Obeticholic acid, a synthetic bile acid agonist of the farnesoid X receptor, attenuates experimental autoimmune encephalomyelitis Peggy P. Hoa and Lawrence Steinmana,1 aDepartment of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford CA 94305 Contributed by Lawrence Steinman, December 22, 2015 (sent for review December 14, 2015; reviewed by Irun R. Cohen and Howard L. Weiner) Bile acids are ligands for the nuclear hormone receptor, farnesoid X (10). Activation of FXR has been reported to inhibit vascular receptor (FXR). The bile acid–FXR interaction regulates bile acid synthe- smooth muscle cell (VSMC) inflammation by down-regulating the sis, transport, and cholesterol metabolism. Recently, bile acid–FXR reg- proinflammatory enzymes inducible nitric oxide synthase and ulation has been reported to play an integral role in both hepatic and cyclooxygenase-2 expression as well as cell migration into VSMCs intestinal inflammation, and in atherosclerosis. In this study, we found (11). This finding suggests that FXR may be a potential target for that FXR knockout mice had more disease severity in experimental the progressive inflammatory disease, atherosclerosis (11, 12). Re- autoimmune encephalomyelitis (EAE), an animal model of multiple cently, 12 nuclear receptors, including FXR, were shown to be + + + + sclerosis (MS). Obeticholic acid (6α-ethyl-chenodeoxycholic acid, expressedonhighlypurifiedCD4 ,CD8 ,CD19 ,andCD14 cells 6-ECDCA), a synthetic FXR agonist, is an orally available drug that is by quantitative real-time PCR, suggesting that FXR may be currently in clinical trials for the treatment of inflammatory diseases coregulatedinhumanimmune cells (13). such as alcoholic hepatitis, nonalcoholic steatohepatitis, and primary Obeticholic acid (6α-ethyl-chenodeoxycholic acid, 6-ECDCA), biliary cirrhosis. When we treated mice exhibiting established EAE with an orally active synthetic FXR agonist, has ∼100-fold greater 6-ECDCA, or the natural FXR ligand chenodeoxycholic acid (CDCA), clin- FXR agonistic activity than CDCA (14) and does not activate ical disease was ameliorated by (i) suppressing lymphocyte activation other nuclear receptors (15). In animal models, 6-EDCA has + and proinflammatory cytokine production; (ii)reducingCD4 T cells been shown to treat cholestatic liver disease and decrease + and CD19 B cell populations and their expression of negative check- insulin resistance and hepatic steatosis (16–18). Obeticholic point regulators programmed cell death protein 1 (PD1), programmed acid is currently in phase I clinical trials for the treatment death-ligand 1 (PD-L1), and B and T lymphocyte attenuator (BTLA); of alcoholic hepatitis (19), phase II clinical trials for non- + (iii) increasing CD8 T cells and PD1, PDl-1, and BTLA expression; and alcoholic steatohepatitis (NASH) (20) and type 2 diabetes (iv) reducing VLA-4 expression in both the T- and B-cell popula- mellitus (18), and phase III clinical trials of primary biliary tions. Moreover, adoptive transfer of 6-ECDCA– or CDCA-treated cirrhosis (PBC) (19, 21). donor cells failed to transfer disease in naive recipients. Thus, we In this study, we explored the role of FXR in an animal model of show that FXR functions as a negative regulator in neuroinflamma- multiple sclerosis (MS), experimental autoimmune encephalomy- tion and we highlight that FXR agonists represent a potential pre- elitis (EAE). EAE is an inducible inflammatory disease of the cen- + viously unidentified therapy for MS. tral nervous system (CNS) mediated by myelin-specific CD4 Tcells that is augmented by circulating myelin autoantibodies (22–24). experimental autoimmune encephalomyelitis | farnesoid X receptor | Here we show that FXR knockout (FXR-KO) mice exhibit en- bile acid agonist | obeticholic acid | multiple sclerosis hanced EAE disease severity. Moreover, we found that the synthetic FXR agonist, 6-ECDCA, inhibits both active and passive EAE more ile acid synthesis plays a major role in cholesterol homeostasis in effectively than the natural FXR ligand, CDCA, in part by reducing Bthe liver and small intestine (1, 2). Within the liver, bile acids are synthesized from cholesterol and serve as the final end products of Significance cholesterol catabolism. Bile acids are also essential for the adsorp- tion and solubilization of dietary cholesterol and fat-soluble vitamins. Bile acids bind to the nuclear hormone receptor, farnesoid X receptor When high levels of bile acids accumulate in the body, the nuclear (FXR). This bile acid–FXR interaction regulates bile acid synthesis, bile acid receptor, farnesoid X receptor (FXR), is activated by its transport, and cholesterol metabolism. Recently, drugs targeting FXR natural bile acid ligands that include chenodeoxycholic acid activation have been reported to treat both liver and intestinal in- (CDCA), deoxycholic (DCA), and lithocholic (LCA) (3, 4). This flammatory diseases in both animal models and human clinical