Listeria monocytogenes infection rewires host metabolism with regulatory input from type I interferons.

Listeria monocytogenes (L. monocytogenes) is a food-borne bacterial pathogen. Innate immunity to L. monocytogenes is profoundly affected by type I interferons (IFN-I). Here we investigated host metabolism in L. monocytogenes-infected mice and its potential control by IFN-I. Accordingly, we used animals lacking either the IFN-I receptor (IFNAR) or IRF9, a subunit of ISGF3, the master regulator of IFN-I-induced genes. Transcriptomes and metabolite profiles showed that L. monocytogenes infection induces metabolic rewiring of the liver. This affects various metabolic pathways including fatty acid (FA) metabolism and oxidative phosphorylation and is partially dependent on IFN-I signaling. Livers and macrophages from Ifnar1-/- mice employ increased glutaminolysis in an IRF9-independent manner, possibly to readjust TCA metabolite levels due to reduced FA oxidation. Moreover, FA oxidation inhibition provides protection from L. monocytogenes infection, explaining part of the protection of Irf9-/- and Ifnar1-/- mice. Our findings define a role of IFN-I in metabolic regulation during L. monocytogenes infection. Metabolic differences between Irf9-/- and Ifnar1-/- mice may underlie the different susceptibility of these mice against lethal infection with L. monocytogenes.

Introduction

Food-borne bacteria cause hundreds of thousands of deaths every year [1]. Several of these pathogenic species induce metabolic changes in the host that determine the establishment of protective immunity [2]. Listeria monocytogenes (L. monocytogenes) is a prime representative of this group of pathogens. L. monocytogenes infection induces metabolic shifts including a reduction of fatty acid oxidation (FAO) and glycolysis [3] and also anorexia [4] in Drosophila melanogaster. Anorexia activates a PPAR-α-driven ketogenic program, and this protects murine hosts during bacterial infection [5]. Furthermore, blocking glycolysis in mice via 2-deoxyglucose (2-DG) protects against detrimental effects of L. monocytogenes infection [6,7]. Thus, available experimental evidence strongly suggests metabolic reprogramming as an important facet of the host response to L. monocytogenes infection and as one of the parameters determining its outcome.

Following recognition and signaling via pattern recognition receptors (PRR), L. monocytogenes causes host cells to produce and release several antimicrobial molecules, cytokines and chemokines, including type I interferons (IFN-I) [8,9]. Engagement of IFN-I with their receptor, a heterodimer of IFNAR1 and IFNAR2 chains, initiates a signaling cascade culminating in activation of the downstream transcription factor ISGF3 (IFN stimulated gene factor 3) and its association with target gene promoters. ISGF3 is assembled from signal transducers and activators of transcription 1 and 2 (STAT 1 and 2) as well as IFN regulatory factor 9 (IRF9), the DNA-binding component of the complex. ISGF3 activity contributes to a robust innate immune response against the invading pathogen. However, unlike their well-known protective effects against viruses, IFN-I have an adverse effect on innate immunity to L. monocytogenes infection, decreasing the resistance of mice against lethal infection [8,10].

Classical M1 activation of macrophages via Toll-like receptor signaling induces glycolysis and FA synthesis (FAS) which are required for enhanced effector functions [11]. In contrast, FAO is a hallmark of alternatively activated macrophages and crucial for their tolerogenic functions. IFN-I were shown to promote glycolysis and oxidative phosphorylation (OXPHOS) in plasmacytoid dendritic cells (pDCs) [12]. The authors of this study concluded that the OXPHOS in CpG-stimulated pDCs is mainly driven by fatty acid oxidation (FAO), which is dependent on IFNAR signaling. In high-fat-diet models IRF9 was shown to regulate expression of fatty acid (FA) metabolism genes and in this way protect from hepatic steatosis and insulin resistance [13]. Together, these studies support the conclusion that IFN-I and their signal transducers regulate metabolism, adding yet another activity to their well described roles in antimicrobial and inflammatory responses.

Once L. monocytogenes crosses the intestinal barrier, it reaches the liver, the body’s major metabolic hub. Reportedly, L. monocytogenes infection reduces liver ATP and NAD+ levels and causes an associated impairment of OXPHOS [14,15]. Further impact of L. monocytogenes on liver metabolism or a potential metabolic control by IFN-I and their transcription factor ISGF3 has not been investigated. Here we show that L. monocytogenes infection induces a gene expression program in the liver that affects various metabolic pathways including FA metabolism and OXPHOS and inhibition of FAO partially protects mice from detrimental effects of L. monocytogenes infection. Both IRF9 and IFNAR1 deficiency alter liver and macrophage metabolism during infection, but in some cases their impacts differ. Our findings indicate that IFN-I are integral to metabolic regulation during L. monocytogenes infection and that some of their activities do not require ISGF3/IRF9-mediated transcription.

Results

Irf9-/- and Ifnar1-/- mice differ in the level of protection from L. monocytogenes infection

IFN-I govern their target genes through formation of the ISGF3 complex. The IRF9 subunit of this complex is essential for its association with DNA. We sought to determine whether the protection from the adverse effects of L. monocytogenes infection of Ifnar1-/- mice [16-18], resulted entirely from ISGF3-dependent activities of IFN-I or whether ISGF3-independent effects contribute [19]. We thus infected mice that lack either IRF9 or IFNAR1 with L. monocytogenes for 10 days and monitored their survival rates. Consistent with previous reports, IFNAR1 deficiency enhanced the survival of infected mice. Surprisingly, Irf9-/- mice were protected to an even higher degree. (. Under the experimental parameters of our infection experiments the increase in resistance produced by IRF9-deficiency was highly significant. However, protection of Ifnar1 mice did not reach a 95% confidence interval although it was a highly reproducible phenotype in line with the literature [8,10,20]. Unexpectedly, this difference in survival rates between Irf9-/- and Ifnar1-/- mice was not reflected by the bacterial load of the liver and spleen three days post-infection (. Thus, the benefit of IRF9 versus IFNAR1 deficiency does not result from increased killing of the bacteria during the early innate response.

IFN-I are known to control a subset of inflammatory cytokines and chemokines. To investigate whether IRF9 and IFNAR1 deficiencies differ in the synthesis of inflammatory mediators, we profiled cytokines and chemokines released into the blood after L. monocytogenes infection. Infection led to increases in proinflammatory CCL2, IFNγ, IL6 and CXCL10 in wt mice. (Figs . Furthermore, both IRF9 and IFNAR1 deficiencies reduced the amount of IL6 and the monocyte attractant CCL2, with CCL2 being affected more in absence of IRF9 than IFNAR1. Consistently, we found reduced inflammatory monocyte recruitment to the peritoneal cavity of infected Irf9-/- mice ( while differences between neutrophil recruitment were not significant (. Monocyte recruitment is considered a protective component of the innate response to L. monocytogenes [21]. Thus, reduced monocyte recruitment fails to explain the increased protection observed in IRF9-deficient mice.

