4-Octyl-Itaconate Alleviates Airway Eosinophilic Inflammation by Suppressing Chemokines and Eosinophil Development

Abstract

4-octyl itaconate (4-OI) is a derivative of the Krebs cycle-derived metabolite itaconate and displays an array of antimicrobial and anti-inflammatory properties through modifying cysteine residues within protein targets. We have found that 4-OI significantly reduces the production of eosinophil-targeted chemokines in a variety of cell types, including M1 and M2 macrophages, Th2 cells, and A549 respiratory epithelial cells. Notably, the suppression of these chemokines in M1 macrophages was found to be NRF2-dependent. In addition, 4-OI can interfere with IL-5 signaling and directly affect eosinophil differentiation. In a model of eosinophilic airway inflammation in BALB/c mice, 4-OI alleviated airway resistance and reduced eosinophil recruitment to the lungs. Our findings suggest that itaconate derivatives could be promising therapeutic agents for the treatment of eosinophilic asthma.

Introduction

Asthma is a heterogeneous illness of the lower airways usually associated with bronchial hyper-responsiveness and chronic airway inflammation ¹⁵. It can be classified into different phenotypes based on clinical, physiological, and inflammatory characteristics ¹⁵. One common classification is based on the predominant inflammatory cell type, with eosinophilic and neutrophilic asthma being the two major subtypes ¹⁴. Eosinophilic asthma is characterized by increased levels of eosinophils in the airways, which is associated with a Th2-type immune response, whereas neutrophilic asthma is characterized by increased levels of neutrophils, often with a Th1/Th17-type immune response ¹⁴. In general, eosinophilic asthma is more common, with a prevalence among asthmatics estimated to be 50% in adults and up to 70% in children ³⁸.

T helper (Th)2 cell- and group 2 innate lymphoid cells-derived cytokines, particularly IL-5, are central to the differentiation, recruitment, survival, and degranulation of eosinophils ^26,64,69. IL-5 promotes eosinophil differentiation and maturation from CD34+ hematopoietic progenitor cells ^11,22,71 and synergizes with chemokines to recruit eosinophils to asthmatic airways ^26,39. MIP-1α/CCL3 and macrophage-derived chemokine (MDC)/CCL22 produced by activated macrophages in the lung interstitium and RANTES/CCL5 expressed by epithelial cells of the airway work collaboratively to allow migration of the eosinophils using different chemokine receptors ^40,41. Upon activation, eosinophils release toxic proteins and cytokines which can lead to airway hyperresponsiveness, mucus production, and tissue damage ¹. Additionally, eosinophils interact with other immune cells in the airways, such as T cells and mast cells to further exacerbate airway inflammation ³. Therefore, targeting eosinophils and the associated chemokine pathways has emerged as a promising therapeutic strategy for the treatment of eosinophilic asthma.

Itaconate is a metabolite produced from the Krebs cycle metabolite cis-aconitate in LPS-stimulated macrophages ⁶⁵ by the enzyme cis-aconitase decarboxylase 1 (encoded by the gene ACOD1 or IRG1)⁴³. 4-OI, an esterified form of itaconate, has been extensively utilized as an itaconate mimic because of its increased cell permeability²⁸. Both itaconate and 4-OI display an array of antimicrobial and anti-inflammatory properties ^{28,34,37,44,53,59,66} due to their ability to modify cysteine residues in protein targets through alkylation ^44,56. 4-OI was shown to directly modify cysteine residues on the protein Kelch-like ECH-associated protein 1 (KEAP1), which functions as a negative regulator of the master antioxidant transcription factor nuclear factor erythroid 2-related factor 2 (NRF2) ⁴⁴. Modification of KEAP1 causes its degradation and liberates NRF2 for transcription of antioxidant genes and inhibition of proinflammatory cytokine expression ^32,35,57. Itaconate derivatives have shown therapeutic efficacy in a range of preclinical murine disease models, including psoriasis ⁶, sepsis ⁴⁴, ischemia-reperfusion injury (IRI) ³⁴, gout ²⁸, and pulmonary fibrosis ⁴⁸, as well as in human samples from patients with STING-associated vasculopathy with onset in infancy (SAVI) ⁴⁹, and cryopyrin-associated autoinflammatory syndromes (CAPS) ²⁸.

4-OI has been studied in the context of asthma in which it was effective in alleviating airway resistance in a steroid-resistant, and neutrophilic murine model of lung inflammation, through inhibiting JAK1 signaling ^31,59. In another study, 4-OI was shown to reduce antigen-specific IgE and cytokine levels in a house dust mite (HDM) model of allergic lung inflammation, and the effect was proposed to be through reduced antigen presentation by dendritic cells ³⁰. In a recently published study, underivatized itaconate was also shown to be effective in ameliorating allergic airway inflammation, likely through its suppression of the NLRP3 inflammasome ⁷². Here, we have examined the role of 4-OI in eosinophilic airway inflammation. We showed that 4-OI greatly reduced eosinophil-targeting chemokine production in a range of cell types, including LPS-stimulated M1 macrophages, IL-4-stimulated M2 macrophages, T helper 2 (Th2) cells, as well as IL-1β-activated A549 respiratory epithelial cells. The suppression of these chemokines in M1 macrophages was found to be NRF2-dependent. We also demonstrated that 4-OI directly decreased eosinophil differentiation by interfering with IL-5 signaling. Finally, in an ovalbumin (OVA)-induced eosinophilic lung inflammation mouse model, 4-OI was shown to alleviate airway resistance, OVA-specific IgE, eosinophil recruitment to the lungs, as well as suppress the production of the eosinophil-active chemokines CCL3, CCL5, CCL22, CCL11, and CCL24. Our study indicates the therapeutic potentials of itaconate-targeted derivatives in the treatment of eosinophilic asthma.

