The Expression of Selected Factors Related to T Lymphocyte Activity in Canine Mammary Tumors

Crosstalk between neoplastic and immune cells in the tumor microenvironment (TME) influences the progression of disease in human and canine cancer patients. Given that canine mammary tumors are a useful model to study breast cancer biology, we aimed to evaluate the expression of genes associated with T lymphocyte activity in benign, malignant, and metastatic canine mammary tumors. Interestingly, metastatic tumors exhibit increased expression of CXCR3, CCR2, IL-4, IL-12p40, and IL-17. In particular, we focused on IL-17, a key interleukin associated with the Th17 lymphocyte phenotype. Th17 cells have been shown to play a contradictory role in tumor immunity. Although IL-17 showed a high expression in the metastatic tumors, the expression of RORγt, a crucial transcription factor for Th17 differentiation was barely detected. We further investigated IL-17 expression using immunohistochemistry, through which we confirmed the increased expression of this interleukin in malignant and metastatic mammary tumors. Finally, we compared the plasma levels of IL-17 in healthy and malignant mammary tumor-bearing dogs using ELISA but found no differences between the groups. Our data indicate that the IL-17 in metastatic tumors may be produced by other cell types, but not by Th17 lymphocytes. Overall, our results broaden the available knowledge on the interactions in canine mammary tumors and provide insight into the development of new therapeutic strategies, with potential benefits for human immune oncology.


Introduction
Mammary gland tumors are one of the most common cancers in dogs, especially in non-spayed females, and are still one of the leading causes of death among canines [1][2][3]. Interestingly, canine mammary tumors are an attractive and useful model for studying human breast cancer. Importantly, cancer in dogs occurs spontaneously and naturally with age. These animals share the same environment as humans and are thus exposed to similar carcinogens. Moreover, multiple clinical similarities have been demonstrated concerning the hormonal etiology of tumors, risk factors, age of onset, clinical outcomes, and prognostic factors (e.g., tumor size, metastasis presence, and clinical stage). In addition, significant analogies at the molecular level have been shown in genome-wide comparative studies of mammary cancers between human and canine. It was shown, for example, that both human and canine mammary cancers are associated with brca1 and tp53 gene mutations [4]. The innate and adaptive immune response of canines is also comparable to that of humans. Major subsets of the dog immune system are well characterized to have significant homology to their corresponding the consumption of IL-2, and the production of immunosuppressive metabolites [27]. Since Treg infiltration has already been extensively studied in mammary canine tumors [28,29], we focused on the expression of other T lymphocyte-associated factors in canine mammary tumors.
Tumor-associated macrophages (TAMs) also play an important role in canine mammary cancer development [6,8]. They are different from monocytes, which are actively recruited from bone marrow by cancer cells. TAMs support tumor maintenance by enhancing the proliferation and migration of malignant cells, as well as facilitating angiogenesis extension via the secretion of growth factors, chemokines, and proangiogenic factors and switches. Moreover, it was shown that TAMs might inhibit the T lymphocyte-mediated antitumor immune response and express proteins that initiate effector immune cell death in humans and in dogs (e.g., FasL, PD-L1, and PD-L2) [30][31][32].
Canine mammary tumors are an attractive model for translational oncology and might be useful for the evaluation of novel diagnostic and therapeutic applications, including immunotherapy [4,7,8]. Therefore, our objective was to investigate the expression of genes encoding different molecules related to T cell activity, such as co-stimulatory and co-inhibitory ligands that regulate T cell activation, transcription factors, chemokine receptors, and inflammatory cytokines in canine mammary tumors with different malignancy statuses and metastatic potential (benign, malignant, and metastatic). We selected Th1-, Th2-, and Th17-related factors (partially based on the research by Park et al.) for canine meningoencephalitis [33]. Additionally, our aim was to characterize T lymphocyte activity-related factors, which might have importance in canine mammary tumors but are expressed not only by immune cells but also by tumor cells. Therefore, we analyzed genes whose expression is not limited to only T cells. We chose genes that encode proteins that regulate T cell activation or differentiation, such as PD-L2, Gal9, CD86, CXCR3, CCR2, which have not yet been investigated in canine mammary cancers. We also analyzed cytokine signatures, including IFN-γ, IL-4, and IL-17 (cytokines characteristic for the Tc lineage/Th1, Th2, and Th17 cells, respectively) as well as pro-and anti-inflammatory cytokines, such as IL-10, IL-12A, and IL-12B. Finally, we selected TBX21, GATA3, and RORγt as key transcription factors for differentiation of the Th1, Th2, and Th17 cell populations, respectively. We decided to cover a wide spectrum of proteins present in bulk tumor tissue that are associated with different T cell subsets. In particular, we focused on Th17 lymphocyte biomarkers since the expression and role of Th17 cells in canine tumors has not yet been fully unraveled. We evaluated the expression of IL-17 at the mRNA and protein levels via qPCR and the immunohistochemical staining of tumor samples, respectively. We also analyzed the IL-17 concentrations in the plasma of healthy and mammary tumor-bearing dogs using ELISA. Our goal was to broaden knowledge about the interplay between malignant cells and immune cells, with emphasis on the role of Th17 lymphocytes in canine mammary cancer. We also aimed to assess potential biomarkers and therapeutic targets for canine mammary tumors which might benefit future human breast cancer clinical trials.

