Molecular epidemiology of resistance to trimethoprim in enterobacteria isolated in a Parisian hospital

Ann. Inst. Pasteur/Microbiol.

(D ELSEVIER

Paris

1986

1986, 137 A, 239-251

M O L E C U L A R EPIDEMIOLOGY OF RESISTANCE TO T R I M E T H O P R I M

IN ENTEROBACTERIA ISOLATED IN A PARISIAN HOSPITAL by B. Papadopoulou (1) (*), G. Gerbaud (2), p. Courvalin (2), J.F. Acar (1), and F.W. Goldstein (1) (1) Service de Microbiologie Mddicale, HOpital St-Joseph, 7, rue Pierre-Larousse, 75014 Paris, and (2) Unit~ des Agents Antibact~riens, CNRS UA 271, Institut Pasteur, 75724 Paris Cedex 15

SUMMARY Between January, 1981 and December, 1984, 419 strains of enterobacteria isolated from patients at the H6pital Saint-Joseph were studied for (1) the level of resistance to trimethoprim (Tp) by determination of minimal inhibitory concentration (MIC), (2) transferability of this resistance by conjugation into Escherichia coli, (3) plasmid content of wild-type strains and transconjugants by agarose gel electrophoresis of crude bacterial lysates and by incompatibility grouping, and (4) type of dihydrofolate reductase (DHFR) by colony hybridization with probes specific for D H F R types I and II. Tp resistance was defined as MIC /> 4 ~g/ml and high-level resistance by MIC >i 1000 ~g/ml. Amongst the strains studied, 90 % were resistant to high levels of Tp, while 10 % had low-level resistance. This latter type of resistance was not transferable. Transferable high-level resistance to Tp was detected in 180 strains corresponding to 185 plasmids. In the vast majority of the plasmids, resistance to Tp was associated with resistance to sulphonamide (94 ~ streptomycin (75-90 ~ ampicillin (75-90 %) and chloramphenicol (65-80 %). Plasmids conferring resistance to Tp were often large, most (84 %) ranging in size from 90 to 175 Kb. They belonged to six different incompatibility groups and Inc6-C was the most prevalent (34 to 75 %). The study of the distribution of the dfr genes by colony hybridization in 183 transconjugants and 89 strains with non-transferable Tp resistance revealed the presence

Manuscrit regu le 9 d6cembre 1985, accept6 le 15 f6vrier 1986. (*) To whomreprint requestsshouldbe sent at present address: Unit6 des Agentsantibact&iens, CNRS UA 271, InstitutPasteur, 75724Paris Cedex 15.

240

B. P A P A D O P O U L O U AND COLL.

of dfrI genes in most of these strains (48 and 53 070, respectively). DHFR of types I and II were found in only 3 070 of the transconjugants, but in 15 07o of the strains with non-transferable resistance. DHFR of other types were found equally (15 070)in strains with transferable and non-transferable resistance. The high incidence of the type I enzyme among the Tp-resistant strains probably results from the integration of transposon Tn7 into the chromosome or into a non-transferable plasmid. KEY-WORDS: Trimethoprim, Antibiotic resistance, Enterobacteriaceae; Plasmid, Transposon, DHFR, Molecular epidemiology.

INTRODUCTION The folic acid analog trimethoprim prevents the bacterial production of tetrahydrofolate derivatives by competitive inhibition of dihydrofolate reductase (DHFR) [1, 5, 15, 34]. This enzyme is required for the biosynthesis of folic acid. Trimethoprim, in combination with sulphamethoxazole, has been widely used in medical practice since 1969 in Great-Britain and since 1971 in France. Three years after the introduction of this combination of drugs into clinical use, resistant strains were detected in hospitals [2, 11]. This resistance was of a high level (minimal inhibitory concentrations (MIC) greater than or equal to 1000 ~tg/ml), and mediated by self-transferable resistance (R) plasmids [1,

2, 171. Trimethoprim resistance can be due to different biochemical mechanisms [1, 5, 15, 34]. High-level resistance (MIC /> 1000 ~.g/ml) is generally secondary to the acquisition of a trimethoprim-resistant DHFR which exists under two (I and II) or more types [14, 26, 33, 36]. Resistance to low levels of trimethoprim (4 ~< MIC ~ 512 ~.g/ml) results from mutations which either reduce the susceptibility of the chromosomal D H F R to trimethoprim [34] or impair the penetration of the drug in the cell [18]. Surveys in recent years of the levels of trimethoprim resistance in bacteria isolated from patients in hospitals in foreign countries have indicated an

aad = Ap aph bla bp Cel cfu Cm DHFR Gm Kb

= = = = = = = = = =

aminoglycoside-aminocyclitol a d e n y lyltransferase. ampicillin. aminoglycoside phosphotransferase. beta-lactamase. base p a i r . p r o d u c t i o n o f colicin El. c o l o n y - f o r m i n g unit. chloramphenicol. dihydrofolate reductase. gentamicin. kilobase (pair).

