A Secreted Slit2 Fragment Regulates Adipose Tissue Thermogenesis and Metabolic Function

Abstract

Activation of brown and beige fat can reduce obesity and improve glucose homeostasis through nonshivering thermogenesis. Whether brown or beige fat also secretes paracrine or endocrine factors to promote and amplify adaptive thermogenesis is not fully explored. Here we identify Slit2, a 180 kDa member of the Slit extracellular protein family, as a PRDM16-regulated secreted factor from beige fat cells. In isolated cells and in mice, full-length Slit2 is cleaved to generate several smaller fragments, and we identify an active thermogenic moiety as the C-terminal fragment. This Slit2-C fragment of 50 kDa promotes adipose thermogenesis, augments energy expenditure, and improves glucose homeostasis in vivo. Mechanistically, Slit2 induces a robust activation of PKA signaling, which is required for its prothermogenic activity. Our findings establish a previously unknown peripheral role for Slit2 as a beige fat secreted factor that has therapeutic potential for the treatment of obesity and related metabolic disorders.

Figure 1. Identification of Slit2 as a PRDM16-regulated secreted protein in adipose cells

A. Representative images from UCP1 immunohistochemistry of inguinal subcutaneous adipose tissues from aP2-PRDM16 or wild type mice . Scale bar, 100 µm. B. Normalized thermogenic and beige gene expression in primary inguinal cells from aP2-PRDM16 and wild type mice at day 7 of differentiation. C. Heat map showing relative protein levels in conditioned medium from wild type or ap2-PRDM16 primary inguinal cells (n=2 per group) as determined by TMT labeling and mass spectrometry. Shown is a shortlist of detected proteins. Fold change for each individual sample is color-coded according to the key. For full gene list, see Table S1. D. Gene expression of Slit2 and Slit3 in BAT and iWAT from 6 week-old Balb/c mice housed at 30°C thermoneutrality (TN) or exposed to 4°C for the indicated time points (n=3 per group). E–F. Gene expression of Ap2, Ucp1, Adipsin, F4/80, Slit2 and Slit3 in iWAT (E) and Slit2 and Slit3 in eWAT (F) from C57/b6 mice fed a chow diet or a high fat diet for 16 weeks. G. Primary inguinal cells treated with forskolin for 4h before gene expression analysis of Adiponectin, Ucp1, Slit2 and Slit3. Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 2. Slit2 promotes a thermogenic program in cells and in mice

A–B. Thermogenic gene expression in primary inguinal cells treated for 24h with 1 µg/ml of Slit2 (A) or lysyl oxidase (LOXL1), glypican1 (GPC1), chordin-like 1 (CHL1) or C-X-C motif chemokine 12 (CXCL12) (B) recombinant proteins at day 6 of differentiation. C. Western blotting against Slit2 in primary inguinal cells overexpressing LacZ or Slit2-FL. D. Normalized thermogenic mRNA expression in primary inguinal cells overexpressing Slit2-FL or LacZ. E. Western blotting against Slit2 in primary inguinal cells from Slit2^flox/flox mice transduced with LacZ virus (Slit2^flox/flox) or Cre virus (Slit2^KO). F. Gene expression in primary inguinal cells from Slit2^flox/flox mice transduced with LacZ virus (Slit2^flox/flox) or CRE virus (Slit2^KO). G. Gene expression in BAT tissue from Slit2^flox/flox mice infected with with GFP-AAV8 (Slit2^flox/flox-AAV8-GFP) or Cre virus (Slit2^flox/flox-AAV8-CRE). H. Plasma levels of Slit2 by western blotting in C57/b6 mice injected with adenoviral vectors for Slit2-FL or LacZ at day 7 postinjection. I–L. Normalized mRNA expression levels in liver (I), quadriceps muscle (J), iWAT (K) and BAT (L) at day 7 postinjection. M. Representative images from UCP1 immunohistochemistry of iWAT and BAT from C57/b6 mice injected with Slit2-FL or LacZ at day 7 . Scale bar, 100 µm. Data are presented as mean ±SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3. Identification and characterization of a Slit2 cleavage fragment

A. Western blot of overexpressed full-length C-terminal FLAG-tagged Slit2 detected with a Slit2 antibody (left) and an anti-FLAG antibody (right). Boxed immunoreactive bands were analyzed using mass spectrometry (see methods). B. Matched peptides to Slit2-FL or Slit2-C (bold blue) using C-terminal FLAG-tagged Slit2 overexpression conditioned medium. C. Cloning scheme for Slit2 full length protein, Slit2-N and Slit2-C protein. D–F. Western blotting of overexpressed LacZ, Slit2-N and Slit2-C in primary inguinal cells (D), liver (E) or plasma (F) detected with a V5-HRP antibody.

