Expression of PD-L1 on Canine Tumor Cells and Enhancement of IFN-γ Production from Tumor-Infiltrating Cells by PD-L1 Blockade

Programmed death 1 (PD-1), an immunoinhibitory receptor, and programmed death ligand 1 (PD-L1), its ligand, together induce the “exhausted” status in antigen-specific lymphocytes and are thus involved in the immune evasion of tumor cells. In this study, canine PD-1 and PD-L1 were molecularly characterized, and their potential as therapeutic targets for canine tumors was discussed. The canine PD-1 and PD-L1 genes were conserved among canine breeds. Based on the sequence information obtained, the recombinant canine PD-1 and PD-L1 proteins were constructed; they were confirmed to bind each other. Antibovine PD-L1 monoclonal antibody effectively blocked the binding of recombinant PD-1 with PD-L1–expressing cells in a dose-dependent manner. Canine melanoma, mastocytoma, renal cell carcinoma, and other types of tumors examined expressed PD-L1, whereas some did not. Interestingly, anti-PD-L1 antibody treatment enhanced IFN-γ production from tumor-infiltrating cells. These results showed that the canine PD-1/PD-L1 pathway is also associated with T-cell exhaustion in canine tumors and that its blockade with antibody could be a new therapeutic strategy for canine tumors. Further investigations are needed to confirm the ability of anti-PD-L1 antibody to reactivate canine antitumor immunity in vivo, and its therapeutic potential has to be further discussed.


Introduction
Recently, cancer has been a major cause of death in dogs and has surpassed infectious diseases. According to a recent report, 27% of overall dog deaths is due to cancer [1]. Current clinical approaches for canine cancer are surgical, chemotherapeutic, and radiation therapies, as in humans. In some cases of dog cancers, however, it is difficult to treat them by only using existing therapeutic methods because of the severe stress, adverse effect, and/or difficulties in approaching the tumor sites. In addition, the sensitivities to the chemo/radiotherapies can differ dependent on the tumor types. Therefore, it is worth investigating the efficacy of novel approaches against canine cancer, including immunotherapy, as this may lead to the development of more effective therapies that can induce tumor remission.
Programmed death 1 (PD-1), an immunoinhibitory receptor, and programmed death ligand 1 (PD-L1), its ligand, together can induce the ''exhausted'' status in antigen-specific lymphocytes and are thus involved in the immune evasion of tumor cells [2,3]. In human cancer, tumor cells express PD-L1, and the cells cause Tcell dysfunction in breast cancer [4], pancreatic cancer [5,6], and bladder cancer [7]. Tumor-infiltrating lymphocytes, which are specific to tumor antigens, express PD-1 in melanoma [8], lung cancer [9], and intrahepatic bile duct cancer [10]. Furthermore, in cases of renal cell carcinoma [11] and gastric carcinoma [12], patients with tumor PD-L1 are at a significantly increased risk of rapid disease progression and overall mortality, suggesting that PD-L1 is associated with poor prognosis in patients with tumors. When PD-1 binds to PD-L1, negative signals are transmitted into the lymphocytes, resulting in the suppression of antigen-specific immune responses [13][14][15]. Most importantly, this suppressive status is reported to be reversible, and the blockade of the PD-1/ PD-L1 pathway using anti-PD-L1 antibodies or other molecules can restore the function of exhausted lymphocytes [16,17]. It is also reported that the blockade of the PD-1/PD-L1 pathway results in reactivation of antitumor immunity and in subsequent regression of some tumors in human clinical trials [18,19]. Therefore, this therapeutic strategy may be promising for effective treatment of tumors. In the veterinary field, however, reports on the PD-1/PD-L1 pathway are few and its relevance to diseases is almost unknown.
Recent studies in our laboratory have revealed that immunoinhibitory molecules including PD-1/PD-L1 play critical roles in immune exhaustion and disease progression in case of bovine leukemia virus (BLV) infection [20][21][22][23][24][25][26]. We established the specific antibodies; blocking using the antibodies in the inhibitory pathway in vitro increased cytokine responses and enhanced immune cell function, leading to decrease in the viral load [20,21]. Therefore, evaluation of inhibitory receptor expression kinetics is essential to improve the development of an effective immunotherapy that can induce antitumor responses in dogs. In this study, canine PD-1 and PD-L1 were molecularly characterized. Then, PD-L1 expression on canine tumors and the potential of the PD-1/PD-L1 pathway as a therapeutic target for treatment of dog tumors were assessed in the lab.

