A parapoxviral virion protein targets the retinoblastoma protein to inhibit NF-κB signaling

Poxviruses have evolved multiple strategies to subvert signaling by Nuclear Factor κB (NF-κB), a crucial regulator of host innate immune responses. Here, we describe an orf virus (ORFV) virion-associated protein, ORFV119, which inhibits NF-κB signaling very early in infection (≤ 30 min post infection). ORFV119 NF-κB inhibitory activity was found unimpaired upon translation inhibition, suggesting that virion ORFV119 alone is responsible for early interference in signaling. A C-terminal LxCxE motif in ORFV119 enabled the protein to interact with the retinoblastoma protein (pRb) a multifunctional protein best known for its tumor suppressor activity. Notably, experiments using a recombinant virus containing an ORFV119 mutation which abrogates its interaction with pRb together with experiments performed in cells lacking or with reduced pRb levels indicate that ORFV119 mediated inhibition of NF-κB signaling is largely pRb dependent. ORFV119 was shown to inhibit IKK complex activation early in infection. Consistent with IKK inhibition, ORFV119 also interacted with TNF receptor associated factor 2 (TRAF2), an adaptor protein recruited to signaling complexes upstream of IKK in infected cells. ORFV119-TRAF2 interaction was enhanced in the presence of pRb, suggesting that ORFV119-pRb complex is required for efficient interaction with TRAF2. Additionally, transient expression of ORFV119 in uninfected cells was sufficient to inhibit TNFα-induced IKK activation and NF-κB signaling, indicating that no other viral proteins are required for the effect. Infection of sheep with ORFV lacking the ORFV119 gene led to attenuated disease phenotype, indicating that ORFV119 contributes to virulence in the natural host. ORFV119 represents the first poxviral protein to interfere with NF-κB signaling through interaction with pRb.


Introduction
Orf virus (ORFV), the type member of the genus Parapoxvirus of the Poxviridae, is the causative agent of orf or contagious ecthyma, a ubiquitous disease of sheep and goats characterized by proliferative lesions affecting muco-cutaneous tissues of the mouth and muzzle [1,2]. Orf is a zoonotic disease affecting humans in close contact with infected animals [3][4][5].
Like other parapoxviruses (PPV), ORFV is highly epitheliotropic, and keratinocytes and their ontogenetically related counterparts in the oral mucosa are the most important if not the only cell type to support ORFV replication in vivo [6]. In addition to producing the essential protective stratum corneum of the epidermis, keratinocytes function as immune sentinels and instigators of inflammatory responses in the skin, representing a specialized branch of the innate immune system. Keratinocytes are well equipped for pathogen sensing as they express a broad spectrum of pattern recognition receptors (PRRs), including surface and endosomal toll-like receptors (TLRs), NOD-like receptors (NLRs), and retinoic acid-inducible gene (RIG-I)-like receptors, and rapidly respond to cell injury and infection by releasing critical pro-inflammatory chemokines and cytokines such as tumor necrosis factor α (TNFα) and interleukin 1 (IL-1) [7,8]. Engagement of these receptors initiates downstream pro-inflammatory cascades, including the NF-κB signaling pathway, which mediates innate immune responses and contributes to skin homeostasis by regulating keratinocyte proliferation and differentiation [9].
Many viruses with dissimilar life styles are known to interfere with the NF-κB pathway. In particular, poxviruses have evolved multiple strategies to counteract NF-κB function, indicating that inhibition of NF-κB-mediated transcription is important for successful infection of the host. This is not surprising as poxvirus infections are sensed by NF-κB-activating PRRs such as endosomal TLRs, RIG-I-like receptors, and the inflammasome [19]. General features of poxviral NF-κB inhibitors include, 1-individual viruses encode for multiple inhibitors, with vaccinia virus (VACV) encoding at least twelve [20]. While orthologs of some NF-κB inhibitors are found in viruses belonging to multiple poxvirus genera (e.g. VACV A52R, VACV E3L), others are restricted to a particular genus (e.g. VACV A46R and VACV B14R in Orthopoxvirus) or even selected viruses within a genus (PPV ORFV002) [21][22][23][24][25][26][27][28][29]; 2-in contrast to other classes of poxviral immunomodulators, poxviral NF-κB inhibitors have no or little resemblance to host proteins; 3-although inhibitors target extracellular, membrane, cytosolic, or nuclear events of NF-κB regulation, most inhibitors directly target NF-κB subunits or the proximal IKKs; 4-despite the multiplicity of inhibitors, there seems to be low or no redundancy as judged by the effect of individual gene deletions on viral pathogenesis; 5-with a few exceptions (myxoma virus MYXV150, cowpoxvirus CPXV006) no single gene-deletion rendered complete virus attenuation [30][31][32][33][34]; 6-most inhibitors are expressed early after virus entry into cells.
The retinoblastoma tumor suppressor, pRb, is a multifunctional, predominantly nuclear protein encoded by the RB1 gene. pRb affects cell cycle, differentiation and metabolism, genome stability and apoptosis, mostly but not exclusively through transcription regulation [40]. Central to pRb function is its ability to nucleate complexes containing multiple interacting partners, thus participating in various regulatory circuits. Viruses have evolved functions to modulate those pathways by targeting pRb to their advantage. For example, adenovirus (Ad) protein E1A interaction with pRb and other factors represses select host genes to promote productive virus infection [41]. The human cytomegalovirus pp71 tegument protein and the human papillomavirus E7 oncoprotein bind to pRb and induce its degradation thus driving cells to a cell cycle stage that potentially favors efficient replication of viral genomes [42,43]. pRb has been shown to affect the regulation of the NF-κB pathway following TNFα signaling or vesicular stomatitis virus infection, although mechanisms involved are not yet defined [44,45]. Many viral and cellular pRb-interacting proteins contain the motif LxCxE (where x means any amino acid) that binds to the LxCxE cleft of the pRb pocket domain [46]. Select pRb-binding proteins that interact through LxCxE with pRb also bind p107 and p130, the other two members of the retinoblastoma family of proteins.
Here we show that ORFV119, a LxCxE motif-containing virion protein unique to PPV, interferes with NF-κB signaling in a pRb-dependent manner early in infection by inhibiting IKK complex activation.

