Department of Biological Sciences, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK1
Public Health Laboratory, Royal Sussex County Hospital, Brighton BN2 5BE, UK2
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK3
Author for correspondence: Christopher G. Dowson. Tel: +44 1203 523534. Fax: +44 1203 523568. e-mail: c.g.dowson{at}warwick.ac.uk
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Keywords: molecular typing, carried clones, virulent clones, serotype stability, HIV
Abbreviations: ET, electrophoretic type; IA, index of association; MLEE, multilocus enzyme electrophoresis; MLRT, multilocus restriction typing; MLST, multilocus sequence typing; REP-PCR, repetitive element PCR; RT, restriction type; ST, sequence type; UPGMA, unweighted pair group method with arithmetic means
The GenBank accession numbers for the sequences of the trpA/B alleles determined in this study are AF157817 to AF157826.
a Present address: Laboratoire dEcologie, UMR 7625 CNRS Université Pierre et Marie Curie, 7 quai St Bernard, F-75252 Paris Cedex 05, France.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
To date, population studies of S. pneumoniae have focused largely on disease-associated invasive isolates, with the majority being isolates that have recently acquired resistance to a range of different antibiotics (Sibold et al., 1992 ; Lomholt; 1995
, Hall et al., 1996
; Enright & Spratt, 1998
). However, asymptomatic carriage of pneumococci in the throat or nasopharynx is widespread, with carriage rates being especially high in children, where up to 60% of individuals may be colonized (Austrian, 1986
; Speller et al., 1996
; Sung et al., 1995
; Appelbaum et al., 1996
), and there is clear evidence of spread amongst families (Hendley et al., 1975
). Colonization by multiple pneumococcal capsular types has also been reported (Austrian, 1986
). Some serotypes are particularly associated with disease in children or adults (Scott et al., 1996
) and others with carriage (Hoeprich, 1994
). However, it is only just becoming apparent that amongst invasive isolates there are important virulent pneumococcal clones that are responsible for many cases of disease around the world (Enright & Spratt, 1998
). Little is known about carried isolates; for example, is capsular exchange more likely amongst a population of carried isolates than invasive isolates (Weiser et al., 1994
), and are important invasive clones also frequently carried asymptomatically, or are there many widely carried clones that are infrequently associated with invasive disease?
A diverse range of molecular typing methods has been employed to identify isolates or outbreaks of many important pathogens, although historically multilocus enzyme electrophoresis (MLEE) has been the method of choice to resolve bacterial population structures (Souza et al., 1992 ; Maynard Smith et al., 1993
; ORourke & Stevens, 1993
; Musser, 1996
; Guttman, 1997
). However, because relatively few alleles are identified by MLEE, many loci (typically 20) are required to attain the power of resolution required to identify individuals within a population. During the course of this study a very powerful, portable technique, multilocus sequence typing (MLST) (Maiden et al., 1998
; Enright & Spratt, 1998
) has appeared that offers a direct replacement for MLEE but requires a high sequencing throughput. Here we examine the population structure of a sample of 134 throat or nasopharyngeal penicillin-susceptible isolates from three distinct localities (Oxford, UK, Manchester, UK and Nairobi, Kenya) using multilocus restriction typing (MLRT). Allelic variation at each of nine loci examined was determined by restriction analysis of PCR products derived from the open reading frames of each locus. In addition, since previous work with predominantly penicillin-resistant invasive isolates has shown an epidemic population structure, we also included 20 susceptible and 33 resistant strains isolated from blood, cerebrospinal fluid or the middle ear, to validate the methodology employed, to compare the distribution of alleles among the loci examined for carried and invasive isolates, and to identify frequently carried isolates that may also be associated with invasive disease. To further understand the relationship between carried and invasive isolates, MLST was also used to compare frequently carried clones from this study with previously reported sequence types (STs) from invasive disease.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Multilocus restriction typing (MLRT).
DNA was extracted from confluent overnight plate cultures of S. pneumoniae, originating from a single colony, using the Promega Wizard Genomic DNA Purification Kit according to the manufacturers instructions.
