Veterinary Microbiology, 30 ( 1 9 9 2 ) 2 1 3 - 2 2 2 Elsevier Science Publishers B.V., A m s t e r d a m
213
Conjugation of antibiotic resistance in Streptococcus suis J.G. Stuart a, E.J. Zimmerera and R.L. Maddux b aDepartment of Biological Sciences, bBreathitt Veterinary Center, Murray State University, Murray, Kentucky 42071, USA (Accepted 5 August 1991 )
ABSTRACT Stuart, J.G., Zimmerer, E.J. and Maddux, R.L., 1992. Conjugation of antibiotic resistance in Streptococcus suis. Vet. MicrobioL, 30:213-222. Forty-eight clinical isolates of Streptococcus suis were examined for antibiotic sensitivity and the presence of plasmid DNA. It was determined that isolates from this study showed a substantial increase in resistance to erythromycin (ery), clindamycin, and tetracycline (tet) compared to a similar study conducted five years earlier. Eleven of the 48 isolates contained plasmid DNA as revealed by DNA isolation and gel electrophoresis. Plasmid DNA from four strains resistant to the above three antibiotics was tested for the ability to transform an antibiotic sensitive recipient. No transformation of antibiotic resistance could be demonstrated. In other experiments, the above four strains, along with four plasmid-negative triply resistant strains were tested for the ability to transfer tet or ery resistance to tet and ery sensitive recipients by conjugation. In each mating, antibiotic resistance was transferred at frequencies averaging 2.4X 10 -6 recombinants/recipient for ery and 3.4 X 10 -6 recombinants/recipient for tet resistance. DNA from each clinical specimen, as well as the recombinants mentioned above was probed with tn916. Autoradiographs revealed that several clinical isolates and recombinants bound the probe. It is concluded that conjugation of antibiotic resistance in these clinical strains is possibly mediated by a transposon similar to tn916.
INTRODUCTION
Several reports have appeared in recent years concerning antibiotic sensitivity in Streptococcus suis (Chengappa et al., 1986; Hoffman and Henderson, 1985; Touil et al., 1988; Sanford, 1989). These studies indicated that clinical S. suis isolates were uniformly sensitive to penicillin class antibiotics, trimethoprim-sulfamethoxazole, and closely followed by gentamicin (i.e. approximately 90% sensitive from studies cited above). However, resistance to several antibiotics which usually affect Gram positive organisms has also been reported. Hoffman and Henderson (1985 ) and Chengappa et al. (1986 ) reported high levels of resistance to erythromycin (approximately 50% of isolates), lincomycin (64%), and tetracycline (82%). Resistance to these anti0 3 7 8 - 1 1 3 5 / 9 2 / $ 0 5 . 0 0 © 1992 Elsevier Science Publishers B.V. All rights reserved.
214
J.G. STUART ET AL.
biotics occurs at similar high frequencies in other streptococcal species such as S. faecalis and S. pneumoniae (Clewell, 1981 ). Also, according to Clewell ( 1981 ), erythromycin and lincomycin resistance are frequently controlled by plasmid, while tetracycline resistance may be controlled by a plasmid or transposon. Transfer of antibiotic resistance in streptococci is most commonly mediated by conjugation, but transduction and transformation occur in a few species (Clewell, 1981 ). Conjugation in streptococci was first described by Jacob and Hobbs ( 1974 ), who noted conjugal transfer of multiple antibiotic resistance by a plasmid. Since 1974 many conjugal R plasmids have been described for S. faecalis and other streptococcal species (Clewell, 1981; Clewell and Gawron-Burke, 1986). The first transposon (tn916) discovered in S. faecalis was described by Franke and Clewell ( 1981 ). This transposon is 16.4 MDa in size, and carries a tetracycline resistance (tet) r determinant. Also, according to Franke and Clewell ( 1981 ), this transposon mediated conjugation of tet r at low frequencies. Since 1981 other reports of conjugal transposons have been made within S. faecalis and other streptococcal species (Clewell and Gawron-Burke, 1986 ). Burdett et al. ( 1982 ) described three types oftet determinants from a variety of streptococcal species. The tet M determinant was found to be genetically similar to tn916, and the most common determinant associated with chromosomal tet among a variety of streptococcal species. It should be noted that the tet M probe used by Burdett et al. (1982) was made from a strain containing a conjugative transposon. More recently, LeBouquenec et al. ( 1988 ), working with a clinical isolate of S. pyogenes, discovered a very large conjugative element, tn3707, encoding resistance to erythromycin (cry) and tetracycline/minocycline(min). This transposon was subsequently mapped by LeBouquenec et al. (1990), and found to carry a tn916-like element designated tn3703 which carried the resistance determinants for ery and tet/min. This report includes the first observation of conjugation in S. suis. Also data are included that indicate conjugation of antibiotic resistance may be mediated by a transposon similar to tn916. MATERIALS AND METHODS
