PTER is a N-acetyltaurine hydrolase that regulates feeding and obesity

Chemicals

dl-Dithiothreitol (DTT) (D0632-1G), taurine (T0625-100G), acetate (S2889-250G), isoleucine (I2752-1G), l-methionine (M5308-25G), l-leucine (L8000-25G), l-valine (V-0500), l-serine (S260-0), l-proline (P0380-100G), l-threonine (T8625-1G), l-alanine (A7627-1G), β-alanine (05160-50G), l-arginine (A5006-100G), l-cysteine (168149-25G), l-glutamic acid (49621-250G), l-glutamine (G-3126), l-histidine (H-8000), l-tryptophan (T0254-5G), l-asparagine (A0884-25G), l-lysine (L5501-5G), acetate (S2889-250G), propionate (P1880-100G), butyrate (B5887-1G), palmitate (P9767-5G), oleate (O7501-1G), stearate (S3381-5G), arachidonate (10931), N-acetyl-l-methionine (01310-5G), N-acetyl-l-leucine (441511-25G), N-acetyl-l-phenylalanine (857459-5G), N-acetyl-l-tyrosine (PHR1173-1G), N-acetyl-l-serine (A2638-1G), N-acetyl-l-proline (A0783-1G), N-acetyl-l-alanine (A4625-1G), N-acetyl-l-arginine (A3133-5G), N-acetyl-l-cysteine (A7250-25G), N-acetyl-l-glutamic acid (855642-25G), N-acetyl-l-glutamine (A9125-25G), N-acetyl-l-tryptophan (A6376-10G), N-acetyl-glycine (A16300-5G), N-acetyl-l-asparagine (441554-1G), N-acetyl-l-lysine (A2010-1G), N-acetyl-l-aspartic acid (00920-5G), chloramphenicol (C0378-25G), spectinomycin dihydrochloride pentahydrate (S4014-25G), apramycin sulfate (A2024-5G), tetracycline hydrochloride (T7660-5G) and ampicillin (A9518) were purchased from Sigma. Paraformaldehyde (AAJ19943K2), tryptone (BP1421-500), yeast extract (BP1422-500), l-tyrosine (A11141.22), glycine (G48-212), N-acetyl-β-alanine (H50208.03) and kanamycin (11815032) were purchased from Thermo Scientific. N-acetyltaurine (35169), lithocholate (20253), α-muricholate (20291), taurocholate (16215), N-palmitoyl-taurine (10005611), N-oleoyl-taurine (10005609), N-stearoyl-taurine (10005610), N-arachidonoyl-taurine (10005537), taurolithocholic acid (17275), tauro-α-muricholic acid (20288) and taurocholic acid (16215) were purchased from Cayman. l-Phenylalanine (A13238), N–N-acetyl-l-isoleucine (H66771), N-acetyl-l-valine (H66943) and N-acetyl-l-histidine (J65657) were purchased from Alfa Aesar. l-Aspartic acid (11625) was purchased from United States Biochemical. N-acetyl-l-threonine (03262) was purchased from Chem-Impex Int’l. Heavy N-acetyltaurine and N-propionyl-taurine were synthesized by Acme. GLP-1 (7–37) peptides (CP0005) were purchased from Genescript. Exendin-3 (9–39) amide (2081) was purchased from Tocris. Recombinant GDF15 (957-GD) was purchased from R&D Systems. Anti-GFRAL neutralizing antibody and control IgG antibody³⁵ were obtained from Eli Lilly, a gift provided by P. Emmerson.

Cell line culture

The HEK293T cell line was obtained from the American Type Culture Collection (ATCC) and grown at 37 °C with 5% CO₂. The culture medium consisted of Dulbecco’s modified Eagle’s medium (Corning, 10-017-CV) with 10% FBS (Corning, 35010CV) and 1:1,000 penicillin–streptomycin (Gibco, 15140-122). For transient transfection, cells were transfected in 10 cm² at about 60% confluency using PolyFect (Qiagen, 301107) and washed with complete culture medium 6 h later. The HEK293T cells were negative following testing for mycoplasma contamination.