Canine Samples
Animal use throughout this study was approved by the Institutional Animal Care and Use Committee (the serial number of approval was #1039), Graduate School of Veterinary Medicine, Hokkaido University, which has been fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. Peripheral blood samples were obtained from healthy 5-

Identification of Canine PD-1 and PD-L1 Genes
Total RNA was isolated from the Beagle and the Samoyed PBMCs stimulated with ConA for 4 h, white blood cells of the Labrador retriever, testis tissue of the Japanese Akita, and lung tissue of the Bernese mountain dog, using TRIzol reagent (Life Technologies) according to the manufacturer's instructions. Residual genomic DNA was removed from the total RNA by DNase (Life Technologies) treatment. cDNA was synthesized from 1 mg of the total RNA using Moloney murine leukemia virus reverse transcriptase (Takara, Shiga, Japan) and oligo-dT primer, as recommended by the manufacturer. To amplify the inner sequences of canine PD-1 or PD-L1, canine PD-1-and PD-L1specific primers were designed based on the putative canine PD-1 and PD-L1 mRNA sequence reported in the GenBank database (XM_543338 and XM_541302). Canine PD-1 and PD-L1 cDNA were amplified from Beagle cDNA by PCR using primers 59-AGG ATG GCT CCT AGA CTC CC-39 (PD-1 inner forward), 59-AGA CGA TGG TGG CAT ACT CG-39 (PD-1 inner reverse), 59-ATG AGA ATG TTT AGT GTC TT-39 (PD-L1 inner forward), and 59-TTA TGT CTC TTC AAA TTG TAT ATC-39 (PD-L1 inner reverse). The PCR cycling conditions were as follows: (1) initial denaturation at 94uC for 5 min, (2) 40 cycles of denaturation at 94uC for 1 min, annealing at 58uC (PD-1) or 50uC (PD-L1) for 1 min and extension at 72uC for 1 min 30 s, and (3) final extension at 72uC for 7 min. PCR amplicons were purified using the FastGene gel/PCR extraction kit (Nippon Genetics, Tokyo, Japan), cloned into the pGEM-T Easy vector (Promega, Madison, WI, USA), and sequenced using the CEQ8000 DNA analysis system (Beckman Coulter, Fullerton, CA, USA). 59-RACE and 39-RACE were then performed using the 59-RACE system for rapid amplification of cDNA ends and 39-RACE system for rapid amplification of cDNA ends (Life Technologies), respectively. The gene-specific primers for the canine PD-1/PD-L1 used for 59-RACE were 59-GTT GAT CTG TGT GTT G-39 (PD-1) and 59-TTT TAG ACA GAA AGT GA-39 (PD-L1). The gene-specific primers for canine PD-1/PD-L1 used for 39-RACE were 59-CGG GAC TTC CAC ATG AGC AT-39 (PD-1) and 59-GAC CAG CTC TTC TTG GGG AA-39 (PD-L1). Based on the sequences obtained, new primer sets were designed to amplify the full length of the canine PD-1 and PD-L1 cDNA. PCR was performed using primers 59-GGG GGA GGC GAG CAG G-39 (PD-1 ORF F), 59-GAG TCG AGA GAG GAG AGC CAT GAG-39 (PD-1 ORF R), 59-GCC AGC AGG TCA CTT CAG AAC-39 (PD-L1 ORF F), and 59-GCT GAA CTC AAA CCA CAG GCC-39 (PD-L1 ORF R) as described above, except that the annealing temperature used was 60uC. The resulting amplicons were sequenced as described.
To confirm the polymorphisms of the PD-1 and PD-L1 genes among canine breeds, the blood samples of the Beagle, Samoyed and Labrador retriever, testis tissue of the Japanese Akita, and lung tissue of the Bernese mountain dog were collected and studied using ORF primer and each the cDNAs of each sample. The established sequences were aligned, and unrooted neighborjoining trees were constructed using the Mega version 5 software [34,35].

