FEMS MicrobiologyLetters 96 (19921125-128 © 1992 Federation of European MicrobiologicalSocieties0378-1097/92/$05.00 Published by Elsevier
125
FEMSLE 05018
Esterase electrophoresis compared with biotyping for epidemiological typing of Acinetobacter baumannii strains P. Pou/~dras a, S. G r a s a, J.M. Sire ~, R. M e s n a r d ~, P.Y. D o n n i o ", B. Picard b and J.L. Avril " a Laboratoire de Bactt~riologie-Virologie. Ht~pital Uni~'ersitaire Pontchaillou. Remws. France. and ~'Laboratoire de Microbiologie, Facult~ de Mt:dechwXacier Bichat, Unicersit~ Paris 7. Paris. France
Received 14 April 1992 Revision received I June 1992 Accepted 5 June 1992 Key words: Acinetobacter; Esterase; Biotype; Epidemiology; Zymotype
1. S U M M A R Y Forty-nine Acinetobacter baumannii strains belenging to three biotypes and isolated from four hospitals were differentiated by electrophoretic typing of their esterases. Six main kinds of esterases were distinguished by their spectrt of hydrolytic activity toward seven synthetic substrates. The electrophoretic variations of these enzymes were used to define ten zymotypes among the three biotypes. Esterase electrophoresis appeared to be more sensitive than biotyping, and could represent an additional marker for epidemiological analysis.
cently twelve species in the genus Acinetobacter were delineated by D N A - D N A hybridization [2]. A s Acinetohacter baumannii was the Acinetobacter species most frequently encountered in hospitalized patients, a biotyping system differentiating 19 biotypes was devised for this species [3,4]. Electrophoretic typing of esterases can be used to differentiate strai~:~ on the basis of both specific actwity prtffiles and allelic variations of these enzymes. Thi:~ procedure has been shown to be a molecular tool for epidemiological analysis of nosocomial pathogens [5-8]. The purpose of this study was to compare the csterase electrophoresis with biotyping, which ~g the most commonly used typing method in clinical laboratories, of 49 A. baumannii strains.
2. I N T R O D U C T I O N Nosocomial outbreaks of Acinetobacter strains have increased over the last ten years [1]. Re-
Correspondence to: P. Pou/~dras, Laboratoire de Bactdrio-
Iogie-Virologie, H6pital Universitaire Pontchaillou, 35033 Rennes, France.
3. M A T E R I A L S A N D M E T H O D S 3.1. Bacterial strains Forty-nine Acinetobacter
strains, isolated in 1991, were included in this study. The bacteria, from environmental and clinical sources, were provided from four French hospitals (Table 1)
126
Table I Origin and biotype of the Acinetobacter haumannii strains Strain no.
Source
Origin ~
Biotype
1-I I 12-13
Rennes" Rennes" Rennes" Strasbourg i, St-Antoine. Paris r Rennes ~' Rennes" Rennes ~' Rennes ~' Boucicaut.Paris ,I St-Antoine. Paris ~
C HC ES CJ C Ct ES HC C CJ Cf
9 9 9 9 9 18 18 18 6 6 6
14 15-19 20-23 24-28 29-33 34-41 42-45 46-48 49
T h e forty-nine strains (Table 1 ) w e r e assigned to biotype 9 (23 strains), biotype 18 (18 strains) and biotype 6 (8 strains).
