A novel canine histiocytic sarcoma cell line: initial characterization and utilization for drug screening studies

Histiocytic sarcoma (HS) is a highly aggressive hematopoietic neoplasm in humans and animals. In humans, HS accounts for less than 1% of all hematopoietic neoplasms [1, 2], affects all ages, but predominately adults, and involves lymph nodes and/or a variety of extranodal organs including skin, bone marrow, spleen, the gastrointestinal tract and the central nervous system [3, 4]. This malignancy is often approached with a combination of various modalities of treatment, including multi-drug chemotherapeutic protocols, radiation therapy, surgery, and bone marrow transplantation [5‐7]. However, patients with HS in general respond poorly to any form of treatment, therefore this neoplasm carries a poor prognosis and has a high mortality rate. The low number of clinical cases is a limitation for an extended understanding of the pathogenesis of this disease in people, and restricts the investigation for novel and more efficacious forms of treatment.

In dogs, a similar disorder is commonly found in certain breeds, especially the Bernese mountain dog (BMD), in which the prevalence ranges from 15 to 25% of the population [8‐12]. The genetic susceptibility to HS in dogs has been the focus of several investigations. Current published data of a population of BMD from US and Europe, suggests deregulation of the tumor suppressor genes MTAP/CDKN2A/B located within the region homologous to human chromosome 9p21 [13, 14]. Studying HS in dogs is of high importance as, similarly to people, it is a fatal disease characterized by rapid progression and high metastatic rate [15‐18]. Thus dogs, with spontaneously occurring HS, are a crucial model for development of new approaches to treat this orphan disease in people. Affected canine patients also respond poorly to treatment. The currently most effective drug is N-(2-chloroethyl)-N′-cyclohexyl-N-nitrosourea (CCNU) that provides a reported response rate ranging from 29 to 46% for a median survival time of 85–96 days [16, 19]. A preliminary study using another cytotoxic drug, doxorubicin, showed a similar response rate of 46% for a median survival time of 93 days [20]. Additional alternatives of treatment have been investigated on small cohorts of dogs, including delivery of the human MHC non-restricted T-cell line TALL-104, frameless stereotactic radiosurgery, and chemotherapy using either paclitaxel or pegylated-liposomal doxorubicin; however, none has shown promising results [20‐24].

Research groups have demonstrated the therapeutic potential of drugs as small molecule inhibitors against HS, based on in vitro studies [25‐27]. Screening a large library of small molecules, Ito et al. successfully identified eight compounds of high potency (> 60% inhibition) at concentration of 100 nM [25]. Interestingly, only two drugs shared the same main target, showing the diversity of factors driving tumorigenesis in HS, and the potential benefit of drug combination hitting multiple targets. In humans, there is only a single report of four patients with HS treated with small molecule inhibitors, associated with the overexpression of targets for those drugs within the neoplasm, however efficacy could not be evaluated due to the small cohort of patients [28]. No published studies exist reporting the clinical use of small molecule inhibitors in dogs with HS.

We present here details of a novel HS cell line from a BMD, named the BD cell line, we have successfully established and demonstrated its utility in identification of potential novel therapeutic options [29]. Considering that to date, there is no human HS cell line available for research purposes, we believe that this cell line will provide an essential scientific tool for the study of this disease.

The use of the naturally occurring histiocytic sarcoma in the dog as a translational model for human HS represents a valuable opportunity to further understand this malignancy, and to find better tools for treatment. In this article we have presented the initial characterization of a novel HS cell line - BD cell line - and investigated potential novel treatment approaches.

Histiocytic sarcomas are frequently associated with a high level of cellular pleomorphism, therefore, the definitive diagnosis is determined by a pattern of expression of markers specific for histiocytes. Although, several markers have been used to characterize human HS including CD163, CD68, CD11c, lysozyme, and CD14 [4, 32, 33], two different markers were validated for HS tumors in dogs: CD18 (integrin beta chain beta 2) and CD204 (class A macrophage scavenger receptor) [8, 15, 30, 31, 34]. The novel BD cell line showed positive expression of CD18 and CD204, consistent with the diagnosis of HS. The negative expression of CD3, CD21 and CD79a excluded lymphoid malignancies, while positive vimentin, a type III intermediate filament protein, confirmed the mesenchymal origin, and positive CD45 confirmed the myeloid origin. Interestingly, the stimulation of BD cells with LPS, a lipopolysaccharide component of Gram-negative bacteria walls, resulted in increased expression of MHC class II. A similar increase in MHC Class II expression was reported when murine dendritic cells were exposed to LPS, including CD14 positive dendritic cells [35, 36]. CD14 expression is mainly present in macrophages and monocytes, however it can also be found in DCs, where it has been proposed to play a role on the immunologic responses in DCs [37]. The same effect was not observed when the cells were stimulated with interferon-γ (IFNγ). Due to the fact that the majority of BD cells expressed MHC I and MHC II, we could confirm an antigen presenting cell profile. Both macrophages and DCs are antigen presenting cells that, despite their differences in biological functions, are only distinguished by a few surface markers. Candidate markers that have an increased specificity for DC and DC subsets have been identified in several studies [38‐41]. Two DC-specific markers were tested in the present study, CD11c and CD172a, for which the vast majority of BD cells were positive. Together, these findings indicated a pattern of expression associated with a dendritic cell origin.