Materials and Methods

Reagents

4-OI, Diethyl maleate (DEM), and DMSO were purchased from Sigma-Aldrich. LPS from Escherichia coli EH100 was from Enzo Life Sciences. Recombinant mouse IL-1β and recombinant human IL-4 proteins were purchased from BioLegend.

Mice

6-8 weeks old C57Bl/6J female mice (Harlan UK) were used to isolate murine bone marrow-derived macrophages (BMDMs). Irg1^-/- mice (named C57BL/6N-Acod1^em1(IMPC)J/J) were generated by CRISPR-targeted deletion of exon 4 of Acod1 and were purchased from The Jackson Laboratory. Wild-type littermates were used as controls. Mice were bred and housed in the Comparative Medicine Unit (CMU) at Trinity Biomedical Sciences Institute (TBSI) (Trinity College Dublin, TCD, Ireland). All mice were maintained under specific pathogen-free conditions according to Irish and European Union regulations. All the procedures involving experiments on animals have been approved by the Health Products Regulatory Authority (HPRA, Ireland), and were conducted according to Directive 2010/63/EU of the European Parliament and Council on the protection of animals used for scientific purposes. The in vivo model of allergic lung inflammation was performed with female wild-type BALB/c mice at Centenary institute and all studies were carried out under protocols approved by the Sydney Local Health District Animal Welfare Committee. Female mice were chosen due to more susceptibility to the development of allergic airway inflammation than male mice ⁴².

BMDMs

Bone marrow was obtained by flushing DMEM media through the femur, tibia, and hip bones of mice with a 25-gauge needle. The bone marrow was subsequently resuspended in RBC lysis buffer (Sigma) for 5 min before being centrifuged, resuspended, and passed through a cell strainer (70 μm). Cells were plated in 10 cm Petri dishes and DMEM media containing L-glutamine (Gibco) supplemented with FCS (10%, Sigma), penicillin-streptomycin (1%, Sigma), and L929 supernatant (20%), and cultured at 37° C in a 5% CO₂ incubator for 6 days. On day 6, the cells were scraped, resuspended in DMEM complete media, and seeded into multi-well plates at 0.5 x 10⁶ cells/mL unless otherwise stated.

Bone marrow-derived eosinophils

Bone marrow cells were enumerated to be seeded at 1 x 10⁶ cells/mL and stimulated with Flt-3 ligand (Flt-3L, 100 ng/mL, ImmunoTools) and stem cell factor (SCF, 100 ng/mL, ImmunoTools) in RPMI 1640 containing L-glutamine (Gibco), supplemented with FCS (10%), penicillin-streptomycin (1%), HEPES (25 mM, Sigma), nonessential amino acids (1x, Sigma), sodium pyruvate (1 mM, Sigma), and 2-mercaptoethanol (55 μM, Gibco) from days 0 to 4 at 37° C in a 5% CO₂ incubator. On day 4 and day 8, the medium was replaced with recombinant mouse IL-5 (rmIL-5, 10 ng/mL, ImmunoTools). On day 10, half of the medium was replaced with rmIL-5. 4-OI (125 μM or 250 μM) or an equal amount of DMSO was added to the culture on day 0 or day 4 with IL-5, and cells were harvested for flow cytometry and qPCR analyses on day 4 or day 12.

T cell isolation and culture

Murine resting T cells were purified from total mouse splenocytes by magnetic cell sorting with a CD4⁺ T Cell Isolation Kit (Miltenyi), followed by incubation with CD62L microbeads (Miltenyi), all according to the manufacturer’s instructions. Cells were plated at 0.15 x 10⁶ cells/mL in 96-well round bottom plates and stimulated with plate-bound anti-CD3 (1 mg/mL, BD Biosciences) and anti-CD28 antibodies (2 mg/mL, BD Biosciences), IL-4 (40 ng/ml, Biolegend) and anti-mouse IFN-γ antibody (5 mg/ml, BD Biosciences) in RPMI 1640 containing L-glutamine, supplemented with FCS (10%), penicillin-streptomycin (1%), HEPES (25 mM), nonessential amino acids (1x), sodium pyruvate (1 mM), and 2-mercaptoethanol (55 μM) at 37° C, 5% CO₂ for 48 h. For experiments with 4-OI, cells were pre-incubated at 37° C for 30 min with 4-OI (250 μM), or an equal amount of DMSO (control condition) before activation.

A549 cell culture

Cells were cultured in DMEM media supplemented with FCS (10%) and penicillin-streptomycin (1%) in a T175 flask, at 37°C, 5% CO₂. Cells were grown to 70% confluency before being split or plated out. To detach cells, 10% trypsin-EDTA in PBS was incubated with cells for 15 min at 37°C. The cells were scraped, resuspended in DMEM complete media, and seeded into multi-well plates at 0.25 x 10⁶ cells/mL.

siRNA transfection

Cells were replated at 0.5 x 10⁶ cells/mL in 500 μL serum-free and penicillin-streptomycin-free (SF/PSF) DMEM. The transfection reagent was prepared in another 500 μL SF/PSF DMEM with the required amount of Lipofectamine RNAiMAX transfection reagent (5 μL/mL, Thermo Fischer Scientific) and siRNAs (50 nM) and incubated for 15 mins at room temperature before being added to cells. The Silencer Select siRNA against NRF2 (s70522) and the Silencer Select negative control were from Thermo Fisher Scientific.