Generation of PTER KO cells

The pLentiCRISPRv2 system was used to generate PTER KO HEK293T cells. The single guide RNA (sgRNA) used was 5′-GATGGAACCAGTATCAAGTG-3′. The following oligonucleotides were used to clone the sgRNA into the plentiCRISPRv2 vector: forward, 5′-CACCGGATGGAACCAGTATCAAGTG-3′; reverse, 5′-AAACCACTTGATACTGGTTCCATCC-3′. Lentiviral particles were produced in the HEK293T cell line using PolyFect for the co-transfection of the cloned plentiCRISPRv2 plasmid with the viral packing psPAX2 plasmid and the viral envelope pMD2.G plasmid. A plentiCRISPRv2 plasmid without any sgRNA insert was used as a negative control. Medium containing lentivirus was collected 48 h after transfection and filtered through a 0.45-µm filter. The supernatant was then mixed in a 1:1 ratio with polybrene (Sigma, TR-1003-G) to a final concentration of 8 µg ml^–1 polybrene. The viral mixture was added to HEK293T cells at 40–50% confluence in 6-well plates. Transduced cells were transferred to a 10 cm² plate and subjected to puromycin selection for a period of 3–6 days. Surviving cells were then trypsinized, resuspended and plated at a 10,000× dilution to a new 10 cm² plate. Two weeks later, individually distinguishable colonies were visually identified and then transferred to a 96-well plate using a sterile pipette tip. Finally, single HEK293T cell clones exhibiting complete loss of endogenous PTER protein were confirmed by western blotting using a polyclonal anti-PTER antibody (Invitrogen, TR-1003-G).

Western blotting

For analysis of samples from cell culture, cells were collected and lysed by probe sonication. Cell lysates were centrifuged at 13,000 r.p.m. for 10 min at 4 °C. The supernatant was collected and boiled for 10 min at 95 °C in 4× NuPAGE LDS sample buffer (Thermo Fisher, NP0008) supplemented with 100 mM DTT (Sigma, D0632-1G). For analysis of samples from mice, blood was obtained through submandibular bleeding using a 21 G needle (BD, 305129) into lithium heparin tubes (BD, 365985). Blood was subsequently spun down at 5,000 r.p.m. for 5 min at 4 °C to retrieve the supernatant plasma fractions. All tissue samples were dissected, weighed on a scale, collected into Eppendorf tubes and immediately frozen on dry ice and stored at −80 °C. A stereotaxic device was used to dissect out hypothalamus and brainstem. Adipose tissues were preserved in 4% paraformaldehyde (Fisher Scientific, AAJ19943K2) for histology analysis. Tissues were then mixed with 0.5 ml cold RIPA buffer and homogenized using a Benchmark BeadBlaster homogenizer at 4 °C. The mixture was spun down at 13,000 r.p.m. for 10 min at 4 °C to pellet the insoluble material. The supernatant was quantified using a tabletop Nanodrop One or using a BCA Protein Assay kit (Fisher Scientific, 23250) and analysed by western blotting. Adipose tissues from DIO mice were separately processed using a protein extraction kit to remove lipids (Invent Biotechnologies, AT-022). Proteins were separated on NuPAGE 4–12% Bis-Tris gels and transferred to nitrocellulose membranes. Equal loading was ensured by staining blots with Ponceau S solution. Blots were then incubated with Odyssey blocking buffer for 30 min at room temperature and incubated with primary antibodies (1:1,000 dilution rabbit anti-PTER antibody (Invitrogen, PA5-20750), 1:5,000 dilution rabbit anti-β-actin antibody (Abcam, ab8227), 1:1,000 dilution mouse anti-OxPhoS cocktail antibody (Invitrogen, 45-8099), 1:1,000 dilution rabbit anti-HSL antibody (Novus biologicals, NB110-37253), 1:1,000 dilution rabbit anti-pHSL (Novus biologicals, NBP3-05457), 1:1,000 dilution rabbit anti-ATGL (Cell Signaling, 2138), 1:5,000 dilution mouse anti-α-tubulin antibody (Cell Signaling, 3873S), 1:5,000 dilution mouse anti-Flag antibody (Sigma, F1804-200UG), 1:1,000 dilution rabbit anti-6×His antibody (Abcam, ab9108)) in blocking buffer overnight at 4 °C. Blots were washed three times with PBST (0.05% Tween-20 in PBS) and stained with species-matched secondary antibodies (1:10,000 dilution goat anti-rabbit IRDye 800RD (Li-Cor, 925-68070) and 1:10,000 dilution goat anti-mouse IRDye 680RD (Li-Cor, 925-68070)) at room temperature for 1 h. Blots were further washed three times with PBST and imaged with an Odyssey CLx Imaging System.

