Penicillin Susceptibility and Epidemiological Typing of Invasive Pneumococcal Isolates in the Republic of Ireland

Epidemiology and Molecular Biology Unit, The Children's University Hospital,¹ Perinatal Infections Laboratory, The Rotunda Hospital,² Department of Clinical Microbiology, Royal College of Surgeons in Ireland,³ Beaumont Hospital, Dublin, Republic of Ireland⁴

Received 17 October 2002/ Returned for modification 25 November 2002/ Accepted 2 May 2003

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion References

 
A national study was undertaken to investigate the incidence of invasive pneumococcal disease in the Republic of Ireland and to examine the associated isolates. In 1999, 144 S. pneumoniae isolates, all recovered from cases of invasive disease, were received from 12 microbiology laboratories. The incidence of invasive pneumococcal disease was estimated to be 6.6/100,000 population. All isolates were analyzed for serotype, penicillin susceptibility, chromosomal relatedness (by using pulsed-field gel electrophoresis ), and penicillin-binding protein (pbp) fingerprinting. Several findings of note were observed regarding the pneumococcal population in Ireland. First, isolates of 25 different serotypes were represented, with serotypes 14, 9V, 8, 5, 4, and 3 being the most common. This finding, together with the pbp fingerprinting and PFGE typing results, indicated the clonal spread of strains of these serotypes in Ireland. Second, 27 (18.7%) isolates had reduced susceptibility to penicillin, and 74% of these were serotype 9V. Of these, 80% appeared to belong to the same clone. This could suggest the spread of the international Spanish/French 9V penicillin-resistant clone into Ireland. Third, nine different pbp genotypes were identified, four of which were new. Two pbp genotypes accounted for the majority of isolates dividing them according to their penicillin susceptibility status but irrespective of serotype and PFGE type. This is strong evidence for the occurrence of horizontal transfer of pbp genes between strains, observed with both penicillin-susceptible and penicillin-nonsusceptible isolates. Fourth, there was evidence of serotype transformation since isolates, indistinguishable by pbp fingerprinting and PFGE typing, expressed different capsular types.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

 
Streptococcus pneumoniae remains an important human pathogen of children under 5 years of age and of the elderly, causing infections ranging in severity from otitis media and sinusitis to the more invasive conditions of pneumonia, septicemia, and meningitis. In developed countries this organism is one of the principal causes of bacterial meningitis and the most common cause of community-acquired pneumonia (5). S. pneumoniae accounted for 10.3 and 22.3% of all positive blood and cerebrospinal fluid (CSF) cultures, respectively, reported in England and Wales between 1993 and 1995 (12, 26). The rapid spread of penicillin resistance has increased the difficulty in treating pneumococcal infection, particularly in the central nervous system. There are several reports in the literature of the emergence and spread of several penicillin-resistant clones of S. pneumoniae in various countries worldwide (32, 46).

Molecular studies have shown that these penicillin-resistant pneumococcal populations are highly dynamic and that resistance is a combination of the spread of resistant clones, the acquisition and loss of resistance genes within those clonal lineages, and the spread of resistance genes to new lineages (32, 36). The persistent threat of invasive pneumococcal infections highlights the need for increased surveillance, including phenotyping, genotyping, and antimicrobial susceptibility testing. The value of serological and molecular typing techniques in the epidemiological analysis of pneumococcal isolates has recently been illustrated by reports in the literature (15, 16, 20). These studies have highlighted the usefulness of national data in determining the prevalent strains and their antimicrobial resistance patterns and are also important in planning preventative strategies including the use of new conjugated vaccines.

The participation of Ireland in The European Antimicrobial Resistance Surveillance Survey (EARSS) provided the ideal framework for the collection, antimicrobial testing, and documenting of invasive pneumococcal disease (IPD). Prior to this, there were no data available nationally on the prevalence of IPD. In the present study, hospital laboratories reporting cases of IPD were invited to submit isolates for further analysis. We investigated the epidemiology of pneumococcal disease in Ireland over the first year to determine the incidence of IPD and to examine the Irish population of S. pneumoniae isolates with regard to phenotype, genotype, and antimicrobial profile.

(This work was presented in part as a poster at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Ontario, Canada [abstr. 1814], 17 to 20 September 2000.)

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

 
S. pneumoniae isolates. In 1999 all pneumococcal isolates recovered from patients with invasive disease were included in the present study, i.e., isolates from blood or CSF. A single pneumococcal isolate per patient per infection was analyzed. All isolates were identified as S. pneumoniae according to optochin susceptibility. The S. pneumoniae reference strain recommended by EARSS, ATCC 49619, was also analyzed (29, 30). All isolates were stored in 1:3 glycerol (nutrient broth containing 0.5% glucose and 10% glycerol)-horse serum storage medium at -70°C.

Serotyping. All pneumococci were serotyped by slide agglutination with capsular typing sera (Statens Serum Institute, Copenhagen, Denmark) by using the methodology routinely used by the staff at the Streptococcal Reference Laboratory of the Central Public Health Laboratory, Colindale, London, United Kingdom.

