Differential effects of thiazolidinediones on adipocyte growth and recruitment in Zucker fatty rats

Abstract

Background: Adipose tissue grows by two mechanisms: hyperplasia (cell number increase) and hypertrophy (cell size increase). Thiazolidinediones are insulin-sensitizing peroxisome proliferator-activated receptor gamma agonists that are known to affect the morphology of adipose tissue.

Methodology: In this study, adipose cell-size probability distributions were measured in six Zucker fa/fa rats over a period of 24 days, from four weeks of age, using micro-biopsies to obtain subcutaneous (inguinal) fat tissue from the animals. Three of the rats were gavaged daily with rosiglitazone, a thiazolidinedione, and three served as controls. These longitudinal probability distributions were analyzed to obtain the rate of increase in cell-size diameter in rosiglitazone-treated animals, and the hyperplasia induced by treatment quantitatively.

Conclusions: We found that treatment leads to hypertrophy that leads to an approximately linear rate of cell diameter increase (2 m/day), and that the hyperplasia evident in treated animals occurs largely within the first eight days of treatment. The availability of additional lipid storage due to treatment may alleviate lipotoxicity and thereby promote insulin sensitivity. The hypothesis that a TZD regimen involving repeated treatments of limited duration may suffice for improvements in insulin sensitivity merits further investigation.

Figure 1. Adipose cell-size probability distributions for treated vs. control animals for day 0.

Figure 2. Adipose cell-size probability distributions for treated vs. control animals for day 16.

Figure 3. Adipose cell-size probability distributions for treated vs. control animals for day 24.

Figure 4. Heat-map of adipose cell-size distributions, showing dynamics in both control and treated animals, with the propagation of a rosiglitazone-induced pulse of adipocytes in treated animals.

Figure 5. Differences of adipose cell-size probability distributions at successive time points for control animals.

The curves show the difference between succeeding day adipose cell-size probability distributions at each cell-size bin –-for example, the curve labeled (day 12 – day 8) shows Prob(day 12, ) – Prob(day 8, ).

Figure 6. Differences of adipose cell-size probability distributions at successive time points for treated animals.

The treated animal differences show a hump of cells moving to larger sizes; this hump is not evident in the contol animals.

Figure 7. Log-likelihood values for rates of adipose cell diameter increase for control and treated animals.

The log-likelihood values for control animals are highest close to 0.44 m/day whereas the log-likelihood values are highest around 2 m/day for treated animals.

Figure 8. Relative cell number (Total cell number/Total cell number at day 0) as a function of time for control and treated animals.

Figure 9. Values and standard deviations of cell-size dependence of rates of cell diameter increase for control animals.

Figure 10. Values and standard deviations of cell-size dependence of rates of cell diameter increase for treated animals.

Treated animals show greater rates of cell-diameter increase at intermediate sizes.

Figure 11. Area-under-the-curve for glucose for Zucker fatty rats, treated with rosiglitazone and control, under an oral glucose tolerance test.

The data shown is the mean from three independent experiments with three rats each in control and treated groups. The SEM for the data is 10%.

Figure 12. Area-under-the-curve for insulin for Zucker fatty rats, treated with rosiglitazone and control, under an oral glucose tolerance test.

The data shown is the mean from three independent experiments with three rats each in control and treated groups. The SEM for the data is 10%.