km Iel Inc MIC Mob Sm Su Tc Tm Tp Tra UV

= = = = = = = = = = = =

kanamycin. i m m u n i t y to colicin El. incompatibility group. minimal inhibitory concentration. mobilizable. streptomycin. sulphonamide. tetracycline. tobramycin trimethoprim. transferable. ultraviolet.

TRIMETHOPRIM

RESISTANCE

IN ENTEROBACTERIA

241

increase in the incidence o f this t y p e o f resistance [1, 17, 23, 30]. W e h a v e studied t h e genetic basis a n d the b i o c h e m i c a l m e c h a n i s m o f resistance to trim e t h o p r i m in 419 strains o f e n t e r o b a c t e r i a i s o l a t e d o v e r a f o u r - y e a r p e r i o d at the H 6 p i t a l S a i n t - J o s e p h in Paris.

M A T E R I A L S AND M E T H O D S Bacterial strains and plasmids. - - The plasmids used in this study are listed in table I. Escherichia coli BM13 (pro, met, rpoB), BM694 (phototroph, g y r A ) [20] and C600 (thr, leu, thi, lacY, tonA, strA) were from our laboratory collection. Api-system galleries (La Balme-les Grottes, France) were used for the identification of the strains. Media. - - Brain-heart infusion broth and agar (Difco), and Mueller-Hinton and Drigalski agar (Diagnostic Pasteur) were used. All incubations were at 37~ Antibiotic susceptibility. - - Disk agar diffusion susceptibility tests on MuellerHinton agar low in thymidine were performed as described [15]. The method of Steers et al., with an inoculum o f 105-106 c f u / m l or 103-104 cfu/spot, was used to determine MIC of trimethoprim. Conjugation. - - Conjugation and determination o f incompatibility were performed as described [7]. Selection was on Drigalski agar containing nalidixic acid (25 ~tg/ml), sodium azide (400 ~tg/ml), streptomycin (500 ~tg/ml) or trimethoprim (20 ~g/ml). Preparation o f p l a s m i d D N A . - - High molecular weight [20] and low molecular weight [10] plasmid DNA was prepared as described. Purification o f restriction endonuclease-generated D N A f r a g m e n t s . - - The restriction DNA fragments were separated by electrophoresis in horizontal slab gels (20 • 20 • 0.7cm) containing 0.8 % low-temperature gelling agarose type VII (Sigma). DNA fragments were purified as described by Maniatis et al. [22].

TABLE I. - -

P l a s m i d s a n d their o r i g i n s .

Plasmid

Phenotypic characters

Genotype

Source or reference

ColEI::Tn7 pBR322 pFE872 pFE364 RP4

Tra-,Mob§ Iel Sm Sp Tp Tra- ,Mob- ,Ap Tc Tra- ,Mob - ,Ap Tp Tra- ,Mob- ,Ap Tp Tra,IncP,Ap Km Tc

N. Datta Bolivar and Rodriguez, [4] Fling and Richards, [13] Burchall et al. [5]

piP135

Tra,Inc7-M,GmSm Sp Su Tc Hg

cea aad(3") (9)dfrI bla(Tem-1)tet(class C) bla(T em-1)dfrI bla(T em-1)dfrlI bla(Tem-2)aph(3')-I tet(class A) aac(3)-I aad(3") (9) tet(class C)

Rs-a piP55 Rl-16 plPll2 R64 plPll3

Tra,IncW,Cm Km Sm Su Tra,Inc6-C,Ap Cm Gm Km Su Hg Tra,IncFII,Km Tra,IncI1,Km Tra,InclI,Sm Tc Tra,IncN,Tc

bla(OXA-3)aad(2") aph(3')-I tet(class B) tet(class A)

Datta et al. [9] Witchitz et Gerbaud, [39] Hedges and Datta, [19] Witchitz et Chabbert, [38] Meynell and Cooke, [24] Chabbert et al. [7] Datta et Hedges, [8] Chabbert et al. [7]

Nomenclature of phenotypic characters of plasmids is according to Novick et al. [25]. Genetic symbols are according to Novick et aL [25].

