Effects of granulocyte-colony stimulating factor on obesity and energy metabolism in male Otsuka-Long-Evans-Tokushima fatty rats

Abstract

Purpose: The structure and signaling pathways of G-CSF are similar to those of leptin. G-CSF also has immune-modulator properties, reduces reactive oxygen species in intracellular spaces and protects mitochondrial functions. Given that obesity and diabetes are associated with adipocyte inflammation and mitochondrial dysfunction, we hypothesized that G-CSF was a plausible candidate for reducing body weight. Methods: We treated twelve 38-week-old male Otsuka-Long-Evans-Tokushima fatty rats (OLETF, diabetic phenotype) and twelve age-matched male Long-Evans-Tokushima rats (LETO, healthy control) with 200 μg/day G-CSF or saline for 5 consecutive days and followed them up for 8 weeks. Energy expenditure (EEx) was measured weekly after the treatments by an indirect calorimeter. Body composition was measured by Dual Energy X-ray Absorptiometry (DEXA) 1 week before and 8 weeks after the treatments. Expression of appetite mediators including Proopiomelanocortin (POMC), Neuropeptide Y (NPY), Cocaine-Amphetamine-Related Transcript (CART) and Agouti-Related Peptide (AgRP) in hypothalamus was measured by quantitative PCR. Expression of Uncoupling Protein-1 (UCP-1) in Brown Adipose Tissue (BAT) was measured by both quantitative PCR and western blotting. Results: In OLETF, body weight reduction was larger in the G-CSF-treated group than the saline-treated group 8 weeks after the treatments (6.9% vs. 4.3%, p < 0.05). There was no effect on body weight in LETO, and food intake did not change in any group. The reduction of body fat mass in the G-CSF-treated group was larger than that in the saline-treated group in OLETF (48.14g vs. 36.46g, p < 0.05), but unaffected in LETO. The EEx was higher from 5 weeks after G-CSF treatment in OLETF, but unaffected in LETO. Cholesterol, triglyceride and interleukin-6 decreased after G-CSF treatment only in OLETF. The mRNA expression of POMC and NPY increased in hypothalamus after G-CSF treatment in both OLETF and LETO, while the mRNA expression of CART and AgRP did not differ between the G-CSF-treated groups and the saline-treated groups in both OLETF and LETO. Both mRNA and protein expressions of UCP-1 in BAT of the G-CSF-treated group were higher in OLETF, while not different in LETO. G-CSF receptor was also identified in BAT by RT-PCR. Conclusions: G-CSF reduces body weight in a diabetic model, and this may be associated with the restoration of BAT dysfunction in the diabetic model. Therefore, G-CSF is a possible candidate for treating obesity in diabetic subjects.