Letters

A new type of cancer cell growth

Re: cancer stem cells.1,2 Recently we have reported a novel type of cell division involved in the origin and growth of cancers.3,4 Termed neosis, this type of cell division occurs only in senescent polyploid giant cells and never in normal diploid cells. Up to 10% of tumor cells both in vitro and in vivo are polyploid, and so far there is no explanation of their role in cancer. These resemble senescent cells, which are thought to be part of the tumor suppressor mechanism.

We have shown that such cells have he potential to undergo neosis, a parasexual, somatic reduction division characterized by karyokinesis via nuclear budding followed by asymmetric cytokinesis, giving rise to aneuploid or near diploid daughter cells termed Raju cells (Raju means king or leader in the Telugu language). The latter cells are unique in that they transiently display stem cell properties, have inherited genomic instability, have the potential to differentiate into tumor cells and have extended, but, limited mitotic division potential. At the end of their mitotic life span, they repeat the cycle of senescence, neosis and production of the next generation of Raju cells, which repeat the same cycle of events several times through the life of the tumor. When we expose tumor cells to chemotherapeutic agents or radiation, they undergo premature senescence, and some of them escape senescence via neosis, and produce Raju cells that are resistance to genotoxis.

Thus, neosis is the escape mechanism from senescence that gives rise to tumor cells which may become resistant to therapy and that yields the aneuploid cells characteristic of solid tumors. The discovery of neosis therefore questions the basic principles of our current concept of cancer, namely that cancers arise via mitosis, that cancer cells are immortal, and that cancer cell continuity is due to the unlimited asymmetric mitotic division potential of mutant stem cells or cancer stem cells. We think instead that cancer arises via neosis, cancer cells are not immortal but undergo repeated senescent phases, and that tumor cell continuity is due to the escape from senescent phase, since tumor cells carry mutant genes in the senescence checkpoint pathway.

Our data supports the view that genesis and regenesis of Raju cells via repetitive neotic divisions is responsible for the origin and continuous growth of different types of tumors. This accommodates the role of telomerase in tumor growth, explains a role for the ectopic expression of meiotic genes in tumor cells, and also for the first time explains the role of senescent cells that are found in tumors in vivo and in vitro. We invite others to verify our findings.

Rengaswami Rajaraman

Dalhousie University
Halifax, Nova Scotia, Canada
R.Rajaraman@Dal.Ca

References

1. I. Weissman, M. Clarke, "Leukemia and cancer stem cells," The Scientist, 20(4):35, April 2006.2. P. Dirks, "Stem cells for brain cancer," The Scientist, 20(4):37, April 2006.3. M. Sundaram et al., "Neosis: a novel type of cell division in cancer," Cancer Biol Ther, 3:217-8 2004.4. R. Rajaraman et al., "Neosis - a paradigm of self-renewal in cancer," Cell Biol Int 29:1084-97, 2005.

The problem with microarrays

Re: "In search of microarray standards."1 The root cause of problems with DNA microarrays is the complexity of cDNA probes used. Perhaps microarray practioners, more than 10 years since the invention of the method, could finally perform the two simple and real control experiments that will unmask the magic of DNA microarrays once for all: label just one, instead of tens of thousands of cDNA at a time by reverse transcriptase to see if an abundantly expressed gene (e.g., actin) can be uniquely detected, and a low abundance transcript can be detected at all on a "chip" that contains tens of thousands of gene probes.

Peng Liang
Vanderbilt-Ingram Cancer Center
Nashville, Tenn.
peng.liang@vanderbilt.edu

References

1. J. Perkel, "In search of micrarray standards," The Scientist, 20(3):73, March 2006.2. P. Liang, A.B. Pardee. "Analysing differential gene expression in cancer," Nat Review Cancer, 3:869-76, 2003.

The interactome: Next steps

Re: "Time for a Human Interactome Project?"1 The possibility of organizing proteomic data in a manner which would allow working scientists to rapidly evaluate genes and gene products would be of great value. Unfortunately, the interactome data don't provide such a biologic context because they don't reflect the underlying cellular and molecular biologic processes. By looking at the interactome as a homogeneous landscape, you miss the principles of cell-cell interactions that determine development, homeostasis and repair.

We have suggested a comparative functional genomic approach to the problem of identifying gene regulatory networks, which would then be amenable to '-omics.' By determining ligand/receptor signaling pathways common to development, homeostasis and repair across phyla2 reveals the first principles of physiology.3 This is a more challenging approach, but one which will provide biological relevance for genomic and proteomic data in the future.

John S. Torday
Harbor-UCLA Medical Center Campus
Los Angeles
jtorday@labiomed.org

References

1. M. Vidal, "Time for a human interactome project?" The Scientist, 20(3): 46-9, March 2006.2. J.S. Torday, V.K. Rehan. "Deconvoluting lung evolution using functional/comparative genomics," Am J Respir Cell Mol Biol, 31:8-12, 2004.3. J.S. Torday, "A Periodic table for biology." The Scientist, 18:32-3, June 2004.