Evolutionary Adaptations of IRG1 Refines Itaconate Synthesis and Mitigates Innate Immunometabolism Trade-offs

Reviewer #1 (Public review):

Summary:

The taxonomic analysis of IRG1 evolution is compelling and fills an important gap in the literature. However, the experimental evidence for IRG1 localization requires greater detail and confirmation.

Strengths:

The phylogenetic analysis of IRG1 evolution fills an important gap in the literature. The identification of independent acquisition of metazoan and fungal IRG1 from prokaryotic sources is novel, and the observation that human IRG1 lost mitochondrial matrix localization is particularly interesting, with potentially significant implications for the study of itaconate biology.

Weaknesses:

The protease protection assay was conducted with MTS-IRG1 but not with wild-type IRG1, which should also be tested. Moreover, no complementary methods, such as microscopy, were employed to validate localization. Beyond humans, the structure and localization of mouse IRG1, highly relevant given the widespread use of the mouse as a model for IRG1 functional studies, are not addressed. Finally, if itaconate is indeed synthesized outside the mitochondrial matrix to safeguard metabolic activity, it is not discussed how this reconciles with its reported inhibitory effect on SDH.

Reviewer #2 (Public review):

Summary:

The authors are trying to explain how the metabolite itaconate evolved, since although it's involved in host defense, it can also limit mitochondrial function. They are trying to probe the trade-off between these two functions.

Strengths:

The evolutionary aspect is novel; this is the first time to my knowledge that the evolution of IRG1 has been analysed, and there are interesting findings here. The key finding appears to be that subcellular localisation is an important aspect, allowing host defense in some organisms without compromising bioenergetics. This is an interesting finding in the context of immunomebolism, although it needs extra analysis.

Weaknesses:

The work concerning sub-mitochondrial localisation is confusing and needs better analysis.

Reviewer #3 (Public review):

Summary:

IRG1 is highly expressed in activated human and mouse myeloid cells. It encodes the mitochondrial enzyme cis-aconitate decarboxylase 1 (ACOD1) that generates itaconate. Itaconate has anti-microbial activity and acts immunoregulatory by interfering with cellular metabolism, signaling to cytokine production, and multiple other processes.

The authors perform a phylogenetic analysis of IRG1 to obtain insight into the evolution of itaconate biosynthesis. Combining BLAST with human IRG1 and a MmgE/Ptrp domain search, they find CAD in all domains of life, but the presence of IRG1 homologs is patchy in eukaryotes, indicating that itaconate biosynthesis is not essential. The phylogenetic analysis showed a more distant relationship of fungal and metazoan CAD/IRG1 to many prokaryotic sequences, suggesting independent acquisition of these metazoan and fungal CAD genes. In metazoans, three subbranches of paleo-IRG1 (in mollusks/early chordates) and two paralogous vertebrate forms (IRG1 and IRG1-like) were identified, with the latter derived from paleo-IRG1, and by genome duplication. While most jawed vertebrates have both IRG1 and IRG1L, metatherian and eutherian mammals have lost IRG1L and contain only IRG1.

Interestingly, sequence analysis of both paralogues showed that many IRG1L genes contain an N-terminal mitochondrial targeting sequence (MTS) that is absent from most IRG1 sequences. Limited proteolysis of submitochondrial localization confirmed that zebrafish IRG1L is only sensitive to proteases in the presence of high Triton X-100, indicative of association with mitochondrial matrix. In contrast, a recent paper from the Galan lab (Lian 2003 Nature Microbiology) reported that human IRG1 is not localized to the mitochondrial matrix, although enriched in mitochondria. Here, the authors generated a matrix-targeted human IRG1 by adding the N-terminal MTS and found that it localizes to the matrix based on a limited proteolysis assay. The loss of MTS-containing IRG1L from most mammals appears, therefore, to indicate that itaconate generation is directed to the cytoplasm, potentially reducing inhibition of TCA cycle activity in the mitochondria.

