Weight Loss Results in Increased Expression of Anti-Inflammatory Protein CRISPLD2 in Mouse Adipose Tissue

Abstract

Objective: Obesity is a major risk factor for cardiovascular disease, metabolic syndrome, and type 2 diabetes mellitus, whereas weight loss is associated with improved health outcomes. It is therefore important to learn how adipose contraction during weight loss contributes to improved health. It was hypothesized that adipose tissue undergoing weight loss would have a unique transcriptomic profile, expressing specific genes that might improve health.

Methods: This study conducted an RNA-sequencing analysis of the epididymal adipose tissue of mice fed either a high-fat diet (HFD) or a regular rodent chow diet (RD) ad libitum for 10 weeks versus a cohort of mice fed HFD for the first 5 weeks before being swapped to an RD for the remainder of the study (swapped diet [SWAP]).

Results: The swapped diet resulted in weight loss, with a parallel improvement in insulin sensitivity. RNA sequencing revealed several transcriptomic signatures distinct to adipose tissue in SWAP mice, distinguished from both RD and HFD adipose tissue. The analysis found a unique upregulated mRNA that encodes a secreted lipopolysaccharide-binding glycoprotein (CRISPLD2) in adipose tissue. Whereas cellular CRISPLD2 protein levels were unchanged, plasma CRIPSLD2 levels increased in SWAP mice following weight loss and could correlate with insulin sensitivity.

Conclusions: Taken together, these data demonstrate that CRISPLD2 is a circulating adipokine that may regulate adipocyte remodeling during weight loss.

Figure 1.. Diet SWAP promotes weight loss in obese mice.

A: Schematic of the dietary regimen of the 3 different groups of mice (n=10 mice per feeding group). B and C: Body mass of individual mice was recorded weekly. D: Tissue masses of each mouse were recorded at the time of killing. One-way analysis of variance was used to identify significant differences at each time point. Asterisk (*) indicates different from RD mice; # indicates different from SWAP mice (p < 0.05).

Figure 2:. Diet SWAP recovers insulin resistance in obese mice.

A-C: During the 5 weeks post-SWAP, fasted blood glucose and serum insulin were used to calculate the insulin sensitivity index of each mouse (n=5 mice per feeding group). D: Whole cell lysates were generated from EPI and SUBQ adipose of mice killed two weeks post-SWAP and were separated via SDS-PAGE and immunoblotted for GLUT4 and tubulin (n=5 mice per group). E: mRNA was isolated from EPI and SUBQ adipose of mice killed two weeks post-SWAP and qRT-PCR was used to determine relative expression of leptin (n=5 mice per group). One-way analysis of variance was used to identify significant differences for each weekly time point and for expression data. Calculation of 95% confidence intervals was used to determine statistical differences. Asterisk (*) indicates different from RD mice; # indicates different from SWAP mice (p < 0.05).

Figure 3.. Feeding behavior and body mass levels one week Post-SWAP.

A: Schematic of the dietary regimen of the 3 different groups of mice (n=5 mice per group). B and C: Body mass of individual mice was recorded daily. D and E: Food consumption of individual mice was recorded daily. One-way analysis of variance was used to identify significant differences at each time point. Asterisk (*) indicates different from RD mice; # indicates different from SWAP mice (p < 0.05).

Figure 4.. RNA-Seq Revealed a unique transcriptomic profile at one week post-SWAP.

A: Heat map representation of the differential gene regulation occurring in the VIS adipose of the 3 experimental groups (n=5 mice per RD and HFD group and n=4 mice per SWAP group). B: A Venn diagram representation of 759 genes that were identified to exhibit a 1.25-fold or greater difference among groups. C: Principle component analysis of the transcriptional profiles of the VIS adipose from the 3 experimental groups. D: Heat map representation of 16 genes that have been identified to be differentially regulated within the VIS adipose in response to diet composition and have well-established, critical roles in adipose function. Heat maps illustrate relative expression within a given gene.

Figure 5.. Expression of 21 ECM mRNAs were differentially regulated.