trials. negatively regulates the transcription of the rate-limiting enzyme, Experimental autoimmune encephalomyelitis (EAE) is an in- cholesterol 7-alpha-hydroxylase (CYP7A1) to inhibit further bile acid flammatory demyelinating disease of the central nervous system synthesis and uptake and to increase the export of bile acids out of and serves as an animal model for multiple sclerosis (MS). In this the cells (1). CDCA is the highest affinity natural ligand for FXR (4). study, we found that FXR knockout mice had more severe EAE, and FXR is mainly expressed in liver, intestine, kidneys, and adrenal treatment of mice with established EAE with a synthetic FXR agonist, gland, with less expression in adipose tissue and heart (5). FXR was obeticholic acid (6α-ethyl-chenodeoxycholic acid, 6-ECDCA), reduced originally identified as a receptor for the cholesterol precursor EAE severity. Thus, we provide an FXR target for development of a farnesol, but is now recognized as the primary receptor for bile acid previously unidentified disease-modifying therapy for MS. (3, 6). The bile acid–FXR interaction plays an integral role in Author contributions: P.P.H. and L.S. designed research; P.P.H. performed research; P.P.H. regulating hepatic inflammation and regeneration, and FXR acti- analyzed data; and P.P.H. and L.S. wrote the paper. vation has been shown to attenuate liver injury in a rodent model Reviewers: I.R.C., Weizmann Institute of Science; and H.L.W., Brigham and Women’s of autoimmune hepatitis (7). In the intestinal tract, bile acids and Hospital. FXR prevent bacterial overgrowth and regulate the extent of in- Conflict of interest statement: P.P.H. and L.S. have filed an invention disclosure with the flammatory responses and pathology from a compromised barrier Stanford Office of Licensing and Technology. (8, 9). Furthermore, FXR activation results in preservation of the Freely available online through the PNAS open access option. intestinal barrier in a rodent model of inflammatory bowel disease 1To whom correspondence should be addressed. Email: [email protected]. 1600–1605 | PNAS | February 9, 2016 | vol. 113 | no. 6 www.pnas.org/cgi/doi/10.1073/pnas.1524890113 Downloaded by guest on October 2, 2021 A Weights before O/N fast Cholesterol Triglycerides responded more robustly to MOG35–55 restimulation than from wild- 20 200 * 500 * type mice (Fig. 1D). 18 150 400 300 16 100 Oral Administration of Obeticholic Acid Is Effective in Treating EAE. mg/dL grams mg/dL 200 14 50 100 Given that the lack of FXR expression worsened disease course 12 0 0 Wildtype FXR-KO Wildtype FXR-KO Wildtype FXR-KO of EAE, we wanted to determine if treatment of EAE with an FXR agonist would in turn be beneficial. Because obeticholic acid HDL LDL 150 80 (6-ECDCA) was recently reported as a promising oral therapeutic 60 100 for the treatment of NASH and PBC, both recognized as in- 40 flammatory diseases, we selected this synthetic FXR agonist as our mg/dL 50 mg/dL 20 0 treatment drug. We initially compared the route of administration 0 -20 of 6-ECDCA intraperitoneally versus orally to determine if one or Wildtype FXR-KO Wildtype FXR-KO both routes would be effective in treating established EAE. Addi- B 6 tionally, to keep the study consistent, all treatment groups were WT dosed both orally and intraperitoneally with the vehicle alone or a 5 FXR-KO combination of vehicle and 6-ECDCA. We found that following 4 randomization of mice at the peak of disease, daily oral dosing of * * * * * * * 6-ECDCA was most effective in attenuating established EAE 3 * * * * A * (Fig. 2 ). 2 * T-cell proliferation to MOG35–55 was significantly reduced in EAE Disease Grade EAE * mice treated orally with 6-ECDCA compared with mice treated 1 * with the vehicle control or treated intraperitoneally with 6-ECDCA 0 (Fig. 2B). Splenocytes harvested from both orally and in- 0 2 4 6 8 101214161820222426 traperitoneally 6-ECDCA-treated mice had reduced IFN-gamma Day post ImmunizaƟon C 3500 Wildtype 3000 FXR-KO 2500 AB 5 Vehicle Randomize & dose daily * 2000 10000 Vehicle * * * * 6-ECDCA i.p. 6-ECDCA i.p. * 4 8000 1500 * 6-ECDCA oral 6-ECDCA oral 3 mean cpm * 6000 * * * * * 1000 * * 2 4000 500 mean cpm 1 EAE disease grade EAE 2000 0 0 0 2 4 6 8 10121416182022242628303234 0 01025 01025 Day post immunizaƟon Concentration of MOG35-55 (g/ml) Concentration of MOG35-55 (g/ml) C 5000 Vehicle 600 Vehicle 6-ECDCA i.p. Fig. 1. FXR knockout mice have more severe EAE. (A) Serum lipid panel 4000 500 6-ECDCA i.p. 6-ECDCA oral * 6-ECDCA oral analysis confirms that FXR-KO mice have elevated serum cholesterol and * * 400 3000 * triglyceride levels after fasting overnight. *P < 0.05 by Student’s t test. 300 2000 = 200 * (B) Clinical EAE scores of wild-type (WT, n 8) versus FXR-knockout (FXR-KO, IL-17 (pg/ml) 1000 100 n = 8) mice immunized with MOG35–55 in CFA with pertussis toxin. Each point ± < – (pg/ml)IFN-gamma 0 0 represents mean SEM, *P 0.05 by Mann Whitney u test.