Monday, July 30, 2007

Duesberg's new hypothesis

As late as 2003 Peter Duesberg, discoverer of the src oncogene, has disputed that the HIV virus is the causal agent of AIDS, the "Duesberg hypothesis" apparently it is caused by drug use and other non-contagious factors. Propagation of this hypothesis, assuming it is incorrect, arguably costs lives. A good blog that debates this hypothesis is He was in some good company with this hypothesis, which includes Kary Mullis, the surfing inventor of the polymerase chain reaction.
From a forward by Kary Mullis: (I think there is a whole chapter on this in his book "Dancing naked in the mine field.)
"So I turned to the virologist at the next desk, a reliable and competent fellow, and asked him for the reference. He said I didn't need one. I disagreed. While it's true that certain scientific discov eries or techniques are so well established that their sources are no longer referenced in the contemporary literature, that didn't seem to be the case with the HIV/AIDS connection. It was totally remarkable to me that the individual who had discovered the cause of a deadly and as-yet-uncured disease would not be con tinually referenced in the scientific papers until that disease was cured and forgotten. But as I would soon learn, the name of that individual - who would surely be Nobel material - was on the tip of no one's tongue."
Kenny G (not the saxophonist but a knowledgeable lab rat), informed me that Duesberg is now debunking the idea that mutations initiate cancer. Instead he proposes aneuploidy is the cause.
"Conventional genetic theories have failed to explain why cancer (1) is not found in newborns and thus not heritable; (2) develops only years to decades after 'initiation' by carcinogens; (3) is caused by non-mutagenic carcinogens; (4) is chromosomally and phenotypically 'unstable'; (5) carries cancer-specific aneuploidies; (6) evolves polygenic phenotypes; (7) nonselective phenotypes such as multidrug resistance, metastasis or affinity for non-native sites and 'immortality' that is not necessary for tumorigenesis; (8) contains no carcinogenic mutations."
I actually don't know enough to properly analyze the credibility of this hypothesis, but the history of the scientist immediately makes me skeptical. Is he just good at cherry-picking the literature or is he on to something?
It's good to have people criticize the science, and at least this time he's not potentially costing lives. I wonder why or if he's stopped focusing on the HIV/AIDS issue?


nosugrefneb said...

Whoa. Aneuploidy as the cause of cancer? How risky.

Sure, aneuploidy happens all the time. Probably most of the time. But to completely debunk the association between all mutations and the development of cancer is ludicrous. There is plenty of data - correlative, functional, you name it - to support that association.

Admittedly, which typically comes first - aneuploidy or mutations - is a very interesting one. Certainly there is evidence of cancers with aneuploidy in the absence of other genetic alterations, but there are also those with mutations in the context of euploidy. So, the answer to which comes first is probably most likely: either. Of course, certain mutations can lead to aneuploidy, and aneuploidy leads to mutations, which complicates matters to say the least. One goldmine here is to focus on preneoplastic lesions (Barrett's esophagus, for example).

rob said...

The problem this presents is that given a sample of cancer prove that it has no chromosomal abnormalities. It may be a very small, difficult to detect translocation or something. Therefore by laying the burden of proof of complete chromosomal normality, it is difficult to disprove. No??

rob said...

Umm I just read what I wrote and I realize I should be proof reading my comments.
Basically I'm saying I think that his argument is difficult to disprove because it would be hard to prove complete chromosomal normality.

Anonymous Coward said...

You've got it the wrong way around. Insert mutations in your cell sytem and see if you get transformation, thus proving the mutation comes before aneuploidy.

rob said...

Proving it in your system not in a genuine tumour. I agree, obviously a collection of mutations can cause transformation. I don't think that's in dispute. Transformation is different than being able to form tumours, no?

Bayman said...

Plus there is the scenario where an underlying mutation causes chromosomal abnormalities, which in turn lead to cancer. For example, a mutation in a gene causing disregulation of the cell cycle or mitosis (this would include pretty much any gene implicated in cancer) is likely to lead to missegregation of chromosomes and chromosomal instability, creating a fertile ground for the creation of mutant daughter cells and selection of cancerous ones. This principle has been demonstrated by disruption of genes involved in maintaining the asymmetry of drosophila neural stem cell division.