Bacterial strains Clinical isolates of S. suis were identified biochemically and serotyped as described earlier (Chengappa et al., 1986). Each isolate was taken from a separate animal and designated by serotype and the first two numbers assigned to the clinical case. S. suis type 1 was obtained from Dr. C.H. Armstrong, Animal Disease Diagnostic Laboratory, Purdue University, West Lafayette, IN 47907. Strain SSRF is a rifampin and fusidic acid resistant mutant derived from the type 1 strain, and is devoid of plasmid DNA. Strain Challis
CONJUGATION OF ANTIBIOTIC RESISTANCE IN STREPTOCOCCUS SUIS
215
was a gift from Joe Ferretti, Dept. Microbiol. and Immunol., Univ. Okla., Oklahoma City, OK 73190. Challis is a group H Streptococcus and served as a competent recipient in transformation experiments. Challis (pSM19035) contains a plasmid conferring macrolide, lincomycin streptogramin (MLS) resistance. This strain was used to isolate pSM 19035 which served as a positive control for transformation experiments. Strain CG 120 was obtained from Don Clewell, Dept. of Oral Biology and Microbiology/Immunology, Schools of Dentistry and Medicine, Univ. Mich., Ann Arbor, MI 48109. Strain CG 120 is an E. coli strain constructed by Clewell et al. ( 1981 ) and contains pAM 120, a pBR322 derivative with a tn916 insert generated by Eco R 1. Stock cultures of the bacterial strains were made using 1 ml of horse serum containing 20% sterile glycerol. Cell suspensions were held in sterile screw cap vials at - 80 oC.
Media Serum Yeast Todd-Hewitt (SYTH) was used and consisted of 0.6% yeast extract (Difco), 3% Todd-Hewitt broth (Difco), 0.038% K2HPO4, and 5% horse serum (Gibco). Yeast Todd-Hewitt (YTH) agar was prepared by adding 1.5% bacteriological agar (Sigma) to YTH broth. Blood agar (BA) was prepared by adding 5% defibrinated sheep blood to YTH agar. Antibiotics (Sigma) were used in plates at the following concentrations: tet, 10 #g/ml; fusidic acid, 25/zg/ml; rifampin 50/tg/ml; and ery 5 #g/ml. Susceptibility testing Antimicrobial susceptibility testing of the 48 clinical strains was performed by the standard disk diffusion m e t h o d as described by Lennette et al. ( 1985 ) with a minor modification in the preparation of the inoculum (D'Amato et al., 1982). DNA isolation and gel electrophoresis S. suis DNA was isolated by a procedure modified from Clewell et al. (1974). Ten ml of SYTH broth was inoculated with 100/A from a S. suis frozen suspension culture, and incubated overnight at 30 ° C. Cells from the overnight culture were centrifuged at 8000 rpm for 10 rain in a Sorval RC-5, the supernatant decanted, and the pellet suspended in 3 ml of potassium acetate (200 mM, pH 6.5 ). The cells were centrifuged as before, and suspended in 3 ml Tris-EDTA (TE) (pH 8) buffer plus 0.2 ml of 0.5 M EDTA (pH 8). Freshly prepared lysozyme (5 mg; Sigma) was added to the cell suspension and incubation proceeded for 1 h at 37 ° C. A 0.2 ml aliquot of Sigma Protease I ( 10 m g / m l ) was added to the suspension and incubation continued for another hour at 37°C. The cell suspension was clarified by adding 0.2 ml of a 10% solution of sarcosyl. The suspension was extracted with an equal volume of phenol:chloroform ( 1:1 ), then centrifuged at 10 000 rpm for 10 min at 4 ° C. The upper aqueous layer was removed and 1/ 10 volume of 2.5 M so-
216
J.G. STUART ET AL,
dium acetate was added. DNA was precipitated by adding 2 vol of cold ethanol and holding at - 70 ° C for 30 min. DNA was pelleted by centrifuging at 10 000 rpm for 15 min at 4 ° C. The supernatant was decanted, and the pellet dried and resuspended in 200/tl of TE buffer. DNA samples were held at 4 ° C until ready for analysis by gel electrophoresis. Procedures for agarose gel electrophoresis were taken from Maniatis et al. ( 1982 ). Samples (5/zl; approx. 1 #g DNA) were loaded on a 0.6% agarose gel containing ethidium bromide (5 /zg/ml), and run at 100 volts for 2 h.
