Twelve rhesus macaques (Macaca mulatta) that were previously experimentally infected with HEV were examined in this study. Primary infection of RH623 was previously reported [18]. The Comparative Medicine Branch at the Centers for Disease Control and Prevention (CDC) provided care and husbandry for the animals in accordance federal laws, regulations, and the Guide for the Care and Use of Animals. All animal protocols and procedures involved in this study were reviewed and approved by the CDC Institutional Animal Care and Use Committee (IACUC protocol number 2643CHOMONC).

After clearance of the primary infection (Supplementary Figures 1 and 2), homologous HEV Sar-55 strain (gt1) was reinoculated intravenously to 2 rhesus macaques (RH624 and RH625) with 1000 monkey infectious doses (MID) (6.4 log₁₀ World Health Organization [WHO] U/mL) and 100 MID (4.9 log₁₀ WHO U/mL) for RH637, RH621, RH617, RH636, RH619, RH622, RH631, RH620, RH626, RH627, and monitored for 115 days. RH623, RH633, and RH634 were inoculated with either 100 or 1000 MID and used as primary infection controls (Table 1). A liver biopsy tissue was obtained from each animal (19 days after inoculation [DAI] from RH622, RH624, RH617, RH619, and RH631; 23 DAI from RH625, RH620, RH627, and RH626; and 28 DAI from RH621, RH636, and RH637). Multiple preinoculation and weekly postinoculation blood samples and daily stool samples were collected from all of the animals. The alanine aminotransferase (ALT) activity in sera was measured using a VetScan VS2 (ABAXIS, Union City, CA) according to the manufacturer’s instructions. ALT activity was measured as described before [18]. Reinfection was defined as the detection of a new HEV infection evidenced by detection of viral RNA in stool and serum by real-time polymerase chain reaction (PCR) and elevated serum alanine aminotransferase (ALT) activity. Protection was analyzed based on IgG anti-HEV antibody concentration and IgG anti-HEV avidity index in serum samples.

Total RNA extraction, cDNA synthesis, and real-time PCR were performed as described previously [18]. The HEV RNA level in liver, stool, and serum specimens was determined as WHO international units (WHO U/mL) by comparison with a standard curve of serial log₁₀ dilutions of the WHO standard (6329/10, Paul-Ehrlich-Institut, Langen, Germany) [19].

IgG anti-HEV antibody levels were measured using Wantai HEV IgG enzyme immunoassay (EIA) kits according to the manufacturer’s instructions using the recommended cutoff values (WE-7296, Wantai Biologic Pharmacy Enterprise, Beijing, China). Quantitation of IgG anti-HEV antibody was adapted as described previously [10]. Briefly, each sample was serially diluted 10-fold in the range from 1/10 to 1/8000 with dilution buffer from the kit to determine the levels of IgG anti-HEV antibody. Signal-to-cutoff ratios (S/CO) of IgG anti-HEV antibodies were calculated as sample OD₄₅₀/cutoff OD₄₅₀ according to the manufacturer’s instructions. All samples with a S/CO ratio of ≥1 were considered positive. The WHO IgG anti-HEV reference standard (95/584, National Institute for Biological Standards and Control, South Mimms, United Kingdom) was used to determine the concentrations of anti-HEV antibodies. The detection limit of the Wantai anti-HEV IgG EIA kit was 0.25 WHO U/mL and a linear range was obtained from 0.25 to 5 WHO U/mL. All of the samples were diluted to achieve an OD₄₅₀ value within the linear range of the WHO anti-HEV IgG reference.

IgG anti-HEV antibody avidity was determined as described previously with slight modifications [10]. Briefly, duplicate serum samples were incubated for 30 minutes at 37°C in Wantai anti-HEV IgG EIA enzyme-linked immunosorbent assay (ELISA) plates, 1 sample from the pair followed the standard procedure, and the other sample was exposed to a 7 M urea solution in the washing buffer for 10 minutes at room temperature. The serum samples were diluted to have an OD₄₅₀ in the range between 0.8 and 2.1. IgG anti-HEV antibody avidity was calculated after accounting for dilution effects as: percentageofmeasuresantibodyconcentrationintheurea−treatedwellantibodyconcentrationintheuntreatedwell×100

