Orf Virus-Based Vaccine Vector D1701-V Induces Strong CD8+ T Cell Response against the Transgene but Not against ORFV-Derived Epitopes

The potency of viral vector-based vaccines depends on their ability to induce strong transgene-specific immune response without triggering anti-vector immunity. Previously, Orf virus (ORFV, Parapoxvirus) strain D1701-V was reported as a novel vector mediating protection against viral infections. The short-lived ORFV-specific immune response and the absence of virus neutralizing antibodies enables repeated immunizations and enhancement of humoral immune responses against the inserted antigens. However, only limited information exists about the D1701-V induced cellular immunity. In this study we employed major histocompatibility complex (MHC) ligandomics and immunogenicity analysis to identify ORFV-specific epitopes. Using liquid chromatography-tandem mass spectrometry we detected 36 ORFV-derived MHC I peptides, originating from various proteins. Stimulated splenocytes from ORFV-immunized mice did not exhibit specific CD8+ T cell responses against the tested peptides. In contrast, immunization with ovalbumin-expressing ORFV recombinant elicited strong SIINFEKL-specific CD8+ T lymphocyte response. In conclusion, our data indicate that cellular immunity to the ORFV vector is negligible, while strong CD8+ T cell response is induced against the inserted transgene. These results further emphasize the ORFV strain D1701-V as an attractive vector for vaccine development. Moreover, the presented experiments describe prerequisites for the selection of T cell epitopes exploitable for generation of ORFV-based vaccines by reverse genetics.


Introduction
Induction of strong transgene-specific immune responses and negligible anti-vector immunity are the prerequisites of efficient viral vector-based vaccines [1][2][3]. Recently, Orf virus (ORFV, Parapoxvirus) was reported as a novel viral vector for the expression of various foreign antigens. The very restricted host range (sheep and goats) and the exceptionally strong immune-modulating properties make ORFV a promising viral vector candidate [4][5][6][7][8][9]. Importantly, different derivatives of the Vero cell culture adapted vector virus strain D1701-V were apathogenic even after high dose infection of immunosuppressed sheep [10]. Whether and which of the ORFV genes deleted in D1701-V remains to be clarified [11].
The apathogenic Vero cell culture-adapted ORFV strain D1701-V has been used to generate recombinants by substituting viral genes with the foreign genes, which led to further attenuation [5,10,12]. ORFV-based vaccines have mediated protective immunity against different viral infections, such as rabbit hemorrhagic disease virus [13], classical swine fever [14], Borna disease virus [15], pseudorabies virus [12,16,17], rabies virus [18] and influenza A virus [19]. Lately, the therapeutic application of ORFV vector-based vaccine against tumours induced by cottontail rabbit papillomavirus has been described [20]. The major advantage of using recombinant ORFV for vaccination is based on the fact that ORFV in general elicits only a short-lived virus-specific immunity in its natural hosts allowing frequent reinfections due to the absence of virus neutralizing antibodies [4,5,[21][22][23][24][25][26][27]. In turn, this feature enables repeated immunizations using ORFV recombinants to enhance humoral immune responses against inserted antigens [13,[15][16][17][18][19]. Analyses of human patients accidentally infected with wild-type ORFV, whose lesions generally resolve in short times, suspected a human immune response very similar to that of sheep [28]. However, it is not known whether a vaccination using D1701-V ORFV induces ORFV-specific T lymphocyte responses.
Induced vector virus-specific T cell responses can limit the immunogenicity of vector-encoded transgenes [2,29]. Some studies indicate that cellular immunity against viral epitopes prevents efficient priming of T cell responses to the antigens delivered by recombinant viruses [30][31][32]. Furthermore, cytotoxic T lymphocyte (CTL) responses to immunodominant epitopes of vector proteins have been shown to suppress CD8+ T cell priming to subdominant epitopes of the transgene [29]. While increased numbers of activated CTLs have been found after Orf virus reinfection at the site of lesion in sheep [22], CD8+ T cells did not appear to be essential for viral clearance later in the infection [21]. Hence, we assume that along with the lack of specific ORFV-neutralizing antibodies, cellular immunity against D1701-V ORFV vector is also negligible.
Understanding the specificity of anti-vector responses requires evaluation of both the epitopes presented and the corresponding CD8+ T cells. For this, virus-derived major histocompatibility complex (MHC) class I restricted peptides have to be identified and any elicited CD8+ T lymphocytes detected and quantified. For small viruses, overlapping libraries of synthetic peptides can be used for the epitope discovery [33,34], while this technique is not applicable for large viruses of the poxvirus family including ORFV. Instead, epitope search for those viruses is mostly based on in silico prediction of the MHC I-bound peptides [35][36][37]. However, many of these peptides might not be of physiological relevance if they are not presented on the cells during infection [36,38]. Thus, the identification of specific MHC-associated peptides, or immunopeptidome, which are naturally processed and presented by the virus infected cells employing mass spectrometry has become a feasible alternative [38][39][40][41][42]. For example, by using this approach 73 H-2K b and 97 H-2D b vaccinia virus (VACV)-derived peptides have been described for murine MHC I molecules [43], as well as 10 and 64 peptides for human leukocyte antigen (HLA)-A2 and B7, respectively [44]. For the modified vaccinia virus Ankara (MVA), 98 unique HLA class I associated peptides have been published [40].
In this study we report for the first time the identification of ORFV-specific epitopes in a combined approach of MHC ligandomics and immunogenicity analysis. Using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and database annotation we detected 36 peptides as ligands for mouse MHC class I allele H-2K b , originating from various ORFV proteins. Immunogenicity of the identified peptides was evaluated in mice after two times administration of ORFV recombinants. We demonstrate that D1701-V ORFV does not induce CD8+ T cell responses against identified virus-derived MHC class I restricted peptides, but a strong CTL immune response directed against the encoded transgene.

