Loss of the RNA polymerase III repressor MAF1 confers obesity resistance

MAF1 is a global repressor of RNA polymerase III transcription that regulates the expression of highly abundant noncoding RNAs in response to nutrient availability and cellular stress. Bonhoure et al. show that a whole-body knockout of Maf1 in mice confers resistance to diet-induced obesity and nonalcoholic fatty liver disease by reducing food intake and increasing energy expenditure by several mechanisms.

and β-actin mRNAs from wild type and Maf1 -/mouse embryo fibroblasts (Reina et al., 2006).      Table S2). The output was filtered by requiring a >two-fold difference in gene expression, an adjusted p value < 0.05 (EdgeR) and >50 reads (normalized) per KO sample for up-regulated genes or >50 reads per WT sample for down-regulated genes. The 13 genes satisfying these criteria are highlighted in red. Due to the sensitivity of EdgeR to outliers, differential gene expression was also evaluated using DESeq.
None of the 13 genes scored as statistically significant using this approach. (   This table contains raw and normalized RNA-seq read count data and EdgeR analysis of differential gene expression for reads that were uniquely mapped using HTseq-count. Data are represented graphically in Supplemental Fig. S5A.

Animals, diets and analysis of lifespan
The Maf1 targeting vector ( Figure S1A

Indirect calorimetry
Mice were housed individually at 22°C with a 12 h light/dark cycle and allowed to acclimate for 48 h before data collection. Gas exchange measurements were made under the following Oxymax system settings: air flow, 0.6 l/min; sample flow, 0.5 l/min; settling time, 55 sec; measuring time, 5 sec. Energy expenditure was calculated as recommended by the manufacturer using the following formulas. Heat = CV × VO2 and CV = 3.815 +1.232 × RER 8 where CV is the calorific value and RER is the respiratory exchange ratio. Calculations of energy expenditure were normalized for lean body mass.

Hyperinsulinemic-Euglycemic Clamp
The hyperinsulinemic-euglycemic clamp study was conducted over 180 min. in awake, freely moving mice following a 5 h fast. HPLC-purified [3-3 H]-glucose (NEN Life Sciences, Boston, MA) was prime-infused throughout the clamp [5 µCi bolus, followed by 0.05 µCi/min (basal) and 0.1 µCi/min (clamp)] to estimate the glucose disposal rate and hepatic glucose production. After an 80 min. basal period, a blood sample was collected from the tail tip for determination of basal glucose disposal rate (which equals basal hepatic glucose production in basal conditions). The clamp was initiated by prime-infusion of human insulin (Actrapid, Novo Nordisk, Denmark, 25mU/kg bolus, then 2.5 mU/kg/min), and 50% glucose was infused at variable rates and adjusted every 10 min. to clamp plasma glucose levels around 120 mg/dL as measured by glucometers on 2 µl blood samples (Ascensia Breeze2, Bayer Healthcare, Switzerland). After a 2 h stabilization period, blood was sampled from the tail tip 5 times at 10 minutes intervals for determination of glucose turnover under hyperinsulinemic, steady-state conditions. Hepatic glucose production (HGP) was measured as the difference between the Glucose Disposal Rate (GDR) anf the Glucose Infusion rate (GINF).

Western blotting and antibodies
Cell extracts (50 ug protein) were resolved by 7-11% SDS-PAGE and transferred to nitrocellulose membranes for antibody probing and detection with an Odyssey imager (LI-COR).
Primary antibodies were obtained against HSL, phospho-HSL and LC3 (Cell Signaling

RNA-seq analysis of pol II and pol III transcriptomes and RT-qPCR
Three biologically independent samples (RNA integrity numbers >7.5, Agilent Bioanalyzer) from wild-type and Maf1 -/mice were sequenced. For analysis of the pol II transcriptome, uniquely mapped reads were aligned to the mouse genome (mm9) and counted using GSNAP and HTseq-counts, respectively (Wu and Nacu 2010;Anders et al. 2014).
Normalization and statistical evaluation of differential gene expression was performed using EdgeR and DESeq (Robinson et al. 2010;Anders et al. 2013). In the analysis of the pol III transcriptome, sequence tags were mapped in three sequential steps. The first mapping was performed with Bowtie on the Mm9 Mouse genome assembly release. The tags left unmapped were then aligned with BLAT on the pol III loci listed in Table S3. As twenty four tRNA genes 10 contain introns, the remaining tags were aligned in a third round with BLAT to sequences corresponding to spliced tRNAs. The results of the three alignment steps were pooled and the tags aligning in the pol III loci listed in Table S3 were counted. The tags were scaled to the total number of tags aligning in pol II genes, and tags with multiple matches in the genome were given a weight corresponding to the number of matches divided by the number of times they were sequenced. Scores corresponding to the log2 of tags per gene were then calculated. Tags were considered derived from precursor tRNAs when they extended either upstream or downstream of the mature tRNA coding region or overlapped with tags extending upstream or downstream of the tRNA coding region or for tRNA genes containing an intron, when they contained intron sequences. Sequence tags were considered derived from mature tRNAs when they had 5' and 3' ends that mapped entirely within the mature RNA, and for tRNA genes containing an intron, when they spanned the exon-exon junction. For the differential analyses, two different approaches were used. In the first, we fitted the generalized linear model (GLM) from EdgeR (Robinson et al. 2010) on the scaled tag counts per gene for the three Maf1 -/versus the three WT samples; in the second, we applied the limma linear model fitting (Smyth 2004;Smyth et al. 2005) on the log2 of the scaled tag counts per gene, again for the three Maf1 -/versus the three WT samples. RT-qPCR was performed using SybrGreen detection as described previously (Reina et al. 2006). For the measurement of pre-rRNA, primers specific for the 5' external transcribed spacer were employed as reported previously (Oie et al. 2014). Differential gene expression was calculated using the ΔΔCt method using GAPDH as the internal reference.

Quantitation of total tRNA
Total RNA (5 µg) was resolved on denaturing polyacrylamide gels, stained with ethidium bromide, imaged and quantified using ImageQuant software. Amounts of total tRNA were