Virus purification and virion protein characterization
Semi-purified viruses were used in infection experiments. OFTu cells infected with OV-IA82, OV-IA82-RV119 Flag , OV-IA82-Δ119, or OV-IA82-RV119 LxGxE-Flag (MOI, 0.1) were disrupted by three cycles of freeze and thaw at three days p.i. Cellular debris were removed by centrifugation at 1500 rpm for 5 min, and supernatants pelleted by ultracentrifugation at 25,000 rpm for 1 h. Pellets were resuspended in MEM and aliquots frozen at -80˚C. Viral titers were obtained by the Spearman-Karber's 50% tissue culture infectious dose (TCID 50 ) method.
To examine the effect of protein synthesis inhibitor cycloheximide (CHX) on expression of ORFV119 and control cellular protein p53. OFTu cells mock treated or treated with CHX (50 μg/ml) (Sigma-Aldrich, St. Louis, MO) for 30 min were infected with OV-IA82RV119 Flag in presence or absence of CHX and harvested at 30 min, 1 h, 1.5 h and 2 h p.i. Total protein extracts (50 μg) were resolved by SDS-PAGE, blotted and transferred to nitrocellulose membranes, probed with antibody against Flag, p53 (sc-6243; Santa Cruz) and GAPDH (sc-25778; Santa Cruz) as described above.

ORFV119 protein expression and subcellular localization
To evaluate ORFV119 expression, OFTu cells were infected with OV-IA82 or OV-IA82-RV119 Flag (MOI, 10) for 2 h, 4 h, 6 h, 8 h, 12 h or 24 h p.i. Total protein extracts (50 μg) were resolved by SDS-PAGE, blotted and transferred to nitrocellulose membranes, and probed with primary antibody against Flag and GAPDH. Blots were then incubated with appropriate HRP-labeled secondary antibodies and developed using chemiluminescent reagents.

Growth curves
To assess the effect of ORFV119 in virus replication, one-step growth curves were determined in OFTu cells infected with OV-IA82, OV-IA82-Δ119 or OV-IA82-RV119 Flag (MOI, 10). Virus yields were quantitated at 6 h, 12 h, 24 h, 36 h, and 48 h p.i. as described above.

NF-κB-p65 nuclear translocation assay
The effect of ORFV119 in TNFα-induced activation of NF-κB was assessed by NF-κB-p65 nuclear translocation assay. HeLa cells transfected with control plasmid (pEGFP-N1) or a plasmid encoding ORFV119-GFP fusion protein (pORFV119 GFP ) were treated with TNFα (20ng/ ml) at 12 h post transfection for 30 min, fixed, permeabilized and blocked as described above, incubated with primary antibody against NF-κB-p65 (Cell Signaling, Cat # 8242s) for 2 h, stained with Alexa fluor 594-labeled secondary antibody (Thermo Fisher Scientific, Cat # A-11005) for 1 h, counterstained with DAPI, and examined by confocal microscopy. Numbers of GFP expressing cells (approximately 300 cells/sample) exhibiting nuclear NF-κB-p65 staining were determined in randomly selected fields and results were shown as mean percentage of GFP/ORFV119 GFP expressing cells containing nuclear NF-κB-p65 over three independent experiments. Statistical analysis of data was performed by using the Student's t test.
To examine the role of ORFV119 on NF-κB-p65 nuclear translocation during ORFV infection, OFTu cells were mock infected or infected with OV-IA82, OV-IA82-RV119 Flag , OV-IA82-Δ119 or OV-IA82-RV119 LxGxE-Flag (MOI, 10). Cells were fixed at 30 min, 1 h, 2 h, 4 h, and 6 h p.i. and processed for NF-κB-p65 staining as described above. Cells (approximately 300/sample) were randomly selected and scored as mean percentage of cells containing nuclear NF-κB-p65 over three independent experiments. Statistical analysis of data was performed by using Student's t test.
To investigate the effects of CHX on NF-κB-p65 nuclear translocation during ORFV infection, OFTu cells were pre-treated with CHX for 30 min, infected with OV-IA82 or OV-IA82-Δ119 in presence or absence of CHX, and fixed at 30 min or 1 h p.i. NF-κB-p65 nuclear translocation assay, scoring, quantification and analysis were performed as described above.

