+ MODEL

Microbes and Infection xx (2014) 1e8

www.elsevier.com/locate/micinf

Towards broadly protective polyvalent vaccines against hand, foot and

mouth disease
Qingwei Liu a,1, Xin Tong b,1, Zhong Huang a,*
a
Vaccine Research Center, Key Laboratory of Molecular Virology & Immunology, Institut Pasteur of Shanghai, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences, 320 Yueyang Road, Shanghai 200031, China
b
Shanghai Zerun Biotechnology Co., Ltd., Building 9, 1690 Zhangheng Rd, Zhangjiang, Pudong New District, Shanghai 201203, China
Received 13 November 2014; accepted 21 November 2014

Abstract

Hand, foot, and mouth disease (HFMD) caused by multiple enterovirus infections is a serious health threat to children in the AsiaePacific
region. This article reviews progresses in the development of vaccines for HFMD and discusses the need for polyvalent HFMD vaccines for
conferring broad-spectrum protection.
© 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

Keywords: Hand, foot, and mouth disease; Enterovirus 71; Coxsackievirus A16; Polyvalent vaccine

1. Introduction and China. In China alone, millions of HFMD cases and hundreds
of HFMD-related deaths occurred annually in the last five years.
Hand, foot, and mouth disease (HFMD) is a highly conta- According to China CDC, as of October 7, there were a total of
gious disease that mostly affects children under the age of 5 2,385,764 HFMD cases and 459 deaths reported in 2014 (http://
[1,2]. HFMD patients often present with mild fever and rash or www.chinacdc.cn). Besides Southeastern Asia, sporadic HFMD
blister on the surface of hand, foot and mouth [1]. A propor- epidemics also occurred in other regions, including North
tion of them may develop severe neurological and cardiopul- America, Oceania, Europe, and even Africa [7,8]. It is likely that
monary complications, including aseptic meningitis, brainstem HFMD will soon become a global health problem and thus the
encephalitis, poliomyelitis, encephalomyelitis and pulmonary development of HFMD vaccines is of significant importance for
edema, and ultimately death [1,3]. In addition, convalescent worldwide control of this highly contagious disease.
patients are at the risk of long-term neurological sequelae and A number of the human enterovirus species A (HEV-A)
cognitive impairment [1,4]. within the Picornaviridae family are the major causative
HFMD was first reported in New Zealand in 1957 [5]. How- agents of HFMD, including enterovirus 71 (EV71), coxsack-
ever, a large outbreak has not taken place until 1997 when HFMD ievirus A16 (CA16), coxsackievirus A6 (CA6), and coxsack-
was epidemic in Malaysia [6]. Since then, HFMD epidemics hit ievirus A10 (CA10) [9,10,1,11]. These are non-enveloped
many countries in the AsiaePacific region, including Singapore, viruses with a single-stranded, positive sense RNA genome of
Malaysia, Taiwan, Japan, Korea, Vietnam, Cambodia, Thailand ~7.4 kb encapsidated within an icosahedral protein shell made
of 60 copies each of VP1, VP2, VP3 and VP4 subunit proteins.
Several receptors for EV71 have been identified [12]. Existing
* Corresponding author. Tel.: þ86 21 54923067.
surveillance data indicate that these viruses either circulate in
E-mail address: [email protected] (Z. Huang). alternative years or co-circulate in same years with one of
1
These authors contributed equally to this work. them being the predominant causative agent [9,10,13]. Co-

http://dx.doi.org/10.1016/j.micinf.2014.11.004
1286-4579/© 2014 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
2 Q. Liu et al. / Microbes and Infection xx (2014) 1e8

infections with two viruses were frequently reported [14,15] vaccines. In fact, the first inactivated EV71 vaccine candidate
and they are likely to have contributed to the emergence of was developed in 1975 when EV71 was epidemic in Bulgaria,
the recombinant viral strains that caused recent HFMD out- but it was discontinued for evaluation due to no further outbreaks
breaks [16,17]. in Bulgaria after 1976 [23]. Recent HFMD outbreaks in the
The magnitude and severity of the recent HFMD endemics AsiaePacific region have renewed interest in developing EV71
has drawn the attention of the governments and vaccine vaccines and have driven active research and development pro-
makers in the AsiaePacific regions. In particular, China has grams in both public and private sectors. A large body of pre-
declared the HFMD vaccine development a national priority clinical studies of inactivated whole virus EV71 vaccines has
[18]. Consequently, a few vaccine candidates have rapidly been published [23,24,18], and the protective efficacy of this
progressed into clinical trials through a “green channel” vaccine approach was validated in animal models e thus
mechanism [19], and many more vaccine candidates are at the providing supporting evidence for clinical trials of some of the
preclinical development stage. However, concerns are also leading vaccine candidates. So far, five inactivated EV71 vac-
raised with regard to the breadth of protection conferred by cines developed by different companies/organizations in
these candidate vaccines [20,21]. Below we summarize the Taiwan, Singapore and Mainland China have entered the clinical
recent achievements and challenges in the field of HFMD trial stage (Table 1). In Taiwan, National Health Research
vaccine development. Institute (NHRI) selected a local isolate E59 (B4 subgenotype)
as the vaccine strain and used formalin for inactivation. The
2. Vaccines targeting EV71 immunogenicity of the formalin-inactivated vaccine was vali-
dated in different animal models [25]. Based on the preclinical
Among all HFMD pathogens, EV71 is more often associ- results, a phase I clinical trial with the inactivated EV71 vaccine
ated with severe HFMD cases with neurological complications candidate at 5 mg and 10 mg doses was launched in 2010 and
and even death [9,22]. Therefore, it has been the only target completed in 2012 [26]. The trial not only showed that the
for developing HFMD vaccines for a very long time [23,24]. A neutralizing antibody titers against the B4 vaccine strain
number of approaches have been taken to develop vaccines increased by more than 4 fold even after a single immunization
targeting EV71; those at advanced developmental stages are [26], but also revealed the presence of cross-neutralizing anti-
listed in Table 1. bodies against subgenotypes B1, B5 and C4a, but not C2 [27]. A
Singapore-based company, Inviragen, initiated phase I clinical
2.1. EV71 inactivated vaccine trial with 0.6 mg and 3 mg per dose of inactivated EV71 vaccine
(subgenotype B3) [28], the results of which are pending. In
Because of the past success of inactivated poliovirus vaccine mainland China, three vaccine makers, including Beijing Vigoo,
(IPV), it was naturally a first choice to develop inactivated EV71 Sinovac and Chinese Academy of Medical Science (CAMS),