mRNA isolation and RT-PCR

RNA extraction from in vitro experiments was carried out using the PureLink RNA minikit (Ambion) according to the manufacturer’s protocol. Eluted RNA was quantified using a Nanodrop 2000 spectrophotometer and each RNA sample was diluted to the lowest yield before RT-PCR. cDNA was prepared using the High-capacity cDNA reverse transcription kit (Applied Biosystems), according to the manufacturer’s protocol. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was performed with the PowerUp SYBR Green Master (Applied Biosystems) on a 7500 Fast thermocycler (Applied Biosystems). Lungs from OVA-sensitized mice were homogenized and total RNA was isolated using TRIzol Reagent (Invitrogen). Reverse transcription was performed using BioScript reverse transcriptase in 1 x first-strand buffer according to the manufacturer’s instructions (Bioline). Real-time qPCR assays were performed with SYBR Green Supermix (KAPA Biosystems) and a CFX384 Real-Time PCR detection system (BioRad). Ct values were converted to 2^{-Δ ΔCt} using the Ct of the housekeeping gene.

Primers were purchased from Eurofins Genomics and custom-designed to cross the exon-exon junctions. For in vitro experiments, all genes were normalized to the housekeeping ribosomal protein S18 (Rps18) for mice or ribosomal protein S13 (RPS13) for humans. For the in vivo study (Figure 5), all genes were normalized to the housekeeping gene hypoxanthine-guanine phosphoribosyl transferase (Hprt). The sequences of the primer pairs for murine genes that were used are as follows; Rsp18, 5’-GGA TGT GAA GGA TGG GAA GT-3’ (forward) and 5’-CCC TCT ATG GGC TCG AAT TT-3’ (reverse); Hprt, 5'-AGG CCA GAC TTT GTT GGA TTT GAA-3’ (forward) and 5'-CAA CTT GCG CTC ATC TTA GGC TTT-3’ (reverse); Ccl3, 5’-GGA AGA TTC CAC GCC AAT TC-3’ (forward) and 5’-TCT GCC GGT TTC TCT TAG TC-3’ (reverse); Ccl5, 5’-GCC CAC GTC AAG GAG TAT TT-3’ (forward) and 5’-CGG TTC CTT CGA GTG ACA AA-3’ (reverse); Ccl22, 5’-CCT CTG CCA TCA CGT TTA GT-3’ (forward) and 5’-ATC TCG GTT CTT GAC GGT TAT C-3’ (reverse); Epx, 5’-GCA ACA ACA AGA AGC ATC CC-3’ (forward) and 5’-AGG AAG CAG GAA GCC ATT AC-3’ (reverse); Fizz1, 5’-ACC TTT CCT GAG ATT CTG CCC C-3’ (forward) and 5’-CAG TGG TCC AGT CAA CGA GTA AGC-3’ (reverse); Gclc, 5’-GCA CGG CAT CCT CCA GTT CCT-3’ (forward) and 5’-TCG GAT GGT TGG GGT TTG TCC- 3’ (reverse); Gclm, 5’-TGG AGT TCC CAA ATC AGC CC-3’ (forward) and 5’-TGC ATG GGA CAT GGT GCA TT-3’ (reverse); Gmcsf (Csf2), 5’-GCC ATC AAA GAA GCC CTG AA-3’ (forward) and 5’-GCG GGT CTG CAC ACA TGT TA-3’ (reverse); Il1b, 5’- GGA AGC AGC CCT TCA TCT TT-3’ (forward) and 5’-TGG CAA CTG TTC CTG AAC TC-3’ (reverse); Il5, 5’- GCT TCC TGT CCC TAC TCA TAA A-3’ (forward) and 5’- CCC ACG GAC AGT TTG ATT CT-3’ (reverse); Il13, 5’- GCT GAG CAA CAT CAC ACA AG-3’ (forward) and 5’- AAT CCA GGG CTA CAC AGA AC-3’ (reverse); Il17, 5'-AAG GCA GCA GCG ATC ATC C-3' (forward) and 5'-GGA ACG GTT GAG GTA GTC TGA G-3' (reverse); Nrf2 (Nfe2l2), 5’-TGG AGT AAG TCG AGA AGT GTT TG-3’ (forward) and 5’-GGA GTT GCT CTT GTC TTT CCT-3’ (reverse); Nqo1, 5’-GCT GCA GAC CTG GTG ATA TT-3’ (forward) and 5’-ACT CTC TCA AAC CAG CCT TT-3’ (reverse); and Pparg, 5’-ACG ATC TGC CTG AGG TCT GT-3’ (forward) and 5’-CAT CGA GGA CAT CCA AGA CA-3’ (reverse). The sequences of the primer pairs for human genes that were used are as follows; RPS13, 5’-TCA CCG TTT GGC TCG ATA TT-3’ (forward) and 5’-GGC AGA GGC TGT AGA TGA TT-3’ (reverse); and CCL5, 5’-TGC TGC TTT GCC TAC ATT GC-3’ (forward) and 5’-CAT CCT TGA CCT GTG GAC GA-3’ (reverse).

ELISA

DuoSet ELISA kits for CCL3, CCL5, and CCL22 were purchased from R&D Systems and ELISA was carried out according to the manufacturer’s instructions with appropriately diluted cell supernatants. Absorbance (450 nm) was then quantified using a FLUOstar Optima plate reader. Corrected absorbance values were calculated by subtracting the background absorbance, and cytokine concentrations were subsequently obtained by extrapolation from a standard curve plotted. In vivo ELISA of CCL3, CCL5, CCL22, CCL11, and CCL24 was performed using Quantikine® purchased from R&D Systems. IgE in plasma was measured using IgE Mouse Sandwich ELISA kit purchased from Thermofisher.