Infection of Cells
For flow cytometric analysis, 10 5 cells were seeded into 24-well plate (Greiner CELLSTAR, Kremsmünster, Austria) in 0.5 mL medium and infected by adding appropriate volume of indicated virus to reach the required multiplicity of infection (MOI). For the isolation of ORFV-derived H-2K b -presented peptides 10 8 HeLa-K b cells were infected with V-D12-mCherry (MOI = 5) in 10 mL of medium. Two hours post infection, the cell concentration was adjusted to 10 6 per ml and cells were cultured for additional 18 h. After harvesting, cells were washed twice with phosphate buffer saline (PBS) and the cell pellet was frozen at −80 • C.

Flow Cytometry Analysis of Infected Cells
Cell viability of HeLa-K b cells 20 h post infection was determined by Zombie Aqua staining (BioLegend, San Diego, CA, USA) followed by flow cytometry. The infection rate was calculated by flow cytometric detection of mCherry or Green Fluorescent Protein (GFP) expressing cells. Quantification of H-2K b molecules on the surface of ORFV-infected HeLa-K b cells was done using QIFIKIT (Dako, Glostrup, Denmark) according to the manufacturer's protocol. Detection of Ova 257-264 SIINFEKL peptide on the surface of ORFV-infected cells was performed after staining with Allophycocyanin (APC) anti-mouse H-2K b bound to SIINFEKL antibody (BioLegend, San Diego, CA, USA; clone 25-D1. 16). Samples were acquired on a BD Fortessa flow cytometer (BD Biosciences, San Jose, CA, USA) and data were analyzed using FlowJo software version 10 (Tree Star Inc., Ashland, OR, USA).

Database Search and Filtering
LC-MS/MS results were processed using Proteome Discoverer (v.1.3; Thermo Fisher Scientific, Waltham, MA, USA) and database search was performed with the Sequest HT search engine (University of Washington, Seattle, WA, USA). The human proteome was used as reference database annotated by the UniProtKB/Swiss-Prot, status 27 September 2013 containing 20,279 reviewed sequences plus the ORFV proteome [48]. The search combined data of three technical replicates was not restricted by enzymatic specificity and oxidation of methionine residues was allowed as dynamic modification. Precursor mass tolerance was set to 5 ppm, and fragment mass tolerance to 0.02 Da. False discovery rate was estimated using the Percolator node [49] and was limited to 5%. For MHC class I ligands, peptide lengths were limited to 8-12 amino acids. MHC annotation of peptides fitting to the H-2K b peptide motif was performed using NetMHCpan4 database [50,51].