Penicillin susceptibility. Isolates were assessed for their susceptibility to penicillin by the commercially available E-test method (AB Biodisk, Solna, Sweden) which was used and interpreted according to manufacturer's instructions. MIC breakpoints for defining susceptibility and resistance were in accordance with the EARSS guidelines. Pneumococcal isolates for which MICs of <0.1 µg/ml were defined as penicillin susceptible (PSP) and isolates for which MICs of 0.1 µg/ml were defined as penicillin nonsusceptible (PNSP) to include isolates with intermediate and high-level resistance to penicillin, i.e., for which the MICs were 2 µg/ml.

pbp fingerprinting. Total genomic DNA was prepared by using a modification of the Wizard Genomic DNA extraction kit for gram-positive organisms (Promega Corp., Madison, Wis.). Briefly, the gram-positive specific lytic enzyme step was omitted and cell lysis was performed as described for gram-negative organisms. The pbp2B and pbp2X genes were amplified from chromosomal DNA by PCR with the primers and conditions described by Dowson et al. (10). The primers and PCR conditions used to amplify the pbp1A gene were designed and optimized during the course of the present study. The pbp1A gene was amplified with the primers 5'-GTTATCGCAGCCATTGTCTTAGG-3' and 5'-GACTGTGAAGTTGAACTATCTGATG-3' using the following PCR cycling conditions: 94°C for 5 min, followed by 35 cycles of 94°C for 60 s, 55°C for 60 s, and 72°C for 130 s. The final cycle was run at 72°C for 7 min. In order to perform restriction fragment length polymorphism (RFLP) analysis of pbp genes, PCR products of pbp2B, pbp2X, and pbp1A were each digested separately with 10 U of HinfI (Promega) according to the manufacturer's instructions. The resulting fragments were separated on 3% (wt/vol) agarose gels containing 5 µg of ethidium bromide/ml at 3.5 V/cm for 2 h and visualized on an UV transilluminator.

PFGE fingerprinting. Agarose-embedded chromosomal DNA plugs for PFGE analysis were prepared and digested with SmaI as described by Hall et al. (16). Macrorestriction fragments were separated on a 1.2% SeaKem agarose gel (FMC BioProducts, Rockland, Maine) by electrophoresis on a CHEF DR III apparatus (Bio-Rad Laboratories, Richmond, Calif.) at 5.6 V/cm, with the switch time ramped from 1 to 15 s over a 20-h period, with an included angle of 120°, in 0.5x Tris-borate-EDTA (TBE) buffer at 14°C. Gels were stained for 20 min in 5 mg of ethidium bromide/ml and destained in 0.5x TBE for 1 h.

Computerized analysis of the pbp and PFGE fingerprints. All gel images were captured with a charge-coupled device camera to the AlphaImager computer program (Flowgen, Leicestershire, United Kingdom) and stored as tiff files (AlphaImager gel documentation system; Alpha Innotech Corp., San Leandro, Calif.). Hard copies of the gels were also printed onto thermal imaging paper (UPP-110S; Sony) by using a digital graphic printer (UP-D890; Sony) for manual evaluation of the patterns. Isolates were grouped by visual estimation into molecular types based on the presence of common bands. Essentially, the criteria defined by Tenover et al. (47) were used to analyze the PFGE patterns. Isolates with identical fingerprints and/or PFGE profiles were considered to be the same. Isolates that had similar PFGE profiles (i.e., a 3-band difference) were defined as being clonally related and subtypes of each other. Isolates that differed by more than three bands were considered different and thus represented a different strain. Gels were evaluated independently by two people. Grouping was determined without any previous knowledge of epidemiological data.

In addition to visual inspection and comparison, the pbp fingerprints were also analyzed by using the Phoretix 1D Advanced (v5.2) and 1D Database (v2.0) software packages (Phoretix International, Newcastle upon Tyne, United Kingdom) according to the methods and algorithms documented by the manufacturers. A 100-bp ladder (Gibco-BRL/Life Technologies, Paisley, Scotland) was used as a molecular weight standard on the pbp fingerprint gels. For pulsed-field gel profiles, the position of each band was interpreted as a binary code in to a spreadsheet (Microsoft Excel). This was done by comparing the position of each band in each profile to the position of each band within each of the other profiles and also in relation to the bands of the molecular weight standard included on each gel (within a grid of 34 possible band positions). This binary code reflected the presence or absence of a band in that position within that profile on the gel. This code was converted to a "gel" image by using the Scored Tiff (v1.0) software (Nonlinear Dynamics, Newcastle upon Tyne, United Kingdom) which was then read as an "ALF" gel by the Phoretix 1D Advanced (v5.01) software. A 50-kb Lambda ladder (Bio-Rad laboratories, Hercules, Calif.) was used as the molecular weight standard on all pulsed-field gels.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Clinical and demographic details. In 1999, a total of 144 invasive pneumococcal isolates from 144 individual patients were received from 12 microbiology laboratories. Since these laboratories serve ca. 60% of the total population of the Republic of Ireland, we estimate the incidence of IPD in Ireland in 1999 to be 6.6/100,000 population. The number of isolates contributed by each of the 12 participating centers varied from 1 to 28 isolates, with a mean of 12 isolates per center. The most common specimen source was blood (95.8%), with the remainder from CSF. There was a bimodal distribution of cases according to age; 9.7% of cases were in patients 1 year old, and 49.3% of cases were in patients >60 years old; 72.9% of isolates were recovered from individuals that were 40 years old. Of the 144 patients, 87 (60.4%) were male and 57 (39.6%) were female, yielding a male/female ratio of 1.5:1.