Plasmid isolation and transformation Plasmids from all bacterial strains were isolated by CsC1 density gradient centrifugation as described in Maniatis et al. (1982). Transformation was performed by the procedure described by LeBlanc and Hassel ( 1976 ).
Conjugation The procedure for conjugation was adapted from Franke and Clewell ( 1981 ). An overnight culture (0.2 ml) from a clinical strain was mixed with 1 ml of an overnight culture of recipient strain SSRF. This mixture was filtered through a 0.45/zm filter (Millipore), and the filter was placed cell side down on the surface of a BA plate. Incubation proceeded for approximately 18 h at 30°C. The cells were then suspended in 3 ml of YTH broth and spread on plates containing ery or tet (to select against recipients), as well as rifampin and fusidic acid (to select against donors). Controls consisting of donor and recipient cells alone were treated similarly to determine the frequency of spontaneous drug resistant mutants in the population. Transfer frequency was measured by dividing the n u m b e r of transconjugants per ml by the viable count of the recipient (SSRF) cells in the mating mixture. Colonies were counted after incubation for 48 hrs. at 37°C. Some experiments included DNAse I (Sigma) at a concentration of 10/~g/ml in the mating mixture and BA plates.
Dot blot procedure The tn916 probe was prepared by isolating pAM 120 from strain CG 120 by methods described in Maniatis et al. (1982). The tn916 insert was removed by digestion with Eco R 1 (BRL), and purified by DEAE membrane binding (Schliecher and Schuell). A radioactive o~ 32p labeled copy of tn916 was synthesized by nick translation, and used to probe 1/zg of DNA from each sample spotted onto nitrocellulose sheets. A positive control consisted of probing 50 ng of pAM 120 on nitrocellulose, while 1/~g of lambda DNA was used as a negative control. DNA concentrations were estimated using ethidium bromide fluorescence by the minigel method as described in Maniatis et al. (1982) Approximately 10 6 counts per minute per ml of the radioactive probe were used in the hybridization mixture. Following hybridization, the
CONJUGATION OF ANTIBIOTICRESISTANCEIN STREPTOCOCCUSSUIS
217
filter was subjected to several 30 minute washes; twice in 1 X SSPE (25°C), twice in 0.1 x S S P E (25°C), once in 0.1 xSSPE (42°C). After drying, the filter was exposed to X-ray film for 72 h prior to development. RESULTS
Antibiotic sensitivity Table 1 contains data generated from antibiotic susceptibility testing of the 48 clinical isolates examined. High frequencies of resistance were noted to several antibiotics. Resistance of the aminoglycosides, neomycin, kanamycin, and streptomycin is not unusual since these drugs are commonly more effective against Gram negative organisms. However, resistance to erythromycin, clindamycin, and tetracycline is significant, since most Gram positive streptococci are susceptible to these drugs. The results from Table 1 were compared to resistance frequencies of 38 clinical S. suis isolates collected five years earlier (Chengappa et al., 1986). Substantial increases in the frequency of resistance was noted for four antibiotics: ery (11%), clindamycin (29%), neomycin (20%) and tet (18%). Since the S. suis isolates from this study were collected from the same geographic area (western Kentucky) as the earlier clinical isolates, a gene transfer mechanism was suspected. Plasmid screen and transformation Each clinical isolate was screened for the presence of plasmid DNA by DNA isolation and agarose gel electrophoresis. Eleven of 48 isolates were shown to TABLE1 Antimicrobial Susceptibility of Streptococcus suis Drug
Ampicillin Erythromycin Clindamycin Gentamicin Kanamycin Neomycin Nitrofurantoin Spectinomycin Streptomycin Sulfachloropyridizine SXT Tetracycline S = susceptible I = intermediate R = resistant
Number o f Isolates ~
(%) Resistance
S
I
R
48 16 9 40 4 2 35 42 7 25 47 6
0 4 0 3 19 5 10 2 11 2 0 12
0 28 39 4 25 41 2 4 30 21 1 30
0 58 81 9 52 85 4 8 63 44 2 63
218
J.G. STUART ET AL.
