Hand, foot and mouth disease (HFMD) is generally considered to be a common exanthematous illness mostly seen in children under 5 years old worldwide. HFMD is a febrile illness characterized by a maculopapular rash or blisters on the hands, feet, groin, and buttocks and is associated with painful ulcerative lesions of the mouth [1]. Infection is usually self-limiting but is highly contagious. HFMD often occurs in small epidemics in kindergartens or childcare centers since the virus is highly contagious [2]. The transmission route of HFMD is usually directly via the fecal-oral route or via nasopharyngeal secretions, such as saliva, sputum, or from the runny nose of an infected person [34]. An indirect viral propagation path can be mediated by contaminated material that was touched by an infected person [345]. HFMD mainly affects infants and children because high incidence and severity was shown in young children under the age of five with weakened immune systems, rather than in adults [36]. In rare cases, HFMD develops to produce severe complications in the central nervous system (CNS), respiratory, and cardiovascular systems, including aseptic meningitis, cerebella ataxia, poliomyelitis-like paralysis, acute brainstem encephalitis, cardiopulmonary failure, and fulminant neurogenic pulmonary edema associated with high mortality [7]. Children who recover from brainstem encephalitis can be left with significant neurologic after-effects. HFMD tends to occur in outbreaks during the warm temperatures of spring, summer, and fall, showing a strong seasonal pattern [36]. HFMD incidence is significantly higher when the humidity is increased, since infection with the causative virus increases sharply in hot and humid conditions [8].

HFMD is a common infection caused by a group of viruses. The two major causes of HFMD are coxsackievirus A16 (CVA16) and enterovirus 71 (EV71). EV71, belonging to the enterovirus A group from the Picornaviridae family [9], is the second most common cause of HFMD, after CVA16. Other kinds of enterovirus and other coxsackieviruses such as A5, A6, A7, A9, A10, B1, B2, B3, and B5 can also cause HFMD [3]. The incidence of HFMD caused by EV71 is less than that caused by CVA16; however, EV71 is a neurotropic virus, tends to be more severe, and is more likely to be associated with complications including neurological symptoms and heart disease such as myocarditis that can even be fatal [59].

In April 1957, HFMD was first reported in eight cases of affected children in New Zealand. Subsequently, the main cause of HFMD outbreaks during the 1960s was CVA16 [10]. EV71 was first isolated from a 9-month-old child with encephalitis in 1969, in the United States. After the confirmation of HFMD caused by EV71, HFMD occurred sporadically throughout the world [11]. However, since 1997, HFMD has occurred mostly as epidemics in countries of the Asia-Pacific region [12]. The HFMD outbreaks caused by identified EV71 genotypes in the Asia-Pacific region are summarized in Fig. 1. HFMD in China was first reported in 1981 in Shanghai. In 2008, HFMD occurred in China, with 488,955 cases and 126 deaths reported [13]. Between 2008 and 2012, 7,200,092 cases of HFMD and 2,457 deaths were reported to the Chinese Center for Disease Control and Prevention. Therefore, HFMD in China has become a serious disease and one of the leading causes of death in children [14]. In Taiwan during 1998, 129,101 cases of HFMD were reported, and there were 405 serious cases of neurological complications such as meningitis and encephalitis, leading to 78 deaths [12]. In 2008, HFMD caused by EV71 resulted in 387 severe cases and 14 fatalities in Taiwan [1215]. In Vietnam, HFMD caused by EV71 was first confirmed in 2003. From 2011 to 2012, over 200,000 patients with HFMD were hospitalized and 207 fatalities owing to HFMD were reported [16]. In South Korea, the first case of HFMD was reported in spring 2009 when the first death occurred [17]. In addition to these cases, many people suffered from HFMD in Japan in 1997 and 2000, in Singapore during 2000 and in Malaysia during 1997. Continuing to the present, HFMD has been continuously reported as occurring endemically in many Asia-Pacific countries [18]. C4 is the most prevalent EV71 genotype found recently in China, Hong Kong, Korea, and Vietnam, whereas B5 is the most common genotype circulating in Japan, Malaysia, and Taiwan.