Western Blotting

Cells were washed once with cold PBS and lysed in RIPA buffer (50 mM Tris-HCl pH 7.5, 125 mM NaCl, 1% NP-40, 0.25% Na-deoxycholate, 1mM Na-fluoride, 1 mM Na orthovanadate, phenylmethanesulfonylfluoride [PMSF], and 1% protease inhibitor cocktail), vortexed, and incubated on ice. After 30 min of incubation on ice, lysates were centrifuged at 14.000 rpm for 15 min to remove cell debris. Supernatant containing the protein portion was added with sample buffer (0.125 M Tris [pH 6.8], 10% [v/v] glycerol, 0.02% SDS, and 1 mM dithiothreitol [DTT]) and subsequently heated at 95°C for 5 min. Protein samples were resolved on SDS-PAGE gels and transferred onto a polyvinylidene difluoride (PVDF) membrane via wet transfer. The membranes were probed with primary antibodies at a 1:1000 dilution and secondary HRP-conjugated antibodies at a 1:2000 dilution. Membranes were visualized using WesternBright ECL HRP substrate (Advansta) or SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Scientific) on a ChemiDoc TM Imaging System (Bio-Rad). Images were analyzed with ImageLab (Bio-rad) and ImageJ software. Antibodies used were anti-β-actin (Sigma-Aldrich) and anti-NRF2 (Cell Signaling).

Flow cytometry

For in vitro eosinophil staining, day 4 or day 12 cells were centrifuged, the supernatant was removed, and cells were washed with PBS and resuspended in FACS buffer (0.1% sodium azide, 0.1% BSA, 1.5% FCS in PBS). Eosinophils were then incubated with Fcγ-blocking antibody anti-mouse CD16/32 (1:300, Biolegend) for 10 min before they were stained with fluorochrome-conjugated antibodies for surface proteins in FACS buffer at 4 °C for 30 min in the dark. The antibodies used were: Live/Dead viability dye (1:1000, Thermo Fisher) and eFluoro660 anti-mouse Siglec-F (1:100, Invitrogen). To analyze the proportion of eosinophils in the cell population, cells were gated by FSC vs. SSC to exclude debris, FSC-A vs. FSC-H to exclude duplets, L/D negative to exclude dead cells, and Siglec-F-positive. All samples were analyzed on a FACS Canto II Cell analyzer (BD Biosciences). Analysis of acquired data was performed with the FlowJo software (FlowJo LLC).

In vivo murine model of OVA-induced lung inflammation

6-8 weeks old female BALB/c mice were sensitized to OVA (50μg, intraperitoneal [i.p.] injection, Sigma) in Th2-inducing adjuvant aluminum hydroxide (Alhydrogel, 2%, Jomar) in sterile saline (200 μL). Mice were subsequently challenged intranasally (i.n.) with OVA (10μg/50μL sterile saline) on days 12-15. 4-OI (5 mg/kg or 10 mg/kg) or dexamethasone (DEX) (2 mg/kg, Sigma) was given i.n. on days 13-15. Sham-sensitized controls received saline sensitization with Alhydrogel and the subsequent OVA challenges. Bronchoalveolar lavage fluid (BALF) was obtained by washing the lung lobes twice with 1 mL of HANKS. Lungs were snap-frozen for RNA and ELISA analyses. Blood was recovered and serum-isolated to analyze OVA-specific IgE.

For cell enumeration, BALF samples were centrifuged (300g, 10 min, 4°C), the cell pellet was resuspended in red blood cell lysis buffer (200 μL, 5 min, 4 °C, Tris-buffered NH4Cl) and then centrifuged again. The resulting cell pellet was resuspended in 160 μL of Hank’s Balanced Salt Solution (HBSS, Gibco), and total leukocyte numbers were enumerated using a hemocytometer. Cells were then cytocentrifuged (300 g, 7 min), stained with May Grunwald Giemsa (Sigma), and differential cell counts were performed based on morphologies (~200 cells per cytospin were counted using light microscopy [40x magnification]) ^10,24,62. All slides were coded, and counts were performed in a blinded manner.

Rn was measured by was measured by anesthetizing mice with ketamine (100 mg/kg, Troy Laboratories, Smithfield, Australia) and xylazine (10 mg/kg, Troy Laboratories, Smithfield, Australia) and cannulating their tracheas (tracheostomy with ligation). FlexiVent apparatus (FX1 System, SCIREQ, Montreal, Canada) was used to assess airway-specific resistance (Rn, tidal volume of 8 mL/kg at a respiratory rate of 450 breaths/min) in response to increasing doses of nebulized methacholine (Sigma). Assessments were performed at least three times per dose of saline/methacholine and the average was calculated.

Statistics

Comparisons between two groups were carried out by an unpaired student t-test. Comparisons between multiple groups were performed using one-way ANOVA with Turkey’s multiple comparisons tests, or two-way ANOVA with Šidák’s post-hoc tests. Lung function was analyzed using two-way ANOVA with a Bonferroni post-hoc test. All statistical analyses were performed in Prism software (GraphPad version 9.0) and data are presented as mean ± SEM. P value < 0.05 was considered statistically significant.