3.3. Esterase electrophoresis 3.3.1. Preparation of bacterial extracts. Bacteria
3.2. Biotyping of A. baumannii strahts
w e r e g r o w n at 37°C in L - b r o t h for 18 h with agitation and collected by centrifugation. T h e pellets were w a s h e d with 60 m M Tris-glycine buffer, p H 8.7, r e s u s p e n d e d in the s a m e b u f f e r and disr u p t e d by sonication for 18 min at 4°C. D e b r i s was r e m o v e d at 200{)0 × g for 20 min at 4°C a n d the crude extract s u p e r n a t a n t s w e r e stored at - 2 0 ° C until used [9]. 3.3.2. Electrophoresis. H o r i z o n t a l slab elect r o p h o r e s i s in a c o m p o s i t e polyacrylamide-agarose gel (7% and 1.4%, respectively) was perf o r m e d as described by Uriel [10], using a discont i n u o u s Tris-glycine buffer, p H 8.7. T h e extracts w e r e subjected to a c o n s t a n t voltage g r a d i e n t (7 V / c m ) until the b r o m o p h e n o l blue m a r k e r had run 13 cm. T h e relative e l e c t r o p h o r e t i c mobility ( M E value) was the p e r c e n t a g e of the r u n n i n g distance o f the e s t e r a s e b a n d to that o f the dye front. M F values w e r e c o m p a r e d by r u n n i n g diff e r e n t bacterial extracts side by side in the s a m e gel.
Utilization of six c a r b o n s o u r c e s (levulinate, citraconate, L-phenylalanine, phenYlacetate , 4-hydroxybenzoate, L-tartrate) allowed recognition o f 19 biotypes a m o n g A. baumannii strains [3,4].
T h e e s t e r a s e s w e r e stained o n the gel [ l l ] using the following synthetic substrates: a - n a p h t h y l ace t a t e , / 3 - n a p h t h y l acetate, a - n a p h t h y l p r o p i o n a t e ,
" Ht~pital Pontch:tillou, Rennes° France. h Institut de Bact6riologie. Strasbourg. France. ~" Ht')pital Saint Antoine. Paris. France. a H6pital Boucicaut. Paris. France. ~" C: Clinical isolates; HC; Human Carriage; ES: Environmental sources. t Strains isolated from patients with significant infection.
and w e r e all i n d e p e n d e n t . T h e strains o f biotypc 18 w e r e recovered f r o m a surgical intensive care unit. All strains w e r e identified as A. baumannii by the simplified identification s c h e m e described previously [3].
3.3.3. Detection and identification of esterase.
Table 2 Specific activity profile of the esterases of A. baumannii Substrates hydrolysed " Esterase Ab I Esterase Ab 2 Esterase Ab.~ ~' Esterase Ab4 Esterase Ab~ Esterase Abe, Additional band
aNA +++ ++ + ++ + + -
aNP +++ + + ++ + + -
aNB ++ ++ + + +
IA ++ ++ + .¢ + _ . +
.
flNA ++ + _ + + . _
~NP ++ + _ + _
flNB ++ _ _ + _
_
_
.
" aNA, a-naphthyl acetate; aNP, a-naphthyl propionate; aNB, a-naphthyl butyrate; IA, indo~l acetate; /3NA, /3-naphthyl acetate;/3NP, fl-naphthyl propionate;/3NB,/3-naphthyl butyrate. Relative intensity of staining + + + > + + > +. Undetected -. t, Faint esterase band.
127
/3-naphthyl propionate, a-naphthyl butyrate, /3naphthyl butyrate and indoxTI acetate (Sigma). The various esterases were identified by differences in their activity staining.
um m~ m I:~
~:J
m
t::
i:~
4. RESULTS
4.1. Characterization of esterases Six distinct esterase bands were identified by their activity towards the seven tested substrates. The esterase bands were numbered Ab I to Ab 6 in order of increasing mobility towards the anode (Table 2). All esterases hydrolysed a-naphthyl acetate and a-naphthyl propionate. Bands Ab n and Ab 4 both hydrolysed the four acetyl esters, but band Ab 4 migrated farther toward the anode. Esterases Ab 2 and Ab 3 had the same mobility (except strain 22), but differed both by their profiles and intensity of staining. One additional esterase band, hydrolysing a-naphthyi butyrate and ~ndoxyi acetate, was only detected in strain 23.
4.Z Electrophoretic patterns of esterases
Fig. I. Diagrammatic representation of the esterase patterns of the 49 strains of A. baumannii grouped accordi,~ to zymotype, m. esterase Abfi I1). esterase A b ~ : ' B . esterase Ab3; El, esterase Ab4; m, esterase Ab5; em ~:.:;rase Ab6; m, additional band.