The death of splenic T cells and hepatocytes was shown to increase the susceptibility of mice to L. monocytogenes infection [22,23]. To assess the contribution of IFN-I and IRF9 to the cytotoxic effect of infection and liver damage, we measured serum levels of alanine aminotranferase (ALT). Both knockouts were similarly protected from L. monocytogenes-induced liver damage (. Furthermore, we quantified the total splenocyte death in infected mice. Similar to liver damage, animals from both knockout mice strains showed diminished splenocyte death compared to wt mice (. In contrast, the inflammatory infiltrate in the peritoneal cavity of the same mice did not show differences in cell viability between wt and gene-deficient mice (. Collectively, these results show that IFN-I signaling augments liver damage and splenocyte death in L. monocytogenes-infected mice with similar contributions of IFNAR1 and IRF9.

Macrophages are the most important effector cells in L. monocytogenes infection [8,24]. In line with the literature [17], we detected no change in the number of intracellular viable L. monocytogenes in both Irf9-/- and Ifnar1-/- primary bone-marrow-derived macrophages (BMDMs) compared to wt. This suggests similar killing potential for L. monocytogenes (. Collectively, these data show that the implications accounting for the detrimental effects of IFN-I, such as IFN-I-induced liver damage and splenic cell death, show no difference between the knockouts. Furthermore, defective monocyte recruitment cannot account for the protection. Therefore, these parameters cannot explain the ISGF3-independent role of IFN-I in the fight against L. monocytogenes.

IFN-I signaling interferes with glutaminolysis and enhances FAO in macrophages

Macrophages are critical in establishing the first line of defense against L. monocytogenes. Influence of IFN signaling on the metabolism of immune cells is well established [25]. Mining of recent macrophage RNAseq data [26] showed that metabolic pathways such as fatty acid (FA) metabolism and oxidative phosphorylation (OXPHOS) gene sets are among the most differentially expressed gene sets in Irf9-/- macrophages treated with IFN-I (). To detect whether wt, Irf9-/- and Ifnar1-/- BMDMs experienced changes in FA metabolism when infected with L. monocytogenes, we performed targeted liquid chromatography-tandem mass spectrometry (LC-MS) measurements of free FA. Intracellular FA levels (, as well as expression of Cd36, a protein responsible for FA uptake (, were not affected by the loss of IRF9 or IFNAR1. Acetyl-CoA Carboxylase (ACC) is responsible for conversion of acetyl CoA into malonyl CoA and its activity is inhibited by phosphorylation. Malonyl CoA is the first substrate of de novo lipogenesis and it inhibits CPT1, the enzyme responsible for conjugation of FAs to carnitine, thereby, inhibiting FA oxidation (FAO) [27]. Intensities of phospho-ACC (pACC) bands obtained by western blotting were decreased in both Irf9-/- and Ifnar1-/-, with infected Ifnar1-/- BMDMs bearing significantly less pACC (Figs . FAs are imported into the mitochondria via carnitine for FAO. Whereas conjugated carnitine (stearoyl carnitine) is engaged in FA import into the mitochondria, free carnitine is not. Our targeted LC-MS measurement for these metabolites showed that Ifnar1-/- BMDMs had lower stearoyl carnitine after infection, despite similar levels of free carnitine compared to wt (. This suggests that less FAs are transported into the mitochondria for oxidation in the absence of IFN-I signaling.

Seahorse flux analysis is a tool to assess cellular OXPHOS via oxygen consumption rate (OCR). OXPHOS is fueled mainly by oxidation of glucose, amino acids and FA. In addition, glycolytic rates are determined by measurement of extracellular acidification rates (ECAR). In line with less FA import into the mitochondria in Ifnar1-/-, Seahorse flux analysis demonstrated that infection induced an enhancement in OXPHOS in all genotypes, however, Ifnar1-/- macrophages had lower OXPHOS than wt before and after infection (Figs . In contrast, L. monocytogenes infection increased glycolysis in each genotype without an impact of IRF9 or IFNAR1 (Figs . Taken together, our data led us to assume that despite similar FA levels in all genotypes, transport of FA into the mitochondria for oxidation is enhanced by IFNAR signals. These effects were much less prominent in Irf9-/- macrophages, suggesting the contribution of an IRF9-independent pathway. To strengthen our interpretation, we measured the OXPHOS of uninfected or L. monocytogenes-infected wt, Irf9-/- and Ifnar1-/- BMDMs after inhibiting CPT1 with etomoxir. We treated the cells with etomoxir at the maximal OCR after uncoupling OXPHOS with FCCP (Carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone). Then, we determined the drop in OCR caused by etomoxir prior to complete shut-down of respiration with Rotenone/Antimycin. We observed less reduction in max OCR, i.e., less FAO-dependent OXPHOS, in L. monocytogenes-infected Ifnar1-/- but not in Irf9-/- BMDM. This is consistent with an IRF9-independent contribution of IFN-I to the induction of FAO. (Figs . As OXPHOS was downregulated in the Ifnar1-/- BMDMs (, we also analysed the contribution of glutaminolysis, a mechanism of TCA cycle anaplerosis. To this end, we measured OXPHOS after addition of the glutaminolysis inhibitor BPTES. The drop in glutaminolysis-dependent OCR was slightly more pronounced in Ifnar1-/- compared to wt BMDMs after infection (Figs . These data allow us to conclude that L. monocytogenes-infection-induced IFN-I lead to increased FAO which may be balanced by a reduction in glutaminolysis.

Transcriptome changes in L. monocytogenes-infected mouse livers demonstrate regulation of genes related to FA metabolism and OXPHOS

IFN-I were recently identified as major determinants of metabolic changes occurring in hepatocytes during lymphocytic choriomeningitis virus (LCMV) infection [28,29]. This, together with the liver being a main target organ of L. monocytogenes infection, prompted us to investigate whether and how infection of wt, Irf9-/- or Ifnar1-/- mice changed hepatocyte transcriptomes and, particularly, genes related to metabolic pathways. We carried out RNA sequencing analyses of livers harvested from wt, Irf9-/- and Ifnar1-/- mice that were injected with PBS or L. monocytogenes. Gene set enrichment analysis (GSEA) comparing untreated wt with untreated Irf9-/- or Ifnar1-/- mouse liver transcriptomes resulted in the enrichment of IFN-I-stimulated genes (ISG) in the wt mouse livers, in line with a role for low-level IFN-I in maintaining homeostatic gene expression [26,30] (. Additionally, few metabolic pathways such as oxidative phosphorylation (OXPHOS), fatty acid (FA) metabolism, adipogenesis, cholesterol homeostasis and glycolysis were affected by the lack of IRF9 or IFNAR1 (. Interestingly, under homeostatic conditions only 36% (580 out of 1595) of the genes enriched in Ifnar1-/- compared to wt were common with Irf9-/- enriched genes ( pointing to a role of IFN-I signaling independent of IRF9 activity in mouse liver.

