Original Report

Penicillin susceptibility and molecular characteristics of clinical isolates of Streptococcus pneumoniae at the University of Malaya Medical Center, Kuala Lumpur, Malaysia

Mohd Nasir Mohd Desa,@) Thong Kwai Lin,c2) Rohani Md Yasin(3) and Navaratnam Parasakthic4)

Objective: To determine the prevalence of penicillin resistance and molecular characteristics of pneumococcal isolates at the University of Malaya Medical Center.

Methods: From March 1999 to July 2000,100 clinical isolates of Streptococcus pneumoniae were obtained from 93 patients of various ages and from various body sites. The minimum inhibitory concentrations (MICs) for penicillin and ceftriaxone were determined by E test, and results were interpreted according to guidelines recommended by the National Committee for Clinical Laboratory Standards (NCCLS). Fifty isolates were further serotyped, and analyzed by pulsed-field gel electrophoresis (PFGE) and polymerase chain reaction/restriction fragment length polymorphism (PCR-RFLP) of the penicillin-binding protein (pbp) 2b and 2x genes.

Results: The majority of the isolates were from respiratory sites. Thirty-one isolates showed decreased susceptibility to penicillin (PRSP), and many of these also showed decreased susceptibility to ceftriaxone. Twelve serogroup/types (SGTs) were present, with 19F being the most common. PFGE analysis identified two dominant profiles, consisting mainly of PRSPs that had common serotypes (19F) and pbp gene patterns within their respective groups, although PCR-RFLP analysis showed different patterns of pbp genes among the PRSPs as compared to penicillin-susceptible strains, which had a uniform pattern.

Conclusion: PRSPs were genetically related as shown by PFGE and serotype. The consistency of pbp gene patterns, observed among many of the PRSPs within their respective PFGE profiles, supported their relatedness as established by PFGE.

Int J Infect Dis 2003; 7: 190-197

INTRODUCTION

In the past, most clinical isolates of Streptococcus pneumoniae were highly susceptible to penicillin and a variety of other antimicrobial agents. Penicillin itself has remained the drug of choice for most pneumococcal infections for more than 50 years. Presently, bacterial infections caused by S. pneumoniae strains with

(i)Department of Clinical Laboratory Science, Faculty of Medicine and Health Science, Universiti Putra Malaysia, Malaysia; @)Institute of Biological Science, Faculty of Science, University of Malaya, Malaysia; (3)Bacteriology Division, Institute for Medical Research, Kuala Lumpur, Malaysia; c4)Department of Medical Microbiology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia. E-mail: parasak@ummc.edu.my

Supported by the University of Malaya VoteF grant F0053/2000B and IRPA grant 06-02-03-0759 from the Ministry of Science, Technology and Environment, Malaysia.

Address correspondence to Navaratnam Parasakthi, Medical Microbiology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia. E-mail: parasak@ummc.edu.my

Corresponding Editor: Robert Gilman, Baltimore, USA

decreased susceptibility to penicillin are being increas- ingly reported worldwide. This is compounded by the tendency of such isolates to be multiresistant. Pneumo- coccal resistance has been reported for p-lactam antibiotics, including the third-generation cephalosporins, as well as other non-p-lactam drugs, such as macrolides, chloramphenicol, clindamycin, tetracyclines, and trimethroprim-sulfamethoxazole.1-4 Multiresistance limits treatment options, prolongs hospitalization, and increases the cost of antibiotic therapy. Hence, studies have been carried out throughout the world to determine the prevalence rate of the pneumococcal isolates with decreased susceptibility to penicillin, their serotypes, and their patterns of susceptibility to other commonly used antibiotics. Molecular investigations of the isolates have also been carried out to determine the molecular epidemiology of antimicrobial resistance. One of the useful and widely used techniques is pulsed- field gel electrophoresis (PFGE). This technique, which involves digesting all of the genomic DNA with a rare cutting restriction enzyme (e.g. SmaI), is useful in evaluating the overall genetic relatedness of chromo- somal DNA, and has been shown to be highly

Penicillin susceptibility and molecular characteristics of clinical isolates of Streptococcus pneumoniae I Desa et al 191

discriminatory and reproducible.5-7 Another commonly used technique is restriction fragment length poly- morphism (RFLP) of the penicillin-binding protein (pbp) genes. RFLP analysis involves the PCR amplific- ation of the pbp genes (e.g. pbp 2x or pbp 2b) with specific primers for each of the pbp genes, followed by the digestion of the genes with a restriction enzyme (e.g. Hi@). This method has been shown to be useful in investigating the genetic relatedness of resistant strains&l0 Application of these techniques alone or in combination has provided information on the molecular characteristics that will provide a better understanding of the spread of antibiotic resistance.