Results

4-OI suppresses eosinophil-recruiting chemokine expression and secretion from BMDM

We first examined the effect of 4-OI production on chemokines using LPS as a standard chemokine inducer. The concentrations of 4-OI were chosen within the range of 62.5 to 250 μM, based on previously published studies demonstrating their efficacy and minimal observed toxicity ^28,44,59,60. As shown in Figure 1 A-C, we observed a strong induction of CCL3, CCL5, and CCL22 by LPS, which was reduced by 4-OI. The inhibition of Ccl3 and Ccl5 required 4-OI at a higher dose, however, Ccl22 was repressed by 4-OI at 62.5 μM (Figure 1 A-C). Similar inhibition was observed by ELISA in the supernatant, where 4-OI at 250 μM significantly reduced CCL3, CCL5, and CCL22 (Figure 1 D-F). We next tested whether endogenous itaconate would affect chemokine expression. The transcript levels of these chemokines were not significantly different between BMDMs generated from Irg1^+/+ and Irg1^-/- mice when induced by LPS (Figure 1 G-I). To validate this, we also added itaconate at a supraphysiological concentration of up to 15 mM and could not detect a significant change in chemokine expression (Supplementary Figure 1 A-C). This indicates that 4-OI has an inhibitory role on chemokine production in LPS-stimulated BMDMs, but itaconate itself does not.

Figure 1. 4-OI but not endogenous itaconate inhibits LPS-induced chemokine expression from BMDMs.

BMDMs were pretreated with increasing doses of 4-OI (62.5, 125, and 250 µM) or DMSO control for 1 h before stimulation with LPS (100 ng/ml) for 4 h. (A-C) mRNA levels of chemokines CCL3 (A), CCL5 (B), and CCL22 (C) were measured with qPCR (n = 3). BMDMs were pretreated with 4-OI (250 µM) or DMSO control for 1 h before stimulation with LPS (100 ng/ml) for 4 h. (D-F) The chemokine concentrations of CCL3 (D), CCL5 (E), and CCL22 (F) in the supernatants were quantified by ELISA (n = 3). BMDMs from Irg1^+/+ (WT) and Irg1^−/− mice were stimulated with LPS (100 ng/ml) for 6 h. (G-I) mRNA levels of chemokines CCL3 (G), CCL5 (H), and CCL22 (I) were measured with qPCR (n = 4). Results were obtained from three or four independent experiments. Data are means ± SEM. The p values were calculated using a one-way ANOVA for multiple comparisons and a two-tailed Student t test for paired comparisons.

4-OI suppresses chemokine expression by BMDM in an NRF2-dependent manner

4-OI is known to be a potent activator of NRF2 through modifying crucial cysteine residues on KEAP1, an NRF2 inhibitor, to promote its degradation ⁴⁴. To confirm this finding, we treated BMDMs with increasing doses of 4-OI with or without LPS and observed an enhancement of NRF2 stabilization (Figure 2 A), as well as an activation of a range of NRF2-activated genes, including Nqo1 (Figure 2 B), Gclc (Figure 2 C), and Gclm (Figure 2 D). We next examined whether the inhibitory effect on chemokines by 4-OI was via NRF2. We performed siRNA-guided knockdown of Nrf2 before 4-OI and LPS treatment. As expected, the mRNA levels of NQO1 and NRF2 were significantly reduced with the knockdown (Supplementary Figure 2 A and B). Following Nrf2 knockdown, 4-OI no longer suppressed the expression (Figure 2 E-G) and secretion (Figure 2 H-J) of CCL3, CCL5, and CCL22, however, the loss of inhibition was more prominent at the protein level of CCL3 and CCL22, and less so in CCL5. The NRF2 activator, DEM, also suppressed the secretion of these chemokines (Figure 2 K-M). These results indicate that the capacity of 4-OI to suppress chemokine expression and downstream secretion is dependent on NRF2.

Figure 2. The suppression by 4-OI on LPS-induced chemokines in macrophages is NRF-2-dependent.

BMDMs were pre-treated with increasing doses of 4-OI (62.5, 125, and 250 μM) for 1 h before stimulation with LPS (100 ng/mL) for 4 h. Protein levels of NRF2 and β-actin were analyzed by Western blotting (n = 3) (A). mRNA levels of NRF2-dependent genes Nqo1 (B), Gclc (C), and Gclm (D) were analyzed by qPCR (n = 3). BMDMs were transfected with 50 nM negative control siRNA or Nrf2 siRNA for 8 h. The cells were then treated with 4-OI (250 μM) or DMSO control for 1 h before stimulation with LPS (100 ng/mL) for 4 h. mRNA levels of Ccl3 (E), Ccl5 (F), Ccl22 (G) and protein levels in supernatant of CCL3 (H), CCL5 (I), CCL22 (J) following negative control or Nrf2 siRNA transfection (n = 4 or 5). BMDMs were pre-treated with DEM (100 μM) or DMSO control for 2 h before stimulation with LPS (100 ng/mL) for 4 h. The secretion of CCL3 (K), CCL5 (L), and CCL22 (M) was analyzed by ELISA (n = 3). Results are obtained from 3-5 independent experiments. Data are mean ± SEM. P values were calculated using one-way or two-way analyses of variance (ANOVA) for multiple comparisons.

4-OI inhibits CCL5 and CCL22 in IL-4-stimulated macrophages, Th2 cells, and IL-1β-stimulated A549 cells

Since the principal cell types producing eosinophil-active chemokines include macrophages, lung epithelial cells, and T cells, we next investigated the capacity of 4-OI to repress chemokine production in IL-4 activated M2 macrophages, Th2 cells, and A549 respiratory epithelial cells. In IL-4-stimulated M2 macrophages, CCL22 mRNA and protein were induced by IL-4 and significantly repressed with the addition of 4-OI (Figure 3 A, B). As shown previously, 4-OI also repressed Fizz1, a common M2 marker downstream of the JAK-STAT signaling pathway (Figure 3 C) ⁵⁹. Next, we investigated the effect of 4-OI on IL-4-activated Th2 cells. 4-OI was able to reduce the expression of both GM-CSF (Figure 3 D) and IL-5 (Figure 3 E), two principal cytokines involved in eosinophilic asthma. Additionally, we also found a transcriptional reduction of CCL22 by 4-OI in IL-4-activated Th2 cells (Figure 3 F). A549 cells have previously been shown to produce CCL5 when stimulated with IL-1β ¹⁶ and this response was again repressed by 4-OI (Figure 3 G and H). In summary, we showed here the inhibitory effect of 4-OI on eosinophil-recruiting chemokines extends beyond LPS-stimulated macrophages; it encompasses a diverse range of cell types, including IL-4-stimulated BMDMs and Th2 cells, and respiratory epithelial A549 cells.

