Primary ovine fetal turbinate (OFTu) cells were used for virus propagation and recombinant virus production. OFTu cells were maintained in minimal essential medium (MEM) (Corning, Glendale, AZ) supplemented with 10% fetal bovine serum (FBS) (Atlanta Biologicals, Flowery Branch, GA), 2 mM L-glutamine, gentamicin (50 μg/ml), penicillin (100 IU/ml), and streptomycin (100 μg/ml). Cells were incubated at 37 °C with 5% CO2.

OV-IA82-Δ120 virus was used as the parental virus to construct a revertant virus for pathogenesis studies, OV-IA82-RV120^3XFlag. A recombination cassette was constructed containing the early-late viral gene ORFV120 with a C-terminal 3XFlag, followed by reporter gene red fluorescent protein sequences under the control of vaccinia virus 7.5K early promoter. To promote homologous recombination, these sequences were flanked by ~700 bp of upstream and downstream ORFV120 sequences. Production of OV-IA82-RV120^Flag was performed as previously described (Khatiwada et al., 2021). Briefly, OFTu cells were infected with OV-IA82 (MOI 1), transfected with the recombination plasmid, and recombinant viruses were isolated by limiting dilution and plaque assay using fluorescence microscopy. The identity and integrity of the viral DNA sequences were confirmed by PCR and DNA sequencing. OV-IA82-RV120^Flag exhibited wild type virus virulence characteristics in the infected sheep.

For microscopic examination of ORFV-infected skin, 70-90 lb lambs were anesthetized with ketamine/midazolam, and full-thickness wounds (FTW) were performed via 3mm punch biopsies in the skin of the inner side of the thighs. Then, 20 ul of a virus stock containing 10^7.5 TCID₅₀/ml of OV-IA82-RV120^3XFlag (infected group, sheep # 12, 21, 30, 104, and 374) or sterile PBS (uninfected group; sheep # 34 and 82) were deposited in the wound bed. To monitor disease progression in the lip, the sheep were additionally inoculated with either OV-IA82-RV120^3XFlag (0.5 ml of a 10^7.5 TCID50/ml inoculum) or PBS in a scarified area of the ventral lip near the left labial commissure. At 24, 48, 60, 72, 96, 120, 144, and 168 h pi, a second, 6mm wide punch was performed around pre-anesthetized (2% lidocaine) inoculation sites to collect the entire wound and surrounding normal skin ( Figure 1A ). Five samples from different infected sheep were collected for each time point except for 24h and 60h (three samples each). For uninfected controls, two samples per time point were collected. Samples were fixed and processed for standard histology. For PCR detection of viral nucleic acids at inoculation sites, six additional sheep (#33, 190, 196, 204, 240, and 533) were infected and sampled as above, with the exclusion of a 60 hpi sample and the inclusion of a 10 dpi sample. PCR samples were frozen at -80C until processing.

Tissue samples were fixed in 10% neutral buffered formalin and embedded in paraffin. Serial tissue sections encompassing the entire wound were obtained perpendicular to the skin surface and mounted on SuperFrost Plus Slides (Fisher Scientific, Pittsburgh, PA). One slide from each series was stained with hematoxylin and eosin (H&E) to evaluate and determine the optimal level and number of additional sections.

The spatial distribution of the late epidermal differentiation markers filaggrin and loricrin, the proliferation marker Ki-67, and cytokeratins (CKs) 6 (CK-6, repair keratinocytes), 10 (CK-10, stratum spinosum keratinocytes), and 14 (CK-14, stratum basale keratinocytes) was investigated by immunohistochemistry (IHC). Sections were deparaffinized, rehydrated, and blocked with BLOXALL^® Blocking solution (Vector Laboratories, Newark, CA). Heat-induced antigen retrieval was performed using citrate buffer (pH 6.0) in a pressure cooker for 15 minutes. After blocking with 5% normal goat serum in PBS, sections were incubated with primary antibodies at 4°C overnight (see antibodies and dilutions in Supplementary Table 1 ). The ImmPRESS^® Goat Anti-Mouse IgG polymer kit (Vector Laboratories; #MP-7452) and the DAB substrate kit (Vector Laboratories; #SK-4100) were used for immunolabelling and visualization, followed by counterstaining with Gill’s hematoxylin. Images were captured using a Keyence BZ-X800 microscope (Keyence, Japan) or NanoZoomer 2.0-HT slide scanner (Hamamatsu, Japan).