Preparations of Canine PD-1-and PD-L1-Expressing Cells
To construct the enhanced green fluorescent protein (EGFP) fusion expression vectors, the ORF region of the canine PD-1 and canine PD-L1 cDNA that did not have the stop codons was amplified by PCR using the cDNA obtained from the Beagle PBMCs and gene-specific primers [59-CCG CTC GAG ATG    [36,37]. These vectors were named pCXN2.1-Rabbit IgG Fc-cPD-1 or pCXN-2.1-rabbit IgG Fc-cPD-L1, respectively. For stable expression, 4610 6 CHO-DG44 cells were transfected with 2.5 mg pCXN-2.1-rabbit IgG Fc-cPD-1 or pCXN-2.1-rabbit IgG Fc-cPD-L1 using Lipofectamine LTX (Life Technologies), as recommended by the manufacturer. Forty-eight hours later, the cells were collected and resuspended in the supplemented CD-DG44 medium containing 800 mg/mL G418 (Enzo Life Science, Farmingdale, NY, USA). Stably expressing cells were cloned using the limiting concentration technique, and high expression cell lines were obtained. The cell culture supernatant of these cell lines was collected after 7 days of the shaking culture. The supernatant containing Ig fusion proteins was concentrated by ultrafiltration using Centricon Plus-70 (Merck Millipore, MA, USA) and the Ig fusion proteins were purified by Ab-Capcher Extra (Protenova, Kagawa, Japan). Buffers were exchanged with phosphate buffered saline (PBS) using PD MiniTrap G-25 (GE Healthcare UK). The concentrations of the Ig fusion proteins were evaluated using the rabbit IgG ELISA quantitation set (Bethyl Laboratories, Montgomery, TX, USA). To confirm the expression of these Ig fusion proteins, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analysis with the Immobilon-P transfer membrane (Merck Millipore) were performed as described elsewhere [21]. The membrane was incubated with Immobilon Western chemiluminescent HRP substrate (Merck Millipore) to visualize the signals and analyzed by a Fluor-S MultiImager (Bio-Rad Laboratories, CA, USA). To detect the canine PD-L1 expressed on the cell surfaces, flow cytometry was performed using rat antibovine PD-L1 monoclonal antibodies [6G7-E1; rat IgM (k), 5A2-A1; rat IgG1 (k), 4G12-C1; rat IgG2a (k)], which were established in our laboratory [20]. In brief, 1610 6 cells were incubated with 10 mg/mL antibovine PD-L1 antibodies at room temperature for 30 min and washed twice, followed by incubation with allophycocyanin-conjugated goat antirat Ig antibody (Beckman Coulter) at room temperature for 30 min. The cells were washed twice and analyzed. Rat IgM (k) isotype control (BD Biosciences), rat IgG1 (k) isotype control (BD Biosciences), and rat IgG2a (k) isotype control (BD Biosciences) were used as isotype-matched negative control antibodies. PBS containing 10% goat serum (Sigma-Aldrich) was used in all washing processes and dilutions of Ig fusion proteins or antibodies.