Generation of recombinant mouse PTER proteins

The mouse Pter gene (UniProt Q60866) was codon optimized to ensure bacterial expression and was synthesized as gBlocks by IDT. The gene fragment was then inserted into a pET-20b vector containing a carboxy-terminal hexa-histidine (His) tag. DNA sequences encoding a Strep tag were cloned into the amino terminus of Pter for Strep-Tactin-based purification. BL21 competent bacteria (Thermo Scientific, EC0114) were used to transform pET-20b-mouse Pter plasmids and subsequently cultured in LB medium with ampicillin at 37 °C on a shaker overnight. BL21 cells were then transferred to autoinduction medium, which consisted of the following components: 10 g tryptone (Fisher Scientific, BP1421-500), 5 g yeast extract (Fisher Scientific, BP1422-500), 2 ml MgSO₄ (1 M), 1 ml metal solution (0.05 M ferric citrate, 0.02 M CaCl₂, 0.02 M ZnSO₄, 2 µM CoCl₂, 2 µM CuSO₄, 2 µM NiCl₂, 2 µM Na₂MoO₄ and 2 µM boric acid), 20 ml salt solution (167.5 g Na₂HPO₄, 85 g KH₂PO₄, 53.4 g NH₄Cl and 17.8 g Na₂SO₄ in 500 ml water in total) and 20 ml sugar solution (125 g glycerol, 12.5 g glucose and 50 g α-lactose in 500 ml water in total) in a total volume of 1 litre. The bacteria were cultured until the optical density value reached a range of 0.5–0.7. Bacteria were subsequently incubated at 15 °C overnight before being spun down at 8,000 r.p.m. for 30 min at 4 °C. Bacteria were then lysed in PBS through probe sonication on ice to release cytosolic proteins. Soluble fractions were isolated by high-speed centrifugation at 15,000 r.p.m. for 30 min at 4 °C. They were then run down a nickel column using an ÄKTA pure chromatography system. The elution was performed from 0 mM to 300 mM NaCl in PBS over a gradient involving 60 column volumes. Fractions containing mouse PTER proteins were pooled together before undergoing another round of purification. This step involved running fractions down columns loaded with Strep-Tactin resins (IBA, 2-1208-002) following the manufacturer’s instructions. The bound PTER proteins were eluted by 2.5 mM d-desthiobiotin before passing through a HiPrep 16/60 Sephacryl S-200 size-exclusion column (Sigma, GE17-1166-01) in buffer containing 25 mM Tris and 100 mM NaCl. Finally, fractions containing monomeric PTER recombinant proteins were pooled together and subjected to SDS–PAGE gel electrophoresis to ensure >95% purity was achieved. The recombinant proteins were aliquoted and stored at −80 °C for subsequent enzymatic assays.

Enzymatic assays

A total of 100 μg of proteins derived from cell or tissue lysates, or 100 ng of recombinant mouse PTER proteins or 50 µl of chromatography fractions were subjected to incubation in a 50 µl PBS solution (pH 7.4, 0.144 g l^–1 KH₂PO₄, 9 g l^–1 NaCl and 0.795 g l^–1 Na₂HPO₄, Corning, 21-040-CV) at 37 °C for 1 h. For assays using kidney membrane and soluble fractions, total kidney homogenates were transferred into ultracentrifuge inserts and spun at 100,000g on a Beckman centrifuge I8-70M for 1 h at 4 °C. The supernatant was quantified as the kidney soluble fraction and the pellet was resuspended thoroughly in PBS and measured using a using a tabletop Nanodrop One. For assays using Tris-HCl buffer, the pH and salt composition was as follows: pH 7.4, 25 mM Tris-Cl. Next, 100 µM N-acetyltaurine (Cayman, 35169) was added for assaying hydrolysis. For assays testing the substrate scope of mouse PTER hydrolysis, 100 μM N-acetyl-l-isoleucine (Alfa Aesar, H66771), N-acetyl-l-methionine (Sigma, 01310-5G), N-acetyl-l-leucine (Sigma, 441511-25G), N-acetyl-l-valine (Alfa Aesar, H66943), N-acetyl-l-phenylalanine (Sigma, 857459-5G), N-acetyl-l-tyrosine (Sigma, PHR1173-1G), N-acetyl-l-serine (Sigma, A2638-1G), N-acetyl-l-proline (Sigma, A0783-1G), N-acetyl-l-threonine (Chem-Impex Int’l, 03262), N-acetyl-l-alanine (Sigma, A4625-1G), N-acetyl-β-alanine (Thermo Scientific, H50208.03), N-acetyl-l-arginine (Sigma, A3133-5G), N-acetyl-l-cysteine (Sigma, A7250-25G), N-acetyl-l-glutamic acid (Sigma, 855642-25G), N-acetyl-l-glutamine (Sigma, A9125-25G), N-acetyl-l-histidine (Alfa Aesar, J65657), N-acetyl-l-tryptophan (Sigma, A6376-10G), N-acetyl-glycine (Sigma, A16300-5G), N-acetyl-l-asparagine (Sigma, 441554-1G), N-acetyl-l-lysine (Sigma, A2010-1G), N-acetyl-l-aspartic acid (Sigma, 00920-5G), N-propionyl-taurine (Acme, AB38328), N-palmitoyl-taurine (Cayman, 10005611), N-oleoyl-taurine (Cayman, 10005609), N-stearoyl-taurine (Cayman, 10005610), N-arachidonoyl-taurine (Cayman, 10005537), taurolithocholic acid (Cayman, 17275), tauro-α-muricholic acid (Cayman, 20288) and taurocholic acid (Cayman, 16215) were used. For N-acetyltaurine synthesis, 10 mM taurine (Sigma, T0625-100G) and 10 mM acetate (Sigma, S2889-250G) were added. For assays testing the substrate scope of mouse PTER synthesis, 10 mM l-isoleucine (Sigma, I2752-1G), l-methionine (Sigma, M5308-25G), l-leucine (Sigma, L8000-25G), l-valine (Sigma, V-0500), l-phenylalanine (Alfa Aesar, A13238), l-tyrosine (Thermo Scientific, A11141.22), l-serine (Aldrich Chemical, S260-0), l-proline (Sigma, P0380-100G), l-threonine (Sigma, T8625-1G), l-alanine (Sigma, A7627-1G), β-alanine (Sigma, 05160-50G), l-arginine (Sigma, A5006-100G), l-cysteine (Sigma, 168149-25G), l-glutamic acid (Sigma, 49621-250G), l-glutamine (Sigma, G-3126), l-histidine (Sigma, H-8000), l-tryptophan (Sigma, T0254-5G), glycine (FisherChemical, G48-212), l-asparagine (Sigma, A0884-25G), l-lysine (Sigma, L5501-5G) and l-aspartic acid (United States Biochemical, 11625) were individually incubated with 10 mM acetate (Sigma, S2889-250G); 10 mM propionate (Sigma, P1880-100G), 10 mM butyrate (Sigma, B5887-1G) was incubated with 10 mM taurine; 1 mM palmitate (Sigma, P9767-5G), oleate (Sigma, O7501-1G), stearate (Sigma, S3381-5G), arachidonate (Sigma, 10931), lithocholate (Cayman, 20253), α-muricholate (Cayman, 20291) and taurocholate (Cayman, 16215) were individually incubated with 100 mM taurine. Reactions were then quenched and metabolites were extracted by 150 µl of a 2:1 mixture of acetonitrile and methanol. The mixture was spun down at 15,000 r.p.m. for 30 min at 4 °C. The supernatant was subsequently transferred to MS vials for LC–MS analysis.

