ELSEVIER FEMS Microbiology Letters 130 (1995) 121-128

MICROBIOLOGY LEll-ERS

Alterations in penicillin-binding proteins in strains of Streptococcus suis possessing moderate and high levels of

resistance to penicillin

Dean Cain a, Fraqois Malouin b91, Michkle Dargis b, Jo&e Hare1 a, Marcel0 Gottschalk ay *

a Groupe de Recherche sur les Maladies Infectieuses Porcines (GREMIP), Fact&t? de Mkdecine Ve’t&inaire, Uniuersitk de Mont&al, C.P.

5000, St-Hyacinthe, Qukbec, Canada J2S 7C6

b Laboratoire et Service d’lnfectiologie, Centre Hospitalier de I’llniuersitk Lava1 (CHUL), Qutbec, Qu&bec, Canada GIV 4G2

Received 6 March 1995; revised 21 April 1995; accepted 24 April 1995

Abstract

We examined the penicillin-binding proteins (PBPs) of certain field strains of Streptococcus suis, as well as those from laboratory variants having different degrees of resistance to penicillin. Results indicated that (i) S. suis possesses three distinct groups of PBPs, arbitrarily named here PBP 1, PBP 2, and PBP 3, with approximate molecular weights of 97, 82,

and 45 kDa respectively; (ii) PBP profiles of field strains of S. suis having different MICs ( 5 0.03 to 16.0 pg/ml) were not uniform (PBP 2 being difficult to detect in strains whose MICs exceeded 0.10 pg/ml, and PBP 3 which exhibited shifts in molecular weight of approximately 5 kDa); (iii) laboratory variant PBPs 1 and 2 showed decreased affinity for penicillin as compared to the parent strain in antibiotic competition experiments, even though the PBP profiles of both were similar. We suggest that PBP modifications (altered molecular weight and/or decreased affinity for penicillin) are involved in the mechanism of resistance to penicillin by S. suis.

Keywords: Penicillin-binding proteins; Streptococcus suis; Penicillin; Antibiotic resistance

1. Introduction

Streptococcus suis is an important swine pathogen responsible for causing a variety of severe infections such as meningitis, septicemia, arthritis, endocarditis,

* Corresponding author. Tel.: (514) 773 8521 ext. 8374; Fax:

(514) 778 8108.

’ Present address: Microcide Pharmaceuticals Inc., Mountain

View, CA 94043, USA.

and pneumonia [l]. In addition to being a zoonotic microbe [2-41, various S. suis infections have been described for the equine [5], bovine [6], canine [7], and feline species [5]. To date, 35 serotypes of S.

suis have been identified, with capsular type 2 being the most prevalent in diseased animals [8].

In the swine industry, penicillin is commonly used in the form of feed medication during periods of high risk (e.g. weaning). Treatment of individual cases must be undertaken early in the course of the disease in order to be successful. In the past, such a routine use of penicillin both as a chemotherapeutic

0378-1097/95/$09.50 0 1995 Federation of European Microbiological Societies. All rights reserved

SSDI 0378-1097(95)00182-4

122 D. Cain et al./FEMS Microbiology Letters 130 (1995) 121-128

and chemoprophylactic agent in the battle against S. suis infections seemed warranted and justified, since

(i) it is relatively inexpensive, (ii) it is easy to administer (oral route), (iii) except in the case of a massive overdose, it is non-toxic and without nega-

tive sequelae to the patient, unlike certain aminogly- cosides [9], and (iv) most strains of S. suis were thought to possess a high susceptibility to this /3- lactam (MIC < 0.03 pg/ml). Previous surveys of

penicillin susceptibility amongst clinical isolates showed these bacteria to be uniformly and highly

susceptible to penicillin [I]. However, recent epi-

demiological studies have suggested that the overall susceptibility to penicillin by clinical isolates of S. suis has significantly changed. Furthermore, several

cases have been reported to be difficult to treat with penicillin, despite employing large doses of the an- tibiotic over a prolonged time lapse [4]. In addition,

several moderately susceptible and some resistant strains of S. suis have recently been isolated [lo].

