Genotypes of the circulating rotavirus strains in the seven prevaccine seasons from September 2000 to August 2007 in South Korea

Authors

  • H. S. Jeong,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • K. B. Lee,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • A.-Y. Jeong,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • M. Y. Jo,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • S. Y. Jung,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • J. H. Ahn,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • Y. Jee,

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author
  • J. Kim,

    1.  Department of Pediatrics, College of Medicine, Catholic University of Korea, St Vincent Hospital, Suwon, South Korea
    Search for more papers by this author
  • D.-S. Cheon

    1.  Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, Korea Centers for Disease Control and Prevention, Seoul
    Search for more papers by this author

Corresponding author and reprint requests:D.-S. Cheon, Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, National Institute of Health, 5 Nokbun-dong, Eunpyung-gu, Seoul 122-701, South Korea
E-mail: cheonds@hanmail.net

Abstract

Clin Microbiol Infect 2011; 17: 232–235

Abstract

A Korean nationwide surveillance on circulating rotavirus strains was conducted from September 2000 to August 2007 aiming to obtain prevaccine data for predicting vaccine effectiveness. The predominant strains among the 2779 strains analyzed varied annually and only approximately 50% had either a G or a P antigen present in both RotaTeq (Merck & Co. Inc., Whitehouse Station, NJ, USA) and Rotarix (GlaxoSmithKline, Brentford, UK).

Group A rotaviruses (RV-A) are the leading cause of severe acute gastroenteritis among infants and young children. In developing countries, severe diarrhoea caused by human RV-A results in an estimated 600 000 childhood deaths annually; worldwide, this disease results in 2 million hospitalizations [1–3]. Although the RV-A infection-associated mortality is relatively low in industrialized countries, RV-A diarrhoea remains a major reason for visits to paediatric clinics and emergency departments, as well as inpatient hospitalizations.

Two new RV-A vaccines, Rotarix (GlaxoSmithKline, Brentford, UK) and RotaTeq (Merck & Co. Inc., Whitehouse Station, NJ, USA), have been licensed in approximately 100 countries worldwide and are already part of the national vaccination schedules of several countries [4,5]. RotaTeq, introduced in South Korea (hereafter referred to as Korea) in September 2007, is a pentavalent vaccine constructed by introducing common human RV-A serotype antigen genes (G1–G4, P[8]) into a bovine RV-A parent strain to create five different reassortant strains [4]. Rotarix is a monovalent vaccine based on an attenuated genotype G1P[8] strain that is designed to provide serotype-specific and heterotypic protection against common RV-A serotypes; it was launched in July 2008 in Korea.

RV-A belong to the family Reoviridae. Viral particles are nonenveloped, and triple-layered protein capsids enclose the genome of 11 dsRNA segments. The outer layer of the viral capsid is composed of two structural proteins: VP4, a protease-cleaved or P protein, and VP7, a glycoprotein or G protein [1]. These proteins carry the major antigenic determinants, which elicit neutralization antibodies, form the basis for the dual molecular classification scheme that indicates the viral G and P serotypes, and play key roles in the development of protective immunity [1]. Because VP4 and VP7 genes reassort independently from one another during mixed infection, the G and P genotypes are monitored during surveillance studies. Such surveillance studies help to identify the most important strains in circulation before the introduction of RV-A vaccines and to evaluate the effectiveness of the vaccines against the common genotypes.

The Korea Centers for Disease Control and Prevention, in collaboration with 16 laboratories of local public health institutes and participants (> 100) in a sentinel hospital, initiated an agent surveillance system for acute gastroenteritis in 1999.

Faecal samples testing positive for the group A RV-A antigen by antigen-capturing ELISA (IDEIA Rotavirus; Dako Diagnostics, Ely, UK) were collected for genotyping using reverse transcription-PCR, as described previously [6]. From September 2000 until August 2007, faecal samples from 164 081 patients with acute gastroenteritis were analyzed. Of these samples, 19 845 (12%, range 9.1%–15%; 2000–2001 to 2006–2007) were positive for RV-A by ELISA. The peak seasons of RV-A infection were winter and spring, from December to May (Fig. 1). During the first two seasons, the highest number of RV-A infections was detected in April, whereas the peak was in February, and a marked number of infections was detected in May, from the 2002–2003 season onward.

Figure 1.

 Temporal distribution of rotavirus infections from September 2000 to August 2007 in South Korea.

Using RT-PCR, we determined the G and P genotypes of 2779 RV-A strains received from around the country. Overall, 23.0% of the strains from the period 2000–2007 were G1P[8], 18.8% were G3P[8], 18.3% were G2P[4] and 15.9% were G4P[6] as the neonatal strain (Table 1); these four strains represented 80.1% of all the strains analyzed. There was considerable genotype diversity among strains, and the incidence of predominant strains fluctuated annually.

