No abstract

BackgroundA fundamental problem in cancer research is identifying the cell type that is capable of sustaining neoplastic growth and its origin from normal tissue cells. Recent investigations of a variety of tumor types have shown that phenotypically identifiable and isolable subfractions of cells possess the tumor-forming ability. In the present paper, using two lineage-related human melanoma cell lines, primary melanoma line IGR39 and its metastatic derivative line IGR37, two main observations are reported. The first one is the first phenotypic evidence to support the origin of melanoma cancer stem cells (CSCs) from mutated tissue-specific stem cells; and the second one is the identification of a more aggressive subpopulation of CSCs in melanoma that are CXCR6+.Methods/FindingsWe defined CXCR6 as a new biomarker for tissue-specific stem cell asymmetric self-renewal. Thus, the relationship between melanoma formation and ABCG2 and CXCR6 expression was investigated. Consistent with their non-metastatic character, unsorted IGR39 cells formed significantly smaller tumors than unsorted IGR37 cells. In addition, ABCG2+ cells produced tumors that had a 2-fold greater mass than tumors produced by unsorted cells or ABCG2- cells. CXCR6+ cells produced more aggressive tumors. CXCR6 identifies a more discrete subpopulation of cultured human melanoma cells with a more aggressive MCSC phenotype than cells selected on the basis of the ABCG2+ phenotype alone.Conclusions/SignificanceThe association of a more aggressive tumor phenotype with asymmetric self-renewal phenotype reveals a previously unrecognized aspect of tumor cell physiology. Namely, the retention of some tissue-specific stem cell attributes, like the ability to asymmetrically self-renew, impacts the natural history of human tumor development. Knowledge of this new aspect of tumor development and progression may provide new targets for cancer prevention and treatment.

The stable expression of reporter genes in adult stem cells (ASCs) has important applications in stem cell biology. The ability to integrate a noncytotoxic, fluorescent reporter gene into the genome of ASCs with the capability to track ASCs and their progeny is particularly desirable for transplantation studies. The use of fluorescent proteins has greatly aided the investigations of protein and cell function on short-time scales. In contrast, the obtainment of stably expressing cell strains with low variability in expression for studies on longer-time scales is often problematic. We show that this difficulty is partly due to the cytotoxicity of a commonly used reporter, green fluorescent protein (GFP). To avoid GFP-specific toxicity effects during attempts to stably mark a rat hepatic ASC strain and, therefore, obtain stable, long-term fluorescent ASCs, we evaluated cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP), in addition to GFP. Although we were unable to derive stable GFP-expressing strains, stable fluorescent clones (up to 140 doublings) expressing either CFP or YFP were established. When fluorescently marked ASCs were induced to produce differentiated progeny cells, stable fluorescence expression was maintained. This property is essential for studies that track fluorescently marked ASCs and their differentiated progeny in transplantation studies.

Enzymes are commonly used as a biochemical means to liberate cells from a host of tissues for use in in vitro studies and/or in vivo transplantations. However, very little understanding exists of the biological and functional effects that enzymes have on cells during the process of releasing the native cells from a given tissue. One specific reason for this is that no technology has existed as a nonenzymatic control to compare baseline biology and function for a given processed tissue. We have developed a sterile, onetime use, disposable system (referred to as the AuxoCell Processing System or AC:Px®) that allows for processing of solid tissue in a closed, standardized system using mechanical means to liberate cells without the need and/or use of any biochemical, enzymatic digestion. In this report, for the first time, we directly compare the cellular outputs derived from processing the same umbilical cord tissue (UCT) in the presence and absence of collagenase. In the presence of collagenase, we observed on average, approximately a 2.7-fold reduction in native mesenchymal stem/stromal cell (MSC) yields and a reduction in MSC-specific markers CD90, CD29, CD105, CD73, CD44, CD36, CD49b, CD49a, CD146, CD295, and CD166 and in endothelial marker CD31. These data directly exhibit that the use of collagenase to process UCT to release cells impacts cell recovery with respect to number and cell surface marker expression and, hence, could affect the in vivo function of the recovered native cellular population.

No abstract