Epidemiological characteristics of the pneumococcal isolates. All 144 isolates were analyzed for their penicillin susceptibility pattern, serotype, pbp fingerprint, and PFGE profile. The results obtained by each method are summarized in Tables 1 and 2.

fig.ommitted TABLE 1. Serogroups and/or serotypes and penicillin susceptibilities of 144 S. pneumoniae isolates isolated from individuals with IPD in the Republic of Ireland in 1999

 

fig.ommitted TABLE 2. Distribution of pbp codes and serogroups or serotypes of S. pneumoniae isolates grouped by PFGE typea

 
(i) Penicillin susceptibility. The range of penicillin MICs observed with the 144 isolates tested was <0.016 to 1.0 µg/ml; 117 isolates (81.3%) were PSP (MIC < 0.016 to 0.064 µg/ml), and the remaining 27 (18.7%) were PNSP (MIC  0.1 µg/ml). Of the 27 PNSP isolates, the penicillin MICs for 19 were 0.1 to 0.94 µg/ml, and the penicillin MICs for 8 were 1.0 µg/ml. No strains had penicillin MICs of 2 µg/ml, and so none were considered resistant to penicillin.

(ii) Serotyping. The 144 IPD-associated isolates comprised 25 different serogroups or serotypes overall (Table 1). All isolates were serotypeable based on the available antisera. The ATCC 49619 strain was serotype 19F. The most prevalent serotypes were serotypes 14 and 9V, accounting for 30 (20.8%) and 22 (15.3%) of all isolates examined, respectively. The other most frequent serotypes were 4 (9%), 3 (6.9%), 12B (4.9%), 8 (5.6%), and 5 (4.9%). These predominant serotypes were reflected among the 117 PSP isolates since 26 (22.2%) were serotype 14. The capsular types of the 27 PNSP isolates were restricted to only four serotypes, in order of frequency 9V (74%), 14 (14.8%), 23F (7.4%), and 6B (3.7%).

(iii) pbp fingerprinting. Of the 144 isolates, three to nine distinct patterns were obtained for each of the pbp1A, pbp2B, and pbp2X genes (Fig. 1). For comparative pbp fingerprinting purposes, DNA from a strain representative of each of RFLP fingerprint obtained was sent to M. Sluijter, Erasmus University, Rotterdam, The Netherlands. Therefore, in the present study the number assignments used to differentiate pbp gene fingerprints are arbitrary but correspond to the numbers and RFLP fingerprints of the pbp fingerprinting system documented previously (8, 19, 37, 38, 43). Three distinct patterns were observed with the pbp1A genes; these were designated A1, A2, and A4. RFLP analysis of the pbp2B gene yielded five distinct fingerprints: B1, B2, B3, B7, and B12. Two pbp1A and pbp2B patterns—A2-B2 and A1-B1—accounted for 116 (80.5%) and 26 (18.1%) of all isolates examined. Nine RFLP fingerprints of pbp2X were identified: X1, X2, X3, X5, X71, X98, X99, X100, and X101. As with the pbp1A and pbp2B genes, two pbp2X patterns—X1 and X3—accounted for the majority of the isolates: 26 (18.1%) and 97 (68%), respectively. Of the nine RFLP patterns observed for the pbp2X, four were not present in the database maintained by M. Sluijter. These novel patterns were designated X98 to X101 and were added to that database.

fig.ommitted FIG. 1. DNA fingerprint patterns of the pbp1A (n = 3), pbp2B (n = 5), and pbp2X (n = 10) genes represented among the 144 invasive pneumococcal isolates. Lane numbers indicate the pbp codes. A4, B7, and X10 are the patterns obtained with the ATCC 49619 strain for the pbp1A, pbp2B, and pbp2X genes, respectively. Lane M denotes the 100-bp DNA ladder size markers electrophoresed on each gel, the sizes of which are indicated.

 
The pbp genotype for each isolate was defined by combining the RFLP patterns for each pbp gene, yielding a three-number code for each isolate (Table 2). Among the 144 isolates, two genotypes contained 85.4% of the isolates, dividing them almost according to their penicillin susceptibilities. Of the remaining seven pbp genotype groups, four contained only one isolate each, one contained three isolates, and the other two contained seven isolates each. The most common pbp genotype, A2-B2-X3, contained 97 isolates, was solely comprised of PSP isolates, and contained 83% of all PSP isolates. The pbp genotype A1-B1-X1 contained 26 isolates (all PNSP). All but one (96.3%) of the PNSP isolates had this genotype. The other PNSP isolate (serotype 6B) had a A1-B3-X5 genotype.

No particular association between serotype and pbp genotype A2-B2-X100 was noted since all isolates had different serotypes, i.e., 9V, 19A, 19F, and 29, although four of these were of serotype 19A. A similar situation was observed with isolates with pbp genotype A2-B2-X2, since all three isolates had different serotypes (6A, 7F, and 23F). However, all seven isolates with the pbp genotype A2-B2-X99 were serotype 5.

(iv) PFGE fingerprinting. Using the analysis criteria of a 3-band difference between profiles (47), i.e., with PFGE profiles that were >77% similar and considered to be closely related and probably part of the same outbreak, the 144 isolates yielded 34 individual PFGE types (Fig. 2). Among the 144 isolates studied, 24 PFGE types, each representing two or more isolates were observed (Table 2). A further 10 PFGE types differing by at least 23% from others were also observed, each of which were represented by a single isolate only.

fig.ommitted   FIG. 2. Genetic relatedness of 144 isolates of S. pneumoniae and the ATCC 49619 strain (indicated by an asterisk) based on PFGE fingerprinting. Codes I to XXIV refer to clusters of pneumococcal isolates displaying PFGE types with homologies greater than 77%. The number of isolates within each cluster is also indicated.

