|Home | About | Journals | Submit | Contact Us | Français|
The proportion of meningococcal disease in the United States, South Africa, and Israel caused by Neisseria meningitidis serogroup Y (NmY) was greater than the worldwide average during the period 1999-2002. Genotypic characterization of 300 NmY isolates by multilocus sequence typing, 16S rRNA gene sequencing, and PorA variable region typing was conducted to determine the relationships of the isolates from these three countries. Seventy different genotypes were found. Two groups of ST-23 clonal complex isolates accounted for 88% of the U.S. isolates, 12% of the South African isolates, and 96% of the isolates from Israel. The single common clone (ST-23/16S-19/P1.5-2,10-1) represented 57, 5, and 35% of the NmY isolates from the United States, South Africa, and Israel. The predominant clone in South Africa (ST-175/16S-21/P1.5-1,2-2), and 11 other closely related clones made up 77% of the South African study isolates and were not found among the isolates from the United States or Israel. ST-175 was the predicted founder of the ST-175 clonal complex, and isolates of ST-175 and related sequence types have been described previously in other African countries. Continued active surveillance and genetic characterization of NmY isolates causing disease in the United States, South Africa, and Israel will provide valuable data for local and global epidemiology and allow monitoring for any expansion of existing clonal complexes and detection of the emergence of new virulent clones in the population.
Neisseria meningitidis is an important cause of morbidity and mortality worldwide and a leading cause of bacterial meningitis and septicemia in children and young adults. Serogroups A, B, and C have caused ca. 90% of the cases of sporadic and epidemic meningococcal disease globally, and N. meningitidis serogroup Y (NmY) has, historically, been responsible for only a minor proportion of meningococcal disease worldwide (9, 14). In the United States, the proportion of meningococcal disease caused by NmY increased from 10% in 1992 to 34% in 1996 with an estimated increase in incidence from 0.1/100,000 population in 1992 to 0.2/100,000 in 1996 (3, 15).
In South Africa, meningococcal disease is endemic but, unlike the patterns of disease in the African meningitis belt, the incidence rates are low, and increases of sporadic disease occur during the winter and spring months (1). Serogroup Y has caused little disease historically, although comprehensive national surveillance data on meningococcal serogroups have been lacking. Recent national laboratory-based surveillance data (1999 to 2002) in South Africa have shown a high proportion (~21%) of meningococcal disease caused by NmY (5). In Israel, the incidence of NmY meningococcal disease rose from 0.05/100,000 population in 1990 to 1991 to 0.16/100,000 in 1998, and the proportion of NmY among all isolates increased from 3% in 1991 to 21% in 1998 (data from the National Center for Meningococci [Israel]).
The goals of the present study were to determine whether a particular genetic lineage of NmY had caused the majority of NmY invasive disease in the United States, South Africa, and Israel and to compare the genotypes of the NmYs found in these countries to the genotypes of NmY found globally. We characterized 300 NmY isolates collected during 1999 through 2002 from the United States, South Africa, and Israel by 16S rRNA gene sequence typing, multilocus sequence typing (MLST), and PorA variable region (VR) typing.
A case of NmY meningococcal disease was defined as NmY isolated from a normally sterile site (primarily blood or cerebrospinal fluid). In the United States, meningococcal isolates were obtained through active, population-based, laboratory-based surveillance as part of the Active Bacterial Core Surveillance project (ABCs) (http://www.cdc.gov/ncidod/dbmd/abcs) representing ca. 10% of the U.S. population. From January 1999 through December 2002, a total of 1,104 cases of meningococcal disease were reported. NmB was isolated in 398 cases (36%), 260 (24%) were caused by NmC, 310 (28%) cases were NmY disease, 24 (2%) isolates were W135, 2 were serogroup X, and 110 (10%) were nongroupable or unknown (ABCs [unpublished data]). Of the 310 NmY cases, 292 (94%) isolates were received at the Centers for Disease Control and Prevention (CDC) from the ABCs sites. Systematic sampling of every second isolate resulted in a set of 144 NmY isolates that were characterized; 17 of these isolates were later determined to have been incorrectly categorized as ABCs isolates and were excluded from further analysis. Therefore, the data for 127 isolates (43%) were used for the final analysis. In South Africa, 962 cases of invasive meningococcal disease were voluntarily reported to a central laboratory from July 1999 through December 2002 as part of a national laboratory-based surveillance program. A total of 701 (73%) cases had isolates available for serogrouping. Of these, 150 (21%) were identified as NmY, and all of these isolates were available for molecular testing. In Israel, all invasive isolates of N. meningitidis are submitted to the National Center for Meningococci in Tel Hashomer. In the period from January 1999 through December 2002, 23 (13%) of the 178 N. meningitidis isolates submitted were NmY, and all 23 NmY isolates were included in the present study. The NmY isolates from the United States, South Africa, and Israel characterized in the present study total 300.
