Biochemical Characterization of a Novel Extended-Spectrum β -Lactamase...

Post on 04-Apr-2017

214 views 0 download

Transcript of Biochemical Characterization of a Novel Extended-Spectrum β -Lactamase...

MICROBIAL DRUG RESISTANCEVolume 8, Number 1, 2002© Mary Ann Liebert, Inc.

Biochemical Characterization of a Novel Extended-Spectrumb-Lactamase from Pseudomonas aeruginosa 802

SAMIA REJIBA, FERID LIMAM, CHERIFA BELHADJ, OMRANE BELHADJ, and KAMEL BEN-MAHREZ

ABSTRACT

Pseudomonas aeruginosa 802 was isolated at Rabta hospital in Tunis and was resistant to extended-spectrumcephalosporins and aztreonam. It produced a pI 7.6 extended-spectrum b-lactamase (ESBL). The ESBL,named LBT 802, was purified to homogeneity by filtration on Sephadex G-75 followed by CM-Sepharosechromatography and high-performance liquid chromatography (HPLC) on a TSK-gel SP-5PW column. TheLBT 802 enzyme had a molecular mass of 30 kDa. It showed a broad-substrate profile by hydrolyzing ben-zylpenicillin, ampicillin, cephalothin, cephaloridine, cefotaxime, ceftriaxone, and cefpirome but not cef-tazidime, cefoxitin, imipenem, or aztreonam. The highest hydrolytic efficiency (Vmax/Km) was obtained forampicillin, cephalothin, cephaloridine, and benzylpenicillin. Among extended-spectrum cephalosporins thebest substrate was ceftriaxone followed by cefotaxime and cefpirome. LBT 802 activity was inhibited by clavu-lanic acid, sulbactam, imipenem, cefoxitin, and aztreonam. It showed its lowest Ki values for clavulanic acid,imipenem and sulbactam.

9

INTRODUCTION

b-LACTAMASE PRODUCTION is the most prevalent mechanismof resistance to b-lactam drugs among Gram-negative bac-

teria. Since the introduction into clinical use of extended-spec-trum cephalosporins such as ceftriaxone, cefotaxime, cef-pirome, and ceftazidime, bacteria have reacted to them mainlyby production of extended-spectrum b-lactamase (ESBLs).Most of them are members of Ambler class A and are relatedto the TEM and SHV types of b-lactamases. ESBLs have beenextensively described worldwide in members of the family En-terobacteriaceae.3

In Pseudomonas aeruginosa, the main mechanism of resis-tance to extended-spectrum cephalosporins is derepression ofthe chromosomal cephalosporinase, resulting in its overpro-duction.4 However, within the last years, some ESBLs werefound in P. aeruginosa. In this species, as opposed to membersof the family Enterobacteriaceae, production of TEM- andSHV-type ESBLs is rare. Thus, TEM-24,13 TEM-42,15 andSHV-2a16 are the only TEM- and SHV-type ESBLs reportedin this species. Furthermore, several other ESBLs have beendescribed. These include PER-117 (a class A enzyme), OXA-derived ESBLs of class D,5–7,9,14,19 and IMP-1,11 VIM-1,12 andVIM-2,20 which are class B metalloenzymes. Nevertheless, only

TEM-24,13 TEM-42,15 SHV-2a,16 PER-1,17 and OXA-1819 areinhibited by clavulanic acid.

In a previous study, we have demonstrated the involvementof a b-lactamase-mediated mechanism in b-lactam resistancein a clinical isolate of P. aeruginosa802.1 This strain expressedan inducible-chromosomal cephalosporinase and a secondaryb-lactamase of pI 7.6, LBT 802. A special interest in this sec-ondary b-lactamase came from its hydrolytic activity againstextended-spectrum cephalosporins and its inhibition by clavu-lanic acid and imipenem.

In this paper, we describe the purification protocol of thissecondary b-lactamase and determine its enzymatic character-istics, which suggest that it is a novel ESBL of P. aeruginosa.

