Low genetic diversity of Haemophilus influenzae type b compared to nonencapsulated H. influenzae in a population in which H. influenzae is highly endemic

Low genetic diversity of Haemophilus influenzae type b compared to nonencapsulated H. influenzae in a population in which H. influenzae is highly endemic
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  I NFECTION AND  I MMUNITY ,0019-9567/98/$04.00  0July 1998, p. 3403–3409 Vol. 66, No. 7Copyright © 1998, American Society for Microbiology. All Rights Reserved. Low Genetic Diversity of   Haemophilus influenzae  Type bCompared to Nonencapsulated  H. influenzae  in aPopulation in Which  H. influenzae  IsHighly Endemic HEIDI C. SMITH-VAUGHAN,* KADABA S. SRIPRAKASH, AMANDA J. LEACH,JOHN D. MATHEWS,  AND  DAVID J. KEMP  Menzies School of Health Research, Darwin, Northern Territory, Australia Received 28 May 1997/Returned for modification 8 August 1997/Accepted 14 April 1998 Immunization with  Haemophilus influenzae  type b (Hib) conjugate polysaccharide vaccines has dramaticallyreduced Hib disease worldwide. As in other populations, nasopharyngeal carriage of Hib declined markedly in Absrcinal infants following vaccination, although carriage has not been entirely eliminated. In this study, wedescribe the genetic characteristics and the carriage dynamics of longitudinal isolates of Hib, characterized byusing several typing methods. In addition, carriage rates of nonencapsulated  H. influenzae  (NCHi) are high,and concurrent colonization with Hib and NCHi is common; we also observed NCHi isolates which weregenetically similar to Hib. There is a continuing need to promote Hib immunization and monitor  H. influenzae carriage in populations in which the organism is highly endemic, not least because of the possibility of geneticexchange between Hib and NCHi strains in such populations. Prior to the implementation of   Haemophilus influenzae  typeb (Hib) vaccination, Hib was a major cause of invasive diseasein the Northern Territory of Australia. Incidence rates of in- vasive  H. influenzae  disease in children under 5 years of agefrom mid-1985 to mid-1988 were 529/100,000 in the Absrcinalpopulation and 92/100,000 in the non-Absrcinal population,and 87.5% of these cases were caused by Hib (10). In a longi-tudinal study of otitis media in Absrcinal infants living in aremote community (1992 to 1994), first nasopharyngeal acqui-sition of respiratory bacteria (  H. influenzae ,  Streptococcus pneumoniae , and  Moraxella catarrhalis ) was predictive of earlyonset of otitis media (20). Of   H. influenzae  strains carried inthe nasopharynx of Absrcinal infants, nonencapsulated strainspredominated; the ratio of nonencapsulated  H. influenzae (NCHi) to Hib-positive swabs was approximately 6.6 to 1. Priorto the introduction of the Hib conjugate capsular polysaccha-ride vaccine to this community in mid-1993, the cumulativeincidence of nasopharyngeal carriage of Hib in the infantsunder study was at least 42.8% by age 6 months (12 of 28infants, with up to eight swabs collected from each infant) (19).These rates fell during the first 6 months of vaccine use andreached zero in 1994 (none of 88 swabs collected from 15infants to the age of 6 months), with no change in carriage of NCHi and non-type b encapsulated strains. Subsequently Hibcarriage rates in infants in the same population have risen; thisis of potential concern (21).The polysaccharide capsule of   H. influenzae  is synthesized byenzymes encoded at the capsulation locus (  cap ).  