Next, the authors confirmed that the recombinant IRG1L protein has CAD activity in vitro. The last part of the manuscript addresses the expression of paleo-IRG1 in oysters and amphioxus, where they found high mRNA levels in oyster hemocytes which was further increased by poly(I:C), which was also the case in amphioxus tissues after feeding of LPS or poly(I:C), indicating a role for paleo-IRG1/itaconate in early metazoan innate immunity.

Strengths

(1) Phylogenetic perspective largely lacking so far in the IRG1/itaconate field.

(2) Manuscript clearly written and understandable across disciplines.

(3) Phylogenetic analyses complemented by biochemical and gene expression analyses to link to function.

(4) Lack of MTS in IRG1 and change in localization from mitochondria, highly relevant antimicrobial and cellular effects of itaconate.

Weaknesses:

(1) Biochemical and functional analysis of different CAD mRNA and proteins lacks depth.

(2) The submitochondrial localization assay lacks a native human IRG1 control.

(3) CAD activity shown for IRG1L but not paleo-IRG1.

(4) Itaconate production by early metazoans after PAMP stimulation?

(5) No measurement of energy metabolism (trade-offs?).

I acknowledge that some of these limitations are inevitable because the range of detailed experimental analysis is necessarily limited. However, some of these data would be important to support central claims of the manuscript (further discussed below).

Author response:

Reviewer #1 (Public review):

Summary:

The taxonomic analysis of IRG1 evolution is compelling and fills an important gap in the literature. However, the experimental evidence for IRG1 localization requires greater detail and confirmation.

Strengths:

The phylogenetic analysis of IRG1 evolution fills an important gap in the literature. The identification of independent acquisition of metazoan and fungal IRG1 from prokaryotic sources is novel, and the observation that human IRG1 lost mitochondrial matrix localization is particularly interesting, with potentially significant implications for the study of itaconate biology.

We thank the reviewer for appreciating the novelty of our study in exploring IRG1 evolution.

Weaknesses:

The protease protection assay was conducted with MTS-IRG1 but not with wild-type IRG1, which should also be tested. Moreover, no complementary methods, such as microscopy, were employed to validate localization. Beyond humans, the structure and localization of mouse IRG1, highly relevant given the widespread use of the mouse as a model for IRG1 functional studies, are not addressed.

Regarding submitochondrial localization of IRG1, we want to draw attention to the published data that a protease protection assay for wild-type mammalian IRG1 has been performed by Lian et al. 2023 (Extended Data Fig. 4), which convincingly demonstrated an outer-mitochondrial membrane localization of endogenous mouse IRG1 in mouse DC2.4 cells upon LPS stimulation that induces IRG1 expression.

Regarding complementary microscopy evidence, the same paper performed two-color, DNA-paint super-resolution imaging to demonstrate an enrichment of IRG1 to mitochondria with a lack of co-localization of the inner membrane/matrix marker Cox IV.

Given the direct visualization of sub-mitochondrial localization, we consider applying super-resolution microscopy to revisit the sub-mitochondrial localization of di[erent IRG1 constructs in the study.

Reference:

Lian H, Park D, Chen M, Schueder F, Lara-Tejero M, Liu J, Galán JE. Parkinson's disease kinase LRRK2 coordinates a cell-intrinsic itaconate-dependent defence pathway against intracellular Salmonella. Nat Microbiol. 2023 Oct;8(10):1880-1895. doi: 10.1038/s41564-023-01459-y. Epub 2023 Aug 28. PMID: 37640963; PMCID: PMC10962312.

Finally, if itaconate is indeed synthesized outside the mitochondrial matrix to safeguard metabolic activity, it is not discussed how this reconciles with its reported inhibitory e[ect on SDH.