Analysis of PD-L1 Expression on Single Cells from Canine Tumors
To obtain single cells from solid tumors, freshly excised solid tumor tissues were cut into small pieces and treated with 2 mg/mL collagenase D (Roche Applied Science, Indianapolis, IN, USA) in RPMI 1640 medium supplemented with 10% FCS, 2 mM Lglutamine, 200 mg/mL streptomycin, and 200 U/mL penicillin at 37uC for up to 2 h. The single cell suspension was washed twice with PBS and cultured in supplemented RPMI 1640 medium at 37uC and 5% CO 2 for 24 h; this restored the cell surface PD-L1 expression, which had been degraded by the collagenase. The cells were harvested and the PD-L1 expression was analyzed by flow cytometry, as described above, using the antibovine PD-L1 monoclonal antibody 4G12-C1. In flow cytometric analysis, the cells were plotted using forward-scattered light (FSC) and sidescattered light (SSC), and the cell population with a higher FSC than lymphocytes was gated as the tumor cell population if there was no other population except for the population of tumorinfiltrating lymphocytes. To exclude any dead cells, the cells were stained with 7-amino-actinomycin D (BD Biosciences), and the live cell population was analyzed for PD-L1 expression. The canine PBMCs were used as a negative control under the same conditions.

Immunohistochemical Analysis of PD-L1 on Tumor Cells
Formalin-fixed and paraffin-embedded tumor tissues were cut into 4-mm-thick sections and dried on silane-coated slides. The dried sections were deparaffinized in xylene. Antigen retrieval was performed in citrate buffer (0.37 g/mL of citric acid and 2.4 g/ mL of trisodium citrate dihydrate) by microwave heating for 10 min. Endogenous peroxidase activity was blocked by incubating the sections in methanol containing 0.3% hydrogen peroxide for 15 min. Primary antibody incubation was performed at room temperature for 30 min using antibovine PD-L1 monoclonal antibody 5A2-A1 (1.2 mg/mL) or rat IgG1 isotype-matched control antibody (Biolegend, San Diego, CA, USA). The sections that were washed twice with PBS were incubated with Histofine simple stain MAX PO (rat) (Nichirei, Tokyo, Japan) at room temperature for 30 min, and positive staining was visualized with 3-diaminobenzidine tetrahydrochloride (DAB). The sections were observed under an optical microscope.   Blocking Assay Using Anti-PD-L1 Antibody Single cell suspensions from fresh solid tumor tissues were obtained by collagenase treatment, as described, or by a mechanical method using the 100 mm cell strainer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). The suspensions were underlaid onto 1.055 g/mL Percoll separation solution. The tumor cells were separated by density gradient centrifugation at 8006g for 20 min, and the cell pellets were collected and resuspended in the supplemented RPMI 1640 medium. Then, the cell suspensions were again underlaid onto 1.077 g/mL Percoll separation solution and centrifuged. The tumor-infiltrating lymphocytes in the gradient interfaces were collected and washed with PBS three times. These lymphocytes were resuspended in the supplemented (10% FCS, 2 mM Lglutamine, 200 mg/mL streptomycin, and 200 U/mL penicillin) RPMI 1640 medium.
The tumor-infiltrating lymphocytes and PBMCs obtained from the healthy adult Beagles (2610 6 /mL) were cultured with 20 mg/ mL of antibovine PD-L1 monoclonal antibody 6G7-E1 at 37uC and 5% CO 2 for 48 h, and the culture supernatant was collected. As a negative control, low-endotoxin, Azide-free rat IgM isotype control antibody (Acris Antibodies, Herford, Germany) was used. The concentration of dog IFN-c in the culture supernatant was evaluated by Duoset ELISA canine IFN-c (R&D systems, Minneapolis, MN, USA) according to the manufacturer's protocol. Absorbance was measured using a microplate reader MTP-650FA (Corona Electric, Ibaraki, Japan).

Statistics
In the blocking assay with anti-PD-L1 antibody, Tukey's test was conducted among 0, 0.5, 1.0, 2.5, and 5.0 mg/mL antibody treatment groups. The Wilcoxon signed rank-sum test was conducted to compare the data from the same individuals. For both tests, p,0.05 was considered statistically significant.   Table 6. N.D., Not Detected. doi:10.1371/journal.pone.0098415.g007

Nucleotide Sequence Accession Numbers
The sequences of the canine PD-1 and PD-L1 genes have been submitted to the GenBank database under accession number AB898677 (PD-1) and AB898678 (PD-L1).