Internal fragments of the following genes were obtained by PCR: recP (1568 bp), recA (1042 bp), polI (2032 bp), hexA (1583 bp), hexB (1809 bp), ddl (669 bp), comA (1618 bp), trpA/B (1162 bp) and ciaR/H (1780 bp), using primers recP up 5'-ACCGCGACCGCTTTATTCTTTC-3', recP down 5'-ATGCTGACTACGCGGGATTTTTC-3'; recA up 5'-AGGCCTTGAATGACGCTCTTA-3', recA down 5'-TCTTCTTTCTTTGGCTCATCTTTT-3'; polI up 5'-TCGGGCCAAGACTCCTGATGA-3', polI down 5'-CGCCTCCCGCACCACTTC-3'; hexA up 5'-CAACAAGAATGCCGACAATCC-3', hexA down 5'-GAGCCCCCATAACCTTTTCAAC-3'; hexB up 5'-CCATTGACGCGGGCTCTA-3', hexB down 5'-CCTGAATACGTCGGAACATCTTT-3'; ddl up 5'-AAGTTTTAAGGCTTGACGGAGTT-3', ddl down 5'-GCGGAACGCGAAGTCTCTGTC-3'; comA up 5'-TAAGGCGGATATGACGCT-3', comA down 5'-CATGCGCTCGATATCCTCTCG-3'; trpA/B up 5'-CTAGTAGCCTGTGTTGGT-3', trpA/B down 5'-TCGAACCGACGATAACAC-3'; ciaR/Hup 5'-TTGCGGATGTTATGCAGGTATT-3', ciaR/H down 5'-GCCGGGTTCTAGCCTTGTCT-3'. Products from each amplified locus were tested in order to select a suitably discriminating restriction enzyme, i.e. a panel of five or six enzymes that cut frequently along each of the amplified fragments was examined for the ability to clearly identify allelic variants. Those finally selected were: recP, MnlI; recA, HaeII/MboI; polI, ScrFI; hexA, TaqI; ddl, DdeI/HaeIII; comA, HinPI; hexB, MnlI and HinfI; trpA/B, MseI; and ciaR/ciaH, Hsp92II. Restriction digests were separated by electrophoresis on 4% and 8% acrylamide gels and visualized to determine different alleles by staining with ethidium bromide (10 mg l-1) and recorded using the UVtech Gel Doc analysis system. If no PCR product was obtained for a particular isolate at a particular locus on three successive occasions, this was scored as a null allele and designated an allele number. Strains possessing unique clusters of alleles are referred to as restriction types (RTs).
Repetitive element PCR.
The PCR using repetitive element primers (REP-PCR) was performed as described by Versalovic et al. (1993) .
Pneumolysin and autolysin PCR.
The pneumolysin (ply) and autolysin (lytA) genes were amplified as described by Whatmore et al. (1999) .
Nucleotide sequencing and analysis.
The hexB and trpA/B PCR products generated for MLRT were purified through Microcon 100 columns (Amicon), sequenced using dideoxyterminators and rhodamine dyes, and run on an ABI373 sequencer. Preliminary sequence analysis was performed using the package DNAStar, and MEGA (Kumar et al., 1993 ) was used for analysis of polymorphic sites. Mosaic structure resulting from interspecies recombination events was detected both by the maximum chi-squared test (P<0·001) (Maynard Smith, 1992
) and by Sawyers runs test (Sawyer, 1989
). The programs for chi-squared tests and the Sawyers analysis were written in C++ for the Macintosh by Nick Ross (University of Sussex).
Multilocus sequence typing (MLST).
MLST using the previously described loci (Enright & Spratt, 1998 ) was used to further confirm that frequently carried clones in this study are also involved in serious invasive disease.
Population analysis.
