One of the most significant challenges in controlling cancer is the high degree of heterogeneity within the neoplastic cells of individual tumors. While tumor cell heterogeneity has been described at many levels and an understanding of the processes that create heterogeneity is emerging, the extent to which the heterogeneous subpopulations of tumor cells vary in their functional activities and whether interactions between subpopulation influence tumor initiation and progression is poorly understood. Major barriers to investigating these questions have been the lack of methodologies to maintain the heterogeneity of human tumors in culture and to propagate distinct clonal subpopulations from individual tumors. Together with our collaborators, we have developed methodologies that overcome these barriers and make it feasible to isolate, culture, and characterize clonal populations of ovarian tumor cells and then track barcoded, epitope-tagged clonal populations within tumor xenografts generated from mixtures of transplanted clones. Using these approaches in a pilot study, we obtained evidence for interclonal collaborations that significantly affect tumor progression -- most interestingly, we found that clonal populations of cancer cells which are unable to initiate tumor formation on their own, can strongly promote the expansion of primary tumors and metastasis at specific organ sites when mixed with 'tumor-initiating'clones. This supports the hypothesis that clonal subpopulations within a tumor cooperate with one another to promote tumor expansion and metastasis. In this proposal, we describe plans to use a systematic approach to test this hypothesis in human ovarian tumors by (1) isolating and characterizing the extent of genetic and phenotypic variation among a large set of clonal populations derived from ovarian tumors, (2) comparing the activities of single clonal populations to mixtures of such populations in order to investigate the existence and nature of intratumoral cross talk, (3) tracking intratumoral localization of clonal populations over time, and (4) elucidating the mechanisms responsible for phenotypes generated by intratumoral crosstalk.
The information derived from these studies will provide important insights into the nature of cooperative interactions between tumor cell populations and mutual dependencies among tumor cell variants and reveal how these affect tumor expansion or metastasis. This approach provides a platform to investigate a wide variety of phenotypes that are influenced by intratumoral crosstalk, and information relevant to diagnosis, prognosis and treatment of cancers.