Coxsackievirus B3 HFMD animal models in Syrian hamster and rhesus monkey

Abstract

Coxsackievirus B3 (CVB3) is the pathogen causing hand, foot and mouth disease (HFMD), which manifests across a spectrum of clinical severity from mild to severe. However, CVB3-infected mouse models mainly demonstrate viral myocarditis and pancreatitis, failing to replicate human HFMD symptoms. Although several enteroviruses have been evaluated in Syrian hamsters and rhesus monkeys, there is no comprehensive data on CVB3. In this study, we have first tested the susceptibility of Syrian hamsters to CVB3 infection via different routes. The results showed that Syrian hamsters were successfully infected with CVB3 by intraperitoneal injection or nasal drip, leading to nasopharyngeal colonization, acute severe pathological injury, and typical HFMD symptoms. Notably, the nasal drip group exhibited a longer viral excretion cycle and more severe pathological damage. In the subsequent study, rhesus monkeys infected with CVB3 through nasal drips also presented signs of HFMD symptoms, viral excretion, serum antibody conversion, viral nucleic acids and antigens, and the specific organ damages, particularly in the heart. Surprisingly, there were no significant differences in myocardial enzyme levels, and the clinical symptoms resembled those often associated with common, mild infections. In summary, the study successfully developed severe Syrian hamsters and mild rhesus monkey models for CVB3-induced HFMD. These models could serve as a basis for understanding the disease pathogenesis, conducting pre-trial prevention and evaluation, and implementing post-exposure intervention.

Highlights

• •Syrian hamsters and rhesus monkeys infected with CVB3 displayed symptoms resembling those observed in HFMD.
• •Syrian hamsters exhibit severe symptoms following both intraperitoneal and nasal drip infections.
• •Syrian hamsters with nasal infections exhibit more severe symptoms.
• •Rhesus monkeys infected via nasal drops have viral excretion and pathological damage similar to clinical mild infection.
• •The CVB3 Syrian hamster and rhesus monkey models can simulate different symptoms, offering valuable tools for HFMD studies.

1. Introduction

Coxsackievirus B3 (CVB3), a member of the Enterovirus genus within the family of Picornaviridae, has a genome of approximately 7.4 ​Kb. The virus encodes four capsid proteins and other non-structural proteins. Epidemiological surveys reveal a gradual increase of CVB3 rate, indicating that CVB3 may become the predominant enterovirus pathogen following enterovirus 71 (EV71), coxsackievirus A16 (CVA16), coxsackievirus A6 (CVA6), and coxsackievirus A10 (CVA10). This rising trend emphasizes the importance of vigilance in combating the infectious diseases (Garmaroudi et al., 2015; Zander et al., 2014; Tian et al., 2014; Yao et al., 2015). The typical manifestations of hand, foot, and mouth disease (HFMD) caused by enteroviruses include herpes on the hand, foot, knees, buttocks, mouth, and tongue. Clinically, CVB3 infections have been associated with viral myocarditis and aseptic meningitis (Wu et al., 2020). While the majority of patients can recover successfully, some experience varying degrees of severity. This highlights the urgent need to investigate the complex pathogenic mechanisms of CVB3, with the ultimate goal of developing effective vaccines and medications. Therefore, it is necessary to establish suitable animal models.

Rodent models of CVB3, primarily rats and mice, have been reported. Unfortunately, these models have failed to accurately replicate all clinical manifestations. For instance, intraperitoneal infection studies on mouse and suckling mouse model can recreate various acute complications, such as acute viral myocarditis, pancreatitis, and diabetes mellitus. However, HFMD has not been reported in those models (Xuan et al., 2016; Kong et al., 2022; Basavalingappa et al., 2020). Although mice are cost-effective and stable, their immune systems differ significantly from humans. Numerous studies have shown that Syrian hamsters exhibit greater similarities to humans in terms of disease symptoms, pathogenesis, and immune response compared to mice (Ogg et al., 2013; Prescott et al., 2014). Consequently, Syrian hamsters have been extensively used to study pathogenicity and host immune response, and in vivo pre-screening for antiviral drugs (Wang et al., 2022; Miao et al., 2019; Hirose and Ogura, 2018; Wold et al., 2019; Wahl-Jensen et al., 2012). Notably, Syrian hamsters have been used to study numerous human viral diseases caused by over 70 distinct RNA viruses. In particular, they have exhibited symptoms similar to HFMD after EV71 infection (Phyu et al., 2016). In previous studies, CVB3 was administered intraperitoneally, resulting in rapid onset and significant disease in the infected animals. However, several studies have reported successful models of EV71, CVA16, and CVA10 nasal drop infections (Wang et al., 2014; Wang et al., 2017; Duan et al., 2022). Current studies have suggested that aerosol and respiratory tracts as potential natural transmission routes for enteroviruses (Hoorn and Tyrrell, 1965). Therefore, we initially infected Syrian hamsters with CVB3 through both intraperitoneal injection and nasal drip to assess the extent of disease and the susceptibility of different infection modes.

Considering the physiological and anatomical similarities between non-human primates (NHPs) and humans, especially in their susceptibility to pathogens, NHPs have proven to be advantageous for establishing animal models to study viral infections (Nakamura et al., 2021). Rhesus monkeys play a crucial role in the development of enterovirus-related vaccines, such as poliovirus, EV71, CVA16, CVA10, and others (Sestak et al., 2018). Han's team has established a rhesus monkey model using intraperitoneal injection of the SSM-CVB3 strain, originating from golden monkeys (Han et al., 2012). This model has provided critical insights into the pathogenic mechanism of CVB3 and the development of novel vaccines and therapeutic methods. However, the study used intraperitoneal injection, which does not align with the natural route of enterovirus infection. Additionally, the strain used was derived from golden monkeys, which may exhibit significant differences in host adaptability compared to human-derived viruses. Recent studies indicate that CVB3 is associated with long-term disease, such as type I diabetes (Mone et al., 2023). Therefore, our primary objective was to establish a CVB3 rhesus monkey animal model based on experimental results in Syrian hamsters with different infection modes. We expect that this model will accurately replicate the infectious pathogenesis and facilitate the exploration of the intricate host immune response process, thereby enabling the evaluation of efficacy and adverse effects of potential drugs or vaccines.