A: Heat map representation of 21 genes that were identified to be differentially regulated within the VIS adipose in response to diet composition and are known components or residents of the ECM. Heat maps illustrate relative expression within a given gene. B: VIS and SUBQ adipose from mice one week post-SWAP were fixed and embedded in paraffin. Tissue slices were deparaffinized and stained with picro sirius red to image collagen within the tissue. Black arrows indicate wider and/or more diffuse cell borders relative to RD mouse adipose. Black lines denote 100μm. C: Total cells per frame. One-way analysis of variance was used to identify significant differences at each time point. * - different from RD mice; # - different from SWAP mice.

Figure 6.. SWAP Mice Exhibit Unique Regulation of Adipose Inflammation.

A: Heat map representation of the 21 genes that were identified to be differentially regulated within the epididymal adipose (EPI) of SWAP mice in response to diet composition. Heat maps illustrate relative expression for each gene transcript. B: Crispld2 mRNA was quantified by rtPCR in total RNA samples that were used for RNA-Seq and other mice treated to the identical feeding protocol (n=9 mice per group). C and D: EPI and SUBQ from mice one week post-SWAP were fixed and embedded in paraffin. Tissue slices were deparaffinized and immunostained for crown-like structures via α -Mac2, followed by a secondary antibody conjugated with Alexa-647. Two consecutive tissue slices from all animals per group were used to count crown-like structures (n=5 mice per RD and HFD group and n=4 animals per SWAP). White arrows indicate crown-like structures. White lines denote 200μm. E: Ccl2 mRNA was quantified by rtPCR in total mRNA samples that were used for RNA-Seq (n=5 mice per RD and HFD group and n=4 mice for SWAP group). One-way analysis of variance was used to identify significant differences between groups. Asterisk (*) indicates different from RD mice; # indicates different from SWAP mice (p < 0.05).

Figure 7.. CRISPLD2 secretion is regulated by insulin and TNF-alpha.

A: CRISPLD2 protein was immunolabeled as described for CRISPLD2, GLUT4, adiponectin and tubulin (as a loading control) in extracts from in primary cultured adipocytes and stromal vascular cells isolated from epididymal (EPI) or inguinal (SUBQ) fat pads from lean mice (n=3 independent separations of primary adipocytes from stromal vascular cells). B: immunoblot analysis of CRISPLD2 in epididymal adipose (EPI) and inguinal adipose (SUBQ) from RD, HFD and two weeks post SWAP mice. Samples were blotted for tubulin as a loading control C: Histogram shows quantification of CRISPLD2 protein from n=3 mice per group. D: Plasma CRISPLD2 levels from RD, HFD and SWAP mice at week (WK) 0, 1 and 2 following diet swap. Data were analyzed using two-way ANOVA. Asterisk (*) indicates statistically significant difference (p < 0.05) compared to WK 0. E: Da 6 3T3-L1 adipocyte cell associated and serum-free cell media associated CRISPLD2 in cells treated 48 hours without or with 85 nM insulin, 25 μM forskolin (Fsk) or 5 ng/mL TNF-α. The western blot represents one of three independent replicates. Cell lysates were also blotted for tubulin as a loading control. F: Acute, 4 hr secretion of CRISPLD2 and glycerol from day 6 3T3-L1 adipocytes treated without or with 10 μM isoproterenol and 0.5 mM Isobutylmethyl xanthine (ISO/IBMX), n=4 per group. Asterisk (*) indicates statistically significant difference (p < 0.01) of 4-hr basal sample compared to all others. Pound sign (#) indicates significant difference (p < 0.01) of 4-hr ISO/IBMX sample compare to all others.

Figure 8.. Bivariate correlations between insulin sensitivity and fat metabolism variables and CRISPLD2 in adolescents.

Independent variables shown are A) fasting insulin sensitivity (iHOMA2), B) resting fat oxidation as a percentage of energy expenditure (EE), C) serum triglycerides, and D) non esterified fatty acids (NEFA). Closed circles are values for girls, open circles are boys. CRISPLD2 did not differ between sexes. Correlation coefficients and p-values are for the whole cohort.