We thank the excellent point raised by the reviewer. Indeed, itaconate has been proposed to inhibit matrix SDH exhibiting anti-inflammation function (Lampropoulou, Cell Metab 2016). While the mitochondrial transport of itaconate has not been fully characterized in vivo or in cells, a specific itaconate transport activity has been shown for the mitochondrial 2-oxoglutarate transporter OGC using in vitro proteoliposome system (Mills et al. Nature 2018).

We plan to discuss this important point on mitochondrial itaconate transport in the revision.

Reference:

Lampropoulou V, Sergushichev A, Bambouskova M, Nair S, Vincent EE, Loginicheva E, Cervantes-Barragan L, Ma X, Huang SC, Griss T, Weinheimer CJ, Khader S, Randolph GJ, Pearce EJ, Jones RG, Diwan A, Diamond MS, Artyomov MN. Itaconate Links Inhibition of Succinate Dehydrogenase with Macrophage Metabolic Remodeling and Regulation of Inflammation. Cell Metab. 2016 Jul 12;24(1):158-66. doi: 10.1016/j.cmet.2016.06.004. Epub 2016 Jun 30. PMID: 27374498; PMCID: PMC5108454.

Mills EL, Ryan DG, Prag HA, Dikovskaya D, Menon D, Zaslona Z, Jedrychowski MP, Costa ASH, Higgins M, Hams E, Szpyt J, Runtsch MC, King MS, McGouran JF, Fischer R, Kessler BM, McGettrick AF, Hughes MM, Carroll RG, Booty LM, Knatko EV, Meakin PJ, Ashford MLJ, Modis LK, Brunori G, Sévin DC, Fallon PG, Caldwell ST, Kunji ERS, Chouchani ET, Frezza C, Dinkova-Kostova AT, Hartley RC, Murphy MP, O'Neill LA. Itaconate is an anti-inflammatory metabolite that activates Nrf2 via alkylation of KEAP1. Nature. 2018 Apr 5;556(7699):113117. doi: 10.1038/nature25986. Epub 2018 Mar 28. PMID: 29590092; PMCID: PMC6047741.

Reviewer #2 (Public review):

Summary:

The authors are trying to explain how the metabolite itaconate evolved, since although it's involved in host defense, it can also limit mitochondrial function. They are trying to probe the trade-o[ between these two functions.

Strengths:

The evolutionary aspect is novel; this is the first time to my knowledge that the evolution of IRG1 has been analysed, and there are interesting findings here. The key finding appears to be that subcellular localisation is an important aspect, allowing host defense in some organisms without compromising bioenergetics. This is an interesting finding in the context of immunomebolism, although it needs extra analysis.

Weaknesses:

The work concerning sub-mitochondrial localisation is confusing and needs better analysis.

We thank the reviewer for the constructive feedback. As in our response to reviewer 1, we want to draw attention to the published data in which the outer mitochondrial membrane localization of IRG1 has been demonstrated by protease protection assay and explored using super-resolution imaging by Lian et al. 2023 (Extended Data Fig. 4). Given the direct visualization of sub-mitochondrial localization by super-resolution imaging, we plan to revisit and to apply the method to di[erent IRG1 constructs used in the paper.

Reviewer #3 (Public review):

Summary:

IRG1 is highly expressed in activated human and mouse myeloid cells. It encodes the mitochondrial enzyme cis-aconitate decarboxylase 1 (ACOD1) that generates itaconate. Itaconate has anti-microbial activity and acts immunoregulatory by interfering with cellular metabolism, signaling to cytokine production, and multiple other processes.

The authors perform a phylogenetic analysis of IRG1 to obtain insight into the evolution of itaconate biosynthesis. Combining BLAST with human IRG1 and a MmgE/Ptrp domain search, they find CAD in all domains of life, but the presence of IRG1 homologs is patchy in eukaryotes, indicating that itaconate biosynthesis is not essential. The phylogenetic analysis showed a more distant relationship of fungal and metazoan CAD/IRG1 to many prokaryotic sequences, suggesting independent acquisition of these metazoan and fungal CAD genes. In metazoans, three subbranches of paleo-IRG1 (in mollusks/early chordates) and two paralogous vertebrate forms (IRG1 and IRG1-like) were identified, with the latter derived from paleo-IRG1, and by genome duplication. While most jawed vertebrates have both IRG1 and IRG1L, metatherian and eutherian mammals have lost IRG1L and contain only IRG1.

