CD1D knockout swine are permissive to African swine fever virus genotype II infection Authors: Daniel Madden, Jessie D. Trujillo, Shanmugasundaram Elango, Isaac Fitz, Chester D. McDowell, Patricia Assato, Eulim Lyoo, Taeyong Kwon, Konner Cool, Yonghai Li, Franco Matias Ferreyra, Natasha N. Gaudreault, Igor Morozov, Kiho Lee, John Driver, Juergen A. Richt Published: 2026 ABSTRACT ABSTRACTAfrican swine fever virus (ASFV) is an important pathogen of domestic and wild suids and the causative agent of African swine fever (ASF). ASFV displays a tropism for myeloid cells, predominately of the monocyte/macrophage lineage, which is critical for ASFV pathogenesis. However, the mechanisms that govern the cellular tropism of ASFV are complex and incompletely understood. ASFV can enter susceptible cells through several mechanisms. One way is by clathrin-mediated endocytosis (CME) via interaction with the host glycoprotein CD1d, encoded by the CD1D gene, and disruption of CD1D expression significantly reduces ASFV replication in cell culture. In order to evaluate the role of CD1d in the replication and pathogenesis of ASF in vivo, CD1D knockout (KO) pigs, generated using a CRISPR-Cas9 system, were challenged with the highly virulent genotype II ASFV isolate MNG19. Notably, CD1D KO pigs were highly permissive to ASFV infection and developed severe acute ASF similar to age-matched wildtype (WT) controls. No significant differences in disease severity or mortality between ASFV-infected WT and KO pigs were observed. ASFV DNA levels in blood or in the majority of visceral and lymphoid tissues were also not significantly different between ASFV-infected wild-type and KO pigs. In addition, pathological changes were similar between both groups and typical of acute ASF, such as fibrinous polyserositis, haemorrhagic lymphadenopathy, and systemic coagulopathy. This work demonstrates that porcine CD1d is not critical for genotype II ASFV replication and virulence in vivo, and that CD1D KO piglets possess no resistance to virulent genotype II ASFV infection. Introduction African swine fever virus (ASFV), the causative agent of African swine fever (ASF), is one of the most significant and devastating viral pathogens of swine globally. Domestic pigs and the Eurasian wild boar are highly susceptible to ASFV infections, and outbreaks of ASF are associated with high mortality rates in affected swine herds and severe economic losses [1]. Progress towards the development of a safe and efficacious ASFV vaccine has been hindered by a variety of factors including the biological complexity of the virus and an incomplete understanding of protective immunity against ASFV [2]. With the exception of two modified live virus (MLV) preparations licenced for use in Vietnam, which have raised concerns regarding safety and efficacy, no vaccines are commercially available for ASFV [3,4]. ASFV possesses a tropism for cells of the monocyte/macrophage lineage, and this is an important contributor to viral pathogenesis [5,6]. Infection of these cells promotes their activation and stimulates a massive release of pro-inflammatory cytokines such as IL-1α, IL-1β, and TNFα which are critical mediators of clinical ASF [5–9]. While the mechanisms governing this tropism are not completely understood, ASFV has been shown to enter susceptible cells through both receptor-mediated and receptor-independent mechanisms. Clathrin-mediated endocytosis (CME) is commonly exploited by viruses for receptor-mediated entry into cells, and ASFV internalization by CME has been described in multiple studies [10–13]. However, to date, no specific cellular receptor has been identified as essential for ASFV internalization, though specific macrophage cellular factors, including CD163, dynamin, clathrin, EGFR, actin, sodium/proton exchanger (Na/H), PKC, PI3 K, phosphorylated PI3 K, Pak1, tyrosine kinase, microtubules, Rab7, CD45, CD203a, MHC II, and activation of Rho-GTPase Rac1, have been shown to be involved in virion adsorption and internalization [14,15]. ASFV can also infect cells through receptor-independent macropinocytosis, a process routinely performed by myeloid-derived antigen presenting cells (APCs) as part of their normal biological functions in immune surveillance [11,16]. Phagocytosis of apoptotic cellular debris containing ASFV virions can also cause productive infection of cells, and ASFV utilizes apoptotic mimicry to facilitate uptake by macrophages [17,18]. Reports of antibody-mediated enhancement of disease in ASFV vaccine studies support Fc receptor-mediated uptake of antibody-coated ASFV virions as another potential pathway for ASFV entry, though investigational studies providing evidence of this entry mechanism are limited [19–21]. The glycoprotein CD1d is a nonclassical MHC protein encoded by the CD1D gene that is expressed on the surface of APCs and presents lipid antigens to natural killer T (NKT) cells and gamma delta T cells [22]. Recent research has demonstrated that CD1d facilitates ASFV entry into MA104 cells and primary porcine macrophages by CME through its interaction with the viral protein p72 and host protein EPS15, and that silencing of cellular CD1d expression or knockout (KO) of the CD1D gene significantly reduces replication of a genotype II HLJ/18 ASFV in vitro [23]. These observations lead to the hypothesis that CD1d could be an important host factor for ASFV replication and pathogenesis in vivo. To test this hypothesis, CD1D knockout (KO) pigs, generated using a CRISPR/Cas9 system, and wildtype (WT) control piglets were infected with the highly virulent genotype II ASFV strain MNG19 to compare disease progression and virus replication between the two groups. This study is the first to evaluate the role of host CD1d in ASFV replication and