FEMS MicrobiologyLetters '43 (1'492) 1`45-198 © 1992 Federation of European MicrobiologicalSocieties0378-10`47/`42/$05.110 Published by Elsevier
FEMSLE I)4902
Co-transfer of vancomycin and other resistance genes from
Enterococcus faecalis NCTC 12201 to Staphylococcus aureus W.C. Noble, Z a r i n a Virani and R o s e m a r y G . A . C r c c Department of Microbial Diseases, hlstitute t~f Dermatology. St. l'homa~" th~spital. London. UK
Received 27 February 1992 Accepted 13 March It192
Key words: Staphylococcus aureus; Enterococcus faecalis; Vancomycin resistance: Erythl omycin resistance; Conjugative transfer
1. S U M M A R Y
2. I N T R O D U C T I O N
Conjugative transfer, in the apparent absence of plasmid DNA, of high-level vancomycin resistance from Enterococcusfaecalis NCTC 12201 to Staphylococcus aureus B i l l has been demonstrated in vivo and in vitro. Selection of transconjuga~!ts on media containing erythromycin or chloramphenicol may result in the transfer of resistance to erythromycin, chloramphenieol, gentamicin, streptomycin and vancomycin though these are capable of separate transfer. Vancomycin resistance has not been transmitted from staphylococcus to staphylococcus though transfer of erythromycin and of chloramphenicol resistance has been achieved.
In 1989 Uttley et al. [1] reported an outbreak of infection in renal patients due to Enterococcus species resistant to vancomycin with minimal inhibitory concentrations (MIC) between 64 and > 2000 mg/I. They were able to transfer this resistance together with resistance to erythromycin and sometimes chloramphenicol to a recipient E. faecalis by filter-mating, selecting for transfer of erythromycin resistance. No plasmid DNA was seen in recipients acquiring vancomycin rcsista~lce from their E. faec::lis No. 1 (deposited as NCTC 12201), though plasmids were seen in transfers from another strain of E. faecalis and two strains of E. faechlm. Vancomycin resistance plasmids in enterococci have been variously described as conjugative and 30-40 kb, or conjugative, tr,l-~eromone-responsive and 55 kb [2-4]. Transfer of such plasmids is reported to occur to other streptococcal and enterococcal species and to Listeria rnonocytogenes, but Leclercq et al. [3] were not able to transfer
Correspondence to: W.C. Noble, Department of Microbial Diseases, Institute of Dermatology, St. Thomas" Hospital, London SEI 7EH, UK.
196 vancomycin resistance to Staphylococcus aureus or Bacillus subtilis. Natural high-level resistance to vancomycin has not been reported in S. au-
Table 1 Transfer of resistance from E. faecalis NCTC 12201 to S. attreus Bill
reus .
Experiment
Selective medium
Transfer per 10~' recipients In vitro In vivo 0.6
Here we report co-transfer of vancomycin and other antibiotic resistances from E. faecalis to S. auretts.