It has been clearly demonstrated for Streptococ-

cus pneumoniae that an increase in the resistance to penicillin is due to alterations in the structure of its

p-lactam target proteins [ll], resulting in altered

molecular weight and/or lower antibiotic affinity [12]. To our knowledge, no information was previ- ously available concerning the possible role of such

penicillin-binding proteins (PBPs) in the mechanism of resistance to penicillin by S. suis. In order to understand how such resistance develops, we sought to examine the PBP profiles of various field strains

of S. suis, as well as those obtained from

laboratory-generated derivatives possessing different levels of resistance to penicillin.

2. Materials and methods

2.1. Bacterial strains and MC determination

Initially, a total of 60 field strains of S. suis were chosen from our strain bank for penicillin G MIC determination. Of these, 30 belonged to capsular type 2 (the one most often associated with disease), and the 30 others to capsular types other than capsu- lar type 2. These strains all originated from diseased pigs, with the exception of one strain (91-1804) which had been isolated from a case of human

Table I

List of field strains of Streptococcus suis used for penicillin-bind-

ing protein analysis and corresponding penicillin-G minimum

inhibitory concentrations

Number ’ Strain Capsular Pen-G MIC Degree of

type t @g/ml) resistance h

_ 89-5046 2 < 0.03 S _. 91-1804 2 < 0.03 S

1 89-999 2 0.03 s

2 P201.90 2 0.03 S

3 88-4635 2 0.10 S

4 89-4236 9 0.10 S

5 88-407 9 1.0 MS

6 88-7159 Y 1.0 MS

7 88-2818 9 4.0 R

8 88-31’)s 22 4.0 R

9 89-2321A 22 16.0 R

’ Numbers correspond to those indicated in Fig. 1.

” S = susceptible (MIC 5 0.10 pg/ml 2 1.0 pg/ml); R =

resistant (MIC > 1.0 pg/ml).

endocarditis [3]. MICs were determined according to NCCLS standards [13] with Mueller-Hinton agar (Difco Laboratories, Detroit, MI) supplemented with

5% defibrinated sheep blood and various two-fold

dilutions of penicillin G/potassium salt (Sigma Chemical Co., St. Louis, MO); control organisms

Staphylococcus aureus ATCC 29213, Enterococcus

faecalis ATCC 29212, and S. pneumoniae ATCC 35088 were also included. The total number of S. suis strains used for PBP analysis was then narrowed

down to 11, with these strains representing as wide an MIC range as possible; strains were classified as susceptible (MIC 2 0.10 pg/ml), moderately sus-

ceptible (0.10 pg/ml < MIC I 1.0 pg/ml), or re- sistant (MIC > 1.0 pg/ml) to penicillin, according to NCCLS standards [13] (Table 1). All of the S. suis strains included in this work were tested for the production of p-lactamase by a chromogenic cephalosporin method (nitrocefin; BBL Microbiol-

ogy Systems, Cockeysville, MD).

2.2. Generation of increasingly resistant variants

Serial passage Two strains, subsequently identified as being

highly susceptible to penicillin G, 89-5046 and 91- 1804 (MICs of < 0.03 pg/ml) were chosen for variant isolation by serial passage. Briefly, each

D. Cain et al. /FEMS Microbiology Letters 130 (1995) 121-128 123

strain was inoculated into a series of tubes contain- ing Todd-Hewitt broth (THB; Difco), with each tube containing a lower concentration of penicillin G than the next (0.01 pg/ml step per tube). Following an overnight aerobic incubation at 37°C (without shak- ing), the tube which contained the highest concentra- tion of penicillin G and which also exhibited at least slightly turbid growth within a series was used as an inoculum for a fresh THB tube containing the next highest concentration of the antibiotic. For each derivative so obtained, the MIC assay was once again performed as described above to confirm the increase in the level of resistance to penicillin G. Variants were named according to their new MICs (e.g. 89-5046 MO.20 is a derivative originating from strain 89-5046 possessing a penicillin G MIC of 0.20 pg/ml). Cultures grown from clones selected at each level of resistance exhibited a positive reaction when tested with capsular type-specific antiserum.