Figs. 1 and 2 show the electrophoretic distribution of the six esterases in the three biotypes. Esterase Ab n, hydrolysing all seven substrates, was monomorphic and detected in all strains. In contrast, esterase Ab 4 showed five mobility variants from M r approx. 65 to M F approx. 72, and was not detected in strain 45. Esterase Ab 6 was also polymorphic with five electrophoretic variants (M r approx. 76 to M F approx. 82), and was undetected within biotype 18. The patterns of esterase banding in the strains differed both in the number of bands and in their migration distances. The combination of electrophoretic variation of the six esterases produced a total of 10 different zymotypes in the 49 A. baumannii strains (Fig. 1).
4.3. Comparison o f zymotype with biotype Among A. baumannii biotype 9, 20 strains were zymotype 1; zymotypes 2, 3 and 4 were detected in strain 21, 22 and 23 respectively. Remarkably, the 4 strains of biotype 9 of SaintAntoine Hospital were of 4 different zymotypes.
'21
Ab6
4.-
Ab2 Abl
I
d
3
2
I
5
6
7
zymo~y~s
Fig. 2. Zymotypes revealed among A. baumannii strains of biotypes 9 and 18. Esterase bands were visualized by a-naphthyl acetate.
Most A. baumannii biotype 18 belonged to zymotype 5, but esterase A b 5 was not detected in strain 40 (zymotype 6) and esterases A b 2 and A b 5 in strain 41 (zymotype 7). Three zymotypes were found among the 8 A. baumannii biotype 6 strains. 5. DISCUSSION The poor state of Acb,etobacter taxonomy has prevented the precise identification of species. A new classification of the genus Acinetobacter based on D N A - D N A hybridization and nutritional characteristics [2] allowed the identification of 12 genomic species, which were readily distinguished by analysis of electrophoretic polymorphism of deshydrogenases and esterases [12]. A. baumannii is the species that is often responsible for nosocomial infections [3]. A phage typing was found to be useful for the study of hospital outbreaks [13]. Comparison of cell envelope protein profiles appeared to be a suitable method for the classification of Acinetobacter strains of nosocomial origin [14]. Recently, Alexander et al. [15] described 4 distinct plasmid types among 29 strains of Acinetobacter. In the present study, A. baumannii strains from different sources were studied for esterase eleetrophoresis and biotyping, two typing methods easier to perform routinely than genomic analyses. A m o n g strains of biotype 9 and 18, zymotypes i and 5 respectively were widely predominant. T h e two methods showed an epidemic occurrence of A. baumannii biotype 9 and 18 in Pontchaillou Hospital. Biotype 6 a p p e a r e d more heterogeneous with 3 zymotypes among only 8 strains, without prevalence of a zymotype. Therefore esterase eleetrophoresis did appear to correlate with biotyping within biotypes 9 and 18. However, esterase eleetrophoresis seemed to be more sensitive than biotyping to further characterize strains, since 10 zymotypes were identified among the 3 biotypes tested. Recently [7], esterase electrophoresis was performed for A. baumannii typing (biotype not defined), and all strains studied had a distinct zymotype. Interestingly, 5 of the 6 different esterases described in the present study were previously
detected [7]. This high variability in the esterase profile makes A. baumannii typing quite different from the esterase electrophoretie typing used for E. coli [16], motile Aeromonas [5] and B. catharralis [17] where most of the isolates from a single species have the same csterases. Esterase electrophoresis could represent an additional and potent marker for A. baumannii epidemiological analysis, whereas biotyping seemed an appropriate method for the screening of strains.
ACKNOWLEDGEMENTS T h e authors thank D. Lesage, J.L. Fauch~re and H. Monteil for A. baumannii isolates, and Nicolas Vu for technical assistance.
