Supplementary Data

• Supplementary Material - Supplementary Material (PDF 246 KB)
• Supplementary Table 1 - Micro RNAs possibly affected by TMEM18 single nucleotide polymorphisms associated with metabolism. Abbreviated output from the online tool mrSNP, showing possible miRNAs affected by TMEM18 variants in the 3' untranslated region. These three SNPs are linked to metabolic function or obesity by genome-wide association studies. (PDF 299 KB)
• Supplementary Table 2 - PAML results. A summary analysis showing performed tests and their results from 3 different levels. (PDF 142 KB)
• Supplementary Figure 1 - TMEM18 is widely conserved. 21 organisms were chosen to investigate the evolution of TMEM18. Three methods were used to construct a multiple sequence alignment and a phylogenetic tree: 1) neighbor-joining, 2) maximum likelihood, and 3) Bayesian inference. There was consensus between all three methods for the resulting tree. The sequence for Naegleria gruberi was used as the outgroup. A) Multiple sequence alignment for TMEM18 orthologs. Protein sequences were aligned by the three putative transmembrane regions as indicated by the black lines above the sequences. Overall, sequences showed high shared identity. Black color was used for identical amino acids while similar amino acids are shown in light grey. Orthologs were absent in Amphimedon queenslandica, Caenorhabditis elegans, and Saccharomyces cerevisiae. B) Phylogenetic tree of the TMEM18 orthologs. The gene has no paralogs in any of the species tested. These results also display the gene's ancient origin, as an ortholog is found in both amoeboflagellata (Naegleria gruberi) and water molds (Albugo candida). The tree produced using the neighbor-joining method had the highest bootstrap values (shown). Posterior probabilities from Bayesian inference analysis are shown in parentheses. Scale bar indicates evolutionary distance. (PDF 603 KB)
• Supplementary Figure 2 - Successful knockdown of CG30051. The Drosophila ortholog TMEM18 was successfully knocked down in a transgenic, cross. Expression of an RNAi construct for CG30051 was driven by the GAL4-UAS system using the ubiquitous daughterless da-GAL4 driver. Thus, RNAi to CG30051 was expressed in all cells. Whole, male flies, aged 5-7 days post-eclosion were collected for analysis. A) Compared to two control crosses, the da-GAL4 × UAS-CG30051-RNAi cross had 94% reduced expression of CG30051 B) To test for off-target effects, levels of DUBAI were analyzed in the same crosses with CG30051 knocked down globally. DUBAI was chosen as this gene overlaps with CG30051. Levels of DUBAI were unaffected by CG30051 knockdown (n = number of samples, 3 replicates per sample; da-GAL4 × w¹¹¹⁸, n = 5; w¹¹¹⁸ × CG30051-RNAi, n = 5; da-GAL4>CG30051-RNAi, n = 6; Tukey HSD test, *** = p < 0.001). (PDF 30 KB)
• Supplementary Figure 3 - CG30051 does not affect intake of food or change expression in response to diet. A) Flies lacking CG30051 in the insulin producing cells do not have altered feeding behavior. The CAFE assay was used to measure the amount of food consumed in 24 hours. Flies were male, aged 5-7 days post-eclosion There was no difference in the amount of food eaten between the Ilp2>CG30051-RNAi cross versus either of the 2 control crosses (n = 10 for each strain). B) Expression of CG30051 did not change in response to different diets, except for one group. CSORC flies were fed different concentrations of a sugar:yeast ratio for 5-7 days post eclosion, and then heads were prepared for analysis via qPCR. Flies fed the 40:40 ratio had less expression of CG30051 compared to the 2.5:2.5 ratio (n = number of samples, 3 replicates per sample; n = 5 for all groups; Tukey HSD test, * = p < 0.05). (PDF 23 KB)