Table 1.   Distribution of rotavirus genotypes in seven consecutive seasons of rotavirus infection from 2000 to 2007 in South Korea
Strain group, genotypeNumber (%) of strains by season of rotavirus infection
2000–01 (= 318)2001–02 (= 463)2002–03 (= 433)2003–04 (= 332)2004–05 (= 399)2005–06 (= 280)2006–07 (= 554)Total (= 2779)
  1. Bold indicates a significant (p <0.001) change in the genotype distribution from one year to the next.

Common human strains222 (69.81)316 (68.25)309 (71.36)262 (78.92)274 (68.67)188 (67.14)359 (64.80)1,930 (69.45)
G1P[8]104 (32.70)143 (30.89)79 (18.24)39 (11.75)52 (13.03)72 (25.71)289 (52.17)778 (28.00)
G2P[4]110 (34.59)145 (31.32)54 (12.47)62 (18.67)90 (22.56)37 (13.21)11 (1.99)509 (18.32)
G3P[8]6 (1.89)13 (2.81)124 (28.64)155 (46.69)108 (27.07)71 (25.36)45 (8.12)522 (18.78)
G4P[8]2 (0.63)11 (2.37)10 (2.31)0 (0)6 (1.50)1 (0.36)5 (0.90)35 (1.26)
G9P[8]0 (0)4 (0.86)42 (9.70)6 (1.81)18 (4.51)7 (2.50)9 (1.62)86 (3.09)
Reassortants common human strains14 (4.40)4 (0.86)1 (0.23)6 (1.81)28 (7.02)9 (3.21)22 (3.97)84 (3.02)
G1P[4]7 (2.20)2 (0.43)1 (0.23)0 (0)3 (0.75)3 (1.07)8 (1.44)24 (0.86)
G2P[8]6 (1.89)1 (0.22)0 (0)1 (0.30)5 (1.25)1 (0.36)12 (2.17)26 (0.94)
G3P[4]0 (0)0 (0)0 (0)2 (0.60)10 (2.51)2 (0.71)0 (0)14 (0.50)
G4P[4]1 (0.31)1 (0.22)0 (0)3 (0.90)10 (2.51)3 (1.07)1 (0.18)19 (0.68)
G9P[4]0 (0)0 (0)0 (0)0 (0)0 (0)0 (0)1 (0.18)1 (0.04)
Zoonotic53 (16.67)109 (23.54)108 (24.94)55 (16.57)71 (17.79)55 (19.64)69 (12.45)520 (18.71)
G1P[6]21 (6.60)6 (1.30)3 (0.69)5 (1.51)0 (0)3 (1.07)1 (0.18)39 (1.40)
G2P[6]2 (0.63)3 (0.65)2 (0.46)0 (0)8 (2.01)1 (0.36)0 (0)16 (0.58)
G2P[9]0 (0)0 (0)0 (0)0 (0)0 (0)0 (0)1 (0.18)1 (0.04)
G3P[6]1 (0.31)4 (0.86)2 (0.46)0 (0)0 (0)1 (0.36)0 (0)8 (0.29)
G3P[10]0 (0)1 (0.22)0 (0)0 (0)0 (0)0 (0)0 (0)1 (0.04)
G4P[6]29 (9.12)86 (18.57)98 (22.63)50 (15.06)62 (15.54)49 (17.50)67 (12.09)441 (15.87)
G9P[6]0 (0)9 (1.94)3 (0.69)0 (0)0 (0)1 (0.36)0 (0)13 (0.47)
G10P[6]0 (0)0 (0)0 (0)0 (0)1 (0.25)0 (0)0 (0)1 (0.04)
Partially typed strains8 (2.52)13 (2.80)2 (0.46)0 (0)0 (0)7 (2.50)49 (8.84)79 (2.84)
Mixed types21 (6.60)21 (4.54)13 (3.00)9 (2.71)26 (6.52)21 (7.50)55 (9.93)166 (5.97)

In brief, the most common strains were G2P[4] during the first two seasons, G3P[8] from 2002–2003 until 2004–2005, and G1P[8] from the 2005–2006 season onward. In the 2000–2001 season, 34.59% of the strains had genotype G2P[4], although this percentage decreased from the 2001–02 season. The number of cases with G3P[8] increased (from 1.89% to 46.69%) during the first four seasons and decreased during the last three seasons (from 27.07% to 8.12%). G1P[8] had lower detection rates from 2003 to 2005 (<20%) than those reported in many countries [7]; however, the detection rates sharply increased in the 2006–2007 season (52.17%). G9P[8] exhibited a peak prevalence of 9.7% in 2002–2003 but had much lower detection rates except in 2003.