Luciferase assays
The effect of ORFV119 on poly(I:C), poly(A:T) or ORFV DNA induced NF-κB-mediated transcriptional activity was investigated using a NF-κB promoter luciferase assay. Firefly luciferase gene under the control of NF-κB promoter (pNF-κB-Luc) and with a plasmid encoding sea pansy (Renilla reniformis) luciferase under the control of herpesvirus TK promoter (pRL-TK) were used in studies. HeLa cells cultured in 12-well plates were co-transfected with the vectors pNF-κB-Luc (450 ng; Clontech, Mountain View, CA), and pRL-TK (50 ng; Promega, Madison, WI) and pFlag or pORFV119 Flag . At 24 h after transfection, cells were induced with poly(I:C) (Amersham, Pittsburgh, PA) (500ng), poly(A:T) (Inviogen, San Diego, CA) (750ng) or ORFV DNA (1μg). ORFV DNA was extracted from OV-IA82 virus stock using QIAamp DNA Blood Mini Kit (Qiagen, Germantown, MD). Luciferase activities were determined at 20 h postinduction using the Dual Luciferase Reporter Assay (Promega) and a luminometer. Data were analyzed as previously described [35]. Statistical analysis of the data was performed by using Student's t test.
To investigate the effect of ORFV119 on E2F transcriptional activity in ORFV infected cells, OFTu cells were co-transfected with a pE2F-Luc (450 ng; Signosis, Santa Clara, CA) and pRL-TK (50 ng) plasmids. At 24 h post transfection, cells were mock infected or infected with OV-IA82, OV-IA82Δ119, or OV-IA82-RV119 LxGxE-Flag (MOI = 10). Firefly and sea pansy luciferase activities were measured at 1, 2, 4 and 6 h p.i. and expressed as relative fold changes in luciferase activity as described above.

Co-immunoprecipitation
The interaction of ORFV119 or ORFV119 LxGxE with pRb and ORFV119 with TRAF2 was assessed in virus-infected cells by co-immunoprecipitation. OFTu cells infected with OV-IA82-RV119 Flag or OV-IA82-RV119 LxGxE-Flag (MOI, 10) or mock infected were harvested at 12 h p.i. Total protein extraction and co-immunoprecipitation was performed using nuclear complex Co-IP Kit (Active Motif, Carlsbad, CA) following the manufacturer's instructions. For co-immunoprecipitation, total protein extracts were incubated with antibodies against Flag, TRAF2 and pRb in the high stringency buffer (  ; Millipore) at 4˚C for 2 h. Beads were washed four times with high stringency buffer (described above) and bound proteins were eluted in Laemmli buffer. For immunoblot analysis, eluted proteins and control total protein cell lysates were resolved by SDS-PAGE, blotted and transferred to nitrocellulose membranes, probed with antibodies against Flag, TRAF2 and pRb and developed as described above. Light chain specific antibody against Rabbit IgG (Cat # ab99697; Abcam) was used for TRAF2 blots. A flag-expressing protein ORFV113 Flag was used as a control for specificity of ORFV119 and TRAF2 interaction.
The interaction of ORFV119 with TRAF2 also was assessed in Saos-2 cells, which do not express pRb. Saos-2 cells transfected with control plasmid (pFlag) and pORFV119 Flag or cotransfected with pRb (Origene, Rockville, MD) were harvested at 12h post transfection. Coimmunoprecipitations of total proteins extracts were performed as described above. Antibodies for co-immunoprecipitation included anti-TRAF2, anti-Flag and anti-pRb.
Co-immunoprecipitation efficiency was calculated by normalizing the band intensities of co-immunoprecipitated proteins to those corresponding immunoprecipitated proteins and to the expression of corresponding input lysates as previously described [50].

Animal inoculations
Five-month-old lambs randomly allocated to three experimental groups were inoculated with either OV-IA82-Δ119 (n = 4) or OV-IA82-RV119 Flag (n = 4), or mock-infected (n = 3). Following anesthesia, the mucocutaneous junction of the right lower lip near the labial commissure was scarified along a two-centimeter line, and virus inoculum (0.5 ml) containing 10 7.5 TCID50/ml was applied topically using cotton swabs. In addition, the inner sides of hind limbs were scarified in a five-centimeter line and inoculated as above. Animals were monitored for 21 days for the presence of characteristic orf lesions. Pictures were taken of the labial inoculation sites at days 3, 5, 9, 12, 16 and 21 p.i. and the lesion sizes measured with a ruler. Skin biopsy specimens from hind limb inoculation sites were collected at days 2, 5, 8, 12 and 21 p.i. fixed in 10% buffered formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin using standard methods.

Ethics statement
All animal procedures were approved by University of Nebraska-Lincoln Institutional Animal Care and Use Committee (IACUC; protocol 1318) and were performed in accordance with the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching.