Electroporation S. suis chromosomal DNA originating from the

most resistant variant obtained (MIC = 0.28 pg/ml; see results) was isolated from a lOlo CFU/ml bacte- rial suspension by the method previously described by Sharma et al. [14], and used for transformation of the penicillin resistance trait into the isogenic parental strain 89-5046 (MIC = 0.03 pg/ml; see results). One hundred ~1 of bacterial cells were placed in a pre-chilled Bio-Rad electroporation cuvette (2 mm electrode gap) and l-5 pg of transforming DNA /ml was added and mixed with the cells. A single pulse was delivered to the cell/DNA mix (2.5 kV, 400 0, 25 PF) with a Bio-Rad Gene-Pulser appara- tus. The cell/DNA mix was then immediately trans- ferred to 1 ml of gene expression medium (TI-IB supplemented with 5 g/l yeast extract (Difco) and 5% horse serum) and incubated for 3 h at 37°C. The cells were then plated out on TSA medium (Difcol supplemented with 5% defibrinated sheep blood and containing a graded series of concentrations of peni- cillin G, and incubated aerobically for at least 48 h. Transformants possessing an MIC higher than previ- ously determined were identified as colonies grow- ing on those plates containing penicillin G concentra- tions which would not allow growth of the control organism (i.e., the same strain electroporated with no transforming DNA), and were subsequently used as

the recipients for the generation of increasingly resis- tant derivatives via further electrotransformation pro- cedures.

2.3. Whole-cell labelling of PBPs

Radiolabelling of PBPs was performed by a modi- fication of the method of Spratt [15], as described by Preston et al. [16], with [1251]penicillin V as the labelled p-lactam. p-(Trimethylstannyl)enicillin V (a gift from L.C. Blaszczak, Eli Lilly, Indianapolis, IN) was iodinated with Nalz51 (Amersham Canada Ltd., Oakville, Ont.). One ml of cells (OD,oo of 0.25) was suspended in 25 ~1 of phosphate-buffered saline (PBS), and the suspension was incubated with 10 ~1 of [‘251]penicillin V (37.3 Ci/mmol; 40 pg/ml) for 30 min at room temperature. The reac- tion was then stopped by washing the labelled cells with a solution containing 1 mg of penicillin V (Sigma) per ml of PBS. The cells were lysed in a DNA thermal cycler apparatus (Perkin-Elmer Cetus) as previously described [17], and suspended in 8 ~1 of sample loading buffer, boiled for 10 min, and 10 ~1 samples were loaded for electrophoresis on dis- continuous 0.1% SDS-lo% polyacrylamide gels. PBP profiles were visualized by autoradiography after exposure of the dried gels to Kodak X-Omat film for 2 to 7 days at -20°C. In PBP competition experi- ments, the cells were prelabelled with various con- centrations of non-radioactive competing penicillin G prior to the addition of [ ‘251]penicillin V as de- scribed previously [18]. The relative percent reduc- tion of radiolabelling of a particular PBP was then determined by scanning the PBP competition profile obtained on film with a Pharmacia model 1282 densitometer and analyzing the data with Pharmacia GelScan 2.1 software on an IBM-compatible PC.

3. Results

3.1. Penicillin G minimum inhibitory concentrations (iwcs)

None of the strains used in this study produced p-lactamase, which is in agreement with previous results that clearly indicate that the mechanism of resistance to penicillin by 5. suis is not mediated or

124 D. Cain et al. /FEMS Microbiology Letters 130 (1995) 121-128

Table 2

Variant strains of Streptococcus suis used for penicillin-binding

protein analysis and corresponding penicillin-G minimum in-

hibitory concentrations

Strain Original MIC Variant MlCs ( pg/ml)

duced colonies only barely visible on the same me- dia and with similar growth conditions, once the

level of resistance was approximately 0.08 pg/ml

or greater.