Expression of Ligands Regulating T Cell Activation
Members of the B7-CD28 family play an important role in T cell activation as costimulatory signals that simultaneously modulate the scale of the T cell response [34]. In our study, we first examined the expression of PD-L2, which encodes the coinhibitory ligand B7-DC (PD-L2 and CD273), and Gal9, which encodes galectin-9, a carbohydrate-binding protein that causes T lymphocyte dysfunction and apoptosis via its receptor TIM-3 [35,36].
We found that the expression of both PD-L2 and Gal9 genes was highly upregulated in the malignant mammary tumors compared to the healthy tissue. Interestingly, in the metastatic tumors, PD-L2 and Gal9 expression decreased significantly compared to the malignant tumors that were not metastatic. There were no significant differences between the expression levels of PD-L2 and Gal9 in the healthy tissue and benign and metastatic tumors ( Figure 1A,B). CTLA4 (an important co-inhibitory receptor of T cell response) [37]. Our results show no statistically significant difference in CD86 expression between benign and malignant or metastatic tumors ( Figure  1C).

Expression of Cytotoxic T Lymphocyte Markers
Elevated numbers of infiltrating cytotoxic T cells have been well-established as a positive prognostic factor for various cancer types [17]. Therefore, we have determined the expression of CD8α and IFN-γ in canine mammary tumors. Increased expression of IFN-γ was observed in benign lesions compared to the malignant and metastatic tumors ( Figure 2B), whereas a slight decrease in CD8α expression was observed in metastatic tumors, albeit without statistical significance ( Figure  2A).

Figure 2.
Relative gene expression of CD8α (A) and IFN-γ (B) in healthy mammary gland tissue (Control) and in benign, malignant, and metastatic canine mammary tumors. The results are presented as the mean ± SEM. A one-way ANOVA and Tukey HSD post hoc test were applied. Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.

Th1 Cell-Associated Gene Expression
In order to assess the role of Th1 cells in canine mammary tumors, we investigated the expression of TBX21, a gene that encodes the T-bet protein, a major transcription factor that controls the differentiation of naïve T cells into the Th1 lineage [38]. Our data demonstrate that TBX21 expression was highly upregulated in malignant canine tumors in comparison to healthy tissue as well as benign and metastatic tumors ( Figure 3A). Interestingly, we also found significantly decreased expression of TBX21 in tumors that produce metastasis compared to the malignant nonmetastatic tumors ( Figure 3A). We have also assessed the expression of CD86, which encodes the B7-2 ligand. This protein can, paradoxically, bind to both CD28 (the major co-stimulatory molecule for the activation of T cells) and CTLA4 (an important co-inhibitory receptor of T cell response) [37]. Our results show no statistically significant difference in CD86 expression between benign and malignant or metastatic tumors ( Figure 1C).

Expression of Cytotoxic T Lymphocyte Markers
Elevated numbers of infiltrating cytotoxic T cells have been well-established as a positive prognostic factor for various cancer types [17]. Therefore, we have determined the expression of CD8α and IFN-γ in canine mammary tumors. Increased expression of IFN-γ was observed in benign lesions compared to the malignant and metastatic tumors ( Figure 2B), whereas a slight decrease in CD8α expression was observed in metastatic tumors, albeit without statistical significance (Figure 2A). CTLA4 (an important co-inhibitory receptor of T cell response) [37]. Our results show no statistically significant difference in CD86 expression between benign and malignant or metastatic tumors ( Figure  1C).

Expression of Cytotoxic T Lymphocyte Markers
Elevated numbers of infiltrating cytotoxic T cells have been well-established as a positive prognostic factor for various cancer types [17]. Therefore, we have determined the expression of CD8α and IFN-γ in canine mammary tumors. Increased expression of IFN-γ was observed in benign lesions compared to the malignant and metastatic tumors ( Figure 2B), whereas a slight decrease in CD8α expression was observed in metastatic tumors, albeit without statistical significance ( Figure  2A). Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.

Th1 Cell-Associated Gene Expression
In order to assess the role of Th1 cells in canine mammary tumors, we investigated the expression of TBX21, a gene that encodes the T-bet protein, a major transcription factor that controls the differentiation of naïve T cells into the Th1 lineage [38]. Our data demonstrate that TBX21 expression was highly upregulated in malignant canine tumors in comparison to healthy tissue as well as benign and metastatic tumors ( Figure 3A). Interestingly, we also found significantly decreased expression of TBX21 in tumors that produce metastasis compared to the malignant nonmetastatic tumors ( Figure 3A). The results are presented as the mean ± SEM. A one-way ANOVA and Tukey HSD post hoc test were applied. Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.