Table 1
Selected EV71 vaccine candidates currently at advanced development stage.
Vaccine type EV71 strain Cell substrate Immunogenicity/protective efficacy Clinical trial Developer Reference
(subgenotype)
Inactivated E59 (B4) Vero Induce neutralizing antibodies in humans Phase 1 completed NHRI (Taiwan) [26]
virus
Inactivated INV21 (B) Vero Not reported Phase 1 completed Inviragen www.inviragen.com
virus (Singapore)
Inactivated H07 (C4) Vero 94.8% efficacy against EV71-associated Phase 3 completed Sinovac (China) [30]
virus HFMD or herpangina and 88.0% against
EV71-associated diseases in humans
Inactivated FY7VP5/AH/ Vero 90.0% efficacy against EV71-associated Phase 3 completed Vigoo (China) [29]
virus CHN/2008 (C4) HFMD and 80.4% against EV71-associated
disease in humans
Inactivated FY2008 (C4) KMB17 97.4% efficacy against EV71-associated Phase 3 completed CAMS (China) [31]
virus HFMD in humans
VLP Neu (C2) Sf9 Induce neutralizing antibodies in monkeys National Taiwan [37]
University
VLP G082 (C4) Sf9 Protect mice against lethal challenge IPS-CAS (China) [38]
VLP AH08/08 (C4) Saccharomyces Protect mice against lethal challenge IPS-CAS (China) [40]
cerevisiae
Attenuated EV71 BrCr (A) Vero Protect monkeys against lethal challenge NIID (Japan) [42]
live virus
NHRI: National Health Research Institute, Taiwan.
CAMS: Chinese Academy of Medical Sciences, China.
IPS-CAS: Institut Pasteur of Shanghai, Chinese Academy of Sciences, China.
NIID: National Institute of Infectious Diseases, Japan.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
 Q. Liu et al. / Microbes and Infection xx (2014) 1e8 3

have independently developed C4 subgenotype-based inacti- subgenotype) [34]. The expressed P1 polyprotein was cleaved
vated EV71 vaccine candidates, all of which have gone through by 3CD into VP0, VP1 and VP3 subunit proteins, resulting in
phase 1 to 3 clinical trials. Beijing Vigoo's phase 3 trial, which the assembly of VLPs with morphology similar to that of the
recruited more than 10,000 participants with age from 6 to 35 native EV71 virus [35]. Two doses (10 mg/dose) of EV71
months, showed vaccine efficacy of 90.0% (95% CI 67.1e96.9) VLPs elicited neutralizing antibodies in mice with titers up to
against EV71-associated HFMD and 80.4% (95% CI 213 and protected mice against lethal viral challenge [36]. It is
58.2e90.8) against EV71-associated disease (including her- worth noting that the neutralizing antibody titers induced by
pangina, neurological complications, and non-specific illnesses VLPs were significantly higher than those elicited by the
caused by EV71) [29]. Similarly, the phase 3 trial with Sinovac's inactivated vaccine derived from the same virus strain [36].
inactivated EV71 vaccine candidate showed 94.8% efficacy Subsequently, the VLP experimental vaccine was tested in
(95% CI, 87.2e97.9) against EV71-associated HFMD or her- monkeys at doses of 20 or 100 mg. Sera obtained from VLP-
pangina and 88.0% (95% CI, 78.6e93.2) against EV71- immunized monkey efficiently neutralized EV71 sub-
associated diseases [30]. In addition, CAMS also completed genotypes B4, B5, C2, C4 and C5 [37]. EV71 VLPs derived
their phase 3 trial using a cohort of 12,000 children between 6 from subgenotype C4 have been produced using baculovirus/
and 71 months [31]. This trial showed an efficacy of 97.4% (95% insect cell expression system and shown to induce neutralizing
CI, 92.9e99.0) against EV71-associated HFMD by intention-to- antibodies against all C4 strains tested and the A genotype
treat analysis and 97.3% (95% CI, 92.6e99.0) by per-protocol stain BrCr in mice [38]. Consistently, the neutralizing anti-
analysis, but could not draw a conclusion regarding its efficacy bodies in sera from the VLP-immunized mice were found to
against EV71-associated disease [31]. More significantly, this target mainly an extremely conserved epitope located at the
trial revealed that the EV71 vaccine could not protect against GH loop of the VP1 protein [38]. These results indicate that
CA16 or other non-EV71 HFMD-causing enteroviruses [31], VLP-based vaccines have the potential to provide protection
thus highlighting the necessity of developing multivalent vac- against a broad spectrum of EV71 isolates. Another advantage
cines for broad-spectrum protection against HFMD. of insect cell-produced VLPs as EV71 vaccine is the high
Since EV71 is traditionally classified into 11 subgenotypes yield, which reached 64.3 mg per liter culture under optimal
(A, B1eB5, and C1eC5), it is desirable to have a single EV71 conditions [39].
vaccine that provides protection against as many of the 11 EV71 VLPs have also been successfully produced in other
subgenotypes as possible. Selection of vaccine strains with recombinant expression systems such as yeast. Li et al.
broad protective potential is thus critical to the success of the generated VLPs of EV71 subgenotype C4 from Saccharo-
EV71 inactivated vaccine approach. The C4-based vaccine myces cerevisiae co-expressing P1 and 3CD [40]. The S.
produced by the Chinese companies induced broadly cerevisiae-derived VLPs elicited robust neutralizing antibody
neutralizing antibodies against the B4, B5, C2, C4 and C5 response in mice and passive transfer of the anti-VLP sera
subgenotypes [32], which together caused nearly all endemics protected suckling mice from lethal challenge [40]. However,
worldwide over the past decade. Interestingly, antibodies EV71 VLP production in S. cerevisiae was at relatively low
induced by the NHRI (Taiwan) developed B4 vaccine strongly levels, averaging 0.25 mg of VLPs from 1 L of yeast culture
neutralized B1, B4, B5 and C4a and weakly neutralized C4b, [40] and thus hindered further product development. This
but had no neutralization effect on C2 [27]. The cause of this roadblock was recently lifted by the use of other yeast species.
discrepancy remains to be determined. Recently, the emer- Our unpublished data demonstrate that VLP expression in
gence of new EV71 genotypes D, E and F in India and Africa transgenic Pichia pastoris can reach high levels up to 100 mg
has been reported [8]. Therefore, it is important to evaluate of VLPs per liter culture and the resulting VLPs exhibit
existing vaccine candidates for their protective efficacy against immunogenicity and protective efficacy similar to those of
these newly emerged EV71 genotypes. insect cell- or S. cerevisiae-derived counterparts.
Collectively, VLP-based EV71 vaccine candidates have
2.2. EV71 virus-like particle vaccine attributes that are favorable for vaccine development,
including non-infectivity, potent immunogenicity, broad pro-
Virus-like particles (VLPs) are self-assembled particulate tective potential, and high yield. Indeed, a number of vaccine
structures. They morphologically resemble authentic virions makers in China, including Hualan Biological Bacterin and
but do not contain infectious viral genetic materials. VLPs can Shanghai Zerun Biotechnology, are developing VLP-based
be produced in large quantity in recombinant expression sys- EV71 vaccines. It is conceivable that some of these candi-
tems without the need of virus culture and inactivation. In date vaccines will soon enter clinical trials.
general, VLPs are safe and highly immunogenic, and represent
an important strategy for novel vaccine development. Thus far, 2.3. EV71 attenuated live vaccine
VLPs-based vaccines against three viruses, including hepatitis
B virus, human papillomavirus, and hepatitis E virus, have The success of oral polio vaccines (OPV) has provided
been licensed for human use [33]. guidance to the development of attenuated live EV71 vaccines.
In 2003, Hu et al. first reported the generation of EV71 Arita et al. introduced mutations into the 3D polymerase and
VLPs in insect cell by co-expressing both structural poly- 30 UTR of the genome of the prototype EV71 strain BrCr and
protein P1 and viral protease 3CD of the neu strain (C2 the resultant mutant virus, designated EV71(S1-30 ), was