contain plasmid DNA, with most plasmid positive strains containing two or more plasmids. Efforts were made to associate plasmid carriage with antibiotic resistance. The average number of drug resistance determinants in plasmid positive strains was 5.1 as compared to 4.7 for plasmid negative strains. Six of eleven (54.5%) plasmid positive strains carried the ery, clindamycin, tet resistant phenotype compared to 13 of 37 (35.1%) of the plasmid negative strains. Plasmid D N A from four S. suis clinical isolates which were plasmid positive and resistant to erythromycin, clindamycin, and tetracycline was isolated and purified. These plasmids along with a positive control from Challis (pSM19035 ), which confers erythromycin resistance, were tested for their ability to transform antibiotic resistance into Challis. Control experiments using pSM 19035 produced as many as 500 ery transformants per plate containing 120 ng of donor DNA. However, in similar experiments using at least 1/tg or more donor DNA, none of the S. suis plasmids were able to transform strain Challis to either tetracycline or erythromycin resistance. Conjugation For these experiments eight clinical strains were selected, four contained plasmids and four did not. Each strain was erythromycin, clindamycin, and tetracycline resistant, and served as a donor in filter mating experiments using the plasmid free SSRF as a recipient. Results of these experiments can be TABLE2
Conjugation o f S . suis clinical donors with strain SSRS ~ Donor
Plasmid
SP 2
Recomb. / ml
Frequency 3
7-45 7-45 8-74 8-74
+ + + +
E T E T
2322 892 127 159
1.5×10 5 1.5>( 10 - 5 5.6×10 7 7.1>(10 7
3-23 3-23 4-27
+ + +
E T E
107 392 8
2 . 1 × 10 - 6 9.3×10 6 2 . 3 × 10 - 7
4-27
+
T
13
2 . 2 X 10 - 7
2-59 2-59 2-20 2-20 7-18 7 - 18 2-15 2-15
-
E T E T E T E T
20 30 159 68 150 203 25 24
~Date presented are an average of at least two conjugation experiments. 2Selection Pressure = Erythromycin or Tetracycline, 3Recombinants/recipients.
1 . 6 × 10 - 7 2.5)
213
Conjugation of antibiotic resistance in Streptococcus suis J.G. Stuart a, E.J. Zimmerera and R.L. Maddux b aDepartment of Biological Sciences, bBreathitt Veterinary Center, Murray State University, Murray, Kentucky 42071, USA (Accepted 5 August 1991 )
ABSTRACT Stuart, J.G., Zimmerer, E.J. and Maddux, R.L., 1992. Conjugation of antibiotic resistance in Streptococcus suis. Vet. MicrobioL, 30:213-222. Forty-eight clinical isolates of Streptococcus suis were examined for antibiotic sensitivity and the presence of plasmid DNA. It was determined that isolates from this study showed a substantial increase in resistance to erythromycin (ery), clindamycin, and tetracycline (tet) compared to a similar study conducted five years earlier. Eleven of the 48 isolates contained plasmid DNA as revealed by DNA isolation and gel electrophoresis. Plasmid DNA from four strains resistant to the above three antibiotics was tested for the ability to transform an antibiotic sensitive recipient. No transformation of antibiotic resistance could be demonstrated. In other experiments, the above four strains, along with four plasmid-negative triply resistant strains were tested for the ability to transfer tet or ery resistance to tet and ery sensitive recipients by conjugation. In each mating, antibiotic resistance was transferred at frequencies averaging 2.4X 10 -6 recombinants/recipient for ery and 3.4 X 10 -6 recombinants/recipient for tet resistance. DNA from each clinical specimen, as well as the recombinants mentioned above was probed with tn916. Autoradiographs revealed that several clinical isolates and recombinants bound the probe. It is concluded that conjugation of antibiotic resistance in these clinical strains is possibly mediated by a transposon similar to tn916.
