Novel Adiponectin Variants Identified in Type 2 Diabetic Patients Reveal Multimerization and Secretion Defects

PLOS ONE, October 2011

22046359

Prapaporn Jungtrakoon, Nattachet Plengvidhya, Watip Tangjittipokin, Sarin Chimnaronk, Wanisa Salaemae, Nalinee Chongjaroen, Kanjana Chanprasert, Jatuporn Sujjitjoon, Chatchawan Srisawat, Pa-thai Yenchitsomanus

ADIPOQ, encoding adiponectin, is a candidate gene for type 2 diabetes (T2D) identified by genome-wide linkage analyses with supporting evidence showing the protein function in sensitizing insulin actions. In an endeavor to characterize candidate genes causing T2D in Thai patients, we identified 10 novel ADIPOQ variations, several of which were non-synonymous variations observed only in the patients. To examine the impact of these non-synonymous variations on adiponectin structure and biochemical characteristics, we conducted a structural analysis of the wild-type and variant proteins by in silico modeling and further characterized biochemical properties of the variants with predicted structural abnormalities from the modeling by molecular and biochemical studies. The recombinant plasmids containing wild-type and variant ADIPOQ cDNAs derived from the variations identified by our study (R55H, R112H, and R131H) and previous work (G90S and R112C) were constructed and transiently expressed and co-expressed in cultured HEK293T cells to investigate their oligomerization, interaction, and secretion. We found that the novel R55H variant impaired protein multimerization but it did not exert the effect over the co-expressed wild-type protein while novel R131H variant impaired protein secretion and also affected the co-expressed wild-type protein in a dominant negative fashion. The R131H variant could traffic from the endoplasmic reticulum to the Golgi, trans-Golgi network, and early endosome but could not be secreted. The R131H variant was likely to be degraded through the lysosomal system and inhibition of its degradation rescued the variant protein from secretion defect. We have shown the possibility of using in silico modeling for predicting the effect of amino acid substitution on adiponectin oligomerization. This is also the first report that demonstrates a dominant negative effect of the R131H variant on protein secretion and the possibility of using protein degradation inhibitors as therapeutic agents in the patients carrying adiponectin variants with secretion defect.