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
4 Q. Liu et al. / Microbes and Infection xx (2014) 1e8

temperature-sensitive, less neurovirulent in monkeys, and had demonstrated by different groups [51e53]. To maximize the
limited capacity to invade the central nervous system [41]. vaccine potential of the identified EV71 neutralizing epitopes,
Inoculation of cynomolgus monkeys with EV71 (S1-30 ) pro- a peptide termed mTLNE, which is comprised of tandem
tected against subsequent lethal challenge with a virulent linked SP55, SP70 and VP2-28 epitopes, was expressed in
strain EV71 (BrCr-TR), demonstrating that the attenuated live Escherichia coli and tested for its immunogenicity in mice.
vaccine was protective [41]. Moreover, sera from the This mTLNE vaccine elicited neutralizing antibody that
EV71(S1-30 ) immunized monkeys could neutralize a panel of passively protected suckling mice from lethal infection [54].
EV71 strains belonging to genotypes A, B1, B4, C2, and C4 In general, mouse immunization studies show that the
[42], indicating that the vaccine induced broadly neutralizing neutralizing antibody titers following peptide vaccination are
antibodies. However, EV71 (S1-30 ) caused mild tremor at relatively low levels.
symptoms in monkeys [41]. Due to safety concerns, the
development of this attenuated live vaccine was discontinued. 3. Vaccines targeting CA16

2.4. Other types of EV71 vaccine candidates Next to EV71, CA16 is the second major causative agent of
HFMD. The virus was first identified in South Africa in 1951
The vaccine potential of EV71 subunit proteins has been [55]. The worldwide epidemiology of CA16 was recently
extensively studied. VP1 protein is the primary target that reviewed [10]. In China, CA16 infection accounts for up to
contains most of the linear neutralizing epitopes [43,44]. It has 75% HFMD cases in a single clinical survey in Beijing in 2007
been demonstrated that VP1 could induce neutralizing anti- [56]. Besides mild symptoms, a proportion of the CA16-
bodies that protected mice from low-dose viral challenge [45], infected patients may develop severe neurological complica-
whereas VP2 or VP3 could not elicit neutralizing antibodies tions, including brainstem encephalitis and acute flaccid pa-
even when co-immunized with Freund's adjuvant [46]. VP1 ralysis [57]. Fatal cases associated with CA16 infection have
delivered by a variety of vector systems, including Salmonella also been reported [10]. Therefore, it is necessary to develop
enterica serovar Typhimurium, Bifidobacterium longum, vaccines targeting CA16. Indeed, CA16 vaccine programs
transgenic tomato, milk, and DNA plasmid, was also capable have been launched by several academic institutions and
of inducing neutralizing antibodies [47e49,23]. However, the vaccine companies, and promising results from preclinical
neutralizing antibody titers elicited by VP1 immunization studies have been obtained (Table 2). Thus far, no CA16
were much lower than those induced by inactivated vaccines vaccine candidate has progressed into clinical trials yet.
or VLPs [45,46], probably because VP1 alone does not have
the conformation essential for the elicitation of potent 3.1. CA16 inactivated vaccine
neutralizing antibodies.
Peptide vaccines based on defined epitopes have also been The first proof-of-concept study of inactivated CA16 vac-
tested as potential EV71 vaccine candidate. Two neutralizing cine was reported by Cai et al. [58]. In that study, CA16 vi-
linear epitopes in the VP1, termed SP55 and SP70, respec- rions produced in Vero cells were inactivated with b-
tively, correspond to residues 163e177 and 208e222 of the propiolactone and then formulated with alum. Adult mice
VP1 protein [43]. Another neutralizing epitope within the VP1 immunized with the vaccine developed high-titer neutralizing
was mapped to residues 240e260 [44]. In addition, one pep- antibodies, and passive transfer of the anti-CA16 sera
tide named VP2-28, which consists of residues 136e150 of conferred significant protection against lethal infection with a
VP2, could be recognized by the neutralizing antibody clinical isolate CA16-G08. More importantly, active immu-
MAB979, suggesting VP2-28 represents a neutralizing epitope nization with the experimental vaccine fully protected mice
[50]. The ability of SP55- and SP70-based vaccines to induce from lethal challenge by a mouse-adapted strain CA16-MAV
neutralizing antibody response and confer protection has been [58]. The effectiveness of the inactivated CA16 vaccine