Upon infection, wt livers showed enrichment of 1390 genes (. In addition to the inflammatory and interferon response pathways, these included OXPHOS, FA metabolism, glycolysis and adipogenesis hallmark genes in wt livers ( suggesting pronounced metabolic changes upon infection. Strikingly, when comparing gene sets in Irf9-/- or Ifnar1-/- with wt after infection, we identified OXPHOS and FA metabolism gene enrichment in Irf9-/- and Ifnar1-/- (. However, only 67.4% (93 out of 138) of enriched OXPHOS genes and 48.5% (50 out of 103) of enriched FA metabolism genes were shared between the Irf9-/- and Ifnar1-/- (. Interestingly, validation qPCRs showed increased expression of OXPHOS (Cox5b, Atp5h, Suclg1, Ndufb5) and FA metabolism (Cpt1a, Sdha, Acadl) genes in both Irf9-/- and Ifnar1-/- livers before and after L. monocytogenes infection (Figs . On the other hand, ISGs (Ifit3 and Irf7) displayed lowered expression in the absence of IRF9 and IFNAR1, as expected (. Although increased gene expression patterns seem to contradict the reduction of OXPHOS and FAO observed in BMDMs, it is likely to imply a positive feedback mechanism leading to decreased activity of these pathways. Our differential RNAseq analysis showed that IFN-I signaling and IRF9 regulate metabolic gene expression both under homeostasis and during L. monocytogenes infection in part by distinct mechanisms. This suggests that IFN-I and IRF9 control the immunometabolism of L. monocytogenes infection in the liver.

Transcriptome changes in L. monocytogenes-infected mouse livers demonstrate regulation of genes related to FA metabolism and OXPHOS.

GSEA of RNAseq data from livers of mice one day post-infection. Dot plots of GSEA results showing the first 20 enriched pathways in (A) infected compared to uninfected wt with the heatmaps of top 15 genes with the highest enrichment score of inflammatory response and IFN alpha response pathways, (B) infected Irf9-/- and Ifnar1-/- compared to infected wt. Color and size of the dots represent adjusted p values and the number of the genes represented from a certain gene set, respectively. The x-axes were calculated by the sum of the core-enriched genes divided by its set size and ordered decreasingly. The y-axis represents the corresponding gene set name. GSEA enrichment plots for OXPHOS and FA metabolism of infected (C) Irf9-/- and (D) Ifnar1-/- compared to infected wt with the heatmaps of top 15 genes with the highest enrichment score of OXPHOS and FA metabolism pathways. (E) Venn diagrams showing the number of genes enriched in infected Irf9-/- and Ifnar1-/- compared to infected wt in OXPHOS and FA metabolism pathways. unt: untreated, Lm: L. monocytogenes infected.

L. monocytogenes infection and IFN-I signaling induce metabolic changes in the livers of mice

While the gene expression analysis showed alterations in metabolic genes, the changes are diverse and do not readily provide cues on how innate immunity to L. monocytogenes might be affected. Furthermore, the role of IFN-I signaling and IRF9 in this context is not clear. Therefore, we performed a targeted LC-MS-based metabolic screen of several metabolites including glycolytic, TCA cycle, FA metabolism, glutaminolysis intermediates and amino acids in the livers of mice. We applied principal component analysis (PCA) to the data. PC1 and PC2 showed separation of infected and uninfected samples independent of the genotype and accounted for 21% and 13% of the total variance, respectively (. Directional vectors on the loading plots ( implicate the significance of the metabolites in the clustering observed in the PCA. The bar graphs quantify the strength of contribution of each metabolite to the PC separation (. Relative changes in the amounts of the FAs (palmitic acid, oleic acid, palmitoleic acid), glucose and glycolytic intermediates (glucose-6-phosphate, fructose-6-phosphate and glyceraldehyde-3-phosphate) and amino acids (leucine, isoleucine, serine and valine) were the main drivers of this separation. This is in line with literature showing infection-induced oxidation of major energy sources such as glucose, fatty acids and amino acids [4]. We also detected clustering of three-day-infected mice in PC3 and PC4 ( showing that infection impacts on all basal carbon sources. Although PC1 and PC2 did not show a contribution to the separation of the genotypes, PC3 and PC4 separated the same dataset according to genotype while delineating 11% and 10% of the total variance, respectively (. We identified the Ifnar1-/- clustered distinctly from wt and Irf9-/-, mainly because of the TCA intermediates succinate, malate and fumarate and also γ-aminobutyric acid (GABA) (, suggesting the regulation of TCA cycle intermediates and GABA levels by IFN-I independently of IRF9. This is in line with the finding of TCA cycle replenishment via the GABA shunt [31]. These data show that both L. monocytogenes infection and IFN-I signaling cause metabolic changes in the liver.

Although the PCA plots show the contribution of metabolites to the separation of a certain cluster, they do not depict the specific alterations of those metabolites. As the RNAseq data showed enrichment of FA metabolism genes in both Irf9-/- and Ifnar1-/- livers during infection, we compared FA levels in more detail by hypothesis testing. In line with the PCA analysis, all genotypes showed reduction of palmitic acid, palmitoleic acid and oleic acid after infection whereas stearic acid increased only in wt (. Additionally, we detected an increase in free carnitine levels upon infection that was less pronounced in the knockouts, particularly in Irf9-/- (. In contrast, there was significantly less increase of stearoyl carnitine in both knockout livers compared to wt upon L. monocytogenes infection. Stearoyl carnitine was reduced in both Irf9-/- and Ifnar1-/- livers in three-day-infected mice providing less FA import into the mitochondria for FAO. In line with the data obtained in macrophages, western blot analysis showed less pACC in Ifnar1-/- liver lysates with Irf9-/- having only slightly less pACC (Figs . The data suggest that similar to Ifnar1-/- macrophages, Ifnar1-/- livers perform less FAO than wt upon L. monocytogenes infection, possibly due to less FA import into the mitochondria for oxidation.

L. monocytogenes infection and IFN-I signaling cause changes in TCA cycle, FA and glutamine metabolism intermediates of liver.

Corrected EIC (extracted ion counts) of metabolites of (A) FAO, (C) TCA cycle and (D) GABA plotted in Fig 4. The values and false discovery rate p values were calculated using values corrected for liver weight, sex of the mouse and experimental variation effects. ANCOVA test was performed. (B) Quantification of ACC phosphorylation as determined by western blot. Bars show the mean values. Each data point represents one western blot replicate of one-day infected mouse liver lysates. Error bars show minimum and maximum values. P value was calculated using ANOVA with Dunnett’s and Geisser-Greenhouse correction. (E) Bacterial loads of spleen and liver of etomoxir-treated mice three days post-infection. P values were calculated using two-tailed Mann-Whitney t-test.

Fig 4

L. monocytogenes infection and IFN-I signaling induce metabolic changes in the liver.

PCA plots of metabolites quantified by targeted LC-MS/MS from livers of (A-B) uninfected, one-day- or three-day-infected and (C-D) wt, Irf9-/- and Ifnar1-/- mice. Percentage of total variance is indicated in the axis label. Bar graphs showing contribution of each metabolite to clustering in (E) PC1 versus PC2 and (F) PC3 versus PC4. The values are corrected for liver weight, genotype, sex of the mouse and time point effects. ANCOVA test was performed. Contrib.: contribution, PC: Principal component.

