Genetic characteristics of serogroup  A meningococci circulating in China, 1956–2005

Authors

  • X. Zhang,

    1. State Key Laboratory for Molecular Virology and Genetic Engineering, Institute of Pathogen Biology, Chinese Academy of Medical Sciences
    Search for more papers by this author
  • Z. Shao,

    1. National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, China
    Search for more papers by this author
  • Y. Zhu,

    1. State Key Laboratory for Molecular Virology and Genetic Engineering, Institute of Pathogen Biology, Chinese Academy of Medical Sciences
    Search for more papers by this author
  • L. Xu,

    1. National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, China
    Search for more papers by this author
  • X. Xu,

    1. State Key Laboratory for Molecular Virology and Genetic Engineering, Institute of Pathogen Biology, Chinese Academy of Medical Sciences
    Search for more papers by this author
  • L. W. Mayer,

    1. Meningitis and Vaccine Preventable Diseases Branch, Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, GA, USA
    Search for more papers by this author
  • J. Xu,

    1. National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, China
    Search for more papers by this author
  • Q. Jin

    1. State Key Laboratory for Molecular Virology and Genetic Engineering, Institute of Pathogen Biology, Chinese Academy of Medical Sciences
    Search for more papers by this author

Corresponding author and reprint requests: Q. Jin, State Key Laboratory for Molecular Virology and Genetic Engineering, Institute of Pathogen Biology, Chinese Academy of Medical Sciences, Beijing 100176, China
E-mail: zdsys@sina.com

Abstract

Neisseria meningitidis serogroup  A accounted for 95% of cases of meningococcal disease in China during the last century. To understand the circulation of these organisms in China over a 50-year period, 275 serogroup  A meningococcal isolates collected between 1956 and 2005 were characterised by multilocus sequence typing (MLST) and PorA typing. In total, 44 sequence types (STs), belonging to five hyperinvasive lineages, and ten singletons were identified in this collection. The ST-5 complex and the ST-1 complex represented 52.8% (86/163) and 44.2% (72/163), respectively, of isolates from cases of infection and, overall, 93.1% (256/275) of all isolates. Three prevalent clones (ST-5, P1.5-2,10; ST-3, P1.7-1,10; and ST-5, P1.20,9) were involved in four national epidemics in 1959, 1967, 1977 and 1984. ST-5 was replaced by ST-7 in the late 1980s, such that ST-7 isolates with P1.20,9 represented >86% of isolates from cases of infection after 2000. The data also revealed that the collection contained 19 PorA VR types, of which P1.7-1,10 and P1.20,9 were the predominant types in the ST-1 and ST-5 common lineages, respectively. Three other hyperinvasive lineages (ST-11 complex, ST-32 complex and ST-4821 complex) were isolated only from carriers. It was concluded that serogroup  A meningococci of the ST-5 complex and the ST-1 complex were responsible for most cases of meningococcal disease in China during the past 50 years.

Introduction

Bacterial meningitis caused by Neisseria meningitidis is a serious threat to global health, accounting for c. 100 000 cases worldwide each year, despite the existence of partially effective vaccines [1]. It is now known that >90% of cases of invasive disease are caused by five of 13 serogroups of N. meningitidis, and that the distribution of these serogroups differs among geographical regions [2]. Thus, serogroup  A strains are predominant in Africa and Asia, whereas serogroups  B, C, W135 and Y predominate in Europe, Australia and the Americas [3].

In the last century, serogroup  A meningococci were responsible for >95% of cases of meningococcal disease in China, whereas serogroups B and C caused only sporadic cases [4]. Several pandemic waves and national epidemics were associated with a single clonal group of serogroup  A meningococci, called subgroup III, which is still circulating in Africa and Asia [5–7]. The cyclic nature of epidemic meningitis, with peaks every 8–10 years, and an annual disease rate of up to 500 cases/100 000 inhabitants, resulted in several million cases of meningococcal disease in China between 1950 and 1980. Following the initiation of a national immunisation campaign during the early 1980s, the morbidity rate of meningococcal disease declined to 0.2–1.0/100 000. Starting in 2003, several outbreaks of serogroup  C meningococcal disease, associated with the emergence of a new hyperinvasive lineage (ST-4821 complex), caused an increase in the number of serogroup  C cases [8]. However, despite the emergence of serogroup  C meningococci, serogroup  A meningococci have remained an important cause of meningococcal disease in China throughout the last half century.