Molecular docking

The AlphaFold-predicted structure of mouse PTER (AF-Q60866-F1) was used to search for proteins with structural or sequence homology using FoldSeek and BLAST, respectively. The top-predicted structural match from the Protein Data Bank (PDB) database as identified by FoldSeek was PDB 3K2G, a resiniferatoxin-binding protein isolated from Rhodobacter sphaeroides. This crystal structure, along with annotation in UniProt, and metal binding-site prediction using MIB2, all indicated the presence of two zinc ions in the active site of PTER. Molecular docking was performed with CB-Dock2, an online docking server using curvature-based cavity prediction followed by AutoDock Vina-based molecular docking. The substrate compound N-acetyltaurine was prepared as a SDF file, and the AlphaFold-predicted protein structure for PTER was prepared as a PDB file. Ligand–receptor docking was performed using CB-Dock2 following the standard procedure. Ligand–receptor docking results were visually evaluated for biochemical feasibility, and docking results with the lowest Vina score were accepted. The predicted docking poses were evaluated using PyMol (v.3.7), and the predicted active-site residues were identified for mutation.

Mouse PTER mutagenesis

A Q5 Site-Directed Mutagenesis kit (NEB, E0554S) was used to introduce mutations in amino acid residues predicted to have a role in stabilizing zinc ions, interacting with N-acetyltaurine or spatially constraining the active site of mouse PTER. The introduced mutations were subsequently verified through plasmid sequencing conducted by Genewiz.

Activity-guided fractionation

A total of 6 kidneys from 10–14-week-old male C57BL/6J mice were homogenized using a Benchmark BeadBlaster homogenizer at 4 °C. The cytosolic fraction was obtained using high-speed centrifugation at 15,000 r.p.m. for 30 min at 4 °C. Then the mixture was concentrated using 3 kDa filter tubes (Millipore, UFC900324) by spinning down at 4,000 r.p.m. for 1 h. The concentrated sample was diluted 50× into buffer containing 20 mM Tris pH 7.5 before anion exchange on a 1 ml HiTrap Q column (Cytiva, GE29-0513-25). The elution was performed from 0 mM to 500 mM NaCl in 20 mM Tris pH 7.5 over a gradient involving 30 column volumes. Following anion exchange, each fraction was evaluated for N-acetyltaurine hydrolase activity as described above. Three fractions with the highest enzymatic activities were combined, concentrated and subjected to size exclusion on a Superose 6 Increase 10/300 GL column (Cytiva, GE29-0915-96). Each fraction from size exclusion was again evaluated for N-acetyltaurine hydrolase activity. The most active fraction was subjected to LC–MS analysis at the Vincent Coates Foundation Mass Spectrometry Laboratory, Stanford University Mass Spectrometry.