pathogenesis in vivo. Materials and methods Cells and viruses The virulent genotype II ASFV strain MNG19 was isolated from an infected pig during the 2019 ASFV outbreak in Mongolia, and its pathogenesis in domestic swine has been previously investigated [24,25]. MNG19 was propagated and titrated on primary porcine alveolar macrophages (PAMs) obtained by bronchoalveolar lavage from healthy pigs as previously described [26,27]. PAMs were cultured in RPMI 1640 (Corning, Manassas, VA, USA) supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals, Flowery Branch, GA, USA), 1x MEM nonessential amino acids (Gibco, Waltham, MA, USA), and 1x antibiotic-antimycotic solution (Gibco) and incubated in a humidified 37 °C incubator with 5% CO2 atmosphere. Virus stocks were titrated by hemadsorption (HAD) on 96-well plates, quantified as HAD50/mL, and stored at −80 °C until use. All laboratory work involving the handling of infectious ASFV was carried out at the Biosecurity Research Institute at Kansas State University under BSL-3 conditions. Generation of CD1D knockout pigs The creation and characterization of the CD1D KO pigt line (RRID: NSRRC 0038) used for this study has been previously described [28]. Briefly, in vitro transcribed mRNA encoding a CRISPR-Cas9 vector construct with a gRNA sequence targeting the porcine CD1D primary transcript was injected into zygotes created by in vitro fertilization using gametes derived from WT commercial outbred pigs. Zygotes were cultured for five days, followed by embryo transfer to surrogate sows for gestation. Embryos and offspring were genotyped using CD1D-specific PCR followed by Sanger DNA sequencing. Heterozygous offspring were crossbred to create homozygous founder pigs possessing a 1,599 base pair (bp) deletion in the CD1D gene, including removal of the start codon in exon 3 and deletions of exons 4 and 5 [28]. The circulating leukocytes of previous CD1D KO animals possessing this specific genotype have been characterized by flow cytometry, which demonstrated that CD1D KO pigs fail to develop NKT cells, while other immune cell populations, including conventional, regulatory, and γδ T lymphocytes, B lymphocytes, monocytes, and granulocytes, do not differ from WT pigs [29]. Founder pigs were bred to produce homozygous CD1D KO piglets used for the study. All pigs were confirmed to be homozygous for the CD1D deletion by gel-based PCR (Supplementary Figure 1). Animal experiment Ten 5-week-old CD1D KO piglets obtained from the National Swine Resource and Research Center at the University of Missouri and six 5-week-old conventional outbred WT piglets sourced from Midwest Research Swine (Glencoe, MN, USA) were used for MNG19 infection (Figure 1). Baseline samples, including EDTA and heparinized blood, serum, and oropharyngeal (OP) swabs, were obtained the day before ASFV challenge. On the day of challenge, two CD1D KO piglets were randomly selected and euthanized to serve as negative controls for baseline postmortem evaluation and sample collection; the remaining animals were infected with 180 HAD50 ASFV MNG19 administered by intramuscular injection. Animals were evaluated for clinical symptoms and rectal temperature twice daily and were assigned clinical scores using standardized criteria, which have been previously described [25, 27]; moribund animals meeting clinical criteria for humane euthanasia were euthanized by intravenous pentobarbital administration. EDTA blood and oropharyngeal (OP) swabs were scheduled for collection from animals at –1, 1, 3 , 5 , 7 , 10 , and 14 days post-challenge (DPC), while serum samples were scheduled for collection at –1, 5, 7, 10, and 14 DPC. Necropsy, gross pathological evaluation, and postmortem tissue collection were performed at the time of death/euthanasia, with any animals surviving ASFV infection scheduled to be necropsied at 14 DPC. Figure 1.Timeline of CD1D KO study. Pigs were sampled for EDTA and heparinized blood, serum, and oropharyngeal swabs the day prior to ASFV infection. On the day of challenge (0 DPC), two CD1D KO (CD1D−/−) non-infected pigs were necropsied for baseline evaluation. All other animals were challenged by intramuscular (IM) injection with 180 HAD50 of ASFV strain MNG19. Scheduled post-challenge sampling timepoints for EDTA blood, oropharyngeal swabs, and serum are illustrated. Necropsies were performed at the time of animal death, with any surviving animals scheduled to be necropsied at 14 DPC. Figure created using BioRender. DPC = days post-challenge; green top tube = heparinized blood; purple top tube = EDTA blood; red top tube = serum.An infographic timeline diagram of ASFV MNG19 intramuscular challenge in CD1D knockout and wild type pigs with sampling days.The figure shows an infographic timeline diagram summarizing an African swine fever virus strain MNG19 challenge study in pigs. On the left, a side view of a pig receives an injection into the rear leg. Nearby text states Challenge MNG19, 180 hemadsorbing dose 50, intramuscular. To the right of this pig, a dorsal view pig illustration with exposed internal organs represents baseline necropsy, and notes CD1D knockout necropsy with sample size 2. A horizontal line labeled Days Post Challenge runs from day minus 1 through day 14, with a central dot at day 0. Beneath the line, small groups of collection tubes mark sampling days. At day minus 1, tubes for ethylenediaminetetraacetic acid (EDTA; purple top tube) and heparinized (green top tube) blood, serum (red top tube), and oropharyngeal swabs are shown. At days 1 and 3, groups of two tubes represent EDTA blood and oropharyngeal swabs. At days 5, 7, 10, and 14, sets of three tubes indicate EDTA blood, serum, and oropharyngeal swabs. On the far right, another dorsal pig with exposed organs marks post