cap  is approx-imately 17 to 18 kb in size and is organized into three func-tionally distinct regions (14) as in all capsulated  H. influenzae serotypes (17). Approximately 98% of all type b strains actuallycontain two directly repeated copies of the 17- to 18-kb seg-ment (13). In such strains,  cap  lies between copies of an inser-tion sequence-like element, IS 1016 , allowing regulation of thenumber of copies of   cap  and hence the level of capsule pro-duction (14) after homologous recombination. Increased cap-sule production may provide a selective advantage in certainsituations (15). Alternatively, if there is selection against cap-sule, there would be an advantage for strains in which multiplecopies of the 17- to 18-kb segment at the  cap  locus have beenreduced to a single copy. The single-copy locus is also subjectto loss (32).Encapsulated  H. influenzae  strains have been classified bymultilocus enzyme electrophoresis into two highly divergentphylogenetic divisions, designated I and II (25). Most invasivestrains cluster in division I (strains of serotypes a through e), while less virulent strains of serotypes a, b, and f cluster indivision II. The increased virulence of division I Hib strains worldwide has been attributed to the deletion of a specificsequence (IS-  bexA ) from the tandemly duplicated capsule lo-cus (Fig. 1) (16). Loss of this region inhibits recombination andis believed to maintain the tandem configuration of   cap  withhigh levels of capsule production (3).Nevertheless, nonencapsulated Hib mutants arise at fre-quencies of 0.1 to 0.3% (12) and have been reported in severalstudies (4, 6, 11, 22, 32). Indeed, NCHi isolates are moreadherent to buccal mucosa than capsulated forms (18), sug-gesting that capsule loss is a means of maintaining mucosalcolonization (12). Immune pressure following Hib conjugate vaccination or past infection would also select against capsu-lated forms (4), raising the possibility that the widespread useof the Hib conjugate vaccine could lead to an increase innonencapsulated mutants. Moreover, if vaccination levels fallin the community, it is also possible that nonencapsulated Hib variants can revert to the capsular form and cause new out-breaks of invasive Hib disease. Hence, it is important to char-acterize  H. influenzae  strains in populations where the organ-ism is endemic to determine the relationship, if any, betweenNCHi and Hib.  H. influenzae  uses further means for antigenic variation topromote survival. Most bactericidal and opsonizing antibodiestargeted against membrane proteins of   H. influenzae  are strain * Corresponding author. Mailing address: Menzies School of HealthResearch, P.O. Box 41096, Casuarina NT 0811, Australia. Phone: 61889 228 196. Fax: 61 889 275 187. E-mail:   onF  e b r  u ar  y 1 7  ,2  0 1  6  b  y  g u e s  t  h  t   t   p:  /   /  i   ai  . a s m. or  g /  D  ownl   o a d  e d f  r  om   specific (34). Hence, variation in outer membrane antigens isan effective means to escape antibody-dependent defensemechanisms. The genetic diversity of outer membrane proteinsof NCHi is much greater than that of Hib (23, 24). This mayreflect the dominant role of capsular polysaccharide in Hib asan antigenic determinant of strain structure, or it may be dueto lateral genetic transfers in NCHi, with more of a barrier tosuch transfers in Hib due to encapsulation. Indeed, one hy-pothesis to account for the clonality of Hib, supported byexperimental data (33), is that the capsule acts as a barrier toexogenous DNA (24). Nevertheless, acapsular mutants of Hibcould use mechanisms similar to those used by NCHi to averthost immune pressures against newly exposed surface antigens.The P2 gene of   H. influenzae  encodes the major outer mem-brane protein P2, with typical transmembrane domains andextracellular loops; this porin is an important target of the hostimmune response (9). Hib strains have highly conserved P2sequences (23), unlike NCHi strains (2). A study of four P2sequences from Hib (23) demonstrated that two strains wereidentical in P2 type to the MinnA strain, a frequent and viru-lent Hib strain in the United States, and that another straindiffered by only one nucleotide in loops 4 to 6 of the P2 porin.The  H. influenzae  Rd strain