Nardilysin in adipocyte regulates insulin sensitivity via HIF1α and PPARγ

Fig. 1. Improvement in glucose tolerance and insulin sensitivity in adipocyte-specific NRDC knockout mice on a high-fat diet. (A) Body weight trend of Floxed and Adipo-KO mice on a normal chow diet (NCD) or high-fat diet (HFD) (N = 10–11/group). (B,C) Weights of epididymal WAT (B) and liver (C) normalized to body weight for Floxed and Adipo-KO mice (at 18 weeks of age, after 12 weeks of NCD or HFD feeding) (N = 11–12/group) (D,E) Glucose tolerance test (GTT) results for mice on NCD (N = 6/8) (D) and HFD (N = 9/7) (E). * P < 0.05, ** P < 0.001 (F,G) Insulin tolerance test (ITT) results for mice on NCD (N = 9/11) (F) and HFD (N = 15/15) (G). ** P < 0.01, *** P < 0.001 (H) Plasma insulin levels of Floxed and Adipo-KO mice before and 30 min after glucose loading on NCD (N = 7/6) and HFD (N = 5/7).

Fig. 2. Reduction of crown-like structures (CLS) and F4/80-positive cells in Adipo-KO on HFD. (A) Representative H&E-stained sections of eWAT from Floxed and Adipo-KO mice on NCD or HFD. Scale bars, 50 μm. (B) Quantification of CLS per field in eWAT from Floxed and Adipo-KO mice on NCD or HFD. N = 7/group. *** P < 0.001 (C, D) Histogram showing adipocyte size distribution in eWAT from Floxed and Adipo-KO mice on NCD or HFD. In panel (D), the asterisk indicates that each pair of bars (Floxed vs. Adipo-KO) within the marked size bins are significantly different. N = 7/group.*P < 0.05 (E) Quantification of F4/80-positive cells per field in eWAT from Floxed and Adipo-KO mice on HFD, based on immunostaining (N = 6/7 per group). ***P < 0.001 (F) qPCR analysis of Nrdc (Nrd1) and F4/80 (Adgre1) in eWAT from Floxed and Adipo-KO mice (mean +/- SE, N = 7/7/6/7 per genotype. ****P < 0.0001, **P < 0.01.

Fig. 3. Enhanced insulin-induced phosphorylation of AKT in eWAT of Adipo-KO mice and NRDC-knockdown 3T3-L1 adipocytes. (A) Western blot analysis of phospho-AKT (T308) and total AKT in eWAT from Floxed and Adipo-KO mice with or without insulin stimulation for 30 min on either NCD or HFD. (B, C) Quantification of the intensity of p-AKT signal normalized by that of total AKT signal in (A) using ImageJ software (NIH). N = 3/group. **P < 0.01. (D, E) Western blot analysis for phospho-AKT and total AKT in 3T3-L1 adipocytes in which NRDC is knocked down by siRNA (siNRDC1, siNRDC2), with or without insulin treatment for 15 min (D). Quantitative analysis of the intensity of p-AKT signal normalized by that of total AKT for siNRDC1, for siNRDC2 vs. siControl (E). N = 5. *P < 0.05.

Fig. 4. NRDC knockdown impacts cellular metabolism in 3T3-L1 adipocytes. (A) Representative images of NRDC-knockdown 3T3-L1 adipocytes in which lipid droplets were stained with Lipidye 2 fluorescent dye. Scale bar: 500 μm. (B) Representative images of the culture medium and bright-field images of control or anti-NRDC siRNA-transfected 3T3-L1 adipocytes. (C) Representative results of oxygen consumption rate (OCR) in NRDC-knockdown 3T3-L1 adipocytes measured by a Seahorse XF analyzer. (D) Basal and maximal OCR in NRDC-knockdown 3T3-L1 adipocytes. Basal OCR represents baseline OCR minus rotenone/antimycin A-insensitive OCR; maximal OCR represents FCCP-induced OCR minus rotenone/antimycin A-insensitive OCR. N = 16/group. Data presented as mean ± SD; **** and ††††for maximal OCR vs. siControl. P < 0.0001 for basal P < 0.0001. (E) Representative results of OCR in NRDC-knockdown 3T3-L1 pre-adipocytes measured by a Seahorse XF analyzer. (F) Basal and maximal OCR in 3T3-L1 pre-adipocytes. N = 10–12/group. Data presented as mean ± SD; N.S. denotes not significant. ††††siControl. P < 0.0001 for maximal OCR vs. siControl.

Fig. 5. NRDC in adipocytes regulates insulin signaling via HIF1α. (A) Western blot analysis showing the induction of HIF1α under hypoxic condition (3% of O2 for – hours) in 3T3-L1 preadipocytes or differentiated adipocytes transfected with control or anti-NRDC siRNA. (B) Western blot analysis of NRDC-knockdown 3T3-L1 adipocytes treated with or without Daprodustat under normoxic conditions. (C) Quantitative analysis of the intensity of protein signals described in (B). N = 5, Data presented as mean ± SE; pAkt/tAkt (p = 0.13 for siNRDC1, p = 0.11 for siNRDC2 vs. siControl), NRDC/GAPDH (***P < 0.001 vs. siControl under Control conditions, †P < 0.05 vs. siControl under 50µM Daprodustat, ‡‡P < 0.01 vs. siControl under 100µM Daprodustat).

Fig. 6. NRDC associates with PPARγ and may regulate its target genes. (A) Volcano plot showing transcription factor enrichment analysis of differentially expressed genes (DEGs) based on public ChIP-seq datasets from murine WAT, analyzed using the ChIP-Atlas platform. The DEGs were identified by RNA-seq in NRDC-knockdown 3T3-L1 adipocytes. Each point represents a transcription factor: blue indicates enriched factors, and red indicates those with P < 0.05. P values were calculated using a two-tailed Fisher’s exact probability test. The x-axis shows log2-transformed fold enrichment of transcription factors in the DEGs relative to all coding genes. The y-axis shows the -log10-transformed P values for enrichment. The dashed line indicates a significance threshold of P = 0.05; the dotted line indicates the Bonferroni-adjusted threshold (P = 0.05/53). (B) Immunoprecipitation and Western blot analysis showing the formation of a complex of NRDC, PPARγ, and PGC-1α in COS7 cells transfected with expression vectors for NRDC-V5, HA-PGC1α and FLAG-PPARγ. (C) Relative mRNA expression levels of Nrd1, Pparγ, Hif1a, and key Pparγ and Hif1a target genes in eWAT from HFD-fed Floxed and Adipo-KO mice, which…