Lipid Identification and Transcriptional Analysis of Controlling Enzymes in Bovine Ovarian Follicle

Abstract

Ovarian follicle provides a favorable environment for enclosed oocytes, which acquire their competence in supporting embryo development in tight communications with somatic follicular cells and follicular fluid (FF). Although steroidogenesis in theca (TH) and granulosa cells (GC) is largely studied, and the molecular mechanisms of fatty acid (FA) metabolism in cumulus cells (CC) and oocytes are emerging, little data is available regarding lipid metabolism regulation within ovarian follicles. In this study, we investigated lipid composition and the transcriptional regulation of FA metabolism in 3⁻8 mm ovarian follicles in bovine. Using liquid chromatography and mass spectrometry (MS), 438 and 439 lipids were identified in FF and follicular cells, respectively. From the MALDI-TOF MS lipid fingerprints of FF, TH, GC, CC, and oocytes, and the MS imaging of ovarian sections, we identified 197 peaks and determined more abundant lipids in each compartment. Transcriptomics revealed lipid metabolism-related genes, which were expressed constitutively or more specifically in TH, GC, CC, or oocytes. Coupled with differential lipid composition, these data suggest that the ovarian follicle contains the metabolic machinery that is potentially capable of metabolizing FA from nutrient uptake, degrading and producing lipoproteins, performing de novo lipogenesis, and accumulating lipid reserves, thus assuring oocyte energy supply, membrane synthesis, and lipid-mediated signaling to maintain follicular homeostasis.

Keywords: MALDI MS profiling; bovine; follicular fluid; gene expression; granulosa; lipids; mass spectrometry imaging; oocyte; ovarian follicle; theca.

Figure 1

Quantification of lipids in bovine follicular compartments by Nile red fluorescence (NRF). (A) Nile red staining (left picture) and hematoxylin coloration of ovarian section (right picture). (B) Scatter plot represents Nile red fluorescence quantification in follicular fluid (FF), granulosa cells (GC), and theca cells (TH) in 15 follicles with an average size of 5.5 mm. Different letters mean significant difference at p < 0.05, (Kruskal–Wallis test and Dunn’s multiple comparison test). Representative region of the follicle stained with Nile Red (lipids) and DAPI (DNA) is shown.

Figure 2

Identification of lipids in bovine 3–8 mm follicles by liquid chromatography coupled with high-resolution MS (LC/HRMS). (A) Family distribution of all of the lipids detected in follicular cells and fluid. * indicates significant difference (p < 0.05, Chi-squared test). (B) Venn diagram presents all of the lipids identified in follicular fluid (FF, red circle) and follicular cells (blue circle). Distribution by families of 275 common lipid species is shown. (C) Distribution by classes of the lipids, detected either in follicular cells only (164 species, blue bars) or in FF (163 species, red bars). Number of identified lipids are indicated above the bars. ST—Sterols; SL—Sphingolipids; GPL—Glycerophospholipids; GL—Glycerolipids; FA—Fatty acyls; FFA—Free fatty acid; PA—Phosphatidic acid; CE—Cholesteryl esters; PC—phosphatidylcholines; PE—phosphatidylethanolamines; PG—phosphatidylglycerols; PI—phosphatidylinositols; PS—phosphatidylserine; LPA—Lyso-Phosphatidic acid; LPC—Lyso-phosphatidylcholins; LPE—Lyso-phosphatidylethanolamine; LPI—Lyso-phosphatidylinositols; Cer—Ceramides; SM—Sphingomyelins; SuSM—Sulfoglycosphingolipids; GM—Gangliosides; DG—Diacylglycerols; TG—Triacylglycerols.

Figure 3

High-resolution mass spectrometry imaging (MSI) of individual bovine follicles. (A) Average spectrum of molecular species in the 100–1000 m/z range. (B) Ion density maps of detected lipids in individual follicles. Spatial segmentation map was obtained by hierarchical clustering using the bisecting k-means algorithm. Lipid identification was performed by LC–HRMS and/or by direct infusion and MS/MS. LPC—Lyso-phosphatidylcholine; PC—Phosphatidylcholine; SM—Sphingomyelin.

Figure 4

MALDI-TOF profiling performed on isolated follicular compartments. Examples of spectra and number of peaks obtained from individual oocyte (OO), cumulus cells (CC), follicular fluid (FF), granulosa cells (GC) and theca cells (TH) are shown for both positive (left) and negative (right) ion modes. After alignment of spectra from all of the compartments, 948 m/z were detected from the average spectrum. The number of common peaks between the compartments and the whole follicle are indicated in white italics.

Figure 5

Differential analysis of lipids in bovine TH (theca), GC (granulosa cells), FF (follicular fluid), CC (cumulus cells), and oocytes (OO) by MALDI-TOF MS lipid profiling. (A) Heat map representation of differentially abundant lipids (ANOVA, p < 0.05). Enclosed table showed the number of peaks per cluster, their specific localization to the ovarian compartment, and the main lipid classes of the identified differential lipids. (B) Examples of identified differentially abundant lipids between ovarian compartments (n = 12 per compartment). Different letters indicate significant differences (p < 0.05, Tukey test for multiple pairs comparison). (C) Discrimination of ovarian follicular compartments by their lipid profiles, by principal component analysis. LPE—Lyso-phosphatidylethanolamine, LPC—Lysophosphatidylcholine, PC—Phosphatidylcholine, PE—Pphosphatidylethanolamine, PI—phosphatidylinositol, SM—Sphingomyelin, TG—Triacylglycerol.

Figure 6

RT-qPCR validation of microarray gene expression analysis. Graphs are presented as the mean+/SEM of normalized expression values of several independent RNA samples per tissue (n = 4 for microarray, and n = 6 for RT-qPCR analysis). Different letters indicate difference at p < 0.05 (ANOVA, Tukey post-hoc test). ABHD6—Abhydrolase Domain Containing 6, AGPAT9—1-Acyl-Sn-Glycerol-3-Phosphate O-Acyltransferase 9, CPT1A—Carnitine Palmitoyltransferase 1A, DGAT2—Diacylglycerol O-Acyltransferase 2, FABP3—Fatty Acid Binding Protein 3, FASN—Fatty Acid Synthase, LPL—Lipoprotein Lipase, PLIN2—Perilipin 2, PNPLA2—Patatin Like Phospholipase Domain Containing 2, PPARG—Peroxisome Proliferator Activated Receptor Gamma.

Figure 7

Differential analysis of gene expression in bovine theca (TH), granulosa cells (GC), cumulus cells (CC), and oocytes (OO), using microarray hybridization (n = 4 per each cell type). (A) Discrimination of ovarian follicular compartments by expression profiles of lipid metabolism-related genes, using principal component analysis. (B). Heat map representation of differentially expressed genes (p < 0.05, Benjamini–Hochberg correction) related to lipid metabolism. (C) Examples of differentially expressed genes between ovarian follicular compartments. Bars are the mean values of four independent replicates +/− SEM. Different letters mean significant difference at p < 0.05 (ANOVA, Tukey test for multiple comparisons). LIPA—Lipase A; GJA1—Gap Junction Protein Alpha 1; SCARB2—Scavenger Receptor Class B Member 2, ACSF2—Acyl-CoA Synthase Family Member 2, SCD5—Stearoyl-CoA Desaturase 5, SCD—Stearoyl-CoA Desaturase, OSBP2—Oxysterol Binding Protein 2, LPL—Lipoprotein Lipase, SCARB1—Scavenger Receptor Class B Member 1, APOA1—Apolipoprotein A1, FADS2—Fatty Acid Desaturase 2, LDLR—Low Density Lipoprotein Receptor, STARD7—StAR (Steroidogenic Acute Regulatory Protein) Related Lipid Transfer Domain Containing 7, APOO—Apolipoprotein O, ELOVL4—Elongation Of Very Long Chain Fatty Acids Protein 4, FABP3—Fatty Acid Binding Protein 3, CEPT1—Choline/Ethanolamine Phosphotransferase 1, HMGCS1—3-Hydroxy-3-Methylglutaryl-CoA Synthase 1.