Dendritic cells are professional phagocytes that have an important role in processing antigens for adaptive immune recognition [42]. The majority of cells from the newly established cell line were able to phagocytize pHrodo E. coli bioparticles when coincubated for 2 h. In this assay, the bioparticles would only emit fluorescence once they are inside the phagosomes of the cells, where the pH is low. Therefore, this pH-sensitive fluorescent method permits a clear discrimination of where the particles are located in respect to the cells [43]. As a representation of positive controls for this assay, the two macrophage cell lines, J774.A and DH82, showed a high level of phagocytosis. In contrast, we could appreciate phagocytic activity in only a very small number of canine fibroblasts, which were used as negative control.

When transplanted to an immunodeficient xenograft mouse, BD cell line successfully grew as a palpable soft tissue mass after 29 days, a time-frame comparable to other tumor xenograft studies [44]. We were able to confirm that BD cell line is a tumorigenic cell line, and could potentially be useful for in vivo studies as an experimental tool to study histiocytic sarcoma. Further studies with a larger number of animals are warranted to truly establish the transplantability and metastatic potential. A few xenograft mouse models of canine HS have been successfully established and were used for the evaluation of their metastatic potential, and to study the phenomenon of resistance to CCNU-based treatment [45, 46]; however, none of these cell lines had originated from BMDs, the most frequently affected breed. To the best of our knowledge, across the many existent canine HS cell lines, only two others were reported to be originated from tumors of BMDs [47]; and these two were not part of their drug response report.

Results from the in vitro drug screening experiment showed a variable response between the different drugs, and across the cell lines. Across a total of 13 compounds, dasatinib and doxorubicin effectively inhibited the cell growth of both HS cell lines within a clinically tolerable and achievable plasma concentration according to the veterinary and human literature.

Doxorubicin was the only conventional chemotherapeutic drug that elicited a favorable response against the HS cell lines. Several studies have reported the use of doxorubicin for the treatment of histiocytic disorders, including HS [5, 6, 33, 48‐51]. Due to the low number of cases, the effectiveness of doxorubicin could not be determined in those studies, especially as single agent, as the drug was invariably used in a drug combination protocol. Our results agree with previous in vitro studies that considered doxorubicin an effective drug against canine HS [20, 52]. The potential of doxorubicin for the treatment of dogs with HS was suggested by a preliminary clinical study, where the response to doxorubicin was comparable to the response to the drug with the best response reported to date, CCNU [20].

Our positive results with dasatinib are in accordance with results from a study by Ito et al., where dasatinib was effective against 4 out of 7 canine HS cell lines derived from various breeds. The calculated IC₅₀ values from that study varied from 5.4 to 54.5 nM, while the average IC₅₀ value in our study was 9.5 nM, clearly within their range [25]. Dasatinib is an multi tyrosine kinase inhibitor with multiple main targets, including the SRC family kinases (SRC, LCK, YES, and FYN), the BCR-ABL, and to a lesser extent, c-KIT, PDGFRβ and EPHA2 [53]. We demonstrated that dasatinib inhibited the activation of SRC, as revealed by decrease of p-SRC (phospho-SRC) in both cell lines, BD and DH82. We hypothesize that the downregulation of p-SRC could be associated with the inhibitory effect of dasatinib on the HS cell lines, however, as a multi-kinase inhibitor, many other molecular targets might also be affected. However, the SRC pathway is a major oncogenic driver involved in HS against which dasatinib and other novel compounds may be used. Although dasatinib has been investigated in many human cancers, its clinical therapeutic value for HS has never been documented. [54‐58]. However, the use of other small inhibitors for the treatment of HS has been reported in one study where human patients were treated with imatinib, sorafenib and bevacizumab, based on the pattern of expression of key molecular targets [28]. Another set of studies reported two patients carrying a HS associated with a mutation in BRAF^V600E gene, most commonly seen in melanomas in humans [50, 59]. Neoplasms that are driven by this mutation are suitable for vemurafenib-based treatment, a B-Raf small molecule inhibitor. In fact, a human patient from one of the studies was treated with vemurafenib and experienced therapeutic response for a couple of months [59]. Although the efficacy of these drugs could not be evaluated due to the small cohort of patients, the identification of druggable targets in HS shows the value of targeted therapy for this disease.

We demonstrated that the combination of dasatinib and doxorubicin resulted in a favorable additive effect against one of the two HS cell lines, DH82 cell line. The association of small inhibitors with chemotherapeutic drugs has been strategized as a treatment with broader spectrum, which resulted in synergistic anticancer activity for many tumors [26, 60, 61]. Often the increased efficiency is accompanied with a dose reduction of the drugs, and consequently decreased treatment related side effects. The combination of drugs with different mechanisms of action should be considered as a relevant strategy to optimize the therapeutic effect of each inhibitor against HS.

The present study was able to establish and fully characterize a DC-subtype HS cell line from a tumor in a BMD. This novel HS cell line represents a model available not only for the investigation of potential therapeutic drugs, but also for the studies of gene expression and genetic variability associated with HS. The existence of cancer cell lines has been a critical tool for the understanding of cancer biology and response to therapy. For that matter, panels of human cancer cell lines such as NCI60 and GDSC have become available for researches, so that the collective use of these resources would generate more efficiently information including genomics and drug sensitivity of these cell lines [62, 63]. Nevertheless, it is valuable to mention that cell lines are prone to genetic changes over time, and early passages should be a better model of the tumor of origin. More recently, a panel of canine cancer cell lines has been created covering the most important cancers seen in canine patients [64]. Due to the rarity of HS in humans, studies based on spontaneous HS in dogs can provide important and relevant translational understanding of this malignancy. Moreover, we also propose BD cell line as a useful system for studies where a cell line of dendritic cell origin can be of value.