Seroepidemiology of coxsackievirus B5 in infants and children in Jiangsu province, China

ABSTRACT

Coxsackievirus B5 (CV-B5) is associated with various human diseases such as viral encephalitis, aseptic meningitis, paralysis, herpangina, and hand, foot and mouth disease (HFMD). However, there is currently no effective vaccine against CV-B5.The seroepidemiologic characteristics of CV-B5 remained unknown. A cohort study was carried out in 176 participants aged 6–35 months from January 2012 to January 2014. The serum samples were collected and tested for CV-B5 neutralizing antibodies (NtAbs) four times during these two years. The confirmed enterovirus cases were recorded through the surveillance system, and their throat or rectal swabs were collected for pathogen detection. According to the changes of CV-B5 NtAbs, two CV-B5 epidemics were detected among these participants during the two-year follow-up. Sixty-seven cases out of all participants had seroconversion in CV-B5 NtAbs. During the first epidemic from March 2012 to September 2012, CV-B5 seropositivity rate increased significantly (6.8%, 12/176 vs. 21.6%, 38/176, P = 0.000). The seroconversion rate and geometric mean fold-increase (GMFI) were 18.2% (32/176) and 55.7, respectively; During the second epidemic from September 2012 to January 2014, CV-B5 seropositivity rate also increased (21.6%, 38/176 vs. 38.6%, 68/176, P = 0.000), and the seroconversion rate and GMFI were 19.9% (35/176) and 46.5, respectively. Only one case had CV-B5 associated HFMD during the two-year follow-up, and CV-B5 from the throat swab isolate was GI.D3 subtype, which belonged to the major pandemic strain in mainland China. CV-B5 infection was common in infants and children in Jiangsu province, China. Therefore, it's necessary to strengthen the surveillance on CV-B5 and to understand the epidemic characteristics of CV-B5 infection.

Introduction

Coxsackievirus B5 (CV-B5) is one of the small icosahedral RNA viruses belonging to the Human enterovirus (HEV) B species within the family Picornaviridae.¹ It is associated with many human diseases ranging from mild, herpangina,² and hand foot and mouth disease (HFMD)^2-5 to aseptic meningitis,^6-13 viral encephalitis,^14-17 and acute flaccid paralysis.¹⁸ Moreover, its association with type 1 diabetes was also suggested.^19,20 Generally, CV-B5 is one of the major pathogens responsible for viral encephalitis and aseptic meningitis outbreaks (Table 1) in Europe,^6-8,21 North America,¹⁴ South America,⁹ and Asia,^10-12,15 but so far there were few reports about the CV-B5 vaccine research.

Table 1.

The epidemic data of CV-B5 worldwide.

Author (Pubtime)	Countries and regions	Report time (year)	Age	CV-B5-associated diseases	The ratio between CV-B5 and positive cases	References
Europe
Thoelen I, et al. (2003)	Belgium	2000	2 months-35 years old	Aseptic meningitis	CV-B5:15.6%	6
					(19/122)
Papa A, et al. (2006)	Greece	2003-2004	21 days-73 years old	Aseptic meningitis	CV-B5:20.1%	7
					(4/19)
Antona D, et al. (2007)	France	2000-2004	80.9% less than 14 years old	Central nervous system syndrome, Gastrointestinal disease, Respiratory disease,Neuromuscular disease and Cardiac disease, etc	CV-B5:6.7%	21
					(184/2757)
Trallero G, et al. (2010)	Spain	1998-2007	N/A	Aseptic meningitis, Respiratory disease, Acute flaccid paralysis	CV-B5:4.2%	8
					(118/2787)
America
Dos Santos GP, et al. (2006)	Brazil	1998–2003	28 days-68 years old	Aseptic meningitis	CV-B5:3.7%	9
					(6/162)
Tavakoli NP, et al. (2008)	The United States	2005	N/A	Meningitis/Encephalitis	CV-B5:40%	14
					(14/35)
Asia
Lee ST, et al. (2007)	Korea	2005	2 days-12 years old	Aseptic meningitis	CV-B5:62.5%	10
					(15/24)
Baek K, et al. (2011)	Korea	2009	95.3% less than 10 years old	Aseptic meningitis, Respiratory disease, HFMD	CV-B5:19.6%	11
					(21/107)
Kumar A, et al. (2011)	India	2008	≤15years	Viral encephalitis	CV-B5:53.33%	15
					(8/15)
Yen FB, et al. (2009)	Taiwan (China)	1999–2006	N/A	Aseptic meningitis	CV-B5:12.5%	12
					(24/192)
Mainland China
Chen P, et al. (2013)	Shandong province	2009	98.6% less than 16 years old	Aseptic meningitis	CV-B5:40.5%	16
					(17/42)
Liu N, et al. (2014)	Henan province	2009	1 year-13 years old	Aseptic meningitis	CV-B5:37.2%	13
					(16/43)
Zhang L, et al. (2013)	Zhejiang province	2002–2012	<15 years	Viral encephalitis	CV-B5:17.8%	17
					(40/225)
Hu YF, et al. (2012)	Shandong province	2009	N/A	HFMD	CV-B5:12.7%	5
					(14/110)
Han JF, et al. (2012)	Jilin province	2010	N/A	HFMD	CV-B5:18.8%	4
					(3/16)