EV71 is a positive single-stranded RNA virus that belongs to the Picornaviridae family, genus Enterovirus, species Enterovirus A [19]. Human enterovirus serotypes include four species, enterovirus A, B, C, and D, which are distinguished by sequence differences, genome organization, and biological properties [20]. The EV71 particle is non-enveloped, icosahedral, and 20-30 nm in diameter (Fig. 2). The virus capsid contains a single-stranded, positive-sense, polyadenylated viral RNA. The RNA size is approximately 7.4 kb. The coding region of EV71 is divided into three sub-regions, namely the P1, P2, and P3 regions. The P1 region encodes four structural viral proteins including VP1, VP2, VP3, and VP4. The P2 and P3 regions encode seven non-structural proteins including 2A-2C and 3A-3D that make up the proteases used in proteolytic cleavage to release structural proteins, and the RNA-dependent RNA polymerase [521]. The four structural proteins VP1 to VP4 assemble to form a protomer. Five protomers make up a pentamer, and 12 pentamers together form a virion enclosing the viral genome [22]. VP1, VP2, and VP3 are exposed on the capsid surface to the external environment and these proteins can therefore be easily targeted by the host immune response. In particular, VP1 contains the major neutralization epitopes, which can be used as biomarkers to assess vaccine potency. VP4 is present at an inner location within the capsid [23]. EV71 is classified into the three genotypes A, B, and C and then further divided into 11 sub-genotypes. Genotype A has the prototype strain (BrCr) only, whereas genotypes B and C have five sub-genotypes including B1 to B5, and C1 to C5, respectively. Such sub-classification of EV71 is dependent on sequence variations in VP1 [24]. Unlike eukaryotic mRNAs, all Picornaviridae viral genomes contain an internal ribosomal entry site (IRES) instead of a 5′-cap structure. Following entry of the virus particle into host cells and release of the viral genome from an endosome into the cytoplasm, viral RNA can be translated in an IRES-dependent manner. During translation and genome replication, the viruses require not only IRES-specific trans-acting factors (ITAFs) but also several host factors including T-cell-restricted intracellular antigen 1 (TIA-1) and TIA-1 related protein (TIAR) for effective viral replication [25]. The interaction of TIA-1 and TIAR with the 5′ untranslated region of the viral genome can positively enhance viral replication, although ITAFs usually regulate viral growth at the translational step [26].

