Targeting activin receptor-like kinase 7 ameliorates adiposity and associated metabolic disorders

Abstract

Activin receptor-like kinase 7 (ALK7) is a type I receptor in the TGF-β superfamily preferentially expressed in adipose tissue and associated with lipid metabolism. Inactivation of ALK7 signaling in mice results in increased lipolysis and resistance to both genetic and diet-induced obesity. Human genetic studies have recently revealed an association between ALK7 variants and both reduced waist to hip ratios and resistance to development of diabetes. In the present study, treatment with a neutralizing mAb against ALK7 caused a substantial loss of adipose mass and improved glucose intolerance and insulin resistance in both genetic and diet-induced mouse obesity models. The enhanced lipolysis increased fatty acid supply from adipocytes to promote fatty acid oxidation in muscle and oxygen consumption at the whole-body level. The treatment temporarily increased hepatic triglyceride levels, which resolved with long-term Ab treatment. Blocking of ALK7 signals also decreased production of its ligand, growth differentiation factor 3, by downregulating S100A8/A9 release from adipocytes and, subsequently, IL-1β release from adipose tissue macrophages. These findings support the feasibility of potential therapeutics targeting ALK7 as a treatment for obesity and diabetes.

Keywords: Adipose tissue; Cytokines; Metabolism.

Figure 1. ALK7 mAb treatment reduces adiposity in ALK7-intact TSOD mice but not in ALK7-deficient counterparts.

TSOD mice and their ALK7-deficient counterparts, T.B-Nidd5/3 mice, were treated with ALK7 mAb (10 mg/kg) or PBS from 5 to 11 weeks of age (A–E). Similarly, TSOD mice were treated with ALK7 mAb or PBS from 5 to 20 weeks of age (F–I). Measurements were made of food intake, BW, adiposity as determined by CT (A and F), fat pad weight (B and G), serum leptin concentration (C and H), ATGL protein expression levels in epiWAT (D), and serum concentrations of NEFA and glycerol and those normalized by weights of epiWAT (E and I). Phenotypic measurement and sample preparation were performed at the end of each cohort, except that food intake and CT were analyzed at 1 week before killing the mice (n = 6 each). The quantification of the ATGL and β-actin protein levels was based on densitometric analyses of immunoblots (n = 9). A representative blot of epiWAT extracts (15 μg of protein) from 3 mice per group is shown (D). ^†P < 0.05, ^††P < 0.01, ^†††P < 0.001 by 2-way ANOVA followed by Tukey multiple comparison. *P < 0.05, **P < 0.01, ***P < 0.001 by t test.

Figure 2. ALK7 mAb treatment reduces adiposity in outbred ddY mice fed an HFD but not in mice fed regular chow.

Outbred ddY mice were fed either an HFD or regular chow (RC) from 4 weeks of age. After HFD-fed mice reached a weight of 45 g, on average at 7 weeks of age, mice started to receive s.c. injections of ALK7 mAb (10 mg/kg) or PBS for 6 weeks (A, n = 4 for food intake and n = 9 or 10 for others; B, n = 9 or 10; C–E, n = 7 or 8) or 18 weeks (F–I, n = 5 or 6). Phenotypic analyses were performed as described in Figure 1. (D) A representative ATGL and β-actin blot of 3 mice per group is shown. CT was analyzed at 5 and 14 weeks after the start of injections for the 6- and 18-week cohorts, respectively. Others were analyzed at the end of the experiments. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 by 1-way ANOVA. *P < 0.05; t test.

Figure 3. Long-term ALK7 mAb treatment improves glucose tolerance and insulin sensitivity in obese mice.