Bacterial isolates from the United States and South Africa were grown on Trypticase soy agar II with 5% sheep blood (Becton Dickinson, Franklin Lakes, NJ) overnight at 37°C in 5% CO2. Bacteria were harvested by using a sterile cotton swab, suspended in sterile 10 mM Tris (pH 8.0) and lysed at 95°C for 10 min before being stored at −20°C. The 23 isolates from Israel were grown on Trypticase soy agar with 5% sheep blood (HyLabs, Rehovot, Israel), harvested, washed, and boiled in sterile distilled water. The boiled suspensions were stored at 4°C and subsequently sent to CDC at ambient temperature. All whole-cell preparations were used in PCR without further purification.
The serogroup of all 300 NmY isolates was confirmed by using the serogroup-specific real-time PCR assay as previously described (12).
All 300 NmY isolates were characterized by MLST as described by Maiden et al. (11). A specific fragment of each of seven housekeeping genes was amplified and sequenced in both forward and reverse orientation. Allele designations were determined by comparison to the existing database at the Neisseria MLST website (http://pubmlst.org/neisseria/). New alleles were assigned a number by the curator of the Neisseria Multi Locus website, Keith Jolley. The designation of the new ST-175 clonal complex was approved by the Neisseria MLST Management Team on 5 June 2009.
All 300 NmY isolates were characterized by 16S rRNA gene sequencing as previously described (17). Briefly, the full-length 16S rRNA gene was amplified by PCR, purified by using a PCR purification kit (Qiagen, Inc., Valencia, CA), and sequenced using 16 primers that span the length of the 16S rRNA gene with at least fourfold coverage, including forward and reverse orientation. 16S types were designated in the order in which they were discovered (17).
The 300 NmY isolates were characterized by PorA VR typing as previously described (16). Briefly, the DNA encoding the PorA VR1 and VR2 regions of the porA gene was amplified by PCR and sequenced in both forward and reverse orientations. Allele designations were determined by comparing the nucleotide or putative amino acid sequence to the PorA database (http://neisseria.org/nm/typing/) at the University of Oxford, Oxford, United Kingdom.
The 69 unique allelic profiles that included the seven MLST allele designations, the 16S type, and the PorA VR1 and VR2 designations were used in eBURST (7) to compute the genotype relationships. One allelic profile did not have PorA designations and was omitted from the analysis. Groups were defined using the strict definition, which classifies a group as members with identical alleles at 9 of 10 loci. DNA sequence data from MLST and 16S rRNA gene sequencing were aligned and edited with the Wisconsin Package (version 10.3; Accelrys, Inc., San Diego, CA). Phylogenetic analysis was performed by using Geneious (A. J. Drummond, B. Ashton, M. Cheung, J. Heled, M. Kearse, R. Moir, S. Stones-Havas, T. Thierer, and A. Wilson, Geneious v3.8.).
Three hundred NmY isolates collected in the United States, South Africa, and Israel during 1999 to 2002 were characterized by MLST, 16S rRNA gene typing, and PorA typing. The results are tabulated in Table Table11.
By MLST, there were 18 different STs found among the 127 NmY isolates, and all isolates belonged to one of 5 different ST complexes. ST-23 and ST-23 clonal complex predominated (80% [n = 102] and 94% [n = 119], respectively). There were 16 different 16S types; 108 (85%) of the isolates contained 16S type 19 (16S-19) (Table (Table2).2). Thirteen PorA types were found among the United States' isolates; 71% (n = 90) of the strains had the PorA VR1 family 5 variant in combination with a member of VR2 family 10, and 25% (n = 32) had the PorA VR1 family 5 in combination with VR2 family 2. PorA types P1.18-1,3, P1.7-2,4-2, and P1.7,16 represented 3% of the 127 isolates.