In previous studies, epidemic clones of serogroup  A meningococci were identified on the basis of genotypic characteristics, e.g., electrophoretic types, restriction fragment-length polymorphism types and genoclouds [4,6]. Few data concerning isolates from cases of infection collected during the inter-pandemic periods or the distribution of genotypes among isolates from carriers are available. In the present study, 275 isolates of serogroup  A meningococci collected in China during the past 50 years were characterised by multilocus sequence typing (MLST) and PorA variable region (VR) typing. MLST is currently the approach used most widely to catalogue genetic variation among meningococci, and characterises each isolate according to its sequence type (ST) [9]. This nucleotide sequence-based technique gives unambiguous and highly portable results that allow a high level of discrimination among isolates. Meningococci form clusters of related STs, referred to as clonal complexes, consisting of a founder ST and its variants, while PorA VR typing characterises the antigenic diversity of a surface-exposed porin that is putatively under immune selective pressure. The present study expands the information available concerning isolates from recent epidemics of serogroup  A disease in China, and is also the first clonal analysis of isolates collected from carriers in China during the periods between pandemics.

Materials and methods

Bacteria

A convenience sample of 275 serogroup  A meningococcal isolates obtained from cases or carriers in 26 provinces of China between 1956 and 2005 was studied. Invasive cases (i.e., isolates obtained from normally sterile sites such as cerebrospinal fluid or blood) accounted for 163 (59%) of the isolates collected during the 50-year period, which included four national epidemics of serogroup  A meningitis in 1959, 1967, 1977 and 1984. The second and the fourth Chinese epidemic waves occurred at the same time as the first and second pandemics [4,5]. Another 112 isolates were obtained from healthy carriers during the same period.

MLST

Isolates were characterised by MLST as described previously [9], with an ABI Prism 3730 automated sequencer (Applied Biosystems, Warrington, UK) being used to separate the labelled extension products. The data were assembled using SeqMan  II of the DNASTAR program. Alleles, STs and clonal complexes were assigned by reference to the Neisseria MLST database (http://pubmlst.org/neisseria).

PorA VR typing

Fragments of porA genes (1.2 kb) were amplified and sequenced as described previously [10,11] from all except two isolates. Isolates were assigned to particular families of the VR1 and VR2 regions by reference to the PorA typing website (http://www.neisseria.org/nm/typing/pora).

Clonal complex analysis

Assignment of isolates to a clonal complex was performed by identifying a founder or central genotype with the BURST algorithm, implemented in the computer programs START and SPLITSTREE [12,13]. Relationships among STs were also analysed using eBURST [14]. All of the programs used in this study are available for electronic download (http://mlst.zoo.ox.ac.uk; http://bibiserv.techfak.uni-bielefeld.de/splits; http://evolgen.biol.metro-u.ac.jp/MEGA/).

Results

STs and lineages found in the collection

There were 44 STs among the 275 isolates typed by MLST, 35 of which were novel and represented 15% (40/275) of the collection (Table 1). The other nine STs accounted for 85% (235/275) of isolates. Three STs (ST-3, ST-5 and ST-7) accounted for 80.7% (222) of the total isolates.

Table 1.   Characteristics of serogroup  A meningococci isolated in China, 1956–2005
LineageSTporA VR typeNo. of isolatesDate (n)
  1. ST, sequence type; NA, not amplified.