In the present study, the prevalence and distribution of RV-A genotypes among cases of acute gastroenteritis in Korea were investigated to predict whether the recently introduced RV-A vaccines, Rotarix and RotaTeq, are sufficient for preventing RV-A infection. By nationwide surveillance, we found that only approximately 50% of the strains during the surveillance period had either a G or a P antigen present in both the vaccines; the distribution of the circulating RV-A genotypes and the predominant strains varied from year to year.

One of the hallmarks of such surveillance in the past decade as part of vaccine development programmes is the characterization of a large diversity of serotypes among human RV-A, with 110 P and G serotypes each and 140 P–G antigen combinations being reported [5,7]. Such reports highlight the findings on emerging serotypes (e.g. G9) that have become globally common during recent years, other potentially emerging serotypes (e.g. G12), as well as regionally common serotypes (e.g. G5 and G8) [5,7–9]. The tracking of these strains is a key component of surveillance studies (as immunization programmes are introduced in individual countries) aiming to determine whether any of the strains can evade the protective immunity provided by the vaccines.

Previous studies of the RV-A strains circulating in Korea have shown that G1P[8] (36%) was the most prevalent strain, followed by G3P[8] (16%), G4P[6] (8.9%) and G1P[6] (8.2%), in the period from 2005 until 2007 [10], whereas the reports spanning 2002–2004 indicate that 39% of the strains were G9P[8], 24% were G1P[8], 17% were G3P[8] and 13% were G2P[4] [11]. However, the previous data were relatively limited in size as well as by the number of sites covered, and may not be representative of the whole country (i.e. Korea). The present study is the first extensive nationwide surveillance for circulating RV-A strains in Korea.

In the 2000–2001 season, the prevalence of G2P[4] was as high as 34.6%. Recently, an increase in the incidence of infection with the G2P[4] strain has been reported in Brazil, where the monovalent G1P[8] RV-A vaccine is currently in use [12–15]. This finding raises questions about the ability of this vaccine to confer cross-protection against G2P[4] RV-A -associated disease, although more evidence is required before firm conclusions can be made [13]. Assessment of the effectiveness of a vaccine against disease caused by specific strains after the introduction of the vaccine is important, particularly against disease caused by serotypically unique strains.

Fifteen out of 2779 strains identified during the present study have both serotype antigens (e.g. G9P[4], G9P[6] and G10P[6]), which are different from the serotype antigens of the strains in either vaccine; in addition, they belong to a genogroup that is different from those of the vaccine strains (Table 1). The prevalence of G9P[6] strains is only 0.47%, with a maximum of 1.94%, and single isolates of G9P[4] and G10P[6] were also identified.

Further studies on vaccine effectiveness are needed to determine whether such strains are more likely to evade immunity provided by the new vaccines. As shown in Table 1, several strains detected at very high frequencies (12.5%–24.9%) had common G antigens (G1, G2, G3 or G4) with uncommon P antigens [6] or [9]; G4P[6] was the major strain among these strains. Therefore, unusual reassortants of the common serotypes, such as G1P[4], possibly express epitopes different from those of the parental P–G combinations. Finally, some neutralization epitopes on the P serotype antigen may be masked depending on the composition of the other ten genes [16]. The proportion of reassortants of the common human strains in the present study was approximately 3% (Table 1). It is difficult to predict how vaccine introduction will affect the diversity of the RV-A strains in Korea. Therefore, it is important to monitor the impact of vaccines on the prevalence of such unusual reassortants.

In summary, surveillance conducted over seven consecutive RV-A seasons from 2000 until 2007 showed that only approximately 50% of the strains had either a G or a P antigen present in both RotaTeq and Rotarix. Strains with an uncommon antigenic makeup circulated in all the seasons during which larger numbers of strains were genotyped, indicating that these strains may be endemic to Korea. The maintenance of nationwide surveillance is essential for understanding the impact of new vaccines on the circulating RV-A strains as well as the overall reduction in the prevalence of RV-A diarrhea in Korea.

Acknowledgements

We thank the staff at the 16 laboratories of local public health institutes as well as the participants at a sentinel hospital in the laboratory surveillance system for acute gastroenteritis in Korea.

Transparency Declaration

This work was supported by a grant from the National Institutes of Health (NIH-091-4845-300), National Research and Development Program, Ministry of Health and Welfare, Republic of Korea.

Ancillary