Sequence Analysis of Canine PD-1 and PD-L1
The complete nucleotide sequences and putative amino acid sequences of canine PD-1 and PD-L1 obtained from the Beagle dog are shown in Fig. 1A and Fig. 2A, respectively. The complete nucleotide sequences for canine PD-1 and PD-L1 were found to be 1,781 bp and 1,561 bp in length, and the sequences had an ORF encoding for 288 and 289 amino acids, respectively. Canine PD-1 and PD-L1 consist of a putative signal peptide, an extracellular domain, a transmembrane domain, and an intracellular domain as in other species (Fig. 1B, 1D and Fig. 2B). The cytoplasmic tail of PD-1 contains two structural motifs, an immunoreceptor tyrosinebased inhibitory motif (ITIM: I/L/V/S/TxYxxL/V/I), and an immunoreceptor tyrosine-based switch motif (ITSM) [38]. ITIM and ITSM were conserved in canine PD-1 (Fig. 1B, 1D). Both the phylogenetic analyses revealed that mammalian PD-1 and PD-L1 were divided into two groups, group 1 (Perissodactyla, Artiodactyla, and Carnivora) and group 2 (Primate and Rodentia), with canine PD-1 and PD-L1 clustering in the Carnivora species group (Fig. 1C and Fig. 2C). Comparative analysis of the PD-1 sequences of several species showed that canine PD-1 had 87.8%, 77.1%, 75.7%, and 68.4% amino acid similarity with cat, cow, human, and mouse, respectively (Table 1). On the other hand, PD-L1 sequences of several species showed that canine PD-L1 had 87.9%, 86.2%, and 82.4% amino acid similarity with cow, human, and mouse, respectively (Table 2). To compare the genetic diversity of the canine PD-1 and PD-L1 among other canine breeds, the samples from the Samoyed, Labrador retriever, Japanese Akita, and Bernese mountain dog were collected and compared to that of the Beagle dog. Although some sequences were not obtained because the samples did not express the genes in the tissue, their genetic sequences showed 100% nucleotide identity with the sequence from the Beagle dog ( Fig. 1B and  Fig. 2B).
Canine PD-1-Ig Binds to PD-L1, and Its Binding is Disturbed by Bovine PD-L1 Antibody First, to confirm the binding of canine PD-L1 to canine PD-1, we established an in vitro model by transfecting Cos7 cells with canine PD-1 and PD-L1 and generated transient transfectants. Expression of canine PD-1 and PD-L1 was detected on cPD-1-EGFP and cPD-L1-EGFP transfectants (Fig. 3A), respectively. On the other hand, canine PD-1-Ig and PD-L1-Ig were expressed in vitro using the pCXN2.1 expression vector and CHO-DG44 cell expression system. Affinity-purified canine PD-1-Ig and PD-L1-Ig migrated as molecular weights of around 65 kDa and 85 kDa protein on a 12% polyacrylamide gel, respectively (Fig. 3B). Subsequently, the binding of canine PD-L1 and PD-1 was alternately confirmed by flow cytometry analysis. As expected, the canine PD-1-Ig was bound to canine PD-L1 on cells, while the canine PD-L1-Ig was bound to canine PD-1 on cells (Fig. 3C).