Immunization of Mice
Female C57BL/6 mice (8-12 weeks old) were obtained from Jackson Laboratories (Jackson Labs, Bar Harbor, ME, USA) and were housed in the biosafety level 1 animal facility at the University of Tübingen, Germany. All animals were handled in strict accordance with good animal practice and complied with the guidelines of the local animal experimentation and ethics committee-experiments were conducted under Project License (Nr. IM 1/15). Mice were immunized intramuscularly (i.m.) in the anterior tibialis with 7.5 × 10 6 plaque forming units (PFU) of V12-Ova-D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 days mice received a homologous i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D12-mCherry. Mice were sacrificed one week after the second administration, and the splenocytes were used for immunological testing as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFN-Vaccines 2020, 8, x FOR PEER REVIEW D12-GFP or control V-D12-mCherry ORFV recombinant. Afte i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP o one week after the second administration, and the splenocyte as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFN ƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Im separately from individual mice or pooled as described in splenocytes per well were either stimulated with peptide po Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide synthetic peptides, corresponding to the identified ORFV-der Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzer control, unstimulated cells served as a negative control. Devel using an ImmunoSpot S5 analyzer (Cellular Technology ImmunoSpot software (Cellular Technology Limited, Cle considered to be positive when the mean spot count per wel mean number of spots in negative control wells. Overlapp response hamper the detection of reliable counts, therefore sp 1.000. ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immune splenocytes were tested either separately from individual mice or pooled as described in the "Results" section. In total 2 × 10 5 splenocytes per well were either stimulated with peptide pools (10 µg/mL of each peptide) or with Ova 257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide pool contained eight randomly mixed synthetic peptides, corresponding to the identified ORFV-derived ligands (numbers 1-32 as listed in Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland) (10 µg/mL) was used as a positive control, unstimulated cells served as a negative control. Developed spots were automatically counted using an ImmunoSpot S5 analyzer (Cellular Technology Limited, Cleveland, OH, USA) and ImmunoSpot software (Cellular Technology Limited, Cleveland, OH, USA). Responses were considered to be positive when the mean spot count per well was at least threefold higher than the mean number of spots in negative control wells. Overlapping spots in wells with strong IFN-D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 days mice received a homologous i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D12-mCherry. Mice were sacrificed one week after the second administration, and the splenocytes were used for immunological testing as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFNƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immune splenocytes were tested either separately from individual mice or pooled as described in the "Results" section. In total 2 × 10 5 splenocytes per well were either stimulated with peptide pools (10 µg/mL of each peptide) or with Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide pool contained eight randomly mixed synthetic peptides, corresponding to the identified ORFV-derived ligands (numbers 1-32 as listed in Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland) (10 µg/mL) was used as a positive control, unstimulated cells served as a negative control. Developed spots were automatically counted using an ImmunoSpot S5 analyzer (Cellular Technology Limited, Cleveland, OH, USA) and ImmunoSpot software (Cellular Technology Limited, Cleveland, OH, USA). Responses were considered to be positive when the mean spot count per well was at least threefold higher than the mean number of spots in negative control wells. Overlapping spots in wells with strong IFN-ƴ response hamper the detection of reliable counts, therefore spot counts of >1.000 per well were set to 1.000.

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual mice or pooled as described in the "Results" section; 2 × 10 6 cells per well were seeded into 96-well plate (Corning, NY, USA) and stimulated for 16 h with individual ORFV-derived peptides (10 µg/mL), with 1 µg/mL of Ova257-264 SIINFEKL peptide, or with control ORFV at MOI 1 in the presence of brefeldin A (

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual mice or pooled as described in the "Results" section; 2 × 10 6 cells per well were seeded into 96-well plate (Corning, NY, USA) and Flow cytometric measurements and data analysis were performed as described above.

In Vitro Re-Stimulation of Splenocytes
In total 2 × 10 6 pooled splenocytes from V12-Ova-D12-GFP or from control V-D12-mCherry immunized mice per well were seeded into a 24-well plate (Greiner CELLSTAR, Kremsmünster, Austria). Cells were re-stimulated either with 10 µg/mL of the individual synthetic peptides representing identified ORFV-derived epitopes or with 1 µg/mL of Ova 257-264 SIINFEKL peptide. Unstimulated cells were used as a negative control. IL-2 was added on day 2 and 5 after re-stimulation at a final concentration of 20 U/mL. On day 7, splenocytes were re-stimulated with 4 × 10 6 feeder cells per well (200 Gy irradiated splenocytes from syngeneic nonimmune mice) and loaded with 10 µg/mL of individual synthetic peptides. IL-2 was added on days 9 and 12 as indicated above. After 14 days cells were harvested, and ICS was performed as previously described.

Statistical Analysis
Statistical significance was examined with GraphPad Prism 4.02 software (GraphPad, San Diego, CA, USA). For comparison between two conditions or groups, an unpaired student's t-test was used to determine significance. A value of p < 0.05 was considered significantly different.

ORFV Vector Strain D1701-V Efficiently Induces Transgene-Specific CD8+ T Cell Response
To date, the induction of CD8+ T cell responses by ORFV strain D1701-V has not been analyzed in detail. In order to test whether a homologous immunization regimen with recombinant D1701-V ORFV elicits a specific CD8+ T cell response to the vectored antigen, V12-Ova-D12-GFP encoding Ova was injected to C57BL/6 mice (H-2K b positive) twice by i.m. route. For negative control mice were immunized with the control recombinant V-D12-mCherry. The immune response against the H-2K b -restricted CD8+ T cell epitope SIINFEKL was measured in splenocytes one week after the second immunization We observed that V12-Ova-D12-GFP administration elicited a strong Ova-specific CD8+ T cell response. Ex vivo quantification of CTLs by H-2K b Ova 257-264 dextramer staining showed a high frequency of 42.9% specific CD8+ T cells ( Figure 1A). The functionality of Ova-specific CD8+ T lymphocytes was measured by production of the pro-inflammatory cytokines interferon-gamma (IFN-γ), tumor necrosis factor alpha (TNF-α) and interleukin-2 (IL-2), as well as by the expression of lysosomal-associated membrane protein 1 (LAMP-1) known as CD107a. The results revealed that IFN- D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 days mice received a homologous i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D12-mCherry. Mice were sacrificed one week after the second administration, and the splenocytes were used for immunological testing as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFNƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immune splenocytes were tested either separately from individual mice or pooled as described in the "Results" section. In total 2 × 10 5 splenocytes per well were either stimulated with peptide pools (10 µg/mL of each peptide) or with Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide pool contained eight randomly mixed synthetic peptides, corresponding to the identified ORFV-derived ligands (numbers 1-32 as listed in Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland) (10 µg/mL) was used as a positive control, unstimulated cells served as a negative control. Developed spots were automatically counted using an ImmunoSpot S5 analyzer (Cellular Technology Limited, Cleveland, OH, USA) and ImmunoSpot software (Cellular Technology Limited, Cleveland, OH, USA). Responses were considered to be positive when the mean spot count per well was at least threefold higher than the mean number of spots in negative control wells. Overlapping spots in wells with strong IFN-ƴ response hamper the detection of reliable counts, therefore spot counts of >1.000 per well were set to 1.000.