(A–D) TSOD and HFD-fed ddY mice were treated with ALK7 mAb or PBS for 6 (A, n = 8; B, n = 7 or 8 per group) or 18 weeks (C, n = 9; D, n =5 or 6), as described in Figures 1 and 2. Blood glucose levels from the i.p. glucose tolerance test (GTT) after 14 hours of fasting were examined at 5 and 14 weeks after the start of injections for the 6- and 18-week cohorts, respectively, in TSOD (A and C) and ddY mice fed an HFD (B and D). Shown are the glucose levels, glucose infusion rates (GIRs), and total amounts of glucose infused (AUC) during a hyperinsulinemic-euglycemic clamp test in TSOD mice (E) and ddY mice fed an HFD (F) after 17 to 18 weeks of treatment. *P < 0.05, **P < 0.01 by t test. RC, regular chow.

Figure 4. Long-term ALK7 mAb treatment does not increase ectopic fat accumulation in liver or muscle.

TSOD or HFD-fed ddY mice were treated with ALK7 mAb or PBS for the indicated time, as described in Figures 1 and 2. (A and B) Weight (A), and TG content (B) of liver (n = 6 to approximately 10 per group). (C and D) H&E, Sirius Red, and IHC DAB staining against F4/80 (positive staining: brown) of liver. Shown are pictures representative of TSOD (C; n = 4 in 6-week cohort and n = 3 in 15-week cohort) and ddY mouse samples (D; n = 2 in 6-week cohort and n = 3 in 16-week cohort). Scale bars: 50 μm. (E–G) mRNA levels of fibrosis- and inflammation-related genes in liver (E and F) and serum GOT and GPT levels (G) of TSOD (n = 3–4 per group) and ddY mice (n = 6 in 6-week cohort and n = 3 in 16-week cohort). (H) TG content of type IIb–dominant white muscle in lower limbs; gastrocnemius muscle from TSOD mice (n = 3–6) and gastrocnemius muscle from 6-week cohort of ddY mice and tibialis anterior muscle from 18-week cohort of ddY mice (n = 5–8). *P < 0.05, ***P < 0.001 by t test. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, 1-way ANOVA.

Figure 5. ALK7 mAb treatment increases energy expenditure, fat use, and heat production.

(A) HFD-fed ddY mice were treated with ALK7 mAb or PBS, as described in Figure 2. Oxygen consumption (VO₂), CO₂ production (VCO₂), respiratory exchange rate (RER), heat production, and locomotion activity in a metabolic cage during light, dark, and total 24-hour periods were measured after 5 weeks (see first 5 charts; n = 8) and 15 weeks (see second group of 5 charts; n = 5 or 6) of treatment. (B) FAO and mRNA levels of FAO-related genes in type I–dominant soleus muscle isolated from HFD-fed ddY mice treated with ALK7 mAb or PBS for 8 weeks (n = 4 each). *P < 0.05, **P < 0.01, t test.

Figure 6. ALK7 mAb treatment reduces GDF3 expression in ATMs.

(A and B) TSOD or T.B-Nidd5/3 mice (A) and ddY mice fed either an HFD or regular chow (RC) (B) were treated with ALK7 mAb or PBS for 6 weeks, as described in Figures 1 and 2. Then, the SVF was isolated from epiWAT, and the cell differentials were determined by FACS (see first 4 charts). Shown are percentages of ATMs (CD11b⁺F4/80⁺) in SVF cells, CD11c⁺ cells in ATMs, and CD11c⁺ ATMs in SVF cells, and numbers of CD11c⁺ ATMs normalized by fat weight from TSOD or T.B-Nidd5/3 mice (A, n = 5 or 6) and from HFD- or RC-fed ddY mice (B, n = 7 or 8). mRNA levels of NLRP3 (Nlrp3), IL-1β (Il1b), and GDF3 (Gdf3) were determined by real-time reverse transcription PCR (RT-PCR) in epiWAT and its SVF (see the last 3 charts in A and B; n = 5or 6 per group). ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 by 1-way ANOVA.

Figure 7. IL-1β from ATMs and S100A8/A9 from adipocytes mediate a positive feedback loop in the GDF3/ALK7 axis.