N/A = Not available.

In recent years, the association of CV-B5 with aseptic meningitis and viral encephalitis was also reported in mainland China (Table 1).^4,5,13,16,17 From June to September 2009, a total of 2,104 cases were reported in an aseptic meningitis outbreak in Shandong province. Forty-two cerebrospinal fluid specimens collected from aseptic meningitis patients were tested with 17 (40.5%) enteroviruses isolated and identified as CV-B5.¹⁶ In Henan province, an outbreak of aseptic meningitis caused by CV-B5 also occurred in 2009, and the proportion of severe cases was 33% (14/43).¹³ HFMD caused by CV-B5 was also reported. In 2009, CV-B5 was detected in throat swabs and serum of 14 patients, 12.7% (14/110) of all HFMD patients. 78.6% (11/14) CV-B5 positive HFMD patients showed neurological manifestations.⁵ During a HFMD outbreak in 2010, three of 16 throat swab samples tested positive for HEV were eventually identified to have CV-B5.⁴

The HEV surveillance has been implemented in some countries for many years.^2,21,22 Surveillance data showed that CV-B5 was the third most commonly identified enterovirus in South Korea during 1999–2011,²² the fourth in France during 2000–2004,²¹ and the fourth in the United States during 1970–2005.² However, seroepidemiologic data of CV-B5 were limited worldwide. In China CV-B5 has not even been included in the disease surveillance system and it is difficult to thoroughly evaluate the disease burden.

A cohort study of 176 healthy participants aged 6–35 months from a phase III trial for enterovirus A71 (EV-A71) vaccine was carried out from January, 2012 to January, 2014 in Jiangsu province to better understand the seroepidemiologic characteristics of CV-B5 infection in infants and children. A surveillance was also carried out to identify enterovirus-associated cases and to collect related medical data.

Results

CV-B5 neutralizing antibodies (NtAbs) in infants and children

A total of 319 participants aged 6–35 months were followed for two years, and serum samples from 176 participants were successfully collected at all four scheduled visits (January 2012, March 2012, September 2012 and January 2014) in this study. Among the 176 participants, the male-to-female ratio was 1.9:1. All participants were divided into two groups: the infant group with 38 participants aged 6–11 months and the child group with 138 participants aged 12–35 months.

As shown in Fig. 1 and Table 2, CV-B5 total seropositivity rates (NtAb titers ≥ 8) and total GMTs remained flat at 6.8% and 4.8 respectively from January to March 2012. From March to September 2012, the increase in total seroconversion rate of CV-B5 (18.2%, 32/176 and GMFI = 55.7) was associated with a documented CV-B5 epidemic (Fig. 2 and Table 3), followed by significant increases in total seropositivity rates (Fig. 1, 6.8%, 12/176 vs. 21.6%, 38/176, P = 0.000).

Figure 1.

Changes in seropositivity rates of CV-B5 in infants and children in Jiangsu province during January 2012-January 2014. Cytopathogenic effect (CPE) method was used for the measurement of CV-B5 neutralizing antibodies (NtAbs) in serum. NtAb titer <8 was defined as negative; and ≥8 as positive.

Table 2.

The GMTs of CV-B5 neutralizing antibodies at each time point.

		GMTs (95% CL)
Group	N	January 2012	March 2012	September 2012	January 2014
Infant	38	4.1	4.0	9.9	20.1
		(4.1–4.1)	(4.0–4.0)	(9.1–10.7)	(19.0–21.2)
Child	138	5.0	5.1	10.0	19.0
		(5.0–5.1)	(5.1–5.2)	(9.8–10.2)	(18.8–19.3)
Total	176	4.8	4.8	10.0	19.2
		(4.8–4.8)	(4.8–4.9)	(9.8–10.1)	(19.0–19.5)

N = The number of participants.

95% CL = 95 percent confidence limit.

GMT = Geometric mean titer.

Figure 2.

