Nature Genetics pp 87 - 90 and pp 6 - 7

The old dogma that "you are what you eat" doesn't quite hold, although the central premise is basically true. The regulation of fat mass in the body is now known to be under the control of a myriad of molecular pathways, from the control of appetite to the breakdown of stored fat. Coordination of these processes takes place at many levels, with feedback loops connecting the various sub-systems. Even small miscues in fat metabolism can cause substantial deviation from the optimal level of fat mass, resulting in undesirable consequences. One well-known example is that of leptin: without this appetite-suppressing hormone, humans and mice become extremely obese.

One key step in the regulation of fat mass is the synthesis of triglycerides, the major component of fat (and energy storage) in animals, which are stored in a specialized cell called the adipocyte. This synthesis is carried out by the enzyme acyl CoA:diacylglycerol transferase, or DGAT. One might expect that when DGAT is not present, triglycerides are not produced. But a study by Robert Farese, Jr. (of the Gladstone Institute) and colleagues has revealed that this is not quite true. The researchers inactivated the gene that produces DGAT in mice and found that the mice produced less triglycerides than their normal cousins, but were still healthy. When normal mice were fed a high-fat diet, they become obese, but the DGAT-deficient mice didn't, because they had a higher metabolic rate and expended more energy. At least one other process involving triglycerides was affected, though: mutant female mice couldn't lactate.

As pointed out in an accompanying News & Views article by C. Ronald Kahn (of Harvard Medical School), these results indicate that there must be at least two ways by which to make triglycerides, and the regulation of fat metabolism is even more complicated than was previously thought. The researchers speculate that increased energy expediture induced by lack of DGAT may stave off fatty diet-induced obesity, and that inhibitors of this enzyme may prove useful in treating the condition.

Nature Genetics pp 35 - 41

Gene therapy, the transfer of new genes in the cells of a living animal, offers great promise in treating disease and other physiological conditions. But the biggest problem-how to get the new DNA into the cell, and in particular, to insert it into a chromosome-has been difficult to overcome. The most popular approach has been using disarmed viruses, rendered harmless by mutation and unable to replicate themselves, as vectors to carry the new data into the cells just as they would when infecting normally. Although this has been somewhat successful, there are still serious concerns with using a virus to introduce DNA: these vectors can be difficult to produce, they sometimes induce an unwanted immune response, and they have even been known to reassort into forms that can regain the ability to replicate.

Mark Kay (of the Stanford University School of Medicine) and colleagues have used another genetic tool-the transposon-to introduce new therapeutic genes into animals' cells. Transposons are mobile pieces of DNA, found in virtually all cells, that can be 'cut' from one site in the DNA and re-inserted into a new one. The researchers placed a human gene that promotes blood coagulation inside a transposon, then introduced it into of the livers of a mouse strain with haemophilia. Kay and colleagues were able to successfully integrate the transposon-gene combination into 5-6% of the treated liver cells, and they found that the coagulation gene continued to express for more than five months. These results offer the possibility that transposon-mediated gene transfer may provide another, better, means of introducing therapeutic genes.

Nature Genetics pp 55 - 57

Gliomas are the most common, and among the most deadly, of brain tumours, but what causes them to form is poorly understood. In many other types of tumours, the cause is clear: specific mutations can lead to uncontrolled growth of cells, causing cancer. One well-known example is the protein Ras-a sort of 'Grand Central Station' into which signals controlling cell growth arrive, then are sent off onto the appropriate 'tracks' by turning other genes on-which is mutated in as many as 30% of all human cancers. Mutation of Ras alone, though, does not induce gliomas. A second terminal through which signals are transduced is that of the Akt protein, a factor that serves to prevent cell death. As with Ras, activation of Akt by mutation can lead to many cancers, but not gliomas, by inhibiting cells that are badly damaged (and potentially full of mutations) from dying.

So what causes gliomas to form? In a mouse model system, Eric Holland (of the M. D. Anderson Cancer Center) and colleagues introduced the mutated genes encoding Ras and Akt together into the cells that give rise to gliomas, and found that gliomas formed, with many of the characteristics seen in the human tumours. The researchers also measured the activity of Akt in human gliomas, and found that it was increased. Because it is already known that the activity of Ras is increased in gliomas, these observations offer an explanation for why it takes more than just Ras mutation to cause gliomas. They also support the idea that activation of these two proteins-and possibly disruption of the networks that regulate cell growth and death-induces glioma formation, which may provide the basis for new strategies to combat these deadly tumours.