 
No one PFGE type correlated with the penicillin-resistant or -susceptible status of the isolates examined (Fig. 2 and Table 2) since the 117 PSP isolates belonged to 29 PFGE types and the 27 PNSP isolates belonged to seven PFGE types. All of the serotype 9V and 14 PNSP isolates, as well as the PNSP serotype 6B isolate, were contained within a single cluster at a level of >66% identity (PFGE types XIX to XXI) and was at a distance of >34% from any of the other isolate clusters. Only two PNSP isolates, both of serotype 23F, were not contained within this cluster.

In general, there was some correlation between pattern and serotype, since some isolates of the same serotype usually yielded indistinguishable PFGE profiles and at the very least belonged to no more than three groups, although some exceptions were observed (Fig. 2 and Table 2). Isolates belonging to certain serotypes were found to be more genotypically related as follows (serotype, percentage of isolates of the same PFGE group): 1, 100%; 3, 70%; 4, 61%; 5, 100%; 6B, 75%; 7F, 100%; 8, 100%; 9V, 73%; 12B, 57%; 14, 70%; 18C, 100%; 19A, 83%; 19F, 75%; 20, 100%; 22F, 75%; 23F, 40%; and 31, 100%. Isolates of serotypes 5, 18C, and 7F formed unique and exclusive groups. Isolates of serotypes 1, 8, 20, and 31 also formed single groups but along with isolates of other serotypes. In contrast, there also were instances in which isolates of different serotypes belonged to the same group and cases in which isolates with different serotypes had indistinguishable PFGE profiles (Table 2). No particular association could be inferred between PFGE profile and pbp genotype, given the predominance of the pbp genotypes A2-B2-X3 among the PSP isolates and A1-B1-X1 among the PNSP isolates.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

 
The incidence of IPD in the Republic of Ireland estimated by the present study (6.6/100,000 population) is lower but comparable to the incidence for pneumococcal bacteremia and meningitis reported for England and Wales during the 10-year period ending in 1992 (8.7/100,000 population) (2). Also, it is lower that that for Scotland (9.8/100,000 population) based on the number of cases identified in 1993 and 1994 (33). In addition, these results confirm the age and sex distribution of IPD described in previous studies, demonstrating a higher incidence in individuals at the extremes of age and in males (33). Although, the reason for this trend is unclear, the predisposition to serious infection in males has also been noted in association with a wide range of bacteria, including S. pneumoniae, Staphylococcus aureus, Escherichia coli, and Klebsiella and Enterobacter spp. (2).

Worldwide data on pneumococcal serotype distribution have shown that most pneumococcal disease is caused by 23 serotypes in adults (17), and in children 80% of pneumococcal disease is caused by 7 serotypes (7). In the present study, 25 different serotypes were represented among the 144 isolates analyzed. Isolates of serotypes 14 and 9V were the most common, with isolates of serotypes 3 and 4 also being commonly observed. These four serotypes accounted for 52.1% of all invasive pneumococcal isolates recovered in 1999 in the Republic of Ireland. These serotypes, among others, have been the most prevalent in studies of IPD in other countries (28, 39, 41), including England and Wales (34). However, in some countries serotypes 1, 6, 7, 8, 19, and 23, along with serotype 14, are the most prevalent (1, 5, 12, 22, 43, 46).

It has been documented that certain pneumococcal serotypes are known to be more virulent than others; bacteremia caused by serotypes 3, 6A, and 19A is considered to have the highest mortality rates compared to that caused by other serotypes (21, 33). In the present study, 6.9% of infections were associated with serotype 3 isolates, and 11.8% of all cases were associated with these three serotypes.

The prevalence and distribution of serotypes among a population of pneumococci is an important consideration, especially in the context of vaccine development and estimating vaccine efficacy. When all of the isolates are taken into consideration, both PSP and PNSP, 16 (11.1%) expressed one of eight serotypes that currently is not included in the current formulation of the 23-valent nonconjugated polysaccharide vaccine. These 16 isolates included 7 isolates of serotype 12B; single isolates of serotypes 6A, 15A, 16F, 29, and 34; and two isolates of serotypes 23A and 31. However, of these eight serotypes, protection would be afforded by cross-reactivity for four serotypes with similar, closely related serotypes that are present in the 23-valent vaccine. Although all of the serotypes recovered from infants 1 year old or younger are included in the 23-valent vaccine, this vaccine does not induce a protective immune response in this age group. However, the new 7-valent conjugate vaccine is effective in this population, but only 71.5% of the serotypes found in this age group are contained within this vaccine, findings similar to the rates observed in other European countries (5). Overall, only 56.3% of the 144 isolates expressed serotypes that are contained within the 7-valent vaccine. These included 10 of the 14 (71.4%) isolates that were recovered from infants who were 1 year old or younger. The remaining four isolates were serotypes 3 and 8, and 2 isolates were serotype 7F. However, it is possible that the vaccine may confer cross-protection against a further nine isolates, increasing the overall protection to 62.5% of the isolates. All of the PNSP isolates recovered as part of the present study had serotypes that are covered by the 7-valent vaccine.