REFERENCES [I] Bergogne-Berezin, E., Joly-Guillou, M.L. and Vieu, J.F. (1987) J. Hosp. Infect. 10, 105-113. [2] Bouvet, P.J.M. and Grimont, P.A.D. (1986) Int. J. Syst. Bacteriol. 36, 228-240. [3] Bouvet, P.J.M. and Grimont, P.A.D. (1987) Ann. Inst. Pasteur/Microbiol. 138, 569-578. [4] Bouvet, P.J.M., Jeanjean, S., Vieu, J.F. and Dijkshoorn, L. (1990) J. Clin. Microbiol. 28, 170-176. [5] Picard, B. and Goullet, Ph. (1987) Epidemiol. Infect. 98, 5-14. [6] Picard, B., Bruneau, B. and Goullet, Ph. (1988) J. Hosp. Infect. 11, 194-195. [7] Picard, B. and Goullet, Ph. (1990) FEMS Microbiol. Lett. 72, 229-234. [8] Goullet, Ph. and Picard, B. (1991) FEMS Microbiol. Lett. 78, 195-200. [9] Goullet. Ph. and Picard, B. (1985) Electrophoresis 6, 132-135. [10] Uriel, J. (1966) Bull. Soc. Chim. Biol. 48, 969-982. [11] Uriel, J. (1961) Ann. Inst. Pasteur 101, 104-119. [12] Picard, B., Goullet, Ph., Bouvet, P.J.M., Decoux, G. and Denis, J.B. (1989) Electrophoresis 10, 680-685. [13] Vieu, LF., Bergogne-Berezin, E., Joly, M.L., Berthelot, G. and Fichelle, A. (1980) Nouv. Presse M6d. 9, 35513552. [14] Dijkshoorn, L., Van Vianen, W., Degener, J.E. and Michel, M.F. (1987) Epidemiol. Infect. 99, 659-667. [15] Alexander, M., Rahman, M., Taylor, M. and Noble, W.C. (1988) J. Hosp. Infect. 12, 273-287. [16] Goullet, Ph. and Picard, B. (1984) Presse Med. 13, 1079-1081. [17l Picard, B., Goullet, Ph., Denamur, E. and Suermondt, G. (1989) Epidemiol. Infect. 103, 547-554.
125
FEMSLE 05018
Esterase electrophoresis compared with biotyping for epidemiological typing of Acinetobacter baumannii strains P. Pou/~dras a, S. G r a s a, J.M. Sire ~, R. M e s n a r d ~, P.Y. D o n n i o ", B. Picard b and J.L. Avril " a Laboratoire de Bactt~riologie-Virologie. Ht~pital Uni~'ersitaire Pontchaillou. Remws. France. and ~'Laboratoire de Microbiologie, Facult~ de Mt:dechwXacier Bichat, Unicersit~ Paris 7. Paris. France
Received 14 April 1992 Revision received I June 1992 Accepted 5 June 1992 Key words: Acinetobacter; Esterase; Biotype; Epidemiology; Zymotype
1. S U M M A R Y Forty-nine Acinetobacter baumannii strains belenging to three biotypes and isolated from four hospitals were differentiated by electrophoretic typing of their esterases. Six main kinds of esterases were distinguished by their spectrt of hydrolytic activity toward seven synthetic substrates. The electrophoretic variations of these enzymes were used to define ten zymotypes among the three biotypes. Esterase electrophoresis appeared to be more sensitive than biotyping, and could represent an additional marker for epidemiological analysis.
cently twelve species in the genus Acinetobacter were delineated by D N A - D N A hybridization [2]. A s Acinetohacter baumannii was the Acinetobacter species most frequently encountered in hospitalized patients, a biotyping system differentiating 19 biotypes was devised for this species [3,4]. Electrophoretic typing of esterases can be used to differentiate strai~:~ on the basis of both specific actwity prtffiles and allelic variations of these enzymes. Thi:~ procedure has been shown to be a molecular tool for epidemiological analysis of nosocomial pathogens [5-8]. The purpose of this study was to compare the csterase electrophoresis with biotyping, which ~g the most commonly used typing method in clinical laboratories, of 49 A. baumannii strains.
