Hepatic lipid remodeling in cold exposure uncovers direct regulation of bis(monoacylglycero)phosphate lipids by Pla2g15

Cold exposure is a selective environmental stress that elicits a rapid metabolic shift to maintain energy homeostasis. In response to cold exposure, the liver rewires the metabolic state shifting from glucose to lipid catabolism. By probing the liver lipids in cold exposure, we observed that the lysosomal bis(monoacylglycero)phosphate (BMP) lipids were rapidly increased during cold exposure. BMP lipid changes occurred independently of lysosomal abundance but were dependent on the lysosomal transcriptional regulator transcription factor EB (TFEB). Knockdown of TFEB in hepatocytes decreased BMP lipid levels and led to cold intolerance in mice. We assessed TFEB binding sites of lysosomal genes and determined that the phospholipase Pla2g15 regulates BMP lipid catabolism. Knockdown of Pla2g15 in mice increased BMP lipid levels, ablated the cold-induced rise, and improved cold tolerance. Knockout of Pla2g15 in mice and hepatocytes led to increased BMP lipid levels, that were decreased with re-expression of Pla2g15. Mutation of the catalytic site of Pla2g15 ablated the BMP lipid breakdown. Together, our studies uncover TFEB regulation of BMP lipids through Pla2g15 catabolism. Overall design: Chromatin immunoprecipitation was performed on livers from individually housed male C57BL/6J mice (11 weeks old) that were placed at 4°C without food for 6 hours or kept at RT with ad libitum access to food (n=2 per group). All mice had ad libitum access to water. SimpleChIP Plus Enzymatic Chromatin IP Kit was used according to manufacturer's instructions. Briefly, 100mg of tissue was used per mouse and two immunoprecipitations were performed with rabbit anti-TFEB antibody, and two immunoprecipitations were performed with normal rabbit IgG control. ChIP DNA was purified using Zymo ChIP DNA Clean and Concentrator Kit. Purified immunoprecipitated and input DNA was quantified and assessed for sheering using the Qubit dsDNA HS Assay Kit and Agilent DNA HS chip, respectively. Samples were prepared according the TruSeq ChIP Sample Preparation kit. Libraries were size selected for an average insert size of 350 bp using SPRI-based bead selection. Quality and quantity of the finished libraries were assessed using an Agilent Tapestation and Qubit dsDNA HS Assay Kit, respectively. Paired end 150bp sequencing was performed on the NovaSeq 6000 by the UW-Madison Biotechnology Center. Sequenced reads were checked for: sequencing quality, insert length, read concordance, duplications, and adapters via FastP and then aligned to the mouse genome (UCSC build mm10) via bowtie2 v2.2.5 using the sensitive end-to-end read alignment mode. Aligned reads are sorted and indexed with SAMtools 1.9 and peaks are called at p<0.05 and q<0.05 (separately) using the input/IgG as the background control with the MACS2. Primary and differential peaks were annotated for their positioning relative to genomic elements with Homer v4.11. De novo and known motif analysis were carried out with Homer. Pathway ontology information was generated by inputting genes differentially bound in cold or RT into geneontology.org for 'Reactome' pathway results