( &ml) serial passage electroporation

89-5046 < 0.03 0.06 0.04

0.09 0.06

0.15 0.07

0.18 0.13

0.28

91-1804 < 0.03 0.06

0.09

influenced by such an enzyme [lo]. As shown in

Table 1, the penicillin G MICs of the strains used ranged from I 0.03 pg/ml (highly susceptible) to

> 16.0 pg/ml (highly resistant). It is interesting to note that whereas S. suis capsular type 2 is the one most often isolated from diseased individuals, all of

the type 2 strains used here were at least moderately sensitive to the antibiotic.

3.2. Obtention of laboratory-generated variants

Table 2 lists the various derivatives obtained in and selected for this study, either by serial passage or

by electrotransformation. By neither method was the level of resistance augmented rapidly, but rather it occurred in slow, sequential steps (i.e. an increase in MIC of approx. 0.01 pg/ml per round), such that

our selection media necessarily contained graded concentrations of penicillin G, rather than the twofold

serial dilutions normally used in such experiments [11,19]. Furthermore, only a low number of variants

resulted from electroporation of the recipient strains in the presence of transforming DNA (ranging from 1.2 X lo-’ to 1.5 X 10eh transformants/cell; results not shown). It was also noted for all variants that as the level of resistance to penicillin increased, both the growth rate as well as the degree of cy-hemolysis (as seen on TSA plates supplemented with 5% defib- rinated sheep blood) greatly decreased. Whereas par- ent strains 89-5046 and 91-1804 could produce colonies in the l-2 mm range after an overnigh; incubation at 37°C derivatives of either strain pro-

3.3. Field isolate PBP profiles

The non-competitive radiolabelling of S. suis

PBPs revealed that this microorganism possesses

three distinct ‘groups’ of PBPs, here arbitrarily as- signed as PBP 1, PBP 2, and PBP 3 with approxi-

mate molecular weights of 97, 82, and 45 kDa respectively. In general, the PBP profiles of field

strains of S. suis were not uniform. Whereas PBP 1 was present in PBP profiles throughout, PBP 2 was

increasingly difficult to detect in strains whose MICs exceeded 0.10 pg/ml, and PBP 3 exhibited shifts in

molecular weight in the order of about f5 kDa, especially in those strains labelled as moderately sensitive. In addition, even though relatively few strains belonging to the various capsular types were

tested, there seems to be no strict relationship be- tween S. suis capsular types and PBP profiles, since as depicted in Fig. 1 strain 88-4635, a capsular type

2 strain, possesses a PBP profile which varies greatly, especially at the level of PBPs 2 and 3, from that of the two other capsular type 2 strains investigated,

89-5046 and 91-1804 (Fig. 2). However, it is possi- ble that the slight differences in the MICs of these

strains could account for such variance of the PBP profile within a particular capsular type.

PP

-t- Da

1 2 3 4 6 6 7 8 9 v

Penicillin G MlCs @g/ml)

Fig. 1. Penicillin-binding protein profiles of various clinical iso-

lates of Streptococcus suis (see Table 1 for strain nomenclature).

D. Cain et al./FEMS Microbiology Letters 130 (1995) 121-128 125

lO.03 0.00 O.oeI IO.03 0.04 0.00 0.10 0.18 O.ze(

91-1804 893048

Penlcfll& G M/Cs (@g/in/)

Fig. 2. Penicillin-binding protein profiles of Streptococcus suis

strains 89-5046 and 91-1804 and their corresponding variants

selected by serial passage.