Reportedly, ketogenesis induced during bacterial infection is a means of protecting the host [5] and ketone bodies (KB) are produced from FAO-derived acetyl-CoA in the liver. Based on our finding that FA metabolism is controlled by IFN-I signaling, we speculated that both knockouts might cause increased KB formation to protect the host from L. monocytogenes infection (. However, in both Irf9-/- and Ifnar1-/- livers the KB acetoacetate and hydroxybutyric acid were decreased three days post-infection.

Other than FA metabolism, the RNAseq data showed enrichment of OXPHOS genes in infected Irf9-/- and Ifnar1-/- mouse livers compared to wt. Furthermore, we also detected lower OXPHOS in Ifnar1-/- BMDMs. Since changes in OXPHOS are expected to alter or result from the flow of metabolites through the TCA cycle, we measured TCA cycle intermediates. While infected Irf9-/- livers contained increased fumarate and malate, Ifnar1-/- had lower levels of these metabolites (, in line with the PCA plots (Figs . These data suggest that both IRF9 and IFN-I take part in the regulation of liver TCA metabolite levels but differentially target its metabolites.

Glutaminolysis can be used to replenish the TCA cycle by converting glutamine into α-KG or via GABA shunt (. PCA analysis suggested GABA as a reason for the separation of Ifnar1-/- livers (Figs . Therefore, we measured the levels of intracellular GABA, the intermediate product of glutaminolysis, and found that Ifnar1-/- but not Irf9-/- had higher levels of GABA before and after infection (. We also noted a similar increase in GABA levels in uninfected Ifnar1-/- mice. Increases in expression of Slc25a22, the mitochondrial glutamate carrier, in Ifnar1-/- is in line with this observation (. In conclusion, our data suggest that IFN-I signaling has a role in regulating TCA cycle and glutamine metabolism in the liver. They also strengthen the conclusion that Irf9-/- and Ifnar1-/- mice differ in their impact on liver metabolism, both dependent and independent of infection with L. monocytogenes.

Having identified reduced FAO in BMDMs and livers of Ifnar1-/- mice and also slightly of Irf9-/- mice, we determined whether reduction of FAO provides protection against L. monocytogenes infection. Mice were treated with the FAO inhibitor etomoxir one day prior to and one day post-infection. [12] Our data showed that inhibition of FAO reduced the bacterial loads of spleens and livers of infected wt mice while having no or a mild effect on the knockouts (. These data lead us to conclude that FAO induced during L. monocytogenes infection has an adverse impact on the host and that decreased FAO in the absence of IFN-I signaling might account for part of the protection resulting from IRF9 or IFNAR1 deficiency.

Discussion

Metabolic reprogramming has been widely recognized as an important attribute of cells engaged in various aspects of antimicrobial immunity. Many studies demonstrate that reprogramming occurs upon engagement of PRR by macrophages and by the products of PRR signaling such as IL-1 [31]. IFN-I also contribute to metabolic reprogramming. For example, a recent study of LCMV infection demonstrates that IFN-I impact on antiviral immunity by disrupting the urea cycle in the liver [29]. Here we sought to investigate whether macrophage and liver metabolism is altered by L. monocytogenes infection and whether any metabolic effects of L. monocytogenes infection are influenced by IFN-I and their transcription factor ISGF3 (.

Model of Listeria monocytogenes-induced metabolic rewiring.

(A) Listeria monocytogenes (Lm) induced-IFN-I regulate FA import into the mitochondria. Absence of autocrine IFNAR1 signalling leads to decreased fatty acid oxidation (FAO) in Irf9-/- and Ifnar1-/- BMDMs and livers, and also reduced TCA cycle intermediate levels and oxidative phosphorylation (OXPHOS) as well as increased glutaminolysis exclusively in Ifnar1-/- BMDMs and livers.

Surprisingly, protection of Ifnar1-/- mice from L. monocytogenes infection is less pronounced than that provided by IRF9 deficiency. CFU assays in the target organs show that this survival variation derives from differences in resilience of mice rather than antimicrobial effector mechanisms. There are several alternative or complementary explanations for this finding. The most straightforward assumption is that not all immunoregulation by IFN-I signaling requires de novo gene transcription via ISGF3. The interpretation of our data is mainly guided by this hypothesis. The IFN-I receptor has the potential to activate MAPK and PI3K pathways [32] which may contribute to the influence of IFN-I on innate immunity to L. monocytogenes. For further consideration, IRF9 and IFNAR deficiency may differ in their effects on innate resistance to L. monocytogenes due to diverse effects on transcriptional homeostasis. This notion derives from our recent observation that IRF9 contributes to basal expression of its target genes independently of a tonic signal from the IFN-I receptor [26]. Finally, ISGF3 activity downstream of the IFN-III receptor or yet other, unexplored non-IFN-I pathways, may contribute to diverse effects of the IRF9 and IFNAR knockouts.

Mining of previous macrophage RNAseq data from our lab [26] supported a role of IRF9 in regulating FA metabolism and OXPHOS in response to IFN-I treatment. Targeted metabolite and metabolic pathway measurements in macrophages indicated L. monocytogenes infection to increase both glycolysis and OXPHOS. Ifnar1-/- macrophages showed a decrease in OXPHOS compared to wt. In accordance with the smallest amount of conjugated FA and the largest inhibition of ACC, we noted smallest FAO and also largest glutaminolysis dependency in Ifnar1-/- macrophages. Studies by others in IFN-I or poly I:C treated primary mouse macrophages have shown that IFN-I signaling inhibits cholesterol and FAS while enhancing lipid import for oxidation [33,34]. These changes strengthen the anti-viral response. Another study demonstrated the ability of IFN-I to induce FAO in pDCs [12]. These results concur with our data in showing a role for IFN-I in metabolic adaptation of myeloid cells to infection [25,35]. That being said, the small metabolic effects do not alone contribute to the substantial differences between Irf9-/- and Ifnar1-/- mice in resistance to L. monocytogenes infection, however, our research further elucidates the influence of IFN-I signaling in metabolic reprogramming.

Investigation of livers from infected mice by RNAseq and targeted metabolite measurements indicated transcriptional and metabolic reprogramming, part of which is dependent on IFN-I signaling. Measurement of conjugated-carnitine and ACC phosphorylation showed that particularly IFNAR1 deficiency reduced FAO without affecting total intracellular FA levels. A likely explanation for this finding is that IFN-I induce FA import into the mitochondria, similar to our observations in macrophages. Reduced FAO in Ifnar1-/- livers might also explain the decreased amounts of TCA metabolites. It also in part may explain the protection from the adverse effects of L. monocytogenes infection, as FAO inhibition with etomoxir reduced bacterial loads in spleens and livers of wt mice albeit not to the degree observed in Irf9-/- or Ifnar1-/ mice. 80% of the liver volume is constituted by hepatocytes. Therefore, it appears safe to assume that liver RNAseq as well as the metabolite measurements reflect predominantly hepatocyte transcriptomes and metabolomes. IFN-I-dependent FAO can thus be considered a common facet of macrophage and hepatocyte metabolism and a joint contributor to the detrimental effects of IFN-I on L. monocytogenes infection.