ST-1 complex/subgroup I/IIST-3P1.7-1,10Patient (57)1972 (1); 1973 (5); 1974 (9); 1976 (1); 1977 (1); 1978 (4); 1979 (21); 1980 (5); 1982 (5); 1984 (1); 1985 (3); 2004 (1)
Carrier (24)1974 (2); 1978 (1); 1979 (9); 1980 (1); 1982 (3); 1984 (5);1985 (1);1987 (1); 1991 (1)
P1.7-1,10-3Patient (2)1978 (1); 1982 (1)
P1.7-3,10Patient (2)1976 (2)
Carrier (8)1979 (6);1985 (1); 1986 (1)
P1.7-4,10Patient (2)1985 (1); 1986 (1)
P1.12-6,4-1Patient (1)1984 (1)
ST-4991p1.18-7,34-2Carrier (1)1979 (1)
ST-5080P1.7-1,10Carrier (1)1978 (1)
ST-5239P1.7-1,10Patient (1)1974 (1)
ST-5240P1.7-1,10Patient (1)1974 (1)
ST-5241P1.7-1,10Carrier (1)1978 (1)
ST-5242P1.7-3,10Patient (1)1979 (1)
ST-5243P1.7-1,10Patient (1)1974 (1)
ST-5245P1.7-4,10Patient (1)1978 (1)
ST-5249P1.7-1,10Patient (1)1980 (1)
ST-5250P1.7-1,10Patient (1)1980 (1)
ST-5254P1.7-1,10Carrier (1)1984 (1)
ST-5255P1.7-1,10Carrier (1)1977 (1)
ST-5258P1.7-1,10Patient (1)1985 (1)
ST-5464P1.7-1,10Carrier (1)1984 (1)
ST-5 complex/subgroup IIIST-5P1.5-2,10Patient (3)1956 (1); 1963 (1); 1973 (1)
Carrier (4)1963 (4)
P1.20,9Patient (36)1966 (1); 1977 (1); 1978 (1); 1983 (2); 1984 (10); 1985 (11); 1986 (4); 1987 (4); 1988 (1); 1996 (1)
Carrier (29)1966 (2); 1980 (4); 1982 (1); 1984 (11); 1985 (6); 1986 (1); 1987 (3); 2000 (1)
ST-7P1.20,9Patient (32)1984 (2); 1987 (2); 1991 (1); 1994 (2); 1995 (3); 2000 (1); 2004 (2);2005 (19)
Carrier (17)1984 (3); 1986 (1); 2000 (1); 2005 (12)
P1.20-1,9Patient (1)2005 (1)
P1.20-3,9Patient (3)2002 (3)
Carrier (1)1991 (1);
ST-203P1.20,9Patient (1)1984 (1)
ST-4898P1.20,9Carrier (1)2004 (1)
ST-5079P1.5-2,10Carrier (3)1963 (3)
ST-5082P1.20,9Patient (2)2005 (2)
ST-5083P1.20,9Patient (1)2005 (1)
ST-5084P1.20,9Carrier (1)2005 (1)
ST-5114P1.20-1,9Carrier (1)2005 (1)
ST-5246P1.20,9Patient (2)1983 (1); 1985 (1)
ST-5247P1.20,9Patient (1)1985 (1)
ST-5251P1.20,9Patient (3)1966 (1); 1975 (1); 1984 (1)
Carrier (1)1987 (1)
ST-5256P1.20,9Patient (1)1985 (1)
Carrier (1)1984 (1)
ST-5466NACarrier (1)2005 (1)
ST-11/ ET-37 complexST-11P1.5-1,2-2Carrier (2)1984 (2)
ST-32/ET-5 complexST-1784P1.7,16Carrier (1)1979 (1)
ST-4821 complexST-3200P1.20,23Carrier (1)2005 (1)
ST-4821P1.20, 23-7Carrier (1)2005 (1)
ST-4897P1.21-2,28Carrier (1)2005 (1)
ST-59ST-59P1.7-1,10Patient (2)1980 (1); 1982 (1)
Carrier (2)1980 (2)
ST-685ST-685P1.5-1,2-2Carrier (1)1992 (1)
ST-5113ST-5113P1.21,23-10Carrier (1)2005 (1)
ST-5244ST-5244P1.7-1,10Patient (1)1974 (1)
ST-5248ST-5248P1.19,13Carrier (1)1986 (1)
ST-5252ST-5252P1.22,14-6Patient (1)1977 (1)
ST-5253ST-5253P1.12-13,13-20Patient (1)1984 (1)
ST-5257ST-5257NACarrier (1)1978 (1)
ST-5294ST-5294P1.12-6,4-1Carrier (1)1985 (1)
ST-5465ST-5465P1.20,9Carrier (1)1993 (1)