Table 2
CA16 vaccine candidates.
Vaccine type CA16 strain Cell substrate Antibody response Protective efficacy References
Inactivated virus CA16/SZ05 Vero Induce neutralizing antibodies in mice Protect mice against lethal challenge [58]
Inactivated virus CA16/G20 KMB17 Induce neutralizing antibodies in mice Protect mice against lethal challenge [61]
and non-human primates
Inactivated virus CA16/CC024 Vero Induce neutralizing antibodies in mice Protect mice against lethal challenge [60]
Inactivated virus CA16/419 Vero Induce neutralizing antibodies in mice Protect mice against lethal challenge [59]
and non-human primates
VLP CA16/SZ05 Sf9 Induce neutralizing antibodies in mice Protect mice against lethal challenge [62]
VLP CA16/GD09-119 Saccharomyces Induce neutralizing antibodies in mice Protect mice against lethal challenge [64]
cerevisiae
Peptide CA16/SZ05 Synthetic Induce neutralizing antibodies in mice ND [65]
ND: not determined.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
 Q. Liu et al. / Microbes and Infection xx (2014) 1e8 5

approach was confirmed in mouse models by two other similar region overlapping with the SP70 (residues 208e222) of
studies, in which different virus isolates were used as the EV71, whereas the other five CA16 epitopes have no overlap
vaccine strains [59,60]. Moreover, Yang et al. showed that an with known EV71 neutralizing epitopes [65]. In addition, the
inactivated CA16 experimental vaccine prepared from a sequences of these epitopes are extremely conserved among
human diploid cell line could also elicit neutralizing antibody different CA16 genotypes, suggesting that they could be used
in mice and rhesus monkeys [61]. Notably, the neutralizing to develop peptide-based broadly protective CA16 vaccines
antibody titers of the immunized rhesus monkeys sustained at [65]. However, peptides are usually not very immunogenic.
levels above 32 for at least 6 weeks. In addition, sera from For peptide-based CA16 vaccine development, future efforts
immunized animals could neutralize CA16 strains of both should focus on enhancing the immunogenicity of the epitopes
genotypes A and B, and maternal immunization with the using novel presentation strategies or formulated in novel
experimental vaccine protected newborn mice from lethal adjuvants, in order to maximize their protective potential.
challenge by either genotype A or B strains [61]. These results
demonstrate that the experimental vaccine can provide broad 4. Vaccines targeting other HFMD-causing
protection against heterologous CA16 strains. coxsackieviruses

3.2. CA16 virus-like particles vaccine Following EV71 and CA16, other major viral pathogens on
the list of HFMD-causing enteroviruses are CA6 and CA10
Using the same approach as for the EV71 VLP production, [11]. Before 2011, fewer HFMD cases were found to be
Liu et al. generated CA16 VLPs by co-expression of P1 and associated with CA6 or CA10 than with EV71 or CA16 in
3CD proteins in a baculovirus/insect cell expression system China. However, recent surveillance data indicate that CA6
[62]. CA16 VLPs were observed as spherical particles with a has emerged as the predominant causative agent in Southern
diameter of ~30 nm and were comprised of processed VP0, China and in Thailand since 2012 [66,67]. It has been shown
VP1 and VP3 proteins. Immunization with CA16 VLPs that 8.3% of CA6-infected patients developed meningoen-
induced high-titer neutralizing antibodies in mice. More cephalitis and presented with high fever compared with those
importantly, the anti-VLP mouse sera not only neutralized infected with EV71 [66]. The new epidemic pattern of HFMD
homologous (CA16/SZ05) and heterologous (CA16/GX08) has stimulated interest in developing vaccines for CA6 and
strains in vitro, but also conferred passive protection against CA10 as a part of a preparedness plan. Our group recently
lethal challenge with either homologous or heterologous CA16 produced recombinant VLPs of CA6 and CA10 and evaluated
strains [62]. These results demonstrate that neutralizing anti- their immunogenicity in mice (unpublished data). We found
bodies constitute the major protective immunity against CA16 that CA10 VLPs efficiently induced antibodies capable of
infection and that CA16 VLP is a promising vaccine candi- neutralizing CA10 infection in vitro. CA6 VLPs were also
date. Following this pioneering work, another group confirmed highly immunogenic, but the anti-CA6 VLP sera were not
that insect cell-produced CA16 VLPs were able to efficiently tested for neutralization because of the lack of a CA16 culture
elicit cross-reactive neutralizing antibodies against diverse system in our hands. The protective efficacy of these VLPs
CA16 strains [63]. Moreover, it has been recently shown that remains to be determined in virus challenge animal models.
CA16 VLPs could be produced in yeast, such as S. cerevisiae
[64] and P. pastoris (unpublished data), and these yeast- 5. Towards polyvalent vaccines for broad-spectrum
derived VLPs were as immunogenic as their counterparts protection against HFMD
produced in insect cells. Taken together, the existing data
clearly demonstrate that CA16 VLPs are safe and efficacious A variety of enteroviruses, particularly EV71, CA16, CA6
in preclinical studies, therefore strongly supporting subsequent and CA10, contributed to recent HFMD outbreaks
clinical development of VLP-based CA16 vaccines. [9,10,1,11,66]. Recombination and co-infection of these vi-
ruses increase the disease burden [10]. Although EV71 inac-
3.3. CA16 peptide vaccine tivated vaccines have proven efficacious in preventing EV71-
associated HFMD in clinical trials [19], their licensure faces
The identification of linear neutralizing epitopes of CA16 serious challenges [21,20]. Existing data from preclinical and
makes it possible to develop peptide-based CA16 vaccines. clinical studies indicate that monovalent EV71 vaccines could
Shi et al. screened a panel of 95 synthetic peptides spanning not provide protection against infections with other HFMD-
the VP1 region for reactivity to anti-VLP neutralizing sera and causing coxsackieviruses, such as CA16 [31,68,69]. Failure
the resultant high-binding peptides were further tested for their to provide broad protection against HFMD by monovalent
inhibitory effect on neutralization by anti-VLP sera [65]. After EV71 vaccine may not only significantly reduce the impact of
two rounds of selection, six non-overlapping peptides with its introduction, but also potentially mislead public into
significant neutralization-inhibition effect were identified. questioning the effectiveness of such an HFMD vaccine.
Mouse immunization showed that these peptides were able to Therefore, it is of great importance to develop a polyvalent
induce neutralizing antibodies against CA16, indicating that vaccine that can provide protection against EV71 and preva-
they represent six neutralizing epitopes. Among these six lent HFMD-causing coxsackieviruses [21,20]. Recently,
neutralizing epitopes, PEP71 (residues 211e225) locates at a bivalent EV71/CA16 vaccine candidates have been developed