2. Materials and methods

2.1. Animals

Eighteen SPF-class Syrian hamsters, aged four weeks, female, and weighing 65–85 ​g, were utilized. These Syrian hamsters were obtained from Beijing Vitonlivar Laboratory Animal Technology Co., Ltd. (Laboratory Animal Production License No.: SCXK [Beijing] 2016–0011). All experimental animals were housed in a barrier environment at the Institute of Medical Biology, Chinese Academy of Medical Sciences (IMBCAMS, Kunming, China) (Experimental Animal Use License No.: SYXK [Dian] K2021-0001). Infant rhesus monkeys (3 months old, female) were acquired from IMBCAMS, Yunnan, China [Laboratory Animal Production License No.: SCXK (Dian) K2020-0005]. The study protocol was ethically approved by the Ethics Committee of IMBCAMS (DWSP202107019). Throughout the research, the monkeys were kept at animal biosafety level 2 facility [Experimental Animal Use License No.: SYXK (Dian) K2020-0006], maintained in a climate-controlled room (with a temperature range of 18–25 ​°C and humidity levels between 30% and 70%). They were provided with appropriate nutrition, including food and fruits, and have access to water at their discretion.

All animal experiments were conducted in accordance with the guidelines of the National Institutes of Health (NIH) and the International Council for Laboratory Animal Science (ICLAS). And the 3R principles (Replacement, Reduction, and Refinement) were strictly followed. Throughout the study, efforts were made to reduce the number of used animals and to refine the experimental procedures to minimize pain and distress.

2.2. Cells and virus

Vero cells (ATCC CRL-1586) were cultured in a complete medium consisting 10% fetal bovine serum (Atlanta Biologicals), 1 ​mmol/L glutamine (Gibco), 100 U/mL penicillin (Gibco), and 100 ​μg/mL streptomycin (Gibco). The cells were incubated at 37 ​°C in a 5% CO₂ humidified incubator. CVB3 (Gene Bank serial number: A278741), kindly provided by the IMBCAMS researcher Shaohui Ma, was propagated and its virus titer was determined using the Reed-Muench method. All infection procedures performed in a biosafety level 2 laboratory adhered to procedures approved by the Institutional Biosafety Committee. The animal experiments consisted two parts. The first part involved 18 Syrian hamsters, while the second part involved 6 rhesus monkeys. Prior to the experiment, the neutralizing antibody immunoglobulin G (IgG) against CVB3 was tested to confirm the animals’ negativity. Syrian hamsters and rhesus monkeys were both infected with CVB3. The first part experimental results guided the design of the second part. The first part of the experiment comprised three groups, as detailed in Table 1. In the intraperitoneal injection group (IG), 10^5.75 CCID₅₀/100 ​μL CVB3 virus suspension was administered to Syrian hamsters through an intraperitoneal injection. The intranasal injection group (NG) involved infecting the Syrian hamsters with the same virus suspension via nasal drip. The control group (CG) consisted of six untreated Syrian hamsters. The second part of the experiment included two groups of rhesus monkeys, as shown in Table 2. Group A consisted of three monkeys (A1, A2, A3), infected with CVB3 virus suspension through nasal drip using a 1 ​mL syringe, with a volume of 2 ​× ​10^6.75 CCID₅₀/200 ​μL. In contrast, the three rhesus monkeys in Group B (B1, B2, B3) did not receive any treatment.

Table 1.

Experimental groups for CVB3 infection in Syrian hamster.

Group	Hamster ID	Gender	Agea	Virus doseb	Weight (g)c	Infection route
IG	I1–I6	♀	4	105.75	76.72 ​± ​6.88	Intraperitoneal injection
NG	N1–N6	♀	4	105.75	74.07 ​± ​2.94	Nasal drip
CG	C1–C6	♀	4	0	81.06 ​± ​1.76	None

^aMonths after the birth.

^b50% cell culture infectious doses in 0.1 ​mL inoculation volume.

^cMean weight ± standard deviation.

Table 2.

The Syrian hamsters different organs histopathological damage after CVB3 infection.

Group	IG	NG	CG
Brain	Neuroinflammatory changes	A small amount of neuronal cell body pyknosis	None
Heart	Myocardial cell necrosis, edema, and vacuolation changes (++)	Myocardial cell necrosis, edema, and vacuolation changes (+++)	None
Liver	Hepatocellular ballooning, hepatic sinusoidal stenosis (++)	Hepatocellular ballooning, hepatic sinusoidal stenosis (+++)	None
Lung	Alveolar septal thickening, inflammatory cell infiltration (+)	Alveolar septal thickening, inflammatory cell infiltration (++)	None
Spleen	White marrow damage, reduced numbers, decreased lymphocytes (++)	White marrow damage, reduced numbers, decreased lymphocytes (+)	None
Pancreas	None	None	None

Note: +, mild tissue damage; +++, moderate tissue damage; ++++, severe tissue damage.