To detect OV-IA82-RV120^3XFlag in tissues, immunofluorescence was performed. Sections were deparaffinized, rehydrated, permeabilized with 0.1% triton X-100, and subjected to heat-induced antigen retrieval with citrate buffer (pH 6.0) in a pressure cooker for 15 min. After blocking with 5% BSA in PBS, sections were incubated overnight at 4°C with primary antibody against FLAG, followed by Alexa fluor 488 labeled secondary antibody. Sections were mounted with VECTASHIELD Vibrance Antifade Mounting Medium with DAPI (Vector Laboratories; #H-1800-10). Images were acquired using a Nikon A1 confocal microscope (Nikon, Japan).

Chromogenic visualization of viral replication was performed using RNAscope 2.5 HD brown assay (Advanced Cell Diagnostics, Newark, CA; #322300). To detect ORFV OV-IA82 DNA and RNA, three viral gene targets were selected: immediate early gene ORFV113, early gene ORFV064, and late gene ORFV011 (Joshi et al., 2023). Anti-sense probes for these genes were as follows (targeted genomic segments indicated in parentheses): ORFV113 probe (ACD; #570061), ORFV064 probe (ACD; #1279031-C1), and ORFV011 (ACD; #1180021-C1). Oa-POLR2A, a probe targeting ovine RNA Pol II A subunit (ACD; #615171) and DapB, a probe targeting bacterial dihydrodipicolinate reductase gene (ACD; #310043) served as the positive and negative controls, respectively. All procedures were manually carried out following the manufacturer’s instructions. Briefly, 5-μm-thick sample sections were baked at 60°C for 1 h, deparaffinized in xylene and dehydrated in ethanol. Sections were treated with RNAscope Hydrogen Peroxide reagent (ACD; #322330) for 10 min, followed by antigen retrieval using RNAscope Target Retrieval Reagent (ACD; #322000) at 93°C to 98°C for 15 min. Sections were then incubated with RNAscope Protease Plus Reagent (ACD; #322330) for 30 min at 40°C. To detect viral DNA only, sections were treated with 5mg/ml RNase A (Qiagen; #19101) in PBS. For viral RNA only, sections were treated first with DNase-I (Sigma; #D5319-500UG) diluted 1:50 in 1x DNase buffer, followed by Protease Plus treatment. Probes were hybridized for 2 h at 40°C, followed by signal amplification and detection using DAB. Counterstaining was performed with 50% Gill’s hematoxylin, and images were acquired using a Keyence BZ-X800 microscope.

To characterize the keratinocytes supporting ORFV replication, multiplex RNA-ISH and IF were performed using an RNAscope Multiplex Fluorescent V2 Kit (ACD; #323100) combined with an RNA-Protein Co-detection Ancillary Kit (ACD; #323180). Tissue sections were deparaffinized, dehydrated, and treated with hydrogen peroxide, and antigens were retrieved with 1X Co-Detection Target Retrieval Reagents (ACD; #323165) at 93°C to 98°C for 15 min. Sections were incubated with primary antibodies (CK-10/CK-14, Ki67/CK6, Loricrin/Filaggrin) diluted in Co-Detection Antibody Diluent (ACD; #323160) overnight at 4°C. After washing, sections were post-fixed in 10% neutral buffered formalin, treated with RNAscope protease Plus, and hybridized with probes targeting ORFV064, ORFV113, or ORFV011. The positive and negative controls used were Oa-POLR2A and DapB, respectively. Signal amplification was performed, followed by HRP-C1 and TSA Vivid Fluorophore 520 (Tocris Bioscience, UK; #7523). After RNA-ISH signal development, sections were incubated with the appropriate fluorophore-conjugated secondary antibody diluted with Co-Detection Antibody Diluent for 30 min at room temperature (RT). Antibodies and their dilutions are shown in Supplementary Table 1 . Slides were counterstained with DAPI and mounted with ProLong Gold Antifade Mountant (Thermo Fisher Scientific; #P36930). Images were acquired using a Nikon A1 confocal or Keyence BZ-X800 microscope.