In Malaysia, the increasing incidence of pneumo- coccal isolates with decreased susceptibility to penicillin is also of concern. In this report, we describe the prevalence of penicillin resistance, serotype distribution and the molecular characteristics of pneumococcal isolates at the University of Malaya Medical Center (UMMC), a tertiary-care center in Malaysia serving the metropolitan city of Kuala Lumpur and its surrounding area. Given the fact that a high diversity of the pbp 2b and pbp 2x genes has been observed among pneumo- cocci with decreased susceptibility to penicillin,ll-l4 RFLP analysis of the two pbp genes was also carried out and the results correlated with those obtained by PFGE.

MATERIALS AND METHODS

Pneumococcal strains

One hundred clinical isolates of S. pneumoniae were obtained from various body sites of 93 patients of various ages at the UMMC, from March 1999 to July 2000. Seven pairs of isolates were obtained from seven individuals but from different body sites. All isolates were confirmed as S. pneumoniae by standard bacterio- logic methods and susceptibility to optochin and bile solubility. Susceptibility to penicillin was pretested using l-mg oxacillin disks and confirmed by minimum inhibitory concentration (MIC) determination by the E- test method described below. Strains were then stored at -70°C prior to further analyses. Unfortunately, only 50 strains survived the long-term storage, and serotyping and molecular studies were carried out only on these strains.

Serotyping

Isolates were serotyped with the Pneumotest kit (Statens Seruminstitut, Copenhagen, Denmark) by direct suspension of the antisera with pure colonies of pneumococci emulsified in phosphate-buffered saline (PBS) (pH 7.3) on glass slides. The test was carried out with 12-pooled antisera (A-F, H, and P-T) until a positive reaction was observed.15 The serogroupltypes (SGTs) were then established based on the reaction pattern using the table provided. Further typing of

serogroups into serotypes was then performed with specific factor sera.

Antibiotic susceptibility pattern

MICs of penicillin and ceftriaxone were determined by the E test (AB Biodisk, Solna, Sweden), using Mueller-Hinton agar supplemented with 5% sheep blood, and the inoculum recommended by the National Committee for Clinical Laboratory Standards (NCCLS) for the disk diffusion method.16 The plates were incub- ated at 35°C for 18-24 h under 5% COz. S. pneumoniae ATCC 49619 was used as the quality control strain. MICs that fell between any two doubling dilutions were rounded up to the next higher one, and the susceptibility categories were then assigned according to MIC (mg/mL) breakpoints, as in the NCCLS guidelines: penicillin-resistant (R) = 22.0, intermediate (I) = 0.12- 1.0, susceptible (S) = 10.06; ceftriaxone-R = ~2.0, I = 1.0, s = 50.5.17

DNA preparation

Bacterial cells, grown overnight on blood agar plates in CO2 at 37”C, were scraped from the surface of the plates and washed by suspending in 1 mL of SB (10 mM Tris- HCl (pH 7.5), 1 M NaCl). The cell suspension was then centrifuged at 8000 rev/min at 4°C for 10 min, and the supernatant was discarded. The washing step was repeated twice, and cell density was adjusted to an optical density of 0.4 at 650 nm.18 The washed cells were then resuspended in an equal volume of SB and 1.5% low-melting agarose (Sea-Plaque; FMC Bioproducts, Rockland, ME, USA), before being transferred into a mold (Bio-Rad Laboratories, Hercules, CA, USA). The solidified agarose-cell plugs were incubated in 2 mL of lysis buffer (10 mM Tris-HCI, 100 mM NaCl, 0.5% 20 cetyl ether, 0.2% deoxycholate, 0.5% sarkosyl), 5 mL of RNaseA (10 mg/mL), 20 mL of lysozyme (20 mg/mL) and 20 mL of lysostaphin (1 mg/mL) for 2 h at 37°C. The lysis buffer was discarded, 2 mL of PB (0.5 M EDTA (pH S.O), 1% sarkosyl) containing 20 mL of proteinase K (10 mg/mL) (Sigma Chemicals, St Louis, MO, USA) was added, and the solution was incubated for another 16 h at 50°C. Subsequently, the plugs were thoroughly washed 2-hourly with TE buffer (1 M Tris (pH 8.0), 0.5 M EDTA) at least three times at room temperature with gentle agitation.

