Research ArticleMetabolism Open Access | 10.1172/JCI160463

Yi-Rong Zeng,1 Jun-Bin Song,1 Dezheng Wang,1 Zi-Xuan Huang,1 Cheng Zhang,1 Yi-Ping Sun,1 Gang Shu,2 Yue Xiong,3 Kun-Liang Guan,4 Dan Ye,1 and Pu Wang1

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Zeng, Y. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Song, J. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Wang, D. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Huang, Z. in: JCI | PubMed | Google Scholar |

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Zhang, C. in: JCI | PubMed | Google Scholar |

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Sun, Y. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Shu, G. in: JCI | PubMed | Google Scholar |

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Xiong, Y. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Guan, K. in: JCI | PubMed | Google Scholar |

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Ye, D. in: JCI | PubMed | Google Scholar

1Shanghai Key Laboratory of Clinical Geriatric Medicine, Huadong Hospital, Fudan University, and Key Laboratory of Metabolism and Molecular Medicine (Ministry of Education), and Shanghai Key Laboratory of Medical Epigenetics, International Co-laboratory of Medical Epigenetics and Metabolism (Ministry of Science and Technology), and Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Shanghai Medical College of Fudan University, Shanghai, China.

2College of Animal Science, South China Agricultural University, Guangzhou, China.

3Cullgen Inc., San Diego, California, USA.

4Department of Pharmacology and Moores Cancer Center, UCSD, La Jolla, California, USA.

Address correspondence to: Dan Ye or Pu Wang, 131 Dong An Road, Molecular and Cell Biology Lab, Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, China. Phone: 86.21.5423.7450; Email: yedan@fudan.edu.cn (DY); Email: wangpu@fudan.edu.cn (PW).

Authorship note: YRZ and JBS contributed equally to this work.

Find articles by Wang, P. in: JCI | PubMed | Google Scholar |

Authorship note: YRZ and JBS contributed equally to this work.

Published March 15, 2023 - More info

Published in Volume 133, Issue 6 on March 15, 2023
J Clin Invest. 2023;133(6):e160463. https://doi.org/10.1172/JCI160463.
© 2023 Zeng et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Published March 15, 2023 - Version history
Received: March 31, 2022; Accepted: January 25, 2023 View PDF Abstract

Pathogens and inflammatory conditions rapidly induce the expression of immune-responsive gene 1 (IRG1) in cells of myeloid lineage. IRG1 encodes an aconitate decarboxylase (ACOD1) that produces the immunomodulatory metabolite itaconate (ITA). In addition to rapid intracellular accumulation, ITA is also secreted from the cell, but whether secreted ITA functions as a signaling molecule is unclear. Here, we identified ITA as an orthosteric agonist of the GPCR OXGR1, with an EC50 of approximately 0.3 mM, which was in the same range as the physiological concentration of extracellular ITA upon macrophage activation. ITA activated OXGR1 to induce Ca2+ mobilization, ERK phosphorylation, and endocytosis of the receptor. In a mouse model of pulmonary infection with bacterial Pseudomonas aeruginosa, ITA stimulated Oxgr1-dependent mucus secretion and transport in respiratory epithelium, the primary innate defense mechanism of the airway. Our study thus identifies ITA as a bona fide ligand for OXGR1 and the ITA/OXGR1 paracrine signaling pathway during the pulmonary innate immune response.

Graphical Abstract Introduction

Metabolic reprogramming is a fundamental mechanism for regulating the immune response in both innate and adaptive immune systems (1). Itaconate (ITA), a dicarboxylic acid, is one of the best examples of metabolic reprogramming during the immune response (2). ITA has long been known as one of the major building-block chemicals that can be produced from sugars via biological conversions in fungi Aspergillus for the industrial-scale synthesis of a number of high-value, bio-based chemicals or materials (3). ITA was also discovered in activated macrophages of high eukaryotes (4). The enzyme that produces ITA has been identified as the mitochondrially localized aconitate decarboxylase (ACOD1) (5), encoded by immune-responsive gene 1 (IRG1, NCBI Gene ID: 730249), which was first identified by its rapid transcriptional induction in macrophages challenged by bacterial LPS (6). The expression of IRG1/ACOD1 is restricted to cells of myeloid lineages, in particular in activated macrophages that play major roles in host defense against various microbial pathogens such as bacteria and fungi, supporting the function of ITA in the innate immune defense.

