Functional characterization of the Ucp1-associated oxidative phenotype of human epicardial adipose tissue

Brown fat presence and metabolic activity has been associated with lower body mass index, higher insulin sensitivity and better cardiometabolic profile in humans. We, and others, have previously reported the presence of Ucp1, a marker of brown adipocytes, in human epicardial adipose tissue (eAT). Characterization of the metabolic activity and associated physiological relevance of Ucp1 within eAT, however, is still awaited. Here, we validate the presence of Ucp1 within human eAT and its ‘beige’ nature. Using in-vitro analytical approaches, we further characterize its thermogenic potential and demonstrate that human eAT is capable of undergoing enhanced uncoupling respiration upon stimulation. Direct biopsy gene expression analysis reveals a negative association between thermogenic markers and oxidative stress-related genes in this depot. Consistently, isoproterenol (Iso) stimulation of eAT leads to a downregulation of secreted proteins included in the GO terms ‘cell redox homeostasis’ and ‘protein folding’. In addition, cardiac endothelial cells exhibit a downregulation in the expression of adhesion markers upon treatment with Iso-stimulated eAT derived conditioned media. Overall, these observations suggest that Ucp1- associated metabolic activity plays a significant role in local tissue homeostasis within eAT and can plausibly alter its communication with neighboring cells of the cardiovascular system.


RNA extraction, reverse transcription and Quantitative PCR
Total RNA was isolated from tissue or cells using the RNeasy Lipid Tissue Mini Kit (QIAGEN, Mississauga, Ontario) according to manufacturer's instructions. Purity of total RNA was determined as 260/280 nm absorbance ratio with expected values between 1.8-2.0 using Biodrop Duo (BioDrop, UK). In addition, RNA integrity of randomly selected samples (n =30, in tissue biopsies) was assessed using the Agilent 2100 Bioanalyzer (Agilent technologies, Santa Clara, California). 500ng of extracted total RNA was reverse-transcribed using iScript Advanced cDNA Synthesis kit (Bio-Rad Laboratories Ltd., Canada). The cDNA was diluted 1:20 in DNasefree water and was subjected to quantitative PCR using SsoAdvanced Universal SYBR Green Supermix (Bio-Rad Laboratories Ltd., Canada). The qPCR was performed in CFX384-Touch Real-Time PCR Detection System (Bio-Rad Laboratories Ltd., Canada). At the end of each run, melting curve analysis was performed to validate product specificity, and a few samples representative of each experimental group were run on agarose gel to verify specificity of the amplification followed by sequencing of the amplified product. All samples were amplified in duplicates and mean values were used for further analysis. Expression values were determined using standard curve method for each target and reference gene. A normalization factor calculated from the mean of expression levels of Ppia, Gapdh and Rpl27 was used for the normalization process of the biopsies data (n=53), which was derived from the validation methods described previously 1,2 . All in-vitro data (adipocytes and endothelial cells) were normalized to Gapdh levels alone. Primer details are given as Supplementary Table 7.

Ucp1-immunohistochemical analysis
Ucp1-immunostaining was performed on sections of 5-µm-thick formalin-fixed paraffinembedded adipose tissue samples using an automated immunohistochemistry protocol involving EnVision FLEX mini kit, high pH (Agilent technologies), with UCP1 antibody AC10983 (Abcam, USA) at 1:1000 dilution for use with Dako Autostainer 48 at the Department of Histopathology, IUCPQ.

Oil Red O staining
Cells were washed with PBS followed by incubation with 10% paraformaldehyde for 15 minutes to O/N. Cells were washed 3 times with PBS followed by 100% propylene glycol wash followed by incubation with ORO stain (Sigma-Aldrich) at 37°C for 1 hour. Excess stain was discarded followed by addition of 85% propylene glycol. Cells were washed with water until all excess stain is removed. Pictures were taken at this step. ORO stain retained by adipocytes was next eluted using NP40 in isopropanol and quantified by measuring absorbance at 520nm.

Oxygen consumption analyses
Primary preadipocytes were plated (20,000 cells/well) and differentiated in XF24 V7 PET cell culture microplates using the protocol described above. On the day of assay, cells were switched to XF-OCR assay media (XF base medium + 1mM sodium pyruvate+ 2mM Lglutamine + 1mM HEPES + 2% free-fatty acid free BSA) in the presence or absence of dibutyryl cAMP (Db-cAMP) (1mM) for 4 hours prior to the assay. Mito-stress test using oligomycin (Tocris Biosciences)(5µM), FCCP (10µM) (Tocris Biosciences) and antimycin (5µM)/rotenone (3µM) (Cayman Chemicals) were then conducted using Seahorse XF e Bioanalyzer as per manufacturer's instructions (Agilent). Post XF assay, cells were washed with PBS and fixed using 10% PFA followed by ORO staining. Mitochondrial respiration was calculated by subtracting non-mitochondrial respiration rates (determined as the respiration rate post antimycin/rotenone addition from OCR values. Basal respiration was calculated as the average of the values before the addition of oligomycin. Leak respiration was taken to be the lowest OCR value after the addition of oligomycin. Maximal respiration represented the highest OCR values after the addition of FCCP. Coupling efficiency was calculated as the ratio of basal respiration and ATP turnover (calculated as the difference of leak respiration from basal respiration).
For fatty acid oxidation assay, cells were switched from differentiation media to substratelimited media (XF-base medium+ 0.5mM glucose+ 1mM L-glutamine + 0.5mM carnitine and 1% FBS) 24 hours prior to the assay. 45 minutes prior to the assay, cells were switched to FAO media (KHB+ 2.5mM glucose + 0.5mM carnitine + 5mM HEPES adjusted to pH 7.4 at 37 ο C) and incubated in the non-CO 2 incubator at 37°C. Etomoxir (40µM) (Sigma Aldrich) was added in the specified wells 15 minutes prior to assay and finally XF palmitate-BSA FAO (100µM) (Agilent XF Seahorse) substrate or BSA control were added before running the mito-stress test (as specified above) using XF e 24 Seahorse bioanalyzer as per the protocol.

Mass spectrometry
Samples were analysed by nanoLC/MSMS as triplicates for statistical information. replicate were used to calculate the ratio of intensities averages between the two samples to compare as well as a z-score calculated as follow: z = (x-µ)/σ where x = ratio of intensities averages, µ= median of all ratios of quantifiable proteins, σ = standard deviation of all ratios of quantifiable proteins. A p-value based on a Welch's t-test (modified Student's t-test) was also calculated. When LFQ intensity values were missing, there were replaced by a noise value corresponding to the first percentile of LFQ values of all proteins of the sample replicate. A protein was considered as quantifiable only if at least two replicate values in one of the two samples to compare were present. A protein was considered as variant if the absolute value of its z-score was higher than 1.96 and the associated p-value was lower than 0.05.

Metascape Analysis
Metascape first identified all statistically enriched terms (can be GO/KEGG terms, canonical pathways, hall mark gene sets, etc.), accumulative hypergeometric p-values and enrichment factors were calculated and used for filtering. Remaining significant terms were then hierarchically clustered into a tree based on Kappa-statistical similarities among their gene memberships. Then 0.3 kappa score was applied as the threshold to cast the tree into 'enrichment term' clusters.
Subsets of representative enrichment terms from cluster above are converted into a network layout. More specifically, each term is represented by a circle node, where its size is proportional to the number of input genes falling into that term, and its color representing its cluster identity (i.e., nodes of the same color belong to the same cluster). Terms with a similarity score > 0.3 are linked by an edge (the thickness of the edge represents the similarity score). The network is visualized with Cytoscape (v3.1.2) with "force-directed" layout and with edge bundled for clarity. One term from each cluster is selected to have its term description shown as label.      (Table S2a). Genes were grouped in various categories as described in detail in the methods section and were compared using a multivariate ANOVA model (Table S2b).