Restriction enzyme digestion and PFGE

A slice of the plug (2 x 3 x 3 mm) was digested overnight with 10 U of SmaI (Promega, Madison, WI, USA) in 100 mL of restriction buffer at 25°C. The digested chro- mosomal DNA was then separated by PFGE using the CHEF-DRII apparatus (Bio-Rad Laboratories) in 1.2% agarose gel, with pulse times ramped from 1 to 35 s over 23 h at 200 V and a temperature of 14OC in 0.5xtris-

192 International Journal of Infectious Diseases I Volume 7, Number 3,2003

borate EDTA (TBE). After electrophoresis, the gels were stained with 0.5 mg/mL ethidium bromide, destained in distilled water, and photographed under UV transillumination.

PCWRFLP of pbp 2b and pbp 2x

Bacterial cells were grown overnight on blood agar at 37°C in COz. A few single colonies were picked and placed in 50 mL of sterile dHz0 and heated to 95°C for 5 min in a thermocycler (Mastercycler gradient, Eppendorf, Hamburg, Germany). A 5-mL aliquot was used in a 50-mL PCR reaction.19 The pbp 2b and pbp 2x genes were amplified as 1.5-kb and 2.0-kb fragments respectively, with primers as published by Gillespie et all9 The reaction mix consisted of 1.5 mM MgC12, 1 x PCR buffer, 200 mM of each dNTP, 1 mM of each primer, upstream and downstream, and 1 U of Taq polymerase (Promega). The thermocycling protocol was 95°C for 5 min, 94°C for 1 min, 55°C for 2 min and 72°C for 3 min for 30 cycles, and a final incubation at 72°C for 7 min. The amplified DNA of both genes was directly subjected to RE digestion with Z+zfl (Promega).Twenty microliters of the PCR product was mixed with 5 mL of [multi]10 restriction buffer, 25 mL of dHz0 and 10 U of Hinfl, and incubated at 37°C for 2 h. The digested PCR product was separated on a 1.8% agarose gel, visualized by incorporating ethidium bromide into the gel, and photographed under UV transillumination.

DNA pattern analysis

DNA bands were visually analyzed, and each unique pattern was assigned an arbitrary letter/number. For the DNA patterns generated by PFGE, the interpretation of the relatedness was based on the guidelines recom- mended by Tenover et al, .20 the identical patterns were considered as related, those differing by l-3 DNA bands as closely related, and those differing by 4-6 DNA bands as possibly related. In this study, patterns with differ- ences of 56 DNA bands were given a letter followed by a subscript number 1 or 2 for the closely and possibly related respectively. A lower case letter was also assigned, following the number, if the closely or possibly

related patterns were different from each other. Those with differences of >6 bands were considered unrelated and given a different letter. The PFGE gel was also scanned and analyzed using computer comparisons with the GelCompar (version 4) software (Applied Maths, Kortrijk, Belgium), in which a dendogram, showing the overall relatedness of the strains, was computed by the Dice method and by clustering by the unweighted pair group method with averages.

RESULTS

All the 100 clinical isolates of S. pneumoniae were from patients with suspected bacterial infections that were sporadic and presumably community-acquired. The iso- lates were derived from sputum (27%), nasopharyngeal secretion (NPS) (25%), blood (17%) tracheal secretion (12%) eye (6%) throat (4%), ear (3%), pleural fluid (3%),vagina (1%) and cerebrospinal fluid (CSF) (1%); there was also one of unknown source. Respiratory samples (sputum, NPS, tracheal, and throat) accounted for 68 strains, invasive samples (blood, CSF, and pleural fluid) for 21 strains, and other samples (ear, eye, and vagina) for 10 strains. The ages of patients from whom the isolates were obtained were as follows: ~13 years, 48 strains; 2 13 years, 49 strains; ages not traceable, 3 strains. All of these isolates were tested for susceptibility.