2.3. Animal evaluation and sample collection

In the Syrian hamsters experiment, the animals were monitored daily for various clinical signs, including basic health status, herpes on the limbs, and other symptoms following CVB3 infection. The clinical manifestations of CVB3-infected Syrian hamsters were scored based on the following criteria: General performance (normal/alert/free movement, 0; slow/quiet/alert/free movement, 5; quiet/hunched back/alertness/limited movement, 10; lethargy/inability to move/no alertness, 15); Food consumption (normal, 0; decreased appetite, 2; anorexia, 5). The daily body weight and rectal temperature of the Syrian hamsters were recorded. The rectal temperature was measured using an Omron electronic digital stick thermometer, inserting a soft probe 2 ​cm into the anal verge of the rectum for 1 ​min. Pharyngeal swabs, fecal swabs, and feces were collected daily from days 0–14 for viral copies number testing, with the samples stored in 1.5 ​mL ​EP tubes. On 7 days post-infection (d.p.i), three Syrian hamsters from each of the three groups were randomly selected and euthanized following deep anesthesia. The autopsy collected tissues from the brain, heart, liver, lung, spleen, and pancreas. These tissues were subsequently used for virological, histological, and immunohistochemical (IHC) assays. Heart blood was collected from all animals on 14 ​d.p.i for viral load assays, blood cell count, and biochemical (liver function, cardiac enzymes) assays (Fig. 1A). A 200 ​μL anticoagulated whole blood was used for viral load assays, while 100 ​μL anticoagulated whole blood was analyzed on a benchtop hematology analyzer (Sysmex XT-2000i, Hemavet, Japan) for routine measurements of various biomarkers, such as blood cell counts. Non-anticoagulated whole blood (1 ​mL) was allowed to stand for 30 ​min at 4 ​°C, followed by the collection of serum after centrifugation at 3500 ​rpm for 30 ​min. The collected serum was analyzed for liver function indicators (Glutamic-oxaloacetic transaminase AST and Alanine aminotransferase ALT) and cardiac enzymes (Lactate dehydrogenase LDH, Creatine kinase CK, Creatine kinase isoenzyme CKMB) using a biochemical assay instrument (Mindrzy BS-200 Automatic biochemical analyzer, Shenzhen Meirui Biology Co., Ltd., China).

Fig. 1.

The manifestation of HFMD in CVB3-infected Syrian hamsters. Eighteen Syrian hamsters were categorized into three groups: the intraperitoneal injection group (IG), the nasal drip group (NG), and the control group (CG). The IG and NG groups were infected with 100 ​μL of 10^6.75 CCID₅₀/mL, while the CG group served as the untreated control. Daily pharyngeal, nasal, and fecal swabs were collected following infection. At 7 ​d.p.i, three randomly selected tissue specimens from each group were obtained for pathogenic and histopathological analysis following euthanasia. Moreover, at 14 ​d.p.i, heart blood was collected from the remaining Syrian hamsters after anesthesia for biochemical analysis. A The image visualizes the herpes lesions on the mouth of Syrian hamsters post-CVB3 infection. B The clinical manifestations of CVB3 infection in Syrian hamsters, which were scored based on daily performance and dietary status. The values are mean ​± ​standard error, with n ​= ​6. C The results of hematoxylin & eosin and immunohistochemical analyses on herpes tissue from Syrian hamsters at 3 ​d.p.i are presented. The tissues exhibited prominent inflammatory cell infiltration and CVB3 antigen expression. Brown color indicates DAB-stained positive antigen signal, while blue-purple color represents the nucleus (red dashed box indicating the positive antigen signal). The scale bar is 50 ​μm. D The clinical manifestations (daily performance and dietary status) observed scores of CVB3 infection in Syrian hamsters. Daily performance: normal/alert/active-0; slow/quiet/alert/active-5; quiet/arching back/alert/limited activity-10; depressed/inactive/unaware-15; dietary status: normal diet-0; reduced appetite-2; anorexia-5. The values are mean ​± ​standard error, n ​= ​6; E The body temperature changes in Syrian hamsters post-CVB3 infection. F The alterations in body weight. G The blood cell changes in Syrian hamsters following CVB3 infection. Data are presented as mean ​± ​standard deviation. Statistical analyses were performed using t-test with GraphPad Prism software. ∗∗, P ​< ​0.001.

In the rhesus monkey experiment, the animals underwent daily clinical examination during the 0–30 ​d.p.i. Body temperature and weight measurements were recorded for the first 14 days. Pharyngeal, fecal swabs and feces were collected to monitor viral replication and excretion status. Blood samples, including anticoagulated and non-anticoagulated types, were collected intravenously at multiple time points (day 0, 1, 3, 5, 7, 9, 11, 13, 15, 21, 30) during the experiment. At these 11 time points, the non-anticoagulated blood was analyzed for viral copy number, while the anticoagulated blood was measured for blood cell counts, as previously described (Sysmex XT-2000i, Hemavet, Japan). Neutralizing antibodies against CVB3 were assessed using serum samples collected at 0, 7, 15, and 30 ​d.p.i. Inflammatory markers and biochemical functions were measured at 15, 21, and 30 days. At the 30 ​d.p.i, two rhesus monkeys were euthanized by administering a dose of 0.5 ​mg/kg ketamine for deep anesthesia. Tissue samples from the euthanized animals were removed and fixed in 4% paraformaldehyde for 12 ​h before being embedded in paraffin for histopathological testing. Additionally, another tissue samples were snap-frozen in liquid nitrogen for subsequent RNA isolation.

2.4. Quantitative detection of CVB3 RNA by qRT-PCR

In the study, pharyngeal, nasal, fecal swabs, 200 ​μL blood, and 100 ​mg tissue samples were collected from infected Syrian hamsters and rhesus monkeys. Total RNA was extracted from these samples using the traditional TRIzol-A+ method, and the resulting total RNA was eluted in DEPC water to a final 20 ​μL volume. A One Step TB Green PrimeScript TM PLUS RT-PCR Kit (Perfect Real Time) (Takara, Japan) and a 7500 Fast Real-Time RT-PCR System (Applied Biosystems, Foster City, CA, USA) were employed for quantification. The reaction solution, which comprised a total 20 ​μL volume, consisted of the following components: 2 ​× ​One Step TB Green RT-PCR Buffer 4 (10 ​μL), TaKaRa Ex Taq HS Mix (1.2 ​μL), PrimeScript PLUS RTase Mix (0.4 ​μL), PCR Forward Primer (10 ​μmol/L) (0.8 ​μL), PCR Reverse Primer (10 ​μmol/L) (0.8 ​μL), Total RNA (2 ​μL), and RNase Free dH₂O (4.8 ​μL). According to the instruction steps, real-time PCR was performed under the following conditions: initial steps at 42 ​°C for 5 ​min and 95 ​°C for 10 ​s; followed by 40 cycles at 95 ​°C for 5 ​s, and 60 ​°C for 30 ​s. The primers used were CVB3–VP1–F (379–400) with the sequence 5′-GCTCACCTTCGTTATTACTAG-3′ and CVB3-VP1-R (569–589) with the sequence 5′-AATGCTCAAGAATGGAATGG-3'. A standard reference curve was generated by measuring the sequence dilution of the viral RNA produced by in vitro transcription of CVB3 VP1 DNA. The viral loads were analyzed by qRT-PCR based on the standard curve. The viral RNA relative copy number for each sample was calculated using the following mathematical formula: [(μg RNA/μL)/(molecular weight)] ​× ​Avogadro number ​= ​viral copy number/μL. A viral load below 10 copies was considered negative. RNA from the control group was evaluated simultaneously in each TB Green RT-PCR.