has 94% amino acid identity toMinnA through loops 4 to 6 (  H. influenzae  Rd accession no.1573092).The coexistence of Hib and NCHi and high endemicity inthis population (20) warranted a detailed molecular character-ization of local isolates. PCR ribotyping, highly discriminatoryfor NCHi, revealed a particular ribotype (PRT 1) that is com-mon among both NCHi and Hib isolates. Further analysisshowed that NCHi-PRT 1 isolates are diverse whereas Hib-PRT 1 isolates seem clonal. The evolutionary relationshipsbetween encapsulated and nonencapsulated forms of PRT 1isolates, and vaccine implications, are discussed. MATERIALS AND METHODSCommunity and subjects.  The longitudinal study of otitis media in Absrcinalinfants undertaken over the 1992-to-1994 period has been reported previously(20). Briefly, with ethical approval, 41 Absrcinal infants at a remote islandcommunity, representing 85% of the appropriate age cohort, were monitoredfrom birth to examine the relationship between bacterial colonization of thenasopharynx and the onset of otitis media. Nasopharyngeal and ear dischargeswabs were regularly collected from the infants.The Hib conjugate polysaccharide vaccine, PRP-OMP (Pedvax HIB; MerckSharp & Dohme), was introduced to the community in mid-1993. It was admin-istered to infants at 2, 4, and 6 months; older infants and children under 5 yearsof age were included in a catch-up program. Isolates.  Microbiological methods were as described previously (20). Briefly,nasopharyngeal swabs were smeared for Gram staining and frozen in 1.0 ml of transport broth. Thawed broth was mixed, and 10-  l aliquots were cultured on7% chocolate agar and chocolate agar plus bacitracin, vancomycin, and clinda-mycin. From most swabs there was a heavy growth of   H. influenzae , and fourcolonies were sampled from each plate, taking care to select any colonies that were morphologically distinct. Isolates of   H. influenzae  were identified by theirrequirement for X and V factors. Hib organisms were recognized by theiragglutination with antisera against the type b capsular antigen, while NCHiorganisms were recognized by their lack of agglutination reaction against thecapsular antigens, a to f (Phadebact; KaroBio Diagnostics AB, Huddinge, Swe-den). NCHi and Hib isolates were collected prior to and during the Hib vacci-nation period (1992 to 1994 and in 1996).  Amoxicillin MIC.  The MIC of amoxicillin against each isolate was determinedby Etest (AB Biodisk, Solna, Sweden). The procedure was as recommended bythe manufacturer and the National Committee for Clinical Laboratory Standards(NCCLS), with one exception. Chocolate agar (Oxoid) was used to test theNCCLS-recommended  H. influenzae  reference strain ATCC 49247 and was usedin place of the recommended  Haemophilus  test medium. The reference strain, ATCC 49247, which has an acceptable ampicillin MIC range of 2 to 8   g/ml,consistently gave results in this range when the amoxicillin MIC was determinedby using chocolate agar. The ampicillin equivalent MIC breakpoints to zonediameter interpretive standards for  H. influenzae  species are  4  g/ml for resis-tant and  1  g/ml for sensitive (26). DNA preparation.  Total chromosomal DNA was initially extracted from singlecolonies of   H. influenzae  as described previously (29) but more recently by lysisof bacteria embedded in agarose at the bottom of microtiter wells followed byextraction with proteinase K (8). Molecular characterization of the isolates. (i) Long PCR ribotyping.  