Figure 4. Slit2-C is sufficient to recapitulate the thermogenic activity of full-length Slit2

A–B. Normalized thermogenic mRNA expression in primary inguinal cells expressing LacZ or Slit2-C (A) or primary brown fat cells overexpressing LacZ, Slit2-C or Slit2-FL (B). C–D. Thermogenic mRNA expression in iWAT (C) and BAT (D) in mice injected with LacZ or Slit2-C adenovirus at day 7. E. UCP1 immunohistochemistry of iWAT (upper panel) and BAT (lower panel) from mice injected with Slit2-C or LacZ at day 7 . Scale bar, 100 µm. F. O₂ consumption in iWAT (left panel) and BAT (right panel) from mice injected with Slit2-C or LacZ at day 7. n =10 per group. Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5. Increased circulating Slit2-C augments whole body energy expenditure and improves glucose homeostasis in obese mice

A–E. Whole body energy expenditure in diet-induced obese C57/b6 mice under 6 days after injection with LacZ or Slit2-C adenovirus. (A) O₂ consumption, (B) respiratory exchange ratio (RER), (C) locomotor activity, (D) accumulated food intake, (E) averaged oxygen consumption at days 5–7 in mice with no significant different in body weight between the groups. F. Tissue weights of BAT, iWAT and eWAT at day 7 postinjection with LacZ or Slit2-C adenovirus. G–H. Body weight (G) and intraperitoneal glucose tolerance test (H) in 16 weeks diet-induced obese mice injected with Slit2-C or LacZ performed at day 7 (n=9–10). Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 6. Slit2-C induces a thermogenesis program through the PKA signaling pathway in adipocytes

A. Primary inguinal cells treated with Slit2-C or LacZ adenovirus at day 2 of differentiation (10⁸ pfu/well), starved over night at day 6 and analyzed at day 7 by western blotting for phospho-and total proteins for epidermal growth factor receptor (EGFR), ERK1/2 and AMPK. B. Primary cells treated as in (A) and blotted for phosphorylated PKA substrates, phospho-and total HSL, UCP1 and α-tubulin. As a positive control, similar samples were treated with 100 nM NE for 30 min. C. Primary cells treated as in (A) and blotted for phospho-and total ATGL and phosphorylated PKC substrates. D. Quantification of UCP1 protein levels relative α-tubulin in (B), n=3. E. Western blot analysis for PKA substrate phosphorylation upon acute treatment (30 min) with conditioned medium from cells expressing LacZ, Slit2-FL or Slit2-C. F–H. Thermogenic gene expression in primary inguinal cells overexpressing Slit2-C or LacZ at day and treated with β-receptor antagonist propranolol (100 nM) for 24h (F), PKA inhibitor H89 (30µM) for 2h (G) or adenylyl cyclase inhibitor SQ-22536 (10 µM) for 24h (H). I. Silverstain of immunopurified Slit2-C FLAG protein compared with an albumin standard. J. Western blot of immunopurified Slit2-C FLAG protein using antibodies for FLAG or Slit2. K. Cell surface binding of FLAG peptide or Slit2-C protein to primary inguinal adipocytes. L. Treatment of primary inguinal cells with 20 nM NE or 20 nM Slit2-C protein for 0, 5, 15, 30, 60 and 90 min. M. Normalized gene expression in primary inguinal cells after treatment with Slit2-C protein for 2h. Comparisons are presented as Slit2-C vs. LacZ (*), LacZ vs. Slit2-C with drug treatment (#) or LacZ vs. drug treatment ($).Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001.