Objective: To test the hypothesis that either uncoupling protein-2 UCP2 or UCP3 or both together influence obesity and inflammation in transgenic mice.
Design: We generated 12 lines of transgenic mice for both human UCP2 and 3 using native promoters from a human bacterial artificial chromosome (BAC) clone. The BAC expresses no genes other than UCP2 and 3. Mice used for experiments are N4 or higher of backcross to C57BL/6J (B6). Each experiment used transgenic mice and their nontransgenic littermates.
Results: Northern blots confirmed expression on human UCP2 in adipose and spleen, while human UCP3 expression was detectable in gastrocnemius muscle. Western blots demonstrated a four-fold increase of UCP2 protein in spleens of Line 32 transgenic animals. Heterozygous mice of four lines showing expression of human UCP2 in spleen were examined for obesity phenotypes. There were no significant differences between Lines 1 and 32, but female transgenics of both lines had significantly smaller femoral fat depots than the control (littermate) mice (P=0.015 and 0.005, respectively). In addition, total fat of transgenic females was significantly less in Line 1 (P=0.05) and almost significantly different in Line 32 (P=0.06). Male Line 1 mice were leaner (P=0.04) while male Line 32 mice were almost significantly leaner (P=0.06). Heterozygous mice of Lines 35 and 44 showed no significant differences from the nontransgenic littermate controls. Effects of the UCP2/UCP3 transgene on obesity in Line 32 mice were confirmed by crossing transgenic mice with the B6.Cg-Ay agouti obese mice. B6.Cg-Ay carrying the UCP2/UCP3 transgene from Line 32 were significantly leaner than nontransgenic B6.Cg-Ay mice. Line 32 UCP2/UCP3 transgenics showed increased hypothalamic Neuropeptide (NPY) levels and food intake, with reduced spontaneous physical activity. Transgenic baseline interleukin4 (IL-4) and interleukin6 (IL-6) levels were low with lower or later increases after endotoxin injection compared to wild-type littermates. Endotoxin-induced fever was also diminished in transgenic male animals. Low-density lipoprotein (LDL) cholesterol levels were significantly higher in both Line 1 and 32 transgenics (P=0.05 and 0.001, respectively) after they had been placed on a moderate fat-defined diet containing 32% of calories from fat for 5 weeks.
Conclusion: Moderate overexpression of UCP2 and 3 reduced fat mass and increased LDL cholesterol in two independent lines of transgenic mice. Thus, the reduced fat mass cannot be due to insertional mutagenesis since virtually identical fat pad weights and masses were observed with the two independent lines. Line 32 mice also have altered inflammation and mitochondrial function. We conclude that UCP2 and/or 3 have small but significant effects on obesity in mice, and that their mechanism of action may include alterations of metabolic rate.