Ciliary ARL13B is essential for body weight regulation in adult mice

Abstract

Cilia are near ubiquitous cellular appendages critical for cell-to-cell communication and involved in diverse developmental and homeostatic processes. ARL13B is a regulatory GTPase enriched in cilia. We engineered an Arl13b mouse allele, Arl13b ^V358A , which retains ARL13B biochemical activities but renders ARL13B undetectable in cilia. Surprisingly, these mice are hyperphagic and become obese and insulin resistant. In addition to its GTPase function, ARL13B acts as a guanine nucleotide exchange factor (GEF) for ARL3. To test whether ARL13B's GEF activity is required to regulate body weight, we analyzed the body weight of mice expressing an ARL13B variant lacking ARL3 GEF activity (Arl13b ^R79Q ). We found no difference in body weight, indicating ARL13B is unlikely to regulate weight via its ARL3 GEF activity. Ciliary ARL13B could control energy homeostasis through a role in development or in adult mice. We induced wildtype ARL13B expression, which localizes to cilia, in 4-week-old Arl13b ^V358A/V358A mice and found the obesity phenotype and associated metabolic impairments were rescued, consistent with ARL13B regulating homeostatic signaling within cilia in adult mice. These results show that ciliary ARL13B functions to control body weight. Our ability to genetically control the subcellular localization of ARL13B by removing and introducing it into cilia enables us to define the cilia-specific role of ARL13B and provides key information for understanding how cilia act as a signaling hub critical for energy homeostasis.

Keywords: ARL13B; cilia; energy homeostasis; hyperphagia; obesity.

Figure 1:. ARL13B^V358A protein is undetectable in primary cilia in adult mouse tissues.

(A) Schematic of ARL13B protein domains and relevant amino acid sequence in wildtype and mutant in the cilia localization sequence (CLS). (B) Schematic of ARL13B (green) localization in a wildtype (Arl13b^+/+), (grey) null (Arl13b^hnn/hnn), and cilia-excluded (Arl13b^A/A) models. (C-H) Immunofluorescence of ADCY3 (red) and ARL13B (green) in hypothalamic feeding centers PVN (C-D), VMH (E-F), ARC (G-H) in the mouse brain from Arl13b^+/+ controls and Arl13b^A/A mice. Scale bars 20μm and 5μm for insets indicated by white boxes. Hoechst-stained nuclei are blue.

Figure 2:. Exclusion of ARL13B from cilia leads to obesity.

(A-B) Weekly male and female body weights from 3 to 10 weeks of age: Arl13b^+/+ (males n=14; females n=13), Arl13b^+/A (males n=22; females n=22), Arl13^+/hnn (males n=10; females n=9), Arl13b^A/A (males n=24; females n=26), and Arl13b^A/hnn (males n=15; females n=17). Data are presented as means ± SD. P values were determined by repeated measures ANOVA with Tukey’s multiple comparisons test. *P ≤ 0.01. (C and H) Lean mass and (D and I) fat mass of male and female control (Arl13b^+/A) and Arl13b^A/A. (E and J) Serum leptin levels. (F and K) Food intake. (G and L) Energy expenditure. Data are presented as means ± SD. N numbers indicated on graphs. Lean mass, fat mass, serum leptin, and food intake were analyzed using the Mann-Whitney U test. Energy Expenditure was analyzed by CalR ANCOVA, with body weight included as a covariate. **P < 0.01; ***P < 0.001; ns, not significant with P > 0.05.

Figure 3:. Insulin resistance and impaired glucose metabolism in Arl13b^A/A mice

(A-B) Serum insulin levels in nonfasted 10-week-old male and female mice. Data points represent individual mice. Data are presented as means ± SD. Serum levels were analyzed using the Mann-Whitney U test. *P ≤ 0.05; **P ≤ 0.01. (C-D) Glucose tolerance test in males and females. Blood glucose was measured at indicated times after i.p. glucose injection. (E-F) Insulin tolerance test of males and females. Blood glucose was measured at indicated times after i.p. insulin injection. The control group comprised pooled data from Arl13b^+/+, Arl13b^+/A, and Arl13b^+/hnn mice, and they did not differ in body weight. Data are presented as means ± SD. N numbers indicated on graphs. P values were determined by repeated measures ANOVA with Sidak’s multiple comparisons test. *P ≤ 0.05; **P ≤ 0.01, ***P ≤ 0.001; ****P ≤ 0.0001; ns, not significant with P > 0.05.

Figure 4:. ARL13B’s GEF activity for ARL3 is not required for energy homeostasis.

(A-B) Body weights of adult male and female controls (Arl13b^+/+) and mutants (Arl13b^R79Q/R79Q). Data are presented as means ± SD. N numbers indicated on graphs. ns, not significant with Mann Whitney U test indicating P > 0.05.

Figure 5:. Rescuing ciliary expression of ARL13B in adult mice prevents the metabolic phenotype of Arl13b^A/A mice.

(A) Schematic of the Arl13b-Fucci2a conditional allele with tamoxifen Cre induction timeline. (B-C) Immunofluorescence of ADCY3 (red) and cerulean (cyan). (B) without Cre recombination. (C) with Cre-recombination. (D-E) Longitudinal body weight data of control, Arl13b^A/A, and Rescue (Arl13b^A/A;AF2a; CAGG-CreER^T2). Data are presented as means ± SD. N numbers indicated on graphs. P values were determined by repeated measures ANOVA with Tukey’s multiple comparisons test. *P ≤ 0.01. (F-G) Glucose tolerance test in males and females. Blood glucose was measured at indicated times after i.p. glucose injection. (H-I) Insulin tolerance test in males and females. Blood glucose was measured at indicated times after i.p. insulin injection. N numbers indicated on graphs. Data are presented as means ± SD. P values were determined by repeated measures ANOVA with Tukey’s multiple comparisons test. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; ns, not significant with P > 0.05.