Changes in seroconversion rates of CV-B5 in infants and children in Jiangsu province during January 2012-January 2014. Cytopathogenic effect (CPE) method was used for the measurement of CV-B5 neutralizing antibodies (NtAbs) in serum. Seroconversion is defined if the NtAb titer of less than 8 at the front time point became ≥8 at the next time point, or if the NtAb titer of ≥8 at the front time point increased by at least four-fold at the next time point.

Table 3.

The GMFI of CV-B5 neutralizing antibodies during two CV-B5 epidemics.

		GMFIs (95% CL)
Group	N	March 2012-September 2012	September 2012-January 2014
Infant	38	137.9	193.7
		(137.7–138.2)	(193.1–194.2)
Child	138	43.2	37.0
		(42.8–43.6)	(36.3–37.7)
Total	176	55.7	49.5
		(55.4–56.0)	(48.9–50.1)

N = The number of participants.

95% CL = 95 percent confidence limit.

GMFI = Geometric mean fold-increase.

From September 2012 to January 2014, CV-B5 total seroconversion rate (Fig. 2, 19.9%, 35/176) and CV-B5 GMFI (Table 3, 49.5) showed that another CV-B5 epidemic might take place in this population. Significant increase in CV-B5 seropositivity rate was also observed (Fig. 1, 21.6%, 38/176 vs. 38.6%, 68/176, P = 0.000). Among 35 seroconversion cases, 4 of them with CV-B5 NtAb titers at middle/high levels during the first CV-B5 epidemic had NtAb titers increased by more than 4 times (NTAb: 48 vs. 384; 192 vs.1024; 256 vs.1024; 256 vs.1536).

During the follow-up, two CV-B5 epidemics were observed among these participants, and 67 cases out of all participants (38.1%) were detected to have seroconversion in CV-B5 NtAbs. There was no significant difference in seroconversion rate and GMFI between these two epidemics (P = 0.730, P = 0.627). In addition, there was no significant difference in seroconversion rates between infant group and child group (P = 0.966, P = 0.475) during either CV-B5 epidemic, but GMFI in infant group was significantly higher than that in child group (P = 0.013, P = 0.018).

The impact of EV-A71 vaccination on CV-B5 epidemics

All participants were enrolled in an EV-A71 phase III trial.²³ It was necessary to understand whether vaccination against EV-A71 could have any impact on CV-B5 epidemics. From January 2012 to February 2012, 82 of 176 participants were vaccinated with inactivated EV-A71 whole-virus vaccine (Vero cell), while the rest took placebo (aluminum hydroxide adjuvant). The result showed (Table 4) that CV-B5 seroconversion rates (P = 0.413, P = 0.908) and GMFI (P = 0.234, P = 0.230) were not significantly different between the EV-A71 vaccine group and the placebo group during either CV-B5 epidemic, indicating that the EV-A71 vaccination had no impact on CV-B5 epidemics.

Table 4.

The impact of vaccination against EV-A71 vaccine on CV-B5 epidemics.

		March 2012-September 2012		September 2012-January 2014
		Seroconversion	GMFIs	Seroconversion	GMFIs
Group	N	rate	(95% CL)	rate	(95% CL)
EV71 vaccine	82	17/82*	69.7	16/82	66.6
		(20.7%)#	(69.1–70.4)	(19.5%)	(65.3–67.8)
Placebo	94	15/94	43.2	19/94	34.3
		(16.0%)	(42.7–43.8)	(20.2%)	(33.2–35.4)
P value		P = 0.413	P = 0.234	P = 0.908	P = 0.230

N= The number of participants in each group.

^*= The number of seroconversion cases/ The number of participants in each group.

^#= The number of seroconversion cases/ The number of participants in each group×100%.

GMFI= Geometric mean fold-increase.

95% CL= 95 percent confidence limit.

Surveillance results

Only one case was recorded with CV-B5 associated HFMD from September 2012 to January 2014, representing 2.9% (1/35) of all seroconversion cases in the second CV-B5 epidemic. This case (male, 28 months old) belonged to the EV-A71 vaccine group. Onset time was July 30^th, 2013. Only mild clinical symptoms were observed, including fever (39°C), rash and runny nose. CV-B5 NtAb titers showed 256-fold increase after infection (NTAb: <8 vs. 1024).

Genetic type of isolated CV-B5

Throat swab from the HFMD patient was taken for pathogen detection. The complete genome sequence of the CV-B5 isolate, 319-JS/CHN/2013 (Genebank no. KY744347), was analyzed, and the phylogenetic tree based on VP1 region sequences of different CV-B5 strains was constructed. The results showed that 319-JS/CHN/2013 belonged to genogroup GI.D3, and that it shared 98.7% nucleotide identity with the epidemic strain in Shandong province in 2013 (Genebank no. KU050049).

