To the Editor: Obesity is associated with insulin resistance and type 2 diabetes, with accumulation of intra-abdominal fat carrying a more severe disease risk than accumulation of subcutaneous fat. It remains unclear whether increased visceral fat has an adverse metabolic effect due to its location or to the unique properties of intra-abdominal adipocytes. Konrad et al.[1] reported that increasing intra-abdominal fat mass by transplantation of epididymal fat from normal mice into lean recipients improved fasting glucose tolerance and insulin sensitivity, achieving an effect opposite to the expected metabolic consequence of increased intra-abdominal fat. This suggests that obesity-induced alterations in adipose tissue function rather than mass are responsible for the adverse metabolic consequences of obesity. We hypothesised that the intrinsic properties of adipocytes are responsible for their metabolic effects, irrespective of their anatomical location. We have addressed this using regional adipose tissue cross-transplantation in which subcutaneous (inguinal) and intra-abdominal (epididymal) fat pads from donor mice were transplanted into the subcutaneous (group 1) or intra-abdominal (group 2) compartment of recipient mice on a high-fat diet. Our studies revealed that transplantation of intra-abdominal fat into either the intraabdominal or subcutaneous space had no effect on the metabolism of a recipient animal, whereas transplantation of subcutaneous fat into the intra-abdominal space had a significant protective effect on adiposity, insulin sensitivity and glucose tolerance.

In our experiments mice had free access to a high-fat diet (45% of energy as fat, 20% as protein, 35% as carbohydrates, 19.7 kJ/g,[based on rodent diet no. D12451; Research Diets, New Brunswick, NJ]) commencing 1 week prior to surgery and continued for the study duration. Transplantation was performed at 7 weeks. For transplants into the subcutaneous compartment, grafts were implanted through small incisions in the back as previously described [2]; for intra-abdominal transplantation, grafts were sutured to the visceral side of the peritoneum on the anterior abdominal wall [1]. Sham operated mice received identical surgical treatment without transplant. Glucose tolerance tests were performed 12 weeks after transplantation [3] and body composition was determined when mice were killed (13 weeks after transplantation). Experiments were carried out with approval from the Garvan Institute/St. Vincent’s Hospital Animal Ethics Committee. Equivalent amounts of subcutaneous and epididymal adipose tissue were transplanted in groups 1 and 2 (328±20 vs 326±10 mg). At death adipose grafts were clearly identified in both depots and appeared viable with evidence of angiogenesis. Histologically, grafts retained features of their respective endogenous beds comprising well-defined adipocytes interspersed with stromovascular cells. There was no difference in weight gain in mice receiving transplants and sham operations in both groups (Table 1). Strikingly, mice receiving subcutaneous transplants into the intra-abdominal compartment displayed a significant reduction in the mass of various endogenous white adipose tissue beds compared with controls (Table 1). Inguinal, epididymal and retroperitoneal beds were reduced by 26, 33 and 35% respectively in these mice. This effect did not achieve significance in any of the other groups. No difference was seen in interscapular fat pad size in mice receiving subcutaneous transplants into the intra-abdominal compart-