Several mechanisms have been proposed for the immunomodulatory function of ITA. These include the inhibition of succinate dehydrogenase (SDH), which results in succinate accumulation and metabolic reprogramming (7, 8), and alkylation of protein cysteine residues, which induces the electrophilic stress response mediated by NRF2 and IκBζ (9, 10) and impairs aerobic glycolysis (11). We recently discovered that ITA, which is structurally similar to the Krebs cycle intermediate α-ketoglutaric acid (α-KG), binds to and inhibits α-KG–dependent TET DNA demethylases and thereby downregulates NF-κB and STAT target genes to dampen the inflammatory response (12). Intriguingly, ITA was also noted to secrete from activated macrophages into the inflamed microenvironment (4, 8). Whether secreted ITA plays a role in mediating the function of IRG1 in modulating the immune response remains largely unknown.

Macrophages in the lungs are the first responders to pulmonary bacterial and fungal pathogens, including bacterial Pseudomonas aeruginosa (P. aeruginosa), an opportunistic pathogen known to infect patients during existing diseases or conditions such as cystic fibrosis (CF) and also known for its multiple drug resistance (13). The pulmonary respiratory epithelium acts as a physical barrier to airborne pathogens through a self-cleaning mechanism called mucociliary clearance (MCC) (14). Goblet cells and other cells secret a mucus layer that covers the respiratory epithelium to trap inhaled particles, and the cilia beating of ciliated cells continuously transports the mucus into the pharynx. Dysfunction of MCC is associated with multiple diseases such as CF, asthma, and chronic obstructive pulmonary disease (COPD) (14). Nucleosides sensed by purinergic receptors are major regulators of cilia beating and mucin secretion (15), although many other signals are also important, such as paracrine signaling from innate immune cells during infection (16). Of note, ITA was detected in mouse airways after pathogen infection (17, 18), and its concentration was increased in sputum samples from patients with CF (19). Irg1-KO mice were more vulnerable to pulmonary infection with diverse pathogens, such as Mycobacterium tuberculosis, Brucella, Legionella, influenza A virus, respiratory syncytial virus, and so on (20–24). P. aeruginosa and Staphylococcus aureus can exploit host-derived ITA to fuel biofilm formation and thus protect them from clearance by immune defense systems (18, 19). Together, these findings suggest that airway-secreted ITA is a conserved feature that influences the evolution of many airborne pathogens.

Multiple GPCRs play an important role in sensing metabolites to regulate immune function (25, 26). For instance, OXGR1 (also called GPR99, NCBI Gene ID: 27199) encodes a rhodopsin family GPCR and was reported to be stimulated by α-KG (27). OXGR1 is expressed in the respiratory epithelium and is indispensable for MCC (28). In the studies of respiratory tract infection by Alternaria, deletions of Oxgr1 in mice were reported to cause decreased mucin secretion and brush cell expansion (28, 29). Oxgr1-null mice also spontaneously developed otitis media that was probably caused by impaired airway epithelium function in the eustachian tube (30). Renal α-KG/OXGR1 signaling was found to promote bicarbonate secretion and maintain the acid-base balance of body fluids (31). Despite being highly concentrated in urine (32), α-KG levels in serum and other body fluids are far lower than its EC50 for OXGR1 (33), suggesting that α-KG is not likely to be the physiological ligand for OXGR1 in the respiratory epithelium or most other tissues. In this study, we identified ITA as an orthosteric agonist of OXGR1 and determined that the ITA/OXGR1 signaling pathway plays an important role in the self-cleaning function of the airway during the pulmonary innate immune response.