3.4. Laboratory-generated variant PBP profiles and competition experiments

In order to determine whether variant strains of S. suis possessing increased levels of resistance to penicillin would exhibit altered PBP profiles, the various derivatives obtained in this study (Table 2) were loaded side by side on an SDS-polyacrylamide gel. There was no visible difference between the PBP profiles of the parent and variant strains ob- tained by serial passage (Fig. 2) or by electrotrans- formation (data not shown). However, in competition experiments between parent strains and their most resistant derivatives, a somewhat reduced affinity for

Penicillin G concentration (ug/ml) Penicillin G concentration (ug/ml)

Fig. 3. Affinity of PBPs 1 (solid lines) and 2 (broken lines) of Fig. 4. Affinity of PBPs 1 (solid lines) and 2 (broken lines) of

Streptococcus suis parent strain 89-5046 (0) and its variant, Streptococcus suis parent strain 91-1804 (0) and its variant,

MO.28 (A) for penicillin. PBP band density (as % of maximum) MO.09 (A) for penicillin. PBP band density (as % of maximum)

is plotted against unlabelled penicillin G. is plotted against unlablled penicillin G.

penicillin by certain PBPs was observed. Figs. 3 and 4 illustrate the relative affinities for penicillin G of the PBPs from parent and variant strains of 89- 5046 and 91-1804, respectively. For either strain, variant PBPs 1 and 2 showed reduced affinities for penicillin G, whereas variant PBP 3 showed little or no difference in binding capacity (data not shown). Nonetheless, although strain 89-5046 possesses a derivative with a higher MIC than that of 91-1804 (0.28 pg/ml vs. 0.09 pg/ml), a decrease in peni- cillin-binding activity by PBP 2 of the former deriva- tive is only evident at higher concentrations of the antibiotic, well above its penicillin G MIC (Fig. 3).

4. Discussion

4.1. S. suis PBP profiles

The findings described here confirm that the mechanism of resistance to penicillin by the S. suis strains studied herein is not mediated by enzymatic degradation of the /3-lactam molecule, but rather, involves, at least in part, modifications in the form of altered molecular weight and/or a decrease in the penicillin-binding capacity of its /3-lactam target pro- teins. Indeed, these results strongly suggest, as is the case for S. pneumoniae [20], that an increase in resistance to penicillin by S. suis is brought about by distinct cumulative alterations in its PBPs. This fact

126 II. Cain et al. / FEMS Microbiology Letters 130 (1995) 121-128

was well demonstrated by the increasingly resistant variants obtained by serial passage and by electro- transformation, since the level of resistance achiev-

able in a single step was always very slowly progres-

sive (i.e. a maximum increase in MIC of approx. 0.01 pg/ml per round).

The results obtained from the electrotransforma-

tion experiments suggest that penicillin resistance in S. suis is chromosomally encoded, and it is most likely a multigenic property, given the low number

of transformants obtained in a single round of elec- troporation (data not shown). Though it is conceiv- able that S. suis may be capable of natural transfor- mation given the proper conditions, such resistance

to /3-lactams is likely to be acquired via natural transformation only under very specific growth con-

ditions, as is the case for S. pneumoniae [l 11, since

several attempts at such transformation on our part were unsuccessful (unpublished results). In addition,

the electrotransformation of strain 89-5046, a capsu- lar type 2, using DNA from non-homologous resis- tant strains (i.e. several strains of capsular types 9

and 22 with penicillin G MICs > 8.0 pg/ml) were also unsuccessful, indicating that acquisition of resis- tance by a particular capsular type from a different one would be unlikely, as is the case for other streptococci [20,21]. Given these barriers, it could

therefore be explained, at least in part, why a major- ity of the present-day field strains of S. suis are still

at least moderately sensitive to the bactericidal ac- tion of penicillin.

4.2. Role of other factors in acquisition of resistance

to penicillin by S. suis

To our knowledge, this is the first report which describes the PBP profiles of S. suis. Little or no other information is available concerning the mecha- nism of resistance to penicillin by S. suis. Therefore, it is evident that many other facets of this subject warrant further investigation. As an example, it is still not known why highly resistant strains of S. suis (penicillin G MICs > 1.0 pg/ml) apparently possess normal growth rates and cell division times, whereas the variants obtained in this study failed to grow normally once an MIC threshold of about 0.0: pg/ml was surpassed. It is possible that over a prolonged time lapse (i.e. years) under the constant

selective pressure of subMIC levels of antibiotic,

high levels of resistance are acquired due to PBPs which have greatly reduced affinity for penicillin,

but can, via an as of yet unidentified mechanism,

preferentially bind to their natural substrate, acyl-D- alanyl-D-alanine, instead of the p-lactam molecule, thereby conserving a normal growth rate. Moreover, it has been reported for S. pneumoniae that transfor-

mants possessing an increased level of resistance to penicillin as well as modified PBPs which accom-

pany this increase, also possess a peptidoglycan structure which differs greatly from that of the iso- genie parent strain, both in terms of architecture and of amino acid composition, such that the variant