Cluster analysis revealed that there were five clonal complexes containing 34 of the STs present in the collection (Table 1). The ST-1 and ST-5 complexes were the two most common complexes, containing 93.1% (256/275) of the isolates. The ST-1 complex contained 15 STs, with 11 single-locus variants (SLVs) and three double-locus variants of ST-3, including 14 STs identified for the first time in the present study. The most common complex was the ST-5 complex, consisting of two predominant STs (ST-5 and ST-7). There were four SLVs of ST-5 and eight SLVs of ST-7, with ST-7 being an SLV of ST-5 in the ST-5 complex. Three complexes (ST-11 complex, ST-32 complex and ST-4821 complex) contained a total of only five STs, accounting for 2.2% (6/275) of the total isolates, all of which were from carriers. Ten STs did not belong to any known clonal complex and accounted for 4.7% (13/275) of the isolates.

Among 163 isolates from cases of infection, 96.9% (158/163) belonged to either the ST-1 complex (nine STs) or the ST-5 complex (nine STs); the remaining five isolates belonged to four STs that were not in known complexes (Tables 1 and 2). The 112 isolates from healthy carriers were slightly more diverse, belonging to five clonal complexes and singletons, with 98 (87.5%) isolates belonging to either the ST-1 complex (seven STs) or the ST-5 complex (nine STs), and 14 isolates belonging to other clonal complexes (ST-11, ST-32, and ST-4821 complexes) and seven singletons (eight isolates). There were 22 STs and 28 STs represented among isolates from cases and carriers, respectively; six STs (ST-3, ST-5, ST-7, ST-59, ST-5251 and ST-5256) were seen among isolates from both cases and carriers.

Table 2.   Temporal distribution of sequence types (STs) and lineages found among isolates from cases of infection (bold figures indicate that a high frequency of a particular ST was detected compared with the other periods)
PeriodaST-1 complexST-5 complexOthersNo. of isolates
ST-3VariantsST-5ST-7Variants
  1. aThe periods shown are those listed by Zhu et al. [6].

  2. Values in parentheses are percentages.

Before 1965002 (100)0002
1965–1970 (first pandemic)001 (50)01 (50)02
1971–198149 (76.6)8 (12.5)3 (4.7)01 (1.5)3 (4.7)64
1982–1988 (second pandemic)13 (22.4)1 (1.7)32 (55.2)4 (6.9)6 (10.3)2 (3.5)58
1989–19920001001
1993–2000 (third pandemic)001 (14.3)6 (85.7)007
After 20001 (3.4)0025 (86.2)3 (10.4)029
Total6393936115163

Diversity of housekeeping genes

The total number of alleles at each locus for the 275 isolates ranged from five for adk to 16 for aroE. The total number of alleles at other loci included 12 for abcZ, 15 for fumC and pgm, nine for gdh, and 14 for pdhC. Of these 86 alleles, 15 were identified for the first time in this study: four (abcZ-331, abcZ-333, fumC-368, gdh-379) were present in SLVs of ST-3; three (aroE-382, aroE-383, pdhC-354) were present in SLVs of ST-5; and eight (abcZ-338, aroE-402, fumC-361, fumC-369, pdhC-362, pdhC-370, pgm-358 pgm-366) were present in singletons.

To illustrate the diversity of alleles within the two most common lineages, Fig. 1 shows an analysis of the alleles at each locus for the ST-5 complex. The generation of variants of ST-1 was accounted for by 16 variant alleles at five loci (three abcZ, three adk, two aroE, four fumC, two gdh and two pdh variants). All three double-locus variants of ST-1 had variant alleles in fumC. For the ST-5 complex, 12 variant alleles were detected at six loci, two of which contained more than half of these variants (four aroE and three pgm alleles; Fig. 1).

Figure 1.

 Diversity of isolates from China belonging to the ST-5 complex.