1
RE
2
RE
I
ND 6 × 103
3
RE
0
3.5
RC RV
0 0
RE
0
RC RV
1) 0
2 × 102 I 2 x 10I 5 o
3. M A T E R I A L S A N D M E T H O D S 4
Enterococcus faecalis NCTC 12201 resistant to
penicillin, tetracycline, erythromycin, chloramphenicol, gentamicin, streptomycin, vancomycin and mupirocin and Staphylococcus aureus BI 11 (a wild-type isolate used by us previously in transfer experiments [5]) selected for chromosomal resistance to rifampicin ( M I C > 80 m g / I ) and fusidic acid (MIC > 50 m g / I ) were mated on filters and on the skin of the hairless obese motzse [5,6]. Briefly, strains were grown in Brain heart infusion broth, mixed and placed on a 0.45-p.m filter on Blood agar base (BAB) and incubated at 37°C overnight, or on the skin of the do,.sum of pairs of mice which was then occluded for 24 h. Cultures were recovered from skin or filter and inoculated onto selective BAB containing, as a.ppropriate, rifampicin 80 m g / I plus either erythromycin 15 m g / l , chloramphenicoi 10 m g / I or vancomycin 200 m g / l . S. aureus B l l l with acquired resistance to erythromycin, chloramphenicol, vancomycin and streptomycin was similarly mated on filters and mouse skin with S. aureus NCTC 8325 chromosomally resistant to tetracycline with selection on agar containing tetracycline 10 m g / I and other antibiotics as above. Antibiotic resistance patterns of individual isolates or strains were determined on Mueller Hinton agar using discs containing penicillin (10 IU), tetracycline (30 p.g), erythromycin (Era) (15/zg), gentamicin (Gm) (10/~g), streptomycin (Sm) (10 /zg), fusidic acid (30 v-g), chloramphenicol (Cm) (30 v.g), mupirocin (5/~.g) and vancomycin (Vm) (30 p.g) or on BAB containing the same level of antibiotic per ml except fl~r vancomycin (200 p.g/l). MICs to vancomycin were determined by incorporating vancomycin in BAB at levels of
× 10 2
In vivo results represent the mean of the results from two mice. ND. not done; RE. rifampicin+ erythromycin: RC. rifampicin +chloramphenicol; RV. rifampicin+vancomycin.
250, 500 and 1000 m g / l and spot-inoculating overnight broth cultures. Plasmid profiles were made by standard methods as described previously [7] using conventional and pulsed-field gel electrophoresis. Polyacrylamide gel eleetrophoresis of whole-cell proteins of S. aureus were made by standard methods as described previously [8].
4. RESULTS Transfer from E. faecalis to S. aureus was attempted three times on mouse skin and four timvs Gn filters. The frequency of transfer differed on each occasion and partial results are shown in Table 1. Transfers recovered from mouse skin on rifampicin plus chloramphenicol agar (RC) were less frequent but of the same order as on rifampiein plus erythromycin ( R E ) but poor or absent on rifampicin plus vancomyein (RV). In one experiment the n u m b e r of colonies grown from aliquots of the same suspension recovered from a mouse were: RE, 171 colonies; RC, 99 colonies; RV, 1 colony. S. aureus colonies subcuitured from R E or R C plates following in-vivo transfers were also resistant to erythromycin, streptomycin and chloramphenicol but only 348 of the 401 tested were resistant to vancomycin and only two were resis-
197 tant to gentamicin. None was resistant to tetracycline or mupirocin. All 157 colonies picked from RE plates following in vitro transfers were resistant to erythromycin (EmR), 17 were resistant to streptomycin (SmR), 153 to vancomycin (Vm R) but only 86 to chloramphenicoi (CmR). None was resistant to tetracycline or mupirocin but 19 were resistant to gentamicin. Resistance to penicillin was not explored. Plasmids were seen in 11 of 50 EmRCmRSm R Vm R S. a u r e u s isolates using conventional electrophoresis but no further plasmids were found when pulsed gel electrophoresis was employed. Transfer from S. a u r e u s BI 1 ! EmRCmRSmRVmR to S. a t t r e u s 8325 was carried out three times on filters and twice on mouse skin and gave very different results on each selective medium. No transconjugants were recovered on tetracycline plur vancomycin, in vitro or in vivo. In vivo transfers recovered on tetracycline plus erythromycin agar were sensitive to the remaining three antibiotics (56 tested), in vitro no transfer was achieved. Transeonjugants recovered on tetracycline plus chloramphenicol agar were sensitive to the remaining three antibiotics (143 tested from in-vivo transfer and 418 from in-vitro transfer) with transfer rates of about one in 106 recipients. The MIC of 64 vancomycin-resistant variants showed seven to have a MIC of 250 m g / I , seven a MIC of 500 m g / l and 14 a MIC of 1000 m g / l . P A G E gels showed that when grown on agar containing vancomycin, S. a u r e u s B i l l with acquired vancomycin resistance showed a far denser band at about 40 kDa than the same isolate grown in the absence of vancomycin or the original sensitive parent strain.