To detect CK-14 RNA and viral RNAs, tissue sections were processed as described above with RNAscope Target Retrieval Reagent (ACD, # 322000) for antigen retrieval, without combining with an RNA-Protein Co-detection Ancillary Kit. Sections were hybridized with ORFV probes and Oa-KRT14-O1-C2 (ACD, #1573381-C2). Signals were developed using TSA Vivid Fluorophore 520 (for ORFV) and 570 (for CK-14) (Tocris Bioscience, #7523, #7526). All other steps were as previously described.

Skin samples were collected from six sheep (#33, 190, 196, 204, 240, and 533) at days 1 and 2 pi, from four sheep (#33, 190, 204, and 533) at days 3 and 4 pi, and from two sheep (#190 and 204) at days 5, 6, 7, and 10 pi, and DNA was extracted using a DNeasy Blood & Tissue Kit (Qiagen, Germany; #69504) following the manufacturer’s recommendations. DNA was analyzed for viral DNA using TaqMan real-time PCR with primers and probes for genes ORFV118, ORFV119, and ORFV121 ( Supplementary Table 2 ). The PCR was performed on a QuantStudio 3 instrument (Applied Biosystems, Waltham, MA) with 20 μl reactions containing 10 μl TaqMan Fast Advanced Master Mix (Applied Biosystems; #4444557), 1 μl each specific primer (500 nM) and probe (250 nM) (synthesized by IDT, Newark, NJ), 7 μl nuclease-free water, and 2 μl template DNA. Reactions were run in triplicate with cycling conditions of 2 min at 50°C and 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. DNA from normal skin was used as a negative control. Average CT values were used to generate a virus growth curve. One-way ANOVA and Tukey’s HSD post-hoc analysis were performed to identify significant differences among time points using GraphPad Prism version 9 (GraphPad Software, La Jolla, CA).

Two sheep were infected (right leg) or mock- infected (left leg) as described above, and skin samples were collected at various times pi, cut into 1mm pieces, and fixed in 2% glutaraldehyde and 2% paraformaldehyde in 0.1M cacodylate buffer (pH 7.2) for 2 h at RT. Post-fixation was performed with 1% osmium tetroxide for 90-120 min at RT. The samples were then dehydrated through a graded ethanol series, transferred in a series of ethanol:Spurr resin mixtures, embedded in 100% Spurr, and polymerized at 65°C for 24-36 h. Ultrathin sections (70-80 nm) were obtained with a Leica EM UC7 Ultramicrotome. Sections were collected on copper grids and stained with 2% uranyl acetate and Reynolds’ lead citrate to improve the contrast and examined using Hitachi HT7800 TEM at 80 KV. Digital micrographs were collected using an AMT NanoSprint1200 CMOS camera.

ORFV replication in vivo has been linked to epidermal repair processes (Jenkinson et al., 1990b), yet questions regarding the timing of infection and identity of the targeted keratinocyte subpopulations remain unanswered. Most ORFV pathogenesis studies in the natural host have relied on the experimental infection of scarified skin. Issues of reproducibility, repair model characterization, and virus dosage accuracy associated with the scarified skin wound (SW) model led us to explore the suitability of the full-thickness excisional wound (FTW) as a model to study ORFV infection (Masson-Meyers et al., 2020). We initially examined skin changes following FTW in uninfected sheep. Three mm punches were performed in the inner side of the thighs of sheep # 34 and # 82, followed by the addition of PBS to the wound bed. Six mm punches were then daily collected around the original punches on days 1-7 post-wound (pw), and tissue samples were processed for standard histology and IHC.