PBPs would preferentially recognize precursors with increased hydrophobicity, i.e., those containing di-

alanyl or alanylserine bridges, rather than the usual,

less hydrophobic, stem peptides [22]. In addition, we can only speculate on the apparent relationship be- tween the decrease in affinity of PBP 2 and the shifts

in molecular weight of PBP 3 exhibited by the increasingly resistant field strains of S. suis (Fig. 11.

It is conceivable that PBP 3, which is considered as non-essential to the survival of S. pneumoniae [23],

could possibly assume a compensatory role for PBP 2 in the assembly of the peptidoglycan, such that normal or quasi-normal growth and cell division

rates are maintained. Finally, the importance of the

phenomenon of tolerance must also be elucidated in order to draw any conclusions on the depth of the implication of the PBPs of S. suis in their contribu-

tion to resistance in vivo. Although often loosely described, it is generally accepted that an organism is tolerant when its MBC (minimal bactericidal concen- tration) for a given antibiotic is larger than its MIC by a factor of at least 32, such that at or slightly above MIC levels, the antibiotic would tend to exert

a bacteriostatic rather than a bactericidal effect on the microbe [24]. As previously reported for certain strains of S. pneumoniae identified as penicillin- tolerant, an uncoupling of the primary effects of penicillin (i.e. inhibition of the PBPs) from the trig- gering of autolytic activity has been observed [20]. In preliminary experiments, we have identified several strains of S. suis which are highly susceptible to penicillin (MIC < 0.03 pg/ml) and yet which fail to lyse properly even in the presence of up to 100 times the penicillin G MIC in the culture medium. It is

D. Cain et al. /FEW Microbiology Letters 130 (1995) 121-128 127

possible that lysis-defective strains, even though highly sensitive to penicillin, may persist in vivo even when antibiotic levels are superior to the MIC; this would explain the increasing frequency in failure of treatment and/or relapse of infections due to S. suis once chemotherapy is ceased [17]. Such selec- tive pressure could then possibly create a reservoir from which new highly resistant strains could even- tually emerge.

Acknowledgements

The authors wish to thank Dr. Robert Higgins, Nathalie Charland, and Serge Harpin for their critical review of the manuscript. F.M. was the recipient of an independant research scholarship from the Medi- cal Research Council of Canada.

References

[l] Touil, F. and Higgins, R. (1988) Les infections a Streptococ-

cus suis chez I’esp&ce porcine: revue de la litterature. Med.

Vet. Quebec 18, 25-33.

[2] Arends, J.P. and Zanen, H.C. (1988) Meningitis caused by

Streptococcus suis in humans. Rev. Infect. Dis. 10, 131-137.

[31

141

El

[61

171

k31

191

Trottier, S., Higgins, R., Brochu, G. and Gottschalk, M.

(1991) A case of hunan endocarditis due to Streptococcus

suis in North America. Rev. Infect. Dis. 13, 1251-1252.

Woo, J. (1987) Streptococcus suis meningitis requires pro-

longed treatment with penicillin. Infection 15, 129-130.

Devriese, L.A. and Haesebrouck, F. (1992) Streptococcus

suis infections in horses and cats. Vet. Rec. 130, 380.

Higgins, R., Gottschalk, M., Fecteau, G., Sauvageau, R.,

DeGuise, S. and DuTremblay, D. (1990) Isolation of Strepto-

coccus suis from cattle. Can. Vet. J. 31, 529.

Keymer, I.F., Heath, SE. and Wood, J.G.P. (1983) Strepto-

coccus suis type II infection in a raccoon dog Wycterentes

procyonoides). Vet. Rec. 113, 624.