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
6 Q. Liu et al. / Microbes and Infection xx (2014) 1e8

and shown to efficiently protect against EV71 and CA16 vaccine conferred protection against both EV71 and CA16 in
infection in mice (Table 3). These proof-of-concept studies the passive immunization/virus challenge experiments,
represent the successful first step towards polyvalent vaccines whereas the monovalent VLP vaccine only protected mice
for broad-spectrum protection against HFMD. from infections with the homologous virus, but not the het-
erologous one [68]. The existing data indicate that the bivalent
5.1. Bivalent EV71/CA16 inactivated vaccine EV71/CA16 VLP vaccine candidates are indeed capable of
eliciting balanced protective immunity against both viruses,
Cai et al. combined inactivated EV71 and CA16 together to and thus warrant further preclinical and clinical development.
formulate an experimental bivalent EV71/CA16 vaccine [69]. In both inactivated virus- and VLP-based bivalent EV71/
This bivalent vaccine was compared to monovalent EV71 and CA16 vaccine candidates, the EV71 antigens were derived
monovalent CA16 vaccines for their immunogenicity and from genotype C4 and appeared to induce broadly neutralizing
protective efficacy in mice. The results show that antisera from antibodies against all of the tested EV71 strains, including the
the bivalent vaccine immunized mice neutralized both EV71 prototype BrCr stain of genotype A and a panel of C4 isolates
and CA16 in vitro with titers similar to those induced by the [69,68]. In addition, previous studies have demonstrated that
monovalent EV71 or the monovalent CA16 vaccines, respec- the antibodies induced by the C4-derived monovalent EV71
tively, indicating an absence of immune interference between vaccines exhibited broad neutralization activity against the B4,
the two antigen components of the bivalent vaccine. More B5, C2, C4 and C5 subgenotypes [32], whereas the B4
importantly, the bivalent vaccine conferred protection in vivo vaccine-elicited antibodies weakly neutralized C4b and had no
against both EV71 and CA16 infections, whereas the mono- neutralization effect on C2 [27]. Together, these data indicate
valent vaccines only protected mice from homologous, but not that C4-derived antigens are likely the optimal EV71
heterologous, virus challenge [69]. A recent study by Lin et al. component to be included in polyvalent HFMD vaccine for-
also showed that a bivalent EV71/CA16 inactivated vaccine mulations. For other HFMD-causing viruses such as CA16,
candidate elicited serum antibodies capable of neutralizing extensive evaluation of the immunogenicity and protective
EV71 and CA16 in mice, although CA16-specific neutralizing efficacy and breadth of vaccine candidates are required to
antibody titers were lower than the EV71-specific ones [70]. identify the optimal vaccine strains for inclusion into poly-
Together, these data demonstrates the feasibility and effec- valent vaccines for broad protection against HFMD.
tiveness of the combination vaccine approach, which could be
extended to the development of polyvalent HFMD vaccines. 6. Conclusion

5.2. Bivalent EV71/CA16 VLP vaccine HFMD has emerged as a serious threat to the health of
children living in or traveling to the AsiaePacific regions. A
Gong et al. produced EV71- and CA16 VLPs in insect cells number of enteroviruses, mainly EV71, CA16, CA6 and
using baculovirus vectors and evaluated the immunogenicity CA10, are responsible for the recent large HFMD outbreaks.
of monovalent EV71 VLPs, monovalent CA16 VLPs, and a To safely prevent the disease, a polyvalent vaccine capable of
bivalent vaccine candidate comprising of both types of VLPs protecting against multiple prevalent HFMD pathogens is ur-
[63]. They found that the sera from mice immunized with the gently needed. The completion of phase 3 efficacy trials of
bivalent VLP vaccine formulated in alum and CpG adjuvants monovalent EV71 inactivated vaccines represents a successful
efficiently neutralized EV71 and CA16 infections in vitro [63]. first step towards this direction. Moreover, the promising re-
In another study by Ku et al., a bivalent vaccine comprised of sults from proof-of-concept studies of bivalent EV71/CA16
both EV71- and CA16-VLPs was found to induce high titers vaccine candidates demonstrate that it is feasible to induce
of serum antibodies capable of neutralizing EV71 and CA16 balanced, broadly protective immunity by a vaccine
in mice, and these antibodies sustained for at least 9 weeks combining antigens derived from several HFMD-causing vi-
after the last immunization. Moreover, the bivalent VLP ruses, therefore paving the way for the development of a

