All experiments were conducted at the enhanced animal biosafety level 2 facility of the National Institute of Veterinary Research (Hanoi, Vietnam) in accordance with the guidelines of the Vietnamese government. The animal experiments were approved by the Institutional Animal Care and Use Committee of the National Institute of Animal Science, Republic of Korea (Approval No. NIAS2022-556). The experimental design consisted of three independent trials to evaluate infection outcomes following three ASFV exposure models: (i) ASFV-contaminated dry feed, (ii) ASFV-contaminated wet feed, and (iii) organ homogenate prepared from ASFV-infected porcine organ tissues (Figure 1). A genotype II ASFV strain isolated from blood of pigs during an ASF outbreak in Thanh Hóa province (Vietnam) in 2020 (GenBank accession no. OP615344) was used throughout all three experiments. This strain has been previously characterized in a pig-to-pig transmission study and recently used to compare their pathogenicity with a recombinant genotype I/II ASFV (13, 14) and was propagated in porcine alveolar macrophages and titrated by the 50% hemadsorbing dose (HAD₅₀) assay, as previously described (8).

For each experiment, 6- to 7-week-old Landrace × Yorkshire pigs were randomly allocated to an ASFV-exposed group (n = 5 for dry-feed and organ-homogenate trials; n = 3 for the wet-feed trial) and a separate negative control group (n = 3 per trial). All experimental pigs were confirmed negative for the following endemic pathogens using quantitative polymerase chain reaction (qPCR) and enzyme-linked immunosorbent assay before oral ASFV exposure: foot-and-mouth disease virus, porcine reproductive and respiratory syndrome virus, classical swine fever virus, porcine circovirus 2, and ASFV. Animals were individually housed and acclimated to the facility for 7 days prior to ASFV infection, with commercial feed provided twice daily and water provided ad libitum. All virus exposures through contaminated feed or organ homogenate were administered once on day 0, with clinical monitoring continuing for 28 days post-infection (dpi) or until humane endpoints (15) were reached. Clinical monitoring was standardized across all three trials; however, blood and swab sampling differed according to the exposure model and the onset of clinical signs.

Two independent feeding trials were conducted to evaluate the efficiency of ASFV infection through contaminated dry and wet commercial feed. Commercial complete pig feed was used as the exposure matrix in both models. An ASFV inoculum was diluted in phosphate-buffered saline (PBS, pH 7.4) and thoroughly mixed with the feed to achieve a nominal dose of 2 × 10⁶ HAD₅₀ per 100 g of feed. The inoculum was added dropwise to the feed and manually mixed in a sterile container until homogeneous, and the contaminated feed was offered to the pigs immediately after preparation. This procedure was designed to simulate extrinsic (post-production) contamination of finished commercial feed and does not address virus introduction through feed ingredients during the manufacturing process.

In the dry-feed exposure model, each pig in the exposed group received 100 g of ASFV-contaminated dry feed in an individual feeding trough on day 0 and voluntarily consumed the feed within 15 min, corresponding to an oral dose of 2 × 10⁶ HAD₅₀ per pig. In the wet-feed exposure model, the same amount of commercial feed (100 g) was pre-moistened with sterile water to form a porridge-like consistency before mixing with the ASFV inoculum. Because pigs did not readily consume the wet feed voluntarily, each pig in the exposed group was orally administered 100 g of ASFV-contaminated wet feed using a syringe placed at the corner of the mouth on day 0, corresponding to the same dose of 2 × 10⁶ HAD₅₀ per pig. The material was delivered slowly toward the back of the oral cavity. These pigs were not sedated, and no obvious oral trauma, persistent coughing, or immediate regurgitation was observed after administration. Control pigs in both trials received the respective dry or wet feed prepared with sterile PBS instead of the virus; after the single exposure, all animals received an untreated commercial diet for the remainder of the 28-day observation period.

To model oral exposure to infected pig organ tissue, we prepared organ homogenates from an ASFV-infected donor pig. This model was intended to simulate ingestion of high-titer infected organ material that could be present in contaminated food waste. To prepare the inoculum, a 7-week-old pig was intramuscularly inoculated with 1 mL of ASFV (1 × 10⁷ HAD₅₀/mL) in the neck region. The donor pig showed typical ASF clinical signs including high fever (>41 °C), anorexia, and lethargy, and died at 5 dpi. Immediately following death, necropsy was performed under sterile conditions. The spleen and liver showed splenomegaly and hepatic congestion. Twenty-five grams each of the spleen and liver (total 50 g) were aseptically collected and placed in sterile containers on ice. The tissue samples were homogenized in 100 mL of PBS to create a 33% (w/v) organ suspension. This pooled suspension was designed to mimic ingestion of mixed infected offal from naturally infected carcasses. However, because spleen and liver were pooled before titration, organ-specific viral loads in the spleen and liver were not assessed separately. The viral titer of the collected organ homogenate was determined to be 2.5 × 10⁵ HAD₅₀/mL. Oral administration of 20 mL homogenate corresponded to a dose of 5 × 10⁶ HAD₅₀ per pig.

Each pig in the exposed group received 20 mL of the organ homogenate via oral administration using a needle-free syringe. The homogenate was administered slowly to the back of the oral cavity to ensure complete ingestion. Control pigs received 20 mL of PBS following the same administration protocol. All exposures occurred as a single event on day 0 (considered 0 dpi), and the animals also resumed normal feeding with uncontaminated commercial diet.