242

B. P A P A D O P O U L O U A N D COLL.

E n z y m e s . - - Restriction endonucleases E c o R I and H p a I from Boehringer were used according to the manufacturer's recommendations. Lysozyme was from Sigma. Ribonuclease A (bovine pancreas) was from Calbiochem. Proteinase K was from Merck. C o l o n y hybridization. - - Nick translation of the purified restriction DNA fragments and hybridization in 50 % formamide at 42~ were as described [22]. UVsterilized nitrocellulose filters (BA-85 Schleicher and Shull) were inoculated with a Steers apparatus. After 3-h incubation on Mueller-Hinton agar, bacteria were lysed as described [22]. Chemicals. - - Deoxadenosine 5'-~-32p-triphosphate, triethylammonium salt, (specific activity 400 Cie/mmole) was obtained from the Radiochemical Centre, Amersham. The antibiotics were provided by the following laboratories : ampicillin and kanamycin, Bristol; streptomycin and tetracycline, Pfizer; gentamicin, Schering; tobramycin, Lilly; ch!oramphenicol, Roussel-Uclaf; nalidixic acid, Winthrop; sulfamethoxazole and trimethoprim, Roche. Sarkosyl (sodium lauryl sarcosinate) was provided by Colgate-Palmolive.

RESULTS

Between January, 1981 and December, 1984, 419 strains o f enterobacteria resistant to trimethoprim were isolated f r o m patients at the H6pital SaintJoseph in Paris. Bacterial resistance to trimethoprim was defined as MIC i> 4 ~tg/ml and high-level resistance as MIC >t 1000 ~tg/ml. Approximately 70 070 of the strains originated f r o m urinary tract infections and 45 07o to 50 070 were E . c o l i isolates, depending u p o n the year considered. I. - -

Evolution and transferability of trimethoprim resistance

(tables II and III). A m o n g s t the strains studied between 1981 and 1984, 90 07o were resistant to high levels of trimethoprim and 10 070 to low levels o f the drug. This latter type o f resistance was never transferable. In 1981, two thirds (60 ~ o f highlevel resistance to trimethoprim were transferable to E . c o l i by conjugation, whereas only half (48 070) were in 1984. Transferable high-level resistance to

TABLE II. - - Evolution of resistance to trimethoprim. Year Nb of strains tested Transferable highlevel resistance Non-transferable high-level resistance Low-level resistance

1981 57

1982 115

1983 182

1984

Total

65

419

34 (60)

49 (43)

66 (36)

31 (48)

180 (43)

15 (26) 8 (14)

55 (48) 11 ( 9 )

97 (53) 19 (11)

31 (48) 3 (4)

198 (47) 41 (10)

Number in parenthesesindicates percent.

T R I M E T H O P R I M R E S I S T A N C E IN E N T E R O B A C T E R I A

243

TABLE III. - - Strains harbouring transferable plasmids conferring high level resistance to trimethoprim.

Year E. coli Klebsiella spp. P. mirabilis Enterobacter Serratia Citrobacter Proteus indole + Salmonella spp.

Total nb of strains Total nb of plasmids

1981

1982

1983

1984

Total (~

10 3 6 5 3 3 3 1

23 8 7 4 2 3 2 0

27 8 5 10 7 5 4 0

11 7 4 1 7 0 1 0

71 26 22 20 19 11 10 1

34 38 (1)

49 49

66 67 (z)

31 31

180 185

(40) (14) (12) (11) (10.5) (6) (5.5)

(1) Four bacteria harboured two different plasmids conferring resistance to Tp. (2) One clone harboured two different plasmids conferring resistance to Tp.

trimethoprim was detected in 180 strains and corresponded to 185 plasmids. Five strains, four E. coli and one Citrobacter, h a r b o u r e d two plasmids conferring resistance to trimethoprim. Forty per cent o f the R plasmids were detected in E. coli. II. - - Antibiotic resistance gene linkage (table IV).

The resistance characters co-transferred to E. coli with the trimethoprim R determinant are summarized in table IV. In the vast majority o f the plasmids, resistance to trimethoprim was associated with that to sulphonamide, streptomycin, ampicillin and chloramphenicol. Resistances to trimethoprim and sulphonamide were co-transferred in 94 ~ o f the cases studied, and this figure remained stable between 1981 and 1984. During the same period, association of trimethoprim resistance with that o f streptomycin and ampicillin increased f r o m 75 to 90 070. The association trimethoprim-chloramphenicol was frequently encountered (65 to 80 ~ There were, in total, forty different combinations o f R determinants on plasmids. III. - - Plasmid incompatibility groups (table V). The plasmids were classified according to their incompatibility (Inc) groups [7]. The results were divided into two groups based on the period o f time. In the first group, 1981-1982, 30 out o f 38 (79 070) plasmids belonged to the Inc group 6-C. The very high incidence o f a single Inc group probably reflects a plasmid epidemic in our hospital eco-system. During this period, only two plasmids were non-groupable.