Table 3
Bivalent EV71/CA16 vaccine candidates.
Vaccine type Original virus Vaccine dosage/adjuvant Neutralizing antibody titer In vivo protective effect Reference
strains
Inactivated virus EV71/FY573 1 mg each, formulated with alum 1024e4096 against EV71; Protect mice against lethal [69]
CA16/G08 2048 against CA16. challenge with EV71 or CA16
Inactivated virus EV71/E59 0.2 mg each, formulated ~1000 against EV71; ND [70]
CA16/N5079 with alum or PELC/CpG ~100 against CA16
VLP EV71/G082 1 mg each, formulated with alum 1024e8192 against EV71; Protect mice against lethal [68]
CA16/SZ05 1024e4096 against CA16. challenge with EV71 or CA16
VLP EV71/FY 2.5 mg each, formulated 20e640 against EV71; ND [63]
CA16/09e7 with alum and/or CpG 40e320 against CA16.
ND: not determined.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
 Q. Liu et al. / Microbes and Infection xx (2014) 1e8 7

polyvalent HFMD vaccine. Hence, it is conceivable that a [19] Lu S. EV71 vaccines: a milestone in the history of global vaccine
broadly protective, polyvalent HFMD vaccine will come to the development. Emerg Microbes Infect 2014;3.
[20] Chu SC, Wang ET, Liu DP. Is a monovalent vaccine against enterovirus
market in the near future. 71 sufficient? A review of enterovirus 71 vaccine development based on
enterovirus surveillance in Taiwan. J Formos Med Assoc
Conflict of interest 2013;112:508e9.
[21] Lee MS, Tseng FC, Wang JR, Chi CY, Chong P, Su IJ. Challenges to
X.T. is an employee of Shanghai Zerun Biotechnology Co., licensure of enterovirus 71 vaccines. PLoS Negl Trop Dis 2012;6:e1737.
[22] Wong SS, Yip CC, Lau SK, Yuen KY. Human enterovirus 71 and hand,
Ltd.. The other authors declare no conflict of interest. foot and mouth disease. Epidemiol Infect 2010;138:1071e89.
[23] Lee MS, Chang LY. Development of enterovirus 71 vaccines. Expert Rev
References Vaccines 2010;9:149e56.
[24] Xu J, Qian Y, Wang S, Serrano JM, Li W, Huang Z, et al. EV71: an
[1] Repass GL, Palmer WC, Stancampiano FF. Hand, foot, and mouth dis- emerging infectious disease vaccine target in the far east? Vaccine
ease: identifying and managing an acute viral syndrome. Cleve Clin J 2010;28:3516e21.
Med 2014;81:537e43. [25] Chou AH, Liu CC, Chang CP, Guo MS, Hsieh SY, Yang WH, et al. Pilot
[2] Zeng M, Pu DB, Mo XW, Zhu CM, Gong ST, Xu Y, et al. Children of scale production of highly efficacious and stable enterovirus 71 vaccine
rural-to-urban migrant workers in China are at a higher risk of con- candidates. PLoS One 2012;7:e34834.
tracting severe hand, foot and mouth disease and EV71 infection: a [26] Cheng A, Fung CP, Liu CC, Lin YT, Tsai HY, Chang SC, et al. A phase I,
hospital-based study. Emerg Microbes Infect 2013;2. randomized, open-label study to evaluate the safety and immunogenicity
[3] Abzug MJ. The enteroviruses: problems in need of treatments. J Infect of an enterovirus 71 vaccine. Vaccine 2013;31:2471e6.
2014;68(Suppl. 1):S108e14. [27] Chou AH, Liu CC, Chang JY, Jiang R, Hsieh YC, Tsao A, et al.
[4] Weng KF, Chen LL, Huang PN, Shih SR. Neural pathogenesis of Formalin-inactivated EV71 vaccine candidate induced cross-neutralizing
enterovirus 71 infection. Microbes Infect 2010;12:505e10. antibody against subgenotypes B1, B4, B5 and C4A in adult volunteers.
[5] Robinson CR, Doane FW, Rhodes AJ. Report of an outbreak of febrile PLoS One 2013;8:e79783.
illness with pharyngeal lesions and exanthem: Toronto, summer 1957; [28] Li JX, Mao QY, Liang ZL, Ji H, Zhu FC. Development of enterovirus 71
isolation of group A Coxsackie virus. Can Med Assoc J 1958;79:615e21. vaccines: from the lab bench to phase III clinical trials. Expert Rev
[6] Chua KB, Kasri AR. Hand foot and mouth disease due to enterovirus 71 Vaccines 2014;13:609e18.
in Malaysia. Virol Sin 2011;26:221e8. [29] Zhu FC, Meng FY, Li JX, Li XL, Mao QY, Tao H, et al. Efficacy, safety,
[7] Huang PN, Shih SR. Update on enterovirus 71 infection. Curr Opin Virol and immunology of an inactivated alum-adjuvant enterovirus 71 vaccine
2014;5:98e104. in children in China: a multicentre, randomised, double-blind, placebo-
[8] Bessaud M, Razafindratsimandresy R, Nougairede A, Joffret ML, controlled, phase 3 trial. Lancet 2013;381:2024e32.
Deshpande JM, Dubot-Peres A, et al. Molecular comparison and [30] Zhu F, Xu W, Xia J, Liang Z, Liu Y, Zhang X, et al. Efficacy, safety, and
