Americans are gaining weight at an alarming rate. While much of the obesity epidemic can be attributed to poor diet and lack of exercise, there are other extenuating factors to consider. Scientists have looked to polymorphisms in obesity-related genes, such as those for the leptin hormone and its receptors, to explain why some people pile on the pounds while others remain svelte, but a new explanation may be in sight.
Jorge Plutzky, HMS associate professor of medicine at Brigham and Women’s Hospital, Ouliana Ziouzenkova, HMS instructor in medicine at BWH, and colleagues have discovered that fat production and diet-induced obesity in mice is regulated by the vitamin A derivative retinaldehyde. The finding, reported online May 27 in Nature Medicine, could dramatically change how researchers view obesity and the retinoid itself, which until now was believed to have only one biological role in humans— as a photophore in the eye.
Plutzky became interested in retinaldehyde because of his work on the relationship between metabolism and atherosclerosis. His lab studies the function of a family of steroid hormone receptor proteins called the peroxisome proliferator–activated receptors (PPARs). Together with the retinoid X receptor (RXR), PPARs form heterodimers that activate gene transcription and regulate a whole range of biological processes, including energy metabolism, fat-cell production, and insulin sensitivity. The receptors have been in the news recently because the PPAR-gamma agonist, rosiglitazone, which is used by millions of type 2 diabetes patients, has been linked to increased risk of heart disease. Ironically, “one of the curious things about PPARs is that while drug screening led to the identification of agents that could activate them, the identity of the endogenous natural partners for PPARs have remained largely obscure,” said Plutzky. Similar issues and controversy also exist for RXR.
To address this, Plutzky and colleagues began to reassess the biological role of retinoids, which are derived from beta-carotene in the diet. In the body, beta-carotene is converted to retinol (vitamin A), which is then oxidized in a two-step process, first to retinaldehyde and then to retinoic acid. The latter has long been viewed as the only activator of retinoid receptors. “Most text books suggest that retinaldehyde only serves as a precursor for retinoic acid, with no biological role other than its function in the eye,” said Plutzky.
And yet Ziouzenkova found that alcohol dehydrogenase and retinaldehyde dehydrogenase, the two enzymes responsible for production and breakdown of retinaldehyde, respectively, are differentially expressed in adipocytes and during adipogenesis. She found that while preadipocytes express mainly alcohol dehydrogenase, differentiated adipocytes express predominantly retinaldehyde dehydrogenase. Ziouzenkova also found that alcohol dehydrogenase expression was significantly higher in fat from lean mice compared to that from obese animals.
“We thought that if the enzyme that makes retinaldehyde is present and regulated in fat tissue, then maybe retinaldehyde has an important role to play there, too,” said Ziouzenkova. That is just what they found.
When the researchers adapted techniques to trap and precisely measure retinaldehyde, they found that not only was it present in fat at low levels but that in lean mice, levels were almost fivefold higher than in obese animals, suggesting that the retinoid may function in fatty tissue to reduce adipogenesis. To test this idea, the researchers added retinaldehyde to adipocyte precursors. This suppressed expression of key adipogenic genes, including the major adipogenic cytokine, adiponectin. “What we were seeing in vitro was that we could turn off adipogenesis with retinaldehyde, even in the presence of rosiglitazone, a powerful adipogenic stimulant,” said Plutzky. But would the retinoid have the same effect in vivo?
The researchers turned to a mouse strain with no retinaldehyde dehydrogenase gene (Raldh) to test this possibility. Though Raldh-negative animals have higher retinaldehyde levels than normal mice, they do not have any previously reported metabolic phenotype—under normal conditions. But when Plutzky, Ziouzenkova, and their colleagues fed the mice a high-fat diet, the Raldh1 knockouts were protected against obesity and even diabetes. After only six months on the fatty diet, normal mice gained considerable weight (see images). The Raldh knockouts, on the other hand, while consuming the same amount of food and water as the control animals, gained 93 percent less weight, had smaller adipocytes, much less subcutaneous and visceral fat, and performed better in glucose and insulin tolerance tests. “All the factors we looked at fit with the fact that retinaldehyde was blocking adipogenesis,” said Plutzky.
Still, it could be argued that effects seen in the Raldh-negative mice could be due to changes in retinol or retinoic acid levels. So the researchers turned to a different mouse model to confirm that retinaldehyde was the key factor. They administered retinaldehyde itself or citral, a natural retinaldehyde dehydrogenase inhibitor, to the well-characterized ob/ob mouse, which is deficient in leptin and becomes more than double the size of a normal mouse when fed a regular diet. After three weeks, ob/ob mice on citral or retinaldehyde had about 15 percent subcutaneous body fat compared to the 19 percent subcutaneous fat in mice given a placebo, vitamin A, or retinoic acid. The mice also had better glucose tolerance.
“That retinaldehyde and retinoic acid had different effects on these mice strongly supported the idea that retinaldehyde itself is its own player in metabolism and not simply a precursor for retinoic acid,” said Plutzky.
Exactly how retinaldehyde works is not clear, but it seems to be related to PPARs, retinoid receptors, and general metabolism. The researchers found that Raldh-negative animals have higher metabolic rates and elevated body temperature compared to control animals, which helps explain why they do not gain weight. The retinoid also binds to PPAR-gamma, albeit weakly, and attenuates increases in adiponectin and adipogenesis driven by the PPAR-gamma agonist rosiglitazone. Plutzky and colleagues also found that retinaldehyde has both retinoid X receptor–dependent and –independent effects on lipid cell precursors and it could also bind to retinoid binding protein 4, a retinoid carrier protein that HMS researcher Barbara Kahn has linked to diabetes.
Paradoxically, PPAR-gamma agonists like rosiglitazone are used to treat diabetes because they help drive adipogenesis, thus lowering serum fatty acids and glucose. “Several lines of data, including some of our studies, suggest that rather than pursuing more potent activators of these receptors, modulating or even inhibiting them may also have some therapeutic benefit,” said Plutzky.
As for explaining why some people may be more apt to gain weight, this work opens up a whole new world of possibilities. Plutzky and collaborators are considering whether polymorphisms in the retinaldehyde dehydrogenase gene may be associated with obesity, for example. “One interesting thing about retinaldehyde is that its formation is a control point, because the conversion to retinoic acid is irreversible,” he said. The latest findings provide new insights into the control of energy balance and may lead to new ways to reverse excessive weight gain.