Although the metabolite differences between the livers of wt, Irf9-/- and Ifnar1-/- mice are relatively low, they are in line with the gene expression patterns obtained from the same mice. Moreover, PCA plots show that infection induces clear changes in the liver metabolism of main carbon sources, namely glucose, FA and amino acids. This is again consistent with the genotype-independent glycolytic switch in macrophages. Additionally, both liver and macrophages reflect an activating effect of IFN-I on FA metabolism during L. monocytogenes infection. However, we speculate that the FAO reduction in both Irf9-/- and Ifnar1-/- can be compensated by glutaminolysis only in Ifnar1-/-. This hypothesis is supported by increased GABA in Ifnar1-/- livers and by the slightly increased OXPHOS dependency of Ifnar1-/- macrophages on glutaminolysis compared to wt. Of note, recent reports attribute a protective effect to GABA in both acute liver injury [36] and L. monocytogenes infection [37], although involvement of IFN-I signaling has not been addressed in the latter study. Therefore, we speculate that augmented GABA production in Ifnar1-/- livers might contribute to the protection of Ifnar1-/- mice from L. monocytogenes infection. The mechanism behind the protective effects of systemic GABA is subject to further study and metabolic tracing experiments are needed to validate the relationship between GABA synthesis and the protective effect of the Ifnar1-/-.

Differences in metabolite levels and pathway outputs between Irf9-/- and Ifnar1-/- macrophages and livers are consistent with the assumption that the two genotypes produce differences in the FA metabolism. Most likely, IFN-I regulate FAO to some extent without requiring ISGF3-dependent de novo gene transcription. This is reflected in RNAseq data from infected livers as well. Other than the expected cytokine and inflammatory signaling pathways, we found OXPHOS and FA metabolism pathways among the most differentially regulated pathways in both Irf9-/- and Ifnar1-/- compared to wt, suggesting a metabolic dysregulation in these mice. However, Irf9-/- and Ifnar1-/- enriched gene subsets vary considerably, suggesting an IRF9-independent role for IFN-I in regulating metabolic gene expression in the liver following L. monocytogenes infection. To our surprise, a closer look at the expression pattern of single genes showed increased expression of the OXPHOS and FA metabolism genes in Irf9-/- and Ifnar1-/- livers both before and after infection. This led us to speculate that disturbance of these pathways results in reduced OXPHOS and FAO. This may cause a boost in gene expression due to a positive feedback response to compensate for the reduced pathway activity. Also, post-transcriptional regulatory mechanisms might contribute to the effect we have observed. However, these need to be tested in further detail. In addition, differences in FA metabolism may result from ISGF3-independent IRF9 activity. This is in line with previous studies demonstrating an IRF9 involvement in FA metabolism, insulin resistance [13], cardiac hypertrophy [38] and hepatic-ischemia-reperfusion injury [39]. These roles of IRF9 beyond IFN-I signaling may contribute to the survival differences we detected in L. monocytogenes-infected Irf9-/- and Ifnar1-/- mice.

In conclusion, our data establish metabolic reprogramming of both liver and macrophages upon L. monocytogenes infection. Although IFN-I signaling acts similarly in liver and primary macrophages in regulating FAO and glutaminolysis, the roles acquired by the hepatocytes during infection has to be further elucidated. While IRF9 is downstream counterpart and responsible for most of the of IFN-I-induced gene expression, IFN-I affect L. monocytogenes infection and its characteristic metabolism independently of IRF9 and vice versa in liver and macrophages. Furthermore, we show that IFN-I signaling is an essential player in the regulation of TCA metabolite levels, FAO and thus OXPHOS and inhibition of FAO has a protective role during L. monocytogenes infection. Understanding the metabolic pathways controlled by IRF9 and IFN-I during L. monocytogenes infection might lead the way to a better understanding of listeriosis.

Materials and methods

Ethics statement

All, wt, Irf9-/- and Ifnar1-/- mice in C57BL6/N background, were bred and maintained at the Max Perutz Labs under specific pathogen-free conditions according to Federation of European Laboratory Animal Science Associations (FELASA) guidelines. All animal experiments were approved by the ethics and animal Welfare Committee of the University of Veterinary Medicine and the national authority (Austrian Federal Ministry of Science and Research) according to §§ 26ff. of Animal Experiments Act, Tierversuchsgesetz 2012-TVG 2012 (BMWFW-68.205/0032-WF/II/3b/2014; BMWFW-68.205/0173-V/3b/2019) and conform to the guidelines of FELASA and ARRIVE (Animal Research: Reporting of In Vivo Experiments). Age and sex-matched (8–10 weeks) mice were used for all experiments.

Cell culture

Bones harvested from mice were crashed and filtered using 70 mm cell strainer to isolate bone marrows (BM). All BM cells were grown and differentiated in DMEM + 10% FCS +100 U/ml Penicilin + 100 μg/ml Streptomycin medium with 500 ng/ml MCSF (kind gift of Löms Ziegler-Heitbrock, Helmholtz Center, Munich, Germany) for 10 days. The day before infection, macrophages were seeded in DMEM + 10% FCS + 500 ng/ml MCSF. For western blotting, qPCR and flow cytometry experiments 106 cells, for seahorse flux assays 3.5x104 cells and for targeted metabolite measurements 107 cells were seeded in one well of six-well plates, 96-well seahorse plates and 15 cm plates, respectively.

BMDM infections

For BMDM infections the LO28 strain of L. monocytogenes was inoculated in brain heart infusion (BHI) medium for overnight growth. The calculated volume (1OD600nm = 5x108 viable bacteria) was directly pipetted into the wells with BMDMs. Medium containing L. monocytogenes was changed into 50 μg/ml gentamicin + DMEM + 10% FCS after 1 hour of infection, then, into 10 μg/ml gentamicin + DMEM + 10% FCS after 2 hours of infection, which was kept on the cells until the end of the experiment. For in vitro colony forming unit (CFU) assays, 50000 BMDMs per well were seeded in 96 well plate. The cells were washed twice with PBS after the duration of infection was over and lysed in 50 μl ddH2O twice for 5 min at 37°C. Three 1:10 serial dilutions of these lysates were plated on Brain Heart Infusion (BHI) plates for quantifying bacterial loads.