Th1 Cell-Associated Gene Expression
In order to assess the role of Th1 cells in canine mammary tumors, we investigated the expression of TBX21, a gene that encodes the T-bet protein, a major transcription factor that controls the differentiation of naïve T cells into the Th1 lineage [38]. Our data demonstrate that TBX21 expression was highly upregulated in malignant canine tumors in comparison to healthy tissue as well as benign and metastatic tumors ( Figure 3A). Interestingly, we also found significantly decreased expression of TBX21 in tumors that produce metastasis compared to the malignant non-metastatic tumors ( Figure 3A).
The CXCL9, -10, -11/CXCR3 axis regulates Th1 cell differentiation and immune cell migration to sites of inflammation [39]. Additionally, CXCR3 is expressed primarily on activated T cells, such as Th1 cells [33]. Thus, we examined CXCR3 gene expression as it relates to the Th1 lineage. Our results show that CXCR3 expression can be barely detected in benign tumors and healthy tissues ( Figure 3B). The expression of CXCR3 increased, however, with malignant transformation and achieved its highest expression in metastatic tumors ( Figure 3B).

Th2 Cell-Associated Gene Expression
To determine the Th2 lymphocyte phenotype in the canine mammary tumors, we assessed the expression of CCR2, GATA3, and IL-4. We examined the expression of CCR2, the receptor for MCP-1, which is involved in the development of the Th2 immune response [40]. Our data revealed the highly upregulated expression of CCR2 in metastatic tumors compared to benign and malignant tumors as well as healthy tissue ( Figure 4A).
The expression of GATA3, a transcription factor responsible for promotion of the differentiation of Th2 cells and the inhibition of Th1 development [38], was found to be increased in the benign tumors in comparison to healthy tissue as well as malignant and metastatic neoplasms ( Figure 4B).
Our results also reveal that the expression of the Th2-related cytokine, IL-4, was highly upregulated in metastatic tumors compared to the other groups ( Figure 4C). , and IL-4 (C) in healthy mammary gland tissue (Control) and in benign, malignant, and metastatic canine mammary tumors. The results are presented as the mean ± SEM. A one-way ANOVA and Tukey HSD post hoc test were applied. Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001. The CXCL9, -10, -11/CXCR3 axis regulates Th1 cell differentiation and immune cell migration to sites of inflammation [39]. Additionally, CXCR3 is expressed primarily on activated T cells, such as Th1 cells [33]. Thus, we examined CXCR3 gene expression as it relates to the Th1 lineage. Our results show that CXCR3 expression can be barely detected in benign tumors and healthy tissues ( Figure 3B). The expression of CXCR3 increased, however, with malignant transformation and achieved its highest expression in metastatic tumors ( Figure 3B).

Th2 Cell-Associated Gene Expression
To determine the Th2 lymphocyte phenotype in the canine mammary tumors, we assessed the expression of CCR2, GATA3, and IL-4. We examined the expression of CCR2, the receptor for MCP-1, which is involved in the development of the Th2 immune response [40]. Our data revealed the highly upregulated expression of CCR2 in metastatic tumors compared to benign and malignant tumors as well as healthy tissue ( Figure 4A).
The CXCL9, -10, -11/CXCR3 axis regulates Th1 cell differentiation and immune cell migration to sites of inflammation [39]. Additionally, CXCR3 is expressed primarily on activated T cells, such as Th1 cells [33]. Thus, we examined CXCR3 gene expression as it relates to the Th1 lineage. Our results show that CXCR3 expression can be barely detected in benign tumors and healthy tissues ( Figure 3B). The expression of CXCR3 increased, however, with malignant transformation and achieved its highest expression in metastatic tumors ( Figure 3B).

Th2 Cell-Associated Gene Expression
To determine the Th2 lymphocyte phenotype in the canine mammary tumors, we assessed the expression of CCR2, GATA3, and IL-4. We examined the expression of CCR2, the receptor for MCP-1, which is involved in the development of the Th2 immune response [40]. Our data revealed the highly upregulated expression of CCR2 in metastatic tumors compared to benign and malignant tumors as well as healthy tissue ( Figure 4A).
The expression of GATA3, a transcription factor responsible for promotion of the differentiation of Th2 cells and the inhibition of Th1 development [38], was found to be increased in the benign tumors in comparison to healthy tissue as well as malignant and metastatic neoplasms ( Figure 4B).
Our results also reveal that the expression of the Th2-related cytokine, IL-4, was highly upregulated in metastatic tumors compared to the other groups ( Figure 4C).  , and IL-4 (C) in healthy mammary gland tissue (Control) and in benign, malignant, and metastatic canine mammary tumors. The results are presented as the mean ± SEM. A one-way ANOVA and Tukey HSD post hoc test were applied. Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.
The expression of GATA3, a transcription factor responsible for promotion of the differentiation of Th2 cells and the inhibition of Th1 development [38], was found to be increased in the benign tumors in comparison to healthy tissue as well as malignant and metastatic neoplasms ( Figure 4B).
Our results also reveal that the expression of the Th2-related cytokine, IL-4, was highly upregulated in metastatic tumors compared to the other groups ( Figure 4C).

Anti-Inflammatory and Pro-Inflammatory Cytokine Expression
Crosstalk within the TME relies primarily on cytokine and chemokine signaling [41]. IL-10 was proven to be a potent anti-inflammatory cytokine, and is mainly produced by Treg cells but in also produced by other immune cells such as B cells, TAMs, mast cells, granulocytes, and dendritic cells as well as cancer cells themselves [42]. As shown in Figure 5A, we did not find significant differences in the expression of IL-10 between tumors and healthy tissue or among tumors with different malignancy statuses ( Figure 5A).

