FAMIN Is a Multifunctional Purine Enzyme Enabling the Purine Nucleotide Cycle

Abstract

Mutations in FAMIN cause arthritis and inflammatory bowel disease in early childhood, and a common genetic variant increases the risk for Crohn's disease and leprosy. We developed an unbiased liquid chromatography-mass spectrometry screen for enzymatic activity of this orphan protein. We report that FAMIN phosphorolytically cleaves adenosine into adenine and ribose-1-phosphate. Such activity was considered absent from eukaryotic metabolism. FAMIN and its prokaryotic orthologs additionally have adenosine deaminase, purine nucleoside phosphorylase, and S-methyl-5'-thioadenosine phosphorylase activity, hence, combine activities of the namesake enzymes of central purine metabolism. FAMIN enables in macrophages a purine nucleotide cycle (PNC) between adenosine and inosine monophosphate and adenylosuccinate, which consumes aspartate and releases fumarate in a manner involving fatty acid oxidation and ATP-citrate lyase activity. This macrophage PNC synchronizes mitochondrial activity with glycolysis by balancing electron transfer to mitochondria, thereby supporting glycolytic activity and promoting oxidative phosphorylation and mitochondrial H⁺ and phosphate recycling.

Keywords: C13orf31; Crohn's disease; FAMIN; LACC1; Still's disease; immunometabolism; pH homeostasis; purine metabolism; purine nucleotide cycle; redox homeostasis.

Graphical abstract

Figure 1

FAMIN Is a Purine-Nucleoside-Metabolizing Enzyme (A) Metabolomic library of HepG2 cells after transfection with FAMIN siRNA. Representative total mass spectra (left) separated by molecular weight (m/z), chromatography retention time, and relative levels (right). (B) Change in relative metabolite levels in the library after incubation with recombinant FAMIN^254I or protein buffer control, depicted as volcano plot with unadjusted p values. Red dots, candidate substrates and products whose abundance decreased (a–c) or increased (d–f; n = 3 independent reactions). (C) Representative extracted chromatograms for candidate substrates (top) and products (bottom) by using normalized peak intensity for each given m/z value. (D) Representative mass spectra and extracted chromatograms for compound a and corresponding authentic standard. (E) Levels of adenosine, inosine, hypoxanthine, and ribose-1-phosphate (R1P) within the metabolomic library incubated with FAMIN^254I or protein buffer control (n = 3, mean ± SEM). (F) Levels of adenosine within the metabolomic library incubated with 0.1–100 μg of FAMIN^254I or protein buffer control (n = 3, mean ± SEM). Data representative of at least 3 independent experiments. ^∗p < 0.05 and ^∗∗p < 0.01 (unpaired, two-tailed Student’s t test).

Figure S1

FAMIN Metabolizes Purine Nucleosides, Related to Figure 1 (A) Coomassie SDS-PAGE of recombinant human FAMIN^254I and FAMIN^254V following Strep-Tactin affinity purification. Lanes indicate ladder (L), FAMIN^254I or FAMIN^254V transfected HEK293T lysate input, column flow-through and concentrated protein eluate. (B) Left, size exclusion chromatogram of affinity purified FAMIN that has undergone TEV-cleavage to remove Strep-tag. Blue trace corresponds to A280 (protein) and purple trace to A260 (DNA) signal. Fractions C6-C8 were collected, concentrated, and subjected to Coomassie SDS-PAGE. Inset depicts entire chromatogram. Right, Coomassie SDS-PAGE of fractions obtained from size exclusion chromatography. Lanes indicate ladder (L) and fractions B12, C5, C6, C7, C8 and C9, corresponding to the size exclusion chromatogram, and the concentrated protein from fractions C6-C8. (C) Differential scanning fluorimetry (DSF) of recombinant human FAMIN. (D) Cell proliferation of HepG2 cells silenced for FAMIN (siFAMIN) or transfected with scrambled siRNA (siCtrl) as measured by CyQUANT assay (n = 12). (E) Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of HepG2 cells 48 h after transfection with FAMIN or control siRNA. Basal OCR measurement was followed by sequential treatment (dotted vertical lines) with oligomycin A (Oligo), FCCP, and rotenone plus antimycin A (Rot + ant). Basal ECAR measurement was followed by sequential treatment with oligomycin (Oligo) and 2-deoxyglucose (2-DG) (n = 3). (F) Representative mass spectra and extracted chromatograms for putative FAMIN-catalyzed metabolites and corresponding standards for inosine, hypoxanthine and guanine. (G) Guanosine and guanine levels following incubation of HepG2 cell aqueous extract with 10 μg recombinant FAMIN^254I in 100 μL PBS. (n = 3). (H) Left, Representative extracted chromatograms for FAMIN-catalyzed compound ‘f’ and corresponding standards for ribose-1-phosphate, ribose-5-phosphate, ribulose-5-phosphate and xylulose-5-phosphate. All measurements performed using a BEH amide HILIC column and TSQ Quantiva triple quadrupole. Right, Ratio of selected reaction monitoring (SRM) daughter ions with nominal m/z values of 79 and 97. (I) Inosine, guanosine, cytidine, uridine and ATP levels following incubation of 0.1, 1.0, 10.0 or 100.0 μg of recombinant FAMIN^254I with the complete metabolomic library (aqueous phase of methanol:chloroform extract of FAMIN-silenced HepG2 cells) in 100 μL PBS (n = 3). (J) LC-MS peaks putatively identified as adenine, hypoxanthine, inosine, or ribose-1-phosphate with nominal m/z values of 136, 137, 269 and 229, respectively, were selectively targeted and fragmented using a higher-energy collision dissociation (HCD) collision voltage of 25 eV to give the fragments shown. Data are represented as mean ± SEM or representative of at least 3 independent experiments. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test).

Figure 2

FAMIN Has Adenosine Deaminase, Purine Nucleoside Phosphorylase, and S-Methyl-5′-Thioadenosine (MTA) Phosphorylase Activities (A) Representative extracted chromatograms, using normalized peak intensity, for adenosine (top chromatogram); and inosine, hypoxanthine, and R1P (bottom chromatogram) following incubation of FAMIN^254I or control with 100 μM adenosine. (B) Adenosine, inosine, hypoxanthine, and R1P levels following incubation of recombinant FAMIN^254I or control as per (A) (n = 3). (C) Representative extracted chromatograms for adenine using a modified CSH-C18 method. (D) Fractional conversion of adenosine into its products following incubation of Strep-tagged FAMIN^254I with 100 μM adenosine (n = 3, mean). (E) Adenosine levels in reactions of adenine and R1P in the presence of Strep-tagged FAMIN^254I or control (n = 3). (F) FAMIN-catalyzed enzymatic reactions. (G) FAMIN activity toward purine and pyrimidine nucleosides, measured as substrate (each added at 100 μM) consumption (SAM; S-adenosylmethionine; 2′-dA, 2′-deoxyadenosine; n = 3). (H) Representative extracted chromatograms for inosine, hypoxanthine, and R1P (top chromatogram); and MTA, adenine, and methylthioribose-1-phosphate (bottom chromatogram) upon incubation of inosine and MTA, respectively, with recombinant FAMIN^254I. (I) Adenine and methylthioribose-1-phosphate levels upon incubation of MTA with FAMIN^254I or buffer control (n = 3). (J) Further FAMIN-catalyzed enzymatic reaction. Data represented as mean ± SEM. ^∗p < 0.05 (unpaired, two-tailed Student’s t test).

Figure S2

Characterization of FAMIN Enzymatic Activity, Related to Figure 2 (A) Inosine, hypoxanthine and ribose-1-phosphate levels following incubation of 10 μg recombinant FAMIN^254I or equimolar cholesterol oxidase with 10 μM adenosine for 1 h in 100 μL PBS (n = 3). (B) Adenine, inosine and ribose-1-phosphate levels following incubation of 10 μg recombinant Strep-tagged FAMIN^254I or appropriate controls, including heat-denatured recombinant Strep-tagged FAMIN^254I, with 10 μM adenosine for 1 h in 100 μL PBS or HEPES buffer (n = 3). (C and D) Left, FAMIN-catalyzed enzymatic reactions. Right, levels of guanine or hypoxanthine and ribose-1-phosphate in reactions containing 100 μM guanosine or inosine and recombinant FAMIN^254I or buffer control in 100 μL after 1 h at 37°C (n = 3). (E–I) FAMIN-catalyzed enzymatic reaction with (F) adenine, (G) 2′-deoxyinosine, (H) hypoxanthine and (I) deoxyribose-1-phosphate levels following incubation of 10 μg recombinant Strep-tagged FAMIN^254I or buffer control with 10 μM 2′deoxyadenosine for 1 h in 100 μL PBS (n = 3). (J and K) Adenine (J) and deoxyribose-1-phosphate (K) levels following incubation of 10 μg recombinant Strep-tagged FAMIN^254I or buffer control with 10 μM 5′deoxyadenosine (5′dA) for 1 h in 100 μl PBS (n = 3). (L) MTA levels following incubation of 0.1, 1.0, 10.0 or 100.0 μg of recombinant FAMIN^254I with the complete metabolomic library (aqueous phase of methanol:chloroform extract of FAMIN-silenced HepG2 cells) in 100 μl PBS (n = 3). (M) Phylogenetic tree of FAMIN orthologs using human FAMIN protein sequence as the search input. (N) EC 3.5.4.4 (Adenosine deaminase), EC 2.4.2.1 (purine nucleoside phosphorylase) and EC 2.4.2.28 (MTA phosphorylase) activities of E. coli expressed recombinant full-length FAMIN and FAMIN^Δ176 as measured by inosine, hypoxanthine and adenine production following incubation of protein with 10 μM adenosine, inosine and MTA, respectively, in PBS. Data are represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test).

Figure 3

FAMIN Activities Are Evolutionarily Conserved and Adenosine Phosphorolysis Compromised in I254V (A) Enzyme activities of YlmD and YfiH as measured by inosine, adenine, and hypoxanthine production (n = 3). (B) Crystal structure of YlmD determined in the presence of inosine and phosphate, shown in molecular surface representation with a bound inosine as ball-and-stick. (C and D) Substrate binding site of YlmD, polder Fo-Fc electron density map calculated at 1.5-Å resolution with inosine and bulk solvent omitted. Maps contoured at +3.5 σ (green mesh). (C) Cys¹²⁵-His⁸⁰-His¹⁴² located near inosine’s ribose moiety. The His⁴⁷ side chain inserted into the purine-binding pocket in apo-YlmD (semi-transparent representation). (D) View rotated 45° around the y axis. The hypoxanthine moiety forms a hydrogen bond with the Arg⁵⁹ side chain (dashed line). Selected ordered water molecules (red spheres). (E) 2Fo-Fc electron density map near the bound inosine calculated after refinement with diffraction data to 1.2-Å resolution and contoured at +1.2 σ (blue mesh). Viewing orientation between those in (C) and (D). (F) Representative extracted chromatograms demonstrating oxidation of laccase substrates sinapic and ferulic acid into dimer products after incubation with YlmD, YfiH, Strep-tagged FAMIN^254I, laccase from Trametes versicolor, or appropriate control. (G) Michaelis-Menten kinetics of FAMIN activities for indicated substrates. (H and I) Consumption of adenosine (H) and production of adenine, inosine, and R1P (I) following incubation of Strep-tagged FAMIN^254I, FAMIN^254V, or buffer control with 100 μM adenosine (n = 3). (J) Fractional conversion of adenosine into adenine versus inosine versus hypoxanthine following incubation of adenosine with Strep-tagged FAMIN^254I or FAMIN^254V (n = 3, mean). (K) Inosine monophosphate (IMP), hypoxanthine, and guanine levels in HEK293T cells after transient transfection with FAMIN expression vectors or empty vector (n = 3). Data represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test).

Figure S3

FAMIN Enzymatic Activity Determines Cellular Purine Metabolism and Is Impaired in FAMIN^254V, Related to Figure 3 (A) Laccase enzyme activity of recombinant YlmD, YfiH, FAMIN^254I and Trametes versicolor laccase using 2,2′-azino-bis-3-ethylbenzothiazoline-6-sulphonic acid (ABTS). Please note the right panel’s y axis, which shows a graphical enlargement of the low absorbance readings from the left panel. Data are representative of 2-3 independent experiments. (B) Adenosine levels following incubation of 10 μg Strep-tagged FAMIN^254I, FAMIN^254V or control with 50 μM adenine and 50 μM ribose-1-phosphate for 1 h in 100 μL PBS. (C) Hypoxanthine levels following incubation of 10 μg recombinant Strep-tagged FAMIN^254I or FAMIN^254V with 10 μM adenosine for 1 h in 100 μL phosphate buffered saline (PBS), (n = 3). (D) EC 2.4.2.1 (purine nucleoside phosphorylase) and (E) EC 2.4.2.28 (MTA phosphorylase) activities of Strep-tagged FAMIN^254I and FAMIN^254V as measured by hypoxanthine and adenine following incubation of recombinant protein with 10 μM inosine and methylthioadenosine (MTA), respectively, in PBS (n = 3). (F) Guanosine and S-methyl-5′-thioadenosine levels in control and FAMIN-silenced HepG2 cells 48 h after transfection (n = 6). (G) Immunoblots (IB) for ADA, PNP, MTAP and FAMIN from HepG2 cell lysates silenced for FAMIN (siFAMIN), adenosine deaminase (siADA), purine nucleoside phosphorylase (siPNP), methylthioadenosine phosphorylase (siMTAP) or scrambled siRNA (siCtrl) at 24 h, 48 h or 72 h following transfection; β-ACTIN loading control. (H) Cell proliferation over time of control, FAMIN, ADA, PNP or MTAP silenced HepG2 cells as measured by CyQUANT assay (n = 12). (I) Oxygen consumption rate (OCR) of HepG2 cells transfected with siFAMIN, siADA, siPNP, siMTAP or siCtrl. Basal OCR measurement was followed by sequential treatment (dotted vertical lines) with oligomycin A (Oligo), FCCP, and rotenone plus antimycin A (Rot + ant), (n = 3). Arrow indicates steep decline observed in siFAMIN cells after treatment with FCCP. (J) Cell morphology by light microscopy of HepG2 cells silenced with siFAMIN, siADA, siPNP, siMTAP or siCtrl as observed at 72 h after transfection; scale bar = 125 μm. Data are representative of at least 3 independent experiments. (K–O) A 24h pulse with ¹⁵N₂-glutamine labeled three quarters of AMP in HepG2 cells (n = 3; mean). Since this pulse labeled almost all cellular glutamine and ∼50% of aspartate, the number of incorporated ¹⁵N atoms into AMP allowed (L–M) estimating the proportion of purine de novo synthesis (M+2, M+3, M+4 isotopomers) versus salvage via HPRT (M+1 isotopomer). (N) HepG2 cells engaged both purine salvage and de novo synthesis, but as expected with different kinetics. (O) Terminally differentiated M1Φ, in contrast, exhibited very little de novo synthesis, Levels of M, M+1, M+2, M+3, M+4, M+5 labeled forms of AMP in M1 macrophages after a 24 h pulse with ¹⁵N₂-glutamine (n = 3; mean). The M+1 isotopomer might substantially underestimate salvage, since only half of aspartate is labeled and any HPRT-dependent salvage subsequent to de novo synthesis, or salvage via APRT, would not be captured. This high turnover of the purine ring extends to all essential metabolites and cofactors that contain adenyl groups, i.e., coenzyme A (CoA), acetyl-CoA, flavin adenine dinucleotide (FAD), nicotinamide adenine dinucleotide (NAD), NAD phosphate (NADP), SAM, SAH and MTA (data not shown). (P) Immunoblots (IB) for adenosine deaminase (ADA), purine nucleoside phosphorylase (PNP) and S-methyl-5′-thioadenosine phosphorylase (MTAP) in Famin^+/+ and Famin^–/– and Famin^p.254V and Famin^p.254I M1 macrophages; β-actin, loading control. Data are represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test).

Figure 4

FAMIN Variants Impact on Central Purine Routing (A) IMP levels in control and FAMIN-silenced HepG2 cells 24 h after transfection (n = 3). (B) Inosine and hypoxanthine levels in control and FAMIN-silenced HepG2 cells 48 h after transfection (n = 6). (C and D) Adenine, adenosine, (C) and ATP levels (D) in Famin^p.254I, Famin^p.254V, and Famin^p.284R M1 macrophages (n = 12). (E) Metabolic fate of [¹³C₁₀¹⁵N₅] adenosine after a 3-h pulse of M1 macrophages (n = 6; mean). Schematic representation of central purine metabolism. Adenosine deamination into inosine releases ¹⁵N as ammonia, generating a [¹³C₁₀¹⁵N₄] isotopomer (brown). Phosphorolytic cleavage of inosine into hypoxanthine and [¹³C₅] R1P, yielding the [¹³C₅¹⁵N₄] isotopomer (blue). Adenosine conversion to AMP without loss of label (purple). Phosphorolytic cleavage of fully labeled MTA generates [¹³C₅¹⁵N₅] adenine (green) and [¹³C₅] 5′-methylthioribose-1-phosphate. Fractions of differently labeled states (averaged across Famin^p.254I, Famin^p.254V, and Famin^p.284R genotypes) depicted as pie charts. ADA, adenosine deaminase; ADK, adenosine kinase; APRT, adenine phosphoribosyl transferase; HPRT, hypoxanthine-guanine phosphoribosyl transferase; MTAP, MTA phosphorylase; PNP, purine nucleoside phosphorlyase. (F–H) Fraction of guanosine (F), guanine (G), or GTP (H) labeled as the indicated isotopomer in M1 macrophages after a [¹³C₁¹⁵N₂] guanosine pulse (n = 6). (I) Metabolite levels (gray dots) in Famin^p.254I versus Famin^p.284R M1 macrophages depicted as volcano plot. False discovery rate (FDR)-controlled LC-MS features (black dots), select metabolites in red (n = 6). (J) Immunoblots (IBs) with indicated antibodies in M1 macrophages (n = 3). Data represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test or one-way ANOVA).

Figure S4

FAMIN Affects Routing through Central Purine Metabolism, Related to Figure 4 (A) Methylthioadenosine levels in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages (n = 12). (B) Fraction of inosine, adenine and hypoxanthine labeled as the indicated isotopomer in M1 macrophages after a 3 h pulse with [¹³C₁₀¹⁵N₅] adenosine (n = 6). (C) Metabolic fate of stable isotope-labeled [¹⁵N₅] adenine after a 3 h pulse of M1 macrophages (n = 6; mean). Schematic representation of central purine metabolism. AMP can be generated from adenine via sequential FAMIN and ADK activities or via APRT without loss of label (purple). Labeled [¹⁵N₅] adenosine can be deaminated into inosine by FAMIN or ADA, with a loss of a single ¹⁵N as ammonia, generating a [¹⁵N₄] isotopomer (brown). Routes of interconversion and relationship with other metabolic pathways are also illustrated. Fractions of different labeled states averaged across genotypes following the 3 h pulse with [¹⁵N₅] adenine are depicted as pie charts; asterisks indicate metabolites with significantly altered isotopic labeling across genotypes as depicted in Figures S4D and S4E. Adenine phosphoribosyl transferase (APRT); adenosine kinase (ADK); adenosine deaminase (ADA); cytosolic nucleotidase (cN); hypoxanthine-guanine phosphoribosyl transferase (HPRT); purine nucleoside phosphorylase (PNP); S-methyl-5′-thioadenosine phosphorylase (MTAP); S-adenosylhomocysteine hydrolase (SAHase). (D) [¹⁵N₅] adenine levels in M1 macrophages after a 3 h pulse with [¹⁵N₅] adenine (n = 6). (E) Fraction of adenosine labeled as the [¹⁵N₅] isotopomer in M1 macrophages after a 3 h pulse with [¹⁵N₅] adenine (n = 6). (F) Fraction of ATP, NAD⁺, NADH, FAD, acetyl-CoA, HMG-CoA and succinyl-CoA labeled as the indicated isotopomer in M1 macrophages after a 3 h pulse with [¹³C₁₀¹⁵N₅] adenosine. Fractions of different labeled states (averaged across Famin^p.254I, Famin^p.254V and Famin^p.284R genotypes) are depicted as pie charts. Asterisks indicate significantly altered isotopic labeling across Famin genotypes (n = 6 per genotype). Data are represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test or one-way ANOVA).

Figure S5

FAMIN Activity Affects Purine-Containing Cofactor Turnover, Related to Figure 4 (A) Fraction of ATP, NAD⁺, NADH, FAD, acetyl-CoA, HMG-CoA and succinyl-CoA labeled as the indicated isotopomer in M1 macrophages after a 3 h pulse with [¹⁵N₅] adenine. Fractions of different labeled states (averaged across Famin^p.254I, Famin^p.254V and Famin^p.284R genotypes) are depicted as pie charts. Asterisks indicate significantly altered isotopic labeling across Famin genotypes (n = 6 per genotype). (B) Total, unlabelled and [¹³C₁¹⁵N₂] guanine and guanosine levels in M1 macrophages after a 3 h pulse with [¹³C₁¹⁵N₂] guanosine (n = 6). (C) Differential metabolite levels in Famin^p.254I versus Famin^p.254V M1 macrophages. Data depicted as a volcano plot using p value and log₂ fold change. Grey dots are non-significant, while black dots correspond to LCMS features with significantly altered abundance following Benjamini-Hochberg correction for multiple testing. Differential metabolites shown in red were confirmed and identified as indicated (n = 6). (D) Cytoplasmic pH measured using pHrodo in Famin^p.254I, Famin^p.254V, Famin^p.284R M0 macrophages (n = 18) (E) Cytoplasmic pH measured using BCECF with dual excitation at 440nm and 490nm in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages (n = 18). Reduced 490:440 ratio corresponds to lower (more acidic) pH_c. (F) Cytoplasmic pH measured using pHrodo in control and FAMIN-silenced HepG2 cells 48 h after transfection (n = 9). (G) Heatmap of metabolites in control and FAMIN-silenced HepG2 cells 24 h after transfection (n = 3). Data are represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test or one-way ANOVA).

Figure 5

FAMIN Activity Controls Cellular pH and Enables a Purine Nucleotide Cycle (A) Heatmap of differentially expressed genes by Famin genotype in M0, M1, and M2 macrophages (n = 5). (B) Gene set enrichment analysis in Famin^p.254I compared to Famin^p.284R M0 macrophage transcriptomes; Gene Ontology pH regulation gene set (n = 5). (C) Cytoplasmic pH measured by pHrodo in M1 macrophages (n = 18). (D) Inorganic phosphate levels in M1 macrophages (n = 9). (E) Schematic of the purine nucleotide cycle (PNC; blue boxes and circles) with phosphorolysis, deamination, and salvage routes involving FAMIN. AMPD, AMP deaminase; ADSS, adenylosuccinate synthase; ADSL, adenylosuccinate lyase; AK1, adenylate kinase. (F) Oxygen consumption rate (OCR) of Famin^p.254I M0 macrophages silenced for Adsl or Adss or transfected with siRNA control. Basal OCR followed by (dotted vertical lines) oligomycin A (Oligo), FCCP, and rotenone plus antimycin A (Rot + ant) (n = 3). (G) OCR (left), extracellular acidification rate (ECAR) (middle), and maximal respiratory capacity (right) of Famin^p.254I M0 macrophages treated with L-alanosine. (n = 3). Data represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (one-way ANOVA).

Figure S6

A Purine Nucleotide Cycle Operates in Macrophages and HepG2 Cells, Related to Figure 5 (A) Heatmap of metabolites in control and FAMIN-silenced HepG2 cells 48 h after transfection (n = 6). (B) Ampd2, Ampd3, Adss, Adsl and Aprt expression in Famin^p.254I, Famin^p.254V, Famin^p.284R M0, M1 and M2 macrophages (n = 5). Ampd: AMP deaminase; Adss: Adenylosuccinate synthase; Adsl: Adenylosuccinate lyase; Aprt: Adenine phosphoribosyltransferase. (C) Extracellular acidification rate (ECAR) of Famin^p.254I M0 macrophages silenced for Adss, Adsl and Ampds (Ampd1, 2 and 3) or transfected with a non-targeting scrambled siRNA control. Basal ECAR measurement was followed by sequential treatment (dotted vertical lines) with oligomycin A (Oligo) (n = 3). Ampd: AMP deaminase; Adss: Adenylosuccinate synthase; Adsl: Adenylosuccinate lyase. (D) Oxygen consumption rate (OCR), and extracellular acidification rate (ECAR) with maximal respiratory and glycolytic capacities of control and FAMIN-silenced HepG2 cells treated with 10 μM, 20 μM, 60 μM, 100 μM of L-alanosine or vehicle control for 24 h. Basal OCR and ECAR measurements followed by sequential treatment (dotted vertical lines) with oligomycin A (Oligo), FCCP, and rotenone plus antimycin A (Rot + ant) (n = 3). Data are represented as mean ± SEM.

Figure 6

A FAMIN-Dependent PNC Controls Energy Metabolism (A) Schematic of cellular energy metabolism in context of the PNC. PNC enzymes (blue circles), other enzymes (yellow circles), and transporters (gray circles) with gene names. Electron transfer from glycolysis to mitochondria by the glycerol-3-phosphate (G3PS; connected with red dotted lines) and malate-aspartate shuttle (MAS; connected with red dashed lines). Filled red circles depict fate of [¹³C₁₆] palmitic acid (C16:0)-derived carbons through fatty acid oxidation (FAO) into tricarboxylic acid (TCA) cycle citrate and by ATP citrate lyase (ACLY) into fatty acid synthesis (FASN); empty circles depict route of carbons from TCA oxaloacetate; labeling of α-ketoglutarate and succinyl-CoA depicted for M0 macrophages with intact TCA cycle. Complex II (CII) forward and reverse activity, blue and red arrowed arcs, respectively. CoQ, coenzyme Q; CI, complex I; branched chain amino and keto acids (BCAA, BCKA); DHAP, dihydroxyacetone phosphate; G-6-P, glucose-6-phosphate; F-6-P, fructose-6-phosphate; F-1,6-P, fructose-1,6-bisphosphate; Ga-3-P, glyceraldehyde-3-phosphate; 1,3-BPG, 1,3-bis-phosphoglycerate; 3-PG, 3-phosphoglycerate; 2-PG, 2-phosphoglycerate; 2-PEP, 2-phosphoenolpyruvate. (B) OCR (left) and ECAR (right) of M0 macrophages treated (dotted line + fill) with L-alanosine for 3 h or vehicle (solid line). Basal OCR and ECAR followed by (arrows) oligomycin A (Oligo), FCCP, and rotenone plus antimycin A (Rot + ant) (n = 3). (C) ECAR of M1 macrophages treated (dotted line + fill) with L-alanosine or vehicle (solid line). Basal ECAR followed by (arrows) Oligo and 2-deoxyglucose (2-DG) (n = 3). (D) Cytoplasmic pH (pH_c) measured using pHrodo in M1 macrophages treated with L-alanosine or vehicle (n = 12 from 3 mice per genotype). (E and F) Fraction of aspartate (E) or fumarate (F) labeled as the indicated isotopomer in M0 macrophages after a [¹³C₁₆] palmitate pulse (n = 6). (G) OCR (left) and ECAR (right) of M0 macrophages treated (dotted line + fill) with SB204990 (ACLY inhibitor) or vehicle (solid line). (n = 3). Data represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (one-way ANOVA).

Figure S7

The Purine Nucleotide Cycle Is Linked to FAO, and FAMIN Deficiency Can Be Rescued by Exogenous Fumarate, Related to Figures 6 and 7 (A and B) Fraction of citrate labeled as the indicated isotopomer in Famin^p.254I, Famin^p.254V, Famin^p.284R M0 and M1 macrophages after a 3 h pulse with 100 μM [¹³C₁₆] palmitate conjugated with BSA at 6:1 ratio (n = 6). (C) Ratio of ¹³C₂-fumarate (M+2) to ¹³C₂-aspartate (M+2) levels in M0 macrophages after a 3 h pulse with 100 μM [¹³C₁₆] palmitate (n = 6). (D) Fractions of different labeled states of citrate, aspartate and fumarate (averaged across Famin^p.254I, Famin^p.254V and Famin^p.284R genotypes) in M1 macrophages after a 3 h pulse with 100 μM [¹³C₁₆] palmitate conjugated with BSA at 6:1 ratio (n = 6) depicted as pie chart. (E) Citrate levels in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages (n = 6). (F) Fraction of aspartate labeled as the indicated isotopomer in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages after a 1 or 3 h pulse with 2 g/L [¹³C₆] glucose (n = 6). (G) Fraction of labeled glutamine, glutamate and aspartate in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages after a 3 h pulse with 2 mM [¹³C₅¹⁵N₂] glutamine (n = 6). (H) Cytoplasmic pH (pH_c) measured using pHrodo in Famin^p.254I, Famin^p.254V, Famin^p.284R M0 macrophages treated with 300 μM of malate or vehicle control for 24 h. Same control group as Figure 7D (n = 6). (I) Cytoplasmic pH (pH_c) measured using pHrodo in HepG2 cells treated with vehicle control or 100 μM of L-alanosine for 24 h, supplemented as indicated with 300 μM fumarate or malate (n = 7). (J) Mitochondrial biomass measured using MitoTracker Green in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages (n = 6). (K) Mitochondrial membrane potential measured using TMRE in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages, and collapsed following FCCP treatment as indicated (n = 9/6). (L–O) Labeling of (L) fumarate, (M) malate, (N) succinate and (O) aspartate in M0 and M1 macrophages after a [¹³C₂] fumarate pulse (n = 6). (P) Fraction of hexose-phosphate labeled as the indicated isotopomer in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages after a 1 h pulse with 2 g/L [¹³C₆] glucose (n = 6). (Q) Glycerol-3-phosphate in Famin^p.254I, Famin^p.254V, Famin^p.284R M1 macrophages (n = 6). Data are represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 (unpaired, two-tailed Student’s t test or one-way ANOVA).

Figure 7

Fumarate Rescues the Impaired FAMIN-Dependent PNC (A) OCR of HepG2 cells treated with L-alanosine or vehicle and supplemented with fumarate or malate. Basal OCR followed by (dotted lines) Oligo, FCCP, and Rot + ant (n = 6). (B) OCR of control and FAMIN-silenced HepG2 cells treated with fumarate (dotted line), malate (dashed line), or vehicle control (solid line + fill). (n = 6). (C) OCR (left) and ECAR (right) of M0 macrophages treated with fumarate (dotted line) or vehicle (solid line + fill) for 8 h. (n = 3). (D and E) pH_c measured using pHrodo in M0 (D) or M1 macrophages (E) treated with malate, fumarate, or vehicle. Right and left panel of (E) share same control group (n = 6). (F) Mitochondrial superoxide measured using mitoSOX in M1 macrophages treated with fumarate or vehicle (n = 12). (G–J) Labeling of citrate (G), succinate (H, J, and K), and aspartate (I) in M0 and M1 macrophages after a [¹³C₂] fumarate (G–I) or [¹³C₄] fumarate pulse (J and K) (n = 6). (L) NAD⁺/NADH ratio in M1 macrophages (n = 12). (M) Fractional labeling of snG-3-P in M1 macrophages after a [¹³C₆] glucose pulse (n = 6). Data represented as mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (one-way ANOVA).