Results

A screen of GPCR receptors discovers OXGR1 as an ITA receptor. To gain insight into the function of secreted ITA, we challenged mouse bone marrow–derived macrophages (BMDMs) with LPS and found, as expected, that the induction of Irg1 protein led to increased levels of both intracellular and extracellular ITA (Figure 1, A–C). Notably, a substantial amount of ITA was secreted into the culture medium starting at 4 hours after LPS treatment, and the extracellular ITA continuously accumulated and eventually comprised more than 90% of the newly synthesized ITA (Figure 1D). The absolute concentration of extracellular ITA was 48.9 ± 0.6 μM twenty hours after LPS treatment at a routine plating density (8 × 105 BMDMs in 1 mL medium per well) (Figure 1C).

Figure 1

A screen of GPCR receptors discovers OXGR1 as a potential ITA receptor. (A) In mouse BMDMs challenged with LPS, intracellular and extracellular levels of ITA were collected and quantified. (B) Immunoblot of Irg1 in LPS-challenged mouse BMDMs. (C) Intracellular and extracellular levels of ITA were measured by LC-MS in LPS-challenged BMDMs at the indicated time points. Error bar, SEM. (D) Comparison of the percentage of newly synthesized intracellular and extracellular ITA during LPS stimulation. (E) Schematic illustration of the GPCR screen. A total of 301 plasmids encoding for GPCRs were transfected along with Gα16 and aequorin plasmids. ITA-induced aequorin luminescence was determined. (F) The ITA dose response was determined by the aequorin assay in HEK293 cells overexpressing human OXGR1 (hOXGR1) or mouse OXGR1 (mOXGR1). Vec, empty vector.

In our search for the function and mechanism of secreted ITA, we noted the signaling of the GPCR proteins SUCNR1/GPR91 and OXGR1/GPR99, which sense the TCA metabolites succinate and α-KG, which are structurally similar to ITA (27). To test the hypothesis that ITA may similarly signal through a GPCR, we performed a screening of human GPCRs in transfected HEK293 cells and measured GPCR-dependent calcium mobilization (Figure 1E and Supplemental Figure 1; supplemental material available online with this article; https://doi.org/10.1172/JCI160463DS1). In order to allow more GPCRs to couple with the calcium pathway, we optimized the screening system by inducing coexpression of the promiscuous Gq family α subunit Gα16 (34). We examined 301 GPCRs that cover more than 70% of the nonolfactory GPCRs encoded by the human genome (35). Strikingly, we found that OXGR1 was the only receptor that showed robust and reproducible calcium mobilization in response to ITA treatment (Supplemental Figure 1).

The EC50 of ITA for human OXGR1 was 200–300 μM, which was comparable to the EC50 of α-KG (Figure 1F and Supplemental Figure 2, A–C). The EC50 of ITA for mouse OXGR1 was approximately 400 μM, and the EC50 of α-KG was in the same range (Figure 1F and Supplemental Figure 2D). Considering that secreted ITA would be greatly diluted in the large volume of culture medium, the concentration of extracellular ITA around activated macrophages in vivo could be higher than the EC50 of ITA for OXGR1 and could thus activate this potential receptor.

ITA acts as an agonist of OXGR1 to stimulate cytosolic calcium, ERK activation, and receptor endocytosis. Heterotrimeric G proteins mainly relay the signals from GPCRs to intracellular signal transduction (36). To investigate the intracellular signaling pathways initiated by ITA/OXGR1, we measured and found that ITA stimulated the calcium response in a dose-dependent manner in HEK293 cells overexpressing human OXGR1 and that depletion of Gq/11 by siRNAs greatly attenuated the cellular calcium response initiated by ITA/OXGR1 (Figure 2, A and B). In contrast, ITA/OXGR1 did not affect intracellular cAMP, which is known to be influenced by Gs or Gi (Supplemental Figure 3, A and B). ITA/OXGR1 also failed to affect stress fiber formation, which is a typical response to G12/13 activation (Supplemental Figure 3C). These findings thus suggest that ITA/OXGR1 couples to Gq/11, but not Gs, Gi, or G12/13 family G proteins.