challenge necropsy. Text beside it states Post Challenge Necropsy, CD1D knockout sample size 8, wild type sample size 6. Timeline of CD1D KO study. Pigs were sampled for EDTA and heparinized blood, serum, and oropharyngeal swabs the day prior to ASFV infection. On the day of challenge (0 DPC), two CD1D KO (CD1D−/−) non-infected pigs were necropsied for baseline evaluation. All other animals were challenged by intramuscular (IM) injection with 180 HAD50 of ASFV strain MNG19. Scheduled post-challenge sampling timepoints for EDTA blood, oropharyngeal swabs, and serum are illustrated. Necropsies were performed at the time of animal death, with any surviving animals scheduled to be necropsied at 14 DPC. Figure created using BioRender. DPC = days post-challenge; green top tube = heparinized blood; purple top tube = EDTA blood; red top tube = serum. All animal work was performed in the Biosecurity Research Institute at Kansas State University under BSL-3Ag conditions. All protocols and work involving animals were reviewed and approved by the Kansas State University Institutional Animal Care and Use Committee (IACUC) (protocol 4770, approved 08/31/2022) and Institutional Biosafety Committee (IBC) (protocol 1600, approved 09/14/2021) and were conducted in accordance with the Animal Welfare Act and applicable state and federal regulations. Gross and histological pathology Necropsy and gross pathological evaluation of infected animals were performed by veterinary pathologists with prior experience working with necropsies and tissues of ASFV-infected pigs. Gross pathological evaluation and lesion scoring were performed as previously described [21,25]. Gross lesions for each animal were quantitatively assessed systematically by organ system including body condition (0–3), integument (0–3), cardiovascular (0–6), liver (0–6), lung (0–12), kidney (0–3), spleen (0–6), and gastrointestinal (GI) tract (0–11). Additional components of the lymphoid system that were scored included the tonsil (0–6), mandibular lymph node (ManLN) (0–6), prescapular lymph node (SCLN) (0–6), cranial mediastinal lymph node (CMLN) (0–6), renal lymph node (RLN) (0–6), gastrohepatic lymph node (GHLN) (0–6), ileocecal lymph node (ICLN) (0–6), and mesenteric lymph node (MesLN) (0–6). Base scores included 0 as indicating no lesions, with increasing scores reflecting increasing lesion severity. Specific criteria for tissue lesion scoring and results for individual animals are listed in Supplementary File 1. Total gross pathology scores for each animal were calculated as the sum of scores for every tissue. Disease severity for each animal was classified based on total gross lesion scores as mild (<20), moderate (21–47), or severe (>47). Tissues collected at necropsy were processed for routine histopathology using standard methodology following a validated tissue fixation procedure for ASFV inactivation. Histological evaluation was performed blindly by a veterinary pathologist. Histological lesions were scored quantitatively using previously described methodology, with a score of 0 indicating no lesion and higher numbers indicating increasing lesion severity [21,25]. Tissues collected, lesion categories assessed for each tissue, and scoring results for all animals are listed in Supplementary File 2. Total histological scores for each animal were calculated as the sum of all individual tissue lesion scores and were classified as mild (0–30), moderate (31–60), severe (61–80), and marked (>80). Viral DNA extractions and quantitative PCR (qPCR) Antemortem samples collected from pigs included EDTA blood, heparinized blood, serum, and OP swabs. White blood cells were isolated from heparinized blood and stored at –80 °C. To generate lysates from blood and swab samples for DNA purification, blood or swab media were mixed 1:1 with lysis Buffer AL and heat inactivated for 10 min at 70 oC prior to processing for DNA extraction. Postmortem samples collected for quantitative qPCR of viral DNA loads included lung, liver, heart, kidney, pooled GI tissue, faeces, bone marrow (BM), spleen, tonsil, ManLN, SCLN, CMLN, GHLN, MesLN, RLN, and ICLN. For postmortem samples, a 10% tissue homogenate was made in Buffer ATL containing 40 µl of proteinase K using a 5 mm stainless steel bead. Tissues were homogenized on the QIAGEN LT homogenizer (QIAGEN, Hilden, Germany) at 30 hertz for 30 s intervals, followed by digestion at 56 oC for one hour. Clarified homogenates were then diluted 1:1 in Buffer AL and heat inactivated for 10 min at 70 oC prior to processing for DNA extraction. DNA was purified from lysates using an automated magnetic bead-based system (Taco-Mini, Genereach, Taichung City, Taiwan) and reagents for total nucleic acid extraction (taco™ DNA/RNA Extraction Kit; Genereach), with minor modifications. To detect and quantify ASFV DNA, TaqMan probe-based qPCR targeting the viral B646L (p72) gene was performed using primer and probe sequences developed by Zsak et al., with minor modifications as previously described [21,25,30]. Briefly, qPCR was performed with PerfeCTA FastMix II (QuantiBio, Beverly, MA, USA) in 20 µL reaction volumes using 5 µL of extracted DNA per reaction. Thermocycling conditions included an initial 95 °C denaturation for 5 min, followed by 45 cycles of 95 °C for 10 s and 60 °C for 1 min. Reactions were performed using a CFX96 Real-Time System (Bio-Rad, Hercules, CA, USA), and data analysis was performed using the Bio-Rad CFX Manager (Version 2.3) software. A standard curve based on serial dilutions of plasmid encoding the B646L target sequence from genotype II ASFV strain Armenia 2007 was used to calculate the concentration of ASFV DNA within samples as p72 copy number per mL (CN/mL). This plasmid was also used for extraction and PCR positive