradation of the HIF-1 complex in cells transferred from hypoxia to
normoxia (Salceda and Caro, 1997). The use of calpain inhibitor I in the post-culture harvest
protocol was decided against because the same study cited demonstrated that the same
concentration of calpain inhibitor I will also induce stabilization of HIF-1α and formation of the
HIF-1 complex in cells under normoxic conditions and that there was no readily apparent way to
distinguish this from stabilization induced by the hypoxic mimetic agent (Salceda and Caro, 1997).
An additional reason is that proteasome inhibitors have also been shown to impair hypoxia-
dependent translocation of HIF-1α from the cytoplasm to the nucleus (Kallio et al., 1999). A
proteasome inhibitor was not used to avoid removing this valuable evidence of HIF-1α
functionality from the ICC results.
33
3.4 Conclusion – final cell culture and post-culture harvest protocol
Ten mM CoCl2 stock solution was made up fresh each time and frozen 100 µg/ml LPS aliquots,
thawed once only, were used to ensure consistent stability of both stimulants. Frozen cells were
thawed in a 37°C waterbath and washed three times in PBS. Cells were counted and resuspended in
serum-free medium at 5 x106 cells/ml. Tube cultures for lymphocyte analysis were cultured with
1.25 x106 cells in 250 µl serum-free medium suspension treated with 200 µM cobalt chloride and 5
µg/ml LPS; with a non-stimulated control tube containing only the cells suspended in serum-free
medium. Chamber slides for monocyte analysis were cultured containing 2.5 x 106 cells in 500µl
serum-free medium with or without stimulants. Tubes and chamber slides were cultured for 18
hours in 5% CO2 at 37°C, with chamber slides set up just prior to the tube cultures. At harvest,
chamber slides were harvested first as described in section 3.2.6.5. While the chamber slides were
air-drying after fixation, the tube cultures were harvested and placed straight into wet ice for 60
seconds before incubation with the anti-CD20 antibody for 20 minutes while remaining on ice
(refer to Chapter 4 for further detail). One hundred µl fixative solution Medium A was then added
to each tube and incubated for a further 20 minutes on ice. The subsequent steps are covered in
chapter 4.
34
4 Optimization of flow cytometry protocol for measuring HIF 1α expression in lymphocyte subtypes
4.1 Introduction
HIF-1α protein in leukocytes has traditionally been measured using immunoblot assays and IHC
(Yu et al., 1998, Semenza, 2005, Peyssonnaux et al., 2005) or ICC (Jantsch et al., 2008). Western
blot analysis of HIF-1α has been undertaken in cultured bovine luteal cells (Nishimura and Okuda,
2010) and corneal stromal cells (Xing and Bonanno, 2009). Flow cytometry is able to provide a
sensitive and semi-quantitative measurement of protein expression such as HIF-1α. This quality,
together with the ability to simultaneously phenotype the cell population in the same assay made
flow cytometry an attraction option for investigating the aims of this project. Phenotyping would
indicate if specific populations of leukocytes were responsible for any detected differences in HIF-
1α expression.
This chapter covers the work-up of the specific antibodies for flow cytometry that were investigated
during this project but also discusses basic concepts of antibody titration for flow and interpreting
graphical flow data as relevant. 4.2 Materials, methods and results
4.2.1 Flow cytometer maintenance and flow cytometry controls
The machine in use during this project was a Gallios flow cytometer (Beckman Coulter, Australia).
This cytometer contains a blue laser (488nm), red laser (638nm) and a violet laser (405nm) with ten
available channels (or detectors) for fluorescence detection. Given this wide range of available
excitation and emission detection capability, fluorochromes were chosen to minimise spectral
overlap.
The fluorochromes fluorescein isothiocyanate (FITC) and Dylight-488 are excited by the blue laser
with Dylight-488 having an excitation wavelength of 493nm and a emission wavelength of 518nm.
Allophycocyanin (APC), APC-Cy7 and Alexa Fluor® 647 are excited by the red laser with Alexa
Fluor® 647 having an excitation wavelength of 652nm and an emission wavelength of 668nm.
Lastly, the Pacific Blue fluorochrome is excited by the violet laser, with an excitation wavelength of
403nm and an emission wavelength of 455nm.
In a case of spectral overlap, each fluorochrome will contribute a signal to several detectors,
therefore the contribution in detectors not assigned to that fluorochrome must be subtracted fromt
the total signal in those detectors. This process is called compensation, by which the spectral
overlap between different fluorochromes is mathematically eliminated. (Baumgarth and Roederer,
2000).
35
Instrument controls are used to check the set up of the instrument, including the PMT voltage, gain
and compensation (Maecker and Trotter, 2006) The instrument controls used in this project were
Flow-check Pro beads (Beckman Coulter, Sydney, Australia) for laser alignment and the use of a
master control sample during the running of experimental samples. Cleansing and priming was also
performed between runs to avoid machine blockage. Specificity controls are used to distinguish
specific from non-specific binding, and biological controls are to provide relevant comparison
conditions for a marker stimulated in vitro (Maecker and Trotter, 2006). In this project the cells
cultured without any CoCl2 or LPS (unstimulated cells) but stained with the same antibodies served
as both a specificity and biological control for HIF-1α. Unstained samples served as specificity
controls for the titration of CD3, CD20 and CD14 antibodies as immunophenotypic markers.
Phenotypic antibody titration occasionally also incorporated an isotype control to look for any
obvious affects of non-specific staining of an antibody of a particular isotype, but there are apparent
limitations of such isotype controls to truly match the background staining of the antibody being
tested (Maecker and Trotter, 2006).
4.2.2 Cellular staining method for flow cytometry
Developing an antibody staining protocol for flow cytometry involved optimising the types,
combinations and amounts of antibody used, the length of antibody incubation, incubation
temperature and the omittance or inclusion of wash steps. With the exception of the earliest
cultures, the consistent features of the staining method following hypoxic culture were as follows.
Tube cultures (both stimulated and unstimulated) for flow cytometry were harvested from the CO2
incubator and placed straight into wet ice for 60 seconds before addition of the anti-CD20 antibody.
One hundred µl of fixative solution medium A (Life Technologies, Vic, Australia) was then added
to each tube which was vortexed and incubated on ice for 20 minutes in the dark. Tubes were half-
raised from ice and gently vortexed again at mid-incubation. Two ml of wash medium comprising
3% horse serum (Life Technologies) and 1% sodium azide (NaN3) (BDH Lab Supplies, UK) in
PBS was added to each tube, and cells were pelleted by centrifugation for five minutes at 350 g at
25°C. The supernatant was removed followed by a gentle vortex to resuspend the pellet. One
hundred µl of permeabilisation solution medium B (Life Technologies) was added to each tube
along with CD3, CD14 and HIF-1α antibodies as appropriate to the specific experiment. Each tube
was gentled vortexed and incubated at room temperature. Cells were washed and centrifuged again
as described for the medium A-step. Supernatant was removed and cells were resuspended in 500µl
of 0.1% formalin in PBS for short-term storage in the dark at 4°C. For analysis by flow cytometry,
samples were resuspended in Isoflow sheath fluid (Beckman Coulter).
36
4.2.3 Antibody titration
Flow cytometry can be broadly defined as a system of identifying cells and measuring molecules of
interest by analyzing the scatter of light that result as particles flow in a stream of liquid through a
beam of light (Givan, 2001). Both forward scatter (FS) and the side scatter (SS) of light are
measured with FS providing information on cell size and volume of an event, while SS is an
indication of internal complexity of a cell, taken together, these parameters can be used to identify
cell populations (Givan, 2001).
For each single phenotypic antibody titration discussed, aliquots of 5 x 105 cells were stained with
increasing amounts of antibody starting with neat amounts of 1µl (1/200), 2µl (1/100), 4µl (1/50),
8µl (1/25) and 10µl (1/20 dilution) 12µl (approximately 1/16) and 16µl (approximately 1/12) in a
total of 200µl cell suspension. Two aliquots of cells were fixed and permeabilised without antibody
as unstained cells. All cells used during single antibody titrations were uncultured as it was easier to
visually differentiate populations on FS/SS scatter plot of uncultured cells. A typical FS/SS scatter
plot of the uncultured test cells is shown in figure 4.1, depicting the location of signal clusters of
debris, lymphocytes, monocytes and larger events termed ‘other’ (putative aggregates of cells or
large debris). This scatter plot is consistent with a typical scatter plot of peripheral blood
mononuclear cells (Givan, 2001).
Figure 4.1: A bivariate scatter plot of the forward scatter (FS) and side scatter (SS) of the uncultured unstained test cells as visualised on a linear scale. Debris usually has low FS and SS. Lymphocytes have moderate FS but relatively low amounts of SS in contrast to monocytes. Events with high FS and SS are postulated to be aggregates of cells. To then determine if a phenotypic antibody had stained any of the cells, the stained cells were first
compared to unstained control cells. At least 10,000 events were analysed for every sample. The
37
unstained cells were analysed, with machine voltage and gain of the FS, SS and relevant
fluorochrome channel adjusted such that these unstained cells appeared on the scale in the FS/SS
scatter plot, and in the first decade of a 4-decade logarithmic scale of a histogram for the
fluorochrome being measured (Maecker and Trotter, 2006) and thus becoming the initial negative
boundary. Using the same instrument settings, the aliquots of stained cells were then analysed
starting with the lowest concentration of antibody. The univariate histogram of the fluorochrome
was assessed for the emergence of a distinct positive peak (see figure 4.2B) and the bivariate scatter
plot of SS and the fluorochrome was assessed for an emerging distinctly positive population (figure
4.2B). The instrument voltage and gain was minimally adjusted to keep the entire putative positive
and negative populations visible on the histogram, with the negative population kept to the left half
of the histogram (first two decades) where possible. After samples were run, the resulting machine
settings were saved as a protocol to be used later during the multi-antibody protocol setup.
The light signals created as FS, SS and fluorochrome fluorescence are converted by photodetectors
into electronic signals, the intensity of which is related to the intensity of the light. Photodetectors
have voltages applied so that the electronic signal is of a large enough current to be measured
(Givan, 2001). Changing the voltage is a way of decreasing or increasing the current response of
that detector. The other method to change the response to light is to alter the amplification of that
current after it leaves the detector by using linear or log amplification, with choice of the precise
gain applied (Givan, 2001). Applying discrimination to a detector limits events detected – in this
project only discrimination used was to restrict the FS so that very small events (debris and dead
cells) were excluded, and this was minimal (see Table 4.1). Once an electronic signal has been
created and amplified, the intensity of the signal is analysed and grouped (binned) by an analogue-
to-digital converter (ADC) into 1 of 1024 specific signal ranges called channels (Givan, 2001).
These 1024 equal channels are plotted along the x-axis of the histograms and scatter plots used in
this chapter. In this project, log amplification was used to look at light signals over a wide range of
intensity as this is a common amplification used for immunofluorescence studies (Givan, 2001).
The x-axes are thus divided into four log decades so that each log decade represents a quarter of
available channels. Numerical values from the flow cytometry analysis software Kaluza (Beckman
Coulter, Sydney, Australia) that are given in this chapter (and in the experimental chapter) are in
reference to these channels, with differences in fluorescence intensity expressed as a channel shift
(e.g. channel shift from 5 to 8 is channel shift of 3) rather than relating back to a linear change in
fluorescence intensity.
A list mode data file (LMD) was created for each analysis, which described each cell in the
sequence in which it passed through the cytometer laser beam (Givan, 2001) measuring the
parameters that were selected during the analysis (e.g. FS, SS, and detected fluorescence in
38
channels of interest). LMDs were then used for future computer analysis of data collected during
flow cytometry.
4.2.4 T lymphocyte CD3 antibody selection and titration
The multichain CD3 molecule is located within and the cellular membrane of normal and neoplastic
T cells (Campana et al., 1989, Janossy et al., 1989) and in virtually no other cell type, although it
does appear to be present in small amounts in Purkinje cells (Garson et al., 1982). The expression of
the epsilon chain of CD3 appears to be stable regardless of the activation and differentiation stage
of the T cells and therefore constituted the most robust immunophenotypic marker for T lineage
cells in flow cytometry and IHC. 4.2.4.1 Initial T lymphocyte CD3 antibody
Please refer to appendix (section 9.1) for further details. This antibody was not used in the final
protocol.
4.2.4.2 Alternative T lymphocyte CD3 antibody
An alternative antibody, a commercial rat anti-human CD3 IgG1 (clone CD3-12) conjugated to the
fluorochrome Alexa Fluor® 647 (MCA1477A647, Abd Serotec, Oxford UK) was researched and
selected for testing. The Alexa Fluor® 647 is excited by the red laser (638nm) and is detected in the
FL6 channel on the Gallios (see first figure). This clone has reported cross-reactivity on IHC in a
variety of species (Jones et al., 1993) and stated bovine species cross-reactivity in the manufacturer
datasheet. Subsequent to this experimental work, others have published in the use of this CD3
antibody to identify bovine T cells by flow cytometry (Yang et al., 2012).
4.2.4.2.1 IHC staining to confirm T cell positivity
Reactivity of this antibody with bovine T cells was confirmed by immunohistochemical staining of
a bovine lymph node at 1/50 dilution. This showed positive signal confined to the cells within the
paracortex, a T lymphocyte zone (data not shown).
4.2.4.2.2 CD3 antibody titration on flow cytometry
The CD3 antibody was raised against a synthetic peptide of an intracytoplasmic epitope of the
transmembrane part of the CD3ε molecule and so cellular membrane permeabilisation was required
to allow the antibody to enter the cytoplasm and bind.
Titration showed a distinct population of positive events within the ‘cells’ gate. This population
appeared at 2µl (20.5% of gated events) and was considered best visualised using 8µl of neat
39
antibody where 19.8% of the cells were positive for CD3 (Figure 4.2B). This percentage was
relatively steady within increasing amounts of antibody and was considered to represent a
consistent population of T cells.
The previously tested CD3-APC antibody was used as an isotype control. Based on titration, 8µl
was considered the most optimal amount of this antibody when used as a single stain.
Figure 4.2: A: Peripheral blood mononuclear cells - histogram of CD3 fluorescence of events with the cells gate. A peak of cells positive for CD3 can be visualised to the right of the larger negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD3 fluoresence versus SS, log scale. A population positive for CD3 has been gated by eye (black border). This positive population is distinct from and far to the right of the negative population.
40
4.2.5 B lymphocyte CD20 antibody selection and titration
The CD20 is a transmembrane protein expressed on mature B cells but not on plasma cells, and is a
commonly used B cell marker (Nadler et al., 1981).
4.2.5.1 Initial B lymphocyte CD20 antibody
The initial B lymphocyte antibody selected was an APC-Cy7 conjugated mouse anti-human CD20
antibody (ab82017, Abcam). Following the titration process described previously for CD3, this
antibody stained too weakly to be considered for further testing (data not shown).
4.2.5.2 Alternative B lymphocyte CD20 antibody
The next B lymphocyte antibody chosen for investigation was a mouse monoclonal IgG1 anti-
CD20, clone MEM-97 (Abcam, Cambridge, UK). This was recommended for use in flow
cytometry applications and also reported to cross-react with bovine species by the manufacturer.
4.2.5.2.1 IHC staining to confirm B cell positivity
Immunohistochemical staining at a 1/50 dilution of bovine lymph node showed good positive
signal, which was confined to the cells of follicles contained in the outer cortex (B lymphocyte
zones).
4.2.5.2.2 CD20 antibody titration on flow cytometry
Please refer to appendix (section 9.3) for details of an attempt to use a secondary antibody with this
CD20 antibody.
Given that the monocyte population was no longer going to be assessed using flow cytometry, the
Pacific Blue fluorophore (originally allocated for monocyte staining) became available for use with
the CD20 antibody instead. The CD20 antibody was conjugated to a Pacific Blue fluorochrome
using an Apex IgG Antibody Labelling Kit (Life Technologies, Vic, Australia) according to
manufacturer’s instructions. The conjugation process diluted the CD20 antibody from 1mg/ml to a
final concentration of 160µg/ml. Titration showed the best staining at 16 µl (see figure 4.3B)
however the positive population had increased to 31.6% of events within the cells gate. The positive
population peak was not yet fully separate from the negative peak on histogram (Figure 4.3A).
However when gated by eye on the CD20/SS scatter plot (Figure 4.3B) this CD20 positive
population sat within the typical lymphocyte region.
41
Figure 4.3 A: Peripheral blood mononuclear cells - histogram of CD20 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD20 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD20 fluorescence versus SS, log scale. A population of cells positive for CD20 is visualised and gated by eye (black border). 4.2.6 Monocyte CD14 antibody selection and titration
Please refer to appendix (section 9.3) for development details. This antibody was not used in the
final protocol.
4.2.7 HIF-1α antibody selection
4.2.7.1 Initial HIF-1α antibody
A rabbit monoclonal antibody for HIF-1α (Abcam, Cambridge, United Kingdom) was selected for
the initial development of flow cytometry methods. This antibody has been used in ICC and IHC
for human HIF-1α (Kim et al., 2010, Ameri et al., 2010) and was considered likely to react with
bovine HIF-1α due to a 95% homology of deduced amino-acid sequences between human and
bovine HIF-1α (Hara et al., 1999) and its successful use in bovine tissue protocols in other studies
(Bielefeldt-Ohmann et al., 2012). A donkey anti-rabbit polyclonal IgG antibody (ab7698 Abcam)
was selected as a secondary antibody for use in indirect staining of this unlabelled primary
antibody.