Based on the NCCLS breakpoints, 31 isolates showed decreased susceptibility to penicillin, with 20 isolates categorized as resistant (PRSP) and 11 isolates as intermediate (PISP). Many PRSPs also showed decreased susceptibility to ceftriaxone, with three (15%) and 13 (65%) strains being resistant and intermediately resistant respectively; Of PISPs, only three (27%) strains were intermediately resistant, while the rest, as well as all the penicillin-susceptible strains (PSSPs) (69 strains), were susceptible. Most strains that were isolated from the same patients had similar categories of susceptibility for both penicillin and ceftriaxone, with MIC values differing by not more than a one-fold dilution. Among the strains with decreased susceptibility to penicillin and ceftriaxone, a slightly higher percentage was isolated from respiratory and other noninvasive samples than from the invasive ones (Table 1). A higher percentage was also isolated from

Table 1. Distribution of .S. pneumoniae isolates with respect to the site of origin and susceptibility to penicillin and ceftriaxone

Number (‘%)a of isolates

Number of isolates n = 99

Penicillin Ceftriaxone

R I 5 R I 5 n= lgb n= 71 n = 69 n=3 n = lSb n = 81

Respiratory 68 13 (19.1) 9 (13.2) 46 (67.7) 1 (1.5) 12 (17.6) 55 (80.9) Invasive 21 4 (19.0) 0 17 (81.0) 0 3 (14.3) 18 (85.7) Others 10 2 (20.0) 2 (20.0) 6 (60.0) 2 (20.0) 0 8 (80.0)

R, resistant; I, intermediate; S, susceptible. a Percentage was based on 99 strains, excluding one strait+ (R to penicillin and I to ceftriaxone) of unknown site of origin.

Penicillin susceptibility and molecular characteristics of clinical isolates of Streptococcus pneumoniae I Desa et al 193

patients ~13 years old, as compared to those 2 13 years old (against penicillin, 22.9% of strains were resistant as compared to l&4%, and 12.5% of strains were inter- mediate as compared to 10.2%, respectively, while against ceftriaxone, 4.2% of strains were resistant as compared to 2.0%, and 18.7% of strains were intermediate as compared to 14.3%, respectively). However, it is difficult to comment on the significance of these figures, since the sample representatives of the site of origin and the

population size were small, and the data were based only on the percentages, without any statistical analysis.

Fifty strains were serotyped and analyzed further with molecular techniques. Thirty-four of the 50 strains were obtained from respiratory sites, nine were from invasive sites, and seven were from other sites. Forty- eight per cent of these strains came from patients ~13 years old, while 52% of these strains were isolated from patients B 13 years old.

List: Noname Enlnes: 50 Cot-relation: Bands. Dice (Max. tol. 1.2’S Min. Surf. O.O’:S) Zones: [1-400~. Clustering: UPGMA

40 50 60 70 aa 80 Strain

.., ,, ,: ,. '/. . ,_! . . ,..,. I ,- 18

i-f 92

r--J"-+ :5

I ( 85

L ___--r---+ ", i! -77 iI 7

f '

Is'

4

23 12

I 20

- 89 41

1 L-j --+ 93 21

it4 55 72 z 32 53 27 88 98 14 81

95 97 30 6

26

,J K-T. --j 1 . ,- 25 - 35

136 54

-100

SOT PFGE MP~ pbpzr smal mfl

19F ~~ 19F A2d 19F A2c t9F A2C t9F mb ISF A 19F A

238 A2a I

$J 6~ K 7F fi 6A M 6A N N/D 0 :I E lE2 1 p

19F Bli WF B2h 19F Btg 1SF Blf 19F B Z9F 0 1SF B t9F B 14 B2j 14 B2j 14 BZj

23F %2m 23F B2m 15c 1321 19F B2k 7F c 7F c

Nm Q 16A F 15A F2 23A R 19F s

WC T 16136137G 16136137 G2

Pm u V

ii H 15c HZ 23A D 23AD 19F. w

b2 b2 b2 b2 b2 b2 b2 b2 b5 b5 b4 b5

ii b5 b5 W b5 bl bl bl bl bl bl bl bl b5 b5 b5 bl bl b5 bl 65 b5 b5 bs b5 b2 b2 b5 b5 b5 b5 b5 b5 b5 b5 b5 b3

WC (ughl)

xs 0.016 xs eO.016 x3 0.5 x6 0.032 xl3 0.032 x5 0.032 x6 ~0.016 x6 0.016 x6 do.016 6 0.016 xl 2.0 xl 4.0 xl 4.0 xi 4.0 xi 4.0 xl 4.0 Xi 1.0 Xl 0.25 x4 0.5 x4 0.5 x4 0.084 xl 4.0 Xl 4.0 x-5 co.016 xl 4.0 xe 0.032 x6 0.032 E& (1.032 3 0.016 x6 <0.016 xl 2.0 Xl 1.0 x6 0.016 XS qO.016 x6 <0.016 d ~0.016