Several viral receptors that are responsible for entry of EV71 into host cells have been characterized (Fig. 3). These receptors include human scavenger receptor B2 (hSCARB2), human P-selectin glycoprotein ligand 1 (PSGL-1), dendritic cell specific intercellular adhesion molecule-3 grabbing nonintegrin (DC-SIGN), annexin A2 (Anx2), heparan sulfate (HS), and sialylated glycan [272829303132]. SCARB2 belongs to the scavenger receptor class B subfamily and is also known as lysosomal integral membrane protein II, LGP85, or CD36b like-2. SCARB2 is known to participate in endocytosis, membrane transport, and reorganization of the endosomal/lysosomal compartments [33], and usually shuttles between the internal membrane and the plasma membrane. When it is present at the cell surface, SCARB2 is a type III double-transmembrane protein with a large extracellular domain and short cytoplasmic domains. Human SCARB2 on the cell surface binds to EV71, and then EV71 is internalized by the clathrin-mediated pathway. Mouse cells transformed with human SCARB2 are susceptible to all EV71 strains, and are capable of binding, internalization, and uncoating of the virus [34]. However, murine SCARB2 does not act as a receptor for EV71. Human SCARB2 is expressed in a variety of cell types, including neurons in the CNS, which may be involved in direct infection of the brain by EV71 [35]. PSGL-1 is a sialomucin-like protein and is generally expressed in leukocytes, with involvement in the tethering and rolling of leukocytes on the vascular endothelium during leukocyte infiltration. EV71 can infect T cells expressing PSGL-1, and infection of such cells can be inhibited by PSGL-1–specific antibodies [28]. After attachment, viral particles can be internalized by a caveolin-dependent pathway. However, PSGL-1 is unlikely to serve as the receptor on non-leukocytes such as neuronal, epithelial, and fibroblast cells since it is expressed primarily on leukocytes. Instead, leukocytes expressing PSGL-1 that become infected with EV71 might migrate and spread the virus into the CNS. PSGL-1 is not a receptor for all EV71 strains since approximately 80% of EV71 isolates are expected to lack binding to PSGL-1 based on sequence data [36]. DC-SIGN (also known as CD209) is a C-type lectin receptor present on the surface of both macrophages and dendritic cells that is involved in phagocytosis. Monocyte-derived dendritic cells can bind EV71 through DC-SIGN, and these virus particles can be transferred to susceptible cells to spread infection [3237]. Anx2, a member of the annexin family, is a calcium-dependent phospholipid-binding protein that is involved in sorting of endosomes inside the cell and in anticoagulant reactions outside the cell. Anx2 can directly bind to VP1 of EV71 but not CVA16. However, subsequent entry and uncoating of EV71 have not yet been reported. HS is a linear polysaccharide found in all animal tissues, and two or three HS chains are attached to a proteoglycan (HSPG) in close proximity to the cell surface. HS can bind to various ligands and regulates a wide variety of biological processes. HS has been shown to bind a number of viruses including respiratory syncytial virus as well as EV71 [30]. Many glycan-binding proteins in pathogens recognize sialic acid or its modified forms expressed on the glycan chains of glycolipids and glycoproteins. Sialylated glycan is generally enriched in the epithelial cells of the respiratory and gastro-intestinal tracts. EV71 may use the sialylated glycan on intestinal epithelial cells as a receptor, although evidence for a direct interaction has not been reported [31].

EV71 binds viral receptors and enters the host cell via receptor-mediated endocytosis involving clathrin- or caveolin-mediated pathways [38]. Human SCARB2 on the cell surface binds to the EV71 virion and the SCARB2-EV71 complex is then internalized. When the virus-receptor complex reaches the endosome or lysosome, SCARB2 can initiate a conformational change at acidic pH leading to the uncoating of the virion. In contrast to poliovirus and group B coxsackieviruses, low pH is critical for EV71 uncoating. EV71 RNA is released into the cytoplasm following a series of structural changes in the viral capsids [5]. The viral RNA can be directly translated into a large polypeptide followed by prompt cleavage by the viral proteases (Fig. 2). Host cellular protein synthesis is shut down by the viral protease 2A, leaving viral protein synthesis unaffected. Viral replication takes place in the vesicle membrane complex by the RNA-dependent RNA polymerase 3Dpol. Following assembly of viral RNA and capsids, mature infectious virus particles are released when the infected host cell is lysed.

Although mechanisms for viral infection and host defense have been proposed, the details of the propagation process and pathology of EV71 have not been completely elucidated [39]. To obtain more knowledge about EV71 pathogenesis, various mouse and monkey animal models have been used. Neurological complications including ataxia, tremors, and flaccid paralysis are observed in EV71-infected cynomolgus monkeys, similar to the complications that occur in humans [40]. However, it is difficult to justify the use of monkeys as an infection model for EV71 infection owing to ethical and economic reasons. There have been reports of animal models of neonatal mice infected with mouse-adapted EV71 strains [41]. Unfortunately, virus replication in mice occurs mainly in muscle and fat cells, unlike in humans. Mice over 3-weeks-old do not show sensitivity to EV71; therefore, the suckling mouse model cannot be utilized to challenge with EV71 for a vaccine candidate. To allow established immunity in suckling mice, maternal immunization can be performed. The offspring of immunized female mice have high titers of maternal IgG specific to EV71, almost reaching the level seen for the immunized mother. These suckling mice when challenged with EV71 did not develop CNS complications and survived a lethal dose [42]. To mimic a weakened immune system, interferon (IFN) receptor-deficient mice with impaired viral defenses have been used. AG129 mice which lack receptors for both IFN-α/β and IFN-γ were susceptible to EV71 infection [43]. Treatment with neutralizing antibody for type I IFN reduces survival and increases disease progression following EV71 infection [44]. These results suggest that IFNs play an important role in antiviral defense against EV71 infection.