Figure 3. 4-OI inhibits the production of eosinophil recruiting chemokines in IL-4-stimulated BMDMs, Th2 cells, and A549 cells.

BMDMs were pre-treated with DMSO or 4-OI (250 μM) or DMSO control for 1 h before stimulation with IL-4 (20 ng/mL) for 24 h. mRNA (A) and protein secretion (B) levels of CCL22 were measured with qPCR and ELISA, respectively (n = 3). mRNA level of FIZZ1 was measured by qPCR (C) (n = 3). Resting CD4+ T cells were treated with DMSO or 4-OI (250 μM) or DMSO control for 30 min and activated and polarised to Th2 with IL-4 (40 ng/mL) for 48 h. The mRNA levels of chemokines GM-CSF (D), IL-5 (E), and CCL22 (F) were measured with qPCR (n = 3). A549 cells were pre-treated with DMSO or 4-OI (250 μM) or DMSO control for 3 h before stimulation with IL-1β (10 ng/mL) for 16 h. mRNA expression (G) and protein level in the supernatant (H) of CCL5 were quantified by qPCR and ELISA, respectively (n = 3). Results are obtained from 3 independent experiments. Data are mean ± SEM. P values were calculated using one-way analyses of variance (ANOVA).

4-OI directly inhibits eosinophil differentiation

We subsequently examined the impact of 4-OI on eosinophil differentiation through the implementation of an in vitro stimulation protocol using bone marrow cells ²³. To determine the specific effect of 4-OI on eosinophil development, we administered 4-OI (125 μM or 250 μM) to the cells on day 0 or day 4 and the samples were analyzed on day 4 or day 12. We first examined the effect of 4-OI directly on IL-5-induced eosinophil differentiation, in which 4-OI was added on day 4 with IL-5 and the cells were taken down on day 12 (Figure 4 A). 4-OI at a higher dose (250 μM) caused a decrease in cell viability (Figure 4 B), nonetheless, 4-OI markedly reduced the proportion of Siglec F⁺ cells within live cells (Figure 4 C), as well as Epx (Figure 4 D), a prominent eosinophil marker. To confirm the effect of 4-OI happens downstream of IL-5 signaling but not at the progenitor level, we also added 4-OI to the bone marrow progenitor cells before the administration of IL-5. The addition of 4-OI was initiated on day 0, followed by analyses on both day 4 (Figure 4 E) and day 12 (Figure 4 I). At a higher dose, 4-OI-treated samples exhibited cytotoxicity on day 4 (Figure 4 F), however, the cellular viability was recovered by day 12 (Figure 4 J). An initial decline in the proportion of Siglec F⁺ cells was observed on day 4 (Figure 4 G), yet the inhibitory effect of 4-OI was notably attenuated by day 12 (Figure 4 K). Furthermore, it is noteworthy that there was no significant reduction in Epx expression, as indicated in Figures 4 H and L. These findings collectively suggest that the influence of 4-OI on bone marrow progenitor cells is minimal, exerting its effect mainly downstream of IL-5 signaling.

Figure 4. 4-OI reduces eosinophil differentiation from murine bone marrow.

Bone marrow progenitors were isolated on day 0 and stimulated with SCF (100 ng/mL) and Flt-3L (100 ng/mL) for proliferation. On day 4, fresh media was replaced and supplemented with IL-5 (10 ng/mL) to stimulate eosinophil differentiation. The media was fully replaced on day 8, and half replaced on day 10 with IL-5 (10 ng/mL). On day 4, samples were treated with 4-OI (125 or 250 μM) or DMSO control, the cells were taken down on day 12 (n=3) (A). Flow cytometry analyses of the percentage of live cells (B) and Siglec F⁺ cells (C) in the culture. qPCR analysis of an eosinophil marker Epx (D). On day 0, bone marrow progenitors were treated with 4-OI (125 or 250 μM) or DMSO control, the cells were taken down on day 4 (n=3) (E). Flow cytometry analyses of the percentage of live cells (F) and Siglec F⁺ cells (G) in the culture. qPCR analysis of an eosinophil marker Epx (H). Following treatment with 4-OI (125 or 250 μM) or DMSO control on day 0, the cells were subjected to IL-5 administration and media change on day 4, day 8, and day 10, and taken down on day 12 (n=3) (I). Flow cytometry analyses of the percentage of live cells (J) and Siglec F⁺ cells (K) in the culture. qPCR analysis of an eosinophil marker Epx (L). Results are obtained from 3 independent experiments. Data are mean ± SEM. P values were calculated using one-way analyses of variance (ANOVA) for multiple comparisons.