INTRODUCTION
Several reports have appeared in recent years concerning antibiotic sensitivity in Streptococcus suis (Chengappa et al., 1986; Hoffman and Henderson, 1985; Touil et al., 1988; Sanford, 1989). These studies indicated that clinical S. suis isolates were uniformly sensitive to penicillin class antibiotics, trimethoprim-sulfamethoxazole, and closely followed by gentamicin (i.e. approximately 90% sensitive from studies cited above). However, resistance to several antibiotics which usually affect Gram positive organisms has also been reported. Hoffman and Henderson (1985 ) and Chengappa et al. (1986 ) reported high levels of resistance to erythromycin (approximately 50% of isolates), lincomycin (64%), and tetracycline (82%). Resistance to these anti0 3 7 8 - 1 1 3 5 / 9 2 / $ 0 5 . 0 0 © 1992 Elsevier Science Publishers B.V. All rights reserved.
214
J.G. STUART ET AL.
biotics occurs at similar high frequencies in other streptococcal species such as S. faecalis and S. pneumoniae (Clewell, 1981 ). Also, according to Clewell ( 1981 ), erythromycin and lincomycin resistance are frequently controlled by plasmid, while tetracycline resistance may be controlled by a plasmid or transposon. Transfer of antibiotic resistance in streptococci is most commonly mediated by conjugation, but transduction and transformation occur in a few species (Clewell, 1981 ). Conjugation in streptococci was first described by Jacob and Hobbs ( 1974 ), who noted conjugal transfer of multiple antibiotic resistance by a plasmid. Since 1974 many conjugal R plasmids have been described for S. faecalis and other streptococcal species (Clewell, 1981; Clewell and Gawron-Burke, 1986). The first transposon (tn916) discovered in S. faecalis was described by Franke and Clewell ( 1981 ). This transposon is 16.4 MDa in size, and carries a tetracycline resistance (tet) r determinant. Also, according to Franke and Clewell ( 1981 ), this transposon mediated conjugation of tet r at low frequencies. Since 1981 other reports of conjugal transposons have been made within S. faecalis and other streptococcal species (Clewell and Gawron-Burke, 1986 ). Burdett et al. ( 1982 ) described three types oftet determinants from a variety of streptococcal species. The tet M determinant was found to be genetically similar to tn916, and the most common determinant associated with chromosomal tet among a variety of streptococcal species. It should be noted that the tet M probe used by Burdett et al. (1982) was made from a strain containing a conjugative transposon. More recently, LeBouquenec et al. ( 1988 ), working with a clinical isolate of S. pyogenes, discovered a very large conjugative element, tn3707, encoding resistance to erythromycin (cry) and tetracycline/minocycline(min). This transposon was subsequently mapped by LeBouquenec et al. (1990), and found to carry a tn916-like element designated tn3703 which carried the resistance determinants for ery and tet/min. This report includes the first observation of conjugation in S. suis. Also data are included that indicate conjugation of antibiotic resistance may be mediated by a transposon similar to tn916. MATERIALS AND METHODS
Bacterial strains Clinical isolates of S. suis were identified biochemically and serotyped as described earlier (Chengappa et al., 1986). Each isolate was taken from a separate animal and designated by serotype and the first two numbers assigned to the clinical case. S. suis type 1 was obtained from Dr. C.H. Armstrong, Animal Disease Diagnostic Laboratory, Purdue University, West Lafayette, IN 47907. Strain SSRF is a rifampin and fusidic acid resistant mutant derived from the type 1 strain, and is devoid of plasmid DNA. Strain Challis
CONJUGATION OF ANTIBIOTIC RESISTANCE IN STREPTOCOCCUS SUIS
215
was a gift from Joe Ferretti, Dept. Microbiol. and Immunol., Univ. Okla., Oklahoma City, OK 73190. Challis is a group H Streptococcus and served as a competent recipient in transformation experiments. Challis (pSM19035) contains a plasmid conferring macrolide, lincomycin streptogramin (MLS) resistance. This strain was used to isolate pSM 19035 which served as a positive control for transformation experiments. Strain CG 120 was obtained from Don Clewell, Dept. of Oral Biology and Microbiology/Immunology, Schools of Dentistry and Medicine, Univ. Mich., Ann Arbor, MI 48109. Strain CG 120 is an E. coli strain constructed by Clewell et al. ( 1981 ) and contains pAM 120, a pBR322 derivative with a tn916 insert generated by Eco R 1. Stock cultures of the bacterial strains were made using 1 ml of horse serum containing 20% sterile glycerol. Cell suspensions were held in sterile screw cap vials at - 80 oC.