Results
ORFV119 is a nonessential early LxCxE motif-containing protein ORFV119 encodes for proteins of 170 to 206 amino acids, with predicted molecular weights of 18.6 to 22.2 kDa and percentage amino acid identities ranging from 77% to 100% (Fig 1). Homologs in Pseudocowpox virus (PCPV) and Bovine papular stomatitis virus (BPSV) are 89% and 54-56% identical to OV-IA82 ORFV119, respectively, whereas no ORFV119 homologue was found in the HL953 strain of parapoxvirus of red deer in New Zealand [51]. ORFV119 lacks homology to known proteins outside the PPV genus, and domains suggestive of protein function. However, a LxCxE motif (OV-IA82 ORFV119 positions 192-196) and a downstream stretch of acidic amino acids usually found in LxCxE motif-containing proteins are located at the C-terminus of the protein and highly conserved in PPV119 proteins (Fig 1). The LxCxE motif is required by several cellular and viral proteins to interact with members of the retinoblastoma family of proteins, which controls important aspects of cell physiology, including cell cycle progression, differentiation, and apoptosis [52]. The ORFV119 LxCxE motif and surrounding sequences follow the pattern XLXCXEXXX, where X should not be a positively charged amino acid (e.g. Lys or Arg) and X should preferably be a hydrophobic residue, which is predicted to bind pRb with high affinity [46]. A predicted mitochondrial localization sequence is found in the N-terminus of most ORFV strains and in PCPV (underlined in Fig 1), but not in BPSV119.
The kinetics of ORFV119 was assessed during ORFV replication in OFTu cells by Western blot using a recombinant virus expressing N-terminally Flag-tagged ORFV119 (OV-IA82-RV119 Flag ). A protein of approximately 32 kDa was detected at 2 h p.i. with increasing protein levels observed at later time points. At 24 h p.i., the last time point investigated, an additional slightly higher molecular weight species of ORFV119 (approximately 35 kDa) was detected (Fig 2A). The observed protein molecular weight was approximately 10 kDa higher than predicted, suggesting that the protein is post-translationally modified in some manner. Previous reports on ORFV replication have shown that early genes were expressed as early as 1 h p.i., with late genes expressed between 6 to 12 h p.i [35][36][37][38]. Newly replicated viral DNA was first detectable at 4 to 6 h p.i. accumulating rapidly between 8 and 16 h p.i [53]. Infectious virus was detected between 16 and 18 h p.i., with continuous virus production until 40 h p.i. [53]. Thus, ORFV119 is a early viral protein.
OV-IA82-Δ119, a virus lacking the ORFV119 gene, exhibited growth kinetics and virus yields comparable to those of wild type OV-IA82 and revertant OV-IA82-RV119 Flag viruses in single step growth curves in OFTu cells, indicating that the gene is non-essential for growth in these cells (S1 Fig).
To examine the intracellular localization of ORFV119, OFTu cells were infected with OV-IA82-RV119 Flag and examined by confocal microscopy at various times post-infection. ORFV119 staining was not evident at 3 and 6 h time points. At 12 h p.i., weak punctate ORFV119 staining was observed in the cytoplasm and adjacent to the plasma membrane. By 16 h p.i., enhanced ORFV119 staining was evident, and fluorescent circular to ovoid structures (389±30nm) were observed in the cytoplasm and especially in close proximity to the cell membrane. Remarkably, at late times p.i. (24 h), OV-IA82-RV119 Flag infected cells exhibited abundant ORFV119 nuclear staining (Fig 2B). Fluorescence was specific for Flag-tagged ORFV119 as no signal was observed in cells infected with OV-IA82-Δ119 at all examined times p.i. (Fig  2B).