244

B. P A P A D O P O U L O U A N D C O L L . TABLE I V . - - R e s i s t a n c e p h e n o t y p e s o f t h e p l a s m i d s conferring resistance to trimethoprim.

Resistance phenotype Ap Ap Ap Ap Ap Ap Ap Ap

Sm Km Gm Tm Cm Tc Sm Km Gm Tm Cm - CmSm Tc Sm K m - - - - C m - Sm -- G m - - C m - Sm Cm-Sm K m G m T m - Tc Tc Ap Sm K m - - - - C m T c Ap Sm Cm Tc -Sm ---

Sm

Ap-

Km-Sm K m - -

--

~ Su Su Su Su Su Su Su Su Su Su Su

Tp Tp Tp Tp Tp Tp Tp Tp Tp Tp Tp Tp

15.2 14.1 6.5 4.8 4.3 4.3 4.3 3.8 3.2 2.7 2.7 2.7 2.2 2.2 2.2 1.6 1.6

Tc Su Tp

Sm ---

Su Cm--Su Cm-Su Su

Tp Tp Tp Tp

TABLZ V . - - I n c o m p a t i b i l i t y g r o u p s o f t h e p l a s m i d s conferring high level resistance to trimethoprim.

Years

Inc group

Nb of plasmids (07o)

Range of sizes (Kb)

1981-1982

6-C I1 W

30 (79) 5 1

110-126 95-110 39

1983-1984

6-C FII N I1 7-M W

27 (34) 1 3 4 4 1

110-174 79-95 47-63 79-110 79-110 32-47

As already mentioned, m a n y plasmids confer multi-resistance to antibiotics. Amongst the 147 plasmids isolated during the second period (1983-1984), only 40 (27 %) couId be classified by incompatibility grouping. The plasmids belonged to six different groups and, again, group Inc6-C was the most prevalent (34 o7o). The size o f the plasmids in the transconjugants was determined by agarose gel electrophoresis of crude bacterial lysates. Figure 1 provides part of

T R I M E T H O P R I M RESISTANCE IN E N T E R O B A C T E R I A

ref.

11

12

13

14

ref.

15

245

16

126.3 -.J,

45 - chr. ..... 30

FIG. 1. - - Analysis of plasmid DNA. D N A was fractionated by electrophoresis in a 0.6 % agarose gel (18 by 13 by 0.4 cm) for 18 h at 3 V/cm [34]. Odd and even numbers indicate wild types and corresponding transconjugants, respectively. Plasmids pBR322, 4.4 Kb; ColEI::Tn7, 19.3 Kb, Rs-a, 39 Kb: RP4, 54 Kb; piP135, 79.3 Kb; and/or R64, 108 Kb, were used as molecular size standards. Right, the size of the plasmids is expressed in kilobases and the position of chromosomal D N A (Chr.) is indicated.

246

B. P A P A D O P O U L O U A N D COLL.

this analysis and the results are summarized in table V. The results obtained confirm the correlation already observed between the Inc groups and the size classes. Plasmids belonging to the Inc W group were smaller than 45 Kb. IncI 1 plasmids ranged in size between 79 Kb and 110 Kb, whereas Inc6-C plasmids were often larger than 100 Kb. Plasmids conferring resistance to trimethoprim were often large, the majority (84 ~ having a size range between 90 Kb and 175 Kb.

IV. --

C o l o n y h y b r i d i z a t i o n (tables VI and VII).

We tested the presence of the genes encoding D H F R type I or II by colony hybridization. The 500-bp Hpa! fragment of pFE872 contains the dfrI gene of transposon T n 7 [14]. The 800-bp EcoRI fragment of plasmid pFE364 contains the dfrlI gene of plasmid R67 [12]. The genes encoding the two different types of D H F R did not cross-hybridize ([14] and fig. 2). The two 32p_ labelled fragments were hybridized with total D N A of 183 E. coli transconjugants and of 89 strains possessing non-transferable high-level resistance to trimethoprim. The results are summarized in table VI, and figure 2 provides part of this analysis. Under our conditions, we were able to detect a resistance gene in a mixture of trimethoprim-resistant and -susceptible strains in

FIG. 2. - - Analysis of DNA by colony hybridization. Plasmid or total D N A was transferred to a nitrocellulose sheet and hybridized [28] to pFE872 500-bp HpaI (A) or pFE364 800-bp EcoRI (B) probes labelled with 32p in vitro. The absence of hybridization with pFE872 (A19) and pFE364 (B2) indicated that the probes are not contaminated with vector DNA. A : 43, BM694/PFE364; 54, BM694; 55, BM694/ColEl::Tn7. B: 3, BM694; 4, BM694/pFE364; 8, BM694/ColEl::Tn7.