Shotgun proteomics

Samples were reduced with 10 mM DTT for 20 min at 55 °C, cooled to room temperature and then alkylated with 30 mM acrylamide for 30 min. They were then acidified to a pH of about 1 with 2.6 µl of 27% phosphoric acid, dissolved in 165 µl of S-trap loading buffer (90% methanol and 10% 1 M triethylammonium bicarbonate (TEAB)) and loaded onto S-trap microcolumns (Protifi, C02-micro-80). After loading, the samples were washed sequentially with 150 µl increments of 90% methanol and 10% 100 mM TEAB, 90% methanol and 10% 20 mM TEAB, and 90% methanol and 10% 5 mM TEAB solutions, respectively. Samples were digested at 47 °C for 2 h with 600 ng of MS-grade Trypsin/LysC mix (Promega, V5113). The digested peptides were then eluted with two 35 µl increments of 0.2% formic acid in water and two more 40 µl increments of 80% acetonitrile with 0.2% formic acid in water. The four elutions were consolidated in 1.5 ml S-trap recovery tubes and dried by SpeedVac (Thermo Scientific). Finally, the dried peptides were reconstituted in 2% acetonitrile with 0.1% formic acid in water for LC–MS analysis.

MS experiments were performed using an Orbitrap Exploris 480 mass spectrometer (Thermo Scientific) attached to an Acquity M-Class UPLC system (Waters). The UPLC system was set to a flow rate of 300 nl min^–1, for which mobile phase A was 0.2% formic acid in water and mobile phase B was 0.2% formic acid in acetonitrile. The analytical column was prepared in-house with an inner diameter of 100 µm pulled to a nanospray emitter using a P2000 laser puller (Sutter Instrument). The column was packed with Dr. Maisch 1.9 µm C18 stationary phase to a length of approximately 25 cm. Peptides were directly injected onto the column with a gradient of 3–45% mobile phase B, followed by a high-B wash over a total of 80 min. The mass spectrometer was operated in a data-dependent mode using HCD fragmentation for MS/MS spectra generation.

RAW data were analysed using Byonic (v.4.4.1; Protein Metrics) to identify peptides and to infer proteins. A concatenated FASTA file containing UniProt Mus musculus proteins and other probable contaminants and impurities was used to generate an in silico peptide library. Proteolysis with Trypsin/LysC was assumed to be semi-specific allowing for N-ragged cleavage with up to two missed cleavage sites. Both precursor and fragment mass accuracies were held within 12 ppm. Cysteine modified with propionamide was set as a fixed modification in the search. Variable modifications included oxidation on methionine, histidine and tryptophan, dioxidation on methionine and tryptophan, deamidation on glutamine and asparagine, and acetylation on protein N terminus. Proteins were held to a false discovery rate of 1% using standard reverse-decoy technique. Overall, 247 proteins with at least 1 peptide matched in total (Supplementary Table 1). PTER ranked number 6 on the list.

Preparation of mouse tissues for LC–MS analysis

Plasma (50 µl) was mixed with 150 µl of a 2:1 mixture of acetonitrile and methanol and vortexed for 30 s. The mixture was centrifuged at 15,000 r.p.m. for 10 min at 4 °C and the supernatant was transferred to a LC–MS vial. For other mouse tissues, 50 µg of sample was mixed with 150 µl of a 2:1 mixture of acetonitrile and methanol and homogenized using a Benchmark BeadBlaster homogenizer at 4 °C. The mixture was spun down at 13,000 r.p.m. for 10 min at 4 °C to pellet the insoluble material. The supernatant was then transferred to a LC–MS vial.