ORFV119 interacts with the retinoblastoma protein (pRb) in an LxCxE motif-dependent manner
To investigate whether ORFV119 interacts with retinoblastoma protein pRb, 293T and HeLa cells were transfected with plasmids pORFV119 Flag (ORFV119 Flag ), pORFV119 LxGxE-Flag (ORFV119 LxGxE ) or control plasmid (pFlag), and protein extracts were prepared at 12 h posttransfection. Reciprocal co-immunoprecipitation assays with either anti-Flag or anti-pRb antibodies demonstrated that ORFV119 co-immunoprecipitates with pRb in both cell types. Co-immunoprecipitation of ORFV119 and pRb, however, was not observed with pORFV119 LxGxE-Flag , a plasmid encoding ORFV119 in which C in the LxCxE motif was replaced by G, a change shown to abrogate interaction with pRb (Figs 3A and 3B, S2) [54]. To confirm the interaction in the context of the virus infection, OFTu cells were mock-infected or infected with OV-IA82-RV119 Flag or OV-IA82-RV119 LxGxE-Flag (MOI, 10), and cell lysates  prepared at 12 h p.i. Reciprocal co-immunoprecipitation with either anti-Flag or anti pRb antibodies showed that ORFV119 but not ORFV119 LxGxE-Flag co-immunoprecipitates with pRb (Fig 3C-3F). Together, these results indicate that ORFV119 directly or indirectly interacts with pRb. The observation that the integrity of the LxCxE motif is required for the interaction further suggests that ORFV119 might directly bind pRb.
ORFV119 inhibits NF-κB-signaling and NF-κB-p65 nuclear translocation early during infection Preliminary RNA-Seq experiments indicated increased transcription of multiple NF-κB regulated genes in cells infected with OV-IA82-Δ119 compared to cells infected with OV-IA82 virus, suggesting that ORFV119 inhibits NF-κB signaling. To rule out any confounding effect from cytokines that potentially might be present in the virus inocula, viruses used in these studies were semi-purified as described in Materials and Methods. Real-time PCR analysis of gene expression showed increased levels of NF-κB-regulated genes TNFα (6.68-fold), TLR2 (6.48-fold), NF-κB1 (3.26-fold) and IL36α (3.7-fold) in cells infected with OV-IA82-Δ119 compared to OV-IA82 at 2 h p.i (Fig 4A).
To assess the effect of ORFV119 on NF-κB-p65 nuclear translocation, OFTu cells were mock infected or infected with OV-IA82, OV-IA82-Δ119, OV-IA82-RV119 LxGxE-Flag or OV-IA82-RV119 Flag and NF-κB-p65 localization was examined by immunofluorescence. Infection with OV-IA82-Δ119 and OV-IA82-RV119 LxGxE-Flag but not OV-IA82 or OV-IA82-RV119 Flag led to rapid nuclear translocation of NF-κB-p65 as early as 30 min p.i. (Fig 4B  and 4C). The effect was transient as the percentage of cells expressing nuclear NF-κB-p65 returned to those in wild type and revertant virus-infected cells between 1 and 2 h p.i. (Fig  4C, P<0.05). Notably, levels of NF-κB-p65 nuclear translocation observed for OV-IA82-RV119 LxGxE-Flag were significantly reduced compared to those observed with OV-IA82-Δ119 ( Fig 4C). Consistent with the nuclear translocation kinetics, levels of phosphorylated NF-κB-p65 (Ser536), which accumulates in the cytoplasm prior to nuclear translocation, were increased at early times p.i. (30 min and 1 h) with OV-IA82-Δ119 (S3 Fig). Together, these data show that ORFV119 is a poxviral NF-κB inhibitor acting transiently very early in infection and that ORFV119 LxCxE motif is important for the full inhibitory activity of the protein.

ORFV119 is a virion protein and its early NF-κB inhibitory activity does not involve de novo viral protein synthesis
The observation of ORFV119 staining structures of approximate virion size in infected cells at 16 h p.i. (Fig 2B) together with the early inhibitory effect of ORFV119 on NF-κB signaling suggested ORFV119 may be a virion component available during and/or immediately after virus entry. To examine this possibility, extracellular enveloped virus (EEV) and intracellular mature virus (IMV) were purified from OFTu cells infected with an OV-IA82, OV-IA82-Δ119, OV-IA82-RV119 LxGxE-Flag and OV-IA82-RV119 Flag virus. As in infected cell extracts at 24 h p. i. (Fig 2A), western blot analysis showed a protein doublet (~32 and 35 kDa) corresponding to ORFV119 Flag or ORFV119 LxGxE-Flag in both virion fractions. As expected, the ORFV119 protein was not detected in OV-IA82 (ORFV119 lacks Flag tag) and OV-IA82-Δ119 virions ( Fig  6A). A control virion core protein ORFV086 was detected as a predominant 21 kDa band in OV-IA82, OV-IA82-Δ119, OV-IA82-RV119 LxGxE-Flag and OV-IA82-RV119 Flag EEV and IMV virions [49] (Fig 6A). As a control for potential contamination of purified virions with cellular proteins, a GAPDH control was used. GAPDH protein was not detected in OV-IA82, OV-IA82-Δ119, OV-IA82-RV119 LxGxE-Flag and OV-IA82-RV119 Flag purified EEV and IMV virions (Fig 6A).
To assess whether early inhibition of NF-κB-p65 nuclear translocation by ORFV119 involves de novo viral protein synthesis in the infected cells, OFTu cells were pre-treated with the protein synthesis inhibitor cycloheximide (CHX) for 30 min followed by infection with OV-IA82 or OV-IA82-Δ119 for 30 min and 1 h in presence of the drug. If de novo synthesis of ORFV119 is required for inhibiting NF-κB signaling, then increased levels of NF-κB activation should be observed in OV-IA82 infected CHX treated cells. Low levels of ORFV119 proteins were detected at 30 min and 1 h p.i. likely representing input virion-associated protein; however, ORFV119 protein levels declined in CHX-treated OV-IA82-RV119 Flag infected OFTu cells at subsequent times (Fig 6B). Under these treatment conditions, expression of control host protein p53 also was inhibited (Fig 6C). Immunofluorescence analysis showed that inhibition of NF-κB-p65 nuclear translocation was unaltered in OV-IA82 infected cells regardless of CHX treatment (Fig 6D and 6E). Together, these results indicate that ORFV119 is a virion component, and suggest that virion-associated ORFV119 alone is responsible for early inhibition of NF-κB signaling.