Anti-Inflammatory and Pro-Inflammatory Cytokine Expression
Crosstalk within the TME relies primarily on cytokine and chemokine signaling [41]. IL-10 was proven to be a potent anti-inflammatory cytokine, and is mainly produced by Treg cells but in also produced by other immune cells such as B cells, TAMs, mast cells, granulocytes, and dendritic cells as well as cancer cells themselves [42]. As shown in Figure 5A, we did not find significant differences in the expression of IL-10 between tumors and healthy tissue or among tumors with different malignancy statuses ( Figure 5A).
In contrast to IL10, IL-12, which is mainly produced by antigen-presenting cells, has been shown to exhibit pro-inflammatory effects via the induction of IFN-γ, producing Th1 and Tc cell differentiation as well as the enhancement of NK cell activity. IL-12 is a heterodimer that consists of p35 and p40 subunits. While the p35 subunit (IL-12A) is unique to the IL-12 cytokine, the p40 subunit (IL12B) is shared by both the IL-12 and IL-23 cytokines [43]. We observed marginal expression of IL-12p35 in canine mammary tumors compared to IL-12p40 ( Figures 5B and C). We did not find significant differences in the expression of IL-12p35 between different tumors groups and healthy tissue. By contrast, the expression of IL12-p40 was highly increased in metastatic tumors compared to non-metastatic and benign lesions ( Figure 5C).

Th17cell-Associated Gene Expression
The subpopulation of Th17 lymphocytes could play dual and opposing roles in cancer progression [23]. Our study revealed very high IL-17 expression only in the metastatic tumors. The expression of IL-17 observed in metastatic tumors was significantly higher compared to in benign and non-metastatic malignant tumors as well as healthy tissue ( Figure 6A).
Interestingly, expression of the Th17 cell-specific transcription factor ROR-γt was significantly increased in benign neoplastic lesions compared to malignant tumors and healthy tissue. We barely detected ROR-γt gene expression in metastatic canine mammary tumors ( Figure 6B). In contrast to IL10, IL-12, which is mainly produced by antigen-presenting cells, has been shown to exhibit pro-inflammatory effects via the induction of IFN-γ, producing Th1 and Tc cell differentiation as well as the enhancement of NK cell activity. IL-12 is a heterodimer that consists of p35 and p40 subunits. While the p35 subunit (IL-12A) is unique to the IL-12 cytokine, the p40 subunit (IL12B) is shared by both the IL-12 and IL-23 cytokines [43]. We observed marginal expression of IL-12p35 in canine mammary tumors compared to IL-12p40 ( Figure 5B,C). We did not find significant differences in the expression of IL-12p35 between different tumors groups and healthy tissue. By contrast, the expression of IL12-p40 was highly increased in metastatic tumors compared to non-metastatic and benign lesions ( Figure 5C).

Th17cell-Associated Gene Expression
The subpopulation of Th17 lymphocytes could play dual and opposing roles in cancer progression [23]. Our study revealed very high IL-17 expression only in the metastatic tumors. The expression of IL-17 observed in metastatic tumors was significantly higher compared to in benign and non-metastatic malignant tumors as well as healthy tissue ( Figure 6A).
progression [23]. Our study revealed very high IL-17 expression only in the metastatic tumors. The expression of IL-17 observed in metastatic tumors was significantly higher compared to in benign and non-metastatic malignant tumors as well as healthy tissue ( Figure 6A).
Interestingly, expression of the Th17 cell-specific transcription factor ROR-γt was significantly increased in benign neoplastic lesions compared to malignant tumors and healthy tissue. We barely detected ROR-γt gene expression in metastatic canine mammary tumors ( Figure 6B). Interestingly, expression of the Th17 cell-specific transcription factor ROR-γt was significantly increased in benign neoplastic lesions compared to malignant tumors and healthy tissue. We barely detected ROR-γt gene expression in metastatic canine mammary tumors ( Figure 6B).

IL-17 Protein Level in the Canine Mammary Tumors and Tumor-Bearing Dog Plasma
Due to the significant upregulation of IL-17 gene expression in the metastatic tumors in our transcriptomic study and given the controversial role of Th17 cells in tumor progression, we further investigated IL-17 protein expression in canine mammary tumors.
The immunohistochemical analysis demonstrated a significantly higher expression of IL-17 protein in mammary tumors compared to healthy tissue ( Figure 7A-E). However, we did not find differences among the tumors at different stages of malignancy. Our data revealed comparable levels of IL-17 expression in the benign proliferative lesions and malignant tumors regardless of metastatic behavior ( Figure 7E).
In addition, we determined the concentration of IL-17 in the plasma of client-owned healthy dogs and malignant mammary tumor-bearing dogs. We did not find significant differences of plasma IL-17 level between the groups. This cytokine was detected in the plasma of 8 of 21 and 10 of 18 healthy and tumor-bearing dogs, respectively ( Figure 7F).