Higgins, R., Got&chalk, M., Beaudoin, M. and Rawluk, S.A.

(1992) Distribution of Streptococcus suis capsular types in

Quebec and western Canada. Can. Vet. J. 33, 27-30.

Shneerson, J.M., Chattopadhyay, B., Murphy, M.F.G. and

Fawcett, I.W. (1980) Permanent perceptive deafness due to

Steptococcus suis type II infection. J. Laryngol. Otol. 94,

425-427.

[lo] Gottschalk, M., Turgeon, P., Higgins, R., Beaudoin, M. and

Bourgault, A.M. (1991) Susceptibility of Streptococcus suis to penicillin. J. Vet. Diagn. Invest. 3, 170-172.

[ll] Handwerger, S. and Tomasz, A. (1986) Alterations in peni-

cillin-binding proteins of clinical and laboratory isolates of

pathogenic Streptococcus pneumoniae with low levels of

penicillin resistance. J. Infect. Dis. 153, 83-89.

[12] Chalkley, L.J. and Koomhof, H.J. (1988) Penicillin-binding

proteins of Streptococcus pneumoniae. J. Antimicrob.

Chemother. 22, 791-800.

[13] National Committee for Clinical Laboratory Standards (1991)

Methods for dilution antimicrobial susceptibility tests for

bacteria that grow aerobically. NCCLS document M7-A2. In:

Antimicrobial Suceptibility Testing (third edition). Villanova,

PA.

[14] Sharma, R.C., Murphy, A.J.M., Dewald, M.G. and Shimke,

R.T. (1993) A rapid procedure for isolation of RNA-free

genomic DNA from mammalian cells. Biotechniques 14,

176-178.

1151 Spratt, B.G. (1977) Properties of penicillin-binding proteins

of Escherichia coli K-12. Eur. J. Biochem. 72, 341-352.

[16] Preston, D.A., Wu, C.Y.E., Blaszczak, L.C., Seitz, D.E. and

Halligan, N.G. (1990) Biological characterization of a new

radioactive labelling agent for bacterial penicillin-binding

proteins. Antimicrob. Agents Chemother. 34, 718-721.

1171 Ubukata, K., Nakagami, S., Nitta, A., Yamane, A.,

Kawakami, S., Sugiura, M. and Konno, M. (1992) Rapid

detection of the mecA gene in methcillin-resistant staphylo-

cocci by enzymatic detection of polymerase chain reaction

products. J. Clin. Microbial. 30, 1728-1733.

[18] Rousseau, N., Dargis, M., Gourde, P., Beauchamp, D. and

Malouin, F. (1992) Effect of Plactams on peptidoglycan

metabolism of Haemophilus influenzae grown in animals.

Antimicrob. Agents Chemother. 36, 2147-215.5.

[19] Zighelboim, S. and Tomasz, A. (1980) Penicillin-binding

proteins of multiply antibiotic-resistant South African strains

of Streptococcus pneumoniae. Antimicrob. Agents

Chemother. 17, 343-442.

[20] Handwerger, S. and Tomasz, A. (1985) Antibiotic tolerance

among clinical isolates of bacteria. Rev. Infect. Dis. 7,

368-386.

[21] Jabes, D., Nachman, S. and Tomasz, A. (1989) Penicillin-bi-

nding protein families: evidence for the clonal nature of

penicillin resistance in clinical isolates of pneumococci. J.

Infect. Dis. 159, 16-25.

[22] Garcia-Bustos, J.F., Chait, B.T. and Tomasz, A. (1988)

Altered peptidoglycan structure in a pneumococcal transfor-

mant resistant to penicillin. J. Bacterial. 170, 2143-2147.

[23] Chambers, H.F. (1986) Understanding penicillin-binding pro-

teins. Antimicrob. News]. 3, 47-54.

[24] Holbrook, W.P., 6lafsd6ttir, D., Magi&son, H.B. and

Benediktsd6ttir, E. (1988) Penicillin tolerance among oral

streptococci. J. Mol. Microbial. 27, 17-22.