Figure 2

ITA acts as an agonist of OXGR1 to stimulate cytosolic calcium, ERK activation, and receptor endocytosis. (A) Verification of Gq/11 knockdown in HEK293 cells by Western blotting. (B) Aequorin assay showing calcium mobilization by different concentrations of ITA in cells expressing human OXGR1 with or without siRNAs targeting Gq/11. (C) ITA and α-KG induced endocytosis of FLAG-tagged OXGR1, which was detected using anti-FLAG antibody. Nuclei were stained with DAPI. Scale bar: 10 μm. (D) Endocytosis of FLAG-tagged OXGR1 induced by ITA with or without siRNAs targeting β-arrestin 1/2 was detected by anti-FLAG antibody. Nuclei were stained with DAPI. Scale bar: 10 μm. (E) Serum-starved cells were stimulated with different concentrations of ITA (from 5 μM to 1 mM) for 15 minutes (left), or 500 μM ITA for different durations (from 5 to 30 minutes) (right). Cells were harvested, and p-ERK was determined by Western blotting. (F) Serum-starved cells expressing human OXGR1 transfected with siRNAs targeting Gq/11 and β-arrestin 1/2 were stimulated with 500 μM ITA for the indicated durations. Cells were harvested, and p-ERK levels were determined by Western blotting.

Initiation of GPCR signaling by ligand binding is followed by β-arrestin–induced receptor endocytosis, which is a major mechanism for GPCR signaling termination (37). We found that ectopically expressed OXGR1 was localized on the plasma membrane and that rapid endocytosis of OXGR1 was triggered by ITA as well as α-KG within 10 minutes of ligand treatment (Figure 2C). Ablation of β-arrestin 1/2 by siRNAs impaired ITA-induced endocytosis of OXGR1 (Figure 2D), confirming that the internalization of OXGR1 was dependent on β-arrestin. Phosphorylation of ERK (p-ERK) commonly occurs in cells with activated calcium and/or β-arrestin pathways (36, 37). As expected, the phosphorylation of ERK was dose-dependently induced by ITA in OXGR1-expressing cells, and such induction was observed as quickly as 5 minutes after ITA treatment (Figure 2E). Codepletion of β-arrestin and Gq/11, but not either single depletion alone, abolished ITA-induced ERK phosphorylation in OXGR1-expressing HEK293 cells (Figure 2F), indicating that extracellular ITA stimulates OXGR1 to initiate intracellular signaling through Gq and β-arrestin.

ITA activates OXGR1 by interacting with its orthosteric site. Most natural GPCR agonists interact with the orthosteric site of the rhodopsin family GPCRs to induce a conformational change and the guanine nucleotide exchange factor activity of GPCR. To gain additional evidence supporting ITA/OXGR1 signaling and mechanistic insights into ITA-induced OXGR1 activation, we mutated several amino acid residues localized around the orthosteric site of human OXGR1 that are predicted by the AlphaFold structure (38). These mutants are predicted to localize on the transmembrane domains TM3 (C106A, K107A, R110A, H114A, Y118F, I121A), TM6 (H258A, R261A, R264A, I265A), and TM7 (I285A and R288A) (Figure 3A). Among the mutants examined, the H114A, Y118F, I121A, I285A, or R288A mutant of OXGR1 dramatically impaired the effect of ITA or α-KG on stimulation of the calcium response, and the C106A, R110A, H258A, R261A, or I265A mutant completely abolished this response (Figure 3, B and C). Furthermore, the R110A or R261A mutant of OXGR1 did not undergo endocytosis after ITA treatment (Figure 3D). We then conducted a cellular thermal shift assay and found that ITA increased the thermal stability of WT OXGR1, but not of the R110A mutant, which was unresponsive to ITA in a calcium assay. As a negative control, the thermal stability of the cotransfected BirA protein remained unchanged (Supplemental Figure 4). Our data thus support the notion that ITA activates OXGR1 by interacting with its orthosteric site.