Effect of 4-OI on OVA-induced eosinophilic airway inflammation and hyperresponsiveness in mice

We performed an OVA model of eosinophil-high lung inflammation using the protocol illustrated in Figure 5 A. As expected, mice sensitized and challenged with OVA developed enhanced airway resistance (Rn) compared with sham (Sal) control mice (Figure 5 B, C). Treatment with 4-OI improved Rn (48% and 39% at 5 mg/kg and 10 mg/kg, respectively, compared to OVA-treated [100%]) to a similar extent as DEX (38% compared to OVA-treated) (Figure 5 B, C). 4-OI treatment at a higher dose also significantly reduced OVA- specific IgE as illustrated in Figure 5 D. Both DEX and 4-OI caused marked reductions in leukocyte accumulation in the lungs (Figure 5 D), with significant reductions in proportions of macrophages (Figure 5 F) and eosinophils (Figure 5 G). Interestingly, treatment with 4-OI also alleviated neutrophil accumulation, although not to statistical significance (Figure 5 H).

Figure 5. Effects of 4-OI on the disease phenotype of OVA-induced eosinophilic airway inflammation.

BLAB/c mice were i.p. injected with OVA and alum or sham control (Sal) 7 days before the trial, and i.n. challenged with OVA from day 12 to day 15. 4-OI, DEX, or vehicle were i.n. given to the mice on days 13, 14, and 15. The endpoint of the trial was day 16 (A). Airway resistance (Rn) in response to increasing doses of methacholine (Mch) (B) and Rn in response to 10 mg/mL Mch (C). OVA-specific IgE count in plasma (D). Total number of leukocytes (E), macrophages (F), eosinophils (G), and neutrophils (H) in lung bronchoalveolar lavage fluid (BALF). Results are obtained from 6-8 animals per group of treatment in one in vivo study. Data are mean ± SEM. P values were calculated using one-way or two-way analyses of variance (ANOVA) for multiple comparisons. Differences were considered statistically significant at *p < 0.05. **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Since this model drives predominantly type 2 inflammation, we also examined the gene expression of key Th2 cytokines Il5 (Figure 6 A) and Il13 (Figure 6 B) as well as M2 macrophage markers including Fizz1 (Figure 6 C), and Pparg (Figure 6 D) in the lung tissues of OVA-sensitized mice. Mice treated with 4-OI had marked reduction in the expression of the panel of Th2 genes in the lungs compared to control animals. Despite a Th2 dominant inflammatory response, a concurrent induction of type 1 phenotype was discerned. Evaluation of Il1b (Figure 6 E) and Il17 (Figure 6 F) revealed a modest upregulation following OVA exposure and notably, the administration of 4-OI diminished both responses. Lastly, we examined the chemokines responsible for eosinophil recruitment to the lungs, and indeed, we observed strong reductions of CCL11 (Figure 6 G), CCL24 (Figure 6 H), CCL3 (Figure 6 I), CCL5 (Figure 6 J), and CCL22 (Figure 6 K) by 4-OI. Overall, 4-OI exhibited promising anti-inflammatory properties in this in vivo eosinophilic model of airway inflammation by reducing airway hyperresponsiveness, leukocyte accumulation, expression of inflammatory gene markers, and eosinophil-targeted chemokines (Figure 7).

Figure 6. Effects of 4-OI on the immunophenotyping of OVA-induced eosinophilic airway inflammation.

mRNA expression in lung tissue of OVA-sensitized mice of Il5 (A), Il13 (B), Fizz1 (C), Pparg (D), Il1b (E), and Il17a (F) analyzed by qPCR. Protein (pg per μg of lung protein) expression of CCL11 (G), CCL24 (H), CCL3 (I), CCL5 (J), and CCL22 (K) in lung tissue of OVA-sensitized mice analyzed by ELISA. Results are obtained from 6-8 animals per group of treatment in one in vivo study. Data are mean ± SEM. P values were calculated using one-way or two-way analyses of variance (ANOVA) for multiple comparisons. Differences were considered statistically significant at *p < 0.05. **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 7. 4-OI modulates eosinophil-targeted chemokines, and eosinophil differentiation, and alleviates eosinophilic lung inflammation.

4-OI demonstrates significant attenuation in the production of eosinophil-targeted chemokines, including CCL3, CCL5, and CCL22 by M1 macrophages, CCL22 by M2 macrophages and Th2 cells, and CCL5 by A549 cells. Additionally, it suppresses the expression of IL-5 and GM-CSF in Th2 cells. Furthermore, 4-OI disrupts IL-5 signaling and directly impacts eosinophil differentiation. In a BALB/c murine model of eosinophilic airway inflammation, 4-OI effectively reduces airway resistance, and allergen-specific IgE, and diminishes eosinophil recruitment to the lungs.

Discussion

Itaconate and its derivatives have been of significant interest to immunologists due to their broad immunomodulatory effects in various disease models ^{6,21,36,45,48,50,62,63,70,73}. 4-OI was previously shown to have therapeutic effects in a neutrophilic model of allergic lung inflammation by blocking JAK1 signaling in Th2 cells and M2 macrophages ⁵⁹. In HDM-induced lung inflammation, 4-OI also reduced pathology, likely through suppressing antigen presentation by dendritic cells ³⁰. In this study, we built on previous findings and specifically explored the role of 4-OI in chemokines and eosinophils in the context of allergic lung inflammation. We showed that 4-OI suppressed the chemokines CCL3, CCL5, and CCL22, which are known to recruit eosinophils. The inhibitory effect of 4-OI on these chemokines was observed in various cell types, including M1 and M2 macrophages, Th2 cells, and A549 respiratory epithelial cells. In LPS-stimulated M1 macrophages, we also demonstrated that the effect by which 4-OI reduced these chemokines was NRF2-dependent, and that endogenous itaconate did not affect the expression of these chemokines. We also showed that 4-OI directly suppressed IL-5 signaling of the differentiation process of eosinophils. In an OVA-induced eosinophilic lung inflammation model, 4-OI significantly ameliorated airway resistance, paralleling the effects of DEX. Moreover, 4-OI treatment of mice with allergic lung inflammation significantly reduced key inflammatory markers, and eosinophil-recruiting chemokines, underscoring its potential as an anti-inflammatory agent in asthma.