MATERIALS AND METHODS

Bacterial strain

P. aeruginosa 802 was isolated in 1996 from a patient inRabta hospital in Tunis and was kindly provided by Dr. Ab-dellatif Boudabbous (Laboratoire de Microbiologie, Faculté desSciences de Tunis). The strain 802 (O:16 serotype) was resis-tant to extended-spectrum cephalosporins and aztreonam.

Laboratoire de Biochimie et de Technobiologie, Faculté des Sciences de Tunis, Tunis, Tunisia.

Crude extract preparation

A culture of P. aeruginosa802 was grown overnight in tryp-ticase soy broth with shaking at 37°C and was then diluted 25-fold into the same medium, prewarmed at 37°C. Incubation wascontinued for 6 hr to yield late-logarithmic-phase cells. Bacte-ria were harvested for 10 min at 8,000 rpm, washed once in 25mM sodium phosphate buffer at pH 7 (buffer A), and resus-pended in the same buffer (30 ml for a culture of 10 liters). Theresulting suspension was disrupted by sonication twice for 45sec on ice and was subsequently clarified by centrifugation at10,000 rpm for 20 min at 4°C. The protein concentration wasassayed by the method of Bradford using bovine serum albu-min (BSA) as standard.2

Detection of b-lactamases on native polyacrylamide gels

b-Lactamase activities were subjected to starch-7% poly-acrylamide gel electrophoresis and were revealed by the starch-iodine method using cephaloridine at 0.5 mM.18

SDS polyacrylamide gel electrophoresis

Electrophoresis on SDS-12% polyacrylamide gels was per-formed as described by Laemmli.10 Proteins were revealed bythe silver staining procedure.21

b-lactamase kinetics

The activity of b-lactamase was performed spectrophoto-metrically at 37°C in buffer A. Antibiotic solutions were freshlyprepared in buffer A. The following wavelengths were used:235 nm for benzylpenicillin, ampicillin, and carbenicillin; 297nm for imipenem; 255 nm for cephaloridine and ceftriaxone;262 nm for cephalothin; 265 nm for cefoxitin; 260 nm for ce-fotaxime and ceftazidime; 258 nm for cefpirome; and 318 nmfor aztreonam. One unit of b-lactamase was defined as the

amount of enzyme able to hydrolyze 1 mmol of substrate perminute at room temperature.

Kinetic parameters were derived from the initial velocity ob-tained with five substrate concentrations. Km and Vmax valueswere determined according to the Lineweaver-Burke represen-tation. In the case of substrates with undetectable Vmax values,enzyme-substrate affinity was measured as Ki.

Enzyme inhibition was studied with cephaloridine as sub-strate at 50 mM. Five potential inhibitors were tested: clavu-lanic acid, imipenem, sulbactam, cefoxitin, and aztreonam. In-hibitor, at various concentrations, was preincubated withenzyme for 10 min at 37°C before addition of the substrate. Theinhibitor concentration required to inhibit 50% of enzyme ac-tivity (IC50) was determined graphically. Ki values were de-duced from the Dixon representation.

RESULTS

Purification of the LBT 802 b-lactamase

The crude extract obtained from 10 liters of bacterial culturewas concentrated by ammonium sulfate precipitation at 80% onice and resuspended in 4 ml of buffer A. The suspension wascentrifuged at 10,000 rpm for 10 min at 4°C. The supernatantwas then dialyzed overnight against the buffer A. The dialysatewas loaded onto a column (95 3 2 cm) of Sephadex G-75 equi-librated with buffer A. Proteins were eluted with the same bufferinto 2.5-ml fractions, and 20-ml aliquots were assayed spec-trophotometrically for b-lactamase activity with 50 mMcephaloridine as substrate. Two peaks of activity were detectedand analyzed by electrophoresis on native polyacrylamide gel(see Fig. 1). The b-lactamase of basic pI (peak A) correspondedto the chromosomal cephalosporinase, whereas the peak elutedfrom the column more slowly (peak B) contained the LBT 802b-lactamase.