controls. Statistical analysis and images Statistical analysis was performed using GraphPad Prism v9.5.1 (GraphPad Software, Boston, MA, USA). Normality of individual datasets was assessed using the Shapiro-Wilk test. Analyses of post-challenge temperature, blood qPCR, and OP swab qPCR results were performed using multiple Welch’s t tests with correction for multiple comparisons using the Holm-Šídák method, while clinical scores were compared using multiple Mann Whitney tests with correction for multiple comparisons using the Holm-Šídák method. Kaplan-Meier survival curves were compared using the Mantel–Cox test. Tissue qPCR values, gross lesion scores, and histological lesion scores were compared using the Mann–Whitney test for non-normal datasets or Welch’s t test for normal datasets. All graphs were created using GraphPad Prism v9.5.1, with error bars indicating the standard error of the mean where applicable. The diagram of the planned timeline for the animal experiment was created using BioRender (Toronto, Canada). Gross pathological images were attained using a Canon G7X mark II camera (Canon, Tokyo, Japan) in a waterproof case. Histological images were captured using a BX46 microscope (Olympus, Tokyo, Japan) with a DP 28 digital camera (Olympus) and cellSens standard software (Olympus). Results Post-infection clinical progression Following ASFV infection, all animals in both the WT and CD1D KO groups developed clinical illness consistent with acute ASF that rapidly progressed to death or humane euthanasia endpoints (Figure 2). CD1D KO pigs had slightly higher baseline body temperatures than WT pigs; however, these temperatures are within normal limits for 5-week-old pigs (Figure 2A). CD1D KO pigs became febrile 24 h earlier than WT pigs, and body temperature for the CD1D KO group was significantly higher than the WT group at 3 DPC (Figure 2A). No other significant differences in body temperature were observed, with animals in both groups showing declining body temperatures at 6 DPC consistent with advancing ASF hypothermia (Figure 2A). Clinical scores began to rise above baseline at 3 DPC for CD1D KO pigs and 4 DPC for WT pigs; however, this difference was not significant, and clinical scores were not significantly different between groups at any time post-infection (Figure 2B). Animals in both groups began to reach humane endpoints at 5 DPC, and all animals were humanely euthanized or found dead by 6 DPC. No significant difference in survival curves for CD1D KO and WT pigs was detected (Figure 2C). Figure 2.Post-challenge body temperatures, clinical scores, and survival. (A) Body temperatures were compared using multiple Welch’s t tests with correction for multiple comparisons using the Holm-Šídák method. (B) Clinical scores were compared using multiple Mann-Whitney tests with correction for multiple comparisons using the Holm-Šídák method. (C) Survival curves were compared using the Mantel-Cox test. *p < 0.05.Three line graphs comparing WT and CD1D KO piglet temperature, clinical score, and survival over 0 to 8 days post challenge.The figure shows three line graphs labeled A, B, and C, each comparing wild type pigs and CD1D knockout pigs after African swine fever virus infection. Graph A is titled Body Temperature. The x axis is labeled Days Post Challenge, ranging from 0 to 6 in steps of 1. The y axis is labeled Temperature in degrees Celsius, ranging from 38.0 to 42.0 in steps of 1. A dashed horizontal line crosses near 40.5 degrees Celsius indicating the threshold for fever. Two lines with error bars show group means: wild type starts near 39.5 at day 0, rises to about 40.2 at day 2, peaks just above 41.0 at day 4, then falls to about 40.0 at day 6. CD1D knockout starts slightly above 40.0 at day 0, rises by day 1, reaches about 41.3 at day 3, then declines to about 40.0 at day 6. Asterisks appear above days 1 and 3 indicating statistically significant differences determined by multiple Welch's t tests with correction for multiple comparisons using the Holm-Šídák method with a p value less than 0.05. All data are approximate. Graph B is titled Clinical Scores. The x axis is Days Post Challenge from 0 to 6. The y axis is Total score from 0 to 10 in steps of 2. Both groups remain near 0 until day 2, then increase, reaching around 5 for wild type and around 7 for CD1D knockout at day 6. All data are approximate. Graph C is titled Survival. The x axis is Days Post Challenge from 0 to 8. The y axis is Percent Survival from 0 to 100 in steps of 10. Both groups stay at 100 percent until day 5, then drop to 0 by day 6. All data are approximate. Post-challenge body temperatures, clinical scores, and survival. (A) Body temperatures were compared using multiple Welch’s t tests with correction for multiple comparisons using the Holm-Šídák method. (B) Clinical scores were compared using multiple Mann-Whitney tests with correction for multiple comparisons using the Holm-Šídák method. (C) Survival curves were compared using the Mantel-Cox test. *p < 0.05. Viral DNA levels in blood, swabs, and tissues EDTA blood and OP swabs for quantification of ASFV DNA by qPCR were collected pre-challenge and at 1, 3 and 5 DPC. ASFV DNA in blood was detected as early as 1 DPC in 3/8 CD1D KO pigs and 1/6 WT pigs, and by 3 DPC all pigs in both groups were positive for ASFV DNA in blood, with maximum qPCR values occurring at 5 DPC (Supplementary File 3). No significant differences in blood ASFV DNA levels between WT and CD1D KO groups were observed at any time post-infection (Figure 3A). OP swabs became positive for ASFV DNA beginning at 3 DPC for 7/8 CD1D KO pigs and 4/6 WT pigs, and swabs from all animals in both groups were positive at 5 DPC (Supplementary File 3). ASFV DNA levels in OP swabs were not significantly different