2.5. Histological and immunohistochemical examination

Following pathological autopsy of Syrian hamsters, the brain, heart, liver, lung, spleen, and pancreas tissues were collected for histopathological and immunohistochemical (IHC) examination. Similarly, forty-three different tissues from executed rhesus monkeys were selected for the same examinations after pathological autopsy. The tissue samples from Syrian hamsters and rhesus monkeys were collected immediately after anesthetic execution. They were then fixed in 10% neutral buffered formalin for two days, followed by dehydration in a gradient of rising ethanol. Afterward, the samples were embedded in paraffin wax. The paraffin-embedded tissues were sectioned and mounted on poly-l-lysine-coated slides. Histological analysis was performed on de-paraffinized 5 ​μm thick tissue sections stained with hematoxylin and eosin (H&E) to assess cellular injury and inflammation. To further inhibit endogenous peroxidase activity, the tissue sections were dewaxed, dehydrated, and treated with 0.3% H₂O₂ in PBS. Next, a polyclonal rabbit Anti-Coxsackie B Blend Antibody (1:100 dilution; Millipore; USA) was added to the samples, and they were incubated for 1 ​h at 25 °C. Afterward, a peroxidase-conjugated anti-rabbit antibody (1:200 dilution; Cell Signaling Technology, Beverly) was added and incubated for 30 ​min at 25 °C. The viral antigens in the tissue sections were visualized by incubating them with peroxidase stain diaminobezidin 3 (DAB) (Dako, Glostrup, Denmark), following the manufacturer's protocol. The slides were counterstained with hematoxylin, and then sealed. After dehydration, transparency, and neutral resin sealing, the results were observed using light microscopy. Finally, a slide scanner (3DHISTECH, Hungary) was used for scanning observation.

2.6. Neutralizing antibody analysis

Rhesus monkey serum samples, collected on 0, 1, 7, 15, and 30 ​d.p.i, were assessed for neutralizing antibodies against CVB3. Initially, serum samples were heat-inactivated at 56 ​°C for 30 ​min. Next, the inactivated serum was diluted by a factor of 2 in the medium, reaching a final volume of 100 ​μL per well. Subsequently, 100 CCID₅₀ of CVB3 was added to each well containing the serum-containing medium. The antiserum-virus suspension was then pre-incubated for 2 ​h at 37 ​°C. Subsequently, 100 ​μL Vero cells (at 1 ​× ​10⁵ ​cells/mL) were introduced into each well of a 96-well plate and incubated at 37 ​°C for 7 days. The CCID₅₀ values were determined by observing the cytopathic effect using the Reed-Muench method.

2.7. Expression analysis

Rhesus monkey blood samples were collected, and RNA isolated and amplified using TB Green. This process determined the expression levels of various cytokines, including interleukin-2 (IL-2), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ). The employed primers were as follows: IL-2 (forward): 5′-AGCTGCACGGTTCAAAACAGA-3′, reverse: TTCCGGCTCGGTTGGA-3′; IL-6 (forward): 5′-ACAAGTCGGAGGCTTAATTACACAT-3′, reverse: 5′-TTGCCATTGCACAACTCTTTTC-3′; TNF-α (forward): 5′-CCAGAGCCACATGCTCCTAGA-3′, reverse: 5′-GGTCCTTTGTTTGAAAGAAAGTCTTC-3′; IFN-γ (forward): 5′-CATCTTCTCAAAATTCGAGTGACAA, reverse: 5′-CCAGCTGCTCCTCCACTTG-3′.

2.8. Statistical analysis

Data analysis was conducted using GraphPad Prism 9, with mean ​± ​standard deviation (x ​± ​s) as the basis for exploring related data. To compare the experimental group with the control group, the t-test was utilized, revealing a statistically significant difference when the P value was less than 0.05.

3. Results

3.1. CVB3 infects Syrian hamsters with HFMD symptoms

To investigate the susceptibility of CVB3 to different infection modes, 3–4-week-old Syrian hamsters (6 per group) were infected with a 10^5.75 CCID₅₀ CVB3 virus suspension via intraperitoneal (IG) or nasal (NG) route, while the control group (CG) remained untreated (Fig. 1A, Table 1). The clinical signs, temperature, body weight, and blood cell changes in Syrian hamsters from 0 to 14 days post-infection (d.p.i) were recorded. Herpes lesions appeared on the lips and cheeks of infected hamsters in the IG and NG groups at 3 ​d.p.i, turning red after 6 ​d.p.i in the IG group and 7 ​d.p.i in the NG group. However, the CG group did not exhibit any such abnormalities (Fig. 1B).

Interestingly, histological examination using hematoxylin-eosin staining showed increased inflammatory cell infiltration in herpes tissue, and IHC detection using anti-CVB antibodies revealed substantial expression of CVB3 antigen in these tissues (Fig. 1C). Compared to CVB3-infected rhesus monkeys, which did not exhibit symptoms such as pharyngitis and cutaneous herpes (except for HFMD), infected hamsters on day 7 d.p.i did not display any apparent pathological damage in the oral mucosa, lingual tonsils, pharyngeal tonsils, and palatal tonsils, as observed through H&E pathologic staining (Supplementary Fig. S1). These findings imply that CVB3 induced slight pharyngitis-related symptoms or tissue damage in hamsters during the early stage after infection. CVB3-infected Syrian hamsters exhibited HFMD symptoms via both intraperitoneal injection and nasal drip. The nasal drip group had symptoms that lasted longer. Statistical analysis of clinical signs revealed that animals in both groups showed abnormalities such as reduced appetite and restricted activity from 2 ​d.p.i. The IG group exhibited neurological signs until 9 ​d.p.i, while the NG group did so until 11 ​d.p.i (Fig. 1D, Supplementary Videos S1–3, and Supplementary Table S1). This indicated that the neurological signs of the animals in NG group were prolonged. As shown in Fig. 1E and F, all Syrian hamsters in the experimental group tended to have higher body temperatures and lower body weights than the CG group. In addition, blood physiological indices were assessed on 14 ​d.p.i, revealing a significant decrease in leukocytes and a decreasing trend in monocytes in the IG group (Fig. 1G). While no significant differences were observed in the Syrian hamsters of the NG group, fluctuations were noted.