This initial work demonstrated an overall issue with non-specific binding of the chosen secondary
antibody however as some of the unstimulated samples stained only with the FITC polyclonal
antibody had fluorescent signals greater than some of the stimulated samples containing both the
HIF-1α primary antibody and the secondary FITC antibody. Bovine serum albumin was employed
42
as a blocking agent to attempt to decrease non-specific binding of the secondary FITC antibody but
a detectable pattern in FITC fluorescence failed to emerge. The use of a secondary antibody was
concluded to be too problematic, especially for labelling an induced molecule of interest with no
positive control available. 4.2.7.1.1 Initial hypoxic cultures for testing use of initial HIF-1α antibody on flow cytometry
Initial hypoxic cell cultures using CoCl2 and LPS for testing of different dilutions of HIF-1α
primary antibody and secondary FITC antibody were of 4-hour, 6-hour and 8-hour duration (data
not shown). These cultures were performed to see if stabilization of HIF-1α could be detected using
flow cytometry after a short-term culture as described in one instance in literature (Vordermark and
Brown, 2003).
4.2.7.2 Alternative HIF-1α antibody
A mouse IgG1 monoclonal HIF-1α antibody (ESEE122 clone) commercially conjugated to Dylight
488 (Novus Biologicals, Littleton, CO, USA) was chosen for alternative investigation. The neat
concentration of this antibody was 1mg/ml. This HIF-1α antibody had stated cross-reactivity with
bovine cells by the manufacturer and was also listed for use in many immunostaining procedures
such as IHC, ICC and immunofluorescence, with a recommended dilution range of 1/100 to 1/5000.
No specific references of the use of this antibody in bovines were available but the manufacturer
guaranteed the reactivity. Dylight-488 is spectrally similar to FITC and so was not expected to
cause any issues of spectral overlap.
4.2.7.2.1 Hypoxic cultures for testing of alternative HIF-α antibody on flow cytometry
Two sessions of 18-hour hypoxic cell culture were undertaken using established methods (detailed
in chapter 3) including placing cells on ice post-culture with the fixation of cells in the presence of
CoCl2. The two sessions compared staining of stimulated and unstimulated cultures with HIF-1α
antibody, initially at 1µg/ml concentration and 2µg/ml dilution, as well as unstained samples of
stimulated and unstimulated cells. For each session HIF-1α antibody was incubated for a maximum
time of three hours. Every 30 minutes the incubation of an aliquot of cells was stopped and the
staining process finished for those cells in order to assess any affect of antibody incubation time on
subsequent HIF-1α antibody fluorescence measured on flow cytometry.
With increasing incubation time, both 1µg/ml and 2µg/ml staining showed a small increase in mean
HIF-1α antibody fluorescence in unstimulated cells while cells stimulated by hypoxic cell culture
showed very little increase in mean fluorescence (Figure 4.1 and 4.2). Unstained stimulated cells
had a mean fluorescence of 0.54 and a standard deviation of 0.32, and unstained, unstimulated cells
43
had a mean fluorescence of 0.82 with a standard deviation of 0.58. The mean fluorescence of the
unstained cells was therefore, quite close to stimulated stained cells which might indicate a lack of
antibody binding. However, the standard deviation of stained samples (stimulated or unstimulated)
was at least 10.5, demonstrating a broader variability of HIF-1α antibody fluorescence and the
possibility that some staining had taken place.
Figure 4.4: Graph of the mean fluorescence of HIF-1α antibody at 1µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same.
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0.5 1 1.5 2 2.5 3
Mean flu
orescence (cha
nnel)
Incuba2on 2me (hours)
HIF-‐1α an2body staining 1µg/ml
Uns+mulated cells
Cells s+mulated by hypoxic culture
44
Figure 4.5: Graph of the mean fluorescence of HIF-1α antibody at 2µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. Examination of scatter plots of HIF-1α fluorescence versus SS from all incubation lengths showed
the emergence of two distinct populations in the stimulated cells, one being a larger population with
higher SS and the other a smaller population with lower SS (Figure 4.8B). These populations would
consistently appear with stimulation but were not discernable in unstimulated cells (Figure 4.8A). It
was concluded that the channel shift of mean fluorescence of HIF-1α of only 1-2 channels, when
comparing unstimulated to stimulated cells, was a small change, and a doubtful result in terms of
real meaning.
There was, however, a subtle but consistent shift in cell population SS distribution with hypoxic cell
culture. As this subtle shift was the only repeatable response detected during flow method work-up,
and further work-up was not possible during the remaining project time-frame, it was decided to
finish the multicolour flow cytomtery protocol by combining the phenotypic antibodies with the
HIF-1α antibody for use on experimental samples, in the hopes that further information may be
gleaned from looking at cells from other individuals. No further work-up was performed, and it was
decided to use HIF-1α antibody at concentration of 2µg/ml as this equated to approximately 1/500
(the middle of the range recommended by the manufacturer).
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
0.5 1 1.5 2 2.5 3
Mean flu
orescence (cha
nnel)
Incuba2on 2me (hours)
HIF-‐1α an2body staining 2µg/ml
Uns+mulated cells
Cells s+mulated by hypoxic culture
45
Figure 4.6: A. Scatter plot of HIF-1α fluorescence versus SS, log scale. Unstimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. A single population with one central red (i.e. high density) focus is visible. B. Scatter plot of HIF-1α fluorescence versus SS, log scale. Stimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. The mean HIF-1α fluorescence was consistently lower in stimulated cells however stimulation caused a repeatable change in population shape with respect to HIF-1α and SS. A larger population with greater SS is now visible (upper large red foci) situated above a smaller population with lower SS (lower red foci). 4.2.8 Triple staining of cultured cells for flow cytometry
The process of adding antibodies to create a triple-stained sample was informed by empirical
knowledge gained from the single antibody titrations. Samples of stimulated and unstimulated cells
were cultured using the established method and then stained with a combination of 8 µl CD3, 16 µl
CD20 and HIF-1α at 2µg/ml dilution. These volumes were considered optimal based on single
antibody titration results (see previous sections), with additional samples stained singly with each
antibody and unstained samples.
The voltage, gain and discrimination values for channels FL1, FL6 and FL9 were set up based on
the values determined during single antibody titration of HIF-1α , CD3 and CD20, respectively (see
Table 4.2). These values were minimally adjusted during triple-stained sample analysis to ensure all
positively stained populations detected by each channel were on the scale covered by the
histograms and scatter plots (see table 4.2 below for original and adjusted values).
The fluorochrome combination used in the development of this multicolour experiment was chosen
in an attempt to minimize spectral overlap and the need for compensation. After removal of CD14
from the flow cytometry protocol, the fluorochromes that were ultimately used in the developed
protocol were Dylight-488 (for HIF-1α) Pacific Blue (B cell marker) and Alexa Fluor® 647 (T cell
marker). The excitation and emission wavelengths for these fluorochromes are listed in the table
46
below, as well as the signal channel used for detection, the voltage, gain and discrimination values
determined by single antibody titrations and the adjusted values after triple-stained samples were
analysed.
Table 4.1: Detectors used for detection of cell characteristics and the cytometer parameters for each detector output as set using single and then triple-stained cells. Detector Detected cell
characteristic Labelled antibody
Single voltage
Triple voltage
Single gain
Triple gain
Descrim Comp
FS Size n/a 72 72 2 2 8 No SS Complexity n/a 80 80 2 2 OFF No FL1 Dylight-488
fluorescence HIF-1α 400 420 1 1 OFF 2%
with FL9
FL6 Alexor Fluor 647 fluorescence
CD3 477 519 2 2 OFF No
FL9 Pacific Blue fluorescence
CD20 410 424 2 2 OFF 2% with FL1
Using this triple stain in stimulated cells, the T cell population had a mean fluorescence (X-mean)
of 24.61, and the B cell population had a X-mean of 13.64. Both positive populations were visible
graphically (Figure 4.4).
In a repeat experiment of the culture and triple-stain flow analysis lower amounts of CD3 and CD20
antibody were used. Triple-staining with 4µl CD3, 10µl CD20, but with identical HIF-1α antibody
concentrations, gave similar resolution of the T and B cell populations, with the X-mean of the T
cell population mean fluorescence at 23.22 and the B cell at 15.37. The unstimulated cells had a
mean HIF-1α fluorescence of 4.27 and stimulated cells had a mean HIF-1α fluorescence of 5.86.
These final amounts of antibody were used as standard for the rest of the project. Instrument
settings were saved as a triple-stain protocol for use with experimental samples.
47
Figure 4.7: Scatter plot of CD3 fluorescence and CD20 fluorescence, triple-stained stimulated cells, log scale. B and T cell populations are gated (solid black borders) with percentages of gated cells that are CD3 positive, CD20 positive or positive for both phenotypic markers.
4.2.8.1 Master Control samples
Analysis of the experimental samples (see chapter 6) was expected to require more than one session
of flow cytometry. Standardisation of such longitudinal flow studies can be facilitated by use of a
known control cell population, aliquots of which can be thawed, stained and run with each session
(Maecker and Trotter, 2006). In this case, a ‘master control’ population was created based on advice
from a consultant flow cytometrist Dr Nana Satake (personal communication, August 2011).
Five dairy cows residing at the university campus were selected based on their clinical health and
availability to donate 450ml of blood for this project. Three of the animals were clinically healthy at
the time of blood collection, while the remaining two animals had pyrexia at time of sampling and
considered immunologically challenged due to recent surgery. These animals were unfortunately
not beef cattle but their variation in health status was an attempt to account for any phenotypic
variation in the leukocytes of experimental samples due to immune status. Peripheral blood
mononuclear cells were isolated from each animal using methods previously described (see chapter
3). Aliquots of cells from each individual were cultured in both stimulated and unstimulated
conditions as previously described (see chapter 3).
The master control cells varied only slightly from the established staining protocol in that equal
aliquots of cultured cells from all five animals were mixed to create the master control population.
Immediately post-culture the samples were put on ice, followed by rapid combination and vortexing
of stimulated cells from all five animals and 250µl aliquots (representing a mix of cells from the 5
animals combined) were removed as master control samples. This process was repeated for the
unstimulated samples. Staining then proceeded as per the established protocol. Initially, master
control samples were stained with each antibody singly, in addition to a triple-stained sample and
48
an unstained sample. Single-antibody samples were included to facilitate troubleshooting of any
issues arising in the triple-stained samples of this new population being tested, however no obvious
issues arose and subsequent batches of master control cells were triple-stained only (in addition to
the usual unstained samples). All master control samples were analysed using the same triple-stain
flow protocol established for use in the experimental samples (see chapter 6).
4.3 Discussion
Both the CD3 and CD20 antibodies showed positive staining on IHC, and with distinct positive
populations on flow cytometry were considered successful titrations.
HIF-1α expression was measured by the detection of Dylight-488 fluorescence by the flow
cytometer and post-collection data analysis software was used to investigate the presence of
statistically significant trends or differences of HIF-1α across and between sample populations and
between different leukocyte phenotypes. Flow cytometry is not commonly used to measure HIF-1α
due the rapid degradation of the molecules under aerobic conditions, however use of another
technique such as western blotting would remove the phenotypic data gained through flow
cytometry.
The attempt to use a secondary FITC antibody for the HIF-1α primary antibody during flow
cytometry method development yielded confusing results that did not resolve using further
troubleshooting procedures. In this case, the use of 2% horse serum or foetal bovine serum could
have been used to block Fc receptors in an attempt to decreased non-specific binding of the
secondary FITC antibody.
It was decided that secondary antibodies were too problematic and were not used for further method
development. Indirect staining can increase fluorescence intensity as the usually polyclonal
secondary antibody binds to multiple site on the primary antibody, enhancing what might be
otherwise weak fluorescent signal (Givan, 2001). However the use of a secondary antibody
increases the time needed to complete staining and requires the use of extra controls (Givan, 2001).
A relevant disadvantage of indirect staining is that it increases the potential for non-specific staining
(Givan, 2001), which is theorised to have occurred with the secondary labelling of the initial HIF-
1α primary antibody. Indirect staining is also problematic in a multicolour experiment due to
potential cross-reactivity between other primary antibodies and the secondary antibody (Givan,
2001).
To take advantage of the amplification of fluorescent signal by an indirect method, a primary
antibody may be conjugated to biotin, followed by staining with avidin or streptavidin conjugated to
a fluorochrome (Givan, 2001) which provides a specific binding reaction. A biotinylated primary
49
HIF-1α antibody could have been considered in further investigation of a potentially weak HIF-1α
signal.
4.4 Conclusion - Final post-culture staining and flow protocol for 3-colour flow cytometry of T lymphocytes, B lymphocytes and HIF-1α
Immediately following removal from the incubator, tube cultures were placed on ice and allowed to
sit for 60 seconds to facilitate culture medium cooling before antibody incubation. After 60 seconds
the lymphocytes were incubated on ice for 20 minutes with 10 µl mouse IgG1 monoclonal antibody
to CD20 (MEM-97, Abcam, UK), which had previously undergone covalent conjugation to a
Pacific Blue fluorophore using an Apex IgG Antibody Labelling Kit. Tubes were half-raised from
ice and gently vortexed manually at the start and again at mid-incubation.
All tubes were treated with 100 µl of Medium A fixative solution, vortexed manually and allowed
to fix on ice for 20 minutes, with 100 µl fixative adequate for fixation of 1 x 106 cells (total cell
counts at this point during test tube cultures were approximately 2.5 x 105 cells). The CD20
antibody, fixative and CoCl2 were then washed from cells by addition of 3ml of flow wash
composed of 0.1% NaN3 and 3% horse serum in PBS, then centrifuged at 350 g for 5 minutes at
25°C. This temperature allowed cells time to warm in preparation for permeabilisation. The
supernatant was carefully aspirated to 250 µl and the tube gently vortexed to resuspend the cellular
pellet.
All tubes were treated with 100 µl Medium B Permeabilisation medium (Life Technologies, Vic,
Australia) and gently vortexed, then immediately incubated with 4 µl of rat anti-human IgG1
monoclonal CD3 antibody (CD3-12 clone) which was commercially conjugated to Alexa Fluor®
647 and 2 µg/ml of mouse IgG1 monoclonal HIF-1α antibody (ESEE122 clone) commercially
conjugated to Dylight 488. Cells were incubated at room temperature in the dark for 3 hours, then
were again washed, centrifuged and the resultant supernatant aspirated and the pellet resuspended
as previously described. Cells were analysed by flow cytometry using the triple-stain flow protocol
within 2 hours of staining, with refrigeration at 4°C in the dark during this time.
50
5 Development of a protocol to detect HIF-1α expression in monocytes
5.1 Introduction
Flow cytometry was found unsuitable for detecting HIF-1α expression in monocytes using the
developed cell culture protocol. An alternative technique to measure HIF-1α in monocytes was
therefore necessary.
There are a number of approaches to detection and measurement of HIF-1α stabilisation and
subsequent HIF-1 activity within cells, with several of these considered before the final technique
was chosen. HIF-1α protein levels in cell extracts are commonly detected using western blot (Wang
et al., 1995, Jiang et al., 1996, Carmeliet et al., 1998, Jin et al., 2011). An ELISA has also been
developed for measurement of HIF-1α in cultured tumour cell lines (Formento et al., 2005).
Methods have been developed to measure HIF-1 activity by the introduction of transgenes with
hypoxic response elements as promoter sequences coupled to reporter genes (Shibata et al., 2000).
DNA-binding activity of HIF-1 has been measured using electrophoretic mobility shift assays
(Wang and Semenza, 1993b). Expression of HIF-1α mRNA can be detected using quantitative
reverse transcription PCR (Dai et al., 2007) and northern blot (Wiener et al., 1996) as HIF-1α
mRNA has been shown to be upregulated under hypoxic conditions (see chapter 1). HIF-1
transcription activation can also be detected by measurement of downstream gene products such as
VEGF using a variety of the above methods.
HIF-1α expression within cells is also routinely detected using immunolabelling techniques such as
ICC and IHC (Koukourakis et al., 2002, Jin et al., 2011). ICC is the demonstration within individual
cells of a cellular constituent in situ by detecting specific antibody-antigen interactions where the
antibody has been tagged with a visible label (Sternberger, 1979), while IHC performs this
demonstration within whole tissues. ICC using direct labelling is the single layer application of a
labelled antibody to the cell antigen, however, this is less common due to insensitivity and the
requirement for directly-labelled antibodies (Noorden, 2000) and for these reasons was not used in
this project. Indirect labelling is more common and involves application of a primary unlabelled
antibody followed by a labelled secondary antibody raised to the immunoglobulin of the species
providing the primary antibody. Indirect labelling is more sensitive because this allows
accumulation of label at the original reaction site (Noorden, 2000). Commonly used labels include
fluorescent compounds, enzymes and colloidal gold (Noorden, 2000). An enzyme label is an
attached enzyme complex such as horseradish peroxidase (HRP) which activates a subsequently
applied chromagen substrate, producing a coloured reaction at the antigen site that is visible with
light microscopy (Fetsch, 2005).
51
5.2 Materials, methods and results
5.2.1 HIF-1α antibody selection for ICC
A rabbit monoclonal antibody for HIF-1α (Abcam, Cambridge, United Kingdom) was selected.
This antibody has been used for ICC and IHC for human HIF-1α (Kim et al., 2010, Ameri et al.,
2010) and was considered likely to react with bovine HIF-1α due to a 95% homology of deduced
amino-acid sequences between human and bovine HIF-1α (Hara et al., 1999). This antibody has
worked in several different test-modes including IHC and Western Blot, and has been used
previously for IHC on bovine tissue (Bielefeldt-Ohmann et al., 2012). A previous study using
another rabbit polyclonal HIF-1α antibody featuring the same immunogen and human reactivity has
also shown reactivity with bovine HIF-1α (Bielefeldt-Ohmann et al., 2008).
5.2.2 ICC protocol for HIF-1α expression in monocytes
5.2.2.1 Fixation of monocytes
Monocytes fixed in 10% buffered formalin appeared to have overall better cell morphology than
samples fixed in acetone when several slides using each of these fixative solutions were examined
under light microscopy. As previously discussed in chapter 4, the cultured monocytes were
subjected to ICC in-situ after culture. The immunocytochemical protocol was based on use of the
EnVision+ System-HRP rabbit kit (DAKO, Carpentaria, USA).
5.2.2.2 Target antigen retrieval, anti-cross-contamination measures and blocking steps
Prior to ICC, slides were examined by light microscopy to confirm the presence of cells.