REJIBA ET AL.10

FIG. 1. Elution profile of P. aeruginosab-lactamases on Sephadex G-75. Ammonium sulfate fraction was applied to a SephadexG-75 column previously equilibrated with buffer A. Proteins were eluted with the same buffer at a flow rate of 0.26 ml/min andfractions of 2.5 ml were collected. Absorbance of the effluent was monitored at 280 nm. Next 20-ml aliquots were assayed forb-lactamase activities by spectrophotometry with 50 mM of cephaloridine. (Inset) b-lactamase activity analysis by native poly-acrylamide gel electrophoresis. Fractions 42–64 were tested with 0.5 mM cephaloridine.

The LBT 802 b-lactamase fractions were pooled, dialyzedagainst 50 mM sodium acetate buffer at pH 5 (buffer B), andthen applied to a CM-Sepharose column (10.2 3 1.2 cm) equi-librated with buffer B. The adsorbed proteins were eluted witha linear gradient of 0–200 mM NaCl in buffer B. Next, 2.3-mlfractions were collected and 10-ml aliquots were assayed for b-lactamase activity. The LBT 802 activity was eluted at 140 mMNaCl.

The active fractions were pooled and concentrated with aCentricon-10 concentrator at 6,000 rpm for 6 hr at 4°C. NaClwas simultaneously removed by adding distilled water. High-performance liquid chromatography (HPLC) was carried out onTSK-gel SP-5PW column (75 3 7.5 mm) equilibrated withbuffer B. Desorption was obtained with a linear gradient of0–250 mM NaCl in buffer B. The b-lactamase was eluted at190 mM NaCl.

LBT 802 enzyme was purified with a specific activity of1,506 U/mg of protein. The purified preparation of b-lactamaseran as a single band when analyzed by SDS-PAGE, corre-sponding to a molecular weight of approximately 30 kDa (seeFig. 2).

Kinetic properties

The LBT 802 b-lactamase of the strain 802 was separatedfrom the chromosomal cephalosporinase by single-stepSephadex G-75 chromatography. Sephadex G-75 preparationwas also used to determine the kinetic parameters of the LBT802 b-lactamase that gave linear kinetics for all the tested sub-strates.

Vmax values relative to that obtained with benzylpenicillinand Km values for a set of b-lactam antibiotics are presented inTable 1. LBT 802 enzyme is an extended-spectrum b-lacta-mase. It exhibited hydrolytic activity against penicillins, earlycephalosporins, and extended-spectrum cephalosporins. The b-lactamase showed higher Vmax values for ampicillin, ben-zylpenicillin, cephaloridine, and cephalothin than for newercephalosporins. It had the elevated hydrolysis rate for ampi-cillin, Vmax values between 86 and 143 were seen forcephalothin, benzylpenicillin, and cephaloridine. Among ex-tended-spectrum cephalosporins tested, the highest Vmax valuewas for cefpirome, whereas the lowest one was for ceftriaxone.Hydrolysis of carbenicillin, ceftazidime, cefoxitin, and aztre-onam was not detectable.

Although the hydrolysis rates for ceftriaxone and cefotaximewere low, the LBT 802 b-lactamase had a higher affinity forthese drugs with Km values of 3.90 and 16.39 mM, respectively.The highest Km value among the tested b-lactams was for cef-pirome. Penicillins and early cephalosporins showed Km valuesranging from 24 to 86 mM.

The hydrolytic efficiency (Vmax/Km) take both values into ac-count. Hence, the best substrate in terms of Vmax/Km valueswere ampicillin, followed by cephalothin and cephaloridine,whereas Vmax/Km value was approximately two-fold lower forbenzylpenicillin. Because cefotaxime and ceftriaxone were rec-ognized efficiently, they behaved as good substrates withVmax/Km values of 0.80 and 0.98, respectively. Even though theVmax for cefpirome was greater than for the other extended-spectrum cephalosporins, the hydrolytic efficiency was low forthis substrate as a consequence of the low affinity.

Effect of inhibitors

Inhibition parameters (IC50 and Ki) for the LBT 802 b-lacta-mase are compared in Table 2. Analysis of Ki values showed thatthe highest affinity of the enzyme was for clavulanic acid, with aKi of 0.01 mM. This represents a 12- to 35-fold stronger bindingthan for imipenem and sulbactam. Ki values of cefoxitin and aztre-onam were 320- to 750-fold higher than that of clavulanic acid.Thus, the enzyme had the lowest affinity for aztreonam.

