MicroRNAs and BMI

A new paper in PLoS One by Ortega, et al. examines expression of human microRNAs in adipose tissue of lean vs. obese individuals (n is small!) and in differentiating adipocytes. A few interesting observations emerge when one downloads and integrates the data into a larger genome database:

1. MIRN145, which downregulates IRS1 translation, is downregulated during adipocyte differentiation.

2. MIRN23B, whose expression is curtailed by MYC (thereby increasing mitochondrial glutaminase and up-regulating glutamine catabolism, a mechanism behind altered glucose metabolism in cancer cells), is also downregulated during adipocyte differentiation.

3. MIRN337 has been reported to show expression levels negatively correlated with BMI in osteoarthritic chondrocytes and is also downregulated during adipocyte differentiation.

4. Similar to MINR337, MIRN22 expression levels positively correlated with BMI in osteoarthritic chondrocytes. MIRN22 regulated PPARA and BMP7 expression and its inhibition blocked inflammatory and catabolic changes in osteoarthritic chondrocytes. Ortega and colleagues show that MIRN22 is upregulated during adipocyte differentiation.

5. MIRN22, MIRN29A and MIRN337 are all downregulated in subcutaneous fat of obese individuals (Ortega, et al. 2010).

6. MIRN146a positively correlates with triglyceride (TG) levels in subcutaneous fat (Ortega, et al 2010), while MIRN210 and MIRN99B negatively correlate with TG. Nine miRNAs positively correlate with BMI (MIRN10A, MIRN34A, MIRN99A, MIRN100, MIRN125B, MIRN129, MIRN199A, MIRN199B, MIRN221) and five (MIRN92A, MIRN130B, MIRN142, MIRN210, MIRN484) correlate negatively. (Data from Table S3.)

So, it would seem logical that there is a role, a significant one at that, for microRNAs in obesity. What is interesting to me is considering the prospects of small molecules, say from the diet because it abounds with so many different molecules, interacting with miRNAs and altering their interactions with target genes. Furthermore, some of the miRNAs listed here and others contain SNPs that we could easily genotype in any of a number of populations to test for associations to clinical measures of obesity, dyslipidemia, vascular diseases or type 2 diabetes. We would naturally also look for those associations that are modulated by environmental or dietary factors. Now, that would make for a very nice report!