Interestingly, sequence analysis of both paralogues showed that many IRG1L genes contain an N-terminal mitochondrial targeting sequence (MTS) that is absent from most IRG1 sequences. Limited proteolysis of submitochondrial localization confirmed that zebrafish IRG1L is only sensitive to proteases in the presence of high Triton X-100, indicative of association with mitochondrial matrix. In contrast, a recent paper from the Galan lab (Lian 2003 Nature Microbiology) reported that human IRG1 is not localized to the mitochondrial matrix, although enriched in mitochondria. Here, the authors generated a matrix-targeted human IRG1 by adding the N-terminal MTS and found that it localizes to the matrix based on a limited proteolysis assay. The loss of MTS-containing IRG1L from most mammals appears, therefore, to indicate that itaconate generation is directed to the cytoplasm, potentially reducing inhibition of TCA cycle activity in the mitochondria.

Next, the authors confirmed that the recombinant IRG1L protein has CAD activity in vitro. The last part of the manuscript addresses the expression of paleo-IRG1 in oysters and amphioxus, where they found high mRNA levels in oyster hemocytes which was further increased by poly(I:C), which was also the case in amphioxus tissues after feeding of LPS or poly(I:C), indicating a role for paleo-IRG1/itaconate in early metazoan innate immunity.

Strengths

(1) Phylogenetic perspective largely lacking so far in the IRG1/itaconate field.

(2) Manuscript clearly written and understandable across disciplines.

(3) Phylogenetic analyses complemented by biochemical and gene expression analyses to link to function.

(4) Lack of MTS in IRG1 and change in localization from mitochondria, highly relevant antimicrobial and cellular e[ects of itaconate.

We thank the reviewer for the positive comments with the strengths.

Weaknesses:

(1) Biochemical and functional analysis of di[erent CAD mRNA and proteins lacks depth.

We plan to explore two types of experiments:

First, we plan to purify di[erent CAD recombinant proteins; and if successful, we will test their in vitro enzymatic activity in synthesize itaconate. The positive data will also answer question (3) below.

Second, we plan to measure itaconate level in oyster hemocytes after PAMP stimulation, to demonstrate an in vivo itaconate production activity by paleo-IRG1. The data will also address question (4) below.

(2) The submitochondrial localization assay lacks a native human IRG1 control.

As in our response to reviewer 1, we believe Lian et al. 2023. provided strong evidence supporting an outer mitochondrial membrane localization of wild-type endogenous, mouse IRG1. Given the direct visualization using suer-resolution imaging, we plan to revisit submitochondrial localization of di[erent IRG1 constructs using super-resolution imaging.

(3) CAD activity shown for IRG1L but not paleo-IRG1.

We plan to purify di[erent CAD recombinant proteins; and if successful, we will test their in vitro enzymatic activity in producing itaconate.

(4) Itaconate production by early metazoans after PAMP stimulation?

We plan to measure itaconate level in oyster hemocytes after PAMP stimulation, to demonstrate an in vivo itaconate production activity by paleo-IRG1.

(5) No measurement of energy metabolism (trade-o[s?).

Because PAMP signaling might trigger other downstream e[ects that also impair mitochondrial function, for instance nitric oxide that inhibits complex IV, we plan to avoid PAMP condition and direct test the e[ect of itaconate production. We plan to compare the impact on mitochondrial bioenergetics, if the same CAD enzymes (thus with the same activity) can be expressed at the same level intra-mitochondrially and extramitochondrially, for instance in the case of MTS-hACOD1 and hACOD1.