Published online 16 August 2010 | Nature | doi:10.1038/news.2010.412


GlaxoSmithKline strikes back over anti-ageing pills

Drugs do work as thought, says pharmaceutical giant.

Glass of red wineControversy still dogs anti-ageing drugs that mimic resveratrol, found in red wine.

Drugs designed to combat age-related diseases work as claimed, according to research published last week in the Journal of Biological Chemistry1. The authors, most of whom work for the GlaxoSmithKline (GSK) subsidiary Sirtris Pharmaceuticals, hope that the findings will quell debates over whether the drugs activate a key anti-ageing protein called SIRT1.

Sirtris, based in Cambridge, Massachusetts had been working on compounds that activate SIRT1 and showed promising effects, prompting London-based drug giant GSK to buy the company for US$720 million in 2008. The drugs are thought to mimic the effects of the red wine component resveratrol.

Even at that time, a team led by Matt Kaeberlein, a biochemist at the University of Washington in Seattle2 questioned whether resveratrol directly activates SIRT1, which is involved in the anti-ageing effects of dietary restriction. Their experiments showed that resveratrol activated SIRT1 only when a fluorescent molecule used to gauge its activity was present. Scientists at biotechnology company Amgen in Thousand Oaks, California came to a similar conclusion last year3.

That critique resurfaced this year when scientists at Pfizer in Groton, Connecticut, raised a similar question over SIRT1-activating compounds being developed by GSK4. GSK says that these compounds are more potent than resveratrol at activating SIRT1 and therefore more useful as drugs.

"The papers did a lot of damage to a lot of people," says Ross Stein, a scientist at Sirtris and lead author on the company's new rebuttal1.

Test-tube discrepancies

The debate revolves around a laboratory test Sirtris used to assess whether or not a particular drug increases SIRT1 activity. When activated, SIRT1 strips acetyl groups from cellular proteins including histone proteins, which silences gene transcription and is thought to be involved in its anti-ageing effects. To measure this process, researchers used an acetylated protein fragment containing a fluorescent tag. When activated, SIRT1 deacetylates this peptide causing the tag to glow. The Sirtris scientists reasoned that more fluorescence means greater SIRT1 activity.

The critics, however, found that resveratrol and the other Sirtris drugs activate SIRT1 only when the fluorescent tag is present.

After the Pfizer paper came out, GSK defended its SIRT1 test and the drugs the company found using it. The firm claimed that test-tube conditions did not totally capture the complex biochemistry of SIRT1 in living cells.

Moreover, the drugs seemed to work in mice. Resveratrol protected rodents from the ill effects of a high-fat diet5, and a potent resveratrol mimic made obese mice and rats more sensitive to insulin, suggesting that the drug could be effective against type 2 diabetes6.

Off-target effects

To address lingering doubts, the Sirtris researchers took a closer look at how the drugs interact with SIRT1 and that infamous fluorescent tag. The researchers concede that their drugs can bind to the tagged peptide but contend that this interaction can't alone explain why the drugs activate SIRT1.

If the action of the drugs were due only to the fluorescent tag, Stein says, the stronger the activation of SIRT1, the more strongly the drugs would bind to the tagged peptide. "We have a number of compounds that do not bind to the peptide but are great activators of SIRT1," he says.

Stein and his team also found that Sirtris's drugs could activate SIRT1 even when a fluorescent tag wasn't used to measure deacetylation. However, the researchers agree that an extra chemical group on the peptide is important for SIRT1 activation. Stein suggests that some natural cofactor acts in this capacity in living cells.

"Although this is possible, there is no evidence to believe that this might be true," says Ronen Marmorstein, a biochemist at the Wistar Institute in Philadelphia, Pennsylvania, who remains sceptical that the Sirtris drugs directly activate SIRT1.

Joseph Baur, a molecular biologist at the University of Pennsylvania and former member of Sirtris cofounder David Sinclair's lab, doesn't rule out the possibility that the fluorescent tag mimics a natural SIRT1 cofactor. However, he says, "this is difficult to test, and nearly impossible to disprove".

Kaeberlein, whose team found that SIRT1 activation depends on the fluorescent tag, says that the latest paper does not settle his doubts about whether resveratrol and the other Sirtris drugs activate SIRT1. He notes, however, that resveratrol has many targets other than SIRT1 and it need not directly activate SIRT1 to improve health. "It may be that many of these other compounds have similar off-target effects."

Stein remains confident that the Sitris compounds exert their effects by activating SIRT1, but admits that his team doesn't know everything about how the drugs might act.

Internet selling

The continuing doubts over the Sirtris compounds comes on the heels of another setback for the company. Last week, GSK discovered that two of its executives, who had once been involved with Sirtris, sat on the board of a non-profit foundation that sells resveratrol over the Internet.


Michelle Dipp and Christoph Westphal, the cofounder of Sirtris, ended their affiliation with Healthy Lifespan Institute, which is based in Boston, Massachusetts, after reported that the group sold resveratrol as a dietary supplement.

"We were aware of their involvement in the non-profit, but we weren't aware they were selling resveratrol," says Louise Dunn, a GSK spokeswoman. 

  • References

    1. Dai, J. et al. J. Biol. Chem. advance online publication doi:10.1074/jbc.M110.133892 (2010)
    2. Kaeberlein, M. et al. J. Biol. Chem. 280, 17038-17045 (2005). | Article | ChemPort |
    3. Beher, D. et al. Chem. Biol. Drug. Des. 74, 619-624 (2009). | Article | ChemPort |
    4. Pacholec, M. et al. J. Biol. Chem. 285, 8340-8351 (2010). | Article | ChemPort |
    5. Baur, J. et al. Nature 444, 337-342 (2006). | Article | ChemPort |
    6. Milne, J. C. et al. Nature 450, 712-716 (2007). | Article | ChemPort |


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60962

    In addition, research dishonesty is rampant in Universities because there is not enough money to fund every program, so people exaggerate findings or fake findings to ensure they receive grants the following year. I think for pharma / biotech, the cost of entry for new drugs is so high that if researchers cant prove efficacy early on, their project is scrapped nearly immediately.

Commenting is now closed.