pRb affects ORFV119 inhibition of NF-κB signaling
Given that: 1) ORFV119 interacts with pRb (Fig 3), 2) the ORFV119 LxCxE motif is required for that interaction (Fig 3) and 3) OV-IA82-RV119 LxGxE-Flag -a virus containing a mutation in the ORFV119 LxCxE motif that abrogates pRb binding-was unable to efficiently inhibit NF-κB signaling (Fig 5), we examined the involvement of pRb in ORFV119-mediated inhibition of NF-κB signaling and NF-κB-p65 nuclear translocation in cells either lacking or with reduced levels of pRb.
Additional experiments examining involvement of pRb in ORFV119-mediated inhibition of NF-κB signaling were conducted in human osteosarcoma Saos-2 cells, a pRb-deficient cell line [55]. pRb was not detected in Saos-2 cell extracts by western blot. (Fig 9A, Lane1). If pRb is mediating ORFV119 inhibition of NF-κB-p65 nuclear translocation, increased NF-κB-p65 nuclear translocation would be expected in OV-IA82 virus infected Saos-2 cells. Saos-2 cells were mock infected or infected with OV-IA82 or OV-IA82-Δ119 and NF-κB-p65 localization was examined by immunofluorescence at indicated times p.i. No significant differences in NF-κB-p65 positive nuclei in OV-IA82-or OV-IA82-Δ119-infected cells were observed at any time post-infection (Fig 10A and 10B). Thus, in the absence of pRb, the early NF-κB inhibitory phenotype observed for OV-IA82 is lost. Together, data using cells with reduced pRb levels or lacking pRb altogether indicate that ORFV119-mediated inhibition of NF-κB-signaling is largely pRb-dependent.

ORFV119 interacts with TRAF2, a component of the NF-κB signaling pathway
To assess potential interactions of ORFV119 with components of the TNFα-induced NF-κB signaling pathway, co-immunoprecipitation assays were conducted with antibodies against TRAF2, TAK1, RIP1, TRAF6 and NEMO. OFTu cells were mock-infected or infected with (50 μg/ml) for 30 min were mock infected or infected with OV-IA82-RV119 Flag (MOI, 10) in presence of the drug and harvested at 0.5 h, 1 h, 1.5 h, and 2 h p.i. Cell extracts (50 μg) were resolved by SDS-PAGE, blotted, and probed with anti-Flag, anti-p53 and anti-GAPDH antibodies. (D, E) OFTu cells were drug treated and infected as in B with OV-IA82 or OV-IA82-Δ119 viruses, fixed at 30 min (6D) or 1 h p.i (6E), incubated with anti-NF-κB-p65 antibody, and examined by confocal microscopy. Cells were counted (n = 300 cells/slide) and results are shown as percentage of cells expressing nuclear NF-κB-p65. Results are mean values from two independent experiments. NS, non-significant by Student t test. OV-IA82-RV119 Flag (MOI, 10) and harvested at 12 h p.i. Total cell protein extracts were immunoprecipitated with anti-Flag or anti-target cellular protein antibodies. Reciprocal coimmunoprecipitation demonstrated that ORFV119 co-immunoprecipitates with TRAF2 ( Fig  11A and 11B). Similar results were obtained in 293T cells transiently expressing pORFV119 Flag (Fig 11E and 11F); however, no interaction was observed between TRAF2 and transiently  (S7 Fig). As a control for specificity of ORFV119 and TRAF2 interaction, a Flag tagged ORFV113 Flag was used. No interaction was observed between ORFV113-Flag and TRAF2 in OFTu cells infected with OV-IA82-RV113 Flag (Fig 11C and 11D) nor in 293T cells transfected with pORFV113 Flag (Fig 11G and 11H). In the context of viral infection, reciprocal co-immunoprecipitation of ORFV119 with TAK1, RIP1, TRAF6 and NEMO was not observed. Together, these results indicate that ORFV119 directly or indirectly interacts with the scaffold protein TRAF2. Further, dependence on a LxCxE motif for interaction suggested that pRb might play a role in ORFV119-TRAF2 complex formation.
pRb enhances ORFV119-TRAF2 interaction As ORFV119-mediated inhibition of the NF-κB signaling is pRb dependent (Figs 4, 5, 9 and 10) and ORFV119 interacted with TRAF2 in a LxCxE motif-dependent manner (S7 Fig), we hypothesized that a ORFV119-pRb complex may be required to efficiently interact with TRAF2. To examine this possibility, 293T cells were transfected with pFlag or pORFV119 Flag and harvested at 12h p.i. Reciprocal co-immunoprecipitation assays using total cellular protein extracts with either anti-TRAF2 or anti-pRb antibodies demonstrate that TRAF2 and pRb coimmunoprecipitation, while weak or absent in pFlag transfected cells, is enhanced by the presence of ORFV119 in pORFV119 Flag transfected cells (Fig 12A and 12B).
To further evaluate a pRb requirement for efficient ORFV119-TRAF2 interaction, coimmunoprecipitation experiments were performed in Saos-2, a pRb-deficient cell line. In contrast to results described above (Fig 11), where ORFV119-TRAF2 interaction was observed in pRb expressing OFTu and 293T cells, no interaction was detected in Saos-2 cells (Fig 12D and  12E). Notably, transfection of Saos-2 cells with both pORFV119 and pRb plasmids restored the interaction as reciprocal co-immunoprecipitation of ORFV119 and TRAF2 were observed (Fig 12F and 12G). Taken together, these data indicate that pRb is important for ORFV119--TRAF2 interaction and, further, they suggest that a ORFV119-pRb complex may be required for efficient interaction with TRAF2.