T R I M E T H O P R I M R E S I S T A N C E IN E N T E R O B A C T E R I A

247

TABLE VI. - - Transferable (plasmids) and non-transferable (strains) D H F R types conferring high level resistance to trimethoprim.

DHFR type

Nb of plasmids (%)

Nb of strains (o70)

I

88 (48)

47 (53)

II I + II other

61 (33) 5 (3) 29 (16)

7 (8) 22 (25) 13 (14)

a ratio o f one to a hundred. Using this technique, we could therefore easily detect one copy o f resistance gene per c h r o m o s o m e . The gene for the type I enzyme was detected in most o f the strains (48-53 ~ Dihydrofolate reductases o f types I and II were found simultaneously in only 3 ~ o f the transconjugants, but in 25 ~ of the strains with high-level non-transferable trimethoprim resistance. D H F R o f types other t h a n I and II were equally (15 ~ transferable or non-transferable. The distribution o f the D H F R types in the different bacterial species is shown in table VII.

TABLE VII. - - Distribution of transferable and non-transferable D H F R types conferring high level resistance to trimethoprim.

Transferable Donor or host E. coli P. mirabilis Klebsiella spp. Serratia Enterobacter spp. Proteus indole ~ Citrobacter

Non-transferable

DHFR I

DHFR II

DHFR I

DHFR II

46 9 5

9 13 16

27 3 5

1 0 0

15

4

4

4

9 2

5 7

0 5

0 I

1

7

3

1

DISCUSSION Bacterial resistance to trimethoprim has spread into m a n y species and has been detected in numerous countries [1, 17, 23, 28, 29, 30, 32, 34]. The incidence o f trimethoprim resistance depends on local epidemiological factors (such as type of patients, antibiotic pressure, presence o f resistance genes, etc.) and varies with locations and time [1, 17, 23, 28, 34].

248

B. PAPADOPOULOU AND COLL.

Resistance to trimethoprim at the H6pital Saint-Joseph rose from 17.9 070 in 1972 to 25.5 ~ in 1984 with a peak at 36.8 ~ in 1982. This increase in incidence is due to the spread of high level resistance to trimethoprim, an evolution already observed in another ecosystem [1, 17, 37]. High level resistance to trimethoprim is mainly plasmid-mediated [2, 11, 17, 23, 30]. In our study, trimethoprim resistance was transferable to E. coli in 40 to 60 ~ of the strains exhibiting high level resistance to the drug (table II and III). The trimethoprim resistance genes were often linked to other resistance determinants, in particular to sulphonamide, streptomycin, ampicillin, and chloramphenicol (table IV). Co-transfer of resistance to the latter antibiotic was unexpected, since chloramphenicol is not widely used in human therapy. As inferred from the resistance patterns (table IV), the incompatibility groups (table V) and the size of the plasmids (table V and fig. 1), resistance to trimethoprim is borne by many different replicons. This observation suggests that dissemination of resistance to trimethoprim is due to gene, rather than plasmid, epidemics. Incompatibility group C is the most prevalent in France [6], and 34 % of the trimethoprim resistance plasmids fell into that group (table V). The study of the distribution of the dfr genes by colony hybridization in 183 transconjugants revealed the presence of dfrI and dfrII genes alone in 48 and 33 % of the strains, respectively (table VI and fig. 2). Three percent of the transconjugants hybridized with both probes (table VI). These strains harbour single plasmids (fig. 1) which therefore encode the two types of enzymes. The remaining 16 % transconjugants which did not hybridize may owe their trimethoprim resistance to production of DHFR of new types [14]. The ratio of type I versus type II DHFR (6.7) in eighty-nine strains with nontransferable trimethoprim resistance analysed similarly (table VI) was significantly higher than that (1.4) in the transconjugants. This finding confirms that transposon TnT, which encodes a type-I DHFR, has spread into bacterial chromosomes or non-transferable plasmids [3, 21, 28, 32, 37]. However, other transposons closely or distantly related to Tn7 may also be responsible for this situation [16, 31, 35]. The fact that, in four out of eight strains studied, trimethoprim resistance is transposable (B. Papadopoulou, unpublished data) is consistent with this notion. The co-existence of two genetic systems (self-transferable plasmids and transposons) may account for the rapid dissemination of resistance to trimethoprim in bacteria isolated from human and veterinary specimens.