Figure 3

ITA activates OXGR1 by interacting with its orthosteric site. (A) Virtual structure of human OXGR1 predicted by AlphaFold. Amino acid residues C106, R110, H114, Y118, H258, R261, I265, and R288 are labeled in red. (B) Verification of ectopically expressed OXGR1 in HEK293 cells as determined by Western blotting. (C) Aequorin assay showing calcium mobilization in cells expressing WT or mutant OXGR1 in response to the indicated ligands. (D) Endocytosis of FLAG-tagged WT or mutant OXGR1 was detected using anti-FLAG antibody. Nuclei were stained with DAPI. Scale bar: 10 μm.

ITA accumulates in the respiratory epithelium during bacterial infection. A previous study reported that Oxgr1 was expressed in the respiratory epithelium and that Oxgr1–/– mice had impaired respiratory function (28), but the natural ligand of Oxgr1 in the respiratory epithelium has not been defined. Leukotriene E4 (LTE4) is an immunometabolite produced by mast cells and was previously reported to act as a ligand of Oxgr1 (28, 29, 39). As expected, an aequorin assay revealed that LTE4 could activate its well-known receptor CysLTR1 in cells (Supplemental Figure 5A). In contrast, LTE4 failed to show any agonizing effect on OXGR1 in both the aequorin and endocytosis assays (Supplemental Figure 5, A and C). Moreover, addition of LTE4 did not prevent, and even moderately enhanced, α-KG– or ITA-triggered intracellular Ca2+ release in OXGR1-expressing HEK293 cells (Supplemental Figure 5B). These results thus indicate that LTE4 was not a direct agonist of OXGR1 in our experimental system.

We next determined the in vivo concentrations of ITA and α-KG in bronchoalveolar lavage (BAL) in the mouse model of pulmonary infection with the Pseudomonas aeruginosa strain PAO1. Our liquid chromatography–MS (LC-MS) results determined the absolute concentration of ITA to be 0–0.067 nM and 122–689 nM in BAL isolated from uninfected and PAO1-infected mice, respectively (Supplemental Figure 6, A and B). We also estimated the ITA concentration in undiluted epithelial lining fluid (ELF) using the urea concentration in BAL and plasma as a marker of dilution (ITA[ELF] = ITA[BAL] × Urea [plasma] / Urea [BAL]) (40). The ITA concentration in undiluted ELF from PAO1-infected mice was 109.5 ± 69.2 μM (ranging from 18–215 μM), indicating that endogenous ITA could accumulate to a concentration that could robustly activate OXGR1 (Supplemental Figure 6C). KO of Irg1, but not Oxgr1, abolished ITA secretion into the BAL (Supplemental Figure 6D). In contrast, gas chromatography–MS (GC-MS) analysis revealed that the detection threshold for α-KG was approximately 125 nM, but the concentration of α-KG in all BAL samples examined was under this threshold, even after the BAL samples from 4 mice were pooled and concentrated into 20 μL (Supplemental Figure 6, E–G). We thus estimated the α-KG concentration in undiluted epithelial lining fluid to be below 40.7 nM using an average dilution factor determined by the urea concentration, which was too low to activate Oxgr1 (EC50 of α-KG is 241.1 ± 61.1 μM for mouse Oxgr1).

Taken together, these results suggest that ITA, but not LTE4 or α-KG, is the natural ligand of Oxgr1 in respiratory epithelium during bacterial infection.

ITA signals through OXGR1 to stimulate MCC in respiratory epithelium. Accumulation of ITA in the respiratory epithelium prompted us to examine the potential role of ITA/OXGR1 signaling in respiratory mucosal immunity. To this end, we isolated primary respiratory epithelial cells from Oxgr1+/+ and Oxgr1–/– mice, followed by ITA treatment and detection of the cytoplasmic calcium concentration with the calcium dye Fluo-4AM. We observed that both Oxgr1+/+ and Oxgr1–/– cells exhibited normal calcium release in response to ATP, which stimulates the purinergic receptors P2X and P2Y (15), suggesting that loss of Oxgr1 did not have a broadly deleterious effect on the general function of respiratory epithelial cells (Figure 4A). In line with our earlier observation made in transfected HEK293 cells (Supplemental Figure 2A), treatment with ITA led to rapid calcium mobilization in Oxgr1+/+, but not Oxgr1–/– respiratory epithelial cells (Figure 4A). Consistently, treatment of primary human bronchial epithelial cells with 0.5 mM ITA also led to robust calcium mobilization, and such ITA-stimulated calcium mobilization was abolished by depletion of OXGR1 (Figure 4D), providing evidence of the physiological relevance of ITA/OXGR1 signaling in human cells.