When the ST, 16S type, and PorA type results were combined, there were 37 distinct combinations (clones) among the 127 isolates from the United States. The predominant clones possessed ST-23/16S-19/P1.5-2,10-1 (n = 72; 57%) and ST-23/16S-19/P1.5-1,2-2 (n = 13; 10%).
Analysis of MLST results of the 150 South African NmY isolates identified 16 STs and 4 known ST complexes (ST-11, ST-23, ST-167, and ST-865 complexes). However, 115 isolates (77%) belonged to four STs that have not been previously defined as part of a complex. ST-4367 (1 isolate) and ST-4669 (1 isolate) were single-locus variants of ST-175 (112 isolates), and ST-4365 (1 isolate) and ST-175 were different at three of the seven loci. They did not match any other central genotype more closely. As a result of these findings and an examination of the Neisseria MLST database, a new ST-175 clonal complex was designated (http://pubmlst.org/neisseria).
There were 20 different 16S types. Seventy-one percent (n = 107) of the South African isolates were 16S type 21 (Table (Table2).2). 16S type 19 was found in 14% (n = 21) of the isolates. The remaining 18 16S types each represented 2% or less of the total.
Nine different PorA types were found among the 150 South African NmY isolates. Seventy-nine percent (n = 119) of the isolates had PorA types with VR1 family 5 and VR2 family 2. Eighteen percent (n = 27) of the isolates were of PorA VR1 family 5 and VR2 family 10. PorA types P1.7-1,1, and P1.12-1,13-1 were found in one and three isolates, respectively (Table (Table11).
When all genetic markers were combined, 102 of the 150 isolates (68%) from South Africa were of ST-175/16S-21/P1.5-1,2-2 (Table (Table1).1). No other clone represented more than 5% of the total isolates from South Africa.
Three STs were found among the 23 Israeli isolates; 21 (91%) of the isolates were ST-23. Four 16S types were found among the Israeli isolates, two of which, types 19 and 126, represented 91% of the isolates (Table (Table11 and and2).2). Over half of the isolates, 13 of 23 (56%), had PorA sequences in the VR1 family 5 and VR2 family 2, and 10 isolates (43%) were from VR1 and VR2 families 5 and 10. One PorA type, P1.5-1,2-13, was unique to Israel and was found in four isolates (17%). Overall, there were four predominant clones that made up 87% of the isolates (Table (Table1).1). Seventeen of these twenty isolates were ST-23/16S-19 combined with P1.5-1,2-2 (five isolates), P1.5-1,2-13 (four isolates), or P1.5-2,10-1 (eight isolates). Three isolates were ST-23/16S-126/P1.5-1,2-2.
When the combined ST, 16S type, and PorA types were considered for each of the 300 isolates, 70 different combinations (clones) were found (Table (Table1).1). The hypervirulent lineages ST-11 complex, ST-41/44, and ST-32 were found in one or more of the three countries but did not contribute more than 4% of the total. Of the 70 different clones, only the clone with the combination ST-23/16S-19/P1.5-2,10-1 was found in all three countries. This common clone accounted for 56% of the isolates studied from the United States, 5% of the South African isolates, and 35% of the Israeli collection. The United States and Israel also shared another clone that had ST-23 and 16S type 19 as the clone above but had a different PorA type, P1.5-1,2-2. This clone was found in 10 and 22% of isolates from the United States and Israel, respectively, and was not found among the South African isolates. One clone, ST-4245/16S-19/P1.5-2,10-1, was found in both South Africa and Israel but not in the United States. The 67 remaining clones were found exclusively in only one of the three countries.