Whole blood collected in BD Vacutainer Mononuclear Cell Preparation Tubes (CPT; BD Bioscience, San Jose, CA) containing sodium heparin as an anticoagulant was processed according to the manufacturer’s instructions. The CPT tubes were inverted several times after blood drawing and centrifuged for 20 minutes at 1800g at room temperature. The cell interface layer containing peripheral blood mononuclear cells (PBMC) was harvested and isolated PBMCs were resuspended in EmbryoMax 2X freezing medium (Millipore, Billerica, MA). Human interferon-gamma (IFN-γ) enzyme-linked immunospot (ELISPOT) assays were performed using the 96-well ImmunoSpot kit according to manufacturer’s instructions (CTL, Shaker Heights, OH). Three time points (baseline [0 DAI], at the peak of antibody responses [40 DAI for RH617, 619, 622, 624, and 631; 52 DAI for RH620, 625, 626, and 627], and at the end of observation [115 DAI]) were used to determine levels of HEV-specific IFN-γ–producing T cells. Briefly, the PBMCs were washed twice with 5 mL CTL-Test Medium, cells were counted with the Countess Automated Cell Counter (Invitrogen, Carlsbad, CA), and cell numbers adjusted to 4 × 10⁶ cells per mL. HEV ORF2 recombinant protein, HEV-236 (20 μg, 452–617 amino acids) (Prospec-Tany TechnoGene, East Brunswick, NJ) was used to stimulate 2 × 10⁵ cells for 24 hours and a final concentration of 1 μg HEV-236 protein was added to each well and incubated at 37°C in 5% CO₂. Anti-human IFN-γ detection solution was added according to manufacturer’s instructions. For positive control, cells were stimulated with 50 ng/mL phorbol 12-myristate 13-acetate (Sigma, St Louise, MO) and 500 ng/mL ionomycin (Sigma), and 1% dimethyl sulfoxide (Sigma) was used as negative control. All assays were performed in duplicate. Cytokine-producing spots were scanned and counted with a CTL Analyzer (Shaker Heights, OH).

Supernatant from the HEV-236 stimulated PBMCs was used to analyze cytokines/chemokines generated by HEV-specific T-cell responses using the Human Multi-Analyte ELISA Array (QIAGEN, Germantown, MD). Twenty-four hours after stimulation, supernatants were collected from the ELISPOT plates and stored at −80°C until ELISA testing. Thawed supernatants were added to the Human Multi-Analyte ELISA Array containing interleukin (IL)-1B, IL-4, IL-6, IL-10, IL-12, IL-17A, IFN-γ, tumor necrosis factor-alpha (TNF-α), transforming growth factor-beta 1 (TGF-β1), monocyte chemoattractant protein-1 (MCP-1), macrophage inflammatory protein-1 alpha (MIP-1α), and MIP-2α, and followed according to the manufacturer’s protocol. Protein expression levels were measured as absorbance at 450 nm using a Biotek ELISA plate reader (Winooski, VT) and relative levels of cytokines and chemokines were compared to baseline samples.

RNA extraction and synthesis of cDNA were carried out as described before [18]. Rhesus macaque specific innate and adaptive immune response (PAQQ-052Z), Th1 and Th2 response (PAQQ-034Z), and type I interferon response (PAQQ-016Z) RT2 profiler PCR arrays (QIAGEN, Carlsbad, CA) were performed according to the manufacturer’s protocol. Gene expression was normalized using the expression of 5 genes, ACTB, B2M, GAPDH, HGPRT, and RPL13A in each PCR array as housekeeping control genes. Liver tissues obtained from 2 anti-HEV antibody-negative animals were used as baseline control samples for determination of differential gene expression. Changes in gene expression of >2-fold compared to baseline were considered significant.

Statistical analysis were performed with GraphPad version 7.04 (La Jolla, CA). Statistical significance was tested using a 2-tailed t test. P values < .05 were considered significant.

As primary infection controls, 3 anti-HEV antibody-negative rhesus macaques were infected with HEV gt1, Sar-55. One of the primary infection controls (RH623) was reported in a previous study [18]. These animals exhibited the typical course of acute HEV infection (Figure 1). HEV RNA was detected in stools between 3 and 7 DAI, in serum from 10 to 42 DAI, and cleared between 20 and 67 DAI in all of the animals. Serum ALT activity was elevated from 31 to 45 DAI in RH623 and RH633. IgM anti-HEV antibody became detectable from 14 to 35 DAI and disappeared from 52 to 104 DAI. IgG anti-HEV antibody response was observed from 14 to 38 DAI, peaked from 28 to 35 DAI, and declined from 63 to 80 DAI.