Thetechnique has been described previously (30). Briefly, approximately 100 ng of DNA was amplified in a 20-  l reaction mixture consisting of PC2 buffer (1), 0.2mM deoxynucleoside triphosphates (dNTPs), 4 pmol each of 16SG primer and5S primer, and 0.12   l of   Taq-Pfu  (15 U of   Taq  polymerase and 2.5 U of   Pfu DNA polymerase in 16   l). Each mixture was overlaid with 50   l of liquidparaffin and cycled 25 times at 94°C for 10 s and 68°C for 8 min. To the resultingPCR product, 1.5 U of   Hae III was added, and the mixture was incubated at 37°Cfor 2 h. Restriction fragments were separated by electrophoresis on a 1.5%agarose gel in 1   TAE buffer (4 mM Tris acetate, 0.2 mM EDTA [pH 8.0])containing 0.5  g of ethidium bromide/ml and photographed by using Polaroid667 film and a 300-nm transilluminator. The resulting patterns were compared visually. PCR ribotyping using the enzyme  Apo I (4 U per reaction mixture,incubated at 37°C for 14 h) was also performed. (ii) P2 typing.  The region of the P2 gene corresponding to loops 4 to 6 of the P2 protein was amplified in a reaction mix (final volume of 25   l)containing approximately 100 ng of dNTP template DNA, 0.2 mM dNTPs, 25pmol of P2L4 and P2L6 primers (31), 2.5 mM MgCl 2 , and 1 U of   Taq polymerase in a reaction buffer (Bresatec). After 30 cycles at 95°C for 15 s,60°C for 1 min, and 72°C for 30 s, PCR products were purified from 0.8%agarose gels by using Bandpure (Progen) as instructed by the manufacturer.Cycle sequencing was performed in both directions with radioactively labeledprimers P2L4 and P2L6, using an AmpliCycle sequencing kit (Perkin-Elmer)as instructed by the manufacturer. Each sequencing reaction was cycled 30times at 95°C for 1 min, 60°C for 1 min, and 72°C for 30 s. The DNA wasseparated on a 6% polyacrylamide sequencing gel as described by Sambrooket al. (27). (iii) PCR fingerprinting.  Isolates were PCR fingerprinted by using primersdirected to enterobacterial repetitive intergenic consensus (ERIC) sequences asdescribed by van Belkum et al. (36). The PCR mix was as described for P2 typingexcept that primers ERIC 1R (ATGTAAGCTCCTGGGGATTCAC) (35) andERIC 2 (AAGTAAGTGACTGGGGTGAGCG) (35) were used. Samples werecycled 35 times at 94°C for 1 min, 25°C for 1 min, and 74°C for 4 min, followedby a final 10-min incubation at 74°C. (iv) Confirmation of serotype and detection of capsule genes by PCR.  AllNCHi and Hib isolates were subjected to PCR as described for P2 typing, usingprimers directed to the  bexA  region (H1 and H2) and the type b-specific region(b1 and b2) (5). The reaction mix, which was as described for P2 typing exceptfor the primers, was cycled 25 times at 95°C for 15 s, 60°C for 1 min, and 72°Cfor 1 min, followed by a final 10-min incubation at 72°C. A 343-bp product (37) with primers H1 and H2 confirmed capsulation, and generation of a 480-bpproduct with primers b1 and b2 indicated a type b capsule (5). A positive reaction with primers b1 and b2 and a negative reaction with primers H1 and H2 wouldbe expected for a type b nonencapsulated mutant. (v) Detection of the IS-  bexA  deletion.  Isolates which had the IS-  bexA  deletiongenerated a 339-bp product by PCR across the IS 1016/bexA  region, as describedby Kroll et al. (16). The PCR mix was as described for P2 typing except thatprimers IF (ATTAGCAAGTATGCTAGTCTAT) from IS 1016  and IR (CAATGATTCGCGTAAATAATGT) from  bexA  (16) were used. Samples were cycled30 times at 95°C for 1 min, 42°C for 1 min, and 72°C for 30 s. (vi) Southern blotting and hybridization.  Approximately 1  g of chromosomalDNA extracted from  H. influenzae  isolates was digested with the restrictionenzyme  Eco RI (Pharmacia) as instructed by the manufacturer. DNA restrictionfragments were separated by electrophoresis on a 0.8% agarose gel, transferredto a positively charged membrane (Hybond N  ; Amersham) in 0.4 M NaOH,and neutralized in 2  SSC (0.3 M NaCl, 30 mM trisodium citrate) substantiallyas described in reference 27. Approximately 100-ng aliquots of pU038 (kindly provided by E. R. Moxon’slaboratory), IS 1016  (PCR products, as described by St. Geme et al. [32], usingHib DNA as a template), and  Eco RI-digested pBR322 (Pharmacia) probes wereprepared by random primer labeling with [  - 32 P]dATP, using a Gigaprime DNA labeling kit (Bresatec). Plasmid pU038 contains a complete set of   cap  genes fromHib and an IS 1016  element cloned into pBR322. Blots were prehybridized for 4 hat 65°C before addition of the labeled probes. The prehybridization mixturecontained 0.3 M NaCl, 20 mM NaH 2 PO 4 , 2 mM EDTA, 1% sodium dodecylFIG. 