PorA VR types among isolates

Sequencing and PorA VR analysis of 273 isolates revealed 19 different PorA VR types (Table 1). The ST-1 complex contained six types, with P1.7-1,10 being the most prevalent, representing 83.6% (92/110) of the isolates in the ST-1 complex. Three variants of this family (P1.7-4,10, P1.7-1,10-3 and P1.7-3,10) accounted for all but two of the remaining isolates in the ST-1 complex. P1.12-6,4-1 and P1.18-7,34-2 were each represented in the ST-1 complex by a single isolate. Only four types (P1.5-2,10, P1.20,9, P1.20-1,9 and P1.20-3,9) were found among the 146 isolates in the ST-5 complex, of which P1.20,9 was the most prevalent type, representing 89.4% (129/146) of the isolates in this complex. There was only slightly more diversity among isolates from carriers (15 PorA types among 112 isolates, or one type for every 7.5 isolates) than among isolates from cases of invasive disease (11 PorA types among 163 isolates, or one type for every 14.8 isolates; p 0.06) (Table 1).

The remaining isolates were present as only one or two isolates/subtype. Only three PorA types that were seen in the ST-1 or ST-5 complexes were also present among the remaining isolates. The porA gene of two isolates (ST-5257 and ST-5466) could not be amplified.

Invasive clones

Most of the cases of invasive disease were caused by only three clones, with ST-3, P1.7-1,10 of the ST-1 complex, ST-5, P1.20,9 of the ST-5 complex, and ST-7, P1.20,9 of the ST-5 complex accounting for 76.7% (125/163) of all cases. ST-7 is an SLV of ST-5. These three clones also accounted for a significant proportion (62.5%) of all carrier isolates (p 0.01).

eBURST cluster analysis

Fig. 2 summarises the results of the eBURST cluster analysis of the ST-1 complex. Based on the data contained in the MLST database, ST-3 and its variants seem to have emerged in China, whereas the other STs belonging to this complex were more closely related and were isolated elsewhere.

Figure 2.

 eBURST analysis of the ST-1 complex showing the geographical distribution of different sequence types (STs) belonging to this complex.

Discussion

Molecular characterisation of meningococcal isolates is crucial for analysis of the bacterial factors that modulate the epidemic behaviour of N. meningitidis. Molecular typing methods based on DNA sequencing data, e.g., MLST and PorA VR typing, are currently the standard methods for epidemiological analysis of meningococci, as they allow a high level of discrimination among isolates. At the time of the present study, the MLST database (http://pubmlst.org/neisseria) contained >6000 genotypes, represented by isolates from global sources. Analysis of PorA VR types shows the antigenic diversity of N. meningitidis and helps to elucidate the molecular character of the cell surface antigens, reflecting the influence of host immunity on population structure [15].

The present study investigated the distribution and changes in the lineages of serogroup  A meningococci prevalent in China during the last 50 years. The study revealed that two common lineages prevailed in China between 1956 and 2005. One common lineage of the ST-5 complex has persisted continuously in China for at least the last 50 years. As only four isolates from before 1971 were available for subtyping, it is difficult to conclude that this was the predominant genotype isolated from cases of invasive disease at that time, even though all four isolates belonged to the ST-5 complex. In contrast, isolates belonging to ST-3 and variants of the ST-1 complex predominated during the inter-pandemic period between 1971 and 1981. The circulation of this lineage continued in a few provinces (Guangdong, Guangxi, Guizhou, Jiangxi) in the southeast of China during the early 1980s, but few isolates belonging to the ST-1 complex have been recognised since 1989.

The ST-5 complex/subgroup III appears to have been an important lineage in China for 50 years, and still remains the predominant lineage of serogroup  A. Two common STs of this hyperinvasive lineage were associated with a series of outbreaks of serogroup  A disease, including the outbreaks during 1984 and 1994 (Table 2). ST-5 was probably the predominant genotype before 1970 and during the 1980s, causing >50% of cases of infection during these two periods, which included three countrywide epidemics in 1959, 1967 and 1984 [5]. However, although ST-5 isolates predominated during these three epidemics, the PorA VR types varied in each epidemic cycle. P1.5-2,10 was isolated mainly before 1973, and was represented by seven of ten ST-5 isolates collected during 1956–1973. Another common PorA VR type, P1.20,9, was prevalent after 1977, but seems to have first emerged around 1966. ST-5, P1.20,9 isolates were seen in two epidemic waves during the 1960s and 1980s.