Measurements of metabolites by LC–MS

Metabolite measurements were performed using an Agilent 6520 Quadrupole time-of-flight LC–MS instrument as previously described²⁹. MS analysis was performed using electrospray ionization (ESI) in negative mode. The dual ESI source parameters were configured as follows: the gas temperature was maintained at 250 °C with a drying gas flow of 12 l min^–1 and the nebulizer pressure at 20 p.s.i.; the capillary voltage was set to 3,500 V; and the fragmentor voltage set to 100 V. The separation of polar metabolites was conducted using a Luna 5 μm NH2 100 Å LC column (Phenomenex 00B-4378-E0) with normal phase chromatography. Mobile phases were as follows: buffer A, 95:5 water and acetonitrile with 0.2% ammonium hydroxide and 10 mM ammonium acetate; buffer B, acetonitrile. The LC gradient was initiated at 100% B with a flow rate of 0.2 ml min^–1 from 0 to 2 min. The gradient was then linearly increased to 50% A/50% B at a flow rate of 0.7 ml min^–1 from 2 to 20 min. From 20 to 25 min, the gradient was maintained at 50% A/50% B at a flow rate of 0.7 ml min^–1. N-acetyltaurine (Cayman, 35169) eluted around 12 min and taurine (Sigma, T0625-500G) was eluted around 13 min under the above conditions. The list of metabolites detected using LC–MS is summarized in Supplementary Table 2. Metabolite data were analysed using Agilent Qualitative Analysis software (v.B.07.00).

General animal information

All animal experiments were performed according to protocols approved by the Stanford University Administrative Panel on Laboratory Animal Care. Mice were maintained in 12-h light–dark cycles at 22 °C and about 50% relative humidity and fed a standard irradiated rodent chow diet. Where indicated, a high-fat diet (D12492, Research Diets 60% kcal from fat) was used. Male C57BL/6J (stock number 000664), male C57BL/6J DIO mice (stock number 380050) and male Mc4r KO mice (stock number 032518) were purchased from the Jackson Laboratory. Whole-body Pter KO mice (catalogue number C57BL/6N(Jax)-Pter^em1(IMPC)Bay) were obtained from the Baylor KOMP2 group of International Mouse Phenotyping Consortium. For intraperitoneal injections of mice with compounds, compounds were dissolved in saline (Teknova, S5825). Compounds were administered to mice every day by intraperitoneal injections at 10 μl g^–1 body weight at the indicated doses. For chronic intraperitoneal injection, oral gavage and subcutaneous injection experiments, mice were mock treated with saline for 3–5 days until body weights were stabilized. For control IgG or anti-GFRAL antibody treatment, mice were subcutaneously injected with 10 mg kg^–1 antibodies once every 3 days. For GLP-1 and exendin-3 injection, GLP-1 and exendin-3 powder was first dissolved in 18:1:1 saline, DMSO and kolliphore and then injected (GLP-1: 2 mg per kg per day, i.p.; exendin-3, 0.1 mg per kg per day, i.p.). Unless specified, compounds were administered at a time of around 18:00. For measuring known feeding-regulating polypeptide hormones, blood plasma was collected at 9:00 and ELISA kits were used following manufacturer’s instructions (leptin: Crystal Chem, 90030; GLP-1: Sigma, EZGLP1T-36K; GDF15: R&D Systems, MGD150; adiponectin: Crystal Chem, 80569; FABP4: Novus Biologicals, NBP2-82410; insulin: Crystal Chem, 90080; ALT: Cayman, 700260; AST: Cayman, 701640; triglycerides: Cayman, 10010303; and NEFA: Cayman, 700310).

Breeding and genotyping of Pter KO mice

Pter KO and WT animals were generated through heterozygous breeding crosses and weaned around postnatal day 21. Genotyping was performed using the following procedures: tail clippings were collected from littermates and boiled for 30 min at 95 °C in 100 μl of 50 mM NaOH to extract genomic DNA. The solution was neutralized by adding 42 μl of 0.5 M Tris (pH 7.5). PCRs were performed by using primers for either the Pter WT allele (forward, 5′-TCATGTCCCACCTTGACAGGTAAGCGGGTC-3′; reverse, 5′-CAGTTGTAGCAGCCATGAACA CTATTGTGC-3′) or Pter KO allele (forward, 5′-GGGTAATATACTTGTCAAACCATGCT-3′; reverse, 5′-CAGTTGTAGCAGCCATGAACA-3′). Promega GoTaq master mix (Promega, PRM7123) was used for the PCR reaction. Each 25 μl reaction consisted of 12.5 μl of the Promega master mix, 2.5 μl of a 10 μM mixture of forward and reverse primers, 2 μl of genomic DNA and 8 μl of ultrapure water. The thermocycling program on a Bio-Rad C1000 Touch Thermo Cycler began with an initial 90 s at 98 °C, followed by cycles of 30 s at 98 °C, 30 s at 58 °C for KO primers and 50 °C for WT primers and 30 s at 72 °C, followed by 5 min at 72 °C and finally held at 4 °C. PCRs for WT primers consisted of 41 cycles, whereas PCRs for KO primers consisted of 35 cycles. Samples were run on a 1.5% agarose gel with 0.1 mg ml^–1 ethidium bromide. WT alleles were expected to produce a PCR product of 699 bp in size, whereas KO alleles were expected to produce PCR products that are 479 bp in size.

