Somewhere between outright starvation and copious Big Macs lies the optimal diet for a long life - though it is apparently closer to starvation than gluttony. Caloric restriction to roughly two-thirds of normal increases the average lifespan of species from worms to primates to cattle. In fact, the latter research prompted an astute as well as practical-minded Texan to note at the recent American Association for Cancer Research annual meeting that he was at a loss to understand what the advantages of a lean, long-lived cow might be.

But despite the fact that the beneficial lifespan effects of near-starvation have been known for more than 70 years, it is only now, in a May 2, 2007 online publication in Nature, that researchers from the Salk Institute for Biological Studies in La Jolla, Calif., describe the first gene affecting lifespan whose activation is specific to caloric restriction.

Surprisingly, the caloric restriction pathway does not appear to be related to insulin/IGF-1 signaling pathway, another pathway that controls aging and had been many scientists' first choice for the molecular underpinning of starvation-induced longevity.

"There was a large emphasis in the field that maybe insulin/IGF signaling was what was regulating dietary restriction responses. But that turns out not to be the case," senior author Andrew Dillin told reporters at a press conference announcing the findings.

The effects of caloric restriction also are independent of the electron transport chain, another pathway that has been implicated in aging. (See BioWorld Today, July 27, 2006.)

"All three of these pathways function independently of each other," Dillin said.

In C. elegans, the worm the authors used as an animal model, the insulin/IGF-1 pathway regulates the DNA-binding protein DAF-16. DAF-16 and another protein called SMK-1 cooperate to regulate expression of longevity genes, making it reasonable to assume that caloric restriction would activate both DAF-16 and SMK-1.

Perhaps reasonable, but nevertheless incorrect. Knocking out DAF-16 had no effect on dietary restriction-induced longevity, but knocking down SMK-1 did. The finding "really suggested to us that SMK-1 was partnering with a different transcription factor to regulate the response to dietary restriction," Dillin said.

C. elegans has 15 other transcription factors that are similar to DAF-16, and the scientist began by knocking each of them out separately with RNAi techniques. They discovered that knocking down another protein, known as PHA-4, "could completely remove the longevity we see when we reduce feeding in the worms," first author Siler Panowski said at the press conference.

The effects of PHA-4 are specific to dietary restriction. PHA-4 knockdown did not affect longevity due to manipulations of either insulin/IGF signaling or the electron transport chain.

PHA-4 is highly conserved on the evolutionary tree. The equivalent gene in mice and humans goes by the name Foxa. Foxa transcription factors play important parts during development and act later in life to regulate glucagons and glucose levels, particularly in response to fasting. "This may be the primordial gene that regulates nutrient sensing that helps an animal overcome stressful conditions and live a long time" under caloric restriction, Dillin said.

The sirtuins are another family of compounds that play a role in the response to starvation, while its effects on longevity are undisputed, and research published last year showed that activating sirtuins can reverse some of the deleterious effects of a high-calorie diet in middle-aged mice. (See BioWorld Today, Nov. 2, 2006.)

But Dillin said that caloric restriction does not appear to be critical for its effects. Sirtuins "interacts with a lot of different pathways," he said. In mice at least, sir-2 appears to interact with the insulin-IGF pathway.

The research also offers the possibility that pharmacological approaches may enable people to reap the benefits of FoxA activation without having to restrict their calorie intake, literally offering what evolutionary biologists have long held to be nonexistent: the free lunch.

"We're not there yet," Dillin said when asked about the possibility. But "it's a really interesting set of experiments that we're trying to do right now in mice." He noted that the University of California has filed patent applications on "PHA-4 in response to dietary restriction."