Yes, both mutations and chromosomal instability underlie cancer; chromosomes do nothing in themselves except carry genetic information, so damage (mutation) to a gene(s) will always be required for a change in phenotype. Chromosomal instabilities and abnormalities cannot directly cause a change in phenotype, but provide fertile grounds for the production of a "library" of messed up, mutant cells upon which natural selection acts to yield cancerous cells.

I think Duesberg is merely reminding us, in this reductionist gene-centric age, to look at the big picture, where it may be more useful to focus our thinking from the broader perspective of cancer as a chromosomal disease.

The question is, is the frequency of random, environmentally-induced point mutations in human cells high enough to make it likely that enough will accumulate in the proper combination of genes, in the same cell, of the right lineage to lead to a viable cancer cell lineage? Or rather, is it more likely that only one or a small number of certain inherited or somatic mutations cause chromosomal instability that serves as an underlying "variability generating machinery" that generate not just point mutations, but gross chromosomal translocations, fusions of different pieces of genes and ORFs etc?...This would increase the probability of creating a "successful" cancer cell - it seems more like this type of variation that would be required to give rise to cell lineages with the right combination of unlikely genetic lesions needed to make a cancerous cell.

With regards to Duesberg's "dangerous" stance on HIV-AIDS, I think that orthodoxy and mainstream ideas are much more dangerous - particularly if they should happen to be wrong or misguided - because governments businesses, individuals, scientists, etc. actually use them to guide policy, business and culture. Minority dissenting opinions are not dangerous whatsoever, and in fact they are the only way to critically assess and thereby validate or improve the mainstream consensus. In fact, it is much more dangerous to suggest that dissent is dangerous than to label proven scientists as killers because they question the science behind the HIV-AIDS link (kind of like calling war dissidents unpatriotic and unsupportive of the troops).

Anonymous Coward said...

Actually dissenting opinion within science is healthy and necessary, but among policy makers it is very dangerous. The public has a hard time understanding why science doesn't work by consensus, and tend to value opposing opinions equally. Case in fact non-HIV theories of AIDS has caused many deaths in African countries, and opposing views on climate research, stem cell research and evolution is used to drive the republican agenda in the States, never mind that they are on the fringe of science and not supported by as much evidence as their more sound alternatives...

Bayman said...

True. Maybe we should distinguish between intentional ignorance and challenging the scientific facts - questioning the science in a rational manner is healthy - irrational ignorance (ie AIDS is not caused by HIV because it is the wrath of God, or westerners trying to poison developing countries)is certainly damaging.

nosugrefneb said...

Was at a talk yesterday given by George Vande Woude, and he seems to be of similar opinion as Duesberg, if ever so slightly. He showed a lot of data regarding the correlation between extent of aneuploidy and specific transformative phenotypes like proliferation, invasion, and concurrent proliferation and invasion. Also, these can be manipulated only via chromosome number to take a cell from proliferator to invader and back to proliferator in sites of micrometastasis.

Of course, questions came up in the audience about the significance of mutations in this system, and whether they preceded aneuploidy, about which he was extremely dodgey. Clearly, this is a huge question and even the biggest names in cancer research don't have much of a clue.

Bayman said...

Very cool. Thanks for the scoop. First I heard of the work, but I'm reading his PNAS paper right now.

Anonymous said...

"and opposing views on climate research, stem cell research and evolution is used to drive the republican agenda in the States, never mind that they are on the fringe of science and not supported by as much evidence as their more sound alternatives..."

This exposes you as a moron. Anthropogenic global warming is proven to be a scam, it has zero science behind it, zero evidence. CO2 lags behind climate change by 800 years, it has cooled for the past 9 years. Solar/Earth magnetic field axial orientation, distance and interaction with Lunar gravity is proven to be the driver by empirical evidence and predictions by Piers Corbyn. There's a mountain of evidence against AGW. Educate yourself.