Several examples of the international spread of PNSP clones have been reported (18, 46). Among the Irish isolates, 18.7% (27 of 144) had a reduced susceptibility to penicillin (PNSP). None of the isolates had penicillin MICs that were indicative of penicillin resistance. The proportion of PNSP isolates observed in the present study is comparable to that of other European countries, e.g., Portugal, Belguim, and Italy, but is considerably lower than that of Spain (34% [4]), Greece (30.4% [28]), and Turkey (54% [41]) and of some non-European countries, including Taiwan (56.4% [6]), Hong Kong (55% [23, 24]), and the United States (44% [9]).

Specific serotypes, predominantly serotypes 6B, 9V, 14, 19A, 19F, and 23F, have been associated with antibiotic resistance (31, 46). Of the 27 Irish PNSP four serotypes were observed: 9V, 14, 23F, and 6B. Serotype 9V accounted for 20 (74.1%) of the isolates, and serotype 14 accounted for another 4 (14.8%). Only two serotype 23F isolates and a single 6B isolate were observed. These are the same four serotypes as the extensively spread "penicillin-resistant" clones that have emerged in Spain. The "Spanish/French clone" usually classified as serogroup 9 or 14 has spread to several countries in Europe, the Far East, and North and South America (13, 35, 42). The "Spanish/Iceland clone," usually having serotype 6B, has become the dominant pneumococcus type isolated from children in Iceland, where it accounts for 75% of multiresistant pneumococci. Therefore, from a serological analysis, it would appear that the PNSP isolates in Ireland might be members of the already-documented multiresistant bacterial clones that have spread extensively worldwide.

PCR-RFLP analysis of the pbp1A and pbp2B genes yielded three and five distinct fingerprint patterns, respectively. Each of these patterns had been observed and documented previously in an international data library comprising of at least 30 RFLP patterns for each gene (18, 19, 36, 37, 43). Ten different RFLP patterns were observed with the pbp2X gene. Six of these had been observed previously, but the remaining four were novel and are not contained with the database that already contains 97 pbp2X RFLP patterns. This demonstrates the highly variable nature of the pbp2X gene, suggesting that it undergoes more recombination or that point mutations occur more frequently than in either of the other pbp genes (14). Variations in the RFLP patterns of the pbp2X gene only were responsible for six of the seven distinct genotypes observed among the PSP isolates. Genotype A2-B2-X3 was found most frequently among PSP isolates. This is the genotype that has been most commonly observed with PSP isolates in both the United States and The Netherlands (37, 43). There were no specific associations between pbp genotype and serotype except for serotype 5 isolates and the A2-B2-X99 genotype; only serotype 5 isolates had this genotype.

The predominant pbp genotype among the PNSP was A1-B1-X1. Isolates with three distinct serotypes—9V, 14, and 23F—displayed this genotype. As in other reports, this genotype was uniquely associated with the PNSP phenotype, and it is predominantly observed in strains of the pandemic clones 23F and 9V/14 (20, 36, 43). Only one other pbp genotype was observed within the PNSP, A1-B3-X5, a serotype 6B isolate. Previous studies have documented that genetic polymorphism, as observed by pbp fingerprinting, is restricted within isolates of the pandemic 6B-type penicillin resistant clone and that pbp2X type 13 was predominantly and almost exclusively present among members of this clone (18, 36).

It has been documented that, in contrast to the pbp genes of PNSP isolates, those of PSP isolates show very little sequence variation (10, 25, 48). This is not evident from the present study since all but one of the PNSP isolates had the same pbp genotype, whereas seven different genotypes were observed with the PSP isolates, albeit only varying by the pattern of the pbp2X gene. Four of these patterns were not observed previously (M. Sluijter, unpublished data). Studies have shown that alterations in pbp2X gene result in low-level penicillin resistance, whereas high-level penicillin resistance requires alterations in the pbp2B and pbp1A genes (3, 44, 45). Of the seven different RFLP patterns observed with the pbp2X gene (X3, X2, X71, X98, X99, X100, and X101) among the PSP isolates in the present study, the pbp1A and pbp2B gene patterns were A2-B2 for six of these (all but X98).

Several studies have already documented the use of PFGE analysis for studying clonal relationships, but comparisons of results between studies is difficult due to the different criteria used for the interpretation of banding patterns (11, 24, 27, 40, 49). In the present study, we used the criterion described by Tenover et al. (47). However, the criteria devised by Tenover et al. were to assist in the interpretation of types from an outbreak and not intended primarily to study large epidemiologically unrelated collections of isolates such as in the present study, where the isolates were distributed throughout one country with a population of 3.7 million people. Consequently, our conclusions based on these criteria must be considered tentative and may open to question. PFGE types did not appear to have a strict correlation with penicillin susceptibility and pbp genotype. However, the majority of PNSP isolates were clustered within a small number of PFGE types (i.e., 7 types), whereas PSP isolates were contained within many PFGE types (i.e., 29 types). This finding indicates that there is greater genetic diversity among the PSP isolates than among the PNSP isolates or that there are fewer PNSP than PSP clones associated with IPD in Ireland. Only two PFGE types contained both PSP and PNSP isolates.

From the data generated in the present study it is evident that the use of different typing methods for the characterization of pneumococci associated with IPD can yield several different epidemiological profiles. pbp fingerprinting relies on the alterations in the sequence of the genes encoding the binding protein targets of penicillin and, as such, can be related to the penicillin susceptibilities of the organisms examined. PFGE typing is dependent on the genomic distribution of restriction endonuclease cleavage sites, whereas serotyping requires that all of the genetic, enzymatic, and transport components necessary to produce an exopolysaccharide are expressed and functional. Several studies have shown PFGE fingerprinting to be highly informative in various epidemiological studies of pneumococci and of greater epidemiological value than either pbp fingerprinting or serotyping. Likewise, in the present study PFGE typing provided valuable information in relation to the clonal nature of pneumococci in Ireland and formed a basis on which the data from the other typing methods used could be compared and examined.