The formalin-fixed slides were then immersed in Target Retrieval Solution pH 9.0 (DAKO) at 95°C
for 25 minutes, for target antigen retrieval. Target retrieval using citrate pH 6.0 (DAKO) and
proteinase K Ready-to-use (DAKO) were also tested but resulted in no detectable signal. Slides
were allowed to cool for 20 minutes while still immersed in target retrieval solution, then were
thoroughly rinsed with deionised water.
During harvest and monocyte fixation the detachable walls were removed from the chamber slides.
Doing this without disturbing the gasket for ICC proved, in reality, very difficult to achieve
consistently, even when following the instructions for removal. Using test slides it was quickly
discovered that although a gasket appeared well adhered to the slide, application of a stained
solution (e.g. Methyl Green) to an individual well indicated that occasional points of gasket ‘micro-
detachments’ existed (invisible to gross inspection) that allowed reagents to diffuse from one well
into an adjacent well. Gaskets were subsequently discarded from every slide and each of the 4 wells
per slide was carefully drawn around with a delimiting pen (DAKO) creating a water-repellent
52
barrier between wells. During this process slides were immersed or covered in deionised water to
prevent drying out.
Slides were then treated with 0.35% hydrogen peroxide in water-solution for 15 minutes to block
endogenous peroxidase activity and rinsed with a solution of Tris-buffered-saline with Tween 20
(TBST) at a concentration of 0.05 mol/L Tris-HCl and 0.3 mol/L NaCl, with 0.1% Tween 20 and
0.01% preservative (DAKO). Slides were then treated with glycine (glycine binds to and blocks
aldehydes left over from the fixation step) for 15 minutes, and then rinsed briefly with TBST.
5.2.2.3 Primary HIF-1α antibody application
Antibody Diluent solution (DAKO) was applied for 30 minutes. Excess antibody diluent was then
tapped off, followed by application of 150µl of the primary anti-HIF-1α antibody at a titrated 1/100
dilution (diluted in the same Antibody Diluent solution) to wells 3 and 4 only (well 3 containing
cells from the unstimulated culture and well 4 containing cells stimulated by CoCl2 – see Figure
5.1). The primary antibody was incubated for 2 hours. The cell populations of wells 1 and 2
mirrored the populations in wells 3 and 4 (Figure 5.1) but did not receive primary HIF-1α to serve
as negative primary controls for detection of any non-specific staining of the secondary antibody.
These upper 2 wells were deliberately assigned as secondary controls to avoid any contact with
primary antibody during vertical rinsing steps (direction of rinse indicated by the arrow in Figure
5.1). After primary antibody incubation, all wells were rinsed 5 times with TBST.
Figure 5.1: Post-culture in situ ICC of adherent monocytes. Secondary controls Wells 1 and 2 contained unstimulated and stimulated cells, respectively, and did not receive primary antibody. The lower wells 3 and 4 contained duplicate cell populations to 1 and 2 and were treated with primary HIF-1α antibody.
53
5.2.2.4 Secondary antibody application
Real Envision HRP goat anti-rabbit immunoglobulin (DAKO) was applied to all wells including
controls, and incubated for 30 minutes followed by repeated rinsing with TBST. Slides were
incubated with 3-amino-9-ethylcarbazole (AEC+) high sensitivity, ready-to-use substrate
chromagen (DAKO) for 15 minutes duration for development of a detectable signal.
3,3'-Diaminobenzidine (DAB) diluted 1:50 in substrate buffer (DAKO) was tested as a possible
substrate chromagen (due to a greater stability than AEC+), however there were consistently higher
levels of background staining and stain precipitate which interfered with production of a distinct
positive signal. Excess substrate was gently tapped off, immediately followed by application of a
methyl green counterstain for 1 minute before a final rinse with deionised water. Methyl green
consistently appeared to contrast better with AEC than Mayer’s Haematoxylin as it improved
visibility of any HIF-1α signal within the nucleus.
5.2.2.5 Slide mounting and slide examination
Slides were quickly coverslipped with Faramount aqueous mounting medium (DAKO) and left to
dry overnight before examination by light microscopy. The primary controls showed no staining
that would have indicated non-specific binding of the secondary antibody (results not shown) and
background staining was minimal.
5.3 Discussion
The primary HIF-1α antibody used has been previously used in IHC on bovine cells (Bielefeldt-
Ohmann et al., 2012) and has been previously successfully implemented using isotype controls (H.
Bielefeldt-Ohmann, personal communication, November 2012). Other controls to consider for
future work include cells from a knock-out animal for HIF- 1α stained with the primary HIF-1α
antibody, on an absorption control which involves the incubation of the primary antibody with the
antigen used to generate the antibody such that the absorbed antibody can no longer bind to the
antigen present on the slide (Burry, 2011). Given more time and funds, the work-up of such
additional absorption controls for this antibody for use in ICC would further help strengthen the
results of this protocol.
The major issue encountered with this protocol was unreliability of the slide gaskets, requiring
extensive use of a delimiter pen to recreate water-proof barriers between the cells. Although time-
consuming, creation of this barrier was vital to ensure no cross contamination of reagents, given the
close physical proximity of the positive and control wells, while facilitating uniform ICC staining
and a reduction in required reagent volume.
54
The most apparent disadvantages of ICC are that this method produces what can be still only be
considered semi-quantitative results, even when performed with the application of morphometric
measurements (Oberholzer et al., 1987, Hershkovitz et al., 2013). The method can provide
information as to the presence of absence of HIF-1α, but can be argued as relatively insensitive for
distinguishing between any subtle differences in the level of HIF-1α expression between different
animals or culture conditions. There is potential for batch variation when performing the same ICC
multiple times and proper assessment relies on the experience of the slide reader. However using
ICC allows the direct identification and examination of cells expressing HIF-1α. In the context of
this project, this technique was considered additionally useful for specifically detecting intranuclear
HIF-1α due to nuclear translocation occurring subsequent to intracytoplasmic stabilisation in
response to hypoxia. Positive intranuclear HIF-1α can be interpreted as evidence of a more
advanced HIF-1α response occurring in these cells compared to negative or cells with
intracytoplasmic staining only (see section1.2.2.1). ICC was considered a good choice for this
project given its common use for HIF-1α, the prior experience gained in immunostaining (for flow
cytometry) during this project, easy access to indirect enzymatic commercial ICC kits, and a readily
available ICC protocol provided by H. Bielefeldt-Ohmann (personal communication, June 2010). In
addition to these reasons, ICC was ultimately chosen over other potential methods due to prior ICC
experience of this researcher, as a practical way of addressing remaining time-constraints of the
project. Western blot was also considered as an alternative method due to available supervisor
expertise and facilities. Western blot was not used as the large numbers of cells required for
adequate protein yield would be problematic due to the low total cell numbers in many of the
intended experimental samples (see 6.2.1), as well as the low retrievable cell population proportion
and low total cell counts of harvested hypoxic cultures.
5.4 Conclusion
A protocol for detecting HIF-1α in bovine monocytes was successfully developed as an alternative
to measurement of HIF-1α in monocytes using flow cytometry.
55
6 Methodology application to assess HIF-1α expression as a biomarker of bovine respiratory disease
6.1 Introduction
BRD is a multi-factorial inflammatory respiratory disease complex. Due to the high incidence of
pneumonia in cattle, investigations are currently focused on ways to enhance effective and non-
injurious immune responses to these pathogens (Ackermann et al., 2010a), in addition to addressing
the environmental and managerial issues that may predispose cattle to respiratory disease. A way to
enhance effective immune responses is to identify candidate molecules that are expressed
differently between animals that have an effective response to an infectious challenge and other
animals that develop disease. A consistent difference in expression means this molecule could be
involved in and also act as biomarker for a particular immune response and disease outcome. For
BRD, identification of a biomarker would allow potential targeted breeding programs to enhance
resistance to this complex disease.
There have been previous studies investigating possible biomarkers for BRD outcome (Aich et al.,
2009, Eitam et al.). Aich et al 2009 established the value of using combinatorial “omics”
approaches to identify candidate biomarkers and used multimodal analysis of these biomarkers to
predict disease outcome. Eitam et al 2010 aimed to detect individual variations in the stress
responses of 13 newly received young calves through measurement of their leukocyte heat shock
protein response, selected neutrophil-related gene expression and oxidative stress, and relate these
variations to pulmonary adhesions at slaughter as an indicative sign of clinical and subclinical
episodes of BRD at an early age.
The increased production of proinflammatory cytokines that occurs during a primary BHV-1
infection (Bielefeldt Ohmann et al., 1991) suggests a potential role of a HIF-1α positive feedback
loop in the subsequent development of BRD. Alternatively, a sluggish HIF-1α response could
decrease the hypoxia-triggered HIF-1α adenosinergic suppression of proinflammatory cytokines in
CD4+ and CD8+ lymphocytes, an important immunosuppressive mechanism which down-regulates
the immune response to protect local tissue and maximize anti-pathogen cellular response during
pathogen encounter (Sitkovsky and Lukashev, 2005, Zinkernagel et al., 2007, Sitkovsky, 2009).
If an association between HIF-1α levels and case/control status existed and this association was also
present prior to cases actually developing BRD, it might be possible to use HIF-1α as a biomarker
for BRD susceptibility. This information could be used to identify susceptible animals prior to
56
developing clinical disease and assist in the creation of targeted breeding programs to reduce the
incidence of BRD.
The aims of this study were to investigate the value of measuring the HIF-1α expression response
as a predictive biomarker for the outcome of BRD by implementing developed methodologies
experimentally to compare the HIF-1α expression response in a group of clinically healthy beef
feedlot cattle with the HIF-1α expression in feedlot cattle that developed and were treated for
clinical respiratory disease while in the feedlot. The application also tested how practical and user-
friendly the developed methodology is in practice.
6.2 Materials and methods
6.2.1 Study population and animal selection
This study was conducted on two small subsets of animals enrolled in a parent project aimed at
identifying and quantifying risk factors for BRD in feedlot cattle. Blood samples were collected
from cattle in the parent project on the day of induction at the feedlot and after approximately 42
days on feed, induction being the process of tagging, weighing, treating and entering their data
(animal identifiers) into the feedlot computer system. Subset one was selected prospectively from
November 2011 to March 2012. Animals eligible for inclusion were slaughtered at a combined
feedlot/abattoir facility in South East Queensland from November 2011 to March 2012, and came
from five cohorts (pens of cattle assembled at induction to the feedlot and kept together during time
on feed) from two of the 14 feedlots participating in the parent study. Using the following methods,
individual animals were then selected from the eligible cohorts. Cattle slaughtered during the study
period were eligible for selection as cases if they had been pulled from the cohort and treated for
respiratory disease by feedlot staff (treatment records sourced from feedlot database). In contrast,
cattle were eligible for selection as controls if they had not been pulled and treated for respiratory
disease whilst on feed. Eligibility for inclusion also depended on blood samples taken at both
induction and approximately 42 days on feed being of sufficient quality and quantity for serological
analyses.
As a maximum of 15 samples could be processed on any single collection day, controls were
selected from the eligible list using numbers randomly generated by Stata 12.1 (StataCorp LP,
Texas, USA). For most cohorts the selection process used for controls was also applied to cases,
although some cohorts had low case numbers so all eligible cases were sampled. The overall
case:control ratio in subset one was approximately 1:2, although within individual cohorts this ratio
varied considerably due to the variable number of case animals in some cohorts. The overall sample
57
size was small due to time, budget and labour constraints. The project was considered to be a pilot
project for these reasons.
The second subset (subset two) comprised 42 animals previously sampled by another researcher
also working within the parent project (Mr. Andrew Ferguson, School of Veterinary Science, UQ,
Australia). These cattle were slaughtered at the same facility from May to July 2010 and were
sourced from one of the two feedlots included in the prospective sampling. Cattle were eligible as
cases if they had been pulled and treated for respiratory disease, and showed an increase in
serological status of 1 or more to BHV-1, BRSV, BVDV or PI3 virus. Controls were animals that
were not pulled but showed an increase in serological status of 1 or more to BHV-1, BRSV, BVDV
or PI3 virus. Cases in subset two varied from the cases in subset one in that this serological status
information was part of case eligibility. Samples from subset two contained fewer cells and were
only used in the ICC analyses.
6.2.2 Baseline data
Baseline data for each selected animal in both subsets was sourced from participating feedlots to
check for possible confounding variables. These data included each animal’s cohort, breed, gender,
dentition and induction weight. Serology results, using the BIO-K 284 multiplex-ELISA (Bio-X
Diagnostics) for antibodies to BHV-1, BRSV, BVDV, PI3 virus and M. bovis for induction and day
42 samples, were also made available by the parent project. Serological results were expressed on a
0-5 scale, with higher numbers indicative of higher antibody titre.
Within each subset of animals, Fisher’s exact test was used to compare the distribution of
categorical variables between case and control populations and the Mann-Whitney U test was used
to compare induction weights, as the distribution was not normal.
6.2.3 Blood sample collection
Blood samples were obtained from selected cattle immediately post mortem. Post mortem
collection was used due to the large amount of blood required and the need to avoid additional
stress to the animals in the immediate pre-slaughter period. Blood collection was undertaken by
trained abattoir workers under the direct supervision of the primary researcher. Approximately 500
ml of whole blood was collected via gravity from each animal within two minutes of initial jugular
cut; blood samples were thus non-sterile. Blood was collected into sterile, pre-chilled 500 ml
containers (Interpath, Victoria, Australia) containing 50 ml of 6 % sodium citrate tribasic dihydrate
58
(Sigma Alrich, St Louis MO, USA) in deionised, sterile water. Samples were immediately chilled
and transported for 1.5 hours to the lab for processing to purify mononuclear leukocytes.
6.2.4 Mononuclear cell isolation from experimental samples
Refer to section 3.2.2 for this previously described protocol. This procedure was only undertaken
within the confines of this project on the 46 experimental samples of subset one.
6.2.5 Cell culture and post-culture harvest of experimental samples
Refer to section 3.4 for this previously described protocol. Stimulated cultures contained CoCl2 and
LPS and unstimulated cultures did not. A stimulated and unstimulated tube culture was set up for
each animal from subset one. Chamber slide culture was performed on all experimental samples (88
individual animals), with two chamber slides containing stimulated and stimulated wells cultured
for each animal (see Figure 5.1 in chapter 5).
6.2.6 Flow cytometry protocol
Refer to Chapter 4 for previously described protocols for cell staining and flow cytometry.
A stimulated and unstimulated sample was run for each animal. At the start of each session, the
cytometer was calibrated using fluorescent beads as sold and recommended by the manufacturer.
A ‘master control’ sample comprised of leukocytes isolated from five different adult dairy cows
located on the UQ Gatton Campus was also run during each of the flow cytometry sessions to detect
any technical machine-related problems or other session-specific issues causing differences that
would have an impact on the ability to compare results from different sessions during flow
cytometry studies. The use of cells from multiple individuals that were either clinically well or
showing clinical disease was to cover the anticipated phenotypic range of immune cells to be
encountered in the experiment (refer to Chapter 4 for further details). These five animals were
selected based on their clinical health and availability to donate 450ml of blood. Three of the
animals were clinically healthy at the time of blood collection, while the remaining two animals had
pyrexia at time of sampling and considered immunologically challenged due to recent surgery.
6.2.6.1 Gating of cell populations
Analysis software (Kaluza, Beckman Coulter) was used to gate cell populations for stimulated and
unstimulated samples from each animal. The first gate (“Cells”) was applied to the forward
scatter/side scatter plot to select the cell population (Figure 6.1A), excluding debris, stain
precipitate and large events (presumed aggregates of cells). Only events included in the “Cells” gate
were included in subsequent gating procedures. The next scatter plot was generated with the axes
59
CD 3 (T lymphocyte) versus CD20 (B lymphocyte) fluorescent signal (Figure 6.1B). The labelled
“B lymphocytes” and “T lymphocytes” populations were gated individually. For each of the gated
B and T lymphocyte populations, expression of HIF-1α fluorescent signal was then visualized as a
histogram (Figure 6.1C).
Figure 6.1 A: An example of the forward scatter/side scatter plot obtained from Kaluza. Events included in cell population lie within the gate “Cells” comprising a solid black line. This gate excludes debris and aggregate cells. B: An example of a T lymphocyte fluorescent marker/B lymphocyte fluorescent marker scatter plot obtained from Kaluza, based on the events included in the “Cells” gate. “T cells” and “B cells” populations are both gated (solid black lines). C: An example of a histogram of HIF-1α fluorescence detected within the T cell gate on flow cytometry with Kaluza-generated summary statistics. 6.2.6.2 Flow data analysis methods
The difference in HIF-1α fluorescence between stimulated and unstimulated samples in case animal
T and B lymphocyte populations was compared with that of the controls. For each sample
60
(stimulated and unstimulated for each animal) the HIF-1α signal from the B lymphocyte population
and the T lymphocyte population was summarized by the sample mean, median and standard
deviation as provided by the Kaluza flow cytometry analysis software. These summary statistics
were selected to assess central tendency and spread of the HIF-1α fluorescence of each cell
population, and were exported from Kaluza into Stata 12.1 for further analysis. The difference
between the stimulated and unstimulated values was then calculated for each of these summary
statistics for each animal. This resulted in 18 summary statistics for each animal (stimulated B cell
mean, median and standard deviation, unstimulated B cell mean, median and standard deviation,
difference between stimulated B cell and unstimulated B cell mean, median and standard deviation,
stimulated T cell mean, median and standard deviation, unstimulated T cell mean, median and
standard deviation, difference between stimulated T cell and unstimulated T cell mean, median and
standard deviation). The animals were grouped by case/control status and the group mean, median
and standard deviation of each of these summary statistics was calculated for all the stimulated and
all the unstimulated samples in the case group and again for the control group. The distribution of
the differences in sample mean, median and standard deviation within case and control groups were
roughly normally distributed and generally the distributions of the case and control groups were of
equal variance. These differences were then compared across the case and control groups using a 2-
sample t-test assuming equal variance, except for the standard deviation of the B lymphocyte
populations, where a 2-sample t-test not assuming variance was used.