between groups at any timepoint (Figure 3B). ASFV DNA was detected in all postmortem samples evaluated. While mean ASFV DNA levels in most postmortem samples tended to be slightly higher for the CD1D KO group compared to WT controls, a few significant differences were noted (Figure 4, Figure 5). Levels of ASFV DNA were significantly higher in the pooled GI tissue of CD1D KO pigs compared to WT controls, but no other significant differences were found for any other non-lymphoid postmortem samples including liver, lung, heart, kidney, and faeces (Figure 4). For lymphoid tissues, CD1D KO pigs had significantly higher levels of ASFV DNA in the RLN and spleen, but no significant differences were found for tonsil, ManLN, SCLN, CMLN, GHLN, MesLN, and bone marrow (Figure 5). Figure 4.ASFV qPCR on postmortem organs and faeces. ASFV qPCR values for all samples are displayed as viral p72 copy number per mL (CN/mL). qPCR results for each sample type were compared using Welch’s t test. *p < 0.05.Six bar charts comparing ASFV p72 viral copy number per milliliter between WT and KO pigs in homogenates prepared from liver, lung, heart, kidney, GI pool, tissues and feces.The figure shows six grouped bar charts summarizing African swine fever virus deoxyribonucleic acid measured by quantitative polymerase chain reaction in wild type and CD1D knockout pigs. Each chart is titled with a tissue or sample type: Liver, Lung, Heart, Kidney, Gastrointestinal Pool, and Feces. In every chart, the horizontal axis lists 2 categories, WT and KO. The vertical axis is labeled ASFV p72 copy number per milliliter and uses a base 10 logarithmic scale from 10 to 10 to the power of 8, with tick marks at increasing 10 fold steps. For each category, a single bar is shown with several individual points overlaid to indicate values for individual pigs. Across Liver, Lung, Heart, Kidney, and Feces, bars for wild type and knockout appear similar in height. In the Gastrointestinal Pool chart, the knockout bar is visibly taller than the wild type bar. An asterisk appears above a horizontal bracket linking the wild type and knockout bars in the Gastrointestinal Pool chart indicating statistically significant difference with p value less than 0.05 by Welch's t test. All data are approximate. Figure 5.ASFV qPCR on postmortem lymphoid tissues. ASFV qPCR values for all samples are displayed as viral p72 copy number per mL (CN/mL). qPCR results for ManLN were compared using the Mann Whitney test, while all other tissues were compared by Welch’s t test. *p < 0.05.Nine bar charts comparing ASFV p72 copy number per milliliter in lymphoid tissues from WT and KO pigs, with higher ASFV p72 copy number per milliliter in RLN and spleen of KO pigs.The figure shows 9 grouped bar charts for lymphoid tissues labeled Tonsil, ManLN, SCLN, CMLN, GHLN, MesLN, RLN, BM, and Spleen. Each chart compares wild type pigs, labeled WT, with CD1D knockout pigs, labeled KO. For every chart, the horizontal axis lists WT and KO. The vertical axis is labeled African swine fever virus p72 copy number per milliliter, abbreviated CN per milliliter, with a logarithmic scale from 10 to 10 to the power of 8, increasing in 10 fold steps. Each bar represents the mean for that group with small vertical error bars, and individual pigs appear as dots overlaid on the bars. Across most tissues, WT and KO bars are similar in height. For RLN and Spleen charts, a horizontal bracket above the WT and KO bars is marked with an asterisk, indicating a reported p value less than 0.05. In the RLN chart the KO bar is slightly taller than WT. In the Spleen chart the KO bar is also taller than WT but without statistically significant difference. All data are approximate. ASFV qPCR on postmortem organs and faeces. ASFV qPCR values for all samples are displayed as viral p72 copy number per mL (CN/mL). qPCR results for each sample type were compared using Welch’s t test. *p < 0.05. ASFV qPCR on postmortem lymphoid tissues. ASFV qPCR values for all samples are displayed as viral p72 copy number per mL (CN/mL). qPCR results for ManLN were compared using the Mann Whitney test, while all other tissues were compared by Welch’s t test. *p < 0.05. Gross pathology and histology Gross necropsy findings for both groups were consistent with acute ASF resulting from virulent ASFV infection. Gross pathology lesion scoring for all tissues and animals is detailed in Supplementary File 1, and all animals had total lesion scores classified as moderate or severe. Decreased body condition was observed in all animals, and a majority in both groups displayed skin lesions including erythema and cyanosis of varying severity. All animals showed congestion and haemorrhage in lymphoid tissues that are typical of virulent ASFV. Pulmonary congestion, oedema, and consolidation were universally present, as were pathologies in liver and GI tissues. Congestion and epicarditis were found in the hearts of all pigs in both groups. Abdominal effusion was also consistently found. All animals had total gross lesion scores classified as moderate or severe, with 7/8 CD1D KO pigs and 5/6 WT pigs having severe gross lesion scores (Supplementary File 1). While gross cardiovascular lesion scores were significantly greater for CD1D KO pigs, no other significant differences were present in any organ system evaluated (Figure 6). Total gross lesion scores were not significantly different between both groups (Figure 7A). Figure 6.Postmortem gross lesion scores. Gross lesion scores for body condition, integument, liver, tonsil, ManLN, CMLN, MesLN, RLN, GHLN, and GI were compared using the Mann Whitney test, while scores for kidney, lung, cardiovascular, spleen, and SCLN were compared using Welch’s t test. *p < 0.05.Twelve bar charts comparing gross lesion