x8 x5 26 x8 ~o.o16 x6 0.064 xe 0.032 Xl 2.0

1.0 I 1.0 I 1.0 f 0.5 S 0.5 S 0.5 S 0.5 S 2:o R

0.016 S <0.016 S

0.25 S O.U32 S

~0.016 B 0.036 S 0.016 S 0.016 S 0.032 S 0.032 S

i.0 I 1.0 I 2.0 R I .o I I .o I ?.O R I.0 I 0.5 S 03 S

0.25 S 0.064 S

1.0 I 1.0 I

<O.OlS S 1.0 1

0.032 S 0.032 S 0.064 S 0.016 S 0.032 S

1.0 I I .o 1

so.018 S 0.832 S 0.016 S 0.032 S a.25 S

0.016 S <0.016 S 0.032 S 0.032 S

1.0 I

Figure 1. Dendogram of Smal digestion electrophoretic profiles of 5. pneumoniae isolates. On the right side are the strains (represented by numbers), SGTs, PFGE profiles as analyzed visually, pbp Zb and pbp 2x gene patterns, and MlCs for penicillin and ceftriaxone, as well as their category of susceptibility. Each pair of strains, 15 and 17, 88 and 98, and 95 and 97, was isolated from three individuals.

194 International Journal of Znfectious Diseases I Volume 7, Number 3.2003

Forty-five of the 50 strains tested were successfully serotyped. Twelve different SGTs were identified, with serotype 19F being the commonest, accounting for 18 strains, which were mainly PRSPs, followed by 6A, 1,7F, 14, 1X, 23A, 15A, 23F, and 23B, accounting for l-4 strains each. Owing to the limited factor sera, SGTs for one and two strains were determined as only 7B/C and 16/36/37 respectively, and five strains could not be typed because of autoagglutination of the strains in the PBS suspension.

Visual inspection of the SmaI-digested chromo- somal DNAs of the 50 isolates showed a total of 23 distinct PFGE profiles, designated A-W. The dendogram produced by computer analysis also segregated the 50 isolates into clusters that corresponded to their respective PFGE profiles (Figure 1). Profiles A (Figure 2a) and B (Figure 2b) were the most dominant (eight strains with two related and six possibly related, and 15 strains with four related, three closely related and eight possibly related, respectively), and consisted mainly of PRSPs belonging to serotype 19F. Other profiles, consisting of one or two strains each, were mainly associated with PSSPs of various serotypes. Exceptions were noted in profiles K, R, S, V, and W, which were represented by either PRSP or PISP Strains that were isolated from the same patient but from different body sites (only three pairs of strains from three individuals

(a) kb M 1 2 3 4 5 6

were included among the 50 strains tested) showed similar DNA patterns and grouped into the same PFGE profiles accordingly.

The genetic relationship analyzed by PFGE was further reassessed by RFLP analysis of the pbp genes. Generally, it was apparent that strains with decreased susceptibility to penicillin had a number of different RFLP patterns for pbp 2b (pattern bl-b4) and pbp 2x (pattern x1-x4), whereas PSSPs mostly had uniform patterns (pattern b5 and pattern x6, respectively) (Figure 3). However, when analyzing the pbp gene patterns with respect to the PFGE profiles, a common pattern was observed within the respective dominant PFGE profiles A and B, although the majority of the isolates in these PFGE profiles were PRSPs or PISPs (Figure 1); 87.5% of strains of profile A had pattern b2 for pbp 2b and pattern x2 for pbp 2x, whereas 73.3% of strains of profile B had pattern bl and pattern xl respectively. In isolates with other PFGE profiles, pattern b5 forpbp 2b and pattern x6 forpbp 2x were the most common RFLP patterns, particularly among the PSSPs. However, some PRSPs or PISPs with minor PFGE profiles had the pbp patterns commonly found in the dominant PFGE profiles A and B in either one or both of the pbp genes, whereas others had pbp patterns that were either unique or similar to that of the PSSPs.

(b) kb Ml2 3 456 7 8 9

388.0 b 339.5 b

242.5 b 291.0 b

194.0 b

48.5 b

Figure 2. (a) PFGE of representative Smal-digested chromosomal DNA of 5. pneumoniae isolates of profile A consisting of 15 or 16 DNA bands with sizes ranging from 40 to 400 kb: M, lambda DNA concatemer standard marker; lane 1, strains 15 and 17; lane 2, strain 18; lane 3, strain 92; lane 4, strain 85; lane 5, strains 46 and 48; lane 6, strain 77. (b) PFGE of representative Smal-digested chromosomal DNA of 5. pneumoniae isolates of profile B consisting of 15 or 16 DNA bands with sizes ranging from 40 to 400 kb: M, lambda DNA concatemer standard marker; lane 1, strain 79; lane 2, strain 24; lane 3, strain 71; lane 4, strains 28, 29, 55 and 72; lane 5, strain 94; lane 6, strains 27, 32 and 53; lane 7, strains 88 and 98; lane 8, strain 14; lane 9, strain 81.