5. DISCUSSION Leclercq e t a ! . [3] reported that they were unable to transfer vancomycin resistance to S. a u r e u s . In this study only one colony of transconjugant was ever recovered on vancomycin-containing agar when more than 100 colonies bearing vancomycia-resistance genes could be recovered on agar containing erythromycin from the same
sample. This may have a simple explanation. Schlaes et al. [9] reported inducible transferrable vancomycin resistance in enterococci associated with the production of a 39-kDa protein located in the cytoplasmic membrane. Our data are consistent with this. It seems probable that on agar containing vancomycin the single-celled transconjugants are unable to develop this inducible protein before being overwhelmed by the vancomycin. A similar problem exists in trying to select penicillinase-producing staphylococcal transconjugants on agar containing penicillin. Primary selection on erythromycin or chloramphenicol-containing agar yields transconjugants most of which are resistant to erythromyein, chloramphenicol, streptomycin and vancomycin, with some showing sensitivity to chloramphenicol or vancomycin, indicating that these are in fact independently transferrable resistances. Further transfer to a second staphylococcus yields colonies resistant to either erythromycin or chloramphenicol only, confirming the independent nature. Further, transconjugants lost these resistances independently during storage unless maintained on agar containing the appropriate antibiotic (data not shown). Uttley et al. [1] found no plasmid D N A in their transfers of resistance from strain NCTC 12201 to a recipient enterococcus. Although occasional plasmids were seen in our staphylococcal transconjugants, these did not appear to be related to the antibiotic resistances transferred. This would appear to imply that the resistance transfers were transposon-mediated, though this has not been established. Multiple-resistance transposon transfers, including independent loss of some resistance, have been reported associated with tetracycline resistance [10]. In this study no tetracycline resistance transfer was seen in any of the 558 colonies tested. The chief immediate interest in this study lies in the observation that high vancomycin resistance can be transferred from an enterococcus to a staphylococcus under conditions similar to those which exist in nature. Such resistance has not yet been described in wild-type S. a u r e u s and is relatively recently reported in enterococcus. No doubt wild-type S. a u r e u s will acquire resistance in time. The nature of the resistance genes remains to be
198 established and may eventually prove more intere s t i n g t h a n t h e fact o f r e s i s t a n c e itself.
References [1] Uttley, A.H.C.. George. R.C.. Naidoo. J., Woodford, N.. Johnson. A.P.. Collins. C.H.. Morrison, D.. Gilfillan. A.J.. Fitch. L.B. and Heptonstall. J (1989) Epidemiol. Infect. 103, 173-181. [2] Leclerq, R.. Derlot. E.. Dural. J. and Courvalin. P. (1988) N. Engl. J. Med. 319, 157-161. [3] Leclercq, R., Derlot, E.. Weber, R.. Duval, J. and Courvalin. P. (1989) Antimicrob. Agents Chemother. 33, 1015.
[4] Handwerger. S. Pucci, R.J. and Kolokathis, A. 0990) Antimicrob. Agents Chemother. 34, 358-36{I. [5] Naidoo, J. (1984)J. Hyg. Camb. 93, 59-66. [6] Naidoo, J. and Noble. W.C. (t978) J. Gen. Microbiol. 107. 391-393. [7] Rahman. R., Conrolly. S.. Noble. W.C.. Cookson, B. and Phillips, l. (1990) J. Med. Microbiol. 33. 97-100. [8] Jackman, P.J.H. (1987) In: Methods in Microbiology (Colwell, R.R. and Grigorova. R.. Eds.), Vol. 19, pp 2119-225. Academic Press. London. [9] Shlaes. D.M., Bouvet. A.. Devine, C.. Shlaes. J.H., AIObeid, S. and Williamson. R. (1989) Antimicrob. Agents Chemother. 33, 198-203. [10] Inamine. J.R. and Burdett. V. (1985) J. Bacteriol. 161, 620-626.