scores for WT and KO pigs across body condition, organs, and lymph nodes, with one graph (Cardiovascular) statistically significantly higher in the KO group compared to WT.The figure shows twelve grouped bar charts comparing gross lesion scores between wild type pigs and CD1D knockout pigs for multiple tissues. Each chart has wild type bars on the left and knockout bars on the right, with individual pig scores shown as points and error bars extending from each bar. All vertical axes are labeled Gross Score with numeric scales that start at 0 and extend to either 4, 6, 8, or 15, marked in increments of 1 or 2. Horizontal axes list the 2 groups as WT and KO. The first row shows charts for body condition, integument, kidney, lung, and liver. The second row shows cardiovascular, gastrointestinal, spleen, tonsil, and mandibular lymph node, abbreviated as ManLN. The third row shows subscapular lymph node, abbreviated as SCLN, cranial mesenteric lymph node, abbreviated as CMLN, gastrohepatic lymph node, abbreviated as GHLN, renal lymph node, abbreviated as RLN, and mesenteric lymph node, abbreviated as MesLN. In the cardiovascular chart, an asterisk mark appears between the WT and KO bars indicating statistically significant difference by Welch's t test with p value less than 0.05. All data are approximate. Postmortem gross lesion scores. Gross lesion scores for body condition, integument, liver, tonsil, ManLN, CMLN, MesLN, RLN, GHLN, and GI were compared using the Mann Whitney test, while scores for kidney, lung, cardiovascular, spleen, and SCLN were compared using Welch’s t test. *p < 0.05. Histopathology scoring for all tissues and animals is detailed in Supplementary File 2. Similar to gross pathological findings, histopathological lesions were widespread and highly consistent with acute ASF in both groups. Lymphoid tissues were universally affected in both groups and showed characteristic changes including cellular depletion and necrosis, congestion, oedema, haemorrhage, and inflammatory changes. Pulmonary oedema and pneumonia were consistently observed in both groups, as well as renal and hepatic lesions. Lung lesion scores were higher for WT controls than CD1D KO pigs, but no other significant differences in tissue histological scores were present between groups (Figure 8). Total histological scores were not significantly different between groups (Figure 7B). Figure 7.Total gross and total histological lesions scores. (A) Total gross lesion scores for both groups were compared using Welch’s t test. (B) Total histological lesion scores for both groups were compared using the Mann Whitney test.Two bar charts comparing WT and KO pig total lesion scores, with KO bars slightly higher in both gross and histological charts.The figure shows two vertical bar charts labelled A and B, each comparing lesion scores in WT and KO pigs. In the first bar chart, titled Total Gross, the x axis lists groups WT and KO. The y axis is labelled Total Score and ranges from 0 to 80 with tick marks every 20 units. The average of each group is represented by a single bar with individual data points overlaid and short error bars at the top. The WT bar rises to a value slightly above 50, with several circular points scattered around that height. The KO bar reaches a value near the high 50s, with square points distributed above and below a similar region. In the second bar chart, titled Total Histological, the x axis again lists WT and KO. The y axis is labelled Total Score and ranges from 0 to 150 with tick marks every 50 units. The WT bar extends to a value around 120, with circular data points concentrated near that level and one lower value near 80. The KO bar reaches a similar height, slightly above 110, with square points spread around the upper part of the bar and one lower point near 70. Error bars appear at the top of each column. All data are approximate. Figure 8.Histological lesion scores. Histological lesion scores for kidney, lung, liver, tonsil, CMLN, and MesLN were compared using Welch’s t test, while scores for spleen, ManLN, SCLN, GHLN, RLN, and ICLN were compared using the Mann Whitney test. *p < 0.05.Twelve bar charts comparing histological scores for WT and KO piglets across kidney, lung, liver, spleen, tonsil, and lymph nodes.The figure shows twelve grouped bar charts summarizing histological scores for wild type piglets and CD1D knockout pigs after acute African swine fever infection. Each chart has the x axis labeled with 2 groups, WT and KO, and the y axis labeled Histological Score. Kidney, Lung, Liver, Spleen, Tonsil, Mandibular lymph node, Superficial cervical lymph node, Cranial mediastinal lymph node, Gastrohepatic lymph node, Renal lymph node, Mesenteric lymph node, and Iliac lymph node are each shown in a separate chart. Y axis scales range from 0 to either 15 or 20, marked at intervals of 5. In every chart, WT and KO appear as adjacent vertical bars with individual animal scores plotted as points over each bar and error bars indicating variability. Lung shows a marked difference, with the WT bar higher than KO and marked with an asterisk symbol between the 2 groups to denote a reported p value less than 0.05. For the remaining organs and lymph nodes, WT and KO bars appear similar in height, with overlapping error bars and scattered individual data points around comparable score levels. All data are approximate. Total gross and total histological lesions scores. (A) Total gross lesion scores for both groups were compared using Welch’s t test. (B) Total histological lesion scores for both groups were compared using the Mann Whitney test. Histological lesion scores. Histological lesion scores for kidney, lung, liver, tonsil, CMLN, and MesLN were compared using Welch’s t test, while scores for