In conclusion, in the present study serotyping, pbp fingerprinting, and PFGE data suggest that the spread of pneumococci with reduced penicillin susceptibility in Ireland is probably due to both the horizontal spread of altered pbp genes and the clonal spread of certain strains of pneumococci already prevalent throughout Europe.

　

	ACKNOWLEDGMENTS

 
We are grateful to Peadar Clarke, Royal College of Surgeons in Ireland, for considerable assistance in technical aspects of this work. We thank all of the staff of the microbiology laboratories of The Adelaide and Meath Hospital incorporating the National Children's Hospital, Tallaght, Dublin, Ireland; The Bons Secours, Cork, Ireland; Beaumont Hospital, Dublin, Ireland; The Children's University Hospital, Dublin, Ireland; The General Hospital, Mayo, Ireland; The General Hospital, Mullingar, Ireland; The General Hospital, Sligo, Ireland; Mater Misercordiae Hospital, Dublin, Ireland; The Regional Hospital, Waterford, Ireland; St. James's Hospital, Dublin, Ireland; St. Vincent's Hospital, Dublin, Ireland; and University College Hospital, Galway, Ireland, for providing us with pneumococcal isolates and demographic details. We also thank the staff of National Disease Surveillance Center for their assistance. We thank D. J. Platt, Royal Infirmary, Glasgow, Scotland, for help with the computer-assisted analysis of the PFGE profiles; M. Sluijter, University Hospital Rotterdam, Rotterdam, The Netherlands, for assistance with the pbp fingerprinting; R. George and all of the staff of the PHLS, Colindale, London, United Kingdom, for help with serotyping.

This work was partly funded by grants (RP109/99) from the Health Research Board, Republic of Ireland and by a grant from the Charitable Infirmary Charitable Trust of the Republic of Ireland. This work was also partly supported by John Wyeth & Brothers Ltd., Wyeth Laboratories, United Kingdom, and we acknowledge the support of Pfizer (Ireland), Ltd., to the Department of Clinical Microbiology, The Royal College of Surgeons in Ireland.

	REFERENCES

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Alpern, E., E. Alessandrini, K. McGowan, L. Bell, and K. Shaw. 2001. Serotype prevalence of occult pneumococcal bacteremia. Pediatrics 108:E23.

2. Aszkenasy, O., R. George, and N. Begg. 1995. Pneumococcal bacteraemia and meningitis in England and Wales 1982 to 1992. Commun. Dis. Rep. CDR Rev. 5:R45-R50.

3. Barcus, V., K. Ghanekar, M. Yeo, T. Coffey, and C. Dowson. 1995. Genetics of high-level penicillin resistance in clinical isolates of Streptococcus pneumoniae. FEMS Microbiol. Lett. 126:299-304.

4. Bronzwaer, O. S., Cars, U. Buchholz, S. Molstad, W. Goettsch, J. I. Veldhuijzen, Kool, M. Sprenger, J. Degener, et al. 2002. A European study on the relationship between antimicrobial use and antimicrobial resistance. Emerg. Infect. Dis. 8:278-282.

5. Cartwright, K. 2002. Pneumococcal disease in western Europe: burden of disease, antibiotic resistance and management. Eur. J. Pediatr. 161:188-195.

6. Chiou, C., and M. McEllistrem. 2001. Novel penicillin-, cephalosporin-, and macrolide-resistant clones of Streptococcus pneumoniae serotypes 23F and 19F in Taiwan which differ from international epidemic clones. J. Clin. Microbiol. 39:1144-1147.

7. Colman, G., E. Cooke, B. Cookson, P. Cooper, and A. Efstratiou. 1998. Pneumococci causing invasive disease in Britian 1982-1990. J. Med. Microbiol. 47:17-27.

8. Dejsirilert, S., K. Overweg, M. Sluijter, L. Saengsuk, M. Gratten, T. Ezaki, and P. Hermans. 1999. Nasopharyngeal carriage of penicillin-resistant Streptococcus pneumoniae among children with acute respiratory tract infections in Thailand: a molecular epidemiological survey. J. Clin. Microbiol. 37:1832-1838.

9. Doern, G., A. Bruggemann, M. Blocker, M. Dunne, H. Holley, K. Kehl, J. Duval, K. Kugler, S. Putnam, A. Rauch, and M. Pfaller. 1998. Clonal relationships among high-level penicillin-resistant Streptococcus pneumoniae in the United States. Clin. Infect. Dis. 27:757-761.

10. Dowson, C., A. Hutchison, and B. Spratt. 1989. Extensive re-modeling of the transpeptidase domain of penicillin binding protein 2B of a penicillin resistant South African isolate of Streptococcus pneumoniae. Mol. Microbiol. 3:95-102.

11. Dunne, W., K. Kehl, C. Holland-Staley, A. Brueggemann, M. Pfaller, and G. Doern. 2001. Comparison of results generated by serotyping, pulsed-field restriction analysis, ribotyping, and repetitive-sequence PCR used to characterize penicillin-resistant pneumococci from the United States. J. Clin. Microbiol. 39:1791-1795.