Overall, the microscopic features of the WHR resembled those previously described for an FTW model in humans (Patel et al., 2006). Early neutrophil infiltration of the denuded dermis was replaced by mononuclear infiltration by day 3 pw. Reactive epidermal hyperplasia (i.e. acanthosis) produced a proliferative zone (PZ) around the wound margin that resulted in a 4 to 7-fold increase in epidermal thickness by day 3-5 pw ( Figures 1A, C ; Figure 2 , right panels). PZ keratinocytes were stained by antibodies against CK-10 suprabasally, CK-14 and proliferation marker Ki-67 basally and suprabasally, and filaggrin and loricrin apically ( Figure 2 ). This IHC profile was similar to the one found in resting (unwounded) skin, except the latter showed minimal to no reactivity to Ki-67, and staining for cytokeratins and filaggrin was more intense than in the wounded skin PZ ( Figure 2 , compare left and right panels). In addition, while staining for CK-14 was restricted to the basal layer in the normal epidermis, labeling expanded to most suprabasal layers in the PZ as was also shown in wounded human skin (Aden et al., 2008; Mendoza-Garcia et al., 2015). This effect, however, was less pronounced when CK-14 was stained with the same antibody but using immunofluorescence, or when the tissue was probed for CK-14 transcripts ( Figure 1C ). Starting on days 1-2 pw, keratinocytes at the wound margin migrated centripetally over the wound bed, forming a leading edge (LE) that was positive for CK-14 and injury-induced CK-6, and negative for CK-10, Ki-67, loricrin, and filaggrin, confirming that the migratory front is not engaged in proliferation nor differentiation ( Supplementary Figures 1A, B ; (Patel et al., 2006)). The appearance of CK-10 staining in cells left behind in the LE indicated that keratinocytes started to differentiate by day 5 pw. By days 6 to 7 pw, the wound was fully covered by a new epidermis, with little to no evidence of the LE remaining. The new epidermis was 4 to 5-fold thicker than the resting epidermis around the wound, exhibited the four-strata typical of unwounded skin, and showed a similar CK-10/CK-14/loricrin/filaggrin staining pattern as observed for the PZ on day 3 pw ( Figure 2 , right column). Epidermal rete peg development and inflammatory infiltration of the underlying dermis were negligible by day 7 pw (2 to 5 pegs per 10 sections), and no indication of the de novo formation of hair follicles and skin glands was observed beneath the wound bed.

Eleven sheep (ID # 12, 21, 30, 33, 104, 190, 196, 204, 240, 374, and 533) were infected with OV-IA82-RV120^3XFlag, a virus derived from strain OV-IA82 (Delhon et al., 2004) that carries 3Xflag sequences fused to the C-terminus of ORFV120, a non-essential early-late gene. OV-IA82-RV120^3Xflag replication in vitro and virulence in sheep are indistinguishable from the parental virus. OV-IA82-RV120^3Xflag was inoculated in the inner side of the thighs using the FTW approach, and tissue samples for microscopy (sheep # 12, 21, 30, 104, and 374) and qPCR (sheep # 33, 190, 196, 204, 240, and 533) were collected at various times pi. As a clinical control for orf, the lower lip was scarified near the left labial commissure and topically inoculated with the virus. All sheep showed typical orf lesions in the lip starting at day 3 pi, with gross changes that were comparable to those at the thigh inoculation sites ( Supplementary Figure 2 ).