Of the 70 allelic profiles, 69 included complete data for ST, 16S type, and PorA VR1 and VR2 and were analyzed by eBURST (7). When the strictest criterion for a group was used, 9 of 10 alleles shared with at least one other profile in the group, eight groups were formed and 14 profiles were singletons (Fig. (Fig.1)1) The ST-23 profiles were split into four groups, and groups 1 and 2 made up 88 and 96% of the isolates from the United States and Israel, respectively. The genotype of isolates common to the three countries (genotype 23a in Table Table1)1) was the predicted founder of group 1 (bootstrap = 100%), and the clone seen in the United States and Israel (genotype 23m in Table Table1)1) was the founder of the group 2 (bootstrap = 100%). A founder isolate is the profile with the greatest number of single-locus variants. There was no founder predicted for the third and fourth groups. The only other group among the eight that had a bootstrap value of 100% for the predicted founder was the ST-175 group, and the predicted founder was the predominant profile in South Africa (genotype 175a in Table Table11).
A dendrogram was constructed by using the concatenated sequences of the seven MLST gene fragments and the 16S rRNA gene (Fig. (Fig.2).2). The distance from ST-884/16S-3 to ST-4117/16S-19 was 5%. The dendrogram and the eBURST results demonstrate that the ST-175 complex isolates found in South Africa are clearly distinct from and not closely related to isolates of the ST-23 complex that were the predominant agents of serogroup Y disease in the United States and Israel.
The aim of this study was to ascertain and compare genotypic data of NmY isolates causing meningococcal disease in the United States, South Africa, and Israel, where the proportion of NmY disease is higher than the global average. Genotypically, there were 70 different clones among the 300 isolates, as determined by the unique combinations of multilocus ST, 16S type, and PorA type. Overall, the genotypic profiles of NmY isolates in the United States and Israel were similar to each other (ST-23 complex, 16S type 19, and P1.5-1,2-2 or P1.5-2,10-1). This genotypic profile was found in only a minor percentage of the South African isolates. A newly defined clonal complex comprising ST-175 and three closely related STs predominated in the South African NmY population.
NmY isolates in the ST-23 complex have previously been found as carriage isolates in the nasopharynx but were thought to cause invasive disease infrequently (20). In fact, a study of disease-associated and carriage isolates of N. meningitidis collected in The Czech Republic, Greece, and Norway from 1991 to 2000 found a negative association of ST-23 with disease (22). More recent studies, however, have documented the isolation of NmY isolates of the ST-23 complex from patients with invasive meningococcal disease in Taiwan, Canada, and Italy (4, 6, 18). The proportion of meningococcal disease caused by NmY in Taiwan and Italy, however, remains below 10%. In contrast, the proportion of NmY disease increased in Canada from 8.4% in 1999 to 21.5% in 2003. And 65.7% of the NmY isolates were from the ST-23 clonal complex. In the present study, the high proportion of serogroup Y meningococcal disease in the United States and Israel was due to ST-23, whereas ST-23 accounted for only 15% of the NmY isolates collected in South Africa.
In a recent report, Coulson et al. described the clonal population structure of NmY in South Africa by means of pulsed-field gel electrophoresis typing (5). The same South African NmY strains were further characterized in the present study using MLST, 16S typing, and PorA VR typing. The majority of the isolates identified by pulsed-field gel electrophoresis as belonging to the clonal cluster Y-1 in the previous study were genotyped as ST-175/16S-21/P1.5-1,2-2. Similarly, the strains identified as belonging to cluster Y-2 in the previous study were ST-23/16S-19/P1.5-2,10-1. These recent data strongly support the previous finding of a clonal population structure of NmY disease in South Africa, and determined that strains belonging to the ST-175 clonal complex were responsible for three-fourths of NmY disease during this period.
The newly designated ST-175 clonal complex includes 32 STs and 110 isolates in the MLST database. Using the strictest criteria for a clonal complex (six of seven alleles identical), whereby all STs are thought to descend from a common founder, there are 17 STs in the MLST database that group with ST-175, including ST-2881, and two STs that were observed in the present study, ST-4367 and ST-4669. All of the 13 STs are single-locus variants of ST-175, indicating that ST-175 may be the founder ST of the group. There are 38 isolates, mainly serogroups W135 and NmY, listed as ST-2881 in the MLST database from the West African countries of Niger, Benin, Burkina Faso, Chad, Togo, and Cameroon. The NmY isolates were from carriers, while many of the W135 isolates were invasive. W135 ST-2881 has caused both sporadic and a cluster of meningitis cases in Niger (2). Based on genotypic data of carriage and invasive isolates from Niger in 2003, investigators proposed that a capsule switch most likely occurred in the ST-2881 isolates of serogroups W135 and Y (2, 13). Evidence of capsule switching among isolates of ST-175 also exists as both serogroup W135 and Y isolates from Gambia and Rwanda have been deposited in the Neisseria PubMLST database. An isolate of ST-4942 that differs from ST-175 at only one locus was isolated in The Netherlands in 1970. Therefore, ST-175 and some very close relatives have been circulating at least for the past 30 years and have recently caused disease in several countries in Africa.
