we were interested to see the article by Poulain-Godefroy et al. (4) in which a global decrease in lipogenic markers was associated with the enhancement of inflammation in human adipose tissue, a relationship accentuated in diabetic subjects. The authors further show that expression levels of the inflammatory markers CD14 and IL-18 are increased in omental compared with subcutaneous adipose tissue, the depot differences being particularly evident for IL-18. This suggests that subcutaneous fat is less susceptible to inflammation than the visceral tissue, consistent with the link between visceral fat and the diseases associated with obesity.
Poulain-Godefroy et al. (4) also examined the hypothesis that hypoxia may occur in clusters of adipocytes distant from the vasculature in adipose tissue in obesity, thus underpinning the inflammatory response that develops as fat mass expands (6, 7). However, no increase in the level of the mRNA encoding the α-subunit of the key hypoxia-inducible transcription factor HIF-1 was found in adipose tissue of lean and obese subjects; nor were there depot differences in HIF-1α mRNA level (4). At first glance, this appears incompatible with the “hypoxia hypothesis” (6, 7), although a previous study observed increased HIF-1α mRNA levels in adipose tissue of obese subjects, which fell with weight loss (1), and there is now direct evidence of hypoxia in adipose tissue of genetic and dietary-induced obese animals together with an increase in HIF-1α protein (3, 5, 10).
A key question is whether HIF-1α mRNA is a good indicator of hypoxia in adipose tissue. While changes in mRNA level are normally paralleled by subsequent alterations in the amount of the encoded protein, this does not seem to be the case for HIF-1α in human adipocytes (6). Indeed, our studies on human fat cells in culture have shown that hypoxia (1% O2 or chemically-induced) leads to a marked increase (up to 8-fold) in HIF-1α protein, while HIF-1α mRNA level falls (by up to 4-fold) (8, 9). There are parallels in other systems, for example, THP-1 human monocytes where the HIF-1α mRNA level fell in undifferentiated cells (and was unchanged in differentiated cells) in response to hypoxia while HIF-1α protein increased (2). The explanation for this is not clear, but stabilization of HIF-1α protein may feed back to transcriptional regulation of the HIF-1α gene.
There is emerging evidence in support of the hypoxia hypothesis in obese animals (3, 5, 10) and from studies on adipocytes in culture, as recently reviewed (6). We suggest that it is important to measure the amount of HIF-1α protein, rather than the mRNA level, as a key marker in studies examining whether hypoxia occurs in adipose tissue in human obesity.
- Copyright © 2008 the American Physiological Society