Figure 4

ITA signals through OXGR1 to stimulate MCC in respiratory epithelium. (A) Calcium mobilization of primary trachea epithelial cells isolated from mice in response to 670 μM ITA or 100 μM ATP, as determined by Fluo-4AM assay. (B) Tracheas were cultured ex vivo, and cilia beating–derived fluid flow was monitored by tracing fluorescent bead movement under the microscope. The trajectory of 20 individual beads in 1 minute with or without 500 μM ITA is shown. Scale bar: 150 μm. (C) Quantification of bead movement speed in trachea culture from individual Oxgr1+/+ (n = 3) and Oxgr1–/– (n = 3) mice in the presence of 500 μM ITA or 100 μM ATP. Each dot represents an independent biological replicate. *P < 0.05 and **P < 0.01, by paired, 2-tailed Student’s t test. (D) Primary human bronchial epithelium cells were infected by a control lentivirus (Vec) or viruses expressing sgRNAs targeting OXGR1 (sgOXGR1). Calcium mobilization in response to 500 μM ITA was determined by Fluo-4AM assay. (E and F) Comparison of bead movement speed between ALI culture of human bronchial epithelium cells in the presence of 500 μM ITA or 100 μM ATP. **P < 0.01 and ****P < 0.0001, by 1-way ANOVA (E) or unpaired, 2-tailed Student’s t test (F). Data indicate the mean ± SEM. (G–I) Anesthetized mice of the indicated genotype (n = 4–6 per group) were intranasally administered PBS or 500 μM ITA, and their nasal cavities were harvested 1 hour after ITA administration. (G) Sections were taken from the most proximal aspect of the nasal cavities. (H) Representative images of Alcian blue/PAS staining of nasal cavities are shown. (I) Quantification of mucin-containing cells in the indicated groups of animals. ****P < 0.0001, by 2-tailed Student’s t test. Data indicate the mean ± SEM.

Calcium signaling promotes the beating of motile cilia and mucociliary transport by respiratory epithelial cells (15). To test whether ITA and OXGR1 regulate mucociliary transport, we isolated the mouse trachea and then cultured it ex vivo in a medium containing 400 nm diameter fluorescent beads to trace culture medium movement (Figure 4B). We found that ITA treatment significantly increased the movement of beads driven by trachea derived from Oxgr1+/+, but not Oxgr1–/–, mice (Figure 4, B and C), supporting the notion that ITA signals through the Oxgr1 receptor to boost mucociliary transport of the liquid. As a negative control, trachea tissues from both Oxgr1+/+ and Oxgr1–/– mice exhibited comparable speed of bead movement in response to ATP (Figure 4C). In addition, we reconstructed ciliated respiratory epithelium by air-liquid interface (ALI) culturing of primary mouse respiratory epithelial cells and observed that ITA could accelerate the movement of beads in an Oxgr1-dependent manner (Supplemental Figure 7). Likewise, ITA treatment also accelerated the movement of beads in primary human respiratory epithelial cells under ALI culture in an OXGR1-dependent manner (Figure 4, E and F), reaffirming that ITA stimulates endogenous OXGR1 in human bronchial epithelial cells.

GPCR-induced calcium signaling promotes rapid mucin secretions from the respiratory epithelium (15). We found that intranasal administration of ITA for 1 hour led to rapid mucin secretion, as observed by a decrease of mucin-containing cells in the respiratory epithelium around the vomeronasal organ (VO) in Oxgr1+/+ mice (Figure 4, G–I). However, we did not observe an effect of ITA on the reduction of mucin-containing cells in the respiratory epithelium of Oxgr1–/– mice, suggesting that ITA promoted Oxgr1-dependent mucin release in a physiologic setting.