Mouse infections

Overnight cultures of the InlA* mutant LO28 strain of L. monocytogenes [18] were diluted to OD600nm 0.1 and cultured until OD600nm reached a value of 1. Then the bacterial pellets were washed and diluted in PBS. 106 bacteria calculated as above were used to intraperitoneally infect the 8-10-week-old mice. For survival assays, the weights of mice were monitored for 10 days after infection and the mice that lost 20% of their weight were sacrificed due to arrival at the humane endpoint. For CFU assays, spleens and livers were harvested three days after infection, homogenized in PBS and four 1:10 serial dilutions of these homogenates were plated on Oxford agar (Merck, #107004) supplemented with Listeria-selective supplement (Merck, #107006) for quantifying bacterial loads. For the FAO inhibition experiment, 20mg/kg etomoxir or corresponding volume of PBS was injected intraperitoneally into mice one day prior to and post-infection. Etomoxir-treated and PBS-treated mice were held in separate cages. Large lobe of the liver and spleen of the mice were harvested three days post-infection. For measurement of blood glucose, ALT and cytokines, the blood was withdrawn via heart puncture in EDTA tubes and spun at 13000 g for 5 min to obtain the plasma. ALT was measured by Labor Invitro GmbH (Vienna, Austria). Plasma cytokines were measured using Luminex-based, customized mouse 15-plex Invitrogen Procartaplex Immunoassay for mix&match panels according to the manufacturer’s instructions.

RNA sequencing and analysis

30 mg liver tissue from PBS-treated or one-day infected mice were shock-frozen in liquid nitrogen and crushed using a mortar and pestle. Tissues were further homogenized with a syringe and needle and transferred into ice-cold tubes for RNA isolation using Allprep DNA/RNA micro kit (Qiagen, #80284) according to manufacturer’s instructions. The experiment was performed using three biological replicates per condition. The quality controls, RNA sequencing and preliminary analysis were performed by the Biomedical Sequencing Facility (CeMM, Vienna, Austria). Reads were mapped with TopHat to mm10 genome (RefSeq UCSC Mouse Dec. 2011/GRCm38/mm10 assembly). Cufflinks tool was used to assemble transcripts, estimate abundances and for differential expression analysis. All the RNA sequencing data is available in NCBI GEO repository on series record GSE 162448. GSEA analysis was performed with GSEA 4.0.3 using these FPKM values. Dot plots were generated with R software (version 4.0.2) using signal-to-noise (s2n) ratio for every gene set. The s2n was calculated by substracting the means of the gene counts per condition. This result was then divided by the sum of the deviations per condition. P value adjustment was done using the Benjamini hochberg procedure and a p value cutoff of 1 was used. Venn diagrams were generated using Venny 2.1 with the gene lists obtained from the GSEA analysis.

Metabolite measurements

150 mg liver tissue was shock-frozen in liquid nitrogen and homogenized in 1.5 ml ice-cold HPLC grade MeOH:ACN:H2O (2:2:1, v/v) with tissue homogenizer at 30000 rpm for 15 sec. The homogenates were incubated on ice for 5 min and lysed further with two freeze-thaw cycles (5 min liquid nitrogen, 3 min 37°C). The lysates were centrifuged at 4000 g for 10 min at 4°C and supernatants were used for metabolite measurement by the Vienna Biocenter Metabolomics Core Facility. LC-MS data were subjected to statistical analysis using R software. The data are corrected for mouse sex, liver weight and experimental variation and ANOVA was used for analysis.

For macrophage metabolites 107 BMDMs were scraped and washed in PBS and cell pellets were vortexed in 1 ml ice-cold HPLC grade MeOH:ACN:H2O (2:2:1, v/v) for 30 sec. Samples were exposed to two vortex-freeze-thaw-sonication cycles (1 min liquid nitrogen, 10 min thaw at RT sonicator) and incubated at -20°C for 1 hour for precipitating proteins. Following centrifugation at 13000 rpm at 4°C for 15 min, the supernatants were used for targeted metabolite measurement performed by the Vienna Biocenter Metabolomics Core Facility. The precipitated proteins were resuspended in loading buffer (6M urea in 1.5x Laemmli buffer), 20 μl of this resuspension was run in 10% SDS-PAGE for western blotting. Tubulin was detected using anti-tubulin antibody diluted 1:5000 in 2% BSA (Sigma, #T9026) incubated 2 hours at RT followed by incubation in mouse secondary antibody (1:5000 in TBST) for 1 hour at RT. Tubulin band intensities, quantified using Image J64 (version 1.48), were used for normalization of metabolites.

RNA isolation, cDNA synthesis and qPCR

Total RNA was isolated using NucleoSpin RNA kit (Macherey-Nagel, #740955), according to manufacturer’s instructions. For liver tissue, one lobe of liver was homogenized with tissue homogenizer in 700 ml RA1 buffer whereas for BMDMs 106 BMDMs were lysed in 350 ml buffer. The cDNA was generated using 400 ng RNA with oligo (dT18) primer and the RevertAid Reverse Transcriptase (Thermo-Fisher Scientific). Quantitative real-time PCR was performed using GoTaq master mix (Promega, #A600A). qPCR reactions were run on Mastercycler (Eppendorf). B2m and actin genes were used as housekeeping gene controls for liver and BMDMs, respectively. qPCR primers used in this study are listed in .

Seahorse flux analysis

A Seahorse XFe96 Analyzer (Agilent) was used to determine oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) measurements of cultured cells in real-time. The Seahorse XF Cell Mito Stress test kit (Agilent, #103015–100) was used according to the manufacturer’s instructions. BMDMs were infected with L. monocytogenes MOI 10 for 6 hours. After measurement of basal OCR and ECAR, oligomycin (1.5 μM), carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP, 1.5 μM) and rotenone/antimycin A (0.5 μM) were injected subsequently. 8 μM Hoechst 33342 (Invitrogen, #H3570) was injected simultaneously with rotenone/antimycin A from a separate port. For FAO and glutaminolysis inhibition experiments, 4 μM Etomoxir and 3 μM BPTES (Agilent, #103260–100) were injected after FCCP instead of Hoechst. The raw data was analysed using Wave Desktop Software (Agilent, version 2.6.1) and exported and graphed in GraphPad Prism 7 (GraphPad Software).

Staining for flow cytometry analysis

For cell recruitment assays in infected mice, peritoneal exudates were collected by injecting 5 ml PBS into the belly of sacrificed mice twice. After counting, the cells were stained for dead cells using fixable viability dye (eBioscience, #65-0865-14). Then, cells were washed in FC buffer (0.5% BSA in PBS) and blocked using Fcg block at RT for 10 min and stained in FC buffer (0.5% BSA in PBS) at RT for 30 min. Cells were washed with PBS and analyzed by flow cytometry using using BD LSRFortessa. For splenocyte staining, the spleens harvested from mice were passed through the 70 mm cell strainer in 5 ml PBS and exposed to red blood cell lysis buffer (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA, pH 7.3) for 5 min on ice. Then cells were washed with PBS twice and stained with the viability dye and antibodies as above. Antibodies used for flow cytometry are CD3e (BD, #562286), B220 (BD, #553090), Ly6G (BD, #561236), CD11b (BD, #563015), Ly6C (BD, #560594). Flow cytometry was performed using BD LSRFortessa. FlowJo 10.6.1 software was used to analyze the data.