The following are the supplementary data related to this article:

Supplementary Video S1

2

Supplementary Video S2

3

Supplementary Video S3

4

3.2. The viral excretion cycle of CVB3-infected Syrian hamsters

The qRT-PCR method was employed to detect the viral load in the pharyngeal swabs of CVB3-infected Syrian hamsters. The virus were detected at 1 ​d.p.i in both the IG and NG groups, with a declining trend of viral load from 2 to 6 ​d.p.i. The copy number ranged from 10^1.78 to 10^5.72 copies/mL, which was consistent with the previously reported clinical characteristics of HFMD. Interestingly, the virus was detected in pharyngeal swab from the NG group Syrian hamsters at 14 ​d.p.i, suggesting that CVB3 might still be replicating after two weeks of infection (Fig. 2A). Fecal swabs from Syrian hamsters infected with CVB3 were evaluated for viral load over 0–14 days. The results indicated that virus can be detected at 1 ​d.p.i in both the IG and CG groups, persisting until 9 ​d.p.i in the IG group and 14 ​d.p.i in the NG group. The copy numbers ranged from 10^1.36–10^4.25 copies/mL and 10^1.34–10^4.22 copies/mL, respectively (Fig. 2B). The viral load assay in nasal swabs showed that the nucleic acid were detectable in both the IG and NG groups as early as 1 ​d.p.i. Notably, the viral copy number in NG group reached up to 10^6.10 copies/mL (Fig. 2C). At 14 ​d.p.i, CVB3 viral nucleic acid can be detected in the pharyngeal, feces, and nasal swab of the NG group, verifying the infection and shedding of CVB3 in the Syrian hamster model. However, only the nasal swab from the IG group tested positive for viral nucleic acid. These findings suggest a longer excretion cycle for CVB3-infected Syrian hamsters through nasal drops, which align with the clinical observation that patients continue to detoxify two weeks after infection.

Fig. 2.

Virus dynamic distribution in CVB3-infected Syrian hamsters. Syrian hamsters (6 per each goup) were infected with CVB3 (10^5.75 CCID₅₀) via intraperitoneal (IG) or nasal (NG) route respectively. Syrian hamsters in the control group (CG) remained untreated. Samples were collected daily from 0 to 14 ​d.p.i. The qRT-PCR method was employed to detect the viral load in swabs. A standard reference curve was generated by measuring the sequence dilution of the viral RNA produced by in vitro transcription of CVB3 VP1 DNA. The viral RNA relative copy number for each sample was calculated using the following mathematical formula: [(μg RNA/μL)/(molecular weight)] ​× ​Avogadro number ​= ​viral copy number/μL. Viral load below 10 copies was considered negative. The viral loads were normalized to 1 ​mL to evaluate the dynamic of viral replication. A Detection of CVB3 viral RNA in pharyngeal swabs. B Detection of CVB3 viral RNA in fecal swabs. C Detection of CVB3 viral RNA in nasal swabs.

3.3. CVB3 infected Syrian hamsters exhibited severe pathological damage

Viraemia, the presence of virus in the blood, is a hallmark of viral infection. In this study, the blood viral load of Syrian hamsters at 14 ​d.p.i was analyzed by qRT-PCR. The results showed that both the IG and NG groups had viraemia, with copy numbers ranging from 10^0.96–10^6.67 copies/100 ​μL (Fig. 3A). Moreover, the myocardial enzymes in the blood, such as LDH, CK, and CKMB, displayed a significant increase. Additionally, liver function indices, AST and ALT, showed a trend of increase in the NG group. These findings suggested that the heart and liver of Syrian hamsters infected with CVB3 might have suffered pathological damage (Fig. 3B and C).

Fig. 3.

Blood pathogenic, biochemical, and pathological analyses of CVB3-infected Syrian hamsters. Syrian hamsters (6 per each goup) were infected with CVB3 (10^5.75 ​CCID₅₀) via intraperitoneal (IG) or nasal (NG) route respectively. Syrian hamsters in the control group (CG) remained untreated. A At 14 ​d.p.i, the CVB3 viral RNA in the blood of Syrian hamsters was detected (n ​= ​3). B The levels of myocardial enzymes in the serum of hamsters at 10 ​d.p.i was detected. Data are presented as mean ​± ​standard deviation. Statistical analyses were performed using t-test with GraphPad Prism software. ∗, P ​< ​0.05; ∗∗, P ​< ​0.001. C The liver function indicators levels in the serum of Syrian hamsters at 10 ​d.p.i were detected. D The CVB3 viral nucleic acid in Syrian hamster brain, heart, liver, lung, spleen, and pancreatic tissues at 7 ​d.p.i. were detected using qRT-PCR. E, F The brain, heart, liver, lung, spleen, and pancreatic tissues of Syrian hamster were stained by hematoxylin and eosin to assess histopathological changes. Pictures were acquired with a 200 ​× ​microscope. G The immunohistochemical assay of Syrian hamster visceral tissue. The brown regions in IHC images indicate positive CVB3 virus antigen expression. Scale bar, 50 ​μm.