ORFV119 affects ORFV virulence in the natural host
The contribution of ORFV119 to virus virulence was investigated in sheep, a natural ORFV host. Animals were inoculated with OV-IA82-Δ119 (n = 4), OV-IA82-RV119 Flag (n = 4) or PBS (control group, n = 3) in the right labial commissure and the inner side of the thighs, and disease course was monitored for 21 days. All virus-inoculated animals developed clinical orf as evidenced by erythema, papules, pustules, and scabby tissue deposition on the lips (S8A Fig). However, beginning on day 3 p.i., lesions in sheep inoculated with OV-IA82-RV119 Flag were 25% to 90% larger and exhibited more extensive pustules and scabby tissue deposition than those inoculated with deletion mutant virus (S8A and S8B

Discussion
The NF-κB signaling pathway plays key roles in the skin by regulating innate immune responses, inflammation, cell survival and cell proliferation [56][57][58]. Crucial cytoplasmic and nuclear events in the signaling pathway are targeted by various proteins encoded by the highly epitheliotropic ORFV [35][36][37]. Here, we describe an ORFV protein, ORFV119, that interacts with pRb and prevents activation of NF-κB signaling very early during infection. Inhibition of NF-κB-signaling by ORFV119 is largely dependent on its ability to interact with pRb, which prevents activation of the IKK complex and downstream NF-κB signaling.
ORFV119 was shown to interact with pRb in infected and uninfected cells, indicating that other ORFV proteins are not required for the interaction. Similar to oncoproteins of small DNA viruses, binding to pRb was dependent on the ORFV119 LxCxE motif since a CxG substitution in the motif abrogated the interaction completely (Fig 3). This suggests that ORFV119, like small DNA virus oncoproteins, might directly bind pRb. Interestingly, another highly epitheliotropic poxvirus, molluscum contagiosum virus (MCV), encodes a protein (MC007L) unrelated to ORFV119 that localizes to mitochondria and interacts with pRb through an LxCxE motif [59]. The function of MC007L in MCV infection is unknown. pRb has not been previously implicated in antiviral responses against poxviruses; however, it has been associated with NF-κB modulation by other viruses. For example, transiently expressed adenovirus E1A, a pRb-binding oncoprotein, was shown to inhibit NF-κB-dependent transcription induced by TNFα, the effect being dependent on E1A/pRb interaction [44]. And pRb was shown to be required for the activation of the NF-κB pathway in response to vesicular stomatitis virus infection, although viral mechanisms involved were not described [45].
Through association with pRb, ORFV119 inhibits NF-κB signaling by preventing IKK complex activation (Fig 5), with ORFV119-TRAF2 interaction likely underlying the inhibition ( Fig  11). Results suggest that a ORFV119-pRb complex may be required for efficient interaction of ORFV119 with TRAF2, leading to inhibition of NF-κB signaling (Fig 12 and S7 Fig). TRAF2 is a RING finger protein recruited to TNF receptors to regulate NF-κB signaling both positively and negatively [60]. Viral proteins that interact with TRAF2 activate or inhibit NF-κB signaling. For example, the MCV protein MC159 was shown to interact with both TRAF2 and NEMO and to inhibit NF-κB activation [61], while Kaposi's sarcoma associated-herpes virus (KSHV) vFLIP and rotavirus VP4 proteins interact with TRAF2 activating NF-κB signaling [62,63]. The effect of these interactions on TRAF2 function remains unknown. Conceivably, viral proteins might interfere with NF-κB signaling by modulating TRAF2 E3 ubiquitin ligase activity and/or by affecting TRAF2 scaffold functions.
ORFV119 is a virion protein (Fig 6) functioning very early in infection ( 30 min) to inhibit NF-κB signaling. Observed early IKK complex activation on infection of cells with virus lacking ORFV119 (OV-IA82Δ119) indicates ORFV119 inhibits NF-κB signaling induced by an early infection event. Early inhibition of IKK complex activation during infection, examined by Western blot using anti-TRAF2 or anti-Flag antibodies. (E-H) 293T cells were transfected with control plasmid (pFlag), pORFV119 Flag or pORFV113 Flag and harvested at 12 h post transfection. Total cell lysate and protein extracts immunoprecipitated with anti-Flag (E and G) or anti-TRAF2 (F and H) antibodies were examined by Western blot using anti-TRAF2 or anti-Flag antibodies. Results are representative of three independent experiments. Percentage of (A) TRAF2 co-immunoprecipitated by ORFV119 Flag in OV-IA82-RV119 Flag infected cells: 35  ORFV119-TRAF2 interaction in infected cells, and a proposed role for TRAF2 in poxvirus entry [64] suggest that ORFV119 may be inhibiting ORFV entry mediated activation of NF-κB signaling. The importance of viral inhibition of NF-κB activation very early in ORFV infection is further supported by the presence of a second virion-associated NF-κB inhibitor ORFV073 (Fig 13). ORFV073 inhibits NF-κB signaling by preventing activation of IKK complex through interaction with NEMO, the regulatory subunit of the IKK complex [38]. As early infection events, including those related to cell entry, are likely conserved among poxviruses [65], early inhibition of NF-κB signaling in poxvirus infected cells may be of greater biological significance than currently appreciated.