Discussion
T lymphocytes play a crucial role in shaping the immune response against cancer. Multiple studies have demonstrated the presence of T lymphocytes in various types of mammary gland neoplasms [19]. A recent study has shown that lymphocytes comprise the primary population of inflammatory cells infiltrating mammary tumors regardless of their histological type [6]. Moreover, several studies have shown that a high number of infiltrating CD3 + T cells is associated with poor prognoses and lower survival rates [14,18,44]. In both human and canine mammary gland cancers,

Discussion
T lymphocytes play a crucial role in shaping the immune response against cancer. Multiple studies have demonstrated the presence of T lymphocytes in various types of mammary gland neoplasms [19]. A recent study has shown that lymphocytes comprise the primary population of inflammatory cells infiltrating mammary tumors regardless of their histological type [6]. Moreover, several studies have shown that a high number of infiltrating CD3 + T cells is associated with poor prognoses and lower survival rates [14,18,44]. In both human and canine mammary gland cancers, the infiltration of T cells is known to be one of the factors correlated with angiogenesis and metastasis [14,45].
Our goal was to analyze the expression of T lymphocyte-related factors, such as T cell co-stimulatory and co-inhibitory ligands, transcription factors, chemokine receptors, and inflammatory cytokines that had not yet been investigated in canine mammary tumors.
Multiple inhibitory ligands are expressed by either cancer cells or immune cells (mainly macrophages and dendritic cells) in an attempt to suppress T lymphocyte activity [34]. We documented significantly increased expression of the co-inhibitory ligand PD-L2 in malignant mammary neoplasms. These results suggest the action of an immunomodulatory mechanism in aggressive tumors that seeks to avoid immune surveillance. In contrast to PD-L1, which was reported to be wildly expressed in both canine immune cells and cancer cells [46], PD-L2 expression has not yet been reported in canine mammary tumors. PD-L2 is primarily expressed in dendritic cells and macrophages and, also, rarely in lymphocytes [34]. Given that malignant canine mammary tumors have been shown to be largely infiltrated by macrophages [47], we suspect that the increased expression of PD-L2 might be linked with a high number of TAMs in canine mammary tumors.
Interestingly, the expression of PD-L2 and the other co-inhibitory ligand, Gal9, was significantly decreased in metastatic tumors compared to malignant non-metastatic tumors. Similarly, PD-L1 expression was shown to decrease in metastatic human breast cancer [48]. Both PD-L1 and -L2 are important inhibitory molecules in TME, which might have clinical significance in the prognoses of cancer patients and responses to checkpoint blockade immunotherapy [34]. In human breast cancer, PD-L2 expression was observed in 50.8% cases but was not correlated with overall survival [49]. The mechanism and relevance of altered PD-L2 expression in metastatic canine mammary tumors remains to be determined. Gal9 was demonstrated to boost antitumor immunity mediated by Th1 cells and exert anti-metastatic potential on human tumor cells through the mediation of cell aggregation [36,50,51]. Irie et al. showed that 19 out of 21 human patients with distant metastasis of their breast cancer were Gal9 negative. Moreover, the cumulative disease-free survival ratio for the patients from the Gal9-negative group was less favorable than that for the Gal9-positive patients. The reduced expression of Gal9 was also linked with the inhibition of NK cell chemotaxis and subsequent poor prognosis for colon cancer, hepatocellular carcinoma, and cervical cancer [35,[51][52][53][54]. The expression of Gal9 was inversely correlated with the occurrence of distant metastasis but not with other clinical features, including local lymph node metastasis. Thus, Gal9 positive expression is proposed to be a useful prognostic factor of anti-metastatic potential in breast cancer patients [50]. Our study is the first to report a significant downregulation of Gal9 expression in metastatic canine mammary tumors. The present results indicate that, similar to human patients, the expression of Gal9 in dogs could potentially be used as a prognostic marker.
We also examined the expression of transcription factors, such as TBX21, GATA3, and RORγt, which are essential for the differentiation of Th1, Th2, and Th17 cells, respectively. Our results revealed that in malignant tumors compared to benign lesions or healthy tissue, there was a much higher expression of TBX21 encoding the Th1 cell-specific transcription factor T-bet. Interestingly, TBX21 expression was also higher in non-metastatic malignant tumors than those that are metastatic. These odd results might be explained by research performed on a large cohort of women with breast carcinoma. This study demonstrated that although high T-bet expression is associated with adverse clinicopathologic characteristics, such as large tumor size, high histological grade, hormone receptor negativity, EGFR and p53 positivity, and a high Ki-67 index, T-bet +/high tumors have a more favorable outcome and longer disease-free survival time compared to T-bet −/low tumors [55]. Our results indicate that in malignant canine mammary tumors, similar to human breast cancers, TBX21 expression might be associated with a more favorable prognosis related to a low metastatic potential for the cancer. Furthermore, these results underscore the role of infiltrating Th1 cells in antitumor immunity.
Several studies have shown that GATA3 regulates mammary gland morphogenesis and luminal cell differentiation [56]. In addition, GATA3 expression was strongly correlated with the luminal A subtype of breast tumors, which carried the best prognostic outcome for human patients [48,49]. We found increased levels of GATA3 in benign canine mammary tumors compared to healthy tissue as well as malignant and metastatic tumors. Our results contradict a previous study that documented the highest GATA3 expression in canine mixed carcinoma, with the lowest level in a complex adenoma [57]. However, our results are consistent with a study performed on mice demonstrating that the loss of GATA3 expression is associated with mammary tumors with a high grade of malignancy [58]. Likewise, in human breast tumors, low GATA3 expression was shown to be a strong predictor of poor clinical outcome, high tumor grade, positive lymph node status, and large tumor size [56]. The high expression of GATA3, in turn, was correlated with the level of estrogen receptor-α (ER), which is a highly significant marker of sensitivity to hormonal therapy in breast cancer patients [59]. Notably, GATA3 expression might be negatively regulated by T-bet. A recent study has shown that the insulin-dependent increase in the expression of T-bet in breast tumors is associated with a subsequent decrease in the expression of GATA3, followed by resistance to hormonal therapy in human ER + breast cancer [60].
Data concerning ER expression in canine mammary tumors differ depending on the study. Nieto et al. showed that the immune expression of ER in dogs has prognostic value and is significantly higher in benign tumors [61]. However, Tonti et al. demonstrated that more than half of the benign and malignant canine mammary tumors negatively stained for ER, which undermines the clinical significance of ER [62]. The relationship between GATA3, T-bet, and ER expression in mammary tumors in dogs remains to be defined.