RESUMI~ I~PIDI~MIOLOGIE MOLI~CULAIRE DE LA RI~SISTANCEAU TRIMI~THOPRIME CHEZ LES ENTI~ROBACTI~RIESISOLt~ES DANS UN HOPITAL PARISIEN

Entre janvier 1981 et d6cembre 1984, 419 ent6robact6ries ont 6t6 isol6es de produits pathologiques ~ l'hSpital Saint-Joseph. Ont 6t6 6tudi~s, (I) le n i v e a u

T R I M E T H O P R I M RESISTANCE IN E N T E R O B A C T E R I A

249

de r6sistance au trim6thoprime (Tp) par d6termination de la concentration minimale inhibitrice (CMI), (2) la capacit6 de transfert de cette r6sistance par conjugaison ~ Escherichia coli, (3) le contenu plasmidique par 61ectrophor~se en gel d'agarose de lysats bruts des souches sauvages et des transconjugants et par incompatibilit6 et (4) le type de dihydrofolate-r6ductase (DHFR) par hybridation sur colonie avec des sondes sp6cifiques pour les D H F R de type I ou II. La r6sistance au Tp est d6finie par une CMI i> 4 ~.g/ml et la r6sistance de haut niveau par une CMI /> 1000 btg/ml. Le pourcentage de souches de haut niveau de r6sistance au Tp est pass6 de 86 ~ en 1981 h 95,4 ~ en 1984 et, corr61ativement, celui de souches de has niveau a diminu6 de 14 ~t 4 ~ Ce dernier type de r6sistance n ' a jamais 6t6 transf6r6. En 1981, les deux tiers (60 ~ des souches de haut niveau de r6sistance au Tp ont pu transf6rer cette r6sistance et la moiti6 (48 ~ seulement a p u l e faire en 1984. Cent quatre-vingt-cinq plasmides conf6rant la r6sistance au Tp ont 6t6 transf6r6s ~t E. coli. Dans la grande majorit6 des cas, la r6sistance au Tp est associ6e ~t celle aux sulfamides (94 %), h la streptomycine (75-90 %), h l'ampicilline (75-90 070)et au chloramph6nicol (65-80 %). Dix-sept associations de caract6res diff6rents rendent compte des ph6notypes de 3 / 4 des plasmides. Les plasmides dont la taille varie de 35 ~t 170 kilobases appartiennent ~t six groupes d'incompatibilit6 diff6rents, avec une nette pr6dominance d'appartenance au groupe Inc6-C (34 ~t 79 %). Sur la base d'hybridation A D N / A D N , 48 ~ des plasmides poss~dent une D H F R I, 33 ~ une D H F R II, 3 ~ des D H F R I + II et 16 070un enzyme d'un autre type. Chez les 89 souches ayant une r6sistance de haut niveau au Tp non transf6rable, 53 ~ poss~dent une D H F R I, 8 ~ une D H F R II, 25 070 des D H F R I + II et 14 ~ des enzymes d ' u n autre type. L'incidence 61ev6e de l'enzyme de type I refl~te probablement la pr6sence du transposon T n 7 dans le chromosome ou dans des plasmides non transf6rables. MOTS-CLI~S: Antibior6sistance, Trim6thoprime, Enterobacteriaceae; Plasmide, Transposon, DHFR, Epid6miologie mol6culaire. ACKNOWLEDGMENTS

We thank N. Datta for the gift of plasmid ColE1 : Tn7 and L. Elwell for plasmids pFE872 and pFE364.

REFERENCES [1] ACAR, J.F. & GOLSTEIN,F.W., Resistance: genetics and medical implications, in ~ Handbook of experimental pharmacology; inhibition of folate metabolism in chemotherapy~ (G.H. Hintchnings), vol. 107 (pp. 243-258). Springer-Verlag, Berlin, 1983. [2] ACAR, J.F., GOLDSTEIN,F.W., GERBAUD,G.R. & CHABBERT,Y.A., Plasmides de r6sistance au trim6thoprime, transf&abilit6 et groupes d'incompatibilit6. Ann. Microbiol. (Inst. Pasteur), 1977, 128 A, 41-47.

250

B. P A P A D O P O U L O U

AND COLL.