(4

Penicillin susceptibility and molecular characteristics of clinical isolates of Streptococcus pneumoniae I Desa et al 19.5

Ml2345 (b) M 1 2 3 4 5 6

bp

500 )

300 )

100 )

Figure 3. (a) RFLP of representative k/in&digested pbp 2b genes of 5. pneumoniae isolates consisting of 3-5 DNA bands with sizes ranging from 50 to 700 bp: M, IOO-bp ladder; lanes l-5, profiles l-5 respectively. (b) RFLP of representative Hi&l-digested pbp 2x genes of 5. pneumoniae isolates consisting of 3-6 DNA bands with sizes ranging from 50 to 700 bp: M, IOO-bp ladder; lanes 1-6, profiles l-6 respectively.

DISCUSSION

S. pneumoniae is commonly associated with respiratory infections in both children and adults. Globally, it is among the leading causes of community-acquired pneumonia, otitis media, meningitis, and bacteremia, resulting in morbidity and mortality in young children, persons with underlying chronic systemic conditions, the elderly, and patients in whom diagnosis and antibacterial therapy is delayed.21-23

The emergence of penicillin resistance among S. pneumoniae strains has become an important worldwide problem.15”Q4 In this study, 31% (20% were resistant and 11% were intermediate) of the 100 clinical isolates of pneumococci had decreased susceptibility to penicillin. This figure is higher than that reported by the Asian Network for Surveillance of Resistant Pathogens Study (ANSORP), which reported the prevalence of PRSP in Malaysia as only 9%. 25 The higher rate observed in this study might be partly due to the catchment area of this medical center, which is located in the middle of the densely populated metropolitan city of Kuala Lumpur. Under these circumstances, the spread of PRSPs might have taken place more rapidly and frequently than elsewhere. However, despite the overall low prevalence rate of PRSP in Malaysia, the rate had increased by about 10% in a period of about 8 years since 1988.26

It has been reported that PRSPs usually emerge within certain serotypes and are frequently multi- resistant.1%27-29 In the 50 strains that were analyzed in this study, many PRSPs as well as some PISPs were of serotype 19F, and had varying susceptibility to

ceftriaxone. This finding is consistent with other studies, where serotype 19F was seen to be frequently associated with PRSPs,16 while cross-resistance is consistent with the principal mechanism of penicillin resistance in S. pneumoniae, in which the alterations in penicillin- binding proteins (PBPs) greatly decrease the affinity for almost all [beta]-lactam antibiotics, including the third- generation cephalosporin, ceftriaxone.28,30

Considering that penicillin resistance usually parallels resistance to other antibiotics1 selecting the proper antibiotic therapy is of utmost importance. The choice of antimicrobial agents for treatment is usually influenced by the MIC. However, the relevance of the NCCLS MIC breakpoint for penicillin susceptibility in non-meningitic infections is debatab1e.r There is evidence that treatment with adequate doses of peni- cillin G might still be effective in treating pneumococci with decreased susceptibility to penicillin in pneumonia, but not in otitis media and meningitis.31 With respect to this, the Drug-Resistant S. pneumoniae Therapeutic Working Group (DRSPTWG) proposed that the susceptibility breakpoints for penicillin against pneumo- cocci in pneumonia be raised (resistant = 24.0 mg/mL, intermediate = 2.0 mg/mL, susceptible = 51.0 mg/mL), so that penicillin susceptibility was more appropriately reported in this infection. With use of these criteria, the respiratory isolates in this study showed a decrease in the percentage of strains with decreased susceptibility to penicillin to 19.1% (7.3% were resistant and 11.8% were intermediate) as compared to 32.3% when interpreted by the NCCLS criteria. This lower rate might be more

196 International Journal of Infectious Diseases I Volume 7, Number 3.2003

appropriate in correlating the clinical response of pneumococcal pneumonia to penicillin therapy, enabling clinicians to use a narrower-spectrum antibiotic, instead of a broad-spectrum one that might not be necessary.