Clinical signs and rectal temperatures were recorded daily for all pigs until the end of the experiment (28 dpi) or until humane endpoints were reached. Clinical sign scores were calculated based on previously established criteria (8). Rectal temperatures were measured each morning, with fever defined as a rectal temperature ≥40.5 °C. Pigs showing severe clinical scores or moribund conditions were humanely euthanized according to institutional guidelines. In the organ-homogenate trial, samples (blood, nasal, oral, and rectal swab) were collected to characterize infection kinetics at 0, 2, 4, 6, 8, and 10 dpi, or until death or a humane endpoint. In the feed trials, animals were monitored clinically each day; if high fever or ASF-compatible signs had developed, serial sampling would be initiated every 2 days. Because no pig in either feed trial developed fever, blood samples were collected at 28 dpi to assess detectable viremia. Blood samples were collected from the jugular vein, and three swab samples (nasal, oral, and rectal) were obtained using sterile cotton swabs. Each swab was placed immediately in 1 mL of PBS and vortexed for 30 seconds, and the eluate was stored at −80 °C until analysis. DNA was extracted from 200 μL of whole blood and swab eluates using the DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. ASFV DNA detection was performed using the VDx ASFV qPCR kit (Median Diagnostics, Chuncheon, Republic of Korea), and samples with Ct values ≤ 35 were considered positive in accordance with a previous study (8). All dead or euthanized animals were necropsied within 2 h, and the pigs that survived to 28 dpi were also necropsied at the end of the observation period.

Statistical analyses were performed using SPSS version 29.0.2.0 (IBM Corp., Armonk, NY, USA). For experiments where viral loads were detected, the estimated clinical sign scores and Ct values from blood as well as swab samples at different time points were analyzed using repeated measures analysis of variance. Data are expressed as mean ± SD or mean ± SEM, as indicated, with p < 0.05 considered statistically significant.

After a single oral exposure to ASFV-contaminated dry or wet feed, none of the pigs in either group developed clinical signs consistent with ASF during the 28-day observation period, and rectal temperatures remained within the normal range. Throughout the study, ASFV DNA was not detected in blood at 28 dpi from any pig in the dry- or wet-feed groups, indicating that no systemic infection was detected under the present experimental conditions. However, given the small sample size of the wet-feed group, these results should be interpreted as preliminary. All pigs in both feed exposure groups survived to 28 dpi without mortality (Table 1).

All pigs in the negative control group survived until the end of the experiment, whereas a single oral administration of ASFV-infected organ homogenate resulted in 100% mortality in the exposed group (Figure 2A). The first death occurred at 6 dpi, and all five pigs died by 11 dpi, consistent with an acute disease course following a single exposure (Table 1).

Rectal temperatures in negative control pigs remained within the physiological range throughout the study, while pigs in the organ homogenate group developed fever shortly after ingestion (Figure 2B). High fever (≥40.5 °C) first appeared at 3.0 ± 0.4 dpi, with mean peak rectal temperatures of 41.3 ± 0.2 °C at 4 dpi (Table 2). Group mean temperatures in the exposed group remained elevated until 7 dpi, with significantly higher values than in the negative control group at 4 and 7 dpi (two-way analysis of variance, p < 0.05).

Clinical scores in the negative control group remained below 1 throughout the study (Figure 2C). In contrast, pigs in the organ homogenate group first showed clinical signs at 3 dpi, and the majority exceeded the predefined clinical threshold of 3 points between 4 and 7 dpi, presenting with lethargy, reduced feed intake, and respiratory signs including labored breathing. Peak scores reached 6–7 points shortly before death or humane end-points, indicating that a single ingestion of infected organ homogenate was sufficient to induce systemic clinical disease and fatal outcomes in all exposed pigs.

To characterize the virological course of ASFV infection following oral exposure to infected tissue homogenate, ASFV DNA levels in blood and nasal, oral, and rectal swabs were analyzed throughout the experimental period (Figure 3 and Table 2). ASFV DNA was first detected in blood at 2.8 ± 0.5 dpi, with mean Ct values around 33, and viremia rapidly increased to peak levels at 4 dpi (mean Ct: 14.6 ± 0.5) (Figure 3A). Ct values remained low through 6 dpi and ASFV DNA was still detectable at 8 and 10 dpi in pigs that survived to these time points, demonstrating sustained systemic infection following a single oral exposure. Among swab samples, oral shedding showed the earliest onset at 3.2 ± 0.5 dpi, with the lowest Ct values at 6 dpi (mean Ct: 31.0 ± 0.4; Figure 3B and Table 2). Nasal shedding onset was at 4.0 ± 0.0 dpi, with peak viral loads at 6 dpi (mean Ct: 25.2 ± 1.1), yielding the lowest Ct values among the swab sample types (Figure 3C). Rectal shedding was the most delayed, with onset at 4.4 ± 0.4 dpi, and peaked at 6 dpi (mean Ct values of 27.7 ± 1.1) (Figure 3D). Viral DNA in rectal swabs remained detectable through 8 and 10 dpi.

No gross lesions were observed in pigs from the wet- or dry-feed groups. In the wet-feed group, the heart, liver and gallbladder, kidney, spleen, tonsil, mesenteric lymph nodes, colon, and other examined organs showed no ASF-compatible lesions (Figure 4A). Similarly, no ASF-compatible macroscopic lesions were found in the dry-feed group, and their necropsy findings are not presented separately herein.

In contrast, all pigs in the organ homogenate group presented gross lesions compatible with acute ASF at necropsy (Figure 4B and Table 3). The most consistent findings were hemorrhagic lesions in the liver, gallbladder, kidneys, spleen, and lymph nodes (5/5 pigs for each site; Table 3). The liver showed diffuse congestion accompanied by hemorrhages in the gallbladder wall. The kidneys exhibited cortical petechiae, and the spleen exhibited splenomegaly. Focal to multifocal tonsillar hemorrhages and myocardial hemorrhages were each detected in 2 of 5 pigs. No gross lesions were noted in the lungs, colon, or other examined organs (Figure 4B and Table 3).