Taurine water supplementation

Taurine (2.5% (w/v); Sigma, T0625-500G) was dissolved in mouse drinking water and given to 4-week-old male Pter KO mice and WT mice. Taurine water was freshly prepared every 3 days while mice were on a high-fat diet (D12492, Research Diets 60% kcal from fat). Body weights, food intake and water consumption were measured every 3 days. No adverse effects were observed in mice fed with taurine water.

N-acetyltaurine ex vivo kinetic analysis

Kidneys from 8-week-old Pter KO mice and WT mice were dissected and incubated with 9× (v/w) pre-warmed Williams Medium E (Quality Biological, 112-033-101) supplemented with 5 µM heavy N-acetyltaurine (Acme) at 37 °C on a shaker. Supernatant medium (30 µl) was collected at 0, 15, 30, 45, 60, 90, 120 and 240 min of incubation. Metabolites were extracted and analysed by LC–MS as described above.

Adipose lipolysis in vivo and ex vivo

Blood plasma and epidydimal fat were collected from 4-month-old male DIO C57BL/6J mice receiving saline, N-acetyltaurine (15 mg kg^–1, i.p.) or noradrenaline (0.5 mg kg^–1, i.p.) treatment. Blood glycerol contents were determined using a glycerol quantification kit (Sigma, F6428-40ML). For mature adipocyte lipolysis ex vivo, epidydimal fat from 4-month-old male DIO C57BL/6J mice was dissected and dissociated using 2 mg ml^–1 collagenase B (Worthington, CLSAFB) and 1 mg ml^–1 soybean trypsin inhibitor (Worthington, LS003570). Digested adipose tissues were spun down at 500g for 3 min to isolate the floating layer of mature adipocytes. Around 1 million mature adipocytes were collected and incubated with saline, 50 μM N-acetyltaurine or 1 μM noradrenaline at 37 °C on a shaker for 1 h. Then released glycerol was determined using a glycerol quantification kit (Sigma, F6428-40ML).

Indirect calorimetry and physiological measurements

Male Pter KO mice and WT mice (8–9 weeks old; N = 9 per group) were supplemented with 2.5% (w/v) taurine water and fed on a high-fat diet for 4 weeks. Taurine water was freshly prepared every 3 days when body weights and food intake were measured. Before the body weights of Pter KO mice started to be significantly different from WT mice (4 weeks on taurine water), metabolic parameters including oxygen consumption, carbon dioxide production, RER, food intake and ambulatory movement of mice were measured using the environment-controlled home-cage CLAMS system (Columbus Instruments) at the Stanford Diabetes Center. A separate group of 12–13-week-old male Pter KO mice and WT mice (N = 8 per group) were supplemented with 2.5% (w/v) taurine water and fed on a high-fat diet for 8 weeks before placement into the metabolic cages for analysis. Mice were housed in the metabolic chambers for 36 h before the start of the experiment. Data collected during a complete 24-h day–night cycle were used for analysis. Energy expenditure calculations were normalized for body weight. P values were calculated from two-tailed unpaired t-tests.

Mouse exercise training protocols

A Columbus Instrument animal treadmill with six lanes (Columbus, 1055-SRM-D65) was used for the treadmill running experiments. Before commencing the treadmill running, mice were given a 5-min acclimation period. The initial treadmill running phase began at a speed of 7.5 m min^–1 with a 4° incline, following a previously described procedure²⁹. At intervals of 3 min, both the speed and incline were incrementally increased by 2.5 m min^–1 and 2°, respectively. Once the maximum parameters of 40 m min^–1 in speed and a 30° incline were attained, they remained constant until the mice reached a state of exhaustion, defined as when the mice remained on the shocker at the rear of the treadmill for longer than 5 s. Pter KO mice and WT mice were exercised every other day while on a high-fat diet (60% kcal from fat) for a duration of 6 weeks. Running was performed in the mid-morning for all experiments. Body weights and food intake were measured immediately before each exercise training session.

Glucose tolerance and insulin tolerance tests in mice

For glucose tolerance tests, mice were fasted for 6 h (fasting starting at 7:00 in the morning) and then i.p. injected with glucose at 2 g kg^–1 body weight. Blood glucose levels were measured at 0, 20, 40, 60 and 120 min by tail bleeding using a glucose meter. For insulin tolerance tests, mice were fasted for 6 h (fasting starting at 7:00 in the morning) and then i.p. injected with insulin in saline 0.75 U kg^–1 body weight. Blood glucose levels were measured at 0, 20, 40, 60 and 120 min by tail bleeding using a glucose meter.

hCom2 bacterial strains and culture conditions

Individually cultivated hCom2 strains were obtained from the Microbiome Therapies Initiative. All strains were cultured in one of two growth medium: mega medium and chopped meat medium with rumen fluid and carbohydrates. Cultures were incubated at 37 °C in an anaerobic chamber (Coy Laboratories) in an atmosphere of 5% hydrogen, 10% CO₂ and 85% N₂. Cultures were stored in anaerobically prepared 25% glycerol and water (v/v). All medium and reagents used in the anaerobic chamber were pre-reduced for at least 48 h.