RORγt is the principal transcription factor in Th17 cell development [23]. Since the role of Th17 cells in cancer has been specified to be a "double-edged sword" [23,25,26], we aimed to verify the expression of their major transcription factor. Our results are the first to demonstrate that RORγt expression is almost undetectable (very low expression in 4 out of 20 tumors) in metastatic canine mammary tumors but is upregulated in benign canine neoplasms. Interestingly, a recent study has documented that RORγ negatively regulates metastasis and aggressive tumorigenicity in human breast cancer [63]. It was shown that increased levels of RORγ are associated with better clinical outcomes. The mechanisms of RORγ activity are associated with the promotion of DNA repair and inhibition of the TGF-β-mediated epithelial-to-mesenchymal transition and mammary stem cell-related pathways. Thus, it was suggested that specific RORγ agonists may exhibit anticancer activities, targeting critical processes that drive the progression of breast cancer [63]. Further investigations of RORγt activity in canine mammary tumors are needed to assess RORγt's utility as a therapeutic target.
Chemokine signaling controls multiple processes during tumor progression, including tumor growth, angiogenesis, and metastatic spread [64]. We have investigated the expression of chemokine receptors, such as CXCR3 and CCR2, as associated with the development of Th1-and Th2-mediated immune response, respectively. We observed a significantly higher expression of both CXCR3 and CCR2 in canine mammary tumors that gave local or distant metastases. Although it was shown that the expression of CXCR3 on T lymphocytes mediates their migration to the tumor side, the presence of CXCR3 on malignant cells promoted tumor growth and dissemination. Notably, two isoforms of CXCR3 with opposite effects might be present in the TME [65]. Our results are consistent with recent meta-analyses, which have shown that higher CXCR3 expression indicates an advanced tumor stage and is correlated with the occurrence of distant metastasis in solid tumors [66,67]. Moreover, targeting CXCR3 was proposed as a strategy to prevent metastatic formation in breast cancer through the inhibition of CXCR3-dependent cancer cell migration and the enhancement of host antitumor immunity [66].
The CCL2/CCR2 axis in breast cancer is known to be involved in the metastatic process via the recruitment of different myeloid cell subsets, including TAMs and the induction of angiogenesis [68]. In contrast to the study that revealed no differences in CCR2 expression in the most aggressive canine inflammatory mammary cancers compared to non-inflammatory tumors [69], we documented a significant increase in CCR2 expression in the metastatic cancers versus the benign and malignant non-metastatic tumors. Our results indicate that, similar to humans, CXCR3 and CCR2 expression might be associated with metastatic formation in dogs. Targeting these chemokine receptors could potentially be beneficial for canine mammary cancer patients.
A variety of cytokines, chemokines, and growth factors are produced in the TME, thereby affecting antitumor immunity [70]. The goal of our study was to investigate whether the autocrine/paracrine cytokine network is relevant to canine mammary tumors. We determined the expression of INF-γ, IL-4, IL-10, IL-12p35, IL-12p40, and IL-17, cytokines that are related to Tc/Th1, Th2, Treg, macrophage, and Th17 cell phenotypes, respectively. The cytokine profiles of canine malignant and metastatic mammary carcinomas were described, illustrating the upregulation of IL-1, IL-6, TNF-α, and IFN-α expression [71,72]. Our study additionally revealed a significant increase in IL-4, IL-12p40, and IL-17 expression in metastatic canine mammary tumors compared to non-metastatic tumors and healthy tissue. IL12p40 was shown to be highly expressed in breast cancer tissue. Furthermore, by binding to the IL-12 receptor, the IL-12p40 monomer can inhibit its immunological functions, thereby evading detection and reducing the eradication of cancer cells by the immune system [73].
Although IL-10 expression was demonstrated to be upregulated in human breast cancer as well as in inflammatory mammary carcinomas in dogs [74,75], we did not find significant differences between IL-10 expression in healthy and neoplastic lesions, regardless of malignancy stage.
Our study did not include the assessment of the stroma content in tumors. However, our results indicated several molecules that could serve as potential prognostic biomarkers or therapeutic targets in canine mammary tumors.
The role of Th17 cells in cancer development remains controversial [23,25,26]. We found a significantly increased expression of IL-17 among the metastatic tumors. However, we did not detect the expression of the Th17 cell-specific transcription factor (RORγt) in these tumors. Thus, we anticipated the presence of other IL-17 producing cells in the canine mammary tumors. Indeed, we revealed, using immunohistochemical staining, that IL-17 is potentially produced by tumor cells themselves and infiltrates innate immune cells such as TAMs and neutrophils.
The production of IL-17 by breast cancer-associated macrophages in humans has been previously demonstrated [76]. Further, IL-17C and IL-17E expression have been reported in human epithelial cells [77,78]. In addition, the predominance of IL-17-producing cells in breast cancer was revealed to be a poor prognostic factor linked with shorter disease-free survival in human patients [79]. A recent study reported elevated levels of Th17 cell cytokines (IL-17A and IL-17F) in T cell non-inflamed triple-negative breast cancer; thus, IL-17 was proposed as a novel prognostic biomarker for this type of cancer [80].
To evaluate the role of the IL-17 as a diagnostic biomarker in canine metastatic mammary tumors, we investigated the plasma levels of IL-17 in healthy and malignant mammary tumor-bearing dogs. We detected IL-17 in approximately half of the samples from both groups and did not find significant differences between the groups. Our results indicate that although IL-17 expression is upregulated in canine mammary tumors, this upregulation is not reflected by its concentration in the peripheral blood. On the contrary, a recent study has revealed that the plasma levels of IL-17 increase in human patients with atypical breast hyperplasia and ductal carcinoma [81].
The present data suggest that the action of IL-17 is local than systemic, and that IL-17, similarly to human studies, may be associated with the development and metastasis of canine mammary tumors. Based on the results showing increased IL-17 protein expression in canine mammary tumors compared to healthy tissue, we believe that IL-17 supports tumor growth and may exert more proangiogenic effects than the immunomodulatory activity in canine mammary tumors. Nevertheless, further investigations are needed to assess the production of IL-17 by cancer cells and the impact of IL-17 on migration and invasiveness, as well as to determine the expression of IL-17 surface receptors.

