Fogal et al (2010) found that the p32 gene (on human chromosome 17q13.3), which was overexpressed in some cancer cells, had actually promoted the level of oxidative phosphorylation (OXPHOS) in mitochondria. The knockdown of p32 in an experiment then lead to a lower level of complexes III, IV and V composing the electron transport chain (ETC) of OXPHOS, thus making a shift in ATP synthesis from OXPHOS to glycolysis in tumor cells, but meanwhile causing a lower level of tumor growth than before. 

This is contradictory with the well known Warburg Effect (Warburg 1924), i.e. an elevated level of glycolysis and glucose consumption as a hallmark of tumor growth, hypothesized to provide a growth advantage for the tumor cells.

However, another gene p53 (on human chromosome 17p13.1), also promoting OXPHOS, is a well known tumor suppressor. p53 could inhance expression of Cytochrome c Oxidase II (also a part of complex IV of ETC), which is essential for OXPHOS in mitochondria. p53 is found to have mutated in many cancer cells, causing a shift from OXPHOS to glycolysis (Matoba 2006) during tumor growth. This is in turn consistent with Warburg Effect.

As the result we see two genes located on the same chromosome regulating the balance between OXPHOS and glycolysis in the same way. However, they seem to play opposite roles in carcinogenesis. So why are they so different?

One possible reason may lie in their roles in inducing apoptosis. Over-expression of p32 could induce apoptosis only when p53 functions in normal status (Itahana & Zhang 2008). So once p53 is disfunctional in cancer cells as said above, the overexpression of p32 won't cause apoptosis alone, and thus won't give any disadvantage against the tumor cells. On the other hand, overexpression of p32 could produce ATPs for tumor cells in a higher efficiency. In such a hypothesis p32 would not be an oncogene, but is only overexpressed as a consequence of carcinogenesis. And in such a case, the Warburg Effect is not rejected but irrelevant to the mechanisms here.

Anyway, the Warburg Effect has been questioned more than once (Weinhouse et al 1956; Zu & Guppy 2004; Dang 2010). Although the inhibition of OXPHOS and promotion of glycolysis have been correlated to carcinogenesis either as a cause or as a consequence in numerous studies throughout the last 80 years (too many literatures), the underlying mechanisms seem still unsolved. And it is still possible to answer the above question in the context of metabolism regulation based on the framework raised by Warburg.

Many studies have been proposing an evolutionary perspective onto the correlation between ATP synthesis and carcinogenesis (e.g. Gatenby & Vincent 2003; Pfeiffer & Schuster 2005; Vincent 2006), by considering the tumor/normal cells within the same tissue/organ as a population, in which individual cells compete with each other in a series of cell generations within the life span of the human body. Such a micro-evolution process could be investigated with methods from population genetics, adaptation dynamics, theories of competition and coexistence, etc. These Darwinist have provided interesting viewpoints and they never forgot about the important roles of mitochondrial functions and mtDNA mutations in tumor growth. However, they seldom considered the cooperation between the mitochondrial genome and the nuclear genome, as de Bivort et al (2007) did in their effort to correlate such an coevolutionary force behind ATP synthesis with the progression of some mitochondrial diseases.

It is known that many proteins and enzymes involved in mitochondrial functions, including complexes I, III, IV and V of the ETC, are composed of both mtDNA-encoded and nDNA-encoded subunits (Wallace 2005). Interestingly, both p32 and p53 could regulate complex IV, but not complex II, which is encoded solely by nDNA. Considering that the two genomes belong to different hierarchies of life forms, some delicate cooperation mechanisms may have evolved to keep them match in a cell. Such mechanisms could be vulnerable to novel influences in the modern world, either environmental or physical, causing cyto-nuclear conflict. It is worth including such cyto-nuclear mismatch/incompatibility patterns when constructing an evolutionary model to answer the above question.

Vital references

• Fogal, V., Richardson, A. D., Karmali, P. P., Scheffler, I. E., Smith, J. W., & Ruoslahti, E. 2010. Mitochondrial p32 protein is a critical regulator of tumor metabolism via maintenance of oxidative phosphorylation. Molecular and Cellular Biology 30: 1303-1318.
• Matoba, S., Kang, J., Patino, W. D., Wragg, A., Boehm, M., Gavrilova, O., Hurley, P. J., Bunz, F., & Hwang, P. M. 2006. p53 regulates mitochondrial respiration. Science 312: 1650-1653.
• de Bivort, B. L., Chen, C., Perretti, F., Negro, G., Philip, T. M., & Bar-Yam, Y. 2007. Metabolic implications for the mechanism of mitochondrial endosymbiosis and human hereditary disorders. Journal of Theoretical Biology 248: 26-36.