CD44 is associated with proliferation, rather than a specific cancer stem cell population, in cultured canine cancer cells.

BACKGROUND: The cancer stem cell hypothesis proposes that tumours are maintained by a population of cancer stem cells (CSC), which must be eradicated to prevent disease relapse after treatment. Cells expressing high levels of CD44 have been identified as candidate CSC in a variety of human tumours. This study sought to investigate CD44 expression and its potential as a CSC marker in canine cancer. METHODS: CD44 expression in several canine cancer cell lines was determined by flow cytometry. Cells with low and high levels of CD44 expression were examined for differences in growth characteristics, colony forming ability, drug sensitivity and cell cycle profile. RESULTS: CD44(High) cells demonstrated enhanced growth and colony forming capacity, under both adherent and low-density serum free ("tumoursphere") conditions. However, no difference in sensitivity to doxorubicin was seen between the two populations. Moreover, whilst most CD44(Low) cells were in resting or G1 growth phase, an increased proportion of CD44(High) cells were in G2M phase of the cell cycle. Upon proliferation in culture, both populations gave rise to progeny with a full spectrum of CD44 expression. CONCLUSION: CD44 expression is associated with proliferation in cultured canine cancer cells, but transient and fluctuating expression may limit its utility as a CSC marker.

Characterization of feline T cell receptor gamma (TCRG) variable region genes for the molecular diagnosis of feline intestinal T cell lymphoma.

A diagnosis of intestinal lymphoma is currently made on the basis of clinical and morphologic criteria. This can prove problematic for many reasons that include inadequate sample size, the coexistence of lymphoma and inflammation, and the inability to assess architectural integrity of all tissue compartments in biopsy specimens obtained endoscopically. The detection of a clonal population of cells in a lymphoproliferative lesion represents an important criterion for the diagnosis of neoplasia, but this has not been assessed in feline intestinal lymphoma. T cell receptor gamma (TCRG) gene rearrangement analysis using polymerase chain reaction (PCR) is a methodology that can be used to detect clonality in T cell populations. The basis of this assay depends on the assessment of the junctional diversity that results from rearrangement of TCRG V (variable) and J (joining) gene segments. Feline TCRG transcripts from normal small intestine and spleen were obtained using a rapid amplification of cDNA ends (5'RACE) method. Limited diversity of TCRG V and J gene segments was observed. The high degree of sequence homology in the TCRG V and J gene segments was exploited to develop a PCR test for the assessment of TCRG V--J junctional diversity and hence clonality determination of T cell populations in cats. Molecular clonality determination was applied to feline intestinal lymphoplasmacytic inflammatory bowel disease (IBD) (9 cats), and transmural and mucosal T cell lymphoma (28 cats). Clonal rearrangement of the TCRG V--J junction was detected in 22 of 28 intestinal T cell lymphomas, and oligoclonality was detected in 3 intestinal T cell lymphomas. This contrasted with the detection of polyclonal rearrangement in normal intestinal tissues (3 cats) and in lymphoplasmacytic IBD (9 cats). It is proposed that assessment of TCRG V--J junctional diversity for the detection of clonality represents an important adjunctive tool for the diagnosis of T cell lymphoma in the cat.

Evaluation of in vitro cytotoxicity of nonsteroidal anti-inflammatory drugs against canine tumor cells.

Piroxicam and other nonsteroidal anti-inflammatory drugs (NSAID) have antitumor activity against naturally acquired cancer in dogs and human beings, and against experimentally induced tumors in rodents. We are investigating potential mechanisms of NSAID antitumor activity. The direct cytotoxicity of piroxicam indomethacin, and aspirin against 4 canine tumor cell lines (transitional cell carcinoma, squamous cell carcinoma, melanoma, and soft tissue sarcoma) was determined in short-term growth rate assays and in clonogenic assays. Piroxicam was evaluated alone and in combination with the lipoxygenase inhibitor zileuton, and in combination with the chemotherapeutic agents cisplatin and carboplatin. The 50% inhibitory concentrations (IC50) against melanoma cells in short-term growth rate assays were: 530 microM piroxicam, 180 microM indomethacin, and greater than 1 mM aspirin. These IC50 values were over 10 times greater than serum concentrations of these drugs that could safely be achieved in vivo. The IC50 of zileuton combined with piroxicam (280 microM) was not different from the IC50 of zileuton alone (230 microM; ANOVA P = 0.47) in melanoma cells. Similarly, addition of piroxicam did not alter the IC50 of either cisplatin (1.6 microM) or carboplatin (6.1 microM). These results suggest that NSAID, at serum concentrations achievable in vivo, do not have direct cytotoxicity against canine tumor cells tested. It is unlikely that the in vivo antitumor activity of NSAID is attributable to a direct cytotoxic effect.