[3] BARTH,P.T., DATTA,N., HEDGES,R.W. & GRINTER,N.J., Transposition of a [4]

[5] [6]

[7] [8] [9] [10]

[11] [12] [13] [14] [15] [16]

[17] [18]

[19]

[20] [21]

[22]

deoxyribonucleic acid sequence encoding trimethoprim and streptomycin resistance from R483 to other replicons. J. Bact., 1976, 125, 800-810. BOLIVAR,F., RODRIGUEZ,R.L., GREENE,P.J., BETLACH,M.C., HEGRESJER,H.L. & BOYER, H.W., Construction and characterization of new cloning vehicles. - - I. Ampicillin-resistant derivatives of plasmid pMBD. Gene, 1977, 2, 75-93. BURCHALL,J.J., ELWELL,L.P. & FLING,M.E., Molecular mechanisms of resistance of trimethoprim. Rev. infect. Dis., 1982, 4, 246-254. CHABBERT,Y.A., ROUSSEL,A., WITCHITZ, J.L., SANSoN-LEPORS, M.J. & COURVALIN, P., Restriction-endonuclease-generated patterns of plasmids belonging to incompatibility groups II, C, M and N: application to plasmid taxonomy and epidemiology, in ~ Plasmids of medical, environmental and commercial importance)) (Timmis and Piihler) (pp. 183-193). Elsevier/North-Holland, Amsterdam, 1979. CHABBERT,Y.A., SCAVIZZI,M.R., WITCHITZ, J.L., GERBAUD,G.R. & BOUANCHAUD,D.H., Incompatibility groups and the classification of fi--resistance factors. J. Bact., 1972, 112, 666-675. DATTA,N. & HEDGES,R.W., Compatibility groups among fi- R factors. Nature (Lond.), 1971, 234, 222. DATTA,N., HEDGES,R.W., SHAW,E.J., SYKES,R.B. & RICHMOND,M.H., Properties of an R factor from Pseudomonas aeruginosa. J. Bact., 1971, 108, 1244- ! 249. DAVIS,R.W., BOTSTEIN,D. & ROTH, J.R., Advanced bacterial genetics (p. 140). Cold Spring Laboratory, New York, 1980. FLEMING,M.P., DATTA,N. & GRUNEBERG,R.N., Trimethoprim resistance determined by R factors. Brit. reed. J., 1972, 1,726-728. FLING,M.E. & ELWELL,L.P., Protein expression in Escherichia coli minicells containing recombinant plasmids specifying trimethoprim resistance dihydrofolate reductase. J. Bact., 1980, 141, 779-785. FLING, M.E. & RICHARDS,C., The nucleotide sequence of the TMP-resistant dihydrofolate reductase gene harbored by Tn7. Nucl. Acids. Res., 1983, 11, 5147-5158. FLING, M.E., WALTON,L. & ELWELL,L.P., Monitoring of plasmid-encoded, trimethoprim-resistant dihydrofolate reductase genes: detection of a new resistant enzyme. Antimicrob. Agents a. Chemother., 1982, 22, 882-888. GOLDSTEIN,F.W., M6canismes de r6sistance aux sulfamides et au trim6thoprime. Bull. Inst. Pasteur, 1977, 75, 109-139. GOLDSTEIN,F.W., GERBAUD,G.R., ACAR, J.F. & COURVALIN,P., Transposable resistance to Tp and 0/129 in Vibrio cholerae. J. Antimicrob. Chemother., 1985 (Submitted for publication). GOLDSTEIN,F.W., PAPADOPOULOU,B. & ACAR, J.F., The changing pattern of trimethoprim-resistance in Paris with a review of worldwide experience. Rev. infect. Dis. (in press). GUTMANN,L., WILLIAMSON,R., MOREAU,N., KITZIS,M.D., COLLATZ,E., ACAR, J.F. & GOLSTEIN,F.W., Cross-resistance to nalidixic acid, trimethoprim and chloramphenicol associated with alterations in outer membrane proteins of Klebsiella, Enterobacter and Serratia. J. infect. Dis., 1985, 151, 501-507. HEDGES, R.W. & DATTA, N., fi- R factors giving chloramphenicol resistance. Nature (Lond.), 1971, 234, 220. LABIGNE-RouSSEL,A., GERBAUD,G. & COURVALIN,P., Translocation of sequences encoding antibiotic resistance from the chromosome to a receptor plasmid in Salmonella ordonez. )Viol. gen. Genetics, 1981, 182, 390-408. LICHTENSTEIN,C. & BRENNER, S., Site-specific properties of Tn7 transposition into the E. coli chromosome. Mol. gen. Genetics, 1981, 183, 380-387. MANIATIS,T., FRITSCH,E.F. & SAMBROOK,J., Molecular cloning: a laboratory manual (p. 331). Cold Spring Harbor Laboratory, New York, 1982.