1. Example of the IS-  bexA  deletion in a phylogenetic division I Hibstrain with a duplication at  cap . The diagonal line indicates the location of theIS-  bexA  deletion. PCR amplification using primers denoted IF and IR resulted ina strong product of approximately 339 bp which relates to the truncated regionand a lighter product (PCR is preferential to smaller products) of 1.2 kb relatingto the complete form of the IS 1016  bridge region. 3404 SMITH-VAUGHAN ET AL. I NFECT . I MMUN .   onF  e b r  u ar  y 1 7  ,2  0 1  6  b  y  g u e s  t  h  t   t   p:  /   /  i   ai  . a s m. or  g /  D  ownl   o a d  e d f  r  om   sulfate, 0.5% nonfat skim milk powder, and 0.5 mg of herring sperm DNA perml. Hybridization was for 16 h at 65°C, and posthybridization washes were at65°C with 0.1% sodium dodecyl sulfate–0.1   SSC. Membranes were then ex-posed to Kodak X-Omat AR film. RESULTSCharacterization of Hib isolates by molecular typing.  Hib was cultured from nasopharyngeal swabs collected from 20 of the 41 (48.8%) infants in the longitudinal study between 1992and 1994 (20). In 1996, 6 (30%) of 20 infants monitored to age6 months in the same community were Hib colonized; all six  were under 4 months of age and either had received the firstdose only at 2 months of age or were too young for vaccination.Eight Hib isolates collected from these infants were compared with one colony from each of 39 available Hib-positive swabscollected from the 20 infants between 1992 and 1994.Serotyping of Hib isolates was confirmed by PCR usingprimers for type b-specific sequences (b1 and b2) (5); all 47Hib isolates gave positive results. In addition, primers H1 andH2, directed to  bexA , were used to identify encapsulation genes(37).  All Hib isolates had the IS-  bexA  deletion.  PCR across theIS-  bexA  deletion generates a PCR fragment that is diagnostic(Fig. 1). This approach detected the IS-  bexA  mutation in all 47Hib isolates, suggesting that the isolates have the multiple-copy  cap  locus characteristic of strains in phylogenetic division I;they are thus potentially invasive (6). Subtyping of Hib clones.  One colony from each of the 47Hib-positive swabs was analyzed by PCR ribotyping. All Hibisolates from the community and the reference strain MinnA typed as Hib-PRT 1. These results are illustrated in Fig. 2 andsummarized in Table 1.The same isolates were typed by PCR fingerprinting usingprimers directed at ERIC sequences (35, 36). Among the 47Hib-PRT 1 isolates, two distinct ERIC PCR fingerprintingpatterns, A and B, were observed at similar frequencies (Table1 and Fig. 3). Dynamics of Hib colonization.  Hib was cultured on morethan one occasion from 10 infants in the 1992–1994 study and2 infants examined in 1996. Periods of Hib detection in theseinfants varied from 2 weeks to 8 months. ERIC PCR finger-printing demonstrated that five infants in the 1992–1994 lon-gitudinal study carried ERIC types A and B at separate times,suggesting that there was sequential colonization with differentHib strains or potentially dual carriage with only one ERICtype detected (data summarized in Table 1).  -Lactamase production and amoxicillin resistance.  Of 38Hib isolates from 1992 to 1994 tested, 20 were amoxicillinsensitive, with MICs from 0.125 to 0.5  g/ml. For 18 amoxicil-lin-resistant isolates, MICs were greater than 256   g/ml. Of eight Hib isolates from 1996, one was amoxicillin resistant(MIC    256   g/ml) and the remainder were amoxicillin sen-sitive, with MICs from 0.25 to 0.38  g/ml. All resistant isolates were  -lactamase producers. Correlation of ERIC pattern with   -lactamase production.  