Cancers originally develop from normal cells that gain the ability to proliferate aberrantly and eventually turn malignant. These cancerous cells then grow clonally into tumors and eventually have the potential to metastasize. A central question in cancer biology is, which cells can be transformed to form tumors? Recent studies elucidated the presence of cancer stem cells that have the exclusive ability to regenerate tumors. These cancer stem cells share many characteristics with normal stem cells, including self-renewal and differentiation. With the growing evidence that cancer stem cells exist in a wide array of tumors, it is becoming increasingly important to understand the molecular mechanisms that regulate self-renewal and differentiation because corruption of genes involved in these pathways likely participates in tumor growth. This new paradigm of oncogenesis has been validated in a growing list of tumors. Studies of normal and cancer stem cells from the same tissue have shed light on the ontogeny of tumors. That signaling pathways such as Bmi1 and Wnt have similar effects in normal and cancer stem cell self-renewal suggests that common molecular pathways regulate both populations. Understanding the biology of cancer stem cells will contribute to the identification of molecular targets important for future therapies.

It is becoming recognized that screening of oncology drugs on a platform using two-dimensionally (2D)-cultured cell lines is unable to precisely select clinically active drugs; therefore three-dimensional (3D)-culture systems are emerging and show potential for better simulating the in vivo tumor microenvironment. The purpose of this study was to reveal the differential effects of chemotherapeutic drugs between 2D- and 3D-cultures and to explore their underlying mechanisms. We evaluated differences between 2D- and 3D-cultured breast cancer cell lines by assessing drug sensitivity, oxygen status and expression of Ki-67 and caspases. Three cell lines (BT-549, BT-474 and T-47D) developed dense multicellular spheroids (MCSs) in 3D-culture, and showed greater resistance to paclitaxel and doxorubicin compared to the 2D-cultured cells. An additional three cell lines (MCF-7, HCC-1954, and MDA-MB‑231) developed only loose MCSs in 3D, and showed drug sensitivities similar to those found in the 2D-culture. Treatment with paclitaxel resulted in greater increases in cleaved-PARP expression in the 2D-culture compared with the 3D-culture, but only in cell lines forming dense 3D-MCSs, suggesting that MCS formation protected the cells from paclitaxel-induced apoptosis. Hypoxia was observed only in the dense 3D-MCSs. BT-549 had fewer cells positive for Ki-67 in 3D- than in 2D-culture, suggesting that the greater G0-dormant subpopulation was responsible for its drug resistance in the 3D-culture. BT-474 had a lower level of caspase-3 in the 3D- than in the 2D-culture, suggesting that the 3D-environment was anti-apoptotic. Finally, we compared staining for Ki-67 and caspases in the 2D- and 3D-primary‑cultured cells originating from a patient-derived xenograft (PDX), fresh PDX tumor, and the patient's original tumor; 2D-cultured cells showed greater proportions of Ki-67-positive and caspase-3-positive cells, in agreement with the view that 3D-primary culture better represents characteristics of tumors in vivo. In conclusion, 3D-cultured cells forming dense MCSs may be better than 2D-cultured cells in simulating important tumor characteristics in vivo, namely hypoxia, dormancy, anti-apoptotic features and their resulting drug resistance.

Not all cells in a tumor are alike, but our ability to characterize cancer heterogeneity in detail has been limited. Dalerba et al. use high-throughput single-cell expression analysis to define clinically relevant subpopulations in normal and cancerous colon tissue. Cancer is often viewed as a caricature of normal developmental processes, but the extent to which its cellular heterogeneity truly recapitulates multilineage differentiation processes of normal tissues remains unknown. Here we implement single-cell PCR gene-expression analysis to dissect the cellular composition of primary human normal colon and colon cancer epithelia. We show that human colon cancer tissues contain distinct cell populations whose transcriptional identities mirror those of the different cellular lineages of normal colon. By creating monoclonal tumor xenografts from injection of a single (n = 1) cell, we demonstrate that the transcriptional diversity of cancer tissues is largely explained by in vivo multilineage differentiation and not only by clonal genetic heterogeneity. Finally, we show that the different gene-expression programs linked to multilineage differentiation are strongly associated with patient survival. We develop two-gene classifier systems (KRT20 versus CA1, MS4A12, CD177, SLC26A3) that predict clinical outcomes with hazard ratios superior to those of pathological grade and comparable to those of microarray-derived multigene expression signatures.

To examine the role of breast cancer stem cells (BCSCs) in metastasis, we generated human-in-mouse breast cancer orthotopic models using patient tumor specimens, labeled with optical reporter fusion genes. These models recapitulate human cancer features not captured with previous models, including spontaneous metastasis in particular, and provide a useful platform for studies of breast tumor initiation and progression. With noninvasive imaging approaches, as few as 10 cells of stably labeled BCSCs could be tracked in vivo, enabling studies of early tumor growth and spontaneous metastasis. These advances in BCSC imaging revealed that CD44⁺cells from both primary tumors and lung metastases are highly enriched for tumor-initiating cells. Our metastatic cancer models, combined with noninvasive imaging techniques, constitute an integrated approach that could be applied to dissect the molecular mechanisms underlying the dissemination of metastatic CSCs (MCSCs) and to explore therapeutic strategies targeting MCSCs in general or to evaluate individual patient tumor cells and predict response to therapy.