Previous studies have shown that ORFV antigen cannot be detected in sheep during the first 2 days pi (eclipse period) (Osman, 1976; Jenkinson et al., 1990a), and it has been suggested that IHC might not be sensitive enough to detect the small amounts of antigen present at the early stages pi. Here, samples obtained daily for 7 days pi were processed for RNA in situ hybridization (RNA-ISH) using RNA probes targeting early (ORFV113) and late (ORFV011) viral genes. As negative controls, ISH signals were not observed in samples from infected sheep probed with DapB, that targets bacterial RNA sequences, and no signals for ORFV transcripts were observed for uninfected control sheep # 34 and 82 ( Figure 3B , right panel). Importantly, keratinocytes exhibiting ISH signal for viral transcripts were not observed on multiple sections from samples from sheep # 12, 21, 30, 104, and 374 collected on days 1 and 2 pi with either viral probe. However, a small number of ISH-positive cells with dendritic morphology were observed in the dermis on days 1-4 pi ( Supplementary Figure 4 ). Specific ISH signal in keratinocytes for both viral transcripts and viral DNA was observed at 60 hpi in sheep #21 and in samples from other sheep on days 3 through 7 pi. ( Figures 3A, B ). Nuclease digestion controls indicated that the positive ISH signal was mostly due to the detection of viral RNA, while viral DNA was restricted to intracytoplasmic areas likely corresponding to viral factories ( Figure 3B , central panels). In agreement with ISH data, viral antigen was detected by IF at 60 but not 48 hpi ( Figure 3C ). When DNA from four infected wounds was assessed for ORFV DNA by qPCR, relatively small amounts of viral DNA were detected on days 1 and 2 pi. ( Figure 4 , see legend for sheep IDs). Notably, no major differences in the amount of DNA were observed between day 1 and 2 pi. This changed little on day 3pi in sheep #190, #33, and #533, while in sheep #204 a marked increase in viral DNA was observed. By day 4 pi, the increased CT values for the tested sheep with all three probes were significantly different from day 1 and day 2 pi (p<0.002). High viral DNA levels remained until day 10 pi, the last day examined ( Figure 4 ). Together, data indicate that both viral transcription and DNA replication in the epidermis are prevented or blocked during days 1 and 2 pi in the host, and suggests that, in the FTW model, viral DNA replication starts around day 3 pi. The weak, mostly invariable qPCR signal detected on days 1 and 2 pi likely represents the amplification of input viral DNA.

By day 3 pi, the epidermal thickness in the PZ was of the same magnitude as for the uninfected controls (compare, for example, Figure 5C -a with Figure 2 , right column), and a foci of ballooning degeneration of keratinocytes was observed around the wound margin. From here, the infection spread laterally until day 7 pi (last day sampled for microscopic studies) as mononuclear infiltration of the dermis increased. By day 5 pi, pustules and a few epidermal rete pegs developed ( Supplementary Figure 2 ). By day 7 pi, however, rete pegs were approximately one order of magnitude more abundant than in the uninfected wounds at 7 dpw and became elongated and branched ( Supplementary Figure 3A ). In contrast to uninfected sheep, the epidermal closure of infected wounds was incomplete by day 7pi ( Supplementary Figure 3B ) and the leading edge was still identifiable ( Supplementary Figure 1B ), further indicating that virus infection affects epidermal behavior during healing.

The results here support previous studies showing that ORFV antigens are first detected immediately under the stratum corneum in the repair area between 2 and 3 dpi ( Figure 3A ) (McKeever et al., 1988; Jenkinson et al., 1990a). As the infection spreads laterally around the wound margin ( Figure 1B ), keratinocytes at the front of infection (FOI) show little to no cytopathic effect (CPE, i.e., ballooning degeneration) while the keratinocytes left behind show full CPE ( Figure 1D ). The latter cells, likely not viable, have been shown to lack most if not all organelles and to contain mature virions (Kluge et al., 1972). To characterize newly infected keratinocytes without interference from dying infected cells, we focused on the transition between the intact epidermis and the FOI, which encompasses the newly infected cells but excludes cells with full CPE ( Figure 1D , rectangle). Figures 5A–C show serial sections through this region after multiplex RNA-ISH for viral RNA (ORFV011 probe) and double IF for CK-10/CK-14 (A), CK-6/Ki67 (B), and loricrin/filaggrin (C). Keratinocytes expressing viral transcripts (green) overlap keratinocytes that are positive for CK-10, CK-6, loricrin, and filaggrin, and negative for CK-14 and Ki67, in a region encompassing the stratum spinosum (SS)-stratum granulosum (SG) transition. The lack of CK-14 RNA expression in the area is shown in Figure 5D, a. Together, the results indicate that the upper PZ keratinocytes around the SS/SG transition and supporting viral replication are well into their differentiation program. These cells were found to be Ki-67- and CK-14- negative, indicating they are not proliferating cells. In support of the observations above, CPE and mature ORFV particles were only observed in the upper SS and the SG ( Figure 6 ). No viral transcripts were detected in the LE during the experiment ( Supplementary Figure 1B ).