Analysis of IC50 values revealed that clavulanic acid actedas the most powerful inhibitor with IC50 of 0.02 mM, followed

EXTENDED-SPECTRUM b-LACTAMASE IN P. AERUGINOSA 11

TABLE 1. KINETIC PARAMETERS OF

THE LBT 802 b-LACTAMASE

Substrates Vmaxa Km (mM) Vmax /Km

Benzylpenicillin 100 57.50 1.73Ampicillin 328.57 85.50 3.84Cephalothin 86.26 24.96 3.45Cephaloridine 143.54 46.51 3.08Cefotaxime 13.25 16.39 0.80Cefpirome 38.30 100.00 0.38Ceftriaxone 3.86 3.90 0.98

aVmax values are relative to that of benzylpenicillin, whichwas set at 100.

FIG. 2. SDS-PAGE analysis of active fractions collected atdifferent purification steps. Lane 1, Crude extract after precip-itation with ammonium sulfate (7 mg of protein); lane 2,Sepharose G-75 (3 mg of protein); lane 3, CM-Sepharose (1.3mg of protein); lane 4, HPLC fraction (0.3 mg of protein); lane5, molecular weight markers (kDa).

TABLE 2. INHIBITION PROFILE OF THE LBT 802 b-LACTAMASE

Inhibitors Ki (mM) IC50 (mM)

Clavulanic acid 0.01 0.02Sulbactam 0.35 0.98Imipenem 0.12 0.30Cefoxitine 3.20 13.60Aztreonam 7.50 10.35

by imipenem and sulbactam with IC50 values 15- to 50-foldgreater than that of clavulanic acid. IC50 values of these in-hibitors paralleled the order of the Ki values. Aztreonam andcefoxitin with similar IC50 values of 10.35 mM and 13.60 mM,respectively, were relatively poor inhibitors.

DISCUSSION

LBT 802 b-lactamase had a broad-substrate profile, hy-drolyzing ampicillin, benzylpenicillin, cephalothin, cephalori-dine, cefotaxime, ceftriaxone, and cefpirome. Its activity wasinhibited by clavulanic acid, sulbactam, imipenem, cefoxitin,and aztreonam.

Determination of the kinetic parameters of the enzyme al-lowed us to make comparisons with other ESBLs reported inP. aeruginosa and revealed some interesting features of LBT802.

The highest hydrolytic efficiencies were obtained for ampi-cillin, cephalothin, cephaloridine, and benzylpenicillin, whereasVmax/Km values for cefotaxime, ceftriaxone, and cefpiromewere 4- to 10-fold lower than for ampicillin.

For penicillins, the structure of C6 side chain played an im-portant function.11 Ampicillin behaved as a good substrate(Vmax/Km). Substitution of the amino group by a carboxylicfunction for carbenicillin might explain why no hydrolysis ofthis substrate was detected spectrophotometrically. The com-parison of the kinetic parameters for the hydrolysis of earlycephalosporins (cephaloridine and cephalothin) and extended-spectrum cephalosporins (ceftriaxone, cefotaxime, and cef-pirome) suggested that the substitution at C7 decreased the hy-drolytic efficiency against extended-spectrum cephalosporins ofthe LBT 802 enzyme.8

As suggested by its biochemical properties, the LBT 802 en-zyme exhibits features that distinguish it from the previouslyreported ESBLs isolated from P. aeruginosa. These include:the class A enzymes; TEM-24,13 TEM-42,15 SHV-2a,16 andPER-117, OXA-derived ESBLs of class D5–7,9,14,19; and theclass B metalloenzymes IMP-1,11 VIM-1,12 and VIM-2.20 Theabsence of inhibition by EDTA (data not shown) for the LBT802 b-lactamase excludes this enzyme from the metalloenzymegroup.3