Previously, congruence between particular outer membrane protein sequences (PorA, PorB, and FetA) and ST complexes in the major hypervirulent lineages of N. meningitidis was observed (19). Urwin et al. concluded that simple clonal and epidemic clonal models of population structure were inadequate to explain the observed strain structure (19). In our study, we observed a similar congruence between PorA types and clonal complexes. Usually, a PorA type was found among members of a single clonal complex. The exceptions were P1.5,2 that was found among ST-11 and ST-23 clonal complexes, P1.5-1,2-2 that was found among ST-23 and ST-175 clonal complexes, and P1.5-1,10-4 that was found among ST-167 and ST-175 clonal complexes. The clonal complexes that shared PorA types were not found in the same country. Urwin's proposal of a host immunity model of strain structuring whereby shared immunological variants are disadvantaged in the same transmission system (e.g., country) may apply here (19).
Antigenic shift in PorA from P1.5-1,2-2 to P1.5-2,10-1 among NmY isolates from Maryland in the 1990s, as noted by Harrison et al. (10), was confirmed for NmY isolates in the present study collected through the ABCs surveillance program from other states in the United States. Of the 127 isolates (64%) from the United States characterized here from 1999 to 2002, including 9 isolates from Maryland, 81 had PorA type P1.5-2,10-1, and all but one were in the ST-23 complex. The ratio of P1.5-1,2-2 to P1.5-2,10-1 in the NmY ST-23 complex population studied previously by our laboratory decreased from 2.75:1 in 1992 to 0.4:1 in 1998 (21). The ratio remained below 0.5:1 through 2002 (the present study) (Table (Table1).1). A similar change in PorA type has been observed over the same time period in Israel. In our previous study, 47% of the isolates from Israel were P1.5-1,2-2, 33% were P1.5-1,2-13, and 10% were P1.5-2,10-1 (21). The Israeli isolates from 1990 to 2002 showed a decrease to 34% in PorA type P1.5-1, 2-2, a decrease to 17% of P1.5-1,2-13, and an increase in P1.5-2,10-1 to 43%. Historic data were insufficient for a similar analysis of NmY isolates from South Africa. These trends are indicative of cyclical fluctuations of antigenic variants and are predicted by the strain structure model of nonoverlapping antigenic combinations dictated by immune-mediated competition (8, 19).
In summary, 70 different clones were found in a population of NmY isolates collected in the United States, Israel, and South Africa, the high proportion of serogroup Y meningococcal disease in the United States and Israel was caused by genotypically distinct clones of N. meningitidis (ST-23 clonal complex) from those causing disease in South Africa (ST-175 clonal complex). ST-175 appears to be the founder of a clonal complex that has recently caused disease in other countries in Africa. Continued active surveillance and genetic characterization of invasive and carriage NmY isolates in the United States, Israel, and South Africa will provide valuable data for local and global epidemiology and allow monitoring for the expansion of existing clonal complexes and detection of the emergence of new virulent clones in the population.
We gratefully acknowledge the Group for Enteric, Respiratory and Meningeal Disease Surveillance in South Africa (GERMS-SA) for their efforts in collecting the South African isolates, and the Active Bacterial Core Surveillance (ABCs)/Emerging Infections Program (EIP) Network for the U.S. isolates in this study. We also acknowledge the contributions of Gwen Barnett, Jordan Theodore, Susanna Schmink, Travis Wheeling, and Linda de Gouveia for data used in this study; Claudio Sacchi for critical review of the manuscript; and Tanja Popovic (CDC) for her leadership.
Published ahead of print on 1 July 2009.