The Effects of Acromegaly on Adipose Tissue
Hochberg, I., Tran, Q., Barkan, A., Saltiel, A., Chandler, W., & Bridges, D. (2015). Gene Expression Signature in Adipose Tissue of Acromegaly Patients PLOS ONE, 10 (6) DOI: 10.1371/journal.pone.0129359
A collaborative team including our group, a team at the University of Michigan and the an investigator at the Rambam Health Care Campus in Haifa, Israel recently published this paper in PLOS One. This project started when Irit Hochberg and I were postdoctoral fellows in Alan Saltiel's laboratory and continued after we took positions in Tennessee and Israel respectively. This article was the subject of a recent F1000 recommendation by Giovanni Tulipano at the University of Brescia, in Italy. That review can be found here.
What Were We Looking At?
Acromegaly is a rare endocrine disorder affecting 3 in a million people. It is caused by a tumor of the anterior pituitary that causes excessive release of growth hormone. Normally growth hormone causes bone and muscle growth, while promoting lipid breakdown and insulin resistance. It is high in the young and declines as you age. It also cycles during the day with peaks at night, when growth is most rapid.
In acromegalics, growth hormone stays high and the patients have several distinctive phenotypes including protruded bones (jaw and brow), joint pain, high blood pressure and heart disease. Metabolically, in addition to increased bone and muscle, patients have reduced fat mass and exhibit insulin sensitivity.
Acromegaly is typically treated by by surgical removal of the tumor through the nose under guidance of a microscope or camera, which removes the tumor. While the typical external features are not completely reversed, many of the symptoms areimproved. As a lab, we are interested in the ways by which endocrine systems can alter lipid metabolism. As acromegaly induces both insulin resistance and reduced fat mass, we were interested in the molecular changes in the adipose tissue of these patients. To do this, Irit took advantage of the fact that the surgical field is often sealed with a piece of subcutaneous adipose tissue. We acquired a piece of this fat tissue and did analysed that fat pad.
What Did We Do?
The first part of this study was to identify control patients. We used patients who also had a pituitary tumor, but one that did not secrete either growth hormone or ACTH. These non-secreting adenoma patients were generally older than our acromegaly population, which was an important limitation of our study.
The acromegaly subjects were taller and slightly leaner than the controls, as we expected. They also had higher HOMA-IR score, a measure of insulin resistance. The fat pads exhibited higher rates of lipolysis ex vivo suggesting that the one reason that acromegalics are lean is due to increased triglyceride breakdown.
To understand more about the molecular changes we performed RNA sequencing analysis of the adipose tissue. This technique is used to evaluate changes in mRNA levels in an unbiased manner across the genome. We found 418 genes with differential expression between the controls and the acromegalics. Interestingly, we found that as a person ages the amount of the change was reduced, potentially due to reduced GH/IGF1 levels as one ages, even in acromegalics.
The major pathways that were changed included upregulation of GH/IGF1 negative feedback genes (probably to compensate for increased activation) and higher expression of lipogenesis genes w(maybe to resynthesize the effluxing fatty acids). The best bet to understand the increased lipolysis was an upregulation of lipoprotein lipase (LPL) and two activators of the adipocyte triglyceride lipase (ATGL encoded by the PNPLA2 gene), CGI58 (encoded by ABHD5) and RIP140 (NRIP1). Finally we saw a dramatic downregulation of HSD11B1, an enzyme that promotes glucocorticoid signaling.
What Does it Mean?
The mechanisms by which adipose tissue lipolysis and systemic insulin resistance occurs in acromegaly or growth hormone signaling remain unknown. Transcriptionally, we saw no evidence that at the mRNA level, insulin signaling genes are downregulated, or that mediators of insulin resistance (inflammation or glucocorticoids) are upregulated. It is possible that adipose tissue is not in fact insulin resistant in times of GH over-production, and that this is primarily a muscle phenotype. As far as lipolysis is concerned, the upregulation of LPL, NRIP1 and ABHD5 and how they promote GH-induced lipolysis requires further study.