(A) CD11c^– ATMs purified by FACS from epiWAT SVF of 7- to 9-week-old TSOD mice were plated in a 24-well dish and incubated with or without the recombinant mature form of mouse IL-1β (10 ng/mL) in the absence of FBS. The PI3K inhibitor wortmannin (100 nM) was added 10 minutes before the addition of IL-1β (see third chart). After a 24-hour culture, GDF3 protein concentrations in the supernatants (sups) (first chart, n = 3) and GDF3 mRNA levels in the cell pellets (second chart, n = 5; third chart, n = 3) were measured by real-time ELISA and RT-PCR, respectively. (B) EpiWAT (0.5 g) isolated from 7-week-old TSOD mice were incubated with or without rrhGDF3 (400 ng/mL) in the presence of 20% FBS. The NLRP3 inhibitor, MCC950 (10 μM), was also added into some wells 30 minutes before the addition of GDF3. After a 24-hour culture, the mRNA levels in the SVF isolated from approximately 0.35 g of epiWAT were determined (n = 3). (C) TSOD mice were treated with ALK7 mAb (10 mg/kg) or PBS for 6 weeks, as described in Figure 1. The expression levels of S100A8/A9 heterodimer protein and those of S100A8 and S100A9 mRNA in epiWAT were determined by ELISA and real-time RT-PCR, respectively (n = 5). (D) CD11c^– ATMs (1.0 × 10⁶ cells/24-well dish) isolated from epiWAT SVF of 7- to 8-week-old TSOD mice were cultured with or without S100A8 or S100A8/A9 at 10 μg/mL. After a 24-hour culture, the GDF3 protein and the mRNA levels were measured (n = 3), as described in A and B. *P < 0.05, **P < 0.01 by t test. ^#P < 0.05, ^##P < 0.01 by 1-way ANOVA.

Figure 8. Signaling pathways of S100A8/A9-induced GDF3 upregulation in ATMs.

(A–G) EpiWAT (0.5 g) isolated from 7-week-old TSOD or T.B-Nidd5/3 mice was incubated with GDF3, as described in Figure 7B. (C–G) ALK7 mAb (C and D; 30 minutes), the S100A8/A9 inhibitor, paquinimod (E; 900 μg/mL, 10 minutes), the RAGE antagonist, FPS-ZM1 (F; 30 μg/mL, 2 hours), or wortmannin (G; 100 nM, 10 minutes) was added at the indicated concentrations prior to the addition of GDF3. After a 24-hour culture, RNA and protein were extracted from 0.05 g and 0.1 g of epiWAT, respectively. SVF was isolated from the remaining approximately 0.35 g of epiWAT. The S100A8/A9 protein levels were measured as described in Figure 7C (A, n = 3; C, n = 4). The mRNA levels in SVF were determined (B, n = 3; D, n = 4; E, n = 4; F, n = 3; G, n = 6). (H) CD11c^– ATMs isolated from epiWAT SVF of 7- to 10-week-old TSOD mice were cultured with or without MCC950 (10 μM) for 30 minutes followed by S100A8/A9 (10 μg/mL) for 24 hour. The GDF3 protein and its mRNA levels were measured (n = 3), as described in Figure 7A. (I) Schematic summary. GDF3 increases production of S100A8/A9 by adipocytes through its receptor ALK7. S100A8/A9 increases production of pro–IL-1β in ATMs. Pro–IL-1β can be cleaved to form bioactive mature IL-1β by NLRP3 inflammasome. S100A8/A9 may also directly enhance GDF3 production by ATMs via activation of PI3K. Secreted mature IL-1β increases GDF3 production by ATMs in autocrine and/or paracrine manners via a PI3K-dependent pathway. As such, GDF3 released from ATMs and ALK7 signals in adipocytes forms a positive feedback loop to drive fat accumulation. ^#P < 0.05, ^##P < 0.01 by 1-way ANOVA. **P < 0.01 by t test.