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual mice or pooled as described in the "Results" section; 2 × 10 6 Figure 1B), whereas no Ova-specific response was detected in mice immunized with negative control ORFV ( Figure 1A,B). Notably, the CTL response against Ova-derived epitope was dominated by multifunctional CD8+ T cells producing simultaneously IFN-Vaccines 2020, 8, x FOR PEER REVIEW D12-GFP or control V-D12-mCherry ORFV recombinant. Afte i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP o one week after the second administration, and the splenocyte as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFN ƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Im separately from individual mice or pooled as described in splenocytes per well were either stimulated with peptide poo Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide synthetic peptides, corresponding to the identified ORFV-der Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerl control, unstimulated cells served as a negative control. Devel using an ImmunoSpot S5 analyzer (Cellular Technology ImmunoSpot software (Cellular Technology Limited, Cle considered to be positive when the mean spot count per well mean number of spots in negative control wells. Overlapp response hamper the detection of reliable counts, therefore sp 1.000.

MHC Dextramer Staining
Splenocytes of individual mice were stained with H-2 Kopenhagen, Denmark) according to the manufacturer's prot was performed with anti-CD90.2-APC (BioLegend, San Die Alexa 700 (BioLegend, San Diego, CA, USA; clone 53-6.7) Diego, CA, USA; clone 6D5) antibodies for 30 min at 4 °C. C viability staining was performed using Zombie Aqua (BioLeg the manufacturer's instructions. Samples were acquired on Biosciences, San Jose, CA, USA) and data were analyzed using Inc., Ashland, OR, USA).

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individu "Results" section; 2 × 10 6 cells per well were seeded into 9 stimulated for 16 h with individual ORFV-derived peptides SIINFEKL peptide, or with control ORFV at MOI 1 in the pres  These results demonstrate that ORFV strain D1701-V mediated strong transgene-specific CD8+ T cell immunity in mice after two intramuscular immunizations, which was not induced by the vector virus without foreign immunogen.

HeLa-K b Cells Are Suitable Targets for the Identification of ORFV-Derived Peptides
No ORFV-specific T cell epitopes have been identified so far. To generate a comprehensive inventory of ORFV-derived MHC class I restricted peptides that might be presented to CD8+ T cells during immunization of the H-2K b positive C57BL/6 mice, HeLa cell line stably transfected with the mouse MHC class I allele H-2K b (HeLa-K b cells) was used. HeLa-K b cells are easily expandable in cell culture and exhibited high infection efficiency with ORFV in combination with high cell viability (Figure 2A). For infection the V-D12-mCherry virus expressing mCherry under control of the early eP2 promoter was used, which allows early detection of infected cells without the production of viral infectious progeny [11]. Twenty hours after infection (MOI 5), 48.1% of cells expressed mCherry (Figure 2A). Increasing the MOI up to 50 resulted in 87.1% of infected cells (Figure 2A). Notably, high cell viability of 95.3-90.2% was found after 20 h of infection with all tested MOI (Figure 2A).  Figure 2B). Higher MOI (10-50) resulted in the absolute numbers of H-2K b molecules from 4.2-3.6 × 10 5 that were comparable with the non-infected cells ( Figure 2B). These data indicate that infected HeLa-K b cells provide sufficient amounts of H-2K b molecules for the elution of ORFV ligands.
Moreover, we showed that HeLa-K b cells enable efficient presentation of HeLa-K b restricted peptide SIINFEKL after infection with V12-Ova-D12-GFP recombinant. Thus, a complex of H-2K b molecules with SIINFEKL peptide was detected on the HeLa-K b cells by flow cytometry 20 h post infection with MOI 5 (Figure 2C,D). Importantly, nearly all ORFV-infected cells exhibited H-2K b -SIINFEKL complexes on their surface ( Figure 2C). No specific surface staining was observed after exposure to the control virus ( Figure 2C,D). Fluorescence intensity of the H-2K b -SIINFEKL staining of V12-Ova-D12-GFP infected cells was even higher than of the cells, loaded with synthetic SIINFEKL peptide ( Figure 2D). This can be explained by increased numbers of H-2K b molecules on the cell surface after infection with MOI 5 as described above. The recombinant V12-Ova-D12-GFP virus might also express a higher amount of SIINFEKL peptide as compared to the restricted number of loaded peptide molecules.
Altogether, these results indicate the ability of HeLa-K b cells to endogenously process ORFV expressed proteins and efficiently present H-2K b restricted epitopes. Hence, HeLa-K b cells were chosen as target cells for ORFV infection and subsequent MS analysis.

