We continue our collaboration with the Cushman lab (NIDDK) and several extramural labs to analyze cell-size distributions in adipose tissue in order to elucidate relationships between fat cell size and insulin resistance. We previously reported that such distributions are roughly bi-model, with a Gaussian peak of large, mature cells and an exponential tail of small cells. In contrast to prior hypotheses, we were unable to find an association between the size of the large fat cells per se and insulin resistance (IR) when moderately obese subjects are matched for obesity. Rather, we found a correlation between the proportion of large cells and IR, with resistant subjects having a deficit of large cells. We proposed that this reflects an impairment of adipocyte differentiation and leads to insufficient fat storage capacity and ectopic fat deposition in other organs, such as liver, pancreas, and muscle, that are not well equipped to handle large volumes of fat. A reanalysis of these questions with a much larger group of subjects spanning a much wider range of BMI has confirmed the correlation between the proportion of small cells and IR but also showed a correlation between large cell size and IR. We attribute the failure to observe this in the earlier pilot study to the smallnumber of subjects the restricted range of BMI. A paper is in preparation. In collaboration with the Smith lab (Gothenburg) we have found a correlation between the size of the large cells and IR in a group of leaner, younger subjects who were first degree relatives of type 2 diabetics (paper in revision). We suggest that this is because the leaner subjects are still able to expand their adipose cells whereas the more obese subjects have reached the limit of cell expansion (hypertrophy) and must recruit new cells (hyperplasia). Furthermore, administration of insulin-sensitizing drugs, such as pioglitazone or rosiglitazone, leads to both recruitment of new cells, which increases the proportion of small cells, and expansion of existing large cells (paper in preparation). Thus, in all these cases the distribution of adipose cell sizes is related to metabolic status, but the particular response is dependent on the metabolic status and history of the subject. Considering these results together with those from the Stanford group (previous paragraph) and work in other sections of LBM, we conclude that large cell size is a primary risk factor for insulin resistance, along with an excess of small cells, reflecting impaired differentiation. Indeed, the two properties may be related, as a defect in potential to recruit and enlarge new adipocytes may be compensated by enlargement of existing adipocytes beyond their healthy operating range. In collaboration with the McLaughlin lab (Stanford) we have also tested the hypothesis that small adipocytes per se are subject to impaired differentiation and function. Small adipocytes were isolated from epididymal adipose tissue of Zucker Obese (ZO) and Lean (ZL) rats and separated by sequential filtration through nylon meshes. Using quantitative real-time PCR for cell differentiation and inflammatory genes we found that the small cels represented a greater proportion in ZO than ZL rats. The small cells in the ZO rats had decreased adiponectin, and increased visfatin and IL-6 levels. The small adipocytes in the ZO rats also had lower adiponectin and PPARγlevels than in the ZL rats but increased IL-6. We thus confirmed for these animal models that the small adipocytes exhibited impaired cell differentiation and pro-inflammatory activity in addition to being present in higher proportions. We suggest that these properties contribute to insulin resistance in the animals. See Ref. # 1.