2. I N T R O D U C T I O N Nosocomial outbreaks of Acinetobacter strains have increased over the last ten years [1]. Re-
Correspondence to: P. Pou/~dras, Laboratoire de Bactdrio-
Iogie-Virologie, H6pital Universitaire Pontchaillou, 35033 Rennes, France.
3. M A T E R I A L S A N D M E T H O D S 3.1. Bacterial strains Forty-nine Acinetobacter
strains, isolated in 1991, were included in this study. The bacteria, from environmental and clinical sources, were provided from four French hospitals (Table 1)
126
Table I Origin and biotype of the Acinetobacter haumannii strains Strain no.
Source
Origin ~
Biotype
1-I I 12-13
Rennes" Rennes" Rennes" Strasbourg i, St-Antoine. Paris r Rennes ~' Rennes" Rennes ~' Rennes ~' Boucicaut.Paris ,I St-Antoine. Paris ~
C HC ES CJ C Ct ES HC C CJ Cf
9 9 9 9 9 18 18 18 6 6 6
14 15-19 20-23 24-28 29-33 34-41 42-45 46-48 49
T h e forty-nine strains (Table 1 ) w e r e assigned to biotype 9 (23 strains), biotype 18 (18 strains) and biotype 6 (8 strains).
3.3. Esterase electrophoresis 3.3.1. Preparation of bacterial extracts. Bacteria
3.2. Biotyping of A. baumannii strahts
w e r e g r o w n at 37°C in L - b r o t h for 18 h with agitation and collected by centrifugation. T h e pellets were w a s h e d with 60 m M Tris-glycine buffer, p H 8.7, r e s u s p e n d e d in the s a m e b u f f e r and disr u p t e d by sonication for 18 min at 4°C. D e b r i s was r e m o v e d at 200{)0 × g for 20 min at 4°C a n d the crude extract s u p e r n a t a n t s w e r e stored at - 2 0 ° C until used [9]. 3.3.2. Electrophoresis. H o r i z o n t a l slab elect r o p h o r e s i s in a c o m p o s i t e polyacrylamide-agarose gel (7% and 1.4%, respectively) was perf o r m e d as described by Uriel [10], using a discont i n u o u s Tris-glycine buffer, p H 8.7. T h e extracts w e r e subjected to a c o n s t a n t voltage g r a d i e n t (7 V / c m ) until the b r o m o p h e n o l blue m a r k e r had run 13 cm. T h e relative e l e c t r o p h o r e t i c mobility ( M E value) was the p e r c e n t a g e of the r u n n i n g distance o f the e s t e r a s e b a n d to that o f the dye front. M F values w e r e c o m p a r e d by r u n n i n g diff e r e n t bacterial extracts side by side in the s a m e gel.
Utilization of six c a r b o n s o u r c e s (levulinate, citraconate, L-phenylalanine, phenYlacetate , 4-hydroxybenzoate, L-tartrate) allowed recognition o f 19 biotypes a m o n g A. baumannii strains [3,4].
T h e e s t e r a s e s w e r e stained o n the gel [ l l ] using the following synthetic substrates: a - n a p h t h y l ace t a t e , / 3 - n a p h t h y l acetate, a - n a p h t h y l p r o p i o n a t e ,
" Ht~pital Pontch:tillou, Rennes° France. h Institut de Bact6riologie. Strasbourg. France. ~" Ht')pital Saint Antoine. Paris. France. a H6pital Boucicaut. Paris. France. ~" C: Clinical isolates; HC; Human Carriage; ES: Environmental sources. t Strains isolated from patients with significant infection.
and w e r e all i n d e p e n d e n t . T h e strains o f biotypc 18 w e r e recovered f r o m a surgical intensive care unit. All strains w e r e identified as A. baumannii by the simplified identification s c h e m e described previously [3].