Identification of ORFV-Derived H-2K b -Presented Peptides by LC-MS/MS
HeLa-K b cells were infected for 20 h with ORFV at MOI 5 and the eluted H-2K b -presented peptides were analyzed by LC-MS/MS. This approach resulted in identification of 1460 unique HeLa-K b -presented peptides, originating from both infected and non-infected cells. Binding prediction with NetMHCpan4 assigned 1294 (88.6%) peptides as ligands fitting to the H-2K b peptide motif. In total 43 (3.3%) unique ORFV-derived HeLa-K b -presented peptides were identified. We further analyzed frequencies of individual amino acids in every epitope position for all 8-10-mer peptides considering the MHC anchor amino acid preferences of the H2-K b allele. Finally, 36 (2.78%) unique peptides were selected as ORFV-derived H-2K b ligands (Table 1).

Characteristics of ORFV-Derived H-2K b -Presented Peptides
The 36 ORFV-derived ligands are encoded by 23 (17.2%) out of 134 putative ORFV genes [48,52]. The majority-eight of the detected ligands (22.2%)-originated from the virion core protein, six peptides (16.7%) were derived from the RNA-polymerase subunit and five peptides (13.9%) emerged from the IMV membrane protein. Epitope occurrence from the various ORFV proteins suggests that all these proteins are available to the antigen presentation pathway. While we performed the ligandome analysis at only one time point after infection (20 h), different kinetic classes of viral gene expression were present among the source proteins of the identified epitopes [53,54].
We observed that in some proteins detected H-2K b -restricted ORFV ligands comprised overlapping peptide cores with extensions at either the amino-or carboxy-terminus (for example, ORF088 Virion core protein and IMV membrane protein, Table 1). In other cases, there were sets of non-overlapping peptides from the same protein (RNA-polymerase subunit protein, Table 1). From all identified ORFV-associated peptides 18 (50%) were 8-mers, 17 (47.2%) 9-mers and 1 (2.8%) was a 10-mer.
Overall, we report for the first time the identification of 36 ORFV-derived unique H-2K b -restricted epitopes using immunopeptidomics.

Characteristics of ORFV-Derived H-2K b -Presented Peptides
All ORFV ligands, identified from HeLa-K b cells by LC-MS/MS analysis were further evaluated for their immunogenicity. C57BL/6 mice received two times administration of V12-Ova-D12-GFP or control ORFV recombinant by i.m. route. Memory T cell responses against synthetic peptide pools were analysed in splenocytes one week after the second immunization by IFN-γ ELISPOT assay. Responses were considered positive when the mean count of spot forming units (SFU) per well was at least threefold higher than the mean number SFUs in the negative control wells. We found that the administration of control or V12-Ova-D12-GFP ORFV recombinants induced mean number of 11.25 and 20.5 SFU in the negative control wells, respectively ( Figure 3A). Stimulation with the peptide pools elicited mean responses ranging between 11 and 20.3 IFN-γ spots in the control immunized mice and D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 days mice received a homologous i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D12-mCherry. Mice were sacrificed one week after the second administration, and the splenocytes were used for immunological testing as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFNƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immune splenocytes were tested either separately from individual mice or pooled as described in the "Results" section. In total 2 × 10 5 splenocytes per well were either stimulated with peptide pools (10 µg/mL of each peptide) or with Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide pool contained eight randomly mixed synthetic peptides, corresponding to the identified ORFV-derived ligands (numbers 1-32 as listed in Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland) (10 µg/mL) was used as a positive control, unstimulated cells served as a negative control. Developed spots were automatically counted using an ImmunoSpot S5 analyzer (Cellular Technology Limited, Cleveland, OH, USA) and ImmunoSpot software (Cellular Technology Limited, Cleveland, OH, USA). Responses were considered to be positive when the mean spot count per well was at least threefold higher than the mean number of spots in negative control wells. Overlapping spots in wells with strong IFN-ƴ response hamper the detection of reliable counts, therefore spot counts of >1.000 per well were set to 1.000.