T R I M E T H O P R I M R E S I S T A N C E IN E N T E R O B A C T E R I A

251

[23] MAYER,K.H., FLING,M.E., HOPKIUS,J.D. & O'BRIEN, T.F., Molecular epidemiology and evolution of a trimethoprim resistance plasmid widely dispersed in one center. J. infect. Dis., 1985, 151, 783-789. [24] MEYNELL,E. c~r COOKE, M., Repressor-minus and operator-constitutive derepressed mutants of F-like R factors ; their effect on chromosomal transfer by Hfr. C. Genet. Res. (Camb.), 1969, 14, 309-313. [25] NOVICK,R.P., CLOWES,R.C., COHEN,S.N., CURTISS,R., DATTA,N. & FALKOW, S., Uniform nomenclature for bacterial plasmids : a proposal. Bact. Rev., 1976, 40, 168-189. [26] PATTISHALL,K.H., ACAR, J.F., BURCHALL,J.J., GOLDSTEIN,F.W. & HARVEY, R.J., Two distinct types of trimethoprim-resistant dihydrofolate reductase specified by R plasmids of different compatibility groups. J. biol. Chem., 1977, 252, 2319-2323. [27] PORTNOY,D.A., MOSELEY,S.L. ~r FALKOW,S., Characterization of plasmids and plasmid-associated determinants of Yersinia enterocolitica pathogenesis. Infect. Immun., 1981, 31, 775-782. [28] PULKKINEN,L., HtJOVINEN,P., VUORIO,E. & TOIVANEN,P., Characterization of trimethoprim resistance by use of probes specific for transposon Tn7. Antimicrob. Agents a. Chemother., 1984, 26, 82-86. [29] ROMERO, E. and PERmJCA, M., Compatibility groups of R-factors for trimethoprim-resistance isolated in Italy. J. Antimicrob. Chemother., 1977, 3 (Suppl. C), 35-38. [30] SAROGLOU,G., PARASKEVOPOULOU,P., PANIARA, O. t~ KONTOMICHALOU,P., Trimethoprim-resistance plasmids from Enterobacteriaceae isolated in Greece, in ~Antibiotic resistance>) (S. Mitshuhuasi, L. Rosival and V. Krcmery). Springer-Verlag, Berlin, 1980. [31] SHAPIRO,J.A. & SPORN, P., Tn402: a new transposable element that inserts in bacteriophage lambda. J. Bact., 1977, 129, 1632-1635. [32] STEEN, R. & SKOLD, O., Plasmid-borne or chromosomally mediated resistance by Tn7 is the most common response to ubiquitous use of trimethoprim. Antimicrob. Agents a. Chemother., 1985, 27, 933-937. [33] T~NNHAMMAR-EKMAN,B. & SKOLD, O., Trimethoprim-resistance plasmids of different origin encode different drug-resistant dihydrofolate-reductases. Plasmid, 1979, 2, 334-346. [34] THEN, R.L. & HERMAN, F., Mechanisms of trimethoprim resistance in enterobacteria isolated in Finland. Chemotherapy, 1981, 27, 192-199. [35] TIETZE, E., PRAGER, R. & TSCHAPE, H., Characterization of the transposon Tn1822 (Tc) and Tn1824 (Tp Sm) and the light they throw on the natural spread of resistance genes. Plasmid, 1982, 8, 253-260. [36] TOWNER,K.J. & PINN, P.A., A transferable plasmid conferring only a moderate level of resistance to trimethoprim. FEMS Microbiol. Letters, 1981, 10, 271-272. [37] TOWNER, K.J., VENNING, B.M. & PINN, P.A., Occurrence of transposable trimethoprim resistance in clinical isolates of Escherichia coli devoid of selftransmissible resistance plasmids. Antimicrob. Agents and Chemother., 1982, 21, 336-338. [38] WITCHITZ,J.L. & CHABBERT,Y.A., R6sistance transf&able h la gentamicine. - I. Expression du caract~re de r6sistance. Ann. Inst. Pasteur, 1971, 121, 733-742. [39] WITCHITZ,J.L. & GERBAUD,G.R., Classification de plasmides conf6rant la r6sistance ~t la gentamicine. Ann. Inst. Pasteur, 1972, 123, 333-339.