FEMSLE I)4902
Co-transfer of vancomycin and other resistance genes from
Enterococcus faecalis NCTC 12201 to Staphylococcus aureus W.C. Noble, Z a r i n a Virani and R o s e m a r y G . A . C r c c Department of Microbial Diseases, hlstitute t~f Dermatology. St. l'homa~" th~spital. London. UK
Received 27 February 1992 Accepted 13 March It192
Key words: Staphylococcus aureus; Enterococcus faecalis; Vancomycin resistance: Erythl omycin resistance; Conjugative transfer
1. S U M M A R Y
2. I N T R O D U C T I O N
Conjugative transfer, in the apparent absence of plasmid DNA, of high-level vancomycin resistance from Enterococcusfaecalis NCTC 12201 to Staphylococcus aureus B i l l has been demonstrated in vivo and in vitro. Selection of transconjuga~!ts on media containing erythromycin or chloramphenicol may result in the transfer of resistance to erythromycin, chloramphenieol, gentamicin, streptomycin and vancomycin though these are capable of separate transfer. Vancomycin resistance has not been transmitted from staphylococcus to staphylococcus though transfer of erythromycin and of chloramphenicol resistance has been achieved.
In 1989 Uttley et al. [1] reported an outbreak of infection in renal patients due to Enterococcus species resistant to vancomycin with minimal inhibitory concentrations (MIC) between 64 and > 2000 mg/I. They were able to transfer this resistance together with resistance to erythromycin and sometimes chloramphenicol to a recipient E. faecalis by filter-mating, selecting for transfer of erythromycin resistance. No plasmid DNA was seen in recipients acquiring vancomycin rcsista~lce from their E. faec::lis No. 1 (deposited as NCTC 12201), though plasmids were seen in transfers from another strain of E. faecalis and two strains of E. faechlm. Vancomycin resistance plasmids in enterococci have been variously described as conjugative and 30-40 kb, or conjugative, tr,l-~eromone-responsive and 55 kb [2-4]. Transfer of such plasmids is reported to occur to other streptococcal and enterococcal species and to Listeria rnonocytogenes, but Leclercq et al. [3] were not able to transfer
Correspondence to: W.C. Noble, Department of Microbial Diseases, Institute of Dermatology, St. Thomas" Hospital, London SEI 7EH, UK.
196 vancomycin resistance to Staphylococcus aureus or Bacillus subtilis. Natural high-level resistance to vancomycin has not been reported in S. au-
Table 1 Transfer of resistance from E. faecalis NCTC 12201 to S. attreus Bill
reus .
Experiment
Selective medium
Transfer per 10~' recipients In vitro In vivo 0.6
Here we report co-transfer of vancomycin and other antibiotic resistances from E. faecalis to S. auretts.
