Cancer and embryonic stem cells share genetic fingerprints

Nature Reports Stem Cells ^ | 17 April 2008 | Monya Baker

[neverdem]

At least two modules of genes promote stemness

Two separate groups of genes seem able to promote stemness. And while most adult tissue stem cells share one program, embryonic stem cells and at least some human cancer cells share another. In fact, one well-known oncogene, myc, was able to both initiate the embryonic stem cell module and generate cancer stem cells. These findings, reported recently by researchers at Stanford University in California and the Weizmann Institute of Science in Israel, support the still-controversial hypothesis that embryonic stem cells and at least some cancer stem cells share a common genetic program.

Surveys searching for genes common to various sorts of stem cells have been stymied by noisy expression data. Howard Chang and colleagues thought they could get above this noise by analyzing genes in sets or 'modules' instead of individually1. Researchers started with 102 genome-wide expression profiles of mouse stem cells (including embryonic, neural, hair bulge, haematopoietic, retinal and neural crest) and differentiated cells. Each cell type was represented by at least two independent expression profiles. They compiled over 3,000 sets of genes based on known stem cell gene expression signatures, plus genes that are bound or otherwise regulated by stem cell regulatory proteins such as Oct4 and Sox2. Adult tissue stem cells fell into two groups: a cluster of bone marrow haematopoietic, neural crest, hair bulge and mammary stem cells; and a cluster of embryonic, retinal, fetal liver haematopoietic and (at least in data from three out of four studies) neural stem cells.

The team repeated their work using a compendium of data on human embryonic stem (ES) and differentiated cells, which indicated that an ES cell signature was evolutionarily conserved. The 335 genes shared between the mouse and human modules were designated as the core ES cell–like gene module.

Next, the researchers looked to see whether either of the modules could be identified in human cancer samples and found that the ES cell module was active in a variety of them. In a sample of 295 primary breast cancers, greater ES cell module activity meant less differentiation and greater rates of patient death than cancers with less activity.

Further, unbiased analysis found that many of the genes in this module had regulatory elements with which to bind the protein c-Myc, and that c-Myc, but not other oncoproteins, could induce expression of the ES cell gene module. This protein is already famous as being one of the four reprogramming factors that can transform differentiated cells to an embryonic-like state. Wong et al. found that activating c-Myc not only induced this gene module but also induced epithelial tumour-initiating cells that readily generated tumours when injected into mice without immune systems. These cells also caused tumours when transplanted to a second set of recipient mice.

Practically, this means that epithelial cancer stem cells can potentially be generated at will rather than sorted from heterogeneous sources. Theoretically, this module map not only confirms that no single module controls stemness, but it also points to different stem cell classes being governed by differing molecular networks that can be initiated by a single gene, at least for the ES program.

Related articles Repressing microRNAs for pluripotency

Reference Wong, D. J. et al. Module map of stem cell genes guides creation of epithelial cancer stem cells. Cell Stem Cell 2, 333–344 (2008).

Author affiliation Monya Baker is news editor of Nature Reports Stem Cells.

[neverdem]

USC stem cell study sheds new light on cell mechanism

[Dr. Scarpetta]

BTTT

[aruanan]

While writing my thesis on a particular neural cell adhesion molecule, I remember reading that loss of adhesion to the ECM can cause some cells to dedifferentiate, leading to carcinogenesis.

[neverdem]

What’s the ECM?

[aruanan]

What’s the ECM?

Extracellular matrix. Cells connect to this via a variety of cell adhesion molecules. The types and locations of attachments can provide positional feedback into the cell that can have signaling and even mechanical transduction capability on nuclear expression. This can affect what the cell "thinks" it is and the way it behaves (or misbehaves). In some transformed cells (which could be precursors to cancerous growth), reestablishment of cell adhesion can cause the cell to redifferentiate and revert back to its old self, so to speak.

[neverdem]

Thanks for the feedback. Leukocyte Adhesion Deficiency is an interesting subject.