ST-7 strains with P1.20.9 emerged as the most common genotype after 1989, representing the majority of isolates from cases of infection. Of 36 disease-associated isolates after 1989, 32 (89%) belonged to this ST/PorA combination. As shown in Table 2, the increase in the proportion of ST-7 isolates occurred after 1989, and ST-7 seems to have become the predominant genotype in China during the 1990s. The emergence of ST-7 in the African meningitis belt also occurred during this period [7,16].

Serogroup  A meningococci belonging to the ST-1 complex/subgroup I/II have circulated worldwide [17]. The first subgroup I strain recognised was isolated in the UK in 1941. At the beginning of the 1960s, this lineage was identified in North Africa and in countries of the African meningitis belt, and subsequently caused outbreaks of disease in the Americas (Brazil, USA and Canada), Europe and Australia. It remained the predominant serogroup  A clone in South Africa in the late 1990s [18]. The present data revealed that 15 members of the ST-1 complex, including ST-3, the common Chinese epidemic ST and its variants, had emerged in China, but that they were slightly different from the prevalent genotypes seen in other countries. An eBURST cluster analysis (Fig. 2) suggested that the ST-3 cluster appeared to have emerged in China, and that there were closer relationships among other members of this complex that occurred elsewhere. Isolates belonging to these STs have similar types of outer-membrane proteins (PorA, FetA) [19]; the most common combination is P1.5-2,10/F5-1, whereas P1.7-1,10/F5-5 was seen in some ST-3 isolates.

The ST-1 complex was prevalent in countries neighbouring China, e.g., India, The Philippines and Pakistan, during the 1960s, but there is no evidence that these strains, other than ST-3, spread to China [19]. ST-3 was the predominant ST, representing up to 77% of isolates from cases of infection between 1971 and 1981 in China. It is striking that fewer ST-3 cluster isolates were obtained before 1970 and after 1988, and that the ST-3 cluster has nearly disappeared today. When compared to the long-term prevalence of the ST-5 complex, the emergence and disappearance of the ST-3 cluster was only a short episode during the natural history of serogroup  A meningococci in China in the last 50 years. P1.7-1,10 was the commonest PorA VR type found among the ST-3 isolates predominant in China since the 1970s, accounting for 83.6% (92/110) of the ST-1 complex/subgroup isolates. Immune selection against P1.7-1,10 may have caused the ST-3 cluster to disappear from the population.

MLST and PorA VR typing have not only provided a wealth of information about the serogroup  A meningococci prevalent in China, but have also revealed the extent of microevolution within the meningococcal population. Among the isolates investigated, 21 (75%) of 28 variants of the prevalent clone were generated by replacing alleles of the prevalent clone at one or two loci, whereas the others represented the emergence of novel alleles. Furthermore, allelic replacement often occurred among variants of disease isolates, whereas novel alleles emerged in those variants represented by isolates from carriers. This suggests that different mechanisms are involved in the evolution of disease-associated and carrier strains, although these collections represented different combinations of polymorphisms from a common gene pool [20].

In conclusion, the present study investigated isolates from both cases of disease and healthy carriers, and revealed the general features of serogroup  A meningococci that have been prevalent in China in the past 50 years. The study also revealed that three clones (ST-5 with P1.5-2,10, ST-3 with P1.7-1,10, and ST-5 with P1.20,9) were the cause of the four previous epidemics in China, and that ST-7 with P1.20,9 is the representative clone of serogroup  A meningococci that is currently found in China. In addition, the molecular characterisation of meningococcal isolates from healthy carriers provided useful information about the evolution of these organisms.

Acknowledgements

X. Zhang and Z. Shao contributed equally to this study. This work was funded by a grant (Key Technologies R&D Program 2005BA711A09) from the Ministry of Science and Technology, People’s Republic of China. This study made use of the Neisseria Multi-Locus Sequence Typing website, developed by K. Jolley and M.-S. Chan at the University of Oxford, UK, funded by the Wellcome Trust and the European Union. No information has been provided by the authors concerning the existence or absence of any conflicting or dual interests.

Ancillary