At 7 ​d.p.i, the viral loads in the visceral tissues of three randomly selected Syrian hamsters from each group were evaluated to explore the pathogenic effects of CVB3. The results revealed that CVB3 viral nucleic acids were present in the brain, heart, liver, lung, spleen, and pancreas tissues. Furthermore, the viral copy numbers in all these tissues were notably higher than 10^3.00 copies/g (Fig. 3D). Additionally, viral loads were detected in 18 tissue organs, including various brain regions and intestines (Supplemental Fig. S2). Pathological observation of the various visceral tissues, as shown in Fig. 3E, revealed varying degrees of pathological damage (Table 2, Fig. 3E and 3F). Except for the spleen tissue, the pathological damage in the NG group's visceral tissues was more severe compared to the IG group. Furthermore, the CVB3 antigen was broadly distributed, corresponding to the lesions in each tissue (Fig. 3G). Those findings suggested that CVB3 exhibits strong tropism for all tissues of Syrian hamster, resulting in severe manifestations after infection.

3.4. Clinical signs of CVB3 infected rhesus monkeys

Syrian hamsters suffered severe internal pathological damage and intensified symptoms of nasal drip infection with the dosage of 10^5.75 CCID₅₀. Clinical data showed that most CVB3-infected patients had mild symptoms. To eliminate the potential impacts of viral dose, the infected dose per body weight was used to standardize virus dosage. Rhesus monkeys were inoculated with a 5-fold lower dose compared to Syrian hamsters. Three infant monkeys (A1, A2, A3), aged 3–4 months, received a virus dosage of 2 ​× ​10^5.75 CCID₅₀ through nasal drops. The control group (B1, B2, B3) did not receive any treatment (Table 3, Fig. 4A). Rhesus monkeys were clinically observed from 0 to 14 ​d.p.i. At 2–5 ​d.p.i, all three rhesus monkeys in infected group developed the characteristic HFMD lesions on the lips, hands, and feet (Fig. 4B). Additionally, the histopathological examining revealed mild edema, loosely arranged connective tissues, and punctate lymphocytic infiltration in herpes. To further confirm that the pathological changes in the herpes were indeed caused by CVB3 infection, IHC staining for CVB3 antigen was performed. Positive antigen signal was observed in the herpes tissue (Fig. 4C). Compared to the healthy rhesus monkeys in control group, monkeys in infected group displayed decreased activity (Fig. 4D–Supplementary Table S2), lowered body temperature (Fig. 4E), and reduced body weight gain index (Fig. 4F) from 2 to 12 ​d.p.i. Furthermore, a reduction in leukocytes and lymphocytes was observed in infected rhesus monkeys at 1–5 ​d.p.i. And a transient increase in monocytes and neutrophils was observed at 7 and 9 ​d.p.i, respectively (Fig. 4G).

Table 3.

Experimental groups for CVB3 infection in rhesus monkey.

Group	Hamster ID	Gender	Agea	Virus doseb	Weight (kg)c	Infection route
A	A1–A3	♀	3	2 ​× ​105.75	0.76 ​± ​0.05	Nasal drip
B	B1–B3	♀	3	0	1.71 ​± ​0.11	None

^aMonths after the birth.

^b50% cell culture infectious doses in 0.2 ​mL inoculation volume.

^cMean weight ± standard deviation.

Fig. 4.

Clinical observation of CVB3-infected rhesus monkeys after 0–14 ​d.p.i. A Experimental design of the procedure. Six infant rhesus monkeys, aged 3–4 months, were organized into various groups. Group A was infected via nasal drip, labeled as A1, A2, A3; while group B served as the control group and remained untreated, labeled as B1, B2, B3. The infective dose for both groups was 2 ​× ​10^6.75 CCID₅₀/monkey. Samples were collected at specified time points post-infection. B The rhesus monkeys in group A exhibited typical HFMD signs in their mouth, hand, and foot regions. Ulcerative blisters accompanied by erythematous lesions were observed between day 2 and 5 post-infection (d.p.i). C Histopathological changes in group A on the 3 ​d.p.i was assessed by hematoxylin and eosin (HE) staining and immunohistochemical (IHC) assay. Red arrows indicate mild edema at the dermal papillae. Yellow arrows indicate lymphocyte punctate infiltration. Herpes tissue IHC results showed CVB3 antigenic expressions. Brown color indicates DAB chromogenic positive antigenic signal and blue-purple color is the nucleus (red box is positive antigenic signal). Scale bar: 100 ​μm. D The clinical manifestations (daily performance and dietary status) of CVB3-infected rhesus monkeys were observed and measured in detail. The scoring criteria for daily performance were as follows: normal/alert/active-0; slow/quiet/alert/active-5; quiet/arching back/alert/limited activity-10; depressed/inactive/unaware-15. Similarly, the dietary status was scored as follows: normal diet-0; reduced appetite −2; anorexia-5. The values represent the mean and standard error, with a sample size of n ​= ​3. E, F The body temperature and weight changes in rhesus monkeys were monitored. G The flow cytometry (FCM) assays were employed to examine the blood cells of experimental rhesus monkeys. Leukocytes, WBC; lymphocytes, LYMPH; monocytes, MONO; neutrophils, NEUT.

3.5. qRT-PCR reaction to examine the CVB3 transmission kinetics in rhesus monkeys from 0 to 14 ​d.p.i

Viral nucleic acids were detected in pharyngeal swabs of all CVB3-infected rhesus monkeys, with the viral copy numbers of 10^1.92, 10^1.98, and 10^2.33 copies/100 μL at day 2 ​post-infection. At 14 ​d.p.i, pharyngeal swabs remained positive for CVB3 viral nucleic acids (Fig. 5A), indicating that CVB3 continued replicating in rhesus monkeys for more than two weeks. Further analyses showed that viral nucleic acids were detected in nasal swabs from infected group from 1 to 14 ​d.p.i, with copy numbers fluctuating between 10^1.48 and 10^6.92 copies/100 μL (Fig. 5B). The detection viral load in fecal sample was correlated with that in fecal viral excretion, with copy numbers ranging from 10^2.60–10^5.88 copies/g, suggesting that viral excretion began at 2 ​d.p.i and continued until 8–14 ​d.p.i. (Fig. 5C). The blood viral nucleic acid at different time points showed that viraemia was present from 3 to 11 ​d.p.i, with copy numbers peaking at 7 ​d.p.i. and ranging from 10^2.11 to 10^4.02 copies/100 μL (Fig. 5D). This was consistent with the clinical scores of rhesus monkeys.