Western blot and its quantification

Infected and gentamicin treated BMDMs were washed with PBS and lysed in Laemmli buffer (2% SDS, 10% glycerol, 62.5 mM Tris, pH 6.8). Protein concentration was determined using BCA assay (ThermoFisher, #23225) and after adding 5% β-mercaptoethanol and bromophenol blue (BPB), the samples were loaded in 10% SDS-PAGE gel.

32.5 mg liver tissue from infected mice was homogenized per ml lysis buffer (10 mM Tris pH7.5, 50 mM NaCl, 50 mM NaF, 2 mM EDTA, 1% Triton X-100, 1 mM DTT, 0.1 mM PMSF, 1 mM Vanadate, cOmplete protease inhibitor cocktail tablet (Roche, 11697498001), 10 mg/ml DNase I). Then the lysates were passed through a 25G syringe (Braun, #9186166) three times, incubated at 4°C on a rotating wheel for 2 hours and spun at 13400 g at 4°C for 20 min. The supernatant was transferred into a new tube and kept at -20°C. Lysates were then mixed with Laemmli buffer for loading on a 10% SDS-PAGE gel.

The samples were loaded twice to detect the phospho- and total forms of the proteins separately. The gels were blotted on nitrocellulose membrane for 16 hours at 200 mA and then 2 hours at 400 mA at 4°C in transfer buffer (3 mM Na2CO3, 10 mM NaHCO3, and 20% ethanol). The membranes were blocked in 5% non-fat dry milk powder in TBST for 1 hour at RT and incubated with primary antibodies detecting total ACC (CST, #3662) or phospho-ACC (CST, #3661) prepared in 5% BSA in 1:1000 dilution for overnight at 4°C while shaking. Next day membranes were washed in TBST three times for 5 min and incubated at RT with rabbit secondary antibody (Jackson Immunoresearch, #111-035-003) in 1:2500 dilution for BMDMs and 1:10000 dilution for liver lysates. For development of ACC and phospho-ACC signals SuperSignal West Pico PLUS (Thermo Scientific, #34580) and ECL Prime Western Blotting Detection Reagent (Amersham, #RPN2236) were used. For detection, the BioRad ChemiDoc imaging system was used.

Quantification of western blots were done with bands in the linear detection range. Serial lysate dilutions used were for BMDMs 1.25 mg, 2.5 mg, 5 mg, 10 mg, 20 mg and for liver tissue lysates 0.3125 ml, 0.625 ml, 1.25 ml, 2.5 ml and 5 ml. Minimum of three of these dilutions were used for the downstream calculations. Slope ratios of phospho-ACC and total ACC were calculated for all genotypes and the knockouts were normalized to wt.

Statistics

All statistical analysis was performed in GraphPad Prism 7. Statistical tests used for the calculation of p values are indicated in the figure legends.

Statistical analysis of liver metabolites

Statistical analysis was performed in R version 3.6.2 [40]. We fitted univariate linear models using function L. monocytogenes for each log10 transformed measured metabolite in turn. Effects of genotype and infection and the interaction between them are of main biological interest in our models. Both were modelled as fixed categorical effects with three levels each; for genotype (wt, Irf9-/- and Ifnar1-/-) and infection (d0, d1, d3), respectively. Dummy coded and centered sex of the mice, dummy coded and centered batch/experiment and z-transformed liver weights were fitted as cofactors to reduce residual variability. Assumptions for linear models were met. Residuals were normally distributed and homoscedastic. We calculated contrasts between least square means (LSM) of genotype and infection levels respectively with package emmeans v1.4.7 [41]. Significance was declared at a multiple testing corrected 10% false discover rate. [42].

We further performed Principal Component Analysis with package factoextra v1.0.7 [43]. We produced biplots using function fviz_pca_biplot, which display PCA scores of samples (shown as dots) and loadings of each metabolite (shown as vectors) simultaneously. Dots that are close to each other represent mice with similar values. The longer a vector of a metabolite the bigger the influence of said metabolite on that principal component. Vectors pointing in similar directions, forming small angles between them can be considered as positively correlated, vectors forming an angle of 90° as uncorrelated and vectors pointing in opposing directions as negatively correlated. We used centered and scaled log10 transformed metabolite measures corrected for batch / experiment effects, liver weight and sex for our PCA. We calculated these ‘customized’ residuals by subtracting estimates for each experiment, sex and liver weight from the same linear model as used for hypothesis testing, from the raw metabolite measures. Numbers of mice used for this screen are for uninfected wt, Irf9 and Ifnar1, 15, 15 and 13; for one-day infected wt, Irf9 and Ifnar1, 15, 15 and 15; for three-day infected wt, Irf9 and Ifnar1, 13, 14 and 14.

(Related to Fig 1).

(A) Cytokine levels of mice uninfected or infected for one-day. (B) Neutrophil recruitment into the peritoneal cavity of PBS-treated mice or one day post-infection. (C) Gating strategy for flow cytometry of PECs in Figs 1 and S1. P values were calculated with two-way ANOVA test.

Fig 1

Irf9-/- and Ifnar1-/- mice differ in the level of protection from L. monocytogenes infection.

(A) Kaplan-Meier plots showing survival of mice after infection with 106 L. monocytogenes for 10 days. P values were calculated using log-rank test and corrected for multiple testing using Bonferroni-Holm correction. (B) Bacterial loads of spleen and liver three days post-infection. (C) CCL2 levels measured from the plasma of mice one day after PBS injection or L. monocytogenes infection using Luminex-based multiplex bead array. The figure shows the pool of two experiments (Number of uninfected mice used wt = 10; Irf9-/- = 10; Ifnar1-/- = 8 and one-day infected mice wt = 9; Irf9-/- = 9; Ifnar1-/- = 8) (D) Flow cytometric analysis of inflammatory monocytes derived from peritoneal lavage of mice one day after PBS injection or L. monocytogenes infection. (E) Alanine aminotransferase (ALT) levels measured in plasma three days post-infection. (F) Percentage of total splenocyte (n = 6 for all genotypes) and (G) PEC (n = 4, 5 and 6, respectively) death gated from single cells of mice one day post-infection, quantified using flow cytometry. Each data point represents one biological replicate. Minimum number of mice used per condition is four. The median values are shown with lines. Whiskers show 5–95 percentile. P values for CFU graphs were calculated using log10 transformed CFU values. For all statistical analysis ANOVA was performed and corrected for multiple testing with Tukey or Dunnett’s post-hoc test. (H) In vitro CFU assay showing log 2 transformed number of viable L. monocytogenes after 1h, 2h or 8h of infection per well. Each data point represents one biological replicate. Bars show the mean values. P values were calculated using two-way ANOVA and corrected for multiple testing with Dunnett’s post-hoc test. n represents the number of mice used for the specified experiment. nd: not detected.

(TIF)

Click here for additional data file.

(Related to Fig 2).