Recently, transgenic (Tg) mice with human SCARB2 were developed [4546]. Three-week-old hSCARB2-Tg mice infected with EV71 via intracranial, intravenous, and intraperitoneal routes exhibited ataxia, paralysis, and death [45]. The pathological features in these mice were similar to those of EV71 encephalitis in humans. hSCARB2-Tg mice independently generated by another group were challenged with clinical isolates of EV71 and CVA16, resulting in an HFMD-like neurological syndrome and lethal paralysis [46]. Treatments with EV71-specific neutralizing antibody or pre-immunization with vaccine candidates were tested in the hSCARB2-Tg mice, and showed protection against subsequent lethal challenge with EV71 [4647]. These results show that hSCARB2-Tg mice can be a useful model to assess anti-EV71 treatments. Human PSGL-1 as well as SCARB2 are thought to be functional receptors for EV71 in human cells [2836]. When human PSGL-1 Tg mice were established, Tg expression of human PSGL-1 failed to enhance infectivity of EV71. However, human PSGL-1 expression did facilitate viral replication and symptom severity, but only at the earlier stages of infection. These results show that human PSGL-1 alone is not sufficient to mediate EV71 infection but may act as a cofactor for viral infection in mice in the early stages of EV71 infection.

There have been several types of EV71 vaccine candidates, including attenuated strains, inactivated whole-virus, virus like particles (VLP), recombinant proteins, recombinant vectors, and peptide vaccines.

In December 2015, the Food and Drug Adminstration (FDA) of China approved the first vaccine against EV71, made by the Institute of Medical Biology at the Chinese Academy of Medical Sciences. In January 2016, the Chinese FDA approved a second EV71 vaccine made by Sinovac Biotech. Approved inactivated whole EV71 vaccines are now in commercial production.

EV71 has caused frequent outbreaks in the Asia-Pacific region during the past two decades and has been considered a significant public health problem for the young generation in the post-poliovirus eradication era. A variety of strategies to develop anti-EV71 agents such as small molecules or antibodies have been investigated [68].

Severe neurological symptoms of HFMD caused by EV71 make HFMD infection a serious public health problem for young children in countries of the Asia-Pacific region. In the absence of effective treatment, the development of efficacious vaccines to prevent EV71 outbreaks has been a national priority in some countries. Currently there are two EV71 vaccines that have been approved and are commercially available in China. To enable the worldwide use of EV71 vaccine, the applicability against various EV71 pandemic strains needs to be demonstrated [79]. Therefore, time is required after the EV71 vaccines enter the market before effective protection against severe HFMD can be achieved. Current formalin-inactivated EV71 vaccines can protect against EV71 but not against CVA16 infection, which is the most common cause of HFMD. The development of a bivalent formalin-inactivated EV71/CVA16 vaccine or multivalent vaccine including other prevalent pathogenic enteroviruses should be the next step [82]. In the future, novel modes of administration need to be investigated to provide a safe needle-free EV71 vaccine for young children. To enhance immunity against infection at mucosal tissue barriers, mucosal delivery of vaccine antigens will be helpful to develop improved vaccines. For such vaccines, delivery technology allowing optimal mucosal delivery without additional immunostimulatory adjuvant materials need to be investigated.