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual mice or pooled as described in the "Results" section; 2 × 10 6 cells per well were seeded into 96-well plate (Corning, NY, USA) and stimulated for 16 h with individual ORFV-derived peptides ( Figure 3A,B). PHA as a positive control induced high lymphocyte responses in both groups of mice, indicating a functional state of the analysed lymphocytes (Figure 3). D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 days i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D12-m one week after the second administration, and the splenocytes were us as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFN-ƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immune sp separately from individual mice or pooled as described in the "Res splenocytes per well were either stimulated with peptide pools (10 µg Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide pool con synthetic peptides, corresponding to the identified ORFV-derived ligan Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland) (10 µ control, unstimulated cells served as a negative control. Developed spot using an ImmunoSpot S5 analyzer (Cellular Technology Limited, ImmunoSpot software (Cellular Technology Limited, Cleveland, O considered to be positive when the mean spot count per well was at le mean number of spots in negative control wells. Overlapping spots response hamper the detection of reliable counts, therefore spot counts 1.000.

MHC Dextramer Staining
Splenocytes of individual mice were stained with H-2K b Ova25 Kopenhagen, Denmark) according to the manufacturer's protocol. Afte was performed with anti-CD90.2-APC (BioLegend, San Diego, CA, U Alexa 700 (BioLegend, San Diego, CA, USA; clone 53-6.7) and anti-Diego, CA, USA; clone 6D5) antibodies for 30 min at 4 °C. Cells were viability staining was performed using Zombie Aqua (BioLegend, San D the manufacturer's instructions. Samples were acquired on a BD F Biosciences, San Jose, CA, USA) and data were analyzed using FlowJo Inc., Ashland, OR, USA).

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual mice o "Results" section; 2 × 10 6 cells per well were seeded into 96-well pl

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse ƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). separately from individual mice or pooled as described splenocytes per well were either stimulated with peptide Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each pept synthetic peptides, corresponding to the identified ORFV- Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Swit control, unstimulated cells served as a negative control. De using an ImmunoSpot S5 analyzer (Cellular Technolo ImmunoSpot software (Cellular Technology Limited, considered to be positive when the mean spot count per w mean number of spots in negative control wells. Overl response hamper the detection of reliable counts, therefor 1.000.

MHC Dextramer Staining
Splenocytes of individual mice were stained with Kopenhagen, Denmark) according to the manufacturer's p was performed with anti-CD90.2-APC (BioLegend, San Alexa 700 (BioLegend, San Diego, CA, USA; clone 53-6 Diego, CA, USA; clone 6D5) antibodies for 30 min at 4 °C viability staining was performed using Zombie Aqua (Bio the manufacturer's instructions. Samples were acquired Biosciences, San Jose, CA, USA) and data were analyzed u Inc., Ashland, OR, USA).

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from indiv "Results" section; 2 × 10 6 cells per well were seeded int

IFN-γ ELISPOT Assay
The ELISPOT assay was perform ƴ ELISPOT kit (Mabtech, Nacka separately from individual mice or p splenocytes per well were either stim Ova257-264 SIINFEKL peptide (1 µg/mL synthetic peptides, corresponding to t Table 1). Phytohemagglutinin (PHA-L control, unstimulated cells served as a using an ImmunoSpot S5 analyzer ImmunoSpot software (Cellular Tec considered to be positive when the m mean number of spots in negative c response hamper the detection of relia 1.000. These results demonstrate the ability of the ORFV strain D1701-V to elicit high magnitude transgene-specific cellular responses in mice after immunization and suggest a lack of T cell immunity against the ORFV vector.

ICS Validates Absence of CD8+ T Cell Responses to the Individual ORFV-Derived Peptides
To exclude the possibility of competitive effects among the different ORFV epitopes within one peptide pool we additionally performed ICS after 16 h ex vivo stimulation of pooled splenocytes from V12-Ova-D12-GFP and control ORFV immunized mice with the individual peptides. We also speculated that ORFV-epitope specific CTLs might produce other cytokines than IFN-Vaccines 2020, 8, x FOR PEER REVIEW D12-GFP or control V-D12-mCher i.m. boost immunization with 10 7 P one week after the second admini as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was perf D12-GFP or control V-D12-mCherry ORFV recombinant. Afte i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or one week after the second administration, and the splenocytes as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFN ƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Imm separately from individual mice or pooled as described in splenocytes per well were either stimulated with peptide poo Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide synthetic peptides, corresponding to the identified ORFV-deri Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerl control, unstimulated cells served as a negative control. Develo using an ImmunoSpot S5 analyzer (Cellular Technology ImmunoSpot software (Cellular Technology Limited, Clev considered to be positive when the mean spot count per well mean number of spots in negative control wells. Overlappi