6.2.7 ICC for monocyte expression of HIF-1α
Refer to Chapter 5 for this previously described protocol.
6.2.7.1 ICC slide reading
ICC slides were read by this project’s primary supervisor and experienced IHC expert, Dr Helle
Bielefeldt-Ohmann. At least 200 cells were counted on each slide and each cell was assigned to one
of three categories. These categories were 1) no HIF-1α signal, 2) cytoplasmic HIF-1α signal only
and 3) cytoplasmic and nuclear HIF-1α signal.
6.2.7.2 ICC slide reading data analysis methods
The ICC results from the two subsets of cattle were analysed separately (see section 6.3.4) using
Stata 12.1. For each animal, the percentage of cells in each HIF-1α signal category was calculated
for both the unstimulated and stimulated culture, after which the absolute difference between these
two percentages was calculated. The percentage differences for the case group were compared to
61
those for the control group using a t-test not assuming equal variance for subset one, and a t-test
assuming equal variance for the subset two.
6.3 Results
6.3.1 Mononuclear cell isolation
From subset one, blood was collected from a total of 46 animals – 19 cases and 27 controls.
Peripheral blood mononuclear cell yields varied with total counts ranging from 1.7 x 108 to 2.17 x
109 cells isolated from 200 ml blood processed during the three sessions. Based on this range, the
amount of whole blood processed per animal was reduced to 100ml thereafter to expedite
processing while still isolating ample numbers of cells for culture, as the experimental design only
required 2.5 x106 cells for tube cell cultures and 2 x 107 cells for chamber cell cultures per animal.
In subset two, mononuclear leukocytes were isolated from 15 case animals and 27 control animals.
The low cellularity of the subset two samples meant these 42 animals were only included in the
chamber cultures for HIF-1α ICC.
6.3.2 Comparison of baseline data
Baseline data for the case and control animals is presented in table 6.1. There was no significant
difference in cohort, breed, gender, dentition or serological categories between cases and controls in
either of the two subsets of animals (table 6.1). Case animals were significantly lighter than controls
in the subset one (controls: mean (sd) 451 (30.7) kg; cases: mean (sd) 431 (21.7) kg (p= 0.028).
However, there was no significant difference in induction weight between cases and controls for
subset two (controls: mean (sd) 467 (27.1) kg; cases: mean (sd) 465 (33.4) kg (p= 0.753).
62
Table 6.1: Baseline data for cases and controls for both subsets
Variable
First Subset Second Subset Case
numbers (%)
Control numbers
(%) P-Value
Case numbers
(%)
Control numbers
(%) P-Value Cohort
0.7
1.00
Cohort 1 6 (31.6) 4 (14.8)
N/A N/A Cohort 2 5 (26.3) 8 (29.6)
N/A N/A
Cohort 3 1 (5.2) 2(7.4)
N/A N/A Cohort 4 6 (31.6) 9 (33.3)
N/A N/A
Cohort 5 1 (5.2) 4 (14.8)
N/A N/A Cohort 6 N/A N/A
7 (46.7) 12 (44.4)
Cohort 7 N/A N/A
8 (53.3) 15 (55.5) Breed
0.19
0.45
Angus 11 (57.9) 19 (70.4)
8 (53.3) 10 (37.0) Angus/Hereford 1 (5.2) 0 (0.0)
0 (0.0) 3 (11.1)
Angus/Santa Gertrudis 1 (5.2) 0 (0.0)
1 (6.6) 4 (14.8) Angus/Shorthorn 0 (0.0) 0 (0.0)
1 (6.6) 3 (11.1)
Brangus 1 (5.2) 0 (0.0)
0 (0.0) 0 (0.0) Charbray 1 (5.2) 1 (3.7)
0 (0.0) 0 (0.0)
Droughtmaster 0 (0.0) 1 (3.7)
0 (0.0) 0 (0.0) Hereford 4 (21.1) 2 (7.4)
4 (26.7) 2 (7.4)
Hereford/Santa Getrudis 0 (0.0) 1 (3.7)
0 (0.0) 1 (3.7) Santa Gertrudis 0 (0.0) 3 (11.1)
1 (6.6) 4 (14.8)
Gender Male 19 (100.0) 27 (100.0)
15 (100.0) 27 (100.0)
Dentition
0.25
0.84 0 14 (73.6) 23 (85.2)
11 (73.3) 22 (81.5)
2 5 (10.5) 3 (11.1)
3 (20.0) 3 (11.1) 4 0 (0.0) 1 (3.7)
1 (6.7) 2 (7.4)
Induction BHV-1 Titre
0.83
1.00 0 15 (79.0) 22 (81.5)
8 (53.3) 15 (55.6)
1 2 (10.5) 2 (7.4)
6 (40.0) 9 (33.3) 2 0 (0.0) 2 (7.4)
0 (0.0) 1 (3.7)
3 1 (5.3) 1 (3.7)
1 (6.7) 1 (3.7) 4 0 (0.0) 0 (0.0)
0 (0.0) 1 (3.7)
5 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0) Missing 1 (5.3) 0 (0.0)
0 (0.0) 0 (0.0)
Day 42 BHV-1 Titre
0.58
0.17 0 1 (5.3) 5 (18.5)
0 (0.0) 4 (14.8)
1 3 (15.8) 3 (11.1)
4 (26.7) 4 (14.8) 2 2 (10.5) 2 (7.4)
3 (20.0) 5 (18.5)
3 12 (63.2) 12 (44.4)
8 (53.3) 9 (33.3) 4 0 (0.0) 1 (3.7)
0 (0.0) 5 (18.5)
5 1 (5.3) 4 (14.8)
0 (0.0) 0 (0.0) Missing 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0)
63
Table 6.1: Baseline data for cases and controls for both subsets (continued)
Variable
First Subset Second Subset Case
numbers (%)
Control numbers
(%) P-Value
Case numbers
(%)
Control numbers
(%) P-Value Induction BVDV Titre
0.36
0.78
0 2 (10.5) 6 (22.2)
10 (66.7) 19 (70.4) 1 0 (0.0) 0 (0.0)
2 (13.3) 1 (3.7)
2 2 (10.5) 0 (0.0)
0 (0.0) 0 (0.0) 3 2 (10.5) 2 (7.4)
0 (0.0) 1 (3.7)
4 7 (36.8) 8 (29.6)
3 (20.0) 6 (22.2) 5 5 (26.3) 11 (40.7)
0 (0.0) 0 (0.0)
Missing 1 (5.3) 0 (0.0)
0 (0.0) 0 (0.0) Day 42 BVDV Titre
0.83
0.85
0 0 (0.0) 1 (3.7)
0 (0.0) 0 (0.0) 1 0 (0.0) 0 (0.0)
1 (6.7) 1 (3.7)
2 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0) 3 3 (15.9) 3 (11.1)
2 (13.3) 6 (22.2)
4 8 (42.1) 9 (33.3)
12 (80.0) 20 (74.1) 5 8 (42.1) 14 (51.9)
0 (0.0) 0 (0.0)
Missing 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0) Induction BSRV Titre
0.21
0.28
0 2 (10.5) 1 (3.7)
4 (26.7) 10 (37.0) 1 8 (42.1) 7 (25.9)
1 (6.7) 7 (25.9)
2 0 (0.0) 5 (18.5)
4 (26.7) 3 (11.1) 3 4 (21.1) 8 (29.6)
5 (33.3) 4 (14.8)
4 4 (21.1) 4 (14.8)
1 (6.7) 3 (11.1) 5 0 (0.0) 2 (7.4)
0 (0.0) 0 (0.0)
Missing 1 (5.3) 0 (0.0)
0 (0.0) 0 (0.0) Day 42 BSRV Titre
0.95
0.15
0 0 (0.0) 1 (3.7)
0 (0.0) 0 (0.0) 1 1 (5.3) 1 (3.7)
0 (0.0) 0 (0.0)
2 2 (10.5) 4 (14.8)
1 (6.7) 5 (18.5) 3 5 (26.3) 4 (14.8)
6 (40.0) 16 (59.3)
4 8 (42.1) 11 (40.7)
8 (53.3) 6 (22.2) 5 3 (15.8) 6 (22.2)
0 (0.0) 0 (0.0)
Missing 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0) Induction PI3 Titre
0.19
0.25
0 3 (15.8) 0 (0.0)
1 (6.7) 0 (0.0) 1 3 (15.8) 2 (7.41)
0 (0.0) 5 (18.5)
2 3 (15.8) 4 (14.8)
5 (33.3) 4 (14.8) 3 2 (10.5) 7 (25.9)
4 (26.7) 10 (37.0)
4 7 (36.8) 13 (48.2)
4 (26.7) 6 (22.2) 5 0 (0.0) 1 (3.7)
1 (6.7) 2 (7.4)
Missing 1 (5.3) 0 (0.0)
0 (0.0) 0 (0.0)
64
Table 6.1: Baseline data for cases and controls for both subsets (continued)
Variable
First Subset Second Subset Case
numbers (%)
Control numbers
(%) P-Value
Case numbers
(%)
Control numbers
(%) P-Value Day 42 PI3 Titre
0.79
0.77
0 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0) 1 1 (5.3) 1 (3.7)
0 (0.0) 3 (11.1)
2 2 (10.5) 2 (7.4)
2 (13.3) 4 (14.8) 3 5 (26.3) 7 (25.9)
3 (20.0) 6 (22.2)
4 10 (52.6) 12 (44.4)
10 (66.7) 13 (48.2) 5 1 (5.3) 5 (18.5)
0 (0.0) 1 (3.7)
Missing 0 (0.0) 0 (0.0)
0 (0.0) 0 (0.0)
6.3.3 Flow cytometry results
The mean and standard deviation of the HIF-1α fluorescent signal of B and T cells from stimulated
and unstimulated cultures for the master control samples from each session were compared, with
session one being different from the next two sessions. For example, sessions two and three resulted
in similar stimulated master control HIF-1α T-cell fluorescence, with a mean (sd) of 2.91 (1.83) and
2.94 (1.86) for these sessions, respectively. However, the session one stimulated master control
HIF-1α T-cell fluorescence mean (sd) was 3.43 (2.32). Session one samples contained larger
amounts of non-specific debris than previously encountered, and this along with the difference in
the master controls suggests that there were some differences in sample processing during this first
session. These differences may have arisen during the culture or subsequent flow procedures. The
fifteen animals from session one of flow cytometry were therefore excluded from further analysis.
Summarized analysis results of the remaining 31 animals from session two and session three are
shown in Table 6.2.
65
Table 6.2: Descriptive statistics for the absolute difference in mean, median and standard deviation of HIF-1α in B cells and T cells between unstimulated and stimulated cultures for case group and control groups.
Cell Type - Summary statistic of HIF-1α fluorescence
First Subset - Sessions two and three Number of
animals Mean Median SD P-Value1 B Cells - Mean
0.97
Difference Cases 16 -0.39 -0.42 0.42 Difference Controls 15 -0.39 -0.35 0.36 B Cells - Median
0.66
Difference Cases 16 -0.42 -0.34 0.44 Difference Controls 15 -0.35 -0.40 0.39 B Cells - Standard Deviation 15
0.162
Difference Cases 16 0.63 -0.05 2.84 Difference Controls 15 -0.43 -0.33 0.70 T Cells - Mean
0.50
Difference Cases 16 -0.44 -0.29 1.02 Difference Controls 15 -0.67 -0.64 0.81 T Cells - Median
0.76
Difference Cases 16 -0.44 -0.40 0.83 Difference Controls 15 -0.51 -0.50 0.27 T Cells - Standard Deviation
0.30
Difference Cases 16 -0.21 -0.07 0.88 Difference Controls 15 -0.58 -0.49 1.06
1 The p-value refers to the result of a 2-sample t-test comparing the difference between unstimulated and stimulated cultures of the case group to those of control group. 2 A t-test for unequal variances was used. 6.3.3.1 Statistical analysis of flow cytometry results
The difference in HIF-1α expression between simulated and unstimulated populations was not
significantly different between cases and controls when the mean, median and standard deviation of
the HIF-1α fluorescence for each of the B and T lymphocyte populations were compared (Table 6.2
above). There was also no significant difference present when data were reanalyzed including
results from the session one animals (data not shown).
6.3.4 ICC results
Six animals (all from subset one) were removed from ICC analysis due to concerns about the
quality of ICC staining for those animals. ICC of slides made from unstimulated cultures of subset
two had consistently greater numbers of cells with positive staining for HIF-1α than the
unstimulated cells of subset one. Due to this difference in baseline activation despite lack of
stimulation, the two subsets were analysed separately (Table 6.3). There was no staining present in
66
the control wells that had not received primary antibody, whereas some staining was detected in all
stimulated (Figure 6.2) and all unstimulated cultures immunolabeled for HIF-1α.
Table 6.3: ICC results for case and control groups from both subsets. The mean, median and standard deviation of the percentage of cells displaying cytoplasmic or combined cytoplasmic and nuclear HIF-1α positive ICC staining in stimulated and unstimulated cell cultures, as well as for the percentage difference of these two types of HIF-1α positive staining between stimulated and unstimulated cultures.
Cell culture status-Case/control status
First Subset Second Subset Number
of animals Mean Median SD Number of
animals Mean Median SD % Cells with Cytoplasmic HIF-1α signal
Unstimulated Cases 15 3.68 3.54 2.85 16 4.13 3.75 3.47 Unstimulated Controls 27 3.28 2.50 2.51 23 6.39 5.00 6.17 Stimulated Cases 14 3.00 2.75 2.14 16 3.97 3.25 3.19 Stimulated Controls 27 3.11 2.50 2.66 24 4.71 3.50 5.31 Difference Cases 14 -0.87 0.00 2.78 16 -0.16 0.00 3.45 Difference Controls 27 -0.17 -0.12 2.99 23 -1.50 1.00 7.63 % Cells with Cytoplasmic and Nuclear HIF-1α signal
Unstimulated Cases 15 24.80 28.00 19.07 16 9.50 0.75 16.50 Unstimulated Controls 27 21.69 17.00 18.20 23 3.02 1.00 4.74 Stimulated Cases 14 27.61 32.00 19.35 16 8.09 4.00 9.62 Stimulated Controls 27 26.93 29.00 18.37 24 3.63 2.50 3.49 Difference Cases 14 3.18 -0.28 13.15 16 -1.41 0.50 8.62 Difference Controls 27 5.24 5.50 13.28 23 0.72 1.00 4.05
Figure 6.2: ICC of stimulated cells showing several monocytes with scattered intracytoplasmic HIF-1α expression (arrow) and a central, activated monocyte with strong intracytoplasmic and intranuclear HIF-1α signal.
67
6.3.4.1 Statistical analysis of ICC results
For neither subset of animals was there any significant difference in the percentage of HIF-1α
positive cells (either cytoplasmic ‘cytoplasmic positive’ or cytoplasmic/nuclear ‘nuclear positive’)
between the stimulated and unstimulated cultures when the results of case group was compared to
the results from the control group. Subset one had a p-value = 0.47 for difference in HIF-1α
cytoplasmic positive staining and p-value = 0.37 for difference in HIF-1α nuclear positive staining,
and the second subset had a p-value = 0.47 for difference in HIF-1α cytoplasmic positive staining
and a p-value = 0.64 for difference in HIF-1α nuclear positive staining (Figure 6.4 A-D).
68
Figure 6.3. A: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. B: Percentage difference in ICC HIF-αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. C: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. D: Percentage difference in ICC HIF-1αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two.
0.0
5.1
.15
-40 -20 0 20 40 -40 -20 0 20 40
Controls Cases
Den
sity
Percentage difference of cytoplasmic positive cells
A
0.0
5.1
.15
-40 -20 0 20 40 -40 -20 0 20 40
Controls Cases
Den
sity
Percentage difference of cytoplasmic positive cells
C
0.0
5.1
.15
-40 -20 0 20 40 -40 -20 0 20 40
Controls Cases
Den
sity
Percentage difference of nuclear positive cells
B
0.0
5.1
.15
-40 -20 0 20 40 -40 -20 0 20 40
Controls Cases
Den
sity
Percentage difference of nuclear positive cells
D
69
6.4 Discussion
The cell culture, flow cytometry and ICC protocols described in previous chapters were
successfully applied to the test samples in this experiment. Using these protocols, the results
showed no significant difference in HIF-1α expression between stimulated and unstimulated bovine
leukocytes of animals that were treated for clinical BRD, when compared to that of control animals.
This outcome does not refute the null hypothesis that there is no significant difference in HIF-1α
regulation in bovine leukocytes of animals with clinical BRD compared with other clinically
healthy animals. These results do not support the idea of a direct effect of increased or decreased
HIF-1α expression in macrophages and lymphocytes contributing to the development of BRD
through cytokine production by upregulation or lack of suppression, and so do not support the
experimental hypothesis. The lack of a significant difference in HIF-1α expression between case
and control groups does not support the potential use of HIF-1α as a biomarker for BRD.
The lack of a detected difference in HIF-1α does not refute the theory of an overactive or
underactive immune response involving HIF-1α contributing to the development of clinical BRD.
This experiment did not expose any bovine cells to actual viral pathogens to account for any
possible HIF-1α modulation of the primary viral infection, in order to assess how this interaction
may alter HIF-1α expression and therefore any role HIF-1α may ultimately play in the pathogenesis
of BRD. Viral agents have a variety of mechanisms by which to utilise HIF-1α during infection.