Population analysis was carried out using the index of association (IA) as defined previously (Brown et al., 1980 ; Souza et al., 1992
; Maynard Smith et al., 1993
; Haubold & Rainey, 1996
). RTs were treated equivalently to electrophoretic types (ETs) as defined by MLEE. We examined whether alleles were randomly associated that is at linkage equilibrium, indicating a freely recombining population or non-randomly associated that is at linkage disequilibrium, implying a clonal population structure. If there is linkage equilibrium, i.e. a random association between alleles of different loci, IA=0. If IA is significantly different from 0, it indicates that recombination has been rare or absent (Maynard Smith et al., 1993
). To calculate whether the IA differed significantly from zero, a Monte Carlo procedure with 10000 resamplings was carried out according to Souza et al. (1992)
and Haubold & Rainey (1996)
, and 95% confidence intervals were constructed around IA values. A single-tailed test was used and the significance set at 0·05. A FORTRAN program written by B. Haubold was used for the analysis (Haubold & Rainey, 1996
). One analysis was performed with all isolates included and another only with representatives of each RT in order to eliminate bias introduced by a possible epidemic structure. Cluster diagrams were constructed by entering the alleles for each strain into an Excel database and importing these into the Statistica package (StatSoft). Cluster analysis was performed using a percentage disagreement measure of distance to produce a distance matrix for all isolates, from which a dendrogram was constructed using the unweighted pair group method with arithmetic means (UPGMA). Differences between allele frequencies were tested with a G-test with the significance level adjusted for the number of tests (9) carried out at 0·006 using SAS (SAS Institute, 1990
) statistical software.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
Amongst the isolates examined there were 19 clonal groups composed of three or more isolates from one location, or two or more isolates from at least two different locations. There was clear evidence of clonal groups B, C, E, G, H, I, K, M, N, O, P, Q and S of serotypes 1, 3, 6B, 9V, 10A, 14, 15B, 15C, 16F, 17F, 18C, 19A, 19F, 22, 23F, 33F and 35 being isolated from asymptomatic carriers in both Oxford and Manchester. Three of these, H (serotypes 9V/14/19F), I (serotype 10A) and M (serotype 16) also possessed members from Uruguay or Spain (Table 1). Of the British carried clones, serotypes 3 and 23F were the most numerically dominant in the study (Table 1
). Clonal groups H, I, J and M were composed of isolates from the throat or nasopharynx, as well as isolates causing invasive disease. Clonal group H also possessed one member from Uruguay, with a distinct serotype (serotype 14), and unlike the other five carried members was resistant to penicillin. The genetic identity of prevalent carried clones was further confirmed by REP-PCR (Versalovic et al., 1993
), discounting the possibility that these groups with a common serotype were clustered only because they coincidentally possessed frequently occurring alleles at the loci examined (data not presented). Because of the restricted number and geographically disparate collection of invasive isolates in this study, MLST was used to compare carried clones with a large MLST database of invasive isolates (Enright & Spratt, 1998
) to determine how many of the frequently carried RTs in this study are also associated with invasive disease (Table 1
). Of the 17 carried clones examined, eight gave identical allelic matches (7/7 loci) to previously studied invasive isolates, seven differed at a single locus, one differed by two loci (clonal group Q) and another (clonal group D) by three. Isolates differing in allelic variants at one or two loci have previously been regarded as belonging to the same clonal complex (Enright & Spratt, 1998
). It therefore appears that almost all (16/17) of the carried clones investigated in this study have members that have been associated elsewhere with invasive disease.
Different serotypes were found to be broadly distributed throughout the whole dendrogram (Fig. 1), e.g. five RTs of serotype 1 separated by a linkage distance of up to 0·4 (equivalent to divergence at 4/9 loci). Other genetically distant RTs with common serotype included five RTs of serotype 3, five RTs of serotype 6B, and four RTs of serotype 23F. Serotype 9V isolates were associated with two distinct RTs separated by a linkage distance of up to 0·4. Serotype 16F isolates were found in two closely related clusters, M and N, and within a third more distantly related RT (Table 1
). Three serotype 12F isolates from Kenya with different RTs appeared to be related, having a common node with a linkage distance of ~0·25. Strain 874 from Kenya was clearly the most distant member of the population, with unique alleles at eight of the nine loci examined. Only the allele for comA (allele 1) was similar to that found for other isolates (Table 1
). However, 874 was shown to possess both pneumolysin and autolysin by PCR amplification of ply and lytA.