Our kinetic study demonstrated that LBT 802 differs fromESBLs of Ambler class D because it is inactive against isox-asolypenicillins (data not shown). Moreover, as opposite to thestrong inhibition of our enzyme by clavulanic acid, the OXA-derived ESBLs are usually poorly inhibited by this inhibitor,except for OXA-18. In fact, LBT 802 and OXA-18 are bothsusceptible to inhibition by clavulanic acid, imipenem, and sul-bactam. Nevertheless, OXA-18 was inhibited by clavulanic acidless efficiently than by imipenem.19 The protective effect ofimipenem on otherwise hydrolyzed b-lactam compounds in P.aeruginosahas also been described for PER-117 and OXA-19.14

According to the functional classification scheme describedby Bush et al.,3 one may classify LBT 802 enzyme within group2be because Vmax values, relative to that of benzylpenicillin, ofcefotaxime and cefpirome were .10%. As opposed to the ESBLs TEM-24,13 TEM-42,15 SHV-2a,16 and PER-1,17 no hy-drolytic activity against ceftazidime and aztreonam were de-tected with the LBT 802 enzyme.

LBT 802 differs from PER-1 by its pI and its low hydrolyticefficiencies, especially regarding extended-spectrum cepha-losporins (ceftriaxone, cefotaxime, and cefpirome). Moreover,inhibition effects of sulbactam, cefoxitin, and imipenem weremore potent in PER-117 than on LBT 802.

Enzymatic properties of TEM-42 were different from thoseexhibited by LBT 802. For instance, TEM-42 had the highestVmax and Km values for cefotaxime, and it was strongly inhib-ited by sulbactam (IC50 0.008 mM). In addition, this enzymehad an acid pI.15

Despite the notable differences between LBT 802 and theother ESBLs reported in P. aeruginosa, LBT 802 shares somesimilarities with them. Thus, Ki values of clavulanic acid wereclose for LBT 802 and PER-1 enzymes. LBT 802 had also thesame pI as SHV-2a16 and OXA-19.14

The potential clinical significance of LBT 802 and theuniqueness of some of its functional properties render it an in-teresting candidate for further studies of the molecular struc-ture and the interaction with potential inhibitor of b-lactamasesuch as imipenem.

ACKNOWLEDGMENTS

This work was supported by grants from the Secrétariat d’Etat à la Recherche Scientifique et à la Technologie and theDirection Générale de la Recherche Scientifique et Technique(Tunisia).

REFERENCES

1. Ben-Mahrez, K., S. Réjiba, C. BelHadj, and O. BelHadj. 1999. b-Lactamase-mediated resistance to extended spectrum cephalo-sporins among clinical isolates of Pseudomonas aeruginosa. Res.Microbiol. 150:403–406.

2. Bradford, M.M. 1976. A rapid and sensitive method for quanti-tation of microgram quantities of protein utilising a principal ofprotein-dye binding. Anal. Biochem. 72:248–254.

3. Bush, K., G.A. Jacoby, and A.A. Medeiros. 1995. A functionalclassification scheme for b-lactamases and its correlation with mo-lecular structure. Antimicrob. Agents Chemother. 39:1211–1233.

4. Colom, K., A. Fdz-Aranguiz, E. Suinaga, and R. Cisterna. 1995.Emergence of resistance to beta-lactam agents in Pseudomonasaeruginosa with group I beta-lactamases in Spain. Eur. J. Clin. Mi-crobiol. Infect. Dis. 14:964–971.

5. Danel, F., L.M.C. Hall, B. Duke, D. Gur, and D.M. Livermore.1999. OXA-17, a further extended-spectrum variant of OXA-10 b-lactamase, isolated from Pseudomonas aeruginosa. Antimicrob.Agents Chemother. 43:1362–1366.

6. Danel, F., L.M.C. Hall, D. Gur, and D.M. Livermore. 1995.OXA-14 another extended-spectrum variant of OXA-10 (PSE-2)b-lactamase from Pseudomonas aeruginosa. Antimicrob. AgentsChemother. 39:1881–1884.

7. Danel, F., L.M.C. Hall, D. Gur, and D.M. Livermore. 1998.OXA-16, a further extended-spectrum variant of OXA-10 b-lacta-mase, from two Pseudomonas aeruginosa isolates. Antimicrob.Agents Chemother. 42:3117–3122.

