LONDON ¿ Is obesity really caused by too many TV dinners? Can those excess pounds be shed by sticking to a strict diet? Does overweight run in the family? Are you fat because you are a slob, or a slob because you are fat?

All kinds of prejudices and dogmas surround the investigation and treatment of obesity, and answers to the questions above are hard to come by. There is undoubtedly a genetic element to obesity, to which many different genes contribute, but environmental factors such as exercise and diet also play a part.

To disentangle problems such as these, and identify the genes involved in obesity and how they interact with environmental factors, Gemini Research Ltd., a wholly owned subsidiary of Gemini Holdings plc, of Cambridge, U.K., is studying thousands of twins and disease-affected sibling relationships. Researchers involved in this project are documenting the clinical phenotypes of these individuals, and measuring features that put them at risk of various diseases. For example, people who have more total-body fat, and more central abdominal fat, are more likely to develop insulin resistance and Type 2 diabetes.

One recently published study by scientists at Gemini and their collaborators examined the influence of a genetic predisposition to obesity, and attempted to separate this from the effect of environmental factors.

This showed that how much fat the average person carries is largely determined by his or her genes. By studying precise measures of body fatness in twins, these researchers have shown that diet plays very little part in determining central abdominal or total body fatness, but that smoking and estrogen replacement both have small but significant impacts.

As part of its gene discovery program, Gemini has extended this work with a study that aimed to find out the relationship between physical activity and total-body and central abdominal fat, while controlling for genetic and other environmental factors. The researchers also wanted to find out whether physical activity had the same effect in people who were genetically susceptible to both general and central obesity as in those who were not.

This study involved 970 women, all white, who comprise a segment of a database established by Gemini Holdings, called Phenobase. In it, 241 monozygotic pairs and 228 dizygotic pairs were used.

The results are published in the Annals of Internal Medicine, dated June 1, in a paper titled ¿Genetic and Environmental Influences on Total-Body and Central Abdominal Fat: The Effect of Physical Activity in Female Twins.¿

Paul Kelly, CEO of Gemini and one of the co-authors of the paper, told BioWorld International: ¿Studies such as this one will allow us to identify whether, for example, there are likely to be genetic markers that will allow us to determine whether people who are obese are more or less responsive to exercise, or whether people who have osteoporosis are more or less responsive to dietary calcium manipulation. These are the sorts of products the market will want in the future, as the intelligent analysis of genetic information revolutionizes disease prediction.¿

The women had their body composition measured using dual-energy x-ray absorptiometry, from which total body fat can be calculated. Their physical activity levels were assessed with two standard questionnaires.

Quadriceps muscle strength was measured, and data obtained on diet, socioeconomic class, smoking, menopausal status and use of hormone replacement therapy.

The researchers found that the women who had the least total-body fat and central abdominal fat were those who performed the most physical activity, whether this was because they were active at home or at work, or did one of the following each week: two hours of sport, 10 miles of walking, or some sweating-associated physical activity.

When researchers examined which of several factors was the most important in influencing total-body and central abdominal fat, they found physical activity was the only major environmental influence.

Identical (monozygotic) twins share the same genes. Therefore, any differences within identical twin pairs are due to environmental factors. The researchers also examined the effect of exercise on body fatness by studying identical twins where both twins had the same levels of smoking or hormone replacement ¿ the other two environmental influences on fatness.

They write: ¿Twin pairs who were discordant for moderate-intensity sporting activity had greater differences in total-body fat; the twin with the lower level of activity had higher body fat. Discordance for two hours of moderate-intensity activity accounted for 1.4 kilograms to 2.74 kilograms of total-body fat.¿

Kelly said: ¿Twins are a very powerful tool for studying where genes and environment interact. Gemini researchers have previously shown that how much fat you carry and where it is distributed is determined by your genes. Therefore, if you are an identical twin and your co-twin is obese, we can assume that you carry genetic risk for obesity. We therefore went on to analyze the effect of exercise on fatness in those with high genetic risk of obesity and those with low genetic risk. We found that middle-aged women with a genetic propensity to obesity were as responsive to exercise as those with low genetic risk. In fact, there was a trend for those with a genetic risk of obesity to be more responsive to exercise.¿

What does the study say about the dogmas that obesity is the result of laziness, or the wrong diet? It rebuts them, Kelly said. ¿It supports the fact that how much weight we carry, whether thin, middle weight or obese, is determined largely by our genes,¿ he said. ¿It shows that obesity is not a lifestyle disorder. While alterations in life style will make definite but small influences, this study confirms that obesity really is a target disease where therapeutics and pharmaceutical interventions will have a potentially major impact.¿

Gemini has also carried out a whole-genome analysis on several thousand non-identical twins selected from its database of twin pairs and their siblings, assessing these individuals for more than 500 risk traits associated with the common age-related diseases. ¿With respect to obesity research, for instance, we believe we have the world¿s largest data set of Caucasian and Asian twins with direct measures of fat mass and its distribution.¿

The company said initial analysis of the data had shown ¿more than 100 genomic regions that potentially harbor genes relevant to common age-related chronic diseases, such as diabetes, asthma, osteoporosis, obesity, hypertension and cardiovascular disease.¿ It has selected more than 50 genomic regions in which to pursue gene discovery. Genes identified in this way will be licensed to partners for the development of novel therapeutic and diagnostic agents.