1
RE
2
RE
I
ND 6 × 103
3
RE
0
3.5
RC RV
0 0
RE
0
RC RV
1) 0
2 × 102 I 2 x 10I 5 o
3. M A T E R I A L S A N D M E T H O D S 4
Enterococcus faecalis NCTC 12201 resistant to
penicillin, tetracycline, erythromycin, chloramphenicol, gentamicin, streptomycin, vancomycin and mupirocin and Staphylococcus aureus BI 11 (a wild-type isolate used by us previously in transfer experiments [5]) selected for chromosomal resistance to rifampicin ( M I C > 80 m g / I ) and fusidic acid (MIC > 50 m g / I ) were mated on filters and on the skin of the hairless obese motzse [5,6]. Briefly, strains were grown in Brain heart infusion broth, mixed and placed on a 0.45-p.m filter on Blood agar base (BAB) and incubated at 37°C overnight, or on the skin of the do,.sum of pairs of mice which was then occluded for 24 h. Cultures were recovered from skin or filter and inoculated onto selective BAB containing, as a.ppropriate, rifampicin 80 m g / I plus either erythromycin 15 m g / l , chloramphenicoi 10 m g / I or vancomycin 200 m g / l . S. aureus B l l l with acquired resistance to erythromycin, chloramphenicol, vancomycin and streptomycin was similarly mated on filters and mouse skin with S. aureus NCTC 8325 chromosomally resistant to tetracycline with selection on agar containing tetracycline 10 m g / I and other antibiotics as above. Antibiotic resistance patterns of individual isolates or strains were determined on Mueller Hinton agar using discs containing penicillin (10 IU), tetracycline (30 p.g), erythromycin (Era) (15/zg), gentamicin (Gm) (10/~g), streptomycin (Sm) (10 /zg), fusidic acid (30 v-g), chloramphenicol (Cm) (30 v.g), mupirocin (5/~.g) and vancomycin (Vm) (30 p.g) or on BAB containing the same level of antibiotic per ml except fl~r vancomycin (200 p.g/l). MICs to vancomycin were determined by incorporating vancomycin in BAB at levels of
× 10 2
In vivo results represent the mean of the results from two mice. ND. not done; RE. rifampicin+ erythromycin: RC. rifampicin +chloramphenicol; RV. rifampicin+vancomycin.
250, 500 and 1000 m g / l and spot-inoculating overnight broth cultures. Plasmid profiles were made by standard methods as described previously [7] using conventional and pulsed-field gel electrophoresis. Polyacrylamide gel eleetrophoresis of whole-cell proteins of S. aureus were made by standard methods as described previously [8].
4. RESULTS Transfer from E. faecalis to S. aureus was attempted three times on mouse skin and four timvs Gn filters. The frequency of transfer differed on each occasion and partial results are shown in Table 1. Transfers recovered from mouse skin on rifampicin plus chloramphenicol agar (RC) were less frequent but of the same order as on rifampiein plus erythromycin ( R E ) but poor or absent on rifampicin plus vancomyein (RV). In one experiment the n u m b e r of colonies grown from aliquots of the same suspension recovered from a mouse were: RE, 171 colonies; RC, 99 colonies; RV, 1 colony. S. aureus colonies subcuitured from R E or R C plates following in-vivo transfers were also resistant to erythromycin, streptomycin and chloramphenicol but only 348 of the 401 tested were resistant to vancomycin and only two were resis-
197 tant to gentamicin. None was resistant to tetracycline or mupirocin. All 157 colonies picked from RE plates following in vitro transfers were resistant to erythromycin (EmR), 17 were resistant to streptomycin (SmR), 153 to vancomycin (Vm R) but only 86 to chloramphenicoi (CmR). None was resistant to tetracycline or mupirocin but 19 were resistant to gentamicin. Resistance to penicillin was not explored. Plasmids were seen in 11 of 50 EmRCmRSm R Vm R S. a u r e u s isolates using conventional electrophoresis but no further plasmids were found when pulsed gel electrophoresis was employed. Transfer from S. a u r e u s BI 1 ! EmRCmRSmRVmR to S. a t t r e u s 8325 was carried out three times on filters and twice on mouse skin and gave very different results on each selective medium. No transconjugants were recovered on tetracycline plur vancomycin, in vitro or in vivo. In vivo transfers recovered on tetracycline plus erythromycin agar were sensitive to the remaining three antibiotics (56 tested), in vitro no transfer was achieved. Transeonjugants recovered on tetracycline plus chloramphenicol agar were sensitive to the remaining three antibiotics (143 tested from in-vivo transfer and 418 from in-vitro transfer) with transfer rates of about one in 106 recipients. The MIC of 64 vancomycin-resistant variants showed seven to have a MIC of 250 m g / I , seven a MIC of 500 m g / l and 14 a MIC of 1000 m g / l . P A G E gels showed that when grown on agar containing vancomycin, S. a u r e u s B i l l with acquired vancomycin resistance showed a far denser band at about 40 kDa than the same isolate grown in the absence of vancomycin or the original sensitive parent strain.