evolutionary analyses of VP1 nucleotide sequences of new African immunogenicity of an enterovirus 71 vaccine in China. N Engl J Med
human enterovirus 71 isolates reveal a wide genetic diversity. PLoS One 2014;370:818e28.
2014;9:e90624. [31] Li R, Liu L, Mo Z, Wang X, Xia J, Liang Z, et al. An inactivated
[9] Solomon T, Lewthwaite P, Perera D, Cardosa MJ, McMinn P, Ooi MH. enterovirus 71 vaccine in healthy children. N Engl J Med
Virology, epidemiology, pathogenesis, and control of enterovirus 71. 2014;370:829e37.
Lancet Infect Dis 2010;10:778e90. [32] Zhang H, An D, Liu W, Mao Q, Jin J, Xu L, et al. Analysis of cross-
[10] Mao Q, Wang Y, Yao X, Bian L, Wu X, Xu M, et al. Coxsackievirus A16: reactive neutralizing antibodies in human HFMD serum with an EV71
epidemiology, diagnosis, and vaccine. Hum Vaccin Immunother pseudovirus-based assay. PLoS One 2014;9:e100545.
2014;10:360e7. [33] Zhao Q, Li S, Yu H, Xia N, Modis Y. Virus-like particle-based human
[11] Blomqvist S, Klemola P, Kaijalainen S, Paananen A, Simonen ML, vaccines: quality assessment based on structural and functional proper-
Vuorinen T, et al. Co-circulation of coxsackieviruses A6 and A10 in hand, ties. Trends Biotechnol 2013;31:654e63.
foot and mouth disease outbreak in Finland. J Clin Virol 2010;48:49e54. [34] Hu YC, Hsu JT, Huang JH, Ho MS, Ho YC. Formation of enterovirus-
[12] Yamayoshi S, Fujii K, Koike S. Receptors for enterovirus 71. Emerg like particle aggregates by recombinant baculoviruses co-expressing P1
Microbes Infect 2014;3. and 3CD in insect cells. Biotechnol Lett 2003;25:919e25.
[13] Yang F, Zhang T, Hu Y, Wang X, Du J, Li Y, et al. Survey of enterovirus [35] Chung YC, Huang JH, Lai CW, Sheng HC, Shih SR, Ho MS, et al.
infections from hand, foot and mouth disease outbreak in China, 2009. Expression, purification and characterization of enterovirus-71 virus-like
Virol J 2011;8:508. particles. World J Gastroenterol 2006;12:921e7.
[14] Zhang HM, Li CR, Liu YJ, Liu WL, Fu D, Xu LM, et al. To investigate [36] Chung YC, Ho MS, Wu JC, Chen WJ, Huang JH, Chou ST, et al. Im-
pathogen of hand, foot and mouth disease in Shenzhen in 2008. munization with virus-like particles of enterovirus 71 elicits potent im-
Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi 2009;23:334e6. mune responses and protects mice against lethal challenge. Vaccine
[15] Pan H, Zhu YF, Qi X, Zhang YJ, Li L, Deng F, et al. Analysis on the 2008;26:1855e62.
epidemiological and genetic characteristics of enterovirus type 71 and [37] Lin YL, Yu CI, Hu YC, Tsai TJ, Kuo YC, Chi WK, et al.
Coxsackie A16 virus infection in Jiangsu, China. Zhonghua Liu Xing Enterovirus type 71 neutralizing antibodies in the serum of macaque
Bing Xue Za Zhi 2009;30:339e43. monkeys immunized with EV71 virus-like particles. Vaccine
[16] Yip CC, Lau SK, Zhou B, Zhang MX, Tsoi HW, Chan KH, et al. 2012;30:1305e12.
Emergence of enterovirus 71 “double-recombinant” strains belonging to [38] Ku Z, Ye X, Huang X, Cai Y, Liu Q, Li Y, et al. Neutralizing antibodies
a novel genotype D originating from southern China: first evidence for induced by recombinant virus-like particles of enterovirus 71 genotype
combination of intratypic and intertypic recombination events in EV71. C4 inhibit infection at pre- and post-attachment steps. PLoS One
Arch Virol 2010;155:1413e24. 2013;8:e57601.
[17] Zhang Y, Zhu Z, Yang W, Ren J, Tan X, Wang Y, et al. An emerging [39] Chung CY, Chen CY, Lin SY, Chung YC, Chiu HY, Chi WK, et al.
recombinant human enterovirus 71 responsible for the 2008 outbreak of Enterovirus 71 virus-like particle vaccine: improved production condi-
hand foot and mouth disease in Fuyang city of China. Virol J 2010;7:94. tions for enhanced yield. Vaccine 2010;28:6951e7.
[18] Liang Z, Mao Q, Gao F, Wang J. Progress on the research and devel- [40] Li HY, Han JF, Qin CF, Chen R. Virus-like particles for enterovirus 71
opment of human enterovirus 71 (EV71) vaccines. Front Med produced from Saccharomyces cerevisiae potently elicits protective im-
2013;7:111e21. mune responses in mice. Vaccine 2013;31:3281e7.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004
8 Q. Liu et al. / Microbes and Infection xx (2014) 1e8