Canine Mammary Tumor Samples
Canine mammary tumor samples were collected during mastectomies from female dogs of various breeds and ages. Half of the tumor samples were fixed in 10% neutral buffered formalin and the other half of the same tumor samples were immersed in RNAlater ® RNA Stabilization Solution (Invitrogen) and stored at −80 • C. For histological examination, the formalin fixed samples were dehydrated, embedded in paraffin, and processed for routine hematoxylin and eosin (HE) staining. The tumors were diagnosed according to the World Health Organization (WHO) Histological Classification of Mammary Tumors of the Dog and the Cat with further modifications [82][83][84]. The tumor samples included in our study were diagnosed as benign (simple adenomas, complex adenomas, basaloid adenoma, fibroadenoma, benign mesenchymal tumors, and benign mixed tumors, n = 14), malignant with no detectable metastasis (solid carcinomas, simple carcinomas, tubulopapillary carcinomas, basaloid carcinomas, and mixed carcinomas) (n = 18), and malignant with local or distant metastases (carcinomas, simple carcinomas, tubulopapillary carcinomas, ductal carcinomas, and mixed carcinomas), referred to as metastatic (n = 20). Radiography was used for the diagnosis of lung metastases. Mammary gland tissues obtained from healthy females were used as control samples (n = 10). In all cases, the owners' consent was obtained for use of the canine patients' tumor samples.