Notably, at late times post-infection (!24 h p.i) ORFV119 also is observed in the nucleus of infected cells (Fig 2), suggesting that in addition to the early virion-associated NF-κB inhibitory function described here, the protein may perform additional functions in the infected cell. The fact that most of pRb localizes to the cell nucleus raises the question as to whether ORFV119 interacts with nuclear pRb at late times post-infection affecting pRb functions such as transcriptional control and cell cycle regulation.
Other ORFV NF-κB inhibitors have been detected in the nucleus. ORFV002 localized to the nucleus and inhibited nuclear phosphorylation of NF-κB-p65 by interacting with mitogen stimulated stress kinase 1 (MSK1) [36,39]. And, ORFV073, a virion-associated NF-κB inhibitor, also was shown to localize to the nucleus yet a nuclear function for the protein has not been described [38]. Among other poxviruses, vaccinia virus protein K1 localized to the nucleus where it prevented acetylation of NF-κB-p65 [66].
Examination of published parapoxviral genomes shows that a genomic region encompassing ORFV119 is affected sporadically by gaps or deletions of as much as 5.7 kbp [67][68][69]. Notably, while the extent of the deletion varied considerably, ORFV119 was always affected. The genomic sequence of the PCPV F00-120R strain, for example, was found to contain a 5.5 kbp DNA deletion that removed genes 116-121 [67]. Retrospective PCR analysis, however, strongly suggested that loss of the 5.5 kbp region occurred during virus passage in tissue culture [68]. Deletions affecting this region were also found in ORFV isolates OV-NP (5.7 kbp deletion) and OV-SJ1 (1.5 kbp deletion), which were isolated from goat lesion material [69]. Although retrospective PCR analysis was not conducted, these virus isolates (cell culture-passage 6) were fully attenuated following inoculation of goats, further supporting the notion that the genomic loss occurred during early passage in cells, and indicating that important virulence functions are encoded within this genomic region. Genomic deletions resulting in complex fragmentation of the BPSV ORFV119 homolog have also been reported [70]. An explanation for instability of this genomic region upon virus passage in cell culture is lacking; control plasmid (pFlag) or pORFV119 Flag . Total cell lysate and proteins extracts immunoprecipitated with antibodies against TRAF2 (A), pRb (B) or Flag (C), were examined by Western blot (WB) using anti-pRb, anti-TRAF2 or anti-Flag antibodies. (D and E) Co-immunoprecipitation of ORFV119 with TRAF2 in Saos-2 cells. Saos-2 cells were transfected with control plasmid (pFlag) or pORFV119 Flag and harvested at 12 h post transfection. Total cell lysate and protein extracts immunoprecipitated with anti-TRAF2 (D) and anti-Flag (E) antibodies were examined by Western blot using anti-Flag and anti-TRAF2 antibodies. (F-H) Coimmunoprecipitation of ORFV119 with TRAF2 in Saos-2 cells transiently expressing ORFV119 Flag and pRb. Saos-2 cells were co-transfected with control plasmid (pFlag) or pORFV119 Flag and pRb plasmids and harvested at 12 h post transfection. Total cell lysate and protein extracts immunoprecipitated with anti-TRAF2 (F), anti-Flag (G) and anti-pRb (H) antibodies were examined by Western blot using anti-Flag, anti-TRAF2 and anti-pRb antibodies. Results are representative of two independent experiments. Percentage of (A) pRb co-immunoprecipitated by TRAF2 in pFlag transfected cells: 5 ORFV119 inhibits NF-κB signaling however, it is possible that loss of these genes may be advantageous for virus growth in cell culture.
Here, infection with a deletion mutant virus lacking ORFV119 resulted in restricted lesion development and reduced time to resolution compared with revertant virus infection in the natural host. The attenuated phenotype of infection likely indicates improved host control of ORFV infection in the absence of ORFV119 (S8 Fig). Our results are at variance with a previous report where no differences in virulence and pathogenesis were observed between wild type and ORFV119 deletion viruses [71]. The discrepancy likely reflects differences in viral strains (IA82 vs SHZ1), extent of genomic deletion, and/or in vitro conditions for virus propagation.
Overall, ORFV pathogenesis studies in the natural host using viruses lacking single NF-κB inhibitor genes have shown a remarkable spectrum of phenotypes, ranging from wild type disease (ORFV002, ORFV024) to moderate (ORFV073, ORFV119), or marked attenuation (ORFV121) [35][36][37][38]. The multiple NF-κB inhibitors encoded by a poxvirus together with the possibility of overlapping or complementing functions may explain these observations. Alternatively, specific poxviral NF-κB inhibitors may exert only subtle and perhaps transient host range effects on specific infected cells or the infected tissue microenvironment. Regardless, the impact of these subtle changes on viral fitness in nature may be difficult to fully ascertain under experimental conditions.