PD-L1 Expression on Canine Tumors
First, we investigated the expression of PD-L1 on canine tumor cell lines. PD-L1 was detected on two cell lines (CM-MC, CoMS) from canine mastocytoma, and the expression was enhanced by IFN-c treatment (Fig. 5A, Table 3). Interestingly, IFN-c induced PD-L1 expression on all tested melanoma cell lines (CMeC, LMeC, CMM-1, and CMM-2) but not on osteosarcoma cell lines (POS, HM-POS) (Fig. 5A, Table 3). Subsequently, using clinical materials from dogs with tumors, we investigated the expression of PD-L1 by flow cytometry and identified PD-L1 expression on cells from angiosarcoma, hepatocellular carcinoma, squamous carcinoma, and breast adenocarcinoma; PD-L1 was not detectable on the lymphocytes from control dogs (Fig. 5B, Table 4). Finally, PD-L1 expression was histologically analyzed. PD-L1 was expressed in most melanoma (69.2%), mastocytoma (66.7%), and renal cell carcinoma (70.0%) cases (Fig. 6, Table 5). Although the number of tested cases was limited, PD-L1 was expressed in all oral melanoma cases (Table 5).

PD-1/PD-L1 Blockade by PD-L1 Antibody Enhances IFN-c Production
To investigate the effect of PD-1/PD-L1 blockade by PD-L1 antibody in cytokine production, PBMCs from healthy dogs were cultivated in the presence of PD-L1 antibody or isotype control antibody. As shown in Fig. 7A, addition of PD-L1 antibody at the beginning of the 48 h cultivation period significantly enhanced the IFN-c production compared with those treated with the control antibody (p,0.05). Finally, to confirm that PD-L1 antibody augments the production of cytokines from tumor-infiltrating cells, PD-L1 antibody was added to the infiltrating cell cultures from hepatocellular carcinoma, myelolipoma and seminoma at the beginning (Table 6). Interestingly, IFN-c production was enhanced in the infiltrated cells from hepatocellular carcinoma and myelolipoma (Fig. 7B).