PFGE profiling of the 50 pneumococcal isolates showed that the majority of those with decreased susceptibility to penicillin had more homogeneous DNA patterns and clustered in only two dominant PFGE profiles, A and B, as compared to those of PSSPs, which were diverse. The relatedness of these strains with decreased susceptibility to penicillin as established by the PFGE analysis was further validated by the demonstration that they had similar serotype and pbp gene patterns within their respective PFGE profiles. Many of the isolates in these two profiles exhibited serotype 19F, thus supporting the close relatedness of the strains. Although the consistency of the pbp gene fingerprint supports the genetic relatedness of these strains, the fact is that these genes in PRSPs were very highly variable in sequence.iO Given this fact, identical RFLP patterns, which indicate the likely similarity of the gene sequences, strongly suggest the relatedness of the resistant isolates. Consistency of RFLP pattern was also observed among the majority of the PSSPs; the unaltered pbp gene sequence in these penicillin- susceptible isolates would result in a uniform RFLP pattern. Thus, although there were few strains in the dominant profile A and B manifesting different sero- types andpbp gene patterns in one or both of the genes, possibly due to multiple mutational events involving point mutations, capsular switches or horizontal gene transfers, the analyses did show that the majority of the pneumococcal isolates with decreased susceptibility to penicillin were genetically related, suggesting that they could have initially been derived from common origins and subsequently spread within the population. This is consistent with many other studies in which PRSPs have been shown to be genetically related and to have spread from limited sources.32-34

ACKNOWLEDGEMENT

E-test strips were kindly provided by SmithKline Beecham Int., Kuala Lumpur, Malaysia.

REFERENCES

1. Campbell GD, Silberman R. Drug-resistant Streptococcus pneumoniae. Clin Infect Dis 1998; 26:1188-1195.

2. Doern GV, Pfaller MA, Kugler K, Freeman J, Jones RN. Prevalence of antimicrobial resistance among respiratory tract isolates of Streptococcus pneumoniae in North America: 1997 results from the SENTRY Antimicrobial Surveillance Program. Clin Infect Dis 1998; 27:764-770.

3. Moellering RC. Emerging resistance with gram-positive aerobic infections: where do we go from here? Clin Infect Dis 1998; 26:1177-1178.

4. Spika JS, Facklam RR, Plikaytis BD, Oxtoby MJ, Pneumococcal Surveillance Working Group. Antimicro-

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

bial resistance of Streptococcuspneumoniae in the United States, 1979-1987. J Infect Dis 1991; 163:1273-1278. Hall LMC. Application of molecular typing to the epidemiology of Streptococcus pneumoniae. J Clin Path01 1998; 51:270-274. Maslow JN, Mulligan ME, Arbeit RD. Molecular epidemiology: application of contemporary techniques to the typing of microorganisms. Clin Infect Dis 1993; 17:153-164. Matushek MG, Bonten MJM, Hayden MK. Rapid preparation of bacterial DNA for pulsed-field gel electrophoresis. J Clin Microbial 1996; 34:2598-2600. Marchese A, Ramirez M, Schito GC,Tomasz A. Molecular epidemiology of penicillin-resistant Streptococcus pneu- moniae isolates recovered in Italy from 1993 to 1996. J Clin Microbial 1998; 3612944-2949. Moissenet D, Valcin M, Marchand V, et al. Molecular epidemiology of Streptococcuspneumoniae with decreased susceptibility to penicillin in a Paris Children’s Hospital. J Clin Microbial 1997; 35:298-301. Song JH. Global spread of resistant pneumococcal clones: nowhere to hide. In: Program of the 2nd International Symposium on Antimicrobial Resistance. Seoul: Korean Society for Chemotherapy, 1999:35-51. Munoz R, Coffey TJ, Daniels M, et al. Intercontinental spread of a multiresistant clone of serotypes 23F Strepto- coccus pneumoniae. J Infect Dis 1991; 164:302-306. Plessis MD, Smith AM, Klugman KP Rapid detection of penicillin-resistant Streptococcus pneumoniae in cerebro- spinal fluid by a seminested-PCR strategy. J Clin Microbial 1998; 36:453-457. Smith AM, Klugman KP, Coffey TJ, Spratt BG. Genetic diversity of penicillin-binding protein 2B and 2X genes from Streptococcus pneumoniae in South Africa. Antimicrob Agents Chemother 1993; 37:1938-1944. Sibold C, Wang J, Henrichsen J, Hakenbeck R. Genetic relationships of penicillin-susceptible and -resistant Streptococcus pneumoniae strains isolated on different continents. Infect Immun 1992; 60:4119-4126. Rohani MY, Raudziah A, Ng AJ, et al. Epidemiology of Streptococcus pneumoniae infection in Malaysia. Epi- demiol Infect 1999; 122:77-82. Soh SWL, Poh CL, Lin RVTI? Serotype distribution and antimicrobial resistance of Streptococcus pneumoniae isolates from pediatric patients in Singapore. Antimicrob Agents Chemother 2000; 44:2193-2196. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing; Tenth Informational Supplement. Aerobic dilution. M7. Villanova, PA: NCCLS, 2000:MlOO-SlO. Thong KL, Mohd Nasir MD, Goh YL, Ngeow YF, Parasakthi N. Rapid pulsed-field gel electrophoresis method for streptococcus spp. Asia Pacific J Mol Biol Biotechnol2001; 9:143-145. Gillespie SH, McHugh TD, Hughes JE, Dickens A, Kyi MS, Kelsey M. An outbreak of penicillin resistant Strepto- coccus pneumoniae investigated by a polymerase chain reaction based genotyping method. J Clin Path01 1997; 50:847-851. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbial 1995; 33:2233-2239.