The α, β-unsaturated carbonyl group of itaconate is electrophilic and can modify cysteine residues in protein targets by Michael reactions to exert versatile biological effects ⁵⁵. 4-OI is an esterified form of itaconate and is widely used as a cell-permeable itaconate surrogate. 4-OI can be hydrolyzed to itaconate by esterases once entering cell ⁴⁴. Although both itaconate and 4-OI modify proteins in the same manner, not all the protein targets are identical, since 4-OI contains increased electrophilicity, which likely explains the difference observed in our study between endogenous itaconate synthesized by ACOD1 or added exogenously and 4-OI ⁴⁴. In the case of 4-OI, it replicates the biological effects of itaconate in several cases, for example, the inhibition of succinate dehydrogenase ^18,34, inactivation of the NLRP3 inflammasome ^7,28,34, and impairment of glycolysis ^37,44,55. However, itaconate and esterified derivatives differ in their regulation of the type I IFN response. Irg1-/- BMDMs were found to have an attenuated type I IFN response to LPS ⁶⁶, but studies using 4-OI found a strong decrease in type I IFN-regulated genes ^6,44.

The identification of a role for NRF2 in the effect of 4-OI is interesting since NRF2 not only activates an antioxidant response but also actively represses proinflammatory gene transcription. Upon oxidative or electrophilic stress, KEAP1 dissociates from NRF2, and NRF2 then translocates to the nucleus to activate the expression of antioxidant genes ^13,20,29,33. NRF2 can also bind to the promoters of proinflammatory genes, such as Il6, Il1b, and Il1a, directly suppressing their expression ³². As such, 4-OI-mediated inhibition of IL-1β required both the critical thiol-reactive KEAP1 cysteines (as being modified by 4-OI) and the presence of NRF2 ⁴⁴. Our data reveal an interesting downstream effect regulated by NRF2 in the context of asthma, whether NRF2 suppressed chemokine expression by activating an antioxidant response, or by actively binding to chemokine genes, the exact mechanism here remains to be further elucidated. The downregulation of Il1b by 4-OI elucidates our findings, wherein it attenuates the expression of Il17a, potentially contributing to the diminished in vivo neutrophil accumulation. In addition, 4-OI-activated NRF2 was previously shown by Olagnier et al to negatively regulate STING and type I IFN production ⁴⁹, a response linked extensively to severe asthma phenotypes ^27,46,58. The overall alleviation of lung inflammation observed in our study is likely a joint effect from 4-OI targeting multiple pathways which might partly involve the type I IFN response.

Our study represents to our knowledge the first report of the inhibition of eosinophil differentiation in response to IL-5 signaling, indicating another mechanism for 4-OI in regulating asthma. The binding of IL-5 to the IL-5 receptor activates JAK1 and JAK2, and JAK2 is believed to phosphorylate STAT1/5 in eosinophils ^47,52,68. JAK2 engages with Lyn and Raf-1 kinases ⁵¹ and JAK1 induces the antiapoptotic protein Bcl-xL via NF-κB ⁶¹, thereby inhibiting eosinophil apoptosis. Directly downstream of JAK1/2, Ras-Raf-1-mediated activation of the ERK family of MAPK drives c-fos gene transcription, which is involved in eosinophil maturation, survival, and proliferation ^2,9,54,67. The finding that 4-OI acts as a JAK inhibitor sheds light on how it can suppress eosinophil differentiation ⁵⁹, ultimately leading to a decrease in airway hypersensitivity in mice.

The heterogeneity of asthma means that there is no one-size-fits-all approach to treatment, and the variability in symptoms makes it challenging to determine the best course of treatment. The traditional approach to treatment often involves the use of corticosteroids, which while effective, can also have significant side effects. It is also estimated that 5-10% of individuals with severe, persistent asthma may not respond adequately to steroid treatment ⁸. Current asthma therapies that involve direct biologically targeted therapy aim to target specific pathways in the development and progression of asthma, for example, the IL-5 pathway in the recruitment and activation of eosinophils ²⁵ and the IL-4 and IL-13 pathways in the activation of Th2 cells ^5,12. By targeting specific proteins, these biologics can more effectively reduce airway inflammation with fewer side effects than steroids and provide additional treatment options for those non-responsive to steroids ¹². While current biologics have proven to be effective in reducing airway inflammation and improving symptoms for some individuals with severe and persistent asthma, not all patients respond well to these treatments. The heterogeneity of asthma and the limitations of treatments highlight the importance of developing more approaches for asthma to provide targeted and effective therapies for patients who do not respond well to current options. Our study, and the inhibitory effect of 4-OI on both Th17-driven ⁴ and Th2-driven responses ⁵⁹, both of which are recognized as pivotal subtypes in the pathogenesis of asthma indicate that derivatives of itaconate might hold promise for the treatment of asthma ^17,19. Additional studies are therefore warranted to comprehensively investigate the therapeutic potential of 4-OI and related derivatives in treating asthma.

Key points.

• 4-OI inhibits eosinophil-recruiting chemokines and eosinophil differentiation.

• 4-OI reduces airway hyperresponsiveness and airway inflammation in mice.

List of abbreviations

4-OI

4-octyl itaconate

DEM

dimethyl maleate

DEX

dexamethasone

HDM

house dust mite

KEAP1

Kelch-like ECH-associated protein 1

MDC

macrophage-derived chemokine

SCF

stem cell factor

SF/PSF

serum-free and penicillin-streptomycin-free