8. Donowitz, G.R., and G.L. Mandell. 1988. Beta-lactam antibiotics.N. Engl. J. Med. 318:419–426.

9. Hall, L.M.C., D.M. Livermore, D. Gur, M. Akova, and H.E.Akalin. 1993. OXA-11, an extended-spectrum variant of OXA-10

REJIBA ET AL.12

(PSE-2) b-lactamase from Pseudomonas aeruginosa. Antimicrob.Agents Chemother. 37:1637–1644.

10. Laemmli, U.K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature. 227:680–685.

11. Laraki, N., N. Franceschini, G.M. Rossolini, P. Santucci, C. Me-unier, E. De Pauw, G. Amicosante, J.M. Frère, and M. Galleni.1999. Biochemical characterization of the Pseudomonas aerugi-nosa 101/1477 metallo-b-lactamase IMP-1 produced by Esch-erichia coli. Antimicrob. Agents Chemother. 43:902–906.

12. Lauretti, L., M.L. Riccio, A. Mazzariol, G. Cornaglia, G. Ami-cosante, R. Fontana, and G.M. Rossolini. 1999. Cloning andcharacterization of blaVIM, a new integron-borne metallo-b-lacta-mase gene from a Pseudomonas aeruginosa clinical isolate. An-timicrob. Agents Chemother. 43:1584–1590.

13. Marchandin, H., H. Jean-Pierre, C. De Champs, D. Sirot, H.Darbas, P.F. Perigault, and C. Carrière. 2000. Production of aTEM-24 plasmid-mediated extented-spectrum b-lactamase by aclinical isolate of Pseudomonas aeruginosa. Antimicrob. AgentsChemother. 44:213–216.

14. Mugnier, P., I. Casin, T.T. Bouthors, and E. Collatz. 1998.Novel OXA10-derived extended-spectrum b-lactamase selected invivo or in vitro. Antimicrob. Agents Chemother. 42:3113–3116.

15. Mugnier, P., P. Dubrous, I. Casin, G. Arlet, and E. Collatz. 1996.A TEM-derived extended-spectrum b-lactamase in Pseudomonasaeruginosa. Antimicrob. Agents Chemother. 40:2488–2493.

16. Naas, T., L. Philippon, L. Poirel, E. Ronco, and P. Nordmann.1999. An SHV-derived extended-spectrum b-lactamase inPseudomonas aeruginosa. Antimicrob. Agents Chemother.43:1281–1284.

17. Nordmann, P., E. Ronco, T. Naas, C. Duport, Y. Michel-Briand,and R. Labia. 1993. Characterization of a novel extended-spec-trum b-lactamase from Pseudomonas aeruginosa. Antimicrob.Agents Chemother. 37:962–969.

18. Perret, C.J. 1954. Iodometric assay of penicillinase. Nature154:1012–1013.

19. Philippon, L.N., T. Naas, A.T. Bouthors, V. Barakett, and P.Nordmann. 1997. OXA-18, a class D acid-inhibited extendedspectrum b-lactamase from Pseudomonas aeruginosa. Antimicrob.Agents Chemother. 41:2188–2195.

20. Poirel, L., T. Naas, D. Nicolas, L. Collet, S. Bellais, J.D. Cav-allo, and P. Nordmann. 2000. Characterization of VIM-2, a car-bapenem-hydrolyzing metallo-b-lactamase and its plasmid- and in-tegron-borne gene from a Pseudomonas aeruginosa clinical isolatein France. Antimicrob. Agents Chemother. 44:891–897.

21. Wray, W., T. Boulikas, V.P. Wray, and R. Hancokc. 1981. Sil-ver staining of proteins in polyacrylamide gels. Anal. Biochem.118:197–203.

Address reprint requests to:Dr. Kamel Ben-Mahrez

Laboratoire de Biochimie et de TechnobiologieFaculté des Sciences de Tunis

2092 EL-Manar IITunis, Tunisia

E-mail: Omrane.belhadj@fst.rnu.tn

EXTENDED-SPECTRUM b-LACTAMASE IN P. AERUGINOSA 13