5. DISCUSSION Leclercq e t a ! . [3] reported that they were unable to transfer vancomycin resistance to S. a u r e u s . In this study only one colony of transconjugant was ever recovered on vancomycin-containing agar when more than 100 colonies bearing vancomycia-resistance genes could be recovered on agar containing erythromycin from the same
sample. This may have a simple explanation. Schlaes et al. [9] reported inducible transferrable vancomycin resistance in enterococci associated with the production of a 39-kDa protein located in the cytoplasmic membrane. Our data are consistent with this. It seems probable that on agar containing vancomycin the single-celled transconjugants are unable to develop this inducible protein before being overwhelmed by the vancomycin. A similar problem exists in trying to select penicillinase-producing staphylococcal transconjugants on agar containing penicillin. Primary selection on erythromycin or chloramphenicol-containing agar yields transconjugants most of which are resistant to erythromyein, chloramphenicol, streptomycin and vancomycin, with some showing sensitivity to chloramphenicol or vancomycin, indicating that these are in fact independently transferrable resistances. Further transfer to a second staphylococcus yields colonies resistant to either erythromycin or chloramphenicol only, confirming the independent nature. Further, transconjugants lost these resistances independently during storage unless maintained on agar containing the appropriate antibiotic (data not shown). Uttley et al. [1] found no plasmid D N A in their transfers of resistance from strain NCTC 12201 to a recipient enterococcus. Although occasional plasmids were seen in our staphylococcal transconjugants, these did not appear to be related to the antibiotic resistances transferred. This would appear to imply that the resistance transfers were transposon-mediated, though this has not been established. Multiple-resistance transposon transfers, including independent loss of some resistance, have been reported associated with tetracycline resistance [10]. In this study no tetracycline resistance transfer was seen in any of the 558 colonies tested. The chief immediate interest in this study lies in the observation that high vancomycin resistance can be transferred from an enterococcus to a staphylococcus under conditions similar to those which exist in nature. Such resistance has not yet been described in wild-type S. a u r e u s and is relatively recently reported in enterococcus. No doubt wild-type S. a u r e u s will acquire resistance in time. The nature of the resistance genes remains to be
198 established and may eventually prove more intere s t i n g t h a n t h e fact o f r e s i s t a n c e itself.
References [1] Uttley, A.H.C.. George. R.C.. Naidoo. J., Woodford, N.. Johnson. A.P.. Collins. C.H.. Morrison, D.. Gilfillan. A.J.. Fitch. L.B. and Heptonstall. J (1989) Epidemiol. Infect. 103, 173-181. [2] Leclerq, R.. Derlot. E.. Dural. J. and Courvalin. P. (1988) N. Engl. J. Med. 319, 157-161. [3] Leclercq, R., Derlot, E.. Weber, R.. Duval, J. and Courvalin. P. (1989) Antimicrob. Agents Chemother. 33, 1015.
[4] Handwerger. S. Pucci, R.J. and Kolokathis, A. 0990) Antimicrob. Agents Chemother. 34, 358-36{I. [5] Naidoo, J. (1984)J. Hyg. Camb. 93, 59-66. [6] Naidoo, J. and Noble. W.C. (t978) J. Gen. Microbiol. 107. 391-393. [7] Rahman. R., Conrolly. S.. Noble. W.C.. Cookson, B. and Phillips, l. (1990) J. Med. Microbiol. 33. 97-100. [8] Jackman, P.J.H. (1987) In: Methods in Microbiology (Colwell, R.R. and Grigorova. R.. Eds.), Vol. 19, pp 2119-225. Academic Press. London. [9] Shlaes. D.M., Bouvet. A.. Devine, C.. Shlaes. J.H., AIObeid, S. and Williamson. R. (1989) Antimicrob. Agents Chemother. 33, 198-203. [10] Inamine. J.R. and Burdett. V. (1985) J. Bacteriol. 161, 620-626.