Media Serum Yeast Todd-Hewitt (SYTH) was used and consisted of 0.6% yeast extract (Difco), 3% Todd-Hewitt broth (Difco), 0.038% K2HPO4, and 5% horse serum (Gibco). Yeast Todd-Hewitt (YTH) agar was prepared by adding 1.5% bacteriological agar (Sigma) to YTH broth. Blood agar (BA) was prepared by adding 5% defibrinated sheep blood to YTH agar. Antibiotics (Sigma) were used in plates at the following concentrations: tet, 10 #g/ml; fusidic acid, 25/zg/ml; rifampin 50/tg/ml; and ery 5 #g/ml. Susceptibility testing Antimicrobial susceptibility testing of the 48 clinical strains was performed by the standard disk diffusion m e t h o d as described by Lennette et al. ( 1985 ) with a minor modification in the preparation of the inoculum (D'Amato et al., 1982). DNA isolation and gel electrophoresis S. suis DNA was isolated by a procedure modified from Clewell et al. (1974). Ten ml of SYTH broth was inoculated with 100/A from a S. suis frozen suspension culture, and incubated overnight at 30 ° C. Cells from the overnight culture were centrifuged at 8000 rpm for 10 rain in a Sorval RC-5, the supernatant decanted, and the pellet suspended in 3 ml of potassium acetate (200 mM, pH 6.5 ). The cells were centrifuged as before, and suspended in 3 ml Tris-EDTA (TE) (pH 8) buffer plus 0.2 ml of 0.5 M EDTA (pH 8). Freshly prepared lysozyme (5 mg; Sigma) was added to the cell suspension and incubation proceeded for 1 h at 37 ° C. A 0.2 ml aliquot of Sigma Protease I ( 10 m g / m l ) was added to the suspension and incubation continued for another hour at 37°C. The cell suspension was clarified by adding 0.2 ml of a 10% solution of sarcosyl. The suspension was extracted with an equal volume of phenol:chloroform ( 1:1 ), then centrifuged at 10 000 rpm for 10 min at 4 ° C. The upper aqueous layer was removed and 1/ 10 volume of 2.5 M so-
216
J.G. STUART ET AL,
dium acetate was added. DNA was precipitated by adding 2 vol of cold ethanol and holding at - 70 ° C for 30 min. DNA was pelleted by centrifuging at 10 000 rpm for 15 min at 4 ° C. The supernatant was decanted, and the pellet dried and resuspended in 200/tl of TE buffer. DNA samples were held at 4 ° C until ready for analysis by gel electrophoresis. Procedures for agarose gel electrophoresis were taken from Maniatis et al. ( 1982 ). Samples (5/zl; approx. 1 #g DNA) were loaded on a 0.6% agarose gel containing ethidium bromide (5 /zg/ml), and run at 100 volts for 2 h.
Plasmid isolation and transformation Plasmids from all bacterial strains were isolated by CsC1 density gradient centrifugation as described in Maniatis et al. (1982). Transformation was performed by the procedure described by LeBlanc and Hassel ( 1976 ).
Conjugation The procedure for conjugation was adapted from Franke and Clewell ( 1981 ). An overnight culture (0.2 ml) from a clinical strain was mixed with 1 ml of an overnight culture of recipient strain SSRF. This mixture was filtered through a 0.45/zm filter (Millipore), and the filter was placed cell side down on the surface of a BA plate. Incubation proceeded for approximately 18 h at 30°C. The cells were then suspended in 3 ml of YTH broth and spread on plates containing ery or tet (to select against recipients), as well as rifampin and fusidic acid (to select against donors). Controls consisting of donor and recipient cells alone were treated similarly to determine the frequency of spontaneous drug resistant mutants in the population. Transfer frequency was measured by dividing the n u m b e r of transconjugants per ml by the viable count of the recipient (SSRF) cells in the mating mixture. Colonies were counted after incubation for 48 hrs. at 37°C. Some experiments included DNAse I (Sigma) at a concentration of 10/~g/ml in the mating mixture and BA plates.
