Longevity researchers say they've shown for the first time that following a strict low-calorie diet can decrease DNA damage linked with aging.

Some people who took part in the six-month diet study ate as little as 890 calories a day. Their insulin levels fell and metabolisms slowed — changes that are thought to increase longevity.

The findings are provocative, but preliminary. Longer-term research will try to sort out whether such changes can meaningfully extend people's lives, said senior author Eric Ravussin of the Pennington Biomedical Research Center at Louisiana State University.

"They are the first proof that what has been observed in rodents seems to be also working in humans," Ravussin said.

The results are from the first phase of research at the Baton Rouge center sponsored by a $12.4 million National Institute on Aging grant. They follow unrelated research reported in January which suggested a very restrictive diet seemed to help the heart age more slowly.

The latest study appears in Wednesday's Journal of the American Medical Association.

"It's very exciting," said Dr. Evan Hadley, director of the NIA's geriatrics and clinical gerontology program.

"It's a step forward but not the whole journey," said Hadley, whose agency is part of the NIH.

The 48 participants, all slightly overweight, were randomly assigned to one of four groups: calorie restriction, which cut usual daily calories by 25 percent; calorie restriction plus exercise, which cut daily calories by 12.5 percent and increased physical activity by 12.5 percent five days a week; very low calories, with an 890-calorie liquid diet for up to about three months followed by a weight-maintenance diet; and a control group that aimed to keep weight steady.

Government dietary guidelines for weight maintenance recommend about 2,000 to 3,000 calories a day, depending on age, gender and activity level, with the higher amount generally for very active men.

The non-liquid diets used in the study were high in fruits and vegetables with less than 30 percent fat.

Average weight loss was about 18 pounds, slightly more in the liquid-diet group.

Blood tests showed substantial decreases in the amount of age-related DNA damage in each of the three dieting groups, compared with their initial levels. That kind of microscopic damage is linked to cancer and other age-related ailments, but it's unknown whether the small changes seen in the study would affect the study volunteers' disease risks.

No changes were seen in the control group.

Insulin levels also decreased after six months in all three reduced calorie groups. Core body temperature also dipped slightly in two low-calorie groups but not in the liquid-diet or control group.

The results show that the diets are safe, and not impossible to follow, Hadley said.

Kacy Collins, 34, a once sedentary Baton Rouge law clerk, said she joined the study to lose weight and hoped that would reduce her risks for age-related ailments that run in her family, including diabetes and heart disease.

Assigned to the exercise group, Collins said the hardest part was eliminating most sweets and exercising daily. But she said eating up to seven servings of fruits and vegetables daily kept her feeling full. She has stuck with the new habits and kept off most of the nearly 30 pounds she lost since the study ended almost two years ago.

Collins said it will be a big bonus if the changes help her live longer, too.