Population analysis of invasive isolates
MLRT of invasive isolates revealed between 2 and 12 alleles at each locus, with a genetic diversity that was similar to that for carried isolates (0·450) (Table 2). For all but two loci, recP and ddl, there was no significant difference (P=0·001) in allelic frequencies from those of carried isolates. For ddl, three alleles (79) were uniquely associated with penicillin-resistant invasive isolates; additionally ddl allele 6, which was found in ~12% of carried isolates, was not found among any of the susceptible or resistant invasive isolates (Table 2
). Invasive isolates were found to be at linkage equilibrium at the level of both individual isolates (IA=0·087, P=0·197) and RTs (IA=0·05, P=0·321). Including both carried and invasive isolates there were 30 RTs composed of two or more isolates.
Evidence for capsular type switching
The results of the population analysis initially revealed 11 RTs that each possessed at least two different capsular types; these are either grouped or placed adjacent to each other within Table 1. Subsequent sequencing of fragments of hexB and trpA/B to confirm these results revealed that only four of these RTs, where switching appeared to occur, were identical at all loci examined: these are represented in Table 1
by clonal groups H, K, Q and R. Therefore clear evidence of serotype exchange (H, 9V/14/19F; K, 15B/C/19A; Q, 22/35; and R, 23F/23A) is restricted to members of these four groups. However, this still suggests that serotype exchange is occurring among 4/29 (14%) of all RTs possessing two or more isolates, or 4/24 (17%) of carried RTs. Alternatively, serotype exchange appears to have occurred among 6/105 (5·7%) of all isolates within these carried and invasive RTs.
Fourteen individuals in Oxford were found to be infected by two different serotypes of S. pneumoniae at the time of sampling. Clearly multiple infections are necessary for horizontal gene transfer to occur between isolates. However, all isolates of distinct capsular type from a single individual were found to be different by MLRT, indicating that they were genetically distinct strains and not genetically similar isolates with different capsular types.
Evidence for recombination at the trpA/B locus
The trpA/B genes exhibited the greatest genetic diversity among the nine loci examined. Sequencing nine of the 15 allelic variants revealed clear evidence of this locus having evolved by recombination (Fig. 2). This reveals tracts of nucleotide alterations in several different alleles that form mosaic genes with different recombination points. In all cases, sequences are consistent with the MLRT patterns obtained and confirm that the Kenyan isolate 874 (trpA/B allele 10) is the most divergent: between 5·2% and 7·0% divergent from the other trpA/B alleles sequenced. Overall genetic divergence within the remaining trpA/B alleles is substantially less, ranging from a minimum of 0·8% to a maximum of 3·0%. The majority of nucleotide changes are silent the mean ds/dn ratio of the trpA/B dataset is 13·59.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The Kenyan sample of isolates from HIV-seropositive patients in this study initially differed from those within the UK in appearing to be at linkage disequilibrium. However, it became clear that the analysis was biased by the inclusion of the non-typable isolate 874, which appeared divergent from most of the other S. pneumoniae isolates used in this study (Fig. 1). This is underlined by the fact that the Kenyan population returned to linkage equilibrium when isolate 874 plus four other typable isolates were excluded. These four typable Kenyan isolates, although possessing uncommon alleles at several loci, did in fact cluster among other pneumococci once all of the isolates were included in the analysis. Isolate 874, which appeared very distinct from all other organisms in this study, possessed unique alleles for eight out of the nine loci examined (the common allele being that for comA, which is involved in the export of competence-stimulating peptide, CSP). Despite this, optochin susceptibility and molecular methods of identification (16S RNA gene probe, PCR amplification of pneumolysin and autolysin genes) repeatedly classified 874 as S. pneumoniae. It is becoming increasingly apparent that isolates such as 874 from an HIV-seropositive patient, and others reported previously as atypical pneumococci (Diaz et al., 1992
), represent organisms that are to some extent genetically isolated from typical capsulate and non-capsulate S. pneumoniae. Such isolates, along with other oral streptococci, are undoubtedly important for the population genetics of S. pneumoniae as they act as a reservoir of allelic variation for the pneumococcal population, as has been shown for the evolution of penicillin resistance (Dowson et al., 1993
) and more recently for the evolution of the pneumococcal neuraminidase gene, nanA (S. J. King & C. G. Dowson, unpublished). Although it is clear that HIV-seropositive patients are prone to high carriage rates of pneumococci and incidents of invasive disease (Paul, 1997
) it is also possible that HIV patients may be more likely than immunocompetent individuals to carry divergent isolates such as 874. If this is the case then HIV patients may be an important reservoir for the generation of allelic variation within pneumococci. This requires further study.