Synthetic community construction

Frozen stocks in 96-well plate matrix tubes were thawed, and 300 µl of each thawed culture was used to inoculate 40 ml of growth medium in 50 ml Falcon tubes. After 72 h, non-normalized cultures of all strains were pooled into a mixture. A 1 ml aliquot of the resulting mixed culture was stored at −80 °C for metagenomic sequencing. The remainder of the mixed culture was subjected to centrifugation (4,700g, 30 min). The cell pellet was washed with an equal volume of pre-reduced sterile PBS and then resuspended in 1/120 of the initial volume of 25% glycerol and water (v/v) solution. Aliquots of the resulting synthetic community were stored in 2 ml cryovials (Corning, 430659) at −80 °C until use.

Gnotobiotic mouse experiments

Germ-free C57BL/6N mice (male, 6–8 weeks of age) were originally obtained from Taconic Biosciences and colonies were maintained in gnotobiotic isolators and fed ad libitum. The Institutional Animal Care and Use Committee at Stanford University approved all procedures involving animals. Glycerol stocks of synthetic communities were thawed and shaken well at room temperature, and mice were orally gavaged with 200 µl of the mixed culture. To ensure efficient colonization by all strains in the community, mice were gavaged using the same procedure twice on different days for all experiments. Mice were fed standard chow (LabDiet, 5k67), fresh faecal pellets were collected weekly at the same time of day and stored at −80 °C before analysis. The mice were maintained on a standard diet (LabDiet, 5k67; 0.2% Trp) for 4 weeks before euthanasia (fed ad libitum). Fresh faecal samples from germ-free mice and hCom2-colonized mice were collected, normalized by weight, homogenized and spun down to isolate live bacteria for in vitro incubation. Mice were euthanized humanely by CO₂ asphyxiation and the plasma was collected in a BD blood tube (BD 365967) and stored on ice. Plasma samples were centrifuged at 16,000g for 20 min, and supernatant was stored in −80 °C until use.

hCom2 in vitro screening

Individually cultivated hCom2 strains were resuspended in a standard amino acid complete (SAAC) medium as previously described³⁶. A volume of 100 µl of cell suspension from each strain was incubated with 10 mM taurine and 10 mM acetate in 300 µl SAAC medium in an anaerobic chamber (Coy Laboratories) in an atmosphere of 5% hydrogen, 10% CO₂ and 85% N₂. Cells were spun down after 48 h of incubation to obtain cell pellets and conditioned medium. Metabolites were extracted and analysed by LC–MS. The optical density at 600 nm before and after incubation was measured. The list of strains screened can be found in Supplementary Table 3.

Antibiotic treatment in mice

Mice (12–14 weeks old) were treated with antibiotic mixture (chloramphenicol (Sigma, C0378-25G), spectinomycin dihydrochloride pentahydrate (Sigma, S4014-25G), apramycin sulfate (Sigma, A2024-5G), tetracycline hydrochloride (Sigma, T7660-5G), kanamycin (Thermo Scientific, 11815032) and ampicillin (Sigma, A9518) at 1 g l^–1 per antibiotic) was administered in drinking water ad libitum and orally gavaged (0.5 ml) every other day for a duration of 2 weeks. Before blood was collected from these mice, fresh faecal samples were collected using sterile pre-weighted Eppendorf tubes and labelled with unique identifiers. Samples were immediately stored at −80 °C until further processing. Faecal samples were normalized by weight, homogenized and filtered before DNA extraction. DNA was extracted from faecal samples using a Qiagen Mini Prep kit following the manufacturer’s protocol. Extracted DNA was stored at −20 °C until quantitative PCR analysis.

Universal bacterial primers targeting the V3 region of the bacterial 16S rRNA were selected (forward HV3-16S primer: 5′CCAGACTCCTACGGGAGGCAG-3′; reverse HV3-16S primer: 5′-CGTATTACCGCGGCTGCTG-3′). Mouse genomic DNA was used as housekeeping gene for quantitative PCR analysis. All reactions were carried out with 10 ng total DNA and SsoAdvanced Universal SYBR Green Supermix (Bio-Rad, 1725274) in CFX Opus 384. Reactions were held at 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 60 s. The number of 16s DNA copies was subsequently determined and normalized to the number of mouse genomic DNA copies in the same faecal sample.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.