3.3.3. Detection and identification of esterase.
Table 2 Specific activity profile of the esterases of A. baumannii Substrates hydrolysed " Esterase Ab I Esterase Ab 2 Esterase Ab.~ ~' Esterase Ab4 Esterase Ab~ Esterase Abe, Additional band
aNA +++ ++ + ++ + + -
aNP +++ + + ++ + + -
aNB ++ ++ + + +
IA ++ ++ + .¢ + _ . +
.
flNA ++ + _ + + . _
~NP ++ + _ + _
flNB ++ _ _ + _
_
_
.
" aNA, a-naphthyl acetate; aNP, a-naphthyl propionate; aNB, a-naphthyl butyrate; IA, indo~l acetate; /3NA, /3-naphthyl acetate;/3NP, fl-naphthyl propionate;/3NB,/3-naphthyl butyrate. Relative intensity of staining + + + > + + > +. Undetected -. t, Faint esterase band.
127
/3-naphthyl propionate, a-naphthyl butyrate, /3naphthyl butyrate and indoxTI acetate (Sigma). The various esterases were identified by differences in their activity staining.
um m~ m I:~
~:J
m
t::
i:~
4. RESULTS
4.1. Characterization of esterases Six distinct esterase bands were identified by their activity towards the seven tested substrates. The esterase bands were numbered Ab I to Ab 6 in order of increasing mobility towards the anode (Table 2). All esterases hydrolysed a-naphthyl acetate and a-naphthyl propionate. Bands Ab n and Ab 4 both hydrolysed the four acetyl esters, but band Ab 4 migrated farther toward the anode. Esterases Ab 2 and Ab 3 had the same mobility (except strain 22), but differed both by their profiles and intensity of staining. One additional esterase band, hydrolysing a-naphthyi butyrate and ~ndoxyi acetate, was only detected in strain 23.
4.Z Electrophoretic patterns of esterases
Fig. I. Diagrammatic representation of the esterase patterns of the 49 strains of A. baumannii grouped accordi,~ to zymotype, m. esterase Abfi I1). esterase A b ~ : ' B . esterase Ab3; El, esterase Ab4; m, esterase Ab5; em ~:.:;rase Ab6; m, additional band.
Figs. 1 and 2 show the electrophoretic distribution of the six esterases in the three biotypes. Esterase Ab n, hydrolysing all seven substrates, was monomorphic and detected in all strains. In contrast, esterase Ab 4 showed five mobility variants from M r approx. 65 to M F approx. 72, and was not detected in strain 45. Esterase Ab 6 was also polymorphic with five electrophoretic variants (M r approx. 76 to M F approx. 82), and was undetected within biotype 18. The patterns of esterase banding in the strains differed both in the number of bands and in their migration distances. The combination of electrophoretic variation of the six esterases produced a total of 10 different zymotypes in the 49 A. baumannii strains (Fig. 1).
4.3. Comparison o f zymotype with biotype Among A. baumannii biotype 9, 20 strains were zymotype 1; zymotypes 2, 3 and 4 were detected in strain 21, 22 and 23 respectively. Remarkably, the 4 strains of biotype 9 of SaintAntoine Hospital were of 4 different zymotypes.
'21
Ab6
4.-
Ab2 Abl
I
d
3
2
I
5
6
7
zymo~y~s
Fig. 2. Zymotypes revealed among A. baumannii strains of biotypes 9 and 18. Esterase bands were visualized by a-naphthyl acetate.