Penicillin susceptibility and molecular characteristics of clinical isolates of Streptococcus pneumoniae I Desa et al 197

21. Jernigan DB, Cetron MS, Breiman RF. Minimizing the impact of drug-resistant Streptococcus pneurnoniae (DRSP). JAMA 1996; 275:206-209.

22. Ross PW. Streptococcus pneumoniae. In: Collee JG, Foaser AG, Marmion BP, Simmons A, eds. Mackie and McCartney practical medical microbiology, 14th edn. New York: Churchill Livingstone, 1996:275-281.

23. Tomasz A. Antibiotic resistance in Streptococcus pneumoniae. Clin Infect Dis 1997; 24(suppl l):S85- SXS.

24. Kucers A, Crowe S, Grayson ML, Hoy J. The use of anti- biotics: a clinical review of antibacterial, antifungal and antiviral drugs, 5th edn. Oxford: Butterworth-Heinemann, 1997:5-g.

25. Song JH, Lee NY, Ichiyama S, et al. Spread of drug- resistant Streptococcus pneumoniae in Asian Countries: Asian Network for surveillance of resistant pathogens (ANSORP) study. Clin Infect Dis 1999; 28:1206- 1211.

26. Rohani MY, Parasakthi N, Raudzah A,Yasim MY. In-vitro susceptibilities of Streptococcus pneumoniae strains isolated in Malaysia to six antibiotics. J Antimicrob Chemother 1999; 44:852-853.

27. Hsueh PR, Teng LJ, Lee LN, Yang PC, Ho SW, Luh KT. Dissemination of high-level penicillin-, extended-spectrum cephalosporin-, erythromycin-resistant Streptococcus pneumoniae clones in Taiwan. J Clin Microbial 1999; 37~221-224.

28. Hermans PWM, Sluijter M, Elzenaar K, et al. Penicillin- resistant Streptococcus pneumoniae in the Netherlands: results of a l-year molecular epidemiologic survey. J Infect Dis 1997; 175:1413-1422.

29. Luna VA, Jernigan DB, Tice A, Kellner JD, Roberts MC. A novel multiresistant Streptococcus pneumoniae sero- group 19 clone from Washington State identified by pulsed-field gel electrophoresis and restriction fragment length patterns. J Clin Microbial 2000; 38:1575-1580.

30. Spratt BG. Resistance to antibiotics mediated by target alterations. Science 1994; 264:388-393.

31. Heffelfinger JD, Dowel1 SF, Jorgensen JH, et al. Management of community-acquired pneumonia in the era of pneumococcal resistance. Arch Intern Med 2000; 160:1399-1408.

32. Margaret IP, Lyon DJ, Yung RWH, Chan C, Cheng AFB. Evidence of clonal dissemination of multidrug-resistant Streptococcuspneumoniae in Hong Kong. J Clin Microbial 1999; 37:2834-2839.

33. Doern GV, Brueggemann AB, Blocker M, et al. Clonal relationship among high-level penicillin-resistant Strepto- coccus pneumoniae in the United States. Clin Infect Dis 1998; 27:757-761.

34. Ferroni A, Nguyen L, Gehanno P, Boucot I, Berche P Clonal distribution of penicillin-resistant Streptococcus pneumoniae 23F in France. J Clin Microbial 1996; 34:2707-2712.