Discussion

To investigate the seroepidemiologic characteristics of CV-B5 infection in infants and children, CV-B5 NtAbs in serum samples were analyzed. During the two-year follow-up, 67 cases out of all participants (38.1%) were detected to have seroconversion in CV-B5 NtAbs and total seropositivity rate reached 38.6% at the end of the follow-up. Two CV-B5 epidemics were observed among these participants.

During both CV-B5 epidemics, the epidemic intensities were almost the same between infant group and child group (P = 0.966, P = 0.475), though the range of CV-B5 NtAbs increase in infant group was much bigger than that in child group (P = 0.013, P = 0.018). In addition, EV-A71 vaccination and sex (Table S1) didn't have significant impacts on either CV-B5 epidemic.

The report of CV-B5 in the United States from 1970 to 2005 showed that CV-B5 had a distinct epidemic pattern of circulation with regular sharp increase every 3–6 years.² The epidemic data of HEV in South Korea from 1999 to 2011showed the similar pattern.²² In this study, we discovered two consecutive peaks of CV-B5 infection in infants and children from March 2012 to January 2014, which might be associated with the regional characteristics of enrolled participants. It is worth noting that because of the high frequency of CV-B5 epidemics, infants and children might have a higher risk to develop severe diseases.

Most CV-B5 infections are asymptomatic,¹⁶ but some patients may develop a wide variety of diseases such as meningitis, encephalitis, paralysis, herpangina, diabetes, and HFMD.^2-20 In this study, only one HFMD patient (1/32) was found to have CV-B5, and no severe disease was recorded during the follow-up. It suggested that the rate of clinically overt infection for CV-B5 was low.

The CV-B5 strains circulated worldwide were clustered into two genogroups (GI and GII) (Fig. 3). The prototype strain of CV-B5 (Faulkner, Genebank no.AF114383)²⁴ belonged to GI. The first HFMD associated strain (UK/1954, Genebank no.X67706)²⁵ was reported to belong to GII. GI consisted of 4 genotypes (GI.A-D), with GI.D divided into three branches (GI.D1-D3), while GII included 3 genotypes (GII.A-C). Most epidemic strains in Europe (France strains for instance) belonged to GI, while GII strains circulated during the same period. However, there has been no report on GII strain in mainland China. Almost all strains emerged in mainland China belonged to GI.D, except the first reported strain during Shandong epidemic in 1998 (Genebank no.GQ329771), which belonged to GI.B. In this study, a GI.D strain named 319-JS/CHN/2013 was isolated from the throat swab of a HFMD case. Because only one case had CV-B5 associated HFMD during the two-year follow-up, the isolation rate of CV-B5 was low. The CV-B5 strains in genotype GI.D circulated in different regions in mainland China, which eventually lead to the outbreaks of severe nervous system diseases such as viral encephalitis, and aseptic meningitis in Shandong¹⁶ and Henan.¹³ Recent studies^3,4 showed that GI.D strain was also one of the significant pathogens that caused HFMD outbreaks and sporadic cases and had become a great risk to the healthy infants and children.

Figure 3.

Phylogenetic tree based on VP1 region of CV-B5 strains. The Neighbor-joining method was used to construct the tree. The analysis model of Neighbor-joining and Kimura 2-parameter was applied, and the 1000 bootstrap was used as the analysis parameter. Bootstrap values less than 70% were hidden.

Inactivated EV-A71 whole-virus vaccines are available in China commercially, and become an effective tool for the prevention of severe HFMD epidemic.²⁶ As a result of increasing HFMD outbreaks caused by different enteroviruses,^27-34 it's getting more difficult to prevent and control HFMD epidemics in China. In the post EV-A71 era, the surveillance on the major HFMD-related enteroviruses should be strengthened, including Coxsackievirus A6,^27-29 Coxsackievirus A16,^30-32 Coxsackievirus B3,^33-34 and CV-B5. Since little attention was paid to CV-B5, there was not that much basic research on it. To provide a scientific basis on CV-B5 control, it's necessary to strengthen the surveillance on CV-B5, to investigate CV-B5 evolution, and to study CV-B5 virological characteristics and diagnostic method.

Materials and methods

Human subjects and serum samples

A total of 319 participants aged 6–35 months, who were previously enrolled in a clinical trial on the efficacy of EV-A71 vaccine in Jiangsu Province (clinical trial No. NCT01508247),²³ were monitored for two years in this research (January, 2012-January, 2014).³⁵ Serum samples were collected at the beginning of the study (January 2012), and then another three times in March 2012, September 2012 and January 2014. Eventually, 176 participants had their serum samples collected successfully at all four scheduled visits and their serum samples were analyzed subsequently. Among the 176 participants, there were 116 males and 60 females, while 38 were infants (aged 6–11 months) and 138 were included in the child group (aged 12–35 months).