IFN-γ ELISPOT Assay
The ELISPOT assay was performed ƴ ELISPOT kit (Mabtech, Nacka Str separately from individual mice or poo splenocytes per well were either stimula Ova257-264 SIINFEKL peptide (1 µg/mL) fo synthetic peptides, corresponding to the Table 1). Phytohemagglutinin (PHA-L) (R control, unstimulated cells served as a ne , TNF-α and IL-2 responses to all tested peptides independently on predicted binding level. Mice responded to peptides 19 and 25, derived from ORFV015 and ORFV056 RNA-polymerase subunit RPO147 respectively, with threefold higher numbers of CD107a expressing CD8+ T cells ( Figure 4C,D). However, both detected responses were not statistically significant compared to the unstimulated control. Moreover, CD107a-producing CD8+ T cells were monofunctional ( Figure 4C,D). Thus, according to selection criteria the observed responses were considered as negative. Moreover, no positive responses were found by the use of ORFV for the stimulation of splenocytes ( Figure 4). As observed previously, strong multifunctional responses were detected against Ova 257-264 SIINFEKL peptide and PHA stimulation (Figure 4), indicating a functional state of the analysed lymphocytes. Coactivation of CD4+ T cells was not observed after cell stimulation with any of the tested ORFV ligands. We further note that the employed immunogenicity assays are subject to a threshold of detection. It can be expected that memory CD8+ T cells specific to identified H-2K b ORFV-derived peptides may represent a very small fraction of the splenic lymphocyte population ex vivo and can be below detection limits. To confirm the absence of memory CD8+ T cells to ORFV-derived peptides in general and of peptide 19 and 25 in particular, we re-stimulated pooled splenocytes from the immunized mice during 14 days with the individual peptides as described in Materials and Methods. Whenever specific cells are present among the splenocytes, peptide re-stimulation would lead to their expansion excluding priming of naïve T cells. If no memory CD8+ T-cells have been re-activated and expanded, cytokine expression would be comparable to the unstimulated control. We performed this assay for all identified ORFV-derived peptides to exclude the possibility that low-frequency memory CD8+ T cells might have been below detection limits of ex vivo ICS (Figure 4). Again, the CTLs did not positively respond in the presence of any of the used peptides. Percentages of IFN-Vaccines 2020, 8, x FOR PEER REVIEW D12-GFP or control V-D12-mCherry ORFV recombinant. After 10 i.m. boost immunization with 10 7 PFU of V12-Ova-D12-GFP or V-D one week after the second administration, and the splenocytes w as described below.

IFN-γ ELISPOT Assay
The ELISPOT assay was performed using the mouse IFNƴ ELISPOT kit (Mabtech, Nacka Strand, Schweden). Immun separately from individual mice or pooled as described in the splenocytes per well were either stimulated with peptide pools ( Ova257-264 SIINFEKL peptide (1 µg/mL) for 21 h. Each peptide poo synthetic peptides, corresponding to the identified ORFV-derived Table 1). Phytohemagglutinin (PHA-L) (Roche, Basel, Switzerland control, unstimulated cells served as a negative control. Developed using an ImmunoSpot S5 analyzer (Cellular Technology Lim ImmunoSpot software (Cellular Technology Limited, Clevela considered to be positive when the mean spot count per well wa mean number of spots in negative control wells. Overlapping response hamper the detection of reliable counts, therefore spot c 1.000.

MHC Dextramer Staining
Splenocytes of individual mice were stained with H-2K b O Kopenhagen, Denmark) according to the manufacturer's protoco was performed with anti-CD90.2-APC (BioLegend, San Diego, Alexa 700 (BioLegend, San Diego, CA, USA; clone 53-6.7) and Diego, CA, USA; clone 6D5) antibodies for 30 min at 4 °C. Cells viability staining was performed using Zombie Aqua (BioLegend, the manufacturer's instructions. Samples were acquired on a Biosciences, San Jose, CA, USA) and data were analyzed using Flo Inc., Ashland, OR, USA).

Intracellular Cytokine Staining (ICS)
Splenocytes were tested either separately from individual m "Results" section; 2 × 10 6 cells per well were seeded into 96-w stimulated for 16 h with individual ORFV-derived peptides (10 SIINFEKL peptide,or  , TNF-α, IL-2 or CD107a producing CD8+ T-lymphocytes were close to the unstimulated control cells ( Figure S1) However, different cell culture conditions as well as the protocols applied might explain variation in the percentages of cytokine expressing cells in ex vivo ICS and after 14 days re-stimulation.
Altogether, we confirmed absence of recognition of the tested ORFV-derived H-2K b -presented peptides identified in this study and induction of strong multifunctional CD8+ T cell response against inserted antigen.