Among the Hib isolates available for testing, all ERIC type Bisolates were non-  -lactamase producers and 18 (81.8%) of 21ERIC type A isolates were  -lactamase producers. This asso-ciation suggests that  -lactamase production is chromosomallycontrolled in these isolates. NCHi isolates exhibited the Hib PCR ribotype.  Previ-ously, NCHi isolates from 10 infants of the 41 enrolled inthe study were PCR ribotyped. Among 432 NCHi isolatesrecovered from 141 nasopharyngeal and ear dischargeswabs, 46 different PCR ribotypes were detected (28). Themarker PCR ribotype for the common Hib clone (Hib-PRT1) was also detected among the nonencapsulated popula-tion, and we refer to these isolates as NCHi-PRT 1. TheNCHi-PRT 1 strain was cultured from the nasopharynx orear discharge of 3 of the 10 infants.To further explore the associations of the PRT 1 pattern(generated by  Hae III) in nonencapsulated and type b isolates, we measured restriction fragment length polymorphisms inrRNA operons, using  Apo I. Although rRNA operons of NCHi were polymorphic for internal  Apo I sites (Fig. 4, lanes 1 to 5),all NCHi-PRT 1 isolates and several Hib-PRT 1 isolates tested were identical by  Apo I PCR ribotype patterns (Fig. 4, lanes 6to 9). These results preclude the possibility of chance identityof NCHi-PRT 1 and Hib-PRT 1 restriction patterns, using  Hae III within the rRNA operons. ERIC PCR fingerprinting of the NCHi-PRT 1 isolates.  Incontrast to the conserved patterns of rRNA operons of NCHi-PRT 1 isolates, ERIC PCR fingerprints were more diverse;among five NCHi-PRT 1 isolates cultured from three infants,three different ERIC PCR patterns were observed (Fig. 3 andTable 1). This result suggests that other loci have diverged toa greater degree than the ribosomal operons in NCHi. Like NCHi, the NCHi-PRT 1 isolates lacked genes for cap-sulation.  To identify remnant  cap  sequences, 11 NCHi-PRT 1isolates and at least 2 representative isolates of the 45 otherNCHi PCR ribotypes identified to date were amplified by usingprimers to the  bexA  region (H1 and H2) and the type b-specificregion (b1 and b2). None of the NCHi isolates generated aproduct with these primers. To confirm this result, genomicdigests of selected NCHi and Hib isolates were probed withpU038, a probe containing capsule gene sequences, an IS 1016 sequence, and a portion of pBR322 (32). We found that onlyHib strains hybridized to  cap  sequences in pU038; NCHistrains which showed any hybridization with this probe alsodemonstrated the same pattern of hybridization when probed with IS 1016  or pBR322 alone and hence were scored as neg-ative for  cap  sequences. Thus, none of the NCHi-PRT 1 iso-lates or other selected NCHi isolates contain detectable rem-nant capsule sequences. P2 gene sequences are highly conserved in Hib isolates,unlike in the NCHi-PRT 1 isolates.  P2 genes of several Hib- FIG. 2. PCR ribotype patterns of representative isolates of   H. influenzae (  Hae III digest). Lane 1, Hib MinnA; lane 2, Hib-PRT 1 (608-1); lane 3, NCHi-PRT 1 (855-1); lane 4, Hib-PRT 2 (Hib isolated from a Darwin infant). Thisphotograph was digitized with Deskscan II software and annotated with CorelPhotopaint. V OL  . 66, 1998 LOW GENETIC DIVERSITY IN Hib 3405   onF  e b r  u ar  y 1 7  ,2  0 1  6  b  y  g u e s  t  h  t   t   p:  /   /  i   ai  . a s m. or  g /  D  ownl   o a d  e d f  r  om   PRT 1 isolates were sequenced through loops 4 to 6 of the P2gene. The Hib-PRT 1 nucleotide sequences differed by twobases from the MinnA strain sequence, with translated regionsidentical to those in MinnA (accession no. J03359). Further-more, the P2 sequence of a Hib-PRT 1 isolate from 1996 wasidentical with that of earlier Hib-PRT 1 isolates collected in1992 to 1994; this finding supports a high level of P2 conser- vation among Hib isolates.In contrast, the P2 gene sequence of one NCHi-PRT 1isolate differed markedly from that of Hib-PRT 1. The P2protein of NCHi is subject to immune pressure (38) and, asa result, demonstrates a great deal of heterogeneity amongstrains. We previously reported that the NCHi-PRT 1 iso-late in question had the same P2 sequence as an NCHi-PRT25 isolate; this was most likely a result of horizontal genetransfer (31). TABLE 1. Characterization of   H. influenzae  isolates  a Group Isolate no. Infant Examination date  b ERIC type   -Lactamaseproduction AmoxicillinMIC (  g/ml) 1992–1994 Hib isolates 2-1 903 07-Jan.