12. George, R., A. Johnson, D. Speller, A. Efstratiou, K. Broughton, and B. Patel. 1997. Serogroups/types and antibiotic resistance of referred isolates of Streptococcus pneumoniae: 1993 to 1995. Commun. Dis. Rep. 7:R159-R164.

13. Gherardi, G., J. Inostrozo, M. O'Ryan, V. Prado, S. Prieto, C. Arellano, R. Facklam, and B. Beall. 1999. Genotypic survey of recent ß-lactam-resistant pneumococcal nasopharyngeal isolates from asymptomatic children in Chile. J. Clin. Microbiol. 37:3725-3730.

14. Hakenbeck, R., K. Kaminski, A. Konig, J. M. Van Der Linden, Paik, P. Reichman, and D. Zahner. 2000. Penicillin-binding proteins in ß-lactam-resistant Streptococcus pneumoniae, p. 433-441. In A. Tomasz (ed.), Streptococcus pneumoniae molecular biology and mechanisms of disease. Mary Ann Liebert, Inc., Larchmont, N.Y.

15. Hall, L. M. 1998. Application of molecular typing to the epidemiology of Streptococcus pneumoniae. J. Clin. Pathol. 51:270-274.

16. Hall, L. M., R. A. Whiley, B. Duke, R. C. George, and A. Efstratiou. 1996. Genetic relatedness within and between serotypes of Streptococcus pneumoniae from the UK: analysis of multilocus enzyme electrophoresis, pulsed field electrophoresis, and antimicrobial resistance patterns. J. Clin. Microbiol. 34:853-859.

17. Hausdorff, W., J. Bryant, P. Paradiso, and G. Siber. 2000. Which pneumococcal serogroups cause the most invasive disease: implications for conjugate vaccine formulation and use: part I. Clin. Infect. Dis. 30:100-121.

18. Hermans, P., K. Overweg, M. Sluijter, and R. De Groot. 2000. Penicillin Streptococcus pneumoniae: an international molecular epidemiology study, p. 457-466. In A. Tomasz (ed.), Streptococcus pneumoniae molecular biology and mechanisms of disease. Mary Ann Liebert, Inc., Larchmont, N.Y.

19. Hermans, P., M. Sluijter, K. Elzenaar, A. Van Veen, J. Schonkeren, F. Nooren, W. van Leeuwen, A. De Neeling, B. Van Klingeren, H. Verbrugh, and R. De Groot. 1997. Penicillin-resistant Streptococcus pneumoniae in The Netherlands: results of a 1-year molecular epidemiologic survey. J. Infect. Dis. 175:1413-1422.

20. Hermans, P., M. Sluijter, T. Hoogenboezem, H. Heersma, A. Van Belkum, and R. De Groot. 1995. Comparative study of five different DNA fingerprint techniques for molecular typing of Streptococcus pneumoniae strains. J. Clin. Microbiol. 33:1606-1612.

21. Hsueh, P.-R., L.-J. Teng, L.-N. Lee, P.-C. Yang, S.-W. Ho, and K.-T. Luh. 1999. Dissemination of high-level penicillin-, extended-spectrum, cephalosporin-, and erythromycin-resistant Streptococcus pneumoniae clones in Taiwan. J. Clin. Microbiol. 37:221-224.

22. Hudspeth, M., T. Smith, C. Barrozo, A. Hawksworth, M. Ryan, and G. Gray. 2001. National Department of Defense Surveillance for invasive Streptococcus pneumoniae: antibiotic resistance, serotype distribution, and arbitrarily primed polymerase chain reaction analyses. J. Infect. Dis. 184:591-596.

23. Ip, M., D. Lyon, R. Yung, C. Chan, and A. Cheng. 1999. Evidence of clonal dissemination of multidrug-resistant Streptococcus pneumoniae in Hong Kong. J. Clin. Microbiol. 37:2834-2839.

24. Ip, M., D. Lyon, R. Yung, L. Tsang, and A. Cheng. 2002. Introduction of new clones of penicillin-nonsusceptible Streptococcus pneumoniae in Hong Kong. J. Clin. Microbiol. 40:1522-1525.

25. Laible, G., B. Spratt, and R. Hakenbeck. 1991. Interspecies recombinational events during the evolution of altered PBP 2X genes in penicillin-resistant clinical isolates of Streptococcus pneumoniae. Mol. Microbiol. 5:1993-2002.

26. Laurichesse, H., O. Grimaud, P. Waight, A. Johnson, R. George, and E. Miller. 1998. Pneumococcal bacteraemia and meningitis in England and Wales, 1993 to 1995. Commun. Dis. Public Health 1:22-27.

27. Louie, M., G. Louie, G. Papia, J. Talbot, M. Lovgren, and A. Simor. 1999. Molecular analysis of the genetic variation among penicillin-susceptible and penicillin-resistant Streptococcus pneumoniae serotypes in Canada. J. Infect. Dis. 179:892-900.

28. Maraki, S., A. Christidou, and Y. Tselentis. 2001. Antimicrobial resistance and serotype distribution of Streptococcus pneumoniae isolates from Crete, Greece. Int. J. Antimicrob. Agents 17:465-469.

29. Marchese, A., S. Mannelli, E. Tonoli, F. Gorlero, M. Toni, and G. Schito. 2001. Prevalence of antimicrobial resistance in Streptococcus pneumoniae circulating in Italy: results of the Italian epidemiology observatory survey (1997-1999). Microb. Drug Resist. 7:277-287.