Viruses can undergo HIF-1α –dependent or HIF-1α-enhanced viral replication (Davis et al., 2001,
Haque et al., 2003). VSV can induce HIF-1α stabilisation in bronchial cells, potentially contributing
to the clinical airway oedema of acute VSV infection (Kilani et al., 2004). Other viruses also induce
HIF-1α stabilisation (see 1.2.3.3.2). In some viruses, viral replication is inhibited by HIF-1α (Pipiya
et al., 2005, Hwang et al., 2006). Following BHV-1 infection, innate immune responses include the
antiviral action of IFN and the recruitment of lymphoid cells, macrophages, neutrophils and natural
killer cells to the site of infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Bovine
bronchial cells that are infected with BHV-1 will release cytokines, which subsequently promote
neutrophil recruitment to the lung (Rivera-Rivas et al., 2009). In vitro, TNF-α, IL-1β and IFN-γ are
secreted in response to BHV-1 infection, causing the increased expression of β2 integrin on bovine
alveolar macrophages and neutrophils (Czuprynski, 2009). The expression of HIF-1α and therefore
its value as a biomarker could still be significant in the context of these acute immune responses if
further investigations incorporated relevant and appropriate viral challenges. In this case, the lack of
a detected difference in HIF-1α could also mean that HIF-1α does not have a significant role in the
pathogenesis of clinical BRD, and further studies involving viral challenge could help to establish
or refute this possibility as well.
70
The overall small sample size tested in this experiment resulted in a low power to detect a possible
difference between the groups, so it is unlikely that a true but subtle difference would have been
statistically identified using such low numbers of animals. Testing larger numbers of animals would
have enhanced the power of the study. However the extent of the within-group variation suggests
that any difference between the groups would need to be even more substantial for such a difference
to be statistically significant, even with a much larger sample size.
Of interest, however, is the large variability in the differences of HIF-1α expression between
stimulated and unstimulated cells within both the cases and control groups. This variability in HIF-
1α expression of cells may be due to individual animal genetics not related to overall BRD
susceptibility, sample quality variation and the influences of the remaining insufficiencies in the
methodologies that will be covered further in the general discussion chapter. Between two
experimental sessions, the stimulated master control mean only varied by 0.04. In one session
however, the master control variation was larger, showing a similar variation in channel shift
between different replicates of stimulated cells as was seen between experimental stimulated and
unstimulated cell cultures. This larger variation in the stimulated master control indicates that
variation between stimulated and unstimulated cells may be due to a deficiency in methodology
rather than true variation. Conducting greater numbers of smaller flow sessions would have allowed
more replicates of the master control to be run. This would have allowed collection of more data
points of the master control values to establish the true repeatability, limits of acceptable variation
and the overall usefulness of the master control. This control, once properly established, could then
be used with revised cell culture and flow cytometry methodologies to more confidently detect true
variation in HIF-1α expression. Given the small changes detected in HIF-1α fluorescence, a more
complex analysis using raw data from flow cytometry LMD files (see Chapter 4) might be more
appropriate, instead of using only the summary results generated by cytometer analysis software.
Given additional resource availability, the use of the raw data LMD files during statistical analysis
would have been implemented.
ICC staining was observed either within the monocytic cytoplasm only, or within both the
cytoplasm and nucleus. This pattern of staining is consistent with the expected two types of
intracellular distribution of HIF-1α before and after translocation into the nucleus. There was no
significant difference in either nuclear positive or cytoplasmic positive staining upon stimulation
between cases and controls, however, within subset one there was consistently greater amounts of
nuclear positive staining in concurrence with slightly less cytoplasmic positive staining in
71
stimulated cells, when compared to the unstimulated cells. These results are consistent with
successful stabilisation and translocation of HIF-1α, however, without a known true positive
control, this cannot be confirmed.
6.5 Conclusion
The methodologies developed in the preceding chapters were applied to experimental samples.
There was no evidence to reject the null hypothesis within the framework of the current study.
However, the low power and methodological issues mean that a definitive statement cannot be
made as to whether HIF-1α can be used as a predictive biomarker for BRD. If a significant
difference in HIF-1α had been detected, there remains an issue regarding the practicality of this
protocol. This will be expanded upon in chapter 7.
72
7 General discussion and future directions
7.1 The importance of finding a biomarker for BRD
BRD `is a multi-factorial, inflammatory respiratory disease complex which poses a significant
problem for the beef industry. The disease complex involves both viral and bacterial pathogens and
invokes both innate and adaptive immune responses as well as the development of clinical
pneumonia in certain feedlot cattle. Investigations are focused on ways to enhance effective and
non-injurious immune responses to these pathogens (Ackermann et al., 2010b) in addition to
combating environmental and managerial issues that may predispose cattle to respiratory disease.
An initial step in the ultimate aim of enhancing effective immune responses is to identify biomarker
molecules that are expressed differently in animals that have an effective response to an infectious
challenge when compared to other animals that develop disease. Such a biomarker would allow the
implementation of targeted breeding programs to enhance resistance to this common yet complex
disease. This project proposed HIF-1α as a potential biomarker for development of clinical disease,
and endeavoured to develop methods to induce and detect HIF-1α expression in bovine
lymphocytes and monocytes. This was followed by implementation of developed methods
experimentally to compare the HIF-1α expression in a group of clinically healthy beef feedlot cattle
(controls) with the HIF-1α expression in beef feedlot cattle that were treated for clinical respiratory
disease (cases) while in the feedlot.
The lack of a significant difference in HIF-1α expression between case and control groups did not
support the potential use of HIF-1α as a biomarker for BRD. However, with sample size dictated by
time and resource constraints, the study only had the power to detect a large difference in HIF-1α
expression between cases and controls. Furthermore, this finding is not considered conclusive due
to a number of weaknesses and limitations in the developed methods. These weaknesses,
limitations, as well as potential solutions and additional ideas for further development will now be
discussed in turn.
7.2 Methodology development issues
7.2.1 Hypoxic cell culture and harvest methods
The inherent weakness of the hypoxic cell culture method is that it was not possible to confirm
successful HIF-1α stabilisation by means other than the detection methods simultaneously being
developed. The stabilisation protocol was developed based on similar studies in the literature,
however, ultimately its efficacy could not be confirmed. Future work could involve use of a better-
established measurement technique such as Western blot and the use of commercially produced
CoCl2-treated cell lysate as a known positive control to confirm accumulation of the protein
73
(Lidgren et al., 2005). A positive control could also aid in further troubleshooting of culture length
and hypoxic mimetic concentration. Additional stimulants such as TNF- α could also be used as an
inducing signal to further increase HIF-1α stimulation during a shorter culture time (Albina et al.,
2001). As discussed in chapter 6, use of BRD viral agents during hypoxic culture would better
mimic the in vivo situation and could also contribute to a better understanding of HIF-1α’s role in
the pathogenesis of clinical BRD. A different cell type could also be considered, given that BHV-1,
PI-3, BRSV and BVDV can infect lung epithelial cells (Ackermann et al., 2010b) and that bovine
bronchial cells that are infected with BHV-1 will release cytokines (Rivera-Rivas et al., 2009). Use
of bovine lung cells for culture in conjunction with these viruses could be an avenue for testing the
potential role of HIF-1α in BRD susceptibility.
A further shortcoming in the protocol was the poor post-culture monocyte viability that meant this
cell population had to be assessed with an alternative technique. This negates, to an extent, one of
the reasons for using flow cytometry, i.e. allowing for simultaneous assessment of multiple cell
types, and also decreases the attractiveness of the protocol for more widespread use, as it would
require multiple sets of reagents, equipment and expertise. A lower concentration of CoCl2 may
improve monocyte viability to the point where these cells could once again be included in flow
cytometry analysis.
Poor post-culture viability was a significant issue with frozen-thawed cells. Unfortunately, due to
the ongoing trouble-shooting and work-up of protocols and the restricted nature of sample
collections, the opportunity to be in the position to culture fresh cells immediately after collection
did not arise. Using fresh cells remains an avenue for future investigation into hypoxic culture of
bovine leukocytes for stabilisation of HIF-1α.
7.2.2 Flow cytometry methods
This project explored the use of flow cytometry to measure HIF-1α protein as a relatively novel
technique that could potentially allow enhanced sensitivity in detecting differences in HIF-1α
expression between individuals. Flow cytometry has the capacity to make individual measurements
from a large number of discrete particles, and can be used for the analysis of any cellular structure
or function as long as an appropriate probe is available (Jaroszeski and Radcliff, 1999). Given the
large potential for variation inherent in the multiple steps of this protocol, a positive control with
clearly defined levels of HIF-1α would have been of great use. Unfortunately such a control was not
available. The difference in HIF-1α fluorescence between unstimulated and stimulated cells was not
pronounced and did not show a consistent increased or decreased fluorescence after stimulation.
The HIF-1α fluorescence was also quite close to the negative controls (1 to 2 channel shifts only) in
74
contrast to the phenotypic marker fluorescence, which often was over 15 channel shifts above the
negative controls. This suggests the fluorochrome used for HIF-1α detection was not bright enough
in the context of the relative lower expression of the molecule. Alternatively the phenotypic
fluorochromes were too bright in comparison, or it could be a combination of the two options, with
subtle changes in HIF-1α possibly masked as a result. The brightness of the HIF-1α signal is
affected not only by the intensity with which the positive cells are stained, but also the background
of the negative cells (Baumgarth and Roederer, 2000). Use of a brighter fluorochrome such as PE,
Cy5PE or APC could be used to investigate what was essentially an unknown level of a molecule of
interest (Baumgarth and Roederer, 2000). In addition, discontinuation of cell phenotyping until
measurement of HIF-1α protein has been established should have been considered, if time was not a
limitation.
It is possible that the significant degree of apoptosis occurring in the monocyte population was also
affecting other cell types, causing aberrant events. Large amounts of cell death in this project during
method development could have influenced flow cytometry results. Ethidium monoazide could
have been used to detect dead cells in a separate sample.
With the developed method there still existed flow cytometry events that were not labelled with
either phenotypic marker. These may represent either remaining monocytes, B cells undergoing
differentiation and therefore down-regulate CD20, dead cells, large debris or contaminating
granulocytes, although the latter is unlikely considering the poor survival of freezing-preservation
of granulocytes. Events that were labelled by both CD20 and CD3 were also noted. The nature of
such events remains uncertain. Neoplastic lymphocytes can show true CD20 and CD3 positivity on
the same cell (Sun et al., 2004), however, in this case the events are considered more likely to be
dead cells with non-specific binding of antibody. Several methods were employed to minimize
variation due to electronic factors. These methods included first establishing and then using fixed
machine settings during the experiment sessions, the diligent use of the recommended quality
assurance beads to assess laser alignment prior to each session and the use of a tailored master
control with each session. Unstained samples (which did not contain signal) addressed possible
cellular auto-fluorescence and variation due to non-specific binding of antibody was minimised by
frequent washing of the cells during preparation and omitting use of problematic secondary
antibodies in the protocol. Gating of populations was performed manually on 2-dimensional dot
plots, the subjectivity of which is a potential source of variation during routine post-processing of
flow data, and is acknowledged as a possible source of data variation in this experiment.
75
7.2.3 ICC methods
As a consequence of methodology development issues, ICC was also implemented as a more
established method of HIF-1α measurement. The method can provide information as to the
presence of absence of HIF-1α andthe nuclear translocation of the molecule, but can be argued as
relatively insensitive for distinguishing between any subtle differences in the level of HIF-1α
expression between different animals or culture conditions.
A possible source of variation in ICC is the subjectivity of human interpretation of staining. This
could be mitigated by applying morphometric computer-based interpretation of the
immunolabelling. Lack of appropriate facilities for this was, however, and issue not resolved within
the timeframe of the project. In this study the ICC was carried out manually and if the approach was
to be applied to a much larger number of animals automated staining should be considered in
combination with a morphometric read-out.
required.
7.3 Future directions for investigation of biomarkers for BRD
In the future, in addition to possible modifications of the methodologies used in this study, other
methodologies for the evaluation of HIF-1α in bovines should be considered, such as Western blot
for HIF-1α protein or real-time polymerase chain reaction for HIF-1α mRNA. Western blot was
also considered as a method in this study due to available supervisor expertise and facilities.
However, the large numbers of cells required for adequate protein yield would be problematic due
to the low total cell numbers in many of the intended experimental samples as well as the low
retrievable cell population proportion and low total cell counts of harvested hypoxic cultures.
Performing Western blot instead of flow cytometry would remove information gained through
phenotyping, and would therefore require isolation of lymphocyte subsets prior to hypoxic culture
conditions. This is unlikely to be practical on a large scale unless cell sorting facilities are available.
Nevertheless, Western blot detection of HIF-1α protein is more established method than flow
cytometry and may represent a more practical alternative for large scale assessment.
Another possibility to consider in further investigations of the role the HIF-1α pathway in BRD is to
measure molecules induced downstream of HIF-1α such as VEGF, TNFα, other inflammatory
mediators and ROS. It is also worthwhile considering other molecules in hypoxic inflammation in
the lungs as potential biomarkers, such as the transcription factor NF-kB. Alveolar epithelial cells
and resident alveolar macrophages recognize various microbial components via pattern recognition
receptors (PRR) with resultant production of proinflamamtory cytokines via the NF-kB pathway,
directly affecting expression of genes involved in innate immunity as well as indirectly causing
amplification of the HIF1α pathway (Schaible et al., 2010).
76
Based on this study, the expression of HIF-1α and its value as a biomarker for the development of
clinical BRD as a disease outcome could still be significant and remains to be established. Further
work is needed to improve these methodologies, or develop alternative methods and undertake a
study in a much larger population of cattle in order to achieve more conclusive results. The
expression of HIF-1α could still be shown as significant in the context of the bovine acute immune
responses to BRD viral pathogens if further investigations incorporated relevant and appropriate
viral challenges.
In this study, the lack of a detected difference in HIF-1α could also mean that HIF-1α does not have
a significant role in the pathogenesis of clinical BRD, and further studies involving viral challenge
could also help to establish or refute this possibility.
77
8 References
[ANON] 1973. Glossary on respiration and gas exchange. Journal of Applied Physiology, 34, 549-58.
ACKERMANN, M. R., DERSCHEID, R. & ROTH, J. A. 2010a. Innate immunology of bovine respiratory disease. The Veterinary clinics of North America. Food animal practice, 26, 215-28.
ACKERMANN, M. R., DERSCHEID, R. & ROTH, J. A. 2010b. Innate immunology of bovine respiratory disease. Vet Clin North Am Food Anim Pract, 26, 215-28.
ADAMIS, A. P., MILLER, J. W., BERNAL, M. T., D'AMICO, D. J., FOLKMAN, J., YEO, T. K. & YEO, K. T. 1994. Increased vascular endothelial growth factor levels in the vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol, 118, 445-50.
AICH, P., BABIUK, L. A., POTTER, A. A. & GRIEBEL, P. 2009. Biomarkers for prediction of bovine respiratory disease outcome. OMICS, 13, 199-209.
ALBINA, J. E., MASTROFRANCESCO, B., VESSELLA, J. A., LOUIS, C. A., HENRY, W. L., JR. & REICHNER, J. S. 2001. HIF-1 expression in healing wounds: HIF-1alpha induction in primary inflammatory cells by TNF-alpha. Am J Physiol Cell Physiol, 281, C1971-7.
AMERI, K., LUONG, R., ZHANG, H., POWELL, A. A., MONTGOMERY, K. D., ESPINOSA, I., BOULEY, D. M., HARRIS, A. L. & JEFFREY, S. S. 2010. Circulating tumour cells demonstrate an altered response to hypoxia and an aggressive phenotype. Br J Cancer, 102, 561-9.
ANAND, R. J., GRIBAR, S. C., LI, J., KOHLER, J. W., BRANCA, M. F., DUBOWSKI, T., SODHI, C. P. & HACKAM, D. J. 2007. Hypoxia causes an increase in phagocytosis by macrophages in a HIF-1alpha-dependent manner. J Leukoc Biol, 82, 1257-65.
AUGSTEIN, A., POITZ, D. M., BRAUN-DULLAEUS, R. C., STRASSER, R. H. & SCHMEISSER, A. 2011. Cell-specific and hypoxia-dependent regulation of human HIF-3alpha: inhibition of the expression of HIF target genes in vascular cells. Cell Mol Life Sci, 68, 2627-42.
BABIUK, L. A., VAN DRUNEN LITTEL-VAN DEN HURK, S. & TIKOO, S. K. 1996. Immunology of bovine herpesvirus 1 infection. Vet Microbiol, 53, 31-42.
BASKIN, C. R., BIELEFELDT-OHMANN, H., TUMPEY, T. M., SABOURIN, P. J., LONG, J. P., GARCIA-SASTRE, A., TOLNAY, A. E., ALBRECHT, R., PYLES, J. A., OLSON, P. H., AICHER, L. D., ROSENZWEIG, E. R., MURALI-KRISHNA, K., CLARK, E. A., KOTUR, M. S., FORNEK, J. L., PROLL, S., PALERMO, R. E., SABOURIN, C. L. & KATZE, M. G. 2009. Early and sustained innate immune response defines pathology and death in nonhuman primates infected by highly pathogenic influenza virus. Proc Natl Acad Sci U S A, 106, 3455-60.
BAUMGARTH, N. & ROEDERER, M. 2000. A practical approach to multicolor flow cytometry for immunophenotyping. J Immunol Methods, 243, 77-97.
BELAIBA, R. S., BONELLO, S., ZAHRINGER, C., SCHMIDT, S., HESS, J., KIETZMANN, T. & GORLACH, A. 2007. Hypoxia up-regulates hypoxia-inducible factor-1alpha transcription by involving phosphatidylinositol 3-kinase and nuclear factor kappaB in pulmonary artery smooth muscle cells. Mol Biol Cell, 18, 4691-7.
BEN-SHOSHAN, J., MAYSEL-AUSLENDER, S., MOR, A., KEREN, G. & GEORGE, J. 2008. Hypoxia controls CD4+CD25+ regulatory T-cell homeostasis via hypoxia-inducible factor-1alpha. Eur J Immunol, 38, 2412-8.
BIELEFELDT OHMANN, H., BABIUK, L. A. & HARLAND, R. 1991. Cytokine synergy with viral cytopathic effects and bacterial products during the pathogenesis of respiratory tract infection. Clin Immunol Immunopathol, 60, 153-70.