-92 B NT  c NT12-1 904 25-Mar.-92 B    0.2514-2 904 20-May-92 A     25617-3 904 27-Oct.-92 A     25638-1 907 08-Sept.-92 B    0.2546-2 908 05-Jan.-93 A     25650-1 909 24-Mar.-92 A     0.12560-1 911 14-Apr.-92 A     25661-3 911 18-June-92 A     25668-1 913 20-May-92 B    NT77-3 914 02-Dec.-92 B    0.25125-1 923 21-July-92 A     256130-1 923 07-Jan.-93 B    0.25309-1 923 04-Mar.-93 B    0.38134-4 924 29-Oct.-92 A     256137-1 925 21-July-92 A     256143-2 926 01-Dec.-92 B    0.38154-1 928 05-Jan.-93 A     256403-1 928 27-May-93 B    0.38363-1 931 28-Apr.-93 A     256171-1 934 02-Dec.-92 B    0.38274-1 934 10-Feb.-93 B    0.38304-1 935 03-Mar.-93 A     256347-4 935 01-Apr.-93 A     256397-1 935 26-May-93 B    0.38602-4 938 04-Oct.-93 B    0.38619-3 938 08-Oct.-93 B    0.38497-3 943 10-Aug.-93 A     256579-1 943 20-Sept.-93 A     256608-1 947 05-Oct.-93 A     256653-4 947 20-Oct.-93 A     256714-1 947 30-Nov.-93 A     256910-1 947 04-May-94 B    0.19585-1 950 21-Sept.-93 B    0.25598-4 950 28-Sept.-93 B    0.19604-1 950 04-Oct.-93 B    0.38640-1 950 19-Oct.-93 B    0.25617-4 950 08-Oct.-93 B    0.38707-4 950 30-Nov.-93 B    0.251996 Hib isolates 0-1 A36 02-Dec.-96 A     0.38387-1 A37 16-Oct.-96 B    0.25918-1 A29 27-Aug.-96 B    0.25928-1 A29 15-Oct.-96 B    0.25938-4 A33 09-July-96 B    0.38945-1 A33 24-July-96 B    0.38474-1 A47 04-Sept.-96 A     256507-1 A21 22-Apr.-96 A     0.25NCHi isolates 273-1 932 10-Feb.-93 D NT NT475-2 944 28-July-93 E NT NT520-1 944 24-Aug.-93 E NT NT855-1 938 21-Mar.-94 F NT NT871-3 938 04-Apr.-94 F NT NT  a  All strains were PRT 1.  b Day-month-year.  c NT, not tested. 3406 SMITH-VAUGHAN ET AL. I NFECT . I MMUN .   onF  e b r  u ar  y 1 7  ,2  0 1  6  b  y  g u e s  t  h  t   t   p:  /   /  i   ai  . a s m. or  g /  D  ownl   o a d  e d f  r  om   DISCUSSIONHib continues to circulate in the community under study. The PRP-OMP Hib vaccine was introduced to the communityin this study in mid-1993 for children under 5 years of age. Although infant carriage of Hib fell dramatically, it was noteliminated. Since early 1996, Hib has been isolated severaltimes from infants under 6 months of age who were not fullyimmunized. We speculate that a reservoir of Hib is maintainedin older children and adults which will be depleted only whenherd immunity levels are sufficient to decrease the number of carriers below the critical level needed to maintain transmis-sion to immunologically naive individuals. Dynamics of Hib carriage.  Hib carriage is a dynamic process.ERIC typing split the Hib isolates into two groups in this study, with the apparent replacement of one variant by the otherduring several episodes of Hib carriage. With available data, we cannot decide whether the two variants of Hib were trans-mitted, and thus carried together, taking turns as the dominantstrain, or whether there were separate episodes of colonizationin infants who do not develop a strong immune response to theHib polysaccharide capsule following natural exposure. Amoxicillin-resistant,  -lactamase-producing Hib strains didnot appear to have a selective advantage over amoxicillin-sensitive strains in this population, even though amoxicillin iscommonly used in the community. Proportions of amoxicillin-resistant and -sensitive isolates did not change dramaticallyduring the study; 7 (46.6%) of 15 Hib isolates from 1992 wereamoxicillin resistant, compared to 10 (41.6%) of 24 in 1993(numbers of Hib isolates in 1994 to 1996 were small). Further-more, during Hib carriage, strain turnover was more oftenfrom an amoxicillin-resistant strain to an amoxicillin-sensitivestrain (four of five episodes) than the converse. Coexistence of    -lactamase-producing and -nonproducingstrains.  