For all but two of the nine loci examined, recP and ddl, there was no significant difference in the distribution of allelic frequencies between carried and invasive isolates. For ddl (P=0·001) the observed differences may be related to the inclusion of penicillin-resistant isolates among the collection of invasive isolates. Penicillin-binding protein (PBP) 2B is known to be important in the evolution of resistance (Dowson et al., 1993 ) and lies within 1 kb of ddl. Therefore selection for the horizontal transfer of PBP2B may have driven allelic variation of ddl among these isolates by hitchhiking along with pbp2b. This appears to be borne out by the presence of three allelic variants of ddl that are uniquely associated with penicillin-resistant isolates. It is unclear what may account for the distribution differences seen for recP (P=0·001), as the location of this gene is not yet clear (http://www.selkov-7.mcs.anl.gov/WIT2/CGI/).
This study has identified numerous carried clones of serotypes 1, 3, 6B, 9V, 10A, 14, 15B, 15C, 16F, 17F, 18C, 19A, 19F, 22, 23F, 33F and 35 present at two distinct locations within the UK (Manchester and Oxford). Of these, the serotype 3 and 23F clonal groups (C and S) were numerically the most dominant clones in the study. Apart from looking at the prevalence and stability of disease-associated serotypes (to determine serotype inclusion within capsular based vaccines), it is also important to identify the frequency and distribution of clonal groups associated with serious disease (and ultimately allelic variation of determinants within these) if we are to identify common pneumococcal proteins for inclusion within future generations of vaccines. Clonal distribution has been addressed in several previous studies (Enright & Spratt, 1998 ; Coffey et al., 1991
, 1998
; Hall et al., 1996
; Lomholt, 1995
; Nesin et al., 1998
; Sibold et al., 1992
). These clonal complexes were predominantly associated with invasive disease in several countries around the world. However, it is not clear how frequently these isolates are also associated with asymptomatic carriage. By including a population of invasive isolates along with the carried isolates in this study, we have clearly identified by MLRT four clones that are associated with both asymptomatic carriage and invasive disease. One of these (clonal group H) is composed of carried and invasive isolates from Manchester, Oxford and Uruguay. Enright & Spratt (1998)
also identified this as an important clone, suggesting that this is a frequently carried clone that is also frequently associated with invasive disease (a serotype 14 isolate from Uruguay was common to both studies). Although it has been reported previously that there is a different distribution of serotypes associated with carriage and invasive disease (Austrian et al., 1976
), our MLST-based comparison of frequently carried isolates from this study with invasive isolates studied previously (Enright & Spratt, 1998
) has shown that almost all (16/17) of the frequently carried clones examined here are also reported to be associated with invasive disease. This observation suggests that there is no obvious distinction between carried and invasive pneumococci, and that nasopharyngeal isolates do not represent a benign population. It is more clearly the case that isolates that are frequently carried are the same as those that frequently cause invasive disease. This may have important implications for vaccination strategies.
As with previous studies we did not find a strict correlation between RT and serotype. Even though isolates of the same RT were in the majority of cases of the same serotype, there is clearly evidence of isolates with the same genetic background possessing heterologous capsular types. Capsular type switching has been described by a number of groups (Coffey et al., 1991 , 1998
; Hermans et al., 1997
; Nesin et al., 1998
). However, estimates of the frequency of serotype exchange among different ETs of pneumococci with two or more members vary from 4% (Lomholt, 1995
), to 7% (Hall et al., 1996
), to 21% (Sibold et al., 1992
). These studies used MLEE. More recently, data presented by Enright & Spratt (1998)
using MLST, which they estimate to have a greater power of resolution than previous MLEE studies, suggests that serotype exchange occurs among 7/34 (20%) of sequence types (STs) with two or more members, or among 5/20 (25%) of STs, if each clonal complex is counted as a single ST and only one example of serotype exchange is counted per clonal complex (Enright & Spratt, 1998
). However, these apparently frequent examples of serotype exchange amongst invasive isolates equate to 7/165 (4·2%) of isolates within these STs. This estimate discounts the change from a typable to a non-typable phenotype. Here we report serotype exchange to occur among 4/29 (14%) of RTs from a predominantly carried population, which equates to 6/105 (5·7%) of isolates within these RTs, and is a similar frequency to that calculated from data by Enright & Spratt (1998)
(4·2%). In this study we found exchanges involving serotypes 9V/14/19F, 22F/35, 15/19 and 23F/23A. Recently published reports of serotype switching have exchanges between serotypes 8/27, 9V/14/19F, 14/19F, 18C/18B and 23F/3/9N/14/19F (Enright & Spratt, 1998
; Nesin et al., 1998
). It has been shown quite clearly that acquisition of heterologous capsular types can alter virulence in murine models of infection (Nesin et al., 1998
). However, it is not yet clear whether the observed capsular exchanges in these isolates are specifically driven by host defence systems, or whether the capsular locus is simply moving in equilibrium with all other loci. This still needs to be formally tested, as there are important implications for the frequency at which exchange may occur under the selective pressure of a widely used capsular-based vaccine. Either way it is clear from this and previous studies (Enright & Spratt, 1998
) that serotype exchange occurs in 46% of isolates possessing a range of different capsular types. It is therefore very likely that the proposed use of conjugate vaccines with restricted valency, and heterospecific conjugate proteins, will impose a strong selective pressure upon pneumococci to acquire novel capsular types that are not covered by the proposed vaccines or to become non-capsulate, to enable persistence during carriage. If vaccination does result in the serotype-specific eradication of carried capsulate pneumococci, time will reveal what will replace these organisms in the human oropharynx. Transportable epidemiological techniques are essential if we are to understand the global spread of infectious disease. Clearly MLST has this capacity. However, it is possible, as shown in this study, to identify important carried and invasive clones from a large number of isolates by the inexpensive use of MLRT followed by MLST and cross-checking against the available database of STs.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Austrian, R. (1986). Some aspects of the pneumococcal carrier state. J Antimicrob Chemother 18, 35-45.[Abstract]
Austrian, R., Douglas, R. M., Schiffman, G. & 6 other authors (1976). Prevention of pneumococcal pneumonia by vaccination. Trans Assoc Am Phys 89, 184194.[Medline]
Baquero, F. & Jesus, B. (1997). Evolution of antibiotic resistance. Trends Ecol Evol 12, 482-487.
Brown, A. H. D., Feldman, M. W. & Nevo, E. (1980). Multilocus structure of natural populations of Hordeum spontaneum. Genetics 96, 523-536.
Coffey, T. J., Dowson, C. G., Daniels, M., Zhou, J., Martin, C., Spratt, B. G. & Musser, J. M. (1991). Horizontal transfer of multiple penicillin-binding protein genes and capsular biosynthetic genes, in natural populations of Streptococcus pneumoniae. Mol Microbiol 5, 2255-2260.[Medline]
Coffey, T. J., Enright, M. C., Daniels, M., Morona, J. K., Morona, R., Hryniewicz, W., Paton, J. C. & Spratt, B. G. (1998). Recombinational exchanges at the capsular polysaccharide biosynthetic locus lead to frequent serotype changes among natural isolates of Streptococcus pneumoniae. Mol Microbiol 27, 73-83.[Medline]
Diaz, E., Lopez, R. & Garcia, J. L. (1992). Role of the major pneumococcal autolysin in the atypical response of a clinical isolate of Streptococcus pneumoniae. J Bacteriol 174, 5508-5515.[Abstract]
Dowson, C. G., Coffey, T. J., Kell, C. & Whiley, R. A. (1993). Evolution of penicillin resistance in Streptococcus pneumoniae; the role of Streptococcus mitis in the formation of a low affinity PBP2B in S. pneumoniae. Mol Microbiol 9, 635-643.[Medline]
Enright, M. C. & Spratt, B. G. (1998). A multilocus sequence typing scheme for Streptococcus pneumoniae: identification of clones associated with serious invasive disease. Microbiology 144, 3049-3060.[Abstract]
Feldman, C. & Klugman, K. P. (1997). Pneumococcal infections. Curr Opin Infect Dis 10, 109-115.
Guttman, D. S. (1997). Recombination and clonality in natural populations of Escherichia coli. Trends Ecol Evol 12, 16-21.
Hall, L. M. C., Whiley, R. A., Duke, B., George, R. C. & Efstratiou, A. (1996). Genetic relatedness within and between serotypes of Streptococcus pneumoniae from the United Kingdom: analysis of multilocus enzyme electrophoresis, pulsed-field gel electrophoresis, and antimicrobial resistance patterns. J Clin Microbiol 34, 853-859.[Abstract]
Haubold, B. & Rainey, P. B. (1996). Genetic and ecotypic structure of a fluorescent Pseudomonas population. Mol Ecol 5, 747-761.
Hermans, P. W., Sluijter, M., Dejsirilert, S., Lemmens, N., Elzenaar, K., van Veen, A., Goessens, W. H. & deGroot, R. (1997). Molecular epidemiology of drug-resistant pneumococci: towards an international approach. Microb Drug Resist 3, 243-251.[Medline]
Hendley, O. J., Sande, M. A., Stewart, P. M. & Gwaltney, J. M. (1975). Spread of Streptococcus pneumoniae in families. I. Carriage rates and distribution of types. J Infect Dis 132, 55-61.[Medline]
Hoeprich, P. D. (1994). Bacterial pneumonias. In Infectious Diseases, pp. 421-433. Edited by P. D. Hoeprich, M. C. Jordan & A. R. Ronald. Philadelphia: Lippincott.
Jordens, Z., Paul, J., Bates, J., Beaumont, C., Kimari, J. & Gilks, C. (1995). Characterization of Streptococcus pneumoniae from human immunodeficiency virus seropositive patients with acute and recurrent pneumonia. J Infect Dis 172, 983-987.[Medline]
Kumar, S., Tamura, K. & Nei, M. (1993). MEGA: molecular evolutionary analysis software for microcomputers. CABIOS 10, 189-191.[Abstract]
Lomholt, H. (1995). Evidence of recombination and an antigenically diverse immunoglobulin A1 protease among strains of Streptococcus pneumoniae. Infect Immun 63, 4238-4243.[Abstract]
McDougal, L. K., Facklam, R., Reeves, M., Hunter, S., Swenson, J. M., Hill, B. C. & Tenover, F. C. (1992). Analysis of multiply antimicrobial-resistant isolates of Streptococcus pneumoniae from the United States. Antimicrob Agents Chemother 36, 2176-2177.[Abstract]
Maiden, M. C. J., Bygraves, J. A., Feil, E. & 10 other authors (1998). Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic micro-organisms. Proc Natl Acad Sci USA 95, 31403145.
Maynard Smith, J. (1992). Analysing the mosaic structure of genes. J Mol Evol 34, 126-129.[Medline]
Maynard Smith, J., Smith, N. H., ORourke, M. & Spratt, B. G. (1993). How clonal are bacteria? Proc Natl Acad Sci USA 90, 4384-4388.[Abstract]
Muñoz, R., Coffey, T. J., Dowson, C. G., Laible, G., Casal, J., Hakenbeck, R., Jacobs, M., Musser, J. M., Spratt, B. G. & Tomasz, A. (1991). Interncontinental spread of a multiresistant clone of serotype 23F Streptococcus pneumoniae. J Infect Dis 164, 302-306.[Medline]
Musser, J. M. (1996). Molecular population genetic analysis of emerged bacterial pathogens: selected insights. Emerg Infect Dis 2, 1-17.[Medline]
Nesin, M., Ramirez, M. & Tomasz, A. (1998). Capsular transformation of a multidrug-resistant Streptococcus pneumoniae in vivo. J Infect Dis 177, 707-713.[Medline]
ORourke, M. & Stevens, E. (1993). Genetic structure of Neisseria gonorrhoeae populations: a non-clonal pathogen. J Gen Microbiol 139, 2603-2611.[Medline]
Paul, J. (1997). HIV and pneumococcal infection in Africa. Trans R Soc Trop Med Hyg 91, 632-637.[Medline]
Paul, J., Bates, J., Kimari, J. & Gilks, C. (1996). Serotypes and antibiotic susceptibilities of Streptococcus pneumoniae in Nairobi, Kenya. J Infect 32, 139-142.[Medline]
Pons, J.-L., Mandement, M.-N., Martin, E., Lemort, C., Nouvellon, M., Mallet, E. & Lemeland, J.-F. (1996). Clonal and temporal patterns of nasopharyngeal penicillin-susceptible and penicillin-resistant Streptococcus pneumoniae strains in children attending a day care center. J Clin Microbiol 34, 3218-3222.[Abstract]
SAS Institute (1990). SAS Users Guide, Version 6. Cary, NC: SAS Institute.
Sawyer, S. (1989). Statistical tests for detecting gene conversation. Mol Biol Evol 6, 526-538.[Abstract]
Scott, J. A. G., Hall, A. J., Dagan, R. & 12 other authors (1996). Serogroup-specific epidemiology of Streptococcus pneumoniae associations with age, sex, and geography in 7000 episodes of invasive disease. Clin Infect Dis 22, 973981.[Medline]
Sibold, C., Wang, J., Henrichsen, J. & Hakenbeck, R. (1992). Genetic relationships of penicillin-susceptible and -resistant Streptococcus pneumoniae strains isolated on different continents. Infect Immun 60, 4119-4126.[Abstract]
Souza, V., Nguyen, T. T., Hudson, R. R., Piñero, D. & Lenski, R. E. (1992). Hierarchical analysis of linkage disequilibrium in Rhizobium populations: evidence for sex? Proc Natl Acad Sci USA 89, 8389-8393.[Abstract]
Speller, D. C. E., Johnson, A. P., Cookson, B. D., Waight, P. & George, R. C. (1996). PHLS surveillance of antibiotic resistance, England and Wales: emerging resistance in Streptococcus pneumoniae. Emerg Infect Dis 2, 57-58.[Medline]
Sung, R. Y. T., Ling, J. M., Fung, S. M., Oppenheimer, S. J., Crook, D. W., Lau, J. T. F. & Cheng, A. F. B. (1995). Carriage of Haemophilus influenzae and Streptococcus pneumoniae in healthy Chinese and Vietnamese children in Hong Kong. Acta Paediatr 84, 1262-1267.[Medline]
Versalovic, J., Kapur, V., Mason, E. O., Shah, V., Keouth, T., Lupski, J. R. & Musser, J. M. (1993). Penicillin-resistant Streptococcus pneumoniae strains recovered in Houston: identification and molecular characterisation of multiple clones. J Infect Dis 167, 850-856.[Medline]
Weiser, J. N., Austrian, R., Screenivasan, P. K. & Masure, H. R. (1994). Phase variation in pneumococcal opacity relationship between colonial morphology and nasopharyngeal colonization. Infect Immun 62, 2582-2589.[Abstract]
Whatmore, A. M., King, S. J., Doherty, N., Sturgeon, D., Chanter, N. & Dowson, C. G. (1999). Molecular characterisation of equine isolates of Streptococcus pneumoniae: natural disruption of genes encoding the virulence factors pneumolysin and autolysin. Infect Immun 67, 2776-2782.
Received 31 March 1999;
revised 20 July 1999;
accepted 28 July 1999.