78
BIELEFELDT-OHMANN, H., SMIRNOVA, N. P., TOLNAY, A. E., WEBB, B. T., ANTONIAZZI, A. Q., VAN CAMPEN, H. & HANSEN, T. R. 2012. Neuro-invasion by a 'Trojan Horse' strategy and vasculopathy during intrauterine flavivirus infection. Int J Exp Pathol, 93, 24-33.
BIELEFELDT-OHMANN, H., TOLNAY, A. E., REISENHAUER, C. E., HANSEN, T. R., SMIRNOVA, N. & VAN CAMPEN, H. 2008. Transplacental infection with non-cytopathic bovine viral diarrhoea virus types 1b and 2: viral spread and molecular neuropathology. J Comp Pathol, 138, 72-85.
BIJU, M. P., NEUMANN, A. K., BENSINGER, S. J., JOHNSON, R. S., TURKA, L. A. & HAASE, V. H. 2004. Vhlh gene deletion induces Hif-1-mediated cell death in thymocytes. Mol Cell Biol, 24, 9038-47.
BILTON, R., MAZURE, N., TROTTIER, E., HATTAB, M., DERY, M. A., RICHARD, D. E., POUYSSEGUR, J. & BRAHIMI-HORN, M. C. 2005. Arrest-defective-1 protein, an acetyltransferase, does not alter stability of hypoxia-inducible factor (HIF)-1alpha and is not induced by hypoxia or HIF. J Biol Chem, 280, 31132-40.
BLOUIN, C. C., PAGE, E. L., SOUCY, G. M. & RICHARD, D. E. 2004. Hypoxic gene activation by lipopolysaccharide in macrophages: implication of hypoxia-inducible factor 1alpha. Blood, 103, 1124-30.
BOONLAYANGOOR, P., TELISCHI, M., BOONLAYANGOOR, S., SINCLAIR, T. F. & MILLHOUSE, E. W. 1980. Cryopreservation of human granulocytes: study of granulocyte function and ultrastructure. Blood, 56, 237-45.
BOVE, P. F., HRISTOVA, M., WESLEY, U. V., OLSON, N., LOUNSBURY, K. M. & VAN DER VLIET, A. 2008. Inflammatory levels of nitric oxide inhibit airway epithelial cell migration by inhibition of the kinase ERK1/2 and activation of hypoxia-inducible factor-1 alpha. J Biol Chem, 283, 17919-28.
BRAUN, R. D., LANZEN, J. L., SNYDER, S. A. & DEWHIRST, M. W. 2001. Comparison of tumor and normal tissue oxygen tension measurements using OxyLite or microelectrodes in rodents. Am J Physiol Heart Circ Physiol, 280, H2533-44.
BROWN, L. F., YEO, K. T., BERSE, B., YEO, T. K., SENGER, D. R., DVORAK, H. F. & VAN DE WATER, L. 1992. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med, 176, 1375-9.
BRUICK, R. K. & MCKNIGHT, S. L. 2001. A conserved family of prolyl-4-hydroxylases that modify HIF. Science, 294, 1337-40.
BRUNE, B. & ZHOU, J. 2007. Nitric oxide and superoxide: interference with hypoxic signaling. Cardiovasc Res, 75, 275-82.
BURRY, R. W. 2011. Controls for immunocytochemistry: an update. J Histochem Cytochem, 59, 6-12.
CAI, Q., MURAKAMI, M., SI, H. & ROBERTSON, E. S. 2007. A potential alpha-helix motif in the amino terminus of LANA encoded by Kaposi's sarcoma-associated herpesvirus is critical for nuclear accumulation of HIF-1alpha in normoxia. J Virol, 81, 10413-23.
CALLAPINA, M., ZHOU, J., SCHMID, T., KOHL, R. & BRUNE, B. 2005. NO restores HIF-1alpha hydroxylation during hypoxia: role of reactive oxygen species. Free Radic Biol Med, 39, 925-36.
CAMPANA, D., COUSTAN-SMITH, E. & JANOSSY, G. 1989. "Cytoplasmic" expression of nuclear antigens. Leukemia, 3, 239-40.
CARMELIET, P., DOR, Y., HERBERT, J. M., FUKUMURA, D., BRUSSELMANS, K., DEWERCHIN, M., NEEMAN, M., BONO, F., ABRAMOVITCH, R., MAXWELL, P., KOCH, C. J., RATCLIFFE, P., MOONS, L., JAIN, R. K., COLLEN, D. & KESHERT, E. 1998. Role of HIF-1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature, 394, 485-90.
CHACHAMI, G., SIMOS, G., HATZIEFTHIMIOU, A., BONANOU, S., MOLYVDAS, P. A. & PARASKEVA, E. 2004. Cobalt induces hypoxia-inducible factor-1alpha expression in airway smooth muscle cells by a reactive oxygen species- and PI3K-dependent mechanism. Am J Respir Cell Mol Biol, 31, 544-51.
79
CHANDEL, N. S., MCCLINTOCK, D. S., FELICIANO, C. E., WOOD, T. M., MELENDEZ, J. A., RODRIGUEZ, A. M. & SCHUMACKER, P. T. 2000. Reactive oxygen species generated at mitochondrial complex III stabilize hypoxia-inducible factor-1alpha during hypoxia: a mechanism of O2 sensing. J Biol Chem, 275, 25130-8.
CHEN, C., PORE, N., BEHROOZ, A., ISMAIL-BEIGI, F. & MAITY, A. 2001. Regulation of glut1 mRNA by hypoxia-inducible factor-1. Interaction between H-ras and hypoxia. J Biol Chem, 276, 9519-25.
CRAMER, T., YAMANISHI, Y., CLAUSEN, B. E., FORSTER, I., PAWLINSKI, R., MACKMAN, N., HAASE, V. H., JAENISCH, R., CORR, M., NIZET, V., FIRESTEIN, G. S., GERBER, H. P., FERRARA, N. & JOHNSON, R. S. 2003. HIF-1alpha is essential for myeloid cell-mediated inflammation. Cell, 112, 645-57.
CZUPRYNSKI, C. J. 2009. Host response to bovine respiratory pathogens. Anim Health Res Rev, 10, 141-3.
DAI, Y., XU, M., WANG, Y., PASHA, Z., LI, T. & ASHRAF, M. 2007. HIF-1alpha induced-VEGF overexpression in bone marrow stem cells protects cardiomyocytes against ischemia. J Mol Cell Cardiol, 42, 1036-44.
DAI, Z. J., GAO, J., MA, X. B., YAN, K., LIU, X. X., KANG, H. F., JI, Z. Z., GUAN, H. T. & WANG, X. J. 2012. Up-regulation of hypoxia inducible factor-1alpha by cobalt chloride correlates with proliferation and apoptosis in PC-2 cells. J Exp Clin Cancer Res, 31, 28.
DALTON-GRIFFIN, L., WILSON, S. J. & KELLAM, P. 2009. X-box binding protein 1 contributes to induction of the Kaposi's sarcoma-associated herpesvirus lytic cycle under hypoxic conditions. J Virol, 83, 7202-9.
DAREKAR, S., GEORGIOU, K., YURCHENKO, M., YENAMANDRA, S. P., CHACHAMI, G., SIMOS, G., KLEIN, G. & KASHUBA, E. 2012. Epstein-Barr virus immortalization of human B-cells leads to stabilization of hypoxia-induced factor 1 alpha, congruent with the Warburg effect. PLoS One, 7, e42072.
DAVIS, D. A., RINDERKNECHT, A. S., ZOETEWEIJ, J. P., AOKI, Y., READ-CONNOLE, E. L., TOSATO, G., BLAUVELT, A. & YARCHOAN, R. 2001. Hypoxia induces lytic replication of Kaposi sarcoma-associated herpesvirus. Blood, 97, 3244-50.
DEHNE, N. & BRUNE, B. 2009. HIF-1 in the inflammatory microenvironment. Exp Cell Res, 315, 1791-7.
DESHMANE, S. L., MUKERJEE, R., FAN, S., DEL VALLE, L., MICHIELS, C., SWEET, T., ROM, I., KHALILI, K., RAPPAPORT, J., AMINI, S. & SAWAYA, B. E. 2009. Activation of the oxidative stress pathway by HIV-1 Vpr leads to induction of hypoxia-inducible factor 1alpha expression. J Biol Chem, 284, 11364-73.
EHRISMANN, D., FLASHMAN, E., GENN, D. N., MATHIOUDAKIS, N., HEWITSON, K. S., RATCLIFFE, P. J. & SCHOFIELD, C. J. 2007. Studies on the activity of the hypoxia-inducible-factor hydroxylases using an oxygen consumption assay. Biochem J, 401, 227-34.
EITAM, H., VAYA, J., BROSH, A., ORLOV, A., KHATIB, S., IZHAKI, I. & SHABTAY, A. 2010. Differential stress responses among newly received calves: variations in reductant capacity and Hsp gene expression. Cell stress & chaperones, 15, 865-76.
ELSON, D. A., THURSTON, G., HUANG, L. E., GINZINGER, D. G., MCDONALD, D. M., JOHNSON, R. S. & ARBEIT, J. M. 2001. Induction of hypervascularity without leakage or inflammation in transgenic mice overexpressing hypoxia-inducible factor-1alpha. Genes Dev, 15, 2520-32.
ELTZSCHIG, H. K. & CARMELIET, P. 2011. Hypoxia and inflammation. N Engl J Med, 364, 656-65.
EMA, M., TAYA, S., YOKOTANI, N., SOGAWA, K., MATSUDA, Y. & FUJII-KURIYAMA, Y. 1997. A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1alpha regulates the VEGF expression and is potentially involved in lung and vascular development. Proc Natl Acad Sci U S A, 94, 4273-8.
80
EPSTEIN, A. C., GLEADLE, J. M., MCNEILL, L. A., HEWITSON, K. S., O'ROURKE, J., MOLE, D. R., MUKHERJI, M., METZEN, E., WILSON, M. I., DHANDA, A., TIAN, Y. M., MASSON, N., HAMILTON, D. L., JAAKKOLA, P., BARSTEAD, R., HODGKIN, J., MAXWELL, P. H., PUGH, C. W., SCHOFIELD, C. J. & RATCLIFFE, P. J. 2001. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell, 107, 43-54.
FETSCH, P. 2005. Immunocytochemistry. In: WALKER, J. & RAPLEY, R. (eds.) Medical Biomethods Handbook. Humana Press.
FOLKMAN, J. & SHING, Y. 1992. Angiogenesis. J Biol Chem, 267, 10931-4.
FORMENTO, J. L., BERRA, E., FERRUA, B., MAGNE, N., SIMOS, G., BRAHIMI-HORN, C., POUYSSEGUR, J. & MILANO, G. 2005. Enzyme-linked immunosorbent assay for pharmacological studies targeting hypoxia-inducible factor 1alpha. Clin Diagn Lab Immunol, 12, 660-4.
FORRISTAL, C. E., WRIGHT, K. L., HANLEY, N. A., OREFFO, R. O. & HOUGHTON, F. D. 2010. Hypoxia inducible factors regulate pluripotency and proliferation in human embryonic stem cells cultured at reduced oxygen tensions. Reproduction, 139, 85-97.
FORSYTHE, J. A., JIANG, B. H., IYER, N. V., AGANI, F., LEUNG, S. W., KOOS, R. D. & SEMENZA, G. L. 1996. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol, 16, 4604-13.
FREDE, S., BERCHNER-PFANNSCHMIDT, U. & FANDREY, J. 2007. Regulation of hypoxia-inducible factors during inflammation. Methods Enzymol, 435, 405-19.
GARSON, J. A., BEVERLEY, P. C., COAKHAM, H. B. & HARPER, E. I. 1982. Monoclonal antibodies against human T lymphocytes label Purkinje neurones of many species. Nature, 298, 375-7.
GERALD, D., BERRA, E., FRAPART, Y. M., CHAN, D. A., GIACCIA, A. J., MANSUY, D., POUYSSEGUR, J., YANIV, M. & MECHTA-GRIGORIOU, F. 2004. JunD reduces tumor angiogenesis by protecting cells from oxidative stress. Cell, 118, 781-94.
GILBERT, S. F. 2010. Developmental biology, Sunderland, Mass., Sinauer Associates.
GIVAN, A. L. 2001. Flow cytometry : first principles, New York ; Chichester, Wiley.
GOLDBERG, M. A., DUNNING, S. P. & BUNN, H. F. 1988. Regulation of the erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science, 242, 1412-5.
GREIJER, A. E. & VAN DER WALL, E. 2004. The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis. Journal of clinical pathology, 57, 1009-14.
GRIFFIN, D. 1997. Economic impact associated with respiratory disease in beef cattle. Vet Clin North Am Food Anim Pract, 13, 367-77.
GU, Y. Z., MORAN, S. M., HOGENESCH, J. B., WARTMAN, L. & BRADFIELD, C. A. 1998. Molecular characterization and chromosomal localization of a third alpha-class hypoxia inducible factor subunit, HIF3alpha. Gene Expr, 7, 205-13.
GUO, M., SONG, L. P., JIANG, Y., LIU, W., YU, Y. & CHEN, G. Q. 2006. Hypoxia-mimetic agents desferrioxamine and cobalt chloride induce leukemic cell apoptosis through different hypoxia-inducible factor-1alpha independent mechanisms. Apoptosis, 11, 67-77.
GUZY, R. D., HOYOS, B., ROBIN, E., CHEN, H., LIU, L., MANSFIELD, K. D., SIMON, M. C., HAMMERLING, U. & SCHUMACKER, P. T. 2005. Mitochondrial complex III is required for hypoxia-induced ROS production and cellular oxygen sensing. Cell Metab, 1, 401-8.
HADDAD, J. J. 2002. Antioxidant and prooxidant mechanisms in the regulation of redox(y)-sensitive transcription factors. Cell Signal, 14, 879-97.
81
HADDAD, J. J. & LAND, S. C. 2001. A non-hypoxic, ROS-sensitive pathway mediates TNF-alpha-dependent regulation of HIF-1alpha. FEBS letters, 505, 269-74.
HAEBERLE, H. A., DURRSTEIN, C., ROSENBERGER, P., HOSAKOTE, Y. M., KUHLICKE, J., KEMPF, V. A., GAROFALO, R. P. & ELTZSCHIG, H. K. 2008. Oxygen-independent stabilization of hypoxia inducible factor (HIF)-1 during RSV infection. PLoS One, 3, e3352.
HAGEN, T., TAYLOR, C. T., LAM, F. & MONCADA, S. 2003. Redistribution of intracellular oxygen in hypoxia by nitric oxide: effect on HIF1alpha. Science, 302, 1975-8.
HAILMAN, E., LICHENSTEIN, H. S., WURFEL, M. M., MILLER, D. S., JOHNSON, D. A., KELLEY, M., BUSSE, L. A., ZUKOWSKI, M. M. & WRIGHT, S. D. 1994. Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS to CD14. J Exp Med, 179, 269-77.
HAQUE, M., DAVIS, D. A., WANG, V., WIDMER, I. & YARCHOAN, R. 2003. Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) contains hypoxia response elements: relevance to lytic induction by hypoxia. J Virol, 77, 6761-8.
HAQUE, M. & KOUSOULAS, K. G. 2013. The Kaposi's sarcoma-associated herpesvirus ORF34 protein binds to HIF-1alpha and causes its degradation via the proteasome pathway. J Virol, 87, 2164-73.
HARA, S., KOBAYASHI, C. & IMURA, N. 1999. Molecular cloning of cDNAs encoding hypoxia-inducible factor (HIF)-1alpha and -2alpha of bovine arterial endothelial cells. Biochim Biophys Acta, 1445, 237-43.
HELLWIG-BURGEL, T., RUTKOWSKI, K., METZEN, E., FANDREY, J. & JELKMANN, W. 1999. Interleukin-1beta and tumor necrosis factor-alpha stimulate DNA binding of hypoxia-inducible factor-1. Blood, 94, 1561-7.
HELMLINGER, G., YUAN, F., DELLIAN, M. & JAIN, R. K. 1997. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med, 3, 177-82.
HERSHKOVITZ, T., SHENHAV, A., SABO, E., BEN-IZHAK, O. & HERSHKOVITZ, D. 2013. Development of a computerized morphometry application for assessment of the tumor fraction in colon carcinoma tissue samples. Appl Immunohistochem Mol Morphol, 21, 54-8.
HERTIG, A. T. 1935. Angiogenesis in the early human chorion and in the primary placenta of the macaque monkey, Washington,.
HODGSON, P. D., AICH, P., MANUJA, A., HOKAMP, K., ROCHE, F. M., BRINKMAN, F. S., POTTER, A., BABIUK, L. A. & GRIEBEL, P. J. 2005. Effect of stress on viral-bacterial synergy in bovine respiratory disease: novel mechanisms to regulate inflammation. Comp Funct Genomics, 6, 244-50.
HU, R., DAI, A. & TAN, S. 2002. Hypoxia-inducible factor 1 alpha upregulates the expression of inducible nitric oxide synthase gene in pulmonary arteries of hyposic rat. Chin Med J (Engl), 115, 1833-7.
HUANG, L. E., ARANY, Z., LIVINGSTON, D. M. & BUNN, H. F. 1996. Activation of hypoxia-inducible transcription factor depends primarily upon redox-sensitive stabilization of its alpha subunit. J Biol Chem, 271, 32253-9.
HUANG, L. E., GU, J., SCHAU, M. & BUNN, H. F. 1998. Regulation of hypoxia-inducible factor 1alpha is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc Natl Acad Sci U S A, 95, 7987-92.
HWANG, II, WATSON, I. R., DER, S. D. & OHH, M. 2006. Loss of VHL confers hypoxia-inducible factor (HIF)-dependent resistance to vesicular stomatitis virus: role of HIF in antiviral response. J Virol, 80, 10712-23.
INTAGLIETTA, M., JOHNSON, P. C. & WINSLOW, R. M. 1996. Microvascular and tissue oxygen distribution. Cardiovasc Res, 32, 632-43.