From 1992 to 1994,  -lactamase-producing and -non-producing strains coexisted in this population in relatively sta-ble proportions, with no increase in amoxicillin-resistantstrains and no evidence of transposition of   -lactamase genesto sensitive strains. This balance may be ecological in srcin, with complementary adaptations of    -lactamase-positive and-negative strains; it is also possible that the genes are chromo-somally determined and therefore less mobile than plasmid-carried genes. Furthermore, lateral movement of genes as inNCHi (for example, the P2 gene [31]) may be hindered in Hib,either because the capsule may act as a barrier to exogenousDNA or because of restriction-modification systems (24). Possible explanations for the existence of Hib and NCHi with the same PCR ribotype.  There are several explanationsfor the existence of Hib and NCHi with the same PCR ri-botype. Horizontal transfer of ribosomal operons is unlikely, asup to six different operons (as found in the  H. influenzae  Rdgenome [7]) would have to be transferred. Second, convergentevolution of rRNA operons of two otherwise diverse  H. influ- enzae  organisms is also unlikely. The third and simpler expla-nation is that PRT 1 exists in both capsulated and nonencap-sulated forms because of either uptake or loss of capsule DNA.The loss of capsule DNA seems more plausible, as reduction of the number of copies of   cap  is a well-recognized phenomenon(32) and the lack of capsule gene remnants indicates that allcopies were deleted. Divergence of NCHi-PRT 1 strains atother loci can be explained in terms of the selective pressure onloci such as P2 once the capsular selective constraint was re-moved. NCHi-PRT 1 isolates existed prior to the vaccination era.  If the NCHi-PRT 1 isolates were derived from Hib, capsule losscannot be attributed to vaccine pressure, as NCHi-PRT 1 iso-lates were collected prior to the introduction of vaccination.Long-term coexistence of both forms may depend on comple-mentary selective advantages. For example, NCHi may be bet-ter adapted for mucosal colonization, whereas Hib may resistphagocytosis or desiccation more effectively (12, 16). Diversity of NCHi genome relative to Hib.  NCHi strainscirculating in this community are highly diverse. We observed46 different PCR ribotypes colonizing 10 infants over a 2- to24-month period (28), and some of these types could be split FIG. 3. ERIC PCR fingerprinting patterns of   H. influenzae . Lanes 1, 2, and 5, Hib-PRT 1/ERIC A; lanes 3, 4, and 6 to 9, Hib-PRT 1/ERIC B; lane 10, NCHi-PRT1/ERIC D; lane 11, NCHi-PRT 1/ERIC E; lanes 12 and 13, NCHi-PRT 1/ERIC F; lane 14, Hib-PRT 2/ERIC C (Hib isolated from a Darwin infant). The image wasdigitally acquired by using Bio-Rad GelDoc 1000 and annotated with Corel Draw.FIG. 4.  Apo I PCR ribotype patterns of   H. influenzae . Lane 1, NCHi 621-1(  Hae III PRT 4); lane 2, NCHi 473-22 (  Hae III PRT 3); lane 3, NCHi 413-1(  Hae III PRT 51); lane 4, NCHi 521-1 (  Hae III PRT 2); lane 5, NCHi 599-4(  Hae III PRT 17); lane 6, NCHi 273-1 (  Hae III PRT 1); lane 7, NCHi 855-1(  Hae III PRT 1); lane 8, Hib MinnA (  Hae III PRT 1); lane 9, Hib 608-1 (  Hae IIIPRT 1). The image was digitally acquired by using Bio-Rad GelDoc 1000 andannotated with Corel Draw. V OL  . 66, 1998 LOW GENETIC DIVERSITY IN Hib 3407   onF  e b r  u ar  y 1 7  ,2  0 1  6  b  y  g u e s  t  h  t   t   p:  /   /  i   ai  . a s m. or  g /  D  ownl   o a d  e d f  r  om 
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks