The immunometabolite itaconate inhibits heme synthesis and remodels cellular metabolism in erythroid precursors

Abstract

As part of the inflammatory response by macrophages, Irg1 is induced, resulting in millimolar quantities of itaconate being produced. This immunometabolite remodels the macrophage metabolome and acts as an antimicrobial agent when excreted. Itaconate is not synthesized within the erythron but instead may be acquired from central macrophages within the erythroid island. Previously, we reported that itaconate inhibits hemoglobinization of developing erythroid cells. Herein we show that this action is accomplished by inhibition of tetrapyrrole synthesis. In differentiating erythroid precursors, cellular heme and protoporphyrin IX synthesis are reduced by itaconate at an early step in the pathway. In addition, itaconate causes global alterations in cellular metabolite pools, resulting in elevated levels of succinate, 2-hydroxyglutarate, pyruvate, glyoxylate, and intermediates of glycolytic shunts. Itaconate taken up by the developing erythron can be converted to itaconyl-coenzyme A (CoA) by the enzyme succinyl-CoA:glutarate-CoA transferase. Propionyl-CoA, propionyl-carnitine, methylmalonic acid, heptadecanoic acid, and nonanoic acid, as well as the aliphatic amino acids threonine, valine, methionine, and isoleucine, are increased, likely due to the impact of endogenous itaconyl-CoA synthesis. We further show that itaconyl-CoA is a competitive inhibitor of the erythroid-specific 5-aminolevulinate synthase (ALAS2), the first and rate-limiting step in heme synthesis. These findings strongly support our hypothesis that the inhibition of heme synthesis observed in chronic inflammation is mediated not only by iron limitation but also by limitation of tetrapyrrole synthesis at the point of ALAS2 catalysis by itaconate. Thus, we propose that macrophage-derived itaconate promotes anemia during an inflammatory response in the erythroid compartment.

Graphical abstract

Figure 1.

Heme biosynthesis and the erythroblastic island. The terminal steps of erythroid development occur while erythroblasts are closely associated with a CM via αβ integrins, VCAM1, erythroblast macrophage protein (EMP), and ICAM4. Heme synthesis in this context is fully activated by the basophilic erythroblast (BasoEB) stage of development. CD163 and CD169 are macrophage-specific cell surface markers. Heme synthesis enzymes are in bold green text. CFU-E, colony forming unit–erythroid; CPOX, coproporphyrinogen III oxidase; HMBS, hydroxymethylbilane synthase; HSC, hematopoietic stem cell; OrthoEB, orthochromatic erythroblast; PBGS, porphobilinogen synthase; PolyEB, polychromatic erythroblast; PPOX, protoporphyrinogen oxidase; ProEB, proerythroblasts; RBC, red blood cell; SCoA, succinyl-CoA; UROD, URO decarboxylase; UROS, uroporphyrinogen III synthase.

Figure 2.

Porphyrin and heme analysis of MEL cells treated with itaconate. Heme (A), PPIX (B), and total carboxyl porphyrins (C,D) were determined by high-performance liquid chromatography for wild-type MEL cells differentiated in dimethyl sulfoxide media for 72 hours. (E) Relative fluorescence of cyclic porphyrins in media from MEL cell cultures overexpressing Homo sapiens ALAS2 (hsALAS2) over 72 hours without dimethyl sulfoxide induction. Itaconate concentrations were 2.5 mM in all experiments shown here. Bars represent means ± 1 standard deviation (n = 3-4 biological replicates as indicated red dots). Multiple Student t tests combined with Bonferroni analysis produced P values <.01 for all phosphate-buffered saline (PBS) vs itaconate comparisons shown. RFU, relative fluorescence units.

Figure 3.

Liquid chromatography–mass spectrometry analysis of differentiating MEL cells treated with labeled itaconate. MEL cultures were induced with dimethyl sulfoxide and treated with phosphate-buffered saline (PBS) or 1 mM labeled itaconate for 72 hours. Select metabolites targeted by liquid chromatography–mass spectrometry analysis are given on the y-axes. Analyses were conducted as described in the Methods section. Bars represent means ± 1 standard deviation (n = 4 biological replicates). Multiple Student t tests combined with Bonferroni analysis produced P values <.01 for all PBS vs ¹³C₅-itaconate for the metabolites (A) ¹³C₅-itaconate, (B) ¹³C₅-itaconyl-CoA, (D) succinyl-CoA, (E) propionyl-CoA, (F) propionyl-carnitine, and (G) inositol, comparisons except (C) succinate (0.024) and (H) AMP (0.034).

Figure 4.

Inhibition of the reverse SCS reaction by itaconate. Succinate but not itaconate at 10 mM concentrations is metabolized by pig heart SCS-GTP (SCS-G) (A) and recombinant SCS-ATP (SCS-A) (B). (C) Itaconate inhibits SCS-A at the indicated ratios using 50 mM succinate in all reactions. Summary bar plots represent mean rates of acyl-CoA formation at 225 to 235 nm. The corresponding UV scans for panels A and B are provided in supplemental Figure 7. Bars represent means and error bars = 1 standard deviation (n = 2-3 biological replicates). Panel A: Student t test P value = .0001. Panel C: one-way analysis of variance P value = .0022 with Bonferroni post hoc test P values = .026, .0022, and .15 (not significant [ns]) for control vs 1:200, control vs 1:50, and 1:200 vs 1:50, respectively.

Figure 5.

SUGCT converts itaconate to itaconyl-CoA. (A) Human SUGCT is homologous to the P aeruginosa itaconyl-CoA transferase (PaIct). Black and gray boxes highlight identical residues and conservative substitutions, respectively. (B) Discontinuous high-performance liquid chromatography (HPLC) assays showed that SUGCT generates itaconyl-CoA from succinyl-CoA and itaconate in vitro. (C) Immunoblotting of MEL mitochondria from cells treated with and without 2.5 mM itaconate and dimethyl sulfoxide (DMSO) for 72 hours. MEL mitochondrial lysates were probed with anti-ALAS2, anti-FECH, and anti-SUGCT primary antibodies. HPLC assays were performed on a C18 reverse-phase column with UV detection (mAU = milli-absorbance units at 254 nm). Optimal autoexposure within the linear dynamic range was performed for each immunoblot on the ChemiDoc imaging system (Bio-Rad). The HPLC chromatogram (B) and western blot (C) shown are representative of n = 2 biological replicates.

Figure 6.

ALAS2 inhibition by itaconyl-CoA. Kinetic progress curves (A) and Lineweaver-Burk plot (B) indicating competitive inhibition of ALAS2 with an inhibitory constant = 100 ± 20 μM (mean ± 1 standard deviation) for itaconyl-CoA. Michaelis constant = 10 ± 2 μM (mean ± 1 standard deviation) for succinyl-CoA. Discontinuous colorimetric assays were conducted for n = 3 biological replicates (separate protein preparations).

Figure 7.

Model of the erythroblastic island during an inflammatory response. Itaconate is produced via IRG1 in immunoactivated blood island macrophages and excreted to the surrounding milieu. Subsequent import and metabolism of itaconate to itaconyl-CoA by proximal red cell precursors result in inhibition of ALA synthesis and cellular hemoglobinization at the point of ALAS2 catalysis. Interleukin 6 (IL6), tumor necrosis factor-α (TNFα), and interleukin 1β (IL1β) are proinflammatory cytokines. GSH, glutathione; IFNGR, interferon-γ (IFNγ) receptor; TLR4, Toll-like receptor 4.