Duration between the primary inoculation and the reinoculation varied from 1.2 to 5.2 years (Table 1). Four macaques were reinfected as evinced by HEV RNA in stools, and 8 animals did not develop infection or shed virus in feces after reinoculation with HEV gt1, and were thus protected (Figure 2). In the reinfected animals, HEV RNA levels in stools was lower (average 3.9log₁₀ WHO U/mL) than the primary HEV infection (average 8.4log₁₀ WHO U/mL). Duration of viral shedding in stools was also shorter (average 17 days) than the primary infection (average 41 days). None of the reinfected macaques showed IgM anti-HEV antibody responses or elevation of ALT activity (Figure 2 and Table 1).

Homologous HEV reinfection after the primary infection was associated with significantly higher levels of HEV-specific T-cell responses. Sufficient numbers of PBMCs were obtained from 2 of 4 reinfected and 7 of 8 protected animals to determine cellular immune responses. Levels of HEV-specific IFN-γ–producing T-cell responses in reinfected animals (RH617, 707/10⁶ PBMCs and RH624, 547/10⁶ PBMCs) were higher than 7 protected animals (average of 77/10⁶ PBMCs) both at elevated levels of antibody responses and at the end of observation (RH617, 292/10⁶ PBMCs and RH624, 342/10⁶ PBMCs in reinfected animals and average 49/10⁶ PBMCs in protected animals; P = .0086) (Figure 3A). High levels of IL1-β, TGF-β, IL-6, MCP-1 (CCL-2), MIP-1α (CCL-3), and MIP-2α (CXCL-2) expression in supernatants from HEV 296 protein-stimulated PBMC cultures were detected in reinfected animals (Figure 3B). Levels of IL-1β (2-way ANOVA, P = .0098), TGF-β (P = .0066), and IL-6 (P = .0033) expression were elevated at the peak of antibody response, and MCP-1 (P = .0069), MIP-1α, and MIP-2α expression peaked at the end of observation in the reinfected animals. In the protected animals, TGF-β and IL-6 expression was downregulated at the peak of antibody response and the end of observation.

In the 3 control animals, IgG anti-HEV antibody response became positive at 38 DAI in RH623, 37 DAI in RH633, and 14 DAI in RH634, and peaked at 56 DAI (578 WHO U/mL) in RH623, 52 DAI (3136 WHO U/mL) in RH633, and 28 DAI (868 WHO U/mL) in RH634 (Figure 1 and Figure 4). Baseline IgG anti-HEV antibody concentration in the 4 reinfected animals ranged from 1.5 to 13.4 WHO U/mL (Table 1). IgG anti-HEV concentration rose between 63- and 285-fold in the reinfected animals (P = .003) (Figure 2 and Figure 4). Baseline IgG anti-HEV antibody concentration ranged from 2.8 to 90.7 WHO U/mL in 8 protected animals (Table 1). IgG anti-HEV antibody levels were unchanged in the protected animals following reinoculation (P = .0172), except a slight increase was observed between 80 and 115 DAI in RH627; 21 to 113 DAI in RH621; and 50 and 115 DAI in RH622 (Figure 2).

IgG avidity in the primary infection controls ranged from 25% to 34% (mean 28%) (Figure 4 and Table 1). Baseline IgG avidity levels in the reinfected animals ranged from 17% to 27% (mean 24%) and 24% to 80% (mean 74%) in the protected animals. During the second infection, IgG avidity ranged from 30% to 42% (mean 34%) in the reinfected animals (P = .0014), and 32% to 88% (mean 62%) in the protected animals (P < .001) (Figure 4). Baseline IgG avidity levels were correlated with IgG avidity during secondary infection and the correlation was statistically significant (r = 0.844, P < .00075) (Figure 4C).

A liver biopsy specimen was obtained at either 19, 23, or 28 DAI from all of seropositive animals and host immune response-related genes associated with the outcome of HEV gt1 reinfection were analyzed. Type 1 IFN response-related genes, CAV1, IFNW1, NOS2, SOCS1, and TNFα were upregulated in protected animals. Th1 and Th2 response-related genes, CCR5, CCR2, CD40LG, FASLG, ICOS, IL1R1, IL1RL1, Il7R, CCL7, and TNFSF4 were upregulated in reinfected animals, and TNFRSF9, TYK2, VEGFA, and IRF7 genes were downregulated in reinfected animals (Figure 5).