Richard L. Atkinson

Obetech Obesity Research Center, Richmond, VA, USA


The global prevalence of obesity began to increase dramatically starting about 1980. In the USA, the prevalence of obesity (BMI > 30 kg/m2) increased from 15% to 31% from 1980 to 2000. Across the world, obesity increased regardless of economic or educational status, but the geographic pattern was irregular. In developed countries fast foods, sugary beverages, and decreased activity were blamed for the epidemic of obesity, but these explanations are not tenable in less developed countries. Animals also have become heavier since 1980 with pets, lab animals, and wild animals all increasing in weight. The pattern of rapid spread cannot be due to genetic changes and fits best with an environmental agent such as an infectious disease or worldwide pollutants. Eight viruses have been shown to increase adiposity in animals. Human adenoviruses are of particular interest and human adenovirus-36 (Ad-36), Ad-37, and Ad-5 have been shown to cause obesity in animals. Ad-36 is best studied and experimental infection results in increased adiposity in chickens, mice, rats, and non-human primates. Visceral or total adipose tissue mass increased 50% to >100% in chickens and mice and about 70% in non-human primates. About 60% to 70% of infected chickens and mice became obese compared to control animals and 100% of monkeys, so the effect of this virus is powerful. Humans cannot be experimentally infected but the presence of antibodies to Ad-36 is specific evidence of past infection because the antibodies do not cross-react with other known human adenoviruses. Presence of Ad-36 DNA in tissues also is direct evidence of past infection. Adv36 is a “common cold” virus that is spread by droplet and fecal/oral routes. Multiple investigators across the world have tested people for Adv36 antibodies and the prevalence has ranged from 6% to 65% in obese humans and 4.5% to ~45% in non-obese people. In initial studies in the USA, about 30% of obese adults and 11% of non-obese adults were infected with Ad-36 and there was a very strong correlation of Adv36 infection and obesity. Later studies in adults have given mixed results on the correlation of Adv36 and body weight, but the prevalence may have increased. Children are much more consistent with six studies from four countries showing a prevalence of about 28% in obese children and 15%-20% in non-obese. A unique study has just been completed in military personnel. Adv36 infection predicts development of overweight/obesity over time. In addition to obesity, preliminary data shows that women with breast cancer have a 2-3 fold higher prevalence of Adv36 infection than noncancer patients. In tissue culture, benign human breast cells infected with Adv36 acquire malignant characteristics of increased growth, greater migration, and the appearance of multiple cancer markers. The effect of Adv36 on diabetes is paradoxical. Diabetics have a lower prevalence of Adv36 infection than normal weight or obese. Adv36 infection ofmildly diabetic mice produces a decrease in hyperglycemia and increased glucose disappearance. The mechanism of Adv36 actions appears to be a direct effect on cells to alter intracellular metabolism, particularly glucose and lipid metabolism. Viral DNA is spread throughout the body during the initial viremia phase of acute infection. Viremia lasts about 2-8 weeks, then Adv36 can no longer be cultured. However, Adv36 DNA can be recovered in multiple organs months after the initial infection. Adv36 DNA assayed by polymerase chain reaction has been found in adipose tissue of humans and multiple animal species who have been naturally infected. Adv36 in tissue culture alters leptin, lipoproteinlipase, multiple lipogenic enzymes, and PPAR-gamma. Glucose transporters are increased in infected cells, particularly Glut 4, and non-insulin dependent glucose transport into cells is increased. Stimulation of the Ras pathway appears to be responsible for the increased glucose disappearance rate. The E4orf1 gene of Adv36 is responsible for most of the effects of the virus. When the E4orf1 gene was deleted or blocked with siRNA, the lipogenic effects were blocked. When the E4orf1 gene was cut from Adv36, inserted into a lentivirus, and transfected into preadipocytes or human breast cells in vitro, this reproduced the lipogenic and glucose transporting effects of the virus. A vaccine has been produced to Adv36 and given to animals. Serum from vaccinated rabbits inhibits growth of virus at 17 serial dilutions. Prior experience in the US military with adenovirus vaccines suggests that the Adv36 vaccine will be safe and effective. For individuals already infected, it will be necessary to find antiviral compounds that are capable of blocking Adv36 from stimulating lipid and glucose metabolic pathways. Simply blocking replication of the virus will not be effective as there is minimal or no viral replication noted in terminal cells. Two compounds have been identified that decrease Adv36 effects on cells in vitro. The combination of these two agents is greater than either alone. A protein has been derived from the E4orf1 gene of Adv36 that facilitates glucose transport in vitro and in vivo. This may represent a new method of treating diabetes. In summary, Adv36 causes obesity in animals and is associated with obesity in humans in multiple studies of populations across the world. It seems very likely that Adv36 infection has played a significant role in the worldwide epidemic of obesity since 1980. Adv36 infection is increased in women with breast cancer and in vitro causes or exacerbates malignant changes in cells. Antiviral agents may be useful for the treatment of obese individuals and women with breast cancer. Adv36 proteins may be useful for treating diabetes. A vaccine that could prevent or reduce the prevalence of obesity and/or breast cancer would be a major public health advance.