spleen, ManLN, SCLN, GHLN, RLN, and ICLN were compared using the Mann Whitney test. *p < 0.05. Discussion The ASFV major capsid protein p72 can directly interact with the porcine membrane glycoprotein CD1d to initiate clathrin-mediated endocytosis (CME) of extracellular virions, and the disruption of cellular CD1d expression significantly reduces in vitro ASFV replication in both primary porcine macrophages and Chlorocebus sabaeus renal epithelium-derived MA104 cells [23]. However, the relevance of CD1d for ASFV infection and disease progression in vivo has never been evaluated. Therefore, we infected CD1D KO pigs generated using a CRISPR-Cas9 approach with an epidemiologically relevant highly virulent ASFV genotype II isolate (MNG19) to assess whether disruption of CD1D in pigs resulted in any effects on virus replication, viral tissue distribution, and disease severity. Disruption of CD1D in these animals has been shown to inhibit development of NKT cells, immune cells which recognize glycolipid antigens presented by the CD1d MHC-like protein, while having minimal impact on other leukocyte populations. This makes these pigs well suited for dissecting the significance of CD1d-ASFV interactions in vivo [29]. Our results clearly demonstrate that host CD1d is not essential for ASFV invasion, replication and virulence in domestic pigs, and that pigs lacking CD1d are not resistant to lethal infection with ASFV. Following ASFV infection, the disease course was similar for WT and CD1D KO pigs. While CD1D KO pigs developed fever 24 h prior to WT pigs, overall disease progression was not different from WT-infected pigs, and no significant differences in clinical score or survival between groups were found at any time post-infection (Figure 2). Disruption of porcine CD1D had a minimal effect on ASFV replication in vivo. ASFV DNA levels in the blood and OP swabs of CD1D KO and WT pigs were not significantly different at any time post-challenge, indicating efficient virus replication within CD1D KO animals (Figure 3). Similarly, the majority of postmortem samples showed no significant difference in viral DNA levels between the two groups. While viral DNA levels were significantly higher in the GI pool, spleen, and renal lymph nodes of CD1D KO pigs, no significant differences were present in faeces or in samples from seven other lymphoid tissues and four other major organs (Figure 4, Figure 5). Consistent with clinical findings and qPCR results, gross and histological lesions found at necropsy were also largely comparable between both groups. Gross pathological lesions were more severe in cardiovascular tissue of CD1D KO animals, but no other organ systems or tissues evaluated showed a significant difference in gross lesion scores; and total gross lesion scores were similar between both groups (Figure 6, Figure 7A). Except for more pronounced pulmonary histological lesions in WT animals, no significant differences in histological lesions were found, and total histological scores were not significantly different between WT and CD1D KO animals (Figure 7B, Figure 8). All animals in both groups developed moderate to severe histopathological changes consistent with acute ASF (Supplementary File 1, Supplementary File 2). These results show that the disruption of the CD1D gene in domestic pigs has a minimal impact on ASFV invasion, replication, tissue distribution, and pathogenesis in vivo. We recognize that subtle differences in ASFV replication and pathogenesis within the CD1D KO pigs may not have been apparent due to the limited number of animals included in this study (eight CD1D KO pigs and six WT pigs), and different experimental conditions such as inoculation route, infectious dose, or infection with other ASFV strains or genotypes could have affected the outcomes we observed in this study. Figure 3.ASFV qPCR results for clinical samples. qPCR targeting the ASFV p72 gene was performed on DNA extracted from (A) EDTA blood and (B) oropharyngeal (OP) swabs, with results calculated as p72 copy number per mL (CN/mL). Results for both datasets were analysed using multiple Welch’s t tests with correction for multiple comparisons using the Holm-Šídák method.Two line graphs comparing ASFV p72 DNA copy number over time in blood and oropharyngeal swabs from WT and KO pigs post challenge.The figure shows 2 line graphs labeled A and B that compare African swine fever virus deoxyribonucleic acid levels in wild type pigs and CD1D knockout pigs over time. Graph A is titled ASFV quantitative polymerase chain reaction Blood. The horizontal axis is labeled Days Post Challenge with 4 tick labels: Pre Challenge, 1, 3, and 5. The vertical axis is labeled ASFV p72 copy number per milliliter with a logarithmic scale from 10 to 10 to the power of 8, increasing by powers of 10. Two lines with error bars represent wild type and knockout groups. Both lines start near 10 at Pre Challenge, rise slightly by day 1, and then jump to about 10 to the power of 6 by day 3. By day 5, both lines rise modestly toward about 10 to the power of 7, closely overlapping across all days. Graph B is titled ASFV quantitative polymerase chain reaction OP Swabs. The horizontal axis again shows Days Post Challenge with 1, 3, and 5. The vertical axis and scale match graph A. Both wild type and knockout lines start near 10 at day 1 and increase steadily through day 3 and day 5 toward between 10 to the power of 4 and 10 to the power of 6, with similar trajectories and overlapping error bars. A legend on each graph labels the wild type and knockout lines. All data are approximate. ASFV qPCR results for clinical samples. qPCR targeting the ASFV p72 gene was performed on DNA extracted from (A) EDTA blood and (B) oropharyngeal (OP) swabs, with results calculated as p72 copy