RNA Isolation and qPCR
qPCR was used to examine mRNA expression. RNA was isolated from tumor samples after disruption and homogenization at 50 Hz for 15 min using the TissueLyser LT (Qiagen, Hilden, Germany). RNA was isolated using a Total RNA kit (A&A Biotechnology, Gdynia, Poland) according to the manufacturer's protocol and stored at −80 • C. The quantity of isolated RNA was measured using NanoDrop 2000 (NanoDrop Technologies, Waltham, MA, USA). The samples with adequate amounts of RNA were treated with DNaseI to eliminate DNA contamination. The samples were subsequently purified using an RNeasy MiniElute Cleanup Kit (Qiagen). Finally, the RNA samples were analyzed on a BioAnalyzer (Agilent Technologies, Santa Clara, CA, USA) to measure the final RNA quality and integrity. cDNA synthesis was performed using High-Capacity cDNA Reverse Transcription Kits (ThermoFisher Scientific, Waltham, MA, USA), according to the manufacturer's protocol, in a Mastercycler thermal cycler (Eppendorf, Hamburg, Germany).
The primers for PD-L2, Gal9, CD86, and CD8α were designed de novo using PRIMER3 software (free online access, http://bioinfo.ut.ee/primer3-0.4.0/) and checked using an Oligo Calculator (free online access, http://biotools.nubic.northwestern.edu/OligoCalc.html) and Primer-Blast (NCBI database, https://www.ncbi.nlm.nih.gov/tools/primer-blast/). The other primers used in our study to evaluate Th1-, Th2-, and Th17-related cytokine and chemokine receptors were reported in a a study of canine meningoencephalitis by Park et al. 2013 [33]. The sequences of primers used for qPCR are listed in Table 1. The qPCR reaction was performed using SYBR Green and a Stratagene Mx3005P QPCR System (Agilent Technologies) with an initial degeneration step at 50 • C for 2 min and 95 • C for 2 min, followed by 35 cycles at 95 • C for 15 s, an annealing step at 58 • C for 15 s, and 72 • C extension for 1 min. At the end of each cycle, the intensity of the fluorescence emitted from SYBR Green was measured. Next, the samples were subjected to a dissociation curve analysis. After the completion of this process, the samples were automatically quantified using the MxPro QPCR Software (Agilent Technologies). A single narrow peak was observed in a dissociation curve analysis at the specific melting temperature for all primers used, confirming the specificity of the observed product. Additionally, a single band of the predicted size was observed by 2% agarose gel electrophoresis for the PD-L2, Gal9, CD86, and CD8α primers, which were designed de novo by our team. RPS19 and HPRT genes were used as a non-regulated reference (housekeeping) for normalization of the target gene expression. The relative mRNA expression was calculated using the comparative Ct method [85] as 2 −∆Ct , (∆Ct = Ct reference − Ct target ).

Immunohistochemistry (IHC)
Sections of 4 µm thickness were deparaffinized in xylene and rehydrated in graded ethanol. Antigen was retrieved by placing the slides in 0.02 M citrate buffer (pH 6.0) and boiling in a decloaking chamber. Subsequently, the samples were treated with a peroxidase blocking reagent (Dako, Glostrup, Denmark) for 10 min at room temperature and blocked with 5% bovine serum albumin (Sigma Aldrich, St. Louis, MO, USA) for 30 min. Next, the sections were incubated with primary rabbit antibody anti-canine IL-17 (1:200 dilution; ab79056, Abcam, Cambridge, UK) diluted in 1% bovine serum albumin overnight at 4 • C. After the PBS washing, the slides were incubated with an HRP-labeled polymer conjugated to secondary anti-rabbit antibodies from the EnVision™+ System kit (Dako). Next, 3,3-diaminobenzidine tetrahydrochloride (DAB, Dako) was used for color development. The slides were then counterstained with hematoxylin and dehydrated. Finally, the sections were mounted with mounting medium (Dako) for microscopic evaluation. A negative control test was performed by omitting the primary antibody on corresponding representative slides in each IHC experiment. Three consecutive tissue sections were examined. Around 10-20 pictures were taken of each slide (depending on the tumor size) using a BX60 microscope (Olympus, Tokyo, Japan). The colorimetric intensity of IL-17 expression, reflected as IHC-stained antigen spots (brown color), was counted by a computer-assisted image analyzer (Olympus Microimage™ Image Analysis, software version 4.0 for Windows, USA). The results are presented as the mean pixel integrated optical density (IOD) related to the color intensities of each IL-17 antigen spot.

ELISA
EDTA-anticoagulated whole blood was collected for diagnostic purposes from client-owned healthy dogs (n = 21) and mammary tumor-bearing dogs (n = 18) during routine veterinary procedures. All tumors were subsequently diagnosed as malignant, with adenosquamous, tubulopapillary, anaplastic, ductal, papillary, simple, complex, and mixed carcinomas. The whole blood samples were centrifuged (15 min at 2500 rpm), and the resulting supernatant (plasma) was collected and stored at −20 • C. Plasma IL-17 levels were measured using a Canine IL-17/IL-17A Quantikine ELISA Kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's recommendations.

Statistical Analysis
The statistical analysis was performed using the Prism software, version 5.0 (GraphPad Software, San Diego, CA, USA). The mRNA expression was analyzed by a one-way ANOVA followed by a Tukey honest significant difference (HSD) post-hoc test. When required (as determined by Bartlett's test), the of data was logarithmically transformed to meet the ANOVA assumption of homogeneity of variance. The IHC data were analyzed by one-way ANOVA and Tukey HSD post hoc tests, while for the ELISA data analysis, Student's t-test was applied. The data were expressed as the mean ± SEM, unless otherwise stated. Significance levels were indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.001.

Conclusions
Lymphocytic infiltration is an important feature of canine mammary tumors. Our data are the first to document the expression of T lymphocyte-associated factor expression, including the co-inhibitory ligands (PD-L2, Gal9), transcription factors (GATA3, TBX21, and RORγt), chemokine receptors (CXCR3 and CCR2), and inflammatory cytokines (IL-4, IL-12p40, and IL-17) relevant to canine mammary tumors. However, detailed study of the stroma content of tumors would be beneficial to enrich the understanding of the specific T cell subpopulations role in canine mammary tumors biology.
Interestingly, we revealed that the upregulation of IL-17 is not related to the presence of Th17 lymphocytes but rather to infiltration by other innate immune cells and, potentially, IL-17 production by tumor cells. Our results may have strong implications for translational oncology and the development of novel cancer immunotherapies.