Discussion
Several immunotherapies against tumors have been recently developed in humans. Among these therapies, an immunotherapy targeting the PD-1/PD-Ls pathway would be considered as one of the most encouraging approaches because, in a Phase I clinical trial, treatment with PD-L1 antibody induced durable tumor regression and prolonged stabilization of disease in patients with advanced malignant cancers, including non-small-cell lung cancer, melanoma, and renal cell cancer [18]. In addition to this, clinical trials targeting the PD-1/PD-L pathway by using PD-1 antibody are also ongoing for cancer therapy [19,39]. Furthermore, several clinical trials have already been planned or are in progress, combining antibodies targeting the PD-1/PD-L pathway with cancer vaccines, antitumor antibodies, or chemotherapies in humans. However, there has been no report demonstrating that the blockade targeting the PD-1/PD-L pathway might show potential for development of new therapies against canine tumors. In the present study, to evaluate this possibility, we investigated the expression of PD-L1 in several canine tumors and found high levels of PD-L1 expression on cells from dogs with tumors. We also found that PD-1/PD-L1 blockade by PD-L1 antibody enhances IFN-c production from tumor-infiltrating cells from dogs with tumors. These observations raise the possibility that the PD-1/PD-L1 pathway could be a therapeutic target for the treatment of dog tumors.
PD-L1 was detected on canine cell lines from mastocytoma and on fresh cells from angiosarcoma, hepatocellular carcinoma, squamous carcinoma and breast adenocarcinoma. Furthermore, the increasing expression of canine PD-L1 on tumor cells was confirmed in clinical cases of melanoma, mastocytoma, and renal cell carcinoma by immunohistochemical analysis. These findings corresponded to those of human cancers. However, the expression of PD-L1 was not observed in some cases, such as osteosarcomaderived cell lines and tumor tissues from clinical cases, melanoma, mastocytoma, and renal cell carcinoma. PD-L1 expression is induced by cytokines such as IFNs type I and type II [40][41][42]. Indeed, the expression of PD-L1 on cell lines from canine tumors was induced and enhanced by IFN-c treatment. According to previous studies, PD-L1 expression is associated with cytokine production within the tumor microenvironment [3]. Some cases without PD-L1 expression might be also influenced by the tumor microenvironment without cytokines. Further detailed analyses needs to be conducted to clarify the reasons for the lack of PD-L1 expression on the tumors.
In the present study, we detected significant PD-L1 expression on tumor tissues from dogs with mastocytoma and renal cell carcinoma, which originates from mast cells and renal tubular epithelial cells. Dogs have a unique risk of development of cutaneous mastocytoma, which accounts for up to 21% of all skin tumors; however, mastocytoma is rare in human and other species [43]. It is known that, in humans, PD-L1 is expressed on mast cells and it can negatively regulate several immune responses [44]. Furthermore, renal tubular epithelial cells can express PD-L1, which is involved in inhibition of proliferation and cytokine synthesis [45]. The high expression of PD-L1 on tumor cells from canine mastocytoma and renal cell carcinoma might correspond to its PD-L1-expressing original cells. Interestingly, de Haij reported that interaction of tubular epithelial cells and kidney-infiltrating Tcells via PD-L1 changed the balance of positive and negative signals to the T-cells, leading to IL-10 production and the limitation of local immune responses. These observations suggest that PD-L1-expressing cells are associated with T-cell dysfunction in the tumor microenvironment and can result in tumor formation. It will be important to clarify whether the PD-L1 expression level is related with an increased risk of disease progression. At least, canine mastocytoma and renal cell carcinoma might be candidate target cancers for immunotherapy using the PD-L1 antibody.
Melanoma is a common tumor in dogs; it is a locally invasive and frequently malignant type of cancer that can affect dogs [46,47]. Different forms of melanoma are classified by location: skin (cutaneous melanoma), eyelids (ocular melanoma), nail bed (subungual melanoma), and oral cavity (oral melanoma). Among the various forms, canine oral melanoma is a more highly aggressive and fatal tumor [47,48]. In addition, canine oral melanoma is frequently resistant to chemotherapy [49][50][51] and not affected by radiotherapy [52]. Thus, researchers hope for the development of novel therapy against melanoma. In this study, PD-L1 was detected in all cases of canine oral melanoma. The findings from the present study could lead to the design of novel therapeutics against canine oral melanoma, although it remains to be determined whether PD-L1 is expressed on cells from other types of melanomas.
As a preliminary test of the hypothesis that PD-L1 may be a therapeutic target for the treatment of canine tumors, we investigated the effects of blockade of PD-1/PD-L1 by PD-L1 antibody in tumor-infiltrating T-cells from dogs with seminoma, hepatocellular carcinoma and myelolipoma. Similar to previous findings in human or mice models, inhibition of the PD-1/PD-L1 pathway had upregulated the production of IFN-c from lymphocytes from hepatocellular carcinoma and myelolipoma. However, although the effects of PD-L1 antibody on IFN-c production seem to be profound, it still remains speculative whether the blockade of PD-1/PD-L1 by PD-L1 antibody will be sufficient for tumor regression in vivo. Furthermore, the level of PD-1 expression on the tumor-infiltrating T-cells is still unknown because of the lack of a specific antibody for canine PD-1. These preliminary results must therefore be investigated more in detail.
In conclusion, we here presented aberrant expression of PD-L1 on tumors in dogs and discussed their potential as therapeutic targets for canine tumors. PD-L1 might contribute to the progression of canine tumors via antitumor T-cell dysfunction. In the clinical trials targeting the PD-1/PD-L1 pathway in humans, none of the patients with PD-L1-negative tumors had a positive response; however, reactivation of the antitumor immunity and subsequent regression of some tumors were induced in patients with PD-L1-positive tumors, including malignant cases [19]. The findings indicate that detection of PD-L1 on tumor biopsies might be a powerful and effective method for prediction of prognosis after treatments. At least, it may be worth investigating the antitumor effects of an immunotherapy targeting the PD-1/ PD-L1 pathway on canine tumors that were PD-L1 positive in this study. Studies are underway to evaluate the possible clinical application of the PD-L1 antibody as a novel therapy against canine tumors.