Measurement of CV-B5 NtAbs

CV-B5 NtAb in serum samples was measured with the cytopathogenic effect (CPE) method. Two-fold serial dilutions of serum (started at l:8) were incubated with equal volume of virus culture (50 μl) containing 100 TCID₅₀/well (50% of tissue culture infective dose) of CV-B5-3-3-2 (GenBank accession no. KY303900) at 37°C for 2 h. Vero cell suspension (final concentration: 8.0 × 10⁴ cells/ml) was then added and the mixture was incubated in a CO₂ incubator at 35°C for 7 days before CPE observation was carried out using microscopy. Neutralizing titers were defined as the highest dilution to achieve 50% inhibition of CPEs.

Surveillance

A surveillance of enterovirus-associated disease in EV-A71 trial centers (clinical trial No. NCT01508247)^23,35 was carried out. We regularly contacted the guardians of participants and instructed them to seek medical treatment if any symptom was noticed. All patients were screened for the following symptoms: HFMD-like illness, herpangina, acute respiratory symptoms or gastrointestinal symptoms with or without other systemic complications such as CNS complications and cardio pulmonary failure. All potential cases were reported by the clinics or hospitals and throat or rectal swabs were taken for pathogen detection within 24 h.

Molecular biological detection

One CV-B5 strain was isolated from clinical specimens. Total viral RNA was extracted from patient samples using MagMAX™-96 Viral RNA Isolation Kit (Invitrogen, AM1836). Primers targeting 5′-NTR of EVs were used in nested RT-PCR to amplify 5′-NTR region. After the virus from the positive sample was propagated in cell culture, the overlapping fragments were sequenced using degenerate primers and were assembled to construct the complete genome sequence. Reference strains, including prototype strain and strains from other countries at different times, were selected from GeneBank. Phylogenetic analysis based on VP1 region sequences was carried out using Mega 5.2 program. The analysis model of neighbor-joining and Kimura 2-parameter was applied, and the 1000 bootstrap was used as the analysis parameter.

Statistical analysis

CV-B5 NtAb titer below 8 was considered negative, while CV-B5 NtAb titer ≥ 8 was considered positive. Seroconversion was defined if CV-B5 NtAb titer increased to ≥ 8 from less than 8 at beginning of the study, or if CV-B5 NtAb titer increased by more than 4 times while it was ≥ 8 at the beginning of the study. Seropositivity rates and seroconversion rates were analyzed with Chi-square test, respectively. In this study, CV-B5 epidemic was defined that the total seroconversion rate was at least 10-fold higher than the seroconversion rate of sporadic infection in the area. NtAb titers for all participants were log-transformed to calculate the geometric mean titer (GMT). NtAb titers from seroconversion cases were log-transformed to calculate geometric mean fold-increase (GMFI). All titers below 8 were set as 4 during calculation. GMFIs in each group were compared using Student's t-test.

Hypothesis testing was conducted by two-sided test, with the alpha value set at 0.05 for statistical significance. All statistical analyses were carried out with Microsoft Excel 2007, SPSS 19.0 and GraphPad Prism 5.

Abbreviations

CV-B5

Coxsackievirus B5

HEV

Human enterovirus

HFMD

Hand foot and mouth disease

NtAbs

Neutralizing antibodies

CPE

Cytopathic effect

GMT

Geometric mean titer

GMFI

Geometric mean fold-increase

EV-A71

Enterovirus A71

TCID₅₀

50% of tissue culture infective dose

Acknowledgments

We thank Jiangsu provincial center for disease control and prevention for the help on sample collection. We thank the Major Special Projects Funding Program (No. 2016ZX09101120) from the Ministry of Science and Technology of the People's Republic of China for their support.

Disclosure of potential conflicts of interest

The authors report no conflict of interest.

Funding

The current study was sponsored by the Major Special Projects Funding Program (No. 2016ZX09101120) from the Ministry of Science and Technology of the People's Republic of China. The funders had no role in study design, data collection and analysis, data interpretation, decision to publish, or preparation of the manuscript.

Ethics approval and consent to participate

The study was approved by the institutional review board of Jiangsu Provincial Center of Disease Control and Prevention, and done in accordance with the Declaration of Helsinki, Good Clinical Practice, and Chinese regulatory requirements.

Consent for publication

All guardians of participants provided written informed consent.