Depletion of either Oxgr1 or Irg1 disrupts the clearance of respiratory bacteria. MCC of inhaled pathogens is the primary innate defense mechanism of the airway. A previous study reported that ITA was readily detected in the BAL from mice infected with P. aeruginosa and that the clearance of this Gram-negative bacterium was impaired in Irg1-KO mice (18). To determine whether ITA/Oxgr1 signaling participates in airway pathogen clearance, WT, Oxgr1–/–, and Irg1–/– mice were subjected to intranasal infection with the P. aeruginosa strain PAO1, and BAL and lung tissue were collected 12 hours after infection to count CFU (Figure 5A). In accord with the impaired MCC observed in a previous study (18), Irg1–/– mice exhibited a remarkable increase of bacteria in BAL and lung tissue as compared with WT controls (Figure 5B). Notably, the phenotype of defective airway pathogen clearance was recapitulated in the same infection model in Oxgr1–/– mice (Figure 5B). To provide further direct evidence supporting the role of IRG1/ITA/OXGR1 signaling in the innate bacterial defense, we sought to determine whether ITA supplementation rescues the bacterial overgrowth phenotype. The dose of supplemented ITA was adjusted to reach 200–300 nM in the BAL 0.5 hours after ITA supplementation, which mimics the endogenous ITA concentration in WT mice stimulated by PAO1 infection for 12 hours, although the concentration of exogenous ITA was rapidly decreased because of self-cleaning of the respiratory epithelium (Supplemental Figure 8A). ITA was intranasally supplemented in WT, Oxgr1–/–, and Irg1–/– mice infected with P. aeruginosa. We found that ITA supplementation rescued the bacterial overgrowth phenotype in Irg1–/–, but not Oxgr1–/–, mice (Figure 5, C and D).

Figure 5

Depletion of either Oxgr1 or Irg1 disrupts the clearance of respiratory bacteria. (A and B) Anesthetized mice of the indicated genotype (n = 4–10 per group) were intranasally administrated the P. aeruginosa strain PAO1 (1 × 107 CFU). BAL and lung tissue were collected 12 hours after infection. Representative images of PAO1 colonies on CTAB-selective plates after overnight culturing of BAL and lung from infected mice (dilution: 104-fold) (A) and quantification of CFU (B) are shown. *P < 0.05 and **P < 0.01, by 1-way ANOVA. Data indicate the mean ± SEM. (C and D) Anesthetized mice of the indicated genotype (n = 4–10 per group) were intranasally administrated 20 μL saline containing PAO1 (1 × 107 CFU) with or without 1 mM ITA. The CFU in BAL (C) and lung tissues (D) were quantified. **P < 0.01, ***P < 0.001, and ****P < 0.0001, by 1-way ANOVA. Data indicate the mean ± SEM.

Mucins and antimicrobial polypeptides are essential immune defense components secreted by the respiratory tract (41, 42). To test the hypothesis that ITA/OXGR1 signaling may alter the concentration of mucins and antimicrobial polypeptides, we performed an ELISA to quantify the concentration of mucin 5b and β-defensin 1 in the mouse models of P. aeruginosa infection. Neither ITA intranasal supplementation nor Irg1 or Oxgr1 KO altered the levels of mucin 5b or β-defensin 1 in the BAL (Supplemental Figure 8, B and C). Nevertheless, we found that mucin 5b or β-defensin 1 levels were positively correlated with the ITA concentration in BAL from WT mice after P. aeruginosa infection (Supplemental Figure 8, D and E). It should be noted that Oxgr1 activation by ITA can simultaneously facilitate mucus secretion and mucus transport (Figure 4). These 2 “opposite” effects of ITA/OXGR1 on mucus may explain the observation of a lack of significant change in mucin 5b and β-defensin 1 in the airway’s surface liquid.

Collectively, these results provide in vivo evidence to support the role of Oxgr1 and IRG1/ACOD1/ITA in pathogen clearance in the airway (Supplemental Figure 9).