82
IVAN, M., KONDO, K., YANG, H., KIM, W., VALIANDO, J., OHH, M., SALIC, A., ASARA, J. M., LANE, W. S. & KAELIN, W. G., JR. 2001. HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science, 292, 464-8.
IYER, N. V., KOTCH, L. E., AGANI, F., LEUNG, S. W., LAUGHNER, E., WENGER, R. H., GASSMANN, M., GEARHART, J. D., LAWLER, A. M., YU, A. Y. & SEMENZA, G. L. 1998. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev, 12, 149-62.
JAAKKOLA, P., MOLE, D. R., TIAN, Y. M., WILSON, M. I., GIELBERT, J., GASKELL, S. J., KRIEGSHEIM, A., HEBESTREIT, H. F., MUKHERJI, M., SCHOFIELD, C. J., MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 2001. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science, 292, 468-72.
JANOSSY, G., COUSTAN-SMITH, E. & CAMPANA, D. 1989. The reliability of cytoplasmic CD3 and CD22 antigen expression in the immunodiagnosis of acute leukemia: a study of 500 cases. Leukemia, 3, 170-81.
JANTSCH, J., CHAKRAVORTTY, D., TURZA, N., PRECHTEL, A. T., BUCHHOLZ, B., GERLACH, R. G., VOLKE, M., GLASNER, J., WARNECKE, C., WIESENER, M. S., ECKARDT, K. U., STEINKASSERER, A., HENSEL, M. & WILLAM, C. 2008. Hypoxia and hypoxia-inducible factor-1 alpha modulate lipopolysaccharide-induced dendritic cell activation and function. J Immunol, 180, 4697-705.
JAROSZESKI, M. J. & RADCLIFF, G. 1999. Fundamentals of flow cytometry. Mol Biotechnol, 11, 37-53.
JEONG, J. W., BAE, M. K., AHN, M. Y., KIM, S. H., SOHN, T. K., BAE, M. H., YOO, M. A., SONG, E. J., LEE, K. J. & KIM, K. W. 2002. Regulation and destabilization of HIF-1alpha by ARD1-mediated acetylation. Cell, 111, 709-20.
JIANG, B. H., SEMENZA, G. L., BAUER, C. & MARTI, H. H. 1996. Hypoxia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J Physiol, 271, C1172-80.
JIANG, B. H., ZHENG, J. Z., LEUNG, S. W., ROE, R. & SEMENZA, G. L. 1997. Transactivation and inhibitory domains of hypoxia-inducible factor 1alpha. Modulation of transcriptional activity by oxygen tension. J Biol Chem, 272, 19253-60.
JIN, W. S., KONG, Z. L., SHEN, Z. F., JIN, Y. Z., ZHANG, W. K. & CHEN, G. F. 2011. Regulation of hypoxia inducible factor-1alpha expression by the alteration of redox status in HepG2 cells. J Exp Clin Cancer Res, 30, 61.
JONES, M., CORDELL, J. L., BEYERS, A. D., TSE, A. G. & MASON, D. Y. 1993. Detection of T and B cells in many animal species using cross-reactive anti-peptide antibodies. J Immunol, 150, 5429-35.
JUNG, F., PALMER, L. A., ZHOU, N. & JOHNS, R. A. 2000. Hypoxic regulation of inducible nitric oxide synthase via hypoxia inducible factor-1 in cardiac myocytes. Circ Res, 86, 319-25.
JUNG, Y., ISAACS, J. S., LEE, S., TREPEL, J., LIU, Z. G. & NECKERS, L. 2003. Hypoxia-inducible factor induction by tumour necrosis factor in normoxic cells requires receptor-interacting protein-dependent nuclear factor kappa B activation. Biochem J, 370, 1011-7.
KALCKAR, H. M. 1941. The Nature of Energetic Coupling in Biological Syntheses. Chemical Reviews, 28, 71-178.
KALLIO, P. J., PONGRATZ, I., GRADIN, K., MCGUIRE, J. & POELLINGER, L. 1997. Activation of hypoxia-inducible factor 1alpha: posttranscriptional regulation and conformational change by recruitment of the Arnt transcription factor. Proc Natl Acad Sci U S A, 94, 5667-72.
KALLIO, P. J., WILSON, W. J., O'BRIEN, S., MAKINO, Y. & POELLINGER, L. 1999. Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway. J Biol Chem, 274, 6519-25.
83
KECK, P. J., HAUSER, S. D., KRIVI, G., SANZO, K., WARREN, T., FEDER, J. & CONNOLLY, D. T. 1989. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science, 246, 1309-12.
KEMPF, V. A., LEBIEDZIEJEWSKI, M., ALITALO, K., WALZLEIN, J. H., EHEHALT, U., FIEBIG, J., HUBER, S., SCHUTT, B., SANDER, C. A., MULLER, S., GRASSL, G., YAZDI, A. S., BREHM, B. & AUTENRIETH, I. B. 2005. Activation of hypoxia-inducible factor-1 in bacillary angiomatosis: evidence for a role of hypoxia-inducible factor-1 in bacterial infections. Circulation, 111, 1054-62.
KIELIAN, T. L. & BLECHA, F. 1995. CD14 and other recognition molecules for lipopolysaccharide: a review. Immunopharmacology, 29, 187-205.
KILANI, M. M., MOHAMMED, K. A., NASREEN, N., TEPPER, R. S. & ANTONY, V. B. 2004. RSV causes HIF-1alpha stabilization via NO release in primary bronchial epithelial cells. Inflammation, 28, 245-51.
KIM, J. O., KIM, J. Y., KWACK, M. H., HONG, S. H., KIM, M. K., KIM, J. C. & SUNG, Y. K. 2010. Identification of troglitazone responsive genes: induction of RTP801 during troglitazone-induced apoptosis in Hep 3B cells. BMB Rep, 43, 599-603.
KITTLICK, P. D. 1986. Inflammation, glycolytic metabolism, and glycosaminoglycans. Exp Pathol, 30, 1-19.
KNOWLES, H. J., RAVAL, R. R., HARRIS, A. L. & RATCLIFFE, P. J. 2003. Effect of ascorbate on the activity of hypoxia-inducible factor in cancer cells. Cancer Res, 63, 1764-8.
KOIVUNEN, P., HIRSILA, M., GUNZLER, V., KIVIRIKKO, K. I. & MYLLYHARJU, J. 2004. Catalytic properties of the asparaginyl hydroxylase (FIH) in the oxygen sensing pathway are distinct from those of its prolyl 4-hydroxylases. J Biol Chem, 279, 9899-904.
KOJIMA, H., GU, H., NOMURA, S., CALDWELL, C. C., KOBATA, T., CARMELIET, P., SEMENZA, G. L. & SITKOVSKY, M. V. 2002. Abnormal B lymphocyte development and autoimmunity in hypoxia-inducible factor 1alpha -deficient chimeric mice. Proc Natl Acad Sci U S A, 99, 2170-4.
KONG, T., ELTZSCHIG, H. K., KARHAUSEN, J., COLGAN, S. P. & SHELLEY, C. S. 2004. Leukocyte adhesion during hypoxia is mediated by HIF-1-dependent induction of beta2 integrin gene expression. Proc Natl Acad Sci U S A, 101, 10440-5.
KOUKOURAKIS, M. I., GIATROMANOLAKI, A., SIVRIDIS, E., SIMOPOULOS, C., TURLEY, H., TALKS, K., GATTER, K. C. & HARRIS, A. L. 2002. Hypoxia-inducible factor (HIF1A and HIF2A), angiogenesis, and chemoradiotherapy outcome of squamous cell head-and-neck cancer. International journal of radiation oncology, biology, physics, 53, 1192-1202.
LANDO, D., PEET, D. J., GORMAN, J. J., WHELAN, D. A., WHITELAW, M. L. & BRUICK, R. K. 2002a. FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes Dev, 16, 1466-71.
LANDO, D., PEET, D. J., WHELAN, D. A., GORMAN, J. J. & WHITELAW, M. L. 2002b. Asparagine hydroxylation of the HIF transactivation domain a hypoxic switch. Science, 295, 858-61.
LEUNG, D. W., CACHIANES, G., KUANG, W. J., GOEDDEL, D. V. & FERRARA, N. 1989. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science, 246, 1306-9.
LIDGREN, A., HEDBERG, Y., GRANKVIST, K., RASMUSON, T., VASKO, J. & LJUNGBERG, B. 2005. The expression of hypoxia-inducible factor 1alpha is a favorable independent prognostic factor in renal cell carcinoma. Clin Cancer Res, 11, 1129-35.
LINDER, N., MARTELIN, E., LAPATTO, R. & RAIVIO, K. O. 2003. Posttranslational inactivation of human xanthine oxidoreductase by oxygen under standard cell culture conditions. Am J Physiol Cell Physiol, 285, C48-55.
LIPMANN, F. 1941. Metabolic Generation and Utilization of Phosphate Bond Energy. Advances in Enzymology and Related Areas of Molecular Biology.
84
LUKASHEV, D., KLEBANOV, B., KOJIMA, H., GRINBERG, A., OHTA, A., BERENFELD, L., WENGER, R. H. & SITKOVSKY, M. 2006. Cutting edge: hypoxia-inducible factor 1alpha and its activation-inducible short isoform I.1 negatively regulate functions of CD4+ and CD8+ T lymphocytes. J Immunol, 177, 4962-5.
MAECKER, H. T. & TROTTER, J. 2006. Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A, 69, 1037-42.
MAHON, P. C., HIROTA, K. & SEMENZA, G. L. 2001. FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev, 15, 2675-86.
MAKINO, Y., CAO, R., SVENSSON, K., BERTILSSON, G., ASMAN, M., TANAKA, H., CAO, Y., BERKENSTAM, A. & POELLINGER, L. 2001. Inhibitory PAS domain protein is a negative regulator of hypoxia-inducible gene expression. Nature, 414, 550-4.
MANGAN, D. F. & WAHL, S. M. 1991. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and pro-inflammatory cytokines. J Immunol, 147, 3408-12.
MASSON, N., WILLAM, C., MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 2001. Independent function of two destruction domains in hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation. EMBO J, 20, 5197-206.
MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 1993. Inducible operation of the erythropoietin 3' enhancer in multiple cell lines: evidence for a widespread oxygen-sensing mechanism. Proc Natl Acad Sci U S A, 90, 2423-7.
MAXWELL, P. H., WIESENER, M. S., CHANG, G. W., CLIFFORD, S. C., VAUX, E. C., COCKMAN, M. E., WYKOFF, C. C., PUGH, C. W., MAHER, E. R. & RATCLIFFE, P. J. 1999. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature, 399, 271-5.
MELILLO, G., TAYLOR, L. S., BROOKS, A., MUSSO, T., COX, G. W. & VARESIO, L. 1997. Functional requirement of the hypoxia-responsive element in the activation of the inducible nitric oxide synthase promoter by the iron chelator desferrioxamine. J Biol Chem, 272, 12236-43.
MORINET, F., CASETTI, L., FRANCOIS, J. H., CAPRON, C. & PILLET, S. 2013. Oxygen tension level and human viral infections. Virology, 444, 31-6.
MUKHOPADHYAY, C. K., MAZUMDER, B. & FOX, P. L. 2000. Role of hypoxia-inducible factor-1 in transcriptional activation of ceruloplasmin by iron deficiency. J Biol Chem, 275, 21048-54.
NADLER, L. M., RITZ, J., HARDY, R., PESANDO, J. M., SCHLOSSMAN, S. F. & STASHENKO, P. 1981. A unique cell surface antigen identifying lymphoid malignancies of B cell origin. J Clin Invest, 67, 134-40.
NAJAFIPOUR, H. & FERRELL, W. R. 1995. Comparison of synovial PO2 and sympathetic vasoconstrictor responses in normal and acutely inflamed rabbit knee joints. Exp Physiol, 80, 209-20.
NAKAYAMA, K., FREW, I. J., HAGENSEN, M., SKALS, M., HABELHAH, H., BHOUMIK, A., KADOYA, T., ERDJUMENT-BROMAGE, H., TEMPST, P., FRAPPELL, P. B., BOWTELL, D. D. & RONAI, Z. 2004. Siah2 regulates stability of prolyl-hydroxylases, controls HIF1alpha abundance, and modulates physiological responses to hypoxia. Cell, 117, 941-52.
NIELSEN, H. 1987. Isolation and functional activity of human blood monocytes after adherence to plastic surfaces: comparison of different detachment methods. Acta Pathol Microbiol Immunol Scand C, 95, 81-4.
NISHIMURA, M., MITSUNAGA, S. & JUJI, T. 2001. Frozen-stored granulocytes can be used for an immunofluorescence test to detect granulocyte antibodies. Transfusion, 41, 1268-72.
NISHIMURA, R. & OKUDA, K. 2010. Hypoxia is important for establishing vascularization during corpus luteum formation in cattle. J Reprod Dev, 56, 110-6.
85
NIZET, V. & JOHNSON, R. S. 2009. Interdependence of hypoxic and innate immune responses. Nat Rev Immunol, 9, 609-17.
NOORDEN, S. V. 2000. Immunocytochemistry. Methods in molecular medicine, 40, 415-37.
O'ROURKE, J. F., PUGH, C. W., BARTLETT, S. M. & RATCLIFFE, P. J. 1996. Identification of hypoxically inducible mRNAs in HeLa cells using differential-display PCR. Role of hypoxia-inducible factor-1. Eur J Biochem, 241, 403-10.
O'ROURKE, J. F., TIAN, Y. M., RATCLIFFE, P. J. & PUGH, C. W. 1999. Oxygen-regulated and transactivating domains in endothelial PAS protein 1: comparison with hypoxia-inducible factor-1alpha. J Biol Chem, 274, 2060-71.
OBERHOLZER, M., ETTLIN, R. A., KLOPPEL, G., LOHR, M., BUSER, M. W., MIHATSCH, M. J., CHRISTEN, H., MARTI, U. & HEITZ, P. U. 1987. Morphometry and immunocytochemistry. Anal Quant Cytol Histol, 9, 123-32.
OHMANN, H. B., FILION, L. G. & BABIUK, L. A. 1983. Bovine monocytes and macrophages: an accessory role in suppressor-cell generation by Con A and in lectin-induced proliferation. Immunology, 50, 189-97.
PEYSSONNAUX, C., CEJUDO-MARTIN, P., DOEDENS, A., ZINKERNAGEL, A. S., JOHNSON, R. S. & NIZET, V. 2007. Cutting edge: Essential role of hypoxia inducible factor-1alpha in development of lipopolysaccharide-induced sepsis. J Immunol, 178, 7516-9.
PEYSSONNAUX, C., DATTA, V., CRAMER, T., DOEDENS, A., THEODORAKIS, E. A., GALLO, R. L., HURTADO-ZIOLA, N., NIZET, V. & JOHNSON, R. S. 2005. HIF-1alpha expression regulates the bactericidal capacity of phagocytes. J Clin Invest, 115, 1806-15.
PIPER, E. K., JONSSON, N. N., GONDRO, C., LEW-TABOR, A. E., MOOLHUIJZEN, P., VANCE, M. E. & JACKSON, L. A. 2009. Immunological profiles of Bos taurus and Bos indicus cattle infested with the cattle tick, Rhipicephalus (Boophilus) microplus. Clin Vaccine Immunol, 16, 1074-86.
PIPIYA, T., SAUTHOFF, H., HUANG, Y. Q., CHANG, B., CHENG, J., HEITNER, S., CHEN, S., ROM, W. N. & HAY, J. G. 2005. Hypoxia reduces adenoviral replication in cancer cells by downregulation of viral protein expression. Gene Ther, 12, 911-7.
PISANI, D. F. & DECHESNE, C. A. 2005. Skeletal muscle HIF-1alpha expression is dependent on muscle fiber type. J Gen Physiol, 126, 173-8.
POUYSSEGUR, J. & MECHTA-GRIGORIOU, F. 2006. Redox regulation of the hypoxia-inducible factor. Biol Chem, 387, 1337-46.
QUINN, P. J., MARKEY, B. K., LEONARD, F. C., FITZPATRICK, E. S., FANNING, S. & HARTIGAN, P. J. 2011. Veterinary microbiology and microbial disease, Chichester, West Sussex, UK, Wiley-Blackwell.
RICHARD, D. E., BERRA, E. & POUYSSEGUR, J. 2000. Nonhypoxic pathway mediates the induction of hypoxia-inducible factor 1alpha in vascular smooth muscle cells. J Biol Chem, 275, 26765-71.
RIUS, J., GUMA, M., SCHACHTRUP, C., AKASSOGLOU, K., ZINKERNAGEL, A. S., NIZET, V., JOHNSON, R. S., HADDAD, G. G. & KARIN, M. 2008. NF-kappaB links innate immunity to the hypoxic response through transcriptional regulation of HIF-1alpha. Nature, 453, 807-11.
RIVERA-RIVAS, J. J., KISIELA, D. & CZUPRYNSKI, C. J. 2009. Bovine herpesvirus type 1 infection of bovine bronchial epithelial cells increases neutrophil adhesion and activation. Vet Immunol Immunopathol, 131, 167-76.
ROBINSON, A., KEELY, S., KARHAUSEN, J., GERICH, M. E., FURUTA, G. T. & COLGAN, S. P. 2008. Mucosal protection by hypoxia-inducible factor prolyl hydroxylase inhibition. Gastroenterology, 134, 145-55.
86
ROLFS, A., KVIETIKOVA, I., GASSMANN, M. & WENGER, R. H. 1997. Oxygen-regulated transferrin expression is mediated by hypoxia-inducible factor-1. J Biol Chem, 272, 20055-62.
SAADI, S., WRENSHALL, L. E. & PLATT, J. L. 2002. Regional manifestations and control of the immune system. FASEB J, 16, 849-56.