Fig. 5.

Virus dynamic distribution in CVB3-infected rhesus monkeys on day 0–14 post-infection (d.p.i). Total RNA was extracted by Trizol method from pharyngeal (A), nasal swabs (B), fecal (C) and blood samples (D) of rhesus monkeys. The CVB3 viral load in rhesus monkeys was assessed at various time points post-infection (0–14 ​d.p.i.) using fluorescence quantitative PCR. To evaluate the dynamic replication of the virus, the viral loads were normalized to 100 ​μL or 1 ​g. Group A was infected via nasal drip, while group B served as the control group and remained untreated.

3.6. Persistent observation of CVB3-infected rhesus monkeys exhibited mild infection

According to the enterovirus development timeline, the acute infection period is established to begin from 0 to 14 ​d.p.i. In this study, we conducted persistent observation for over 15–30 days. The results revealed that CVB3-infected rhesus monkeys experienced a remission of clinical signs and normalization of behavioral indicators, as evidenced by restored normal clinical scores (Fig. 6A). Subsequently, viral nucleic acids in pharyngeal swabs, nasal swabs, feces, and blood samples collected from the monkeys were detected. Lower viral copy numbers were detected in infected rhesus monkeys at 15 and 21 ​d.p.i; however, all showed negative at 30 ​d.p.i (Fig. 6B). These findings suggest a continuous viral excretion process lasting approximately one month during the observation period, despite the absence of clinical signs.

Fig. 6.

Persistent observation of CVB3-infected rhesus monkeys for 15–30 days after infection. Group A was infected via nasal drip, while group B served as the control group and remained untreated. A The clinical symptom scores of rhesus monkeys were recorded for 15–30 days after CVB3 infection. B Viral nucleic acids in pharyngeal, nasal, feces swabs and blood samples from CVB3-infected rhesus monkeys were detected. C The liver function indicators levels in the serum of CVB3-infected rhesus monkeys were detected. D Dynamic myocardial enzyme indexes in CVB3-infected rhesus monkeys were measured. E The mRNA levels of inflammatory factors in serum in CVB3-infected rhesus monkeys were thoroughly investigated. F The serum neutralizing antibody levels in CVB3-infected rhesus monkeys were assessed. G Viral load analysis in 43 tissues of A3 monkeys was performed at 30 ​d.p.i. Virus was detected in the heart, optic nerve, peroneal nerve, and pharyngeal tonsil. H Histological and immunohistochemical (IHC) assays on A3 monkey tissues, Histological pictures were acquired with a 200 ​× ​microscope. Red arrow indicates mild edema. Yellow arrow indicates lymphocyte punctate infiltration. The brown regions in IHC images represent the positive CVB3 viral antigen expressions. Scale bar, 50 ​μm.

Liver function and cardiac enzyme indices in the monkeys revealed significant differences in the liver function indices of infected monkeys at 15, 21, and 30 ​d.p.i. Although cardiac enzyme indices tended to increase, no significant differences were noted (Fig. 6C and D). Analysis of various inflammatory factors in the monkeys’ blood showed that each factor remained elevated at 15 ​d.p.i, but normalized after 21 and 30 ​d.p.i, with no significant fluctuations (Fig. 6E). Additionally, serum neutralizing antibodies were detected at 15 and 30 ​d.p.i., with 1:16 and 1:32 neutralizing antibodies titers detectable in the serum of infected monkeys at 30 ​d.p.i (Fig. 6F).

Considering animal experimental ethics, the A3 monkey with a high viral load was euthanized at 30 ​d.p.i. Necropsy tissues from the A3 monkey were subjected to etiological and pathological analysis. The nucleic acid detection revealed that the heart, optic nerve, peroneal nerve, and pharyngeal tonsil tissues were positive, with the highest results found in pharyngeal tonsils at 10^3.43 copies/g (Fig. 6G). Interestingly, histopathological examination revealed minor aqueous degeneration of myocardial fibers in the heart, hepatocytes in the liver, alveolar wall thickening with inflammatory cell infiltration in the lung, and localized necrotic cell debris in the pharyngeal tonsil. Furthermore, to confirm that the pathological changes in these tissues were indeed caused by CVB3 infection, antigen detection by IHC using CVB-specific antibodies revealed the expression of CVB3 antigen in these tissues (Fig. 6H). The above results confirmed the persistent presence of viral nucleic acids and proteins in the tissues at 30 ​d.p.i, reflecting a more comprehensive chronic development process of pathogen-host interactions. Interestingly, viral nucleic acids and proteins were detectable in the heart and other tissues of CVB3-infected monkeys, yet no significant abnormalities were observed in cardiac enzymes and other indicators. This suggests that the mild clinical manifestations of CVB3 infection were indirectly reflected in the rhesus monkeys.

4. Discussion

As the epidemiological spectrum of enteroviruses shifts, the detection of CVB3, a key pathogen causing HFMD, has been increasingly, resulting in variations in clinical manifestations. Consequently, it becomes imperative to investigate the pathogen's immunological and pathological mechanisms using appropriate preclinical models. While various CVB3 animal models have been successfully established in mice, previous studies have mainly focused on myocarditis, neglecting a comprehensive analysis of the host's viral infection mechanism. Given the limited information in previous studies, we suspected that infection routes might affect the success of CVB3 model establishment. For instance, nasal drop infection has been frequently used and produced positive results in preclinical enterovirus studies (Zheng et al., 2017; Liu et al., 2011; Duan et al., 2022; Sun et al., 2019). Furthermore, recent research indicates that enterovirus infections via the respiratory tract and nasal drops can accurately replicate natural infections. The reproducibility of respiratory infections is enhanced, enabling more precise predictions of infection and pathology distribution.

