As a follow-up to yesterday's post, F. Hlubek et al. from the University of Munich presented during last week's AACR Annual Meeting a poster session (Abstract #2152) results of their study on mutations in the Wnt signaling pathway. These mutations result in differential expression of the β-catenin target genes which have been associated with tumor progression and metastasis in colorectal adenocarcinomas. The investigators noted that in most aggressive forms of colorectal cancers there is a loss of epithelial differentiation and an increase in the mesenchymal phenotype of the tumor cells. This EMT in colorectal tumors is believed to be the result of mutations in the APC (Adenmatosis polyposis coli) tumor suppressor. APC mutations give rise to an accumulation of nuclear β-catenins concomitant with increased expression of its target genes. However, as the tumor progresses there is greater heterogeneity among the tumor cells along with a relocation of the β-catenins to the membrane and cytoplasm as the tumor becomes more polarized and metastatic. With large-scale microarray analyses, the investigators found differential expression of the β-catenin target genes in which overexpression of the genes are associated with both tumor growth and invasive tumor progression. The results also showed two distinct expression profiles for cells comprising the invasive front of the tumor in contrast to the cells found within the central tumor. The study results also showed that genes encoding for lamin-5γ2 are overexpressed in cells located at the invasive front, but not in cells in the central tumor. The authors concluded that mutations in the Wnt signaling pathway results in differential expression of β-catenin target genes that lead to intratumorous heterogeneity for colorectal cancers.,

During the AACR Annual Meeting on April 17th , a forum was held to discuss the translational research on cancer stem cells and the potential to correlate their presence in a tumor to the clinical status of the patient. Ruggero De Maria, M.D. from Fondazione IOM, Viagrande, Italy presented his work on characterizing cancer stem cells in solid tumors. With a cell suspension dissociated from solid human tumors, De Maria and his research group identified a subpopulation of cells within the tumors that were CD133+. The initial studies were conducted on thyroid tumors in which the CD133+ were found to have unlimited proliferative potential to self-renew. Xenografts by these cells in SCID mice demonstrated their tumorigenic properties concomitant with their potential to differentiate into different tissue lineages both in vitro and in vivo, thus satisfying the current criteria for defining a cancer stem cell. In vitro expansion of the CD133+ cells was established through the formation of spheres from a cell suspension derived from dissociated tumors which were subsequently cultured in a serum-free media supplemented with EGF and FGF-2. The investigator also presented data showing that sphere-forming CD133+ cells were more tumorigenic than the nonsphere-forming cells. Similarly, De Maria's group also found that CD133+ cells from tumors of both lung and brain (glioblastoma) produced the same results. In glioblastomas, they found that the number of CD133+ cells isolated from the tumor correlated with prognosis, i.e. the presence of higher numbers of CD133+ cells was predictive of a poor prognosis for the patient with that particular tumor.

In yesterday's poster session at AACR Annual Meeting, X-N Li et al. from the Baylor College of Medicine reported (Abstract # 3888) the results of their study on isolating 18 primary pediatric brain tumors of medulloblastoma (MB) and glioblastoma multiforme (GBM) and forming xenografts in SCID mice. Among the transplants, 54% MB (7/13) and 75% (3/4) formed xenografts. Immunohistochemical analysis with the antibody to Ki-67 was used to analyze the xenografts. The investigators demonstrated that they were able to serially passage the human tumors (IC) with 10k cells. Survival in animals receiving the passaged cells was directly correlated with the number of the cells that were transplanted. Within the xenografts, the investigators identified a subpopulation of cells that were positive for the neural stem cell marker CD133 which actively effluxed the Hoescht 33342 dye (ruling out the presence of a side population). In subsequent passages (p5), the presence of neural stem cells in the xenografts were confirmed by their ability to form neurospheres and the capacity to differentiate into various neuronal lineages by immunohistochemical staining of lineage specific proteins in the differentiated cells. The investigators concluded that their xenograft animal model "should provide an avenue for biological study and preclinical drug screening of MBs and GBMs.