number per mL (CN/mL). Results for both datasets were analysed using multiple Welch’s t tests with correction for multiple comparisons using the Holm-Šídák method. The data presented here highlight the complex biology of ASFV’s cellular tropism and cell entry. ASFV displays an in vivo tropism for myeloid-derived monocytes and macrophages, though it can also be detected in other cell types, including lymphocytes, megakaryocytes, and neutrophils; it is also capable of replicating in cell lines derived from Chlorocebus sabaeus (green monkey) kidney epithelial cells (5,26). A receptor-mediated entry mechanism exploiting host cell molecules expressed on both monocytes and macrophages would provide a simple explanation for the observed in vivo cellular tropism of ASFV; it is well established that ASFV can utilize CME, a process exploited by a variety of viruses for receptor-mediated entry to infected host cells [10–13,31]. However, to date, no specific cellular receptor for ASFV entry has been identified [14]. CD163, a marker specific for cells of the monocyte/macrophage lineage, which is most abundantly expressed on terminally differentiated macrophages and tissue dendritic cells, confers susceptibility to porcine reproductive and respiratory syndrome virus (PRRSV) and was previously hypothesized to be a receptor necessary for ASFV infection. This was based on experimental data showing increased susceptibility in monocytes/macrophages with higher levels of CD163 expression, as well as the ability of an anti-CD163 monoclonal antibody to inhibit ASFV binding and infection of PAMs in a dose-dependent manner [32–34]. However, CD163 KO pigs generated using a CRISPR-Cas9 system showed no resistance to infection with ASFV genotype II strain Georgia 2007/1, and virus replication in macrophage cultures derived from these animals showed no difference compared to WT primary cells, indicating that the CD163 protein by itself is not essential for ASFV replication [35]. Multiple ASFV proteins have previously been identified to function in virus binding to susceptible cells, including p12, p54, and p72 [36–39]. However, antibodies targeting these proteins are unable to completely neutralize ASFV in vitro, suggesting virus attachment and entry can occur through mechanisms independent of the interaction between these viral proteins and their respective cellular ligands [37–39]. The current consensus is that ASFV can utilize multiple distinct pathways for cell entry, including receptor-dependent CME as well as receptor-independent processes involving phagocytosis and macropinocytosis [13,14,16–18]. An important corollary that can be drawn from this model of ASFV entry is that, while receptor-mediated mechanisms may facilitate ASFV infection, they may not be an essential requirement. Furthermore, fundamental differences between in vitro and in vivo infection might indicate that receptor-independent entry mechanisms seem more important for ASFV replication in vivo. Both of these conclusions are compatible with previously described observations of CD1d facilitating ASFV entry and replication in vitro and the current data showing disruption of porcine CD1D has a minimal impact on ASFV invasion, replication and virulence in vivo. CD1D KO pigs fail to develop invariant NKT (iNKT) cells, and it is plausible this would impact cytokine production and cellular immune responses to ASFV infection. However, the detailed characterization of the immune responses of ASFV-infected CD1D KO pigs compared to ASFV-infected WT pigs was beyond the scope of this study. Follow-up research is needed to characterize immune responses to ASFV infection in CD1D KO pigs and assess whether the lack of NKT-mediated immune responses influences disease progression. It is also possible that specific ASFV entry mechanisms may have variable importance among different ASFV genotypes or strains. ASFV can modulate host immune responses through multiple mechanisms including suppression of type I interferon responses, apoptosis, and blocking NF-κB- mediated cytokine induction, and this ability is an important determinant of ASFV virulence [15]. Importantly, a significant amount of variability in genome content and virulence exists among high-, moderate-, and low-virulence or attenuated ASFV isolates, and ASFV studies should be interpreted in the context of the specific virus strains used for infection [5,40]. However, it is unknown whether this variability has an impact on the interaction between ASFV and host CD1d. Additionally, it is possible that highly virulent ASFVs, such as the strain used in this study, may induce death of infected animals before small differences in viral replication and tissue distribution could manifest at a detectable level, and that host CD1d may have greater importance during infection with low-virulence ASFV isolates. Consequently, additional challenge studies using different ASFV isolates, including strains of moderate and low virulence, are necessary to determine whether these results are specific for the highly virulent MNG19 strain or more broadly applicable to other ASFVs. Overall, our study demonstrates that CD1d is not essential to the invasion, replication and virulence of a genotype II ASFV in vivo, and the lack of CD1d expression does not significantly alter the pathogenesis of ASFV infection in domestic pigs. Further studies with different ASFV isolates of other genotypes as well as moderate and low virulence are necessary to determine whether this phenomenon is specific to virulent genotype II ASFVs or is more broadly applicable. Supplementary Material Supplementary_file_2_histopathology_Final.xlsx supplementary_file_1 _gross_pathology_Final.xlsx supplementary_file_3_clinical_qPCR_summary_Final.xlsx supplementary_figure_1.png