The ideal CVB3 animal model should encompass the essential features of human infection, including clinical manifestations, viral excretion cycle, and pathological damage. Previous studies have reported that the duration of HFMD is mostly within 7 days (Alsop et al., 1960). However, the existing CVB3 animal models did not exhibit clinical manifestations of HFMD. Our study observed the presence of HFMD in infected Syrian hamsters for 3–7 days and rhesus monkeys for 2–5 days, with symptom duration consistent with clinical patients (Fu et al., 2019; Han et al., 2019; Li et al., 2019). In addition, it was noted that 94.6% of the 65 patients with enterovirus infection had positive viral nucleic acid in feces during the second week, indicating a potential positive correlation between the disease severity and the duration of viral excretion (Teng et al., 2015). In our study, viral loads up to 10^2.3 copies/100 μL were detected in pharynx, feces, and nasal swabs of Syrian hamsters in the NG group at 14 ​d.p.i. Compared to the IG group animals, animals in NG group exhibited a longer viral excretion cycle. Thus, it was suggested that nasal drip infection might more closely resemble clinical infection routes than intraperitoneal infection, leading to a stronger host response (Meenakshi et al., 2022). This discovery prompted us to consider how variations in infection patterns may impact CVB3 infection in hosts, highlighting the need for further comprehensive studies to improve our understanding of CVB3-host interactions and infection immune mechanisms. In our research, the viral loads of pharyngeal, fecal, and nasal swabs in rhesus monkeys were 10^1.35, 10^3.19, and 10^3.20 copies/100 μL or copies/g, respectively, after two weeks of CVB3 nasal drip infection. This discovery confirmed the existence of excretion cycle in rhesus monkeys following CVB3 infection, aligning with clinical observations and indirectly reflecting host-pathogen interactions.

Enteroviruses cause four different HFMD outcomes: asymptomatic (12.7%), mild (86.2%), severe and critical (1.1%), and death (0.03%) (Xing et al., 2014; Esposito and Principi, 2018). It is noteworthy that Syrian hamsters infected with CVB3 exhibited significantly severe pathological damage. Furthermore, among the visceral pathological injuries, except for the spleen, the pathological injuries induced after the nasal drop infection were significantly more severe compared to those observed after intraperitoneal infection. The primary reason for the spleen pathological injury after intraperitoneal injection could be attributed to the stronger antiviral response of the stimulated splenic immune cells. Previous studies have indicated that rhesus and crab-eating monkeys can exhibit susceptibility to enteroviruses after the nasal drop inoculation (Shen et al., 2017). Compared to Syrian hamsters, pathological damage caused by CVB3 infection in rhesus monkeys was significantly milder, which was more consistent with the mild symptoms typically observed in clinical patients. The main reason for this difference possibly relates to the fact that NHP have a variety of different simian enteroviruses and are up to 72% amino acid identical to human enteroviruses (Nix et al., 2008).

It is noteworthy that in both the domestic and international research progress, the mechanism of viral myocarditis caused by CVB3 has primarily been studied in animal models. However, the majority of these studies were conducted in mice by intraperitoneal infection, and the observed pathological damage primarily includes cardiomyocyte deformation, hemorrhage, and inflammatory cell infiltration (Fairweather and Rose, 2007; Lasrado and Reddy, 2020), exhibiting a degree of discrepancy from the clinical patient condition. In contrast, our current study observed necrosis, edema, and vacuolization of cardiomyocytes in the hearts of the rhesus monkey model. This model exhibited hydropic degeneration of myocardial fibers, which was more akin to the clinical condition, reflecting a more accurate representation of the interaction between the virus and the host.

This study had some limitations. We aimed to reduce the use of animals in our laboratory experiments while still maintaining scientific feasibility by following the 3R principle of laboratory animal ethics—Replacement, Reduction, and Refinement. However, the lack of statistical significance for some indicators in this study can be attributed to the small sample size: only six Syrian hamsters were used as subjects, with three being dissected. This small sample size may have resulted in insufficient statistics. Furthermore, our study on Syrian hamsters and rhesus monkeys did not provide an in-depth analysis of sustained pathological damage to different tissues. The relationship between the observed pathological damage in CVB3-infected animals and factors such as increased viral replication, prolonged viral replication, or intensified damage produced by infection-induced cytokines remains unknown. This represents an important area for future research. In-depth investigation of the mechanisms contributing to various pathogenic and pathological changes could be facilitated. The establishment of CVB3 Syrian hamster and rhesus monkey animal models contributes to a better understanding of complex pathogenic mechanisms and lays the groundwork for relevant drug and vaccine development.

5. Conclusions

In this study, we initially established a severe CVB3 infection model by infecting Syrian hamsters using various methods. Our findings revealed that nasal drip infection was more effective than intraperitoneal infection. While Syrian hamsters are capable of exhibiting multiple organ damage, they hold limited significance in vaccine evaluation. Subsequently, we developed a mild CVB3 NHP animal model using rhesus monkeys, systematically investigated the replication kinetics and tissue adaptability of CVB3. Furthermore, the noticeably less severe damage in monkeys deserves further investigation. These findings provide insights for subsequent small animal and NHP models in pathogenesis studies, vaccine, and drug research.

Data availability

All the data generated during the current study are included in the manuscript.

Ethics statement

The experimental animal procedure, conducted in accordance with the guidelines set by the Office of Laboratory Animal Management of Yunnan Province, China, was designed to ensure animal welfare. The entire process was approved by the Institutional Animal Care and Use Committee of the Institute of Medical Biology, Chinese Academy of Medical Sciences (Ethics numbers: DWSP202107021 and DWSP202107019).

Author contributions

Suqin Duan: conceptualization; data curation; formal analysis; writing-original draft; writing-review& editing. Wei Zhang: investigation; methodology; project administration. Yongjie Li: software; supervision. Yanyan Li: methodology; visualization. Yuan Zhao: software; supervision; validation. Weihua Jin: investigation. Quan Liu: project administration. Mingxue Li: visualization. Wenting Sun: software. Lixiong Chen: validation. Hongjie Xu: validation. Jie Tang: software. Jinghan Hou: supervision. Zijun Deng: visualization. Fengmei Yang: project administration; resources; software; supervision. Shaohui Ma: writing-original draft; resources; software; supervision; validation. Zhanlong He: resources; funding acquisition; investigation; methodology; project administration; writing-original draft; writing-review & editing.

Conflict of interest

The authors of this study declared that they do not have any conflict of interest.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Fig. S1.

Fig. S2.