For the remainder of the week, we will be posting papers and symposia summaries related to cancer and stem cells from the AACR Annual Meeting being held in Los Angeles from April 14th to the 18th. In a period of only two years, there has been a tremendous amount of research being conducted in isolating, characterizing, and determining the tumorigenic potential of these putative cancer stem cells. For example in a yesterday's poster session, University of Melbourne's B. I. Eckhardt et al. reported (Abstract #2177) on bone morphogenic protein (BMP) as novel suppressors of metastasis in breast cancers. This finding resulted from a study of array profiles of mouse breast tumor cell lines and correlating gene expression to metastasis. The investigators found that a higher percentage of genes that encode proteins for cell signaling and extracellular matrix in metastatic tumors. Conversely, BMP-4, a cytokine for tissue patterning and morphogenesis, was down-regulated in these highly metastatic tumor lines (e.g. 4T1.2 cell lines). Overexpression of BMP-4 in 4T1.2 cells (4T1.2NB-BMP4) resulted in a loss of metastasis when the transfected cells were implanted into BALB/c recipients concomitant with enhanced survival of the tumor-bearing mice. Additionally, administration (IP) of recombinant BMP-4 enhanced survival of mice with the 4T1.2 tumors. The investigators noted that a possible mechanism for this loss in metastasis may be the result of activation of Smad 7, a gene associated with suppressing metastasis in breast cancers. The investigators concluded from the results of their study that BMPs may have a novel role as suppressors of breast cancer metastasis. (Perhaps, BMP may drive the putative cancer stem cell to a more differentiated phenotype with a loss in the capacity to metastasize.)

In a poster session at yesterday's American Association for Cancer Research (AACR) held in Los Angeles, California, E. E. Bar et al. from Johns Hopkins University School of Medicine presented data (Abstract #1327) on the effect of a Hedgehog (Hh) inhibitor, cyclopamine, on both cell lines and fresh tissue isolates of glioblastoma multiforme (GBM) expressing Gli1, Ptch, and Smo. Cyclopamine is known to inhibit the expression of the Gli1 expression---a marker for GBM. The investigators found that cyclopamine inhibit the growth of GBM-derived cells in colony-forming in vitro assays in a dose-dependent manner. During neurosphere formation, mRNA expression of Shh and Gli1 mRNA are up-regulated from GBM-derived cells. However, in the presence of cyclopamine, Gli1 mRNA expression is down-regulated (in 3mM concentration of drug, 50-70% reduction in expression) with up-regulation with GFAP expression and a reduction of nestin and abrogration of the ability to form neurospheres...suggesting differentiation of the cancer stem cells by the drug. Since the Hedgehog pathway maintains "stem cellness," inhibition of the pathway by cyclopamine was shown to reduce the expression of Ki67 and CD133 as markers for neural stem cell progenitors. The investigators also demonstrated primary cell cultures pretreated with cyclopamine lose their ability to engraft and initiate tumors in the mouse recipients. In contrast, untreated cells resulted in massive infiltration of the glioma and 100% death 2 months following injection of the cells. The investigators concluded that cyclopamine targets cancer stem cells in GBM as well as a side population of GBM (Aldefluor+ population) and eliminates the clonogenic potential of GBM-derived neurospheres.

In the March 1st online edition of Stem Cells, a research team from the National Centre for Cell Science at Pune, India, A. Shiras et al., published the results of their study on genomic instability of glioma cells producing cancer stem cells. From human adult glioma tissues, the investigators isolated and propagated in vitro CD133 positive neural stem cells (NSCs). Replicative senescence spontaneously generated an immortalized clone with properties reflecting a cancer stem cell (CSC). Thus, the cells were CD133+ which were reflected in their ability to self-renew, form neurospheres, and develop intracranial tumors. The data also provided evidence for the role of genomic instability as the underlying cause for the transformed state as well as inducing these neural stem cells to progress to highly tumorigenic subpopulation of CSCs. It was also suggested that significant overexpression of Notch and Hes transcription factor in the brain tumor stem cells may result in the transformed state and tumorigenicity.