Discussion
Virus-derived vectors, facilitating the induction of strong transgene-specific immune responses, offer advantages over traditional vaccine technologies. However, anti-vector immunity represents a key limitation of current recombinant viral delivery platforms [55,56]. The recognition of virus-derived peptides, presented by MHC class I complexes on the surface of infected cells to CD8+ T lymphocytes, plays an important role in mediating antiviral immune responses against the viral vector [57]. A comprehensive view of the specificity of the anti-vector responses requires evaluation of both the epitopes presented and the corresponding CD8+ T cells. In our approach we employed LC-MS/MS analysis for ORFV for the first time to profile a viral immunopeptidome on mice where detailed evaluation of immunogenicity is possible.
The lack of sufficient ORFV uptake and vigorous cell death after virus exposure did not allow the use of murine primary bone marrow-derived macrophages and dendritic cells as well as the murine macrophage cell lines J774 and RAW264.7 (unpublished data), although they would provide a larger set of ORFV-derived peptides other than K b ligands. Therefore, we used a transgenic HeLa cell line encoding the murine H-2K b gene as a model of MHC class I epitope processing and presentation of ORFV infected cells. The combination of very good ORFV infection efficiency, unaffected cell viability and high H-2K b expression levels with efficient model peptide presentation validated HeLa-K b cells as suitable targets for MS analysis. Detection of the peptide presentation other than SIINFEKL would be of interest, however no antibodies are available to reveal ORFV-derived epitopes in the context of H-2K b .
In total, we were able to identify 36 naturally presented ORFV-derived H-2K b peptides from HeLa-K b cells. All discovered peptides are reported for ORFV for the first time. We were focusing on the early ORFV gene expression until 20 h post infection based on the observation that major VACV immunogenic epitopes are originating from early expression gene products and late viral proteins were recognized by fewer CD8+ T cells [43,58]. Hence, we supposed that the majority of immunogenic peptides would be derived from the ORFV proteins expressed early after infection. Interestingly, we detected peptides from all kinetic classes of ORFV proteins. However, we cannot discriminate whether early presentation of ORFV-derived epitopes resulted directly from entering virion proteins or from newly synthesized gene products as reported for MVA [40].
Immunogenicity analysis of the identified ORFV-specific epitopes revealed that none of the tested H-2K b peptides is immunogenic in mice. Lack of specific memory CD8+ T lymphocyte responses to ORFV is in accordance with the previous observation that CD8+ T cells are not essential for the virus clearance after reinfection [21]. While the cell-depletion study proposed a role of CD4+ T lymphocytes for the control of viral replication [21], duration of anti-ORFV immunity is short-lived [24]. Hence, our data prove that cellular immunity against ORFV strain D1701-V is negligible. It is important to mention that the epitope discovery is limited to the sensitivity of the LC-MS/MS method, which may lead to an inefficient detection of low-abundance MHC class I presented ORFV peptides. However, major CD8+ T cell responses were shown to be associated with the high affinity for MHC class I complexes, even among the peptides of sufficient affinity to be presented [43]. Other reasons could partly explain the absence of memory CD8+ T cell responses against identified ligands in ORFV-immunized mice including virus dose and route of administration [59,60], antigen presentation pattern in infected HeLa-K b cells and possible requirements for posttranslational modifications of ORFV epitopes [43,61]. Despite the nonimmunogenic H-2K b ligands identified, all these peptides are naturally presented on MHC molecules, providing evidence for epitope screening. Further infection of other cell lines expressing distinct MHC haplotypes will enable broader insight into the repertoire CD8+ T cell epitopes of ORFV. Considering the importance of CD4+ T cells in the control of viral replication, characterization of MHC class II ORFV ligands would be of interest.
In contrast to the lack of CTL immunity against ORFV-derived peptides, strong multifunctional transgene-specific CD8+ T cell response was elicited by the ORFV after two intramuscular immunizations. Anti-ORFV antibodies are known to play no or very little role in protection against infection as the passive antibody transfer does not confer protection against a virus challenge [4,21]. Moreover, no correlation between serum antibody titers and severity of viral lesions has been described for ORFV [4,21]. Lack of neutralizing antibodies could be linked to the action of the virus immunomodulatory proteins interfering with the host immune response to avoid virus elimination [21,62,63]. Another reason could be the fact that neutralizing targets are complex and comprised of multisubunit structures, as it has been described for VACV [64]. However, the absence of neutralizing antibodies and negligible ORFV-specific cellular immunity enable multiple vaccinations and make ORFV an attractive vaccine platform for infectious diseases and cancer. In addition, the lack of the memory CD8+ T lymphocytes against ORFV epitopes facilitates efficient priming and further enhancement of the antigen-specific responses after repeated immunizations or heterologous prime-boost regimens. Consequently, the ORFV vector strain D1701-V represents a potent carrier for the development of efficient vaccines.

Conclusions
In this study we investigated the induction of transgene-and vector-specific CD8+ T lymphocyte responses by the ORFV strain D1701-V in mice. We applied a MHC ligandome approach to generate a comprehensive list of ORFV-specific epitopes that might be presented to CD8+ T cells during immunization. For the first time we report the identification of 36 unique D1701-V ORFV-derived H-2K b peptides. Detailed immunogenicity analysis proved that none of the evaluated virus epitopes is immunogenic in mice, but a robust CD8+ T cell immunity is induced against the inserted foreign antigen. These findings reinforce that ORFV strain D1701-V represents an effective and safe recombinant vector for therapeutic and prophylactic vaccines. Additionally, the experiments describe prerequisites for the selection of T cell epitopes exploitable for the generation of ORFV-based vaccines by reverse genetics.