30. Marchese, A., M. Ramirez, G. Schito, and A. Tomasz. 1998. Molecular epidemiology of penicillin-resistant Streptococcus pneumoniae isolates recovered in Italy from 1993-1996. J. Clin. Microbiol. 36:2944-2949.

31. McGee, L., K. Klugman, and A. Tomasz. 2000. Serotypes and clones of antibiotic-resistant pneumococci, p. 375-379. In A. Tomasz (ed.), Streptococcus pneumoniae molecular biology and mechanisms of disease. Mary Ann Liebert, Inc., Larchmont, N.Y.

32. McGee, L., L. McDougal, J. Zhou, B. Spratt, F. Tenover, R. George, R. Hakenbeck, W. Hryniewicz, J. LeFevre, A. Tomasz, and K. Klugman. 2001. Nomenclature of major antimicrobial-resistant clones of Streptococcus pneumoniae defined by the pneumococcal molecular epidemiology network. J. Clin. Microbiol. 39:2565-2571.

33. McKenzie, H., N. Reid, and R. Dijkhuizen. 2000. Clinical and microbiological epidemiology of Streptococcus pneumoniae bacteraemia. J. Med. Microbiol. 49:361-366.

34. Miller, E., P. Waight, A. Efstratiou, M. Brisson, A. Johnson, and R. George. 2000. Epidemiology of invasive and other pneumococcal disease in children in England and Wales 1996-1998. Acta Paediatr. Suppl. 435:11-16.

35. Moissenet, D., M. Valcin, V. Marchand, E.-N. Garabedian, P. Geslin, A. Garbarg-Chenon, and H. Vu-Thein. 1997. Molecular epidemiology of Streptococcus pneumoniae with decreased susceptibility to penicillin in a Paris children's hospital. J. Clin. Microbiol. 35:298-301.

36. Overweg, K., D. Bogaert, M. Sluijter, R. De Groot, and P. Hermans. 2001. Molecular characteristics of penicillin binding protein genes of penicillin-nonsusceptible Streptococcus pneumoniae isolated in The Netherlands. Microb. Drug Resist. 7:323-334.

37. Overweg, K., D. Bogaert, M. Sluijter, J. Yother, J. Dankert, R. De Groot, and P. Hermans. 2000. Genetic relatedness within serotypes of penicillin-susceptible Streptococcus pneumoniae isolates. J. Clin. Microbiol. 38:4548-4553.

38. Overweg, K., P. Hermans, K. Trzcinski, M. Sluijter, R. De Groot, and W. Hryniewicz. 1999. Multidrug-resistant Streptococcus pneumoniae in Poland: identification of emerging clones. J. Clin. Microbiol. 37:1739-1745.

39. Reis, J., S. Cordeiro, S. Coppola, K. Salgado, M. Carvalho, L. Teixeira, T. Thompson, R. Facklam, M. Reis, and A. Ko. 2002. Population-based survey of antimicrobial susceptibility and serotype distribution of Streptococcus pneumoniae from meningitis patients in Salvador, Brazil. J. Clin. Microbiol. 40:275-277.

40. Richter, S., K. Heilman, S. Coffman, H. Huynh, A. Brueggemann, M. Pfaller, and G. Doern. 2002. The molecular epidemiology of penicillin-resistant Streptococcus pneumoniae in the United States, 1994-2000. Clin. Infect. Dis. 34:330-339.

41. Sener, B., S. Arikan, M. Ergin, and A. Gunalp. 1998. Rate of carriage, serotype distribution, and penicillin resistance of Streptococcus pneumoniae in healthy children. Zentbl. Bakteriol. 288:421-428.

42. Sibold, C., J. Wang, J. Henrichsen, and R. Hakenbeck. 1992. Genetic relationships of penicillin-susceptible and -resistant Streptococcus pneumoniae strains isolated on different continents. Infect. Immun. 60:4119-4126.

43. Sluijter, M., H. Faden, R. De Groot, N. Lemmens, W. Goessens, A. van Belkum, and P. Hermans. 1998. Molecular characterization of pneumococcal nasopharynx isolates collected from children during their first years of life. J. Clin. Microbiol. 36:2248-2253.

44. Smith, A., K. Klugman, T. Coffey, and B. Spratt. 1993. Genetic diversity of penicillin-binding protein 2B and 2X genes from Streptococcus pneumoniae in South Africa. Antimicrob. Agents Chemother. 37:1938-1944.

45. Smith, A., and K. Klugman. 1998. Alterations in PBP 1A essential for high-level penicillin resistance in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 42:1329-1333.

46. Stefani, S., and A. Agodi. 2000. Molecular epidemiology of antibiotic resistance. Int. J. Antimicrob. Agents 13:143-153.

47. Tenover, F., R. Arbeit, R. Goering, P. Mickelsen, B. Murray, D. Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33:2233-2239.

48. Whiting, G. 2001. Diagnosis of penicillin resistance by PCR-RFLP. Methods Mol. Med. 48:77-84.

49. Yoshida, R., Y. Hirakata, M. Kaku, K. Tomono, S. Maesaki, Y. Yamada, S. Kamihira, M. Jacobs, P. Appelbaum, and S. Kohno. 1999. Genetic analysis of serotype 23F Streptococcus pneumoniae isolates from several countries by penicillin-binding protein gene fingerprinting and pulsed-field gel electrophoresis. Chemotherapy 45:158-165.