Most A. baumannii biotype 18 belonged to zymotype 5, but esterase A b 5 was not detected in strain 40 (zymotype 6) and esterases A b 2 and A b 5 in strain 41 (zymotype 7). Three zymotypes were found among the 8 A. baumannii biotype 6 strains. 5. DISCUSSION The poor state of Acb,etobacter taxonomy has prevented the precise identification of species. A new classification of the genus Acinetobacter based on D N A - D N A hybridization and nutritional characteristics [2] allowed the identification of 12 genomic species, which were readily distinguished by analysis of electrophoretic polymorphism of deshydrogenases and esterases [12]. A. baumannii is the species that is often responsible for nosocomial infections [3]. A phage typing was found to be useful for the study of hospital outbreaks [13]. Comparison of cell envelope protein profiles appeared to be a suitable method for the classification of Acinetobacter strains of nosocomial origin [14]. Recently, Alexander et al. [15] described 4 distinct plasmid types among 29 strains of Acinetobacter. In the present study, A. baumannii strains from different sources were studied for esterase eleetrophoresis and biotyping, two typing methods easier to perform routinely than genomic analyses. A m o n g strains of biotype 9 and 18, zymotypes i and 5 respectively were widely predominant. T h e two methods showed an epidemic occurrence of A. baumannii biotype 9 and 18 in Pontchaillou Hospital. Biotype 6 a p p e a r e d more heterogeneous with 3 zymotypes among only 8 strains, without prevalence of a zymotype. Therefore esterase eleetrophoresis did appear to correlate with biotyping within biotypes 9 and 18. However, esterase eleetrophoresis seemed to be more sensitive than biotyping to further characterize strains, since 10 zymotypes were identified among the 3 biotypes tested. Recently [7], esterase electrophoresis was performed for A. baumannii typing (biotype not defined), and all strains studied had a distinct zymotype. Interestingly, 5 of the 6 different esterases described in the present study were previously
detected [7]. This high variability in the esterase profile makes A. baumannii typing quite different from the esterase electrophoretie typing used for E. coli [16], motile Aeromonas [5] and B. catharralis [17] where most of the isolates from a single species have the same csterases. Esterase electrophoresis could represent an additional and potent marker for A. baumannii epidemiological analysis, whereas biotyping seemed an appropriate method for the screening of strains.
ACKNOWLEDGEMENTS T h e authors thank D. Lesage, J.L. Fauch~re and H. Monteil for A. baumannii isolates, and Nicolas Vu for technical assistance.
REFERENCES [I] Bergogne-Berezin, E., Joly-Guillou, M.L. and Vieu, J.F. (1987) J. Hosp. Infect. 10, 105-113. [2] Bouvet, P.J.M. and Grimont, P.A.D. (1986) Int. J. Syst. Bacteriol. 36, 228-240. [3] Bouvet, P.J.M. and Grimont, P.A.D. (1987) Ann. Inst. Pasteur/Microbiol. 138, 569-578. [4] Bouvet, P.J.M., Jeanjean, S., Vieu, J.F. and Dijkshoorn, L. (1990) J. Clin. Microbiol. 28, 170-176. [5] Picard, B. and Goullet, Ph. (1987) Epidemiol. Infect. 98, 5-14. [6] Picard, B., Bruneau, B. and Goullet, Ph. (1988) J. Hosp. Infect. 11, 194-195. [7] Picard, B. and Goullet, Ph. (1990) FEMS Microbiol. Lett. 72, 229-234. [8] Goullet, Ph. and Picard, B. (1991) FEMS Microbiol. Lett. 78, 195-200. [9] Goullet. Ph. and Picard, B. (1985) Electrophoresis 6, 132-135. [10] Uriel, J. (1966) Bull. Soc. Chim. Biol. 48, 969-982. [11] Uriel, J. (1961) Ann. Inst. Pasteur 101, 104-119. [12] Picard, B., Goullet, Ph., Bouvet, P.J.M., Decoux, G. and Denis, J.B. (1989) Electrophoresis 10, 680-685. [13] Vieu, LF., Bergogne-Berezin, E., Joly, M.L., Berthelot, G. and Fichelle, A. (1980) Nouv. Presse M6d. 9, 35513552. [14] Dijkshoorn, L., Van Vianen, W., Degener, J.E. and Michel, M.F. (1987) Epidemiol. Infect. 99, 659-667. [15] Alexander, M., Rahman, M., Taylor, M. and Noble, W.C. (1988) J. Hosp. Infect. 12, 273-287. [16] Goullet, Ph. and Picard, B. (1984) Presse Med. 13, 1079-1081. [17l Picard, B., Goullet, Ph., Denamur, E. and Suermondt, G. (1989) Epidemiol. Infect. 103, 547-554.