SAGER, H., DAVIS, W. C., DOBBELAERE, D. A. & JUNGI, T. W. 1997. Macrophage-parasite relationship in theileriosis. Reversible phenotypic and functional dedifferentiation of macrophages infected with Theileria annulata. J Leukoc Biol, 61, 459-68.
SALCEDA, S. & CARO, J. 1997. Hypoxia-inducible factor 1alpha (HIF-1alpha) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem, 272, 22642-7.
SALNIKOW, K., DONALD, S. P., BRUICK, R. K., ZHITKOVICH, A., PHANG, J. M. & KASPRZAK, K. S. 2004. Depletion of intracellular ascorbate by the carcinogenic metals nickel and cobalt results in the induction of hypoxic stress. J Biol Chem, 279, 40337-44.
SANDAU, K. B., FANDREY, J. & BRUNE, B. 2001a. Accumulation of HIF-1alpha under the influence of nitric oxide. Blood, 97, 1009-15.
SANDAU, K. B., ZHOU, J., KIETZMANN, T. & BRUNE, B. 2001b. Regulation of the hypoxia-inducible factor 1alpha by the inflammatory mediators nitric oxide and tumor necrosis factor-alpha in contrast to desferroxamine and phenylarsine oxide. J Biol Chem, 276, 39805-11.
SCHAIBLE, B., SCHAFFER, K. & TAYLOR, C. T. 2010. Hypoxia, innate immunity and infection in the lung. Respir Physiol Neurobiol, 174, 235-43.
SCHOFIELD, C. J. & RATCLIFFE, P. J. 2004. Oxygen sensing by HIF hydroxylases. Nat Rev Mol Cell Biol, 5, 343-54.
SEAGROVES, T. N., RYAN, H. E., LU, H., WOUTERS, B. G., KNAPP, M., THIBAULT, P., LADEROUTE, K. & JOHNSON, R. S. 2001. Transcription factor HIF-1 is a necessary mediator of the pasteur effect in mammalian cells. Mol Cell Biol, 21, 3436-44.
SEMENZA, G. L. 2000. Surviving ischemia: adaptive responses mediated by hypoxia-inducible factor 1. J Clin Invest, 106, 809-12.
SEMENZA, G. L. 2003. Targeting HIF-1 for cancer therapy. Nat Rev Cancer, 3, 721-32.
SEMENZA, G. L. 2005. Pulmonary vascular responses to chronic hypoxia mediated by hypoxia-inducible factor 1. Proceedings of the American Thoracic Society, 2, 68-70.
SEMENZA, G. L., NEJFELT, M. K., CHI, S. M. & ANTONARAKIS, S. E. 1991. Hypoxia-inducible nuclear factors bind to an enhancer element located 3' to the human erythropoietin gene. Proc Natl Acad Sci U S A, 88, 5680-4.
SEMENZA, G. L., ROTH, P. H., FANG, H. M. & WANG, G. L. 1994. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem, 269, 23757-63.
SEMENZA, G. L. & WANG, G. L. 1992. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol, 12, 5447-54.
SHIBATA, T., GIACCIA, A. J. & BROWN, J. M. 2000. Development of a hypoxia-responsive vector for tumor-specific gene therapy. Gene Ther, 7, 493-8.
SHWEIKI, D., ITIN, A., SOFFER, D. & KESHET, E. 1992. Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature, 359, 843-5.
87
SIMMEN, H. P., BATTAGLIA, H., GIOVANOLI, P. & BLASER, J. 1994. Analysis of pH, pO2 and pCO2 in drainage fluid allows for rapid detection of infectious complications during the follow-up period after abdominal surgery. Infection, 22, 386-9.
SITKOVSKY, M. & LUKASHEV, D. 2005. Regulation of immune cells by local-tissue oxygen tension: HIF1 alpha and adenosine receptors. Nat Rev Immunol, 5, 712-21.
SITKOVSKY, M. V. 2009. T regulatory cells: hypoxia-adenosinergic suppression and re-direction of the immune response. Trends Immunol, 30, 102-8.
SOPP, P. & HOWARD, C. J. 1997. Cross-reactivity of monoclonal antibodies to defined human leucocyte differentiation antigens with bovine cells. Vet Immunol Immunopathol, 56, 11-25.
SRIKUMARAN, S., KELLING, C. L. & AMBAGALA, A. 2007. Immune evasion by pathogens of bovine respiratory disease complex. Anim Health Res Rev, 8, 215-29.
STERNBERGER, L. A. 1979. Immunocytochemistry, New York, Wiley.
STROBER, W. 2001. Trypan blue exclusion test of cell viability. Curr Protoc Immunol, Appendix 3, Appendix 3B.
SUN, T., AKALIN, A., RODACKER, M. & BRAUN, T. 2004. CD20 positive T cell lymphoma: is it a real entity? J Clin Pathol, 57, 442-4.
SYNNESTVEDT, K., FURUTA, G. T., COMERFORD, K. M., LOUIS, N., KARHAUSEN, J., ELTZSCHIG, H. K., HANSEN, K. R., THOMPSON, L. F. & COLGAN, S. P. 2002. Ecto-5'-nucleotidase (CD73) regulation by hypoxia-inducible factor-1 mediates permeability changes in intestinal epithelia. J Clin Invest, 110, 993-1002.
TAKEDA, N., O'DEA, E. L., DOEDENS, A., KIM, J. W., WEIDEMANN, A., STOCKMANN, C., ASAGIRI, M., SIMON, M. C., HOFFMANN, A. & JOHNSON, R. S. 2010. Differential activation and antagonistic function of HIF-{alpha} isoforms in macrophages are essential for NO homeostasis. Genes Dev, 24, 491-501.
THIEL, M., CALDWELL, C. C., KRETH, S., KUBOKI, S., CHEN, P., SMITH, P., OHTA, A., LENTSCH, A. B., LUKASHEV, D. & SITKOVSKY, M. V. 2007. Targeted deletion of HIF-1alpha gene in T cells prevents their inhibition in hypoxic inflamed tissues and improves septic mice survival. PLoS One, 2, e853.
THIEL, M., CHOUKER, A., OHTA, A., JACKSON, E., CALDWELL, C., SMITH, P., LUKASHEV, D., BITTMANN, I. & SITKOVSKY, M. V. 2005. Oxygenation inhibits the physiological tissue-protecting mechanism and thereby exacerbates acute inflammatory lung injury. PLoS Biol, 3, e174.
THOMAS, D. D., RIDNOUR, L. A., ISENBERG, J. S., FLORES-SANTANA, W., SWITZER, C. H., DONZELLI, S., HUSSAIN, P., VECOLI, C., PAOLOCCI, N., AMBS, S., COLTON, C. A., HARRIS, C. C., ROBERTS, D. D. & WINK, D. A. 2008. The chemical biology of nitric oxide: implications in cellular signaling. Free Radic Biol Med, 45, 18-31.
TIAN, H., HAMMER, R. E., MATSUMOTO, A. M., RUSSELL, D. W. & MCKNIGHT, S. L. 1998. The hypoxia-responsive transcription factor EPAS1 is essential for catecholamine homeostasis and protection against heart failure during embryonic development. Genes Dev, 12, 3320-4.
TIAN, H., MCKNIGHT, S. L. & RUSSELL, D. W. 1997. Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev, 11, 72-82.
TOLNAY, A. E., BASKIN, C. R., TUMPEY, T. M., SABOURIN, P. J., SABOURIN, C. L., LONG, J. P., PYLES, J. A., ALBRECHT, R. A., GARCIA-SASTRE, A., KATZE, M. G. & BIELEFELDT-OHMANN, H. 2010. Extrapulmonary tissue responses in cynomolgus macaques (Macaca fascicularis) infected with highly pathogenic avian influenza A (H5N1) virus. Arch Virol, 155, 905-14.
VAN UDEN, P., KENNETH, N. S. & ROCHA, S. 2008. Regulation of hypoxia-inducible factor-1alpha by NF-kappaB. Biochem J, 412, 477-84.
88
VASSILAKI, N., KALLIAMPAKOU, K. I., KOTTA-LOIZOU, I., BEFANI, C., LIAKOS, P., SIMOS, G., MENTIS, A. F., KALLIAROPOULOS, A., DOUMBA, P. P., SMIRLIS, D., FOKA, P., BAUHOFER, O., POENISCH, M., WINDISCH, M. P., LEE, M. E., KOSKINAS, J., BARTENSCHLAGER, R. & MAVROMARA, P. 2013. Low oxygen tension enhances hepatitis C virus replication. J Virol, 87, 2935-48.
VORDERMARK, D. & BROWN, J. M. 2003. Evaluation of hypoxia-inducible factor-1alpha (HIF-1alpha) as an intrinsic marker of tumor hypoxia in U87 MG human glioblastoma: in vitro and xenograft studies. Int J Radiat Oncol Biol Phys, 56, 1184-93.
WALMSLEY, S. R., PRINT, C., FARAHI, N., PEYSSONNAUX, C., JOHNSON, R. S., CRAMER, T., SOBOLEWSKI, A., CONDLIFFE, A. M., COWBURN, A. S., JOHNSON, N. & CHILVERS, E. R. 2005. Hypoxia-induced neutrophil survival is mediated by HIF-1alpha-dependent NF-kappaB activity. J Exp Med, 201, 105-15.
WALSHE, T. E. & D'AMORE, P. A. 2008. The role of hypoxia in vascular injury and repair. Annu Rev Pathol, 3, 615-43.
WANG, G. L., JIANG, B. H., RUE, E. A. & SEMENZA, G. L. 1995. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A, 92, 5510-4.
WANG, G. L. & SEMENZA, G. L. 1993a. Characterization of hypoxia-inducible factor 1 and regulation of DNA binding activity by hypoxia. J Biol Chem, 268, 21513-8.
WANG, G. L. & SEMENZA, G. L. 1993b. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood, 82, 3610-5.
WANG, G. L. & SEMENZA, G. L. 1993c. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci U S A, 90, 4304-8.
WANG, G. L. & SEMENZA, G. L. 1995. Purification and characterization of hypoxia-inducible factor 1. J Biol Chem, 270, 1230-7.
WIENER, C. M., BOOTH, G. & SEMENZA, G. L. 1996. In vivo expression of mRNAs encoding hypoxia-inducible factor 1. Biochem Biophys Res Commun, 225, 485-8.
WIESENER, M. S., TURLEY, H., ALLEN, W. E., WILLAM, C., ECKARDT, K. U., TALKS, K. L., WOOD, S. M., GATTER, K. C., HARRIS, A. L., PUGH, C. W., RATCLIFFE, P. J. & MAXWELL, P. H. 1998. Induction of endothelial PAS domain protein-1 by hypoxia: characterization and comparison with hypoxia-inducible factor-1alpha. Blood, 92, 2260-8.
WU, D. & YOTNDA, P. 2011. Induction and testing of hypoxia in cell culture. J Vis Exp.
XING, D. & BONANNO, J. A. 2009. Hypoxia preconditioning protection of corneal stromal cells requires HIF1alpha but not VEGF. Mol Vis, 15, 1020-7.
YANG, J., FU, Z., FENG, X., SHI, Y., YUAN, C., LIU, J., HONG, Y., LI, H., LU, K. & LIN, J. 2012. Comparison of worm development and host immune responses in natural hosts of Schistosoma japonicum, yellow cattle and water buffalo. BMC Vet Res, 8, 25.
YU, A. Y., FRID, M. G., SHIMODA, L. A., WIENER, C. M., STENMARK, K. & SEMENZA, G. L. 1998. Temporal, spatial, and oxygen-regulated expression of hypoxia-inducible factor-1 in the lung. Am J Physiol, 275, L818-26.
YU, F., WHITE, S. B., ZHAO, Q. & LEE, F. S. 2001. HIF-1alpha binding to VHL is regulated by stimulus-sensitive proline hydroxylation. Proc Natl Acad Sci U S A, 98, 9630-5.
YUAN, Y., HILLIARD, G., FERGUSON, T. & MILLHORN, D. E. 2003. Cobalt inhibits the interaction between hypoxia-inducible factor-alpha and von Hippel-Lindau protein by direct binding to hypoxia-inducible factor-alpha. J Biol Chem, 278, 15911-6.
89
ZHANG, S. X., GOZAL, D., SACHLEBEN, L. R., JR., RANE, M., KLEIN, J. B. & GOZAL, E. 2003. Hypoxia induces an autocrine-paracrine survival pathway via platelet-derived growth factor (PDGF)-B/PDGF-beta receptor/phosphatidylinositol 3-kinase/Akt signaling in RN46A neuronal cells. FASEB J, 17, 1709-11.
ZHOU, J. & BRUNE, B. 2006. Cytokines and hormones in the regulation of hypoxia inducible factor-1alpha (HIF-1alpha). Cardiovasc Hematol Agents Med Chem, 4, 189-97.
ZHOU, J., FANDREY, J., SCHUMANN, J., TIEGS, G. & BRUNE, B. 2003. NO and TNF-alpha released from activated macrophages stabilize HIF-1alpha in resting tubular LLC-PK1 cells. Am J Physiol Cell Physiol, 284, C439-46.
ZINKERNAGEL, A. S., JOHNSON, R. S. & NIZET, V. 2007. Hypoxia inducible factor (HIF) function in innate immunity and infection. J Mol Med, 85, 1339-46.
90
9 Appendices
9.1 Initial T lymphocyte CD3 antibody
Given the relative rarity of bovine-specific antibodies, or antibodies with reliable reports of bovine
cross-reactivity, the selection of ideal antibodies was always going to be a challenge. The initial
antibody selected as a T-cell phenotypic marker was an Allophycocyanin (APC) conjugated mouse
anti-human monoclonal CD3 IgG1 antibody (MCA463APC AbD Serotec, Oxford, UK) APC was a
bright fluorochrome and had minimal spectral overlap with FITC, the original fluorochrome choice
for labelling the HIF-1α antibody.
Titrations with increasing concentrations of antibody on flow cytometry unfortunately failed to
show an emerging distinct population of CD3-APC positive cells with all cells being contained
within one peak on the histogram. With titration of this CD3 antibody there was only a gradual,
slight, generalised increase in fluorescence across all cells (i.e., events); an increase which rose in
proportion with increasing antibody concentration. This pattern was suggestive of non-specific
binding. This apparent absence of true staining was confirmed by examining cells stained with
moderate CD3-APC antibody concentration mixed with aliquot of unstained cells from the same
animal at a ratio of 1:1 using the flow cytometer. The absence of a second, slightly higher distinct
peak of APC positive cells, or at least a change in shape of the upper contour of the peak
demonstrated that a known unstained population of cells had the same CD3-APC fluorescence
profile as a known stained population, thereby confirming that the ‘stained’ population was likely
background staining rather than target staining, i.e., the CD3-APC antibody was not binding to any
specific subpopulation of cells present.
9.2 CD20 antibody titration on flow cytometry
Originally, this antibody was used in unconjugated form as a primary antibody, followed by
application of a secondary goat anti-mouse polyclonal IgG antibody (Biolegend, San Diego, CA,
USA) conjugated commercially to the fluorochrome APC-Cy7. However, such use of a secondary
antibody was discontinued due to on-going non-specific staining results when titrating both the
primary and secondary antibodies. Bovine serum albumin was employed as a blocking agent to
block Fc receptors on the cell surface to attempt to decrease non-specific binding of the secondary
antibody in the staining process in an attempt to improve specificity, but results did not improve. It
became apparent that use of secondary conjugated antibodies was too problematic when attempting
to create a multicolour flow cytometry protocol.
91
9.3 Initial CD14 antibody
The initially chosen phenotypic antibody for monocytes was a Pacific Blue conjugated mouse anti-
human monoclonal CD14 IgG2a antibody (clone TÜK4) (AbD Serotec, Oxford, UK). This clone
was selected as it had recorded cross-reactivity with bovine CD14 for flow cytometry (Sopp and
Howard, 1997). Initial titration showed an emerging positive sub-population of cells however
further dual-staining with the already confirmed CD3 - Alexa Fluor® 647 antibody showed that
80% of the cells positive for CD14-Pac Blue were also positive for CD3 - Alexa Fluor® 647 (figure
4.4).
In order to identify the constituent cells in the population, Cells positive for both CD3 and CD14
were gated and plotted on a FS/SS scatter plot in linear scale. The majority of events within the
lymphocyte region were positive, with lesser numbers of cells within the monocyte area staining, as
well as debris (graph not shown). Given that the CD3 antibody had already been titrated
successfully, had specifically stained T lymphocytes on IHC, and that the majority of cells in the
sample were expected to be lymphocytes, a specificity problem with the CD14 antibody seemed the
most likely explanation. This was further supported by lack of positive staining in the monocyte
region. Further IHC testing using this CD14 TÜK4 clone at varying concentrations and following
use of EDTA, citrate and proteinase K target retrieval systems failed to detect clear positivity in
monocytes or other cells.
Figure 4.4. Bivariate scatter plot of CD3 fluorescent signal versus CD14 fluorescence, divided into quadrants, log scale. The negative quadrant was set based on unstained cells. The majority of events fall within the negative quadrant or within the ‘both CD14/CD3’ quadrant, indicating double-staining of these events with both CD14 and CD3 antibody. 9.3.1.1 Alternative monocyte CD14 antibody
An alternative mouse monoclonal IgG1 antibody clone MM61A (VMRD, Pullman WA, USA) had
been used successfully to label bovine monocytes for flow cytometry (Piper et al., 2009) (Sager et
92
al., 1997). Titration on uncultured cells labeled with a Pacific Blue fluorophore (using the same
labeling kit as for CD20 for a labeled antibody concentration of 15 µg/ml) showed a positive peak
emerging at 50 µl of antibody (figure 4.5A) and a relatively distinct positive population on the
SS/CD14 scatter plot (figure 4.5B). Workup was discontinued at this point due to removal of
monocytes from the flow analysis.
Figure 4.5. A: Histogram of CD14 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD14 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.5B. B: Scatter plot of CD14 fluorescence versus SS, log scale. A population of cells positive for CD14 is visualised and gated by eye (black border).