Dot blot procedure The tn916 probe was prepared by isolating pAM 120 from strain CG 120 by methods described in Maniatis et al. (1982). The tn916 insert was removed by digestion with Eco R 1 (BRL), and purified by DEAE membrane binding (Schliecher and Schuell). A radioactive o~ 32p labeled copy of tn916 was synthesized by nick translation, and used to probe 1/zg of DNA from each sample spotted onto nitrocellulose sheets. A positive control consisted of probing 50 ng of pAM 120 on nitrocellulose, while 1/~g of lambda DNA was used as a negative control. DNA concentrations were estimated using ethidium bromide fluorescence by the minigel method as described in Maniatis et al. (1982) Approximately 10 6 counts per minute per ml of the radioactive probe were used in the hybridization mixture. Following hybridization, the
CONJUGATION OF ANTIBIOTICRESISTANCEIN STREPTOCOCCUSSUIS
217
filter was subjected to several 30 minute washes; twice in 1 X SSPE (25°C), twice in 0.1 x S S P E (25°C), once in 0.1 xSSPE (42°C). After drying, the filter was exposed to X-ray film for 72 h prior to development. RESULTS
Antibiotic sensitivity Table 1 contains data generated from antibiotic susceptibility testing of the 48 clinical isolates examined. High frequencies of resistance were noted to several antibiotics. Resistance of the aminoglycosides, neomycin, kanamycin, and streptomycin is not unusual since these drugs are commonly more effective against Gram negative organisms. However, resistance to erythromycin, clindamycin, and tetracycline is significant, since most Gram positive streptococci are susceptible to these drugs. The results from Table 1 were compared to resistance frequencies of 38 clinical S. suis isolates collected five years earlier (Chengappa et al., 1986). Substantial increases in the frequency of resistance was noted for four antibiotics: ery (11%), clindamycin (29%), neomycin (20%) and tet (18%). Since the S. suis isolates from this study were collected from the same geographic area (western Kentucky) as the earlier clinical isolates, a gene transfer mechanism was suspected. Plasmid screen and transformation Each clinical isolate was screened for the presence of plasmid DNA by DNA isolation and agarose gel electrophoresis. Eleven of 48 isolates were shown to TABLE1 Antimicrobial Susceptibility of Streptococcus suis Drug
Ampicillin Erythromycin Clindamycin Gentamicin Kanamycin Neomycin Nitrofurantoin Spectinomycin Streptomycin Sulfachloropyridizine SXT Tetracycline S = susceptible I = intermediate R = resistant
Number o f Isolates ~
(%) Resistance
S
I
R
48 16 9 40 4 2 35 42 7 25 47 6
0 4 0 3 19 5 10 2 11 2 0 12
0 28 39 4 25 41 2 4 30 21 1 30
0 58 81 9 52 85 4 8 63 44 2 63
218
J.G. STUART ET AL.
contain plasmid DNA, with most plasmid positive strains containing two or more plasmids. Efforts were made to associate plasmid carriage with antibiotic resistance. The average number of drug resistance determinants in plasmid positive strains was 5.1 as compared to 4.7 for plasmid negative strains. Six of eleven (54.5%) plasmid positive strains carried the ery, clindamycin, tet resistant phenotype compared to 13 of 37 (35.1%) of the plasmid negative strains. Plasmid D N A from four S. suis clinical isolates which were plasmid positive and resistant to erythromycin, clindamycin, and tetracycline was isolated and purified. These plasmids along with a positive control from Challis (pSM19035 ), which confers erythromycin resistance, were tested for their ability to transform antibiotic resistance into Challis. Control experiments using pSM 19035 produced as many as 500 ery transformants per plate containing 120 ng of donor DNA. However, in similar experiments using at least 1/tg or more donor DNA, none of the S. suis plasmids were able to transform strain Challis to either tetracycline or erythromycin resistance. Conjugation For these experiments eight clinical strains were selected, four contained plasmids and four did not. Each strain was erythromycin, clindamycin, and tetracycline resistant, and served as a donor in filter mating experiments using the plasmid free SSRF as a recipient. Results of these experiments can be TABLE2
Conjugation o f S . suis clinical donors with strain SSRS ~ Donor
Plasmid
SP 2
Recomb. / ml
Frequency 3
7-45 7-45 8-74 8-74
+ + + +
E T E T
2322 892 127 159
1.5×10 5 1.5>( 10 - 5 5.6×10 7 7.1>(10 7
3-23 3-23 4-27
+ + +
E T E
107 392 8
2 . 1 × 10 - 6 9.3×10 6 2 . 3 × 10 - 7
4-27
+
T
13
2 . 2 X 10 - 7
2-59 2-59 2-20 2-20 7-18 7 - 18 2-15 2-15
-
E T E T E T E T
20 30 159 68 150 203 25 24
~Date presented are an average of at least two conjugation experiments. 2Selection Pressure = Erythromycin or Tetracycline, 3Recombinants/recipients.
1 . 6 × 10 - 7 2.5)