[41] Arita M, Shimizu H, Nagata N, Ami Y, Suzaki Y, Sata T, et al. Tem- [57] Xu W, Liu CF, Yan L, Li JJ, Wang LJ, Qi Y, et al. Distribution of en-
perature-sensitive mutants of enterovirus 71 show attenuation in cyn- teroviruses in hospitalized children with hand, foot and mouth disease
omolgus monkeys. J Gen Virol 2005;86:1391e401. and relationship between pathogens and nervous system complications.
[42] Arita M, Nagata N, Iwata N, Ami Y, Suzaki Y, Mizuta K, et al. An Virol J 2012;9:8.
attenuated strain of enterovirus 71 belonging to genotype a showed a [58] Cai Y, Liu Q, Huang X, Li D, Ku Z, Zhang Y, et al. Active immunization
broad spectrum of antigenicity with attenuated neurovirulence in cyn- with a Coxsackievirus A16 experimental inactivated vaccine induces
omolgus monkeys. J Virol 2007;81:9386e95. neutralizing antibodies and protects mice against lethal infection. Vac-
[43] Foo DG, Alonso S, Phoon MC, Ramachandran NP, Chow VT, Poh CL. cine 2013;31:2215e21.
Identification of neutralizing linear epitopes from the VP1 capsid protein [59] Qi An W, Guo Su Z, Wen Pan R, Ping Yang B, Chao Zhang Y, Shi L,
of enterovirus 71 using synthetic peptides. Virus Res 2007;125:61e8. et al. The immunogenicity and protection effect of the BPL-inactivated
[44] Chang GH, Luo YJ, Wu XY, Si BY, Lin L, Zhu QY. Monoclonal antibody CA16 vaccine in different animal systems. Hum Vaccin Immunother
induced with inactivated EV71-Hn2 virus protects mice against lethal 2014;10:628e39.
EV71-Hn2 virus infection. Virol J 2010;7:106. [60] Li J, Chang J, Liu X, Yang J, Guo H, Wei W, et al. Protection from
[45] Wu CN, Lin YC, Fann C, Liao NS, Shih SR, Ho MS. Protection against lethal challenge in a neonatal mouse model by circulating recombi-
lethal enterovirus 71 infection in newborn mice by passive immunization nant form coxsackievirus A16 vaccine candidates. J Gen Virol
with subunit VP1 vaccines and inactivated virus. Vaccine 2014;95:1083e93.
2002;20:895e904. [61] Yang E, Cheng C, Zhang Y, Wang J, Che Y, Pu J, et al. Comparative
[46] Chou AH, Liu CC, Chang JY, Lien SP, Guo MS, Tasi HP, et al. Immu- study of the immunogenicity in mice and monkeys of an inactivated
nological evaluation and comparison of different EV71 vaccine candi- CA16 vaccine made from a human diploid cell line. Hum Vaccin
dates. Clin Dev Immunol 2012;2012:831282. Immunother 2014;10:1266e73.
[47] Chiu CH, Chu C, He CC, Lin TY. Protection of neonatal mice from lethal [62] Liu Q, Yan K, Feng Y, Huang X, Ku Z, Cai Y, et al. A virus-like particle
enterovirus 71 infection by maternal immunization with attenuated Sal- vaccine for coxsackievirus A16 potently elicits neutralizing antibodies
monella enterica serovar Typhimurium expressing VP1 of enterovirus 71. that protect mice against lethal challenge. Vaccine 2012;30:6642e8.
Microbes Infect 2006;8:1671e8. [63] Gong M, Zhu H, Zhou J, Yang C, Feng J, Huang X, et al. Cryo-electron
[48] Yu Z, Huang Z, Sao C, Huang Y, Zhang F, Ma G, et al. Oral immuni- microscopy study of insect cell-expressed enterovirus 71 and coxsack-
zation of mice using Bifidobacterium longum expressing VP1 protein ievirus a16 virus-like particles provides a structural basis for vaccine
from enterovirus 71. Arch Virol 2013;158:1071e7. development. J Virol 2014;88:6444e52.
[49] Tung WS, Bakar SA, Sekawi Z, Rosli R. DNA vaccine constructs against [64] Zhao H, Li HY, Han JF, Deng YQ, Li YX, Zhu SY, et al. Virus-like
enterovirus 71 elicit immune response in mice. Genet Vaccines Ther particles produced in Saccharomyces cerevisiae elicit protective immu-
2007;5:6. nity against Coxsackievirus A16 in mice. Appl Microbiol Biotechnol
[50] Liu CC, Chou AH, Lien SP, Lin HY, Liu SJ, Chang JY, et al. Identifi- 2013;97:10445e52.
cation and characterization of a cross-neutralization epitope of entero- [65] Shi J, Huang X, Liu Q, Huang Z. Identification of conserved neutralizing
virus 71. Vaccine 2011;29:4362e72. linear epitopes within the VP1 protein of coxsackievirus A16. Vaccine
[51] Foo DG, Alonso S, Chow VT, Poh CL. Passive protection against lethal 2013;31:2130e6.
enterovirus 71 infection in newborn mice by neutralizing antibodies [66] Li JL, Yuan J, Yang F, Wu ZQ, Hu YF, Xue Y, et al. Epidemic charac-
elicited by a synthetic peptide. Microbes Infect 2007;9:1299e306. teristics of hand, foot, and mouth disease in southern China, 2013:
[52] Tian X, Su X, Li X, Li H, Li T, Zhou Z, et al. Protection against coxsackievirus A6 has emerged as the predominant causative agent. J
enterovirus 71 with neutralizing epitope incorporation within adenovirus Infect 2014;69:299e303.
type 3 hexon. PLoS One 2012;7:e41381. [67] Puenpa J, Chieochansin T, Linsuwanon P, Korkong S, Thongkomplew S,
[53] Ye X, Ku Z, Liu Q, Wang X, Shi J, Zhang Y, et al. Chimeric virus-like Vichaiwattana P, et al. Hand, foot, and mouth disease caused by cox-
particle vaccines displaying conserved enterovirus 71 epitopes elicit sackievirus A6, Thailand, 2012. Emerg Infect Dis 2013;19:641e3.
protective neutralizing antibodies in mice through divergent mechanisms. [68] Ku Z, Liu Q, Ye X, Cai Y, Wang X, Shi J, et al. A virus-like particle
J Virol 2014;88:72e81. based bivalent vaccine confers dual protection against enterovirus 71 and
[54] Li YX, Zhao H, Cao RY, Deng YQ, Han JF, Zhu SY, et al. Recombinant coxsackievirus A16 infections in mice. Vaccine 2014;32:4296e303.
tandem multi-linear neutralizing epitopes of human enterovirus 71 eli- [69] Cai Y, Ku Z, Liu Q, Leng Q, Huang Z. A combination vaccine
cited protective immunity in mice. Virol J 2014;11:79. comprising of inactivated enterovirus 71 and coxsackievirus A16 elicits
[55] Sickles GM, Mutterer M, Feorino P, Plager H. Recently classified types balanced protective immunity against both viruses. Vaccine
of Coxsackie virus, group A; behavior in tissue culture. Proc Soc Exp 2014;32:2406e12.
Biol Med 1955;90:529e31. [70] Lin CW, Liu CC, Lu TC, Liu SJ, Chow YH, Chong P, et al. Immuno-
[56] Zhu J, Luo Z, Wang J, Xu Z, Chen H, Fan D, et al. Phylogenetic analysis genicity studies of bivalent inactivated virions of EV71/CVA16
of enterovirus 71 circulating in Beijing, China from 2007 to 2009. PLoS formulated with submicron emulsion systems. Biomed Res Int
One 2013;8:e56318. 2014;2014:670506.

Please cite this article in press as: Liu Q, et al., Towards broadly protective polyvalent vaccines against hand, foot and mouth disease, Microbes and Infection
(2014), http://dx.doi.org/10.1016/j.micinf.2014.11.004