(A) GSEA enrichment plots for OXPHOS and FA metabolism of IFN-I treated wt and Irf9-/- BMDMs. (B) Extracted ion counts (EIC) of intracellular free FAs normalized to tubulin (n = 3 per condition), (C) mRNA expression of Cd36 (n = 3 per condition), (E) OCR and (F) ECAR in BMDMs 6h post-infection or medium treatment. (D) Three replicates of Western blots used to quantify ACC phosphorylation in BMDMs 6h post-infection.

(TIF)

Click here for additional data file.

Mean OCR of four replicates of (A) Etomoxir- and (B) BPTES-treated uninfected BMDMs that are shown in Fig 2. The inhibitors were added at the time point indicated by the arrow. OCR difference of (C) Etomoxir-treated and (D) BPTES-treated BMDMs calculated by subtracting OCR at 73 min (addition of Rotenone/Antimycin) from OCR at 53 min. OCR differences were calculated using OCR values measured at time points indicated with dashed lines. Bars show the mean values. P values were calculated using ANOVA corrected for multiple testing with Dunnett’s post-hoc test.

Fig 2

L. monocytogenes infection and Ifnar1 deficiency lead to glutaminolysis and FAO changes in BMDMs.

(A) Quantification of ACC phosphorylation of BMDMs 6h post-infection. P value was calculated using ANOVA with Geisser-Greenhouse correction. (B) Extracted ion count (EIC) levels of carnitine and stearoyl carnitine normalized to tubulin levels as detected by WB of BMDM lysates 6h post-infection. P value was calculated using ANOVA corrected for multiple testing with Tukey’s post-hoc test. (C) OCR (Oxygen Consumption Rate) and (D) ECAR (Extracellular Acidification Rate) of wt, Irf9-/- and Ifnar1-/- BMDMs uninfected or 6h post- infection. Each data point represents one biological replicate. Bars show the mean values. P values were calculated using two-way ANOVA and corrected for multiple testing with Sidak and Dunnett’s post-hoc tests. Bars show the mean values. Mean OCR of four replicates of (E) Etomoxir- and (F) BPTES-treated L. monocytogenes infected BMDMs. The inhibitors were added at the time point indicated by the arrow. OCR differences (shown with braces) were calculated using OCR values measured at 73 min (addition of Rotenone/Antimycin) and 53 min (indicated with dashed lines).

(TIF)

Click here for additional data file.

(Related to Fig 3).

(A) Dot plots of GSEA results showing first 20 enriched pathways in uninfected (A) Irf9-/- and Ifnar1-/- compared to uninfected wt. Color and size of the dots represent adjusted p values and the number of the genes represented from a certain gene set, respectively. The x-axes were calculated by the sum of the core-enriched genes divided by its set size and ordered decreasingly. The y-axis represents the corresponding gene set name. (B) Venn diagram showing the number of genes enriched in uninfected Irf9-/- and Ifnar1-/- compared to uninfected wt.

(TIF)

Click here for additional data file.

Quantitative RT-PCR of (A) OXPHOS, (B) FA metabolism and (C) IFN-stimulated genes from uninfected and one-day infected mouse livers. P values were calculated using ANOVA corrected for multiple testing with Dunnett’s post-hoc test.

(TIF)

Click here for additional data file.

(Related to Fig 5).

(A-B) Loading plots showing the contribution of each metabolite to clustering with vectors. Length and color of the vectors represent their contribution to separation of different clusters. Impact of each metabolite on a certain cluster separation is reflected in the direction of the vectors. Percentage of total variance is indicated in the axis label. The values are corrected for liver weight, genotype, sex of the mouse and time point effects. ANCOVA test was performed. Extracted ion counts (EIC) of intracellular (C) FAs and (D) ketone bodies in wt, Irf9-/- and Ifnar1-/- mice one- or three-days post-infection or PBS injection. False discovery rate p values were calculated using values corrected for liver weight, genotype, sex of the mouse and time point effects. ANCOVA test was performed. (E) Scheme showing glutaminolysis replenishing the TCA cycle. (F) mRNA expression of glutamate carrier Slc25a22 in uninfected wt, Irf9-/- and Ifnar1-/- livers. P values were calculated using ANOVA with Tukey post-hoc test. (G) Three replicates of Western blots used to quantify ACC phosphorylation in livers one day post-infection. Contrib.: contribution, PC: Principal component, PGA: Phosphoglyceric acid, PEP: Phosphoenolpyruvate, Asp: Aspartic acid, AcCoA: Acetyl CoA, Glu: Glutamate, aKG: alpha-ketoglutarate, BHOB: Hydroxybutyric acid, SA: Stearic acid, AcAct: Acetoacetate, Val: Valine, Leu: Leucine, Ile: Isoleucine, PA: Palmitic acid, OA: oleic acid, Ser: Serine, POA: Palmitoleic acid, aGlc: alpha-D-glucose, bGlc: beta-D-glucose, Pyr: Pyruvate, G6P: Glucose-6-phosphate, GA3P: Glyceraldehyde-3-phosphate.

(TIF)

Click here for additional data file.

GSEA of RNAseq from livers showing enrichment in uninfected Irf9 compared to uninfected wt.

(XLSX)

Click here for additional data file.

GSEA of RNAseq from livers showing enrichment in uninfected Ifnar1 compared to uninfected wt.

(XLSX)

Click here for additional data file.

GSEA of RNAseq from livers showing enrichment in infected wt compared to uninfected wt.

(XLSX)

Click here for additional data file.

GSEA of RNAseq from livers showing enrichment in infected Irf9 compared to infected wt.

(XLSX)

Click here for additional data file.

GSEA of RNAseq from livers showing enrichment in infected Ifnar1 compared to infected wt.

(XLSX)

Click here for additional data file.

List of genes in the Venn diagrams in Fig 3E.

(XLSX)

Click here for additional data file.

List of primers used in this study.

(XLSX)

Click here for additional data file.

28 Jan 2021

Dear Prof. Decker,

Thank you very much for submitting your manuscript "Listeria monocytogenes infection rewires host metabolism with regulatory input from type I interferons" for consideration at PLOS Pathogens. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

Overall, the reviewers concurred that your study focused on an interesting and important question, and were generally positive about the data supporting a role for IRF9 and IFNAR in modulating host metabolism during infection. While a number of specific suggestions were provided by the reviewers that would improve the manuscript, two major points should be substantively addressed in any future resubmission.

First, two reviewers noted that while your data supported the conclusions that IRF9 and IFNAR regulated susceptibility to infection, and also contributed to re-wiring of host metabolism, the relationship between metabolic changes and susceptibility to infection was unclear. The study would greatly benefit from experimental approaches that could test the role of one or more of the altered metabolites in susceptibility of these mice to Listeria infection. Second, validation of prominent hits from the transcriptomics data would enhance the rigor of the study.

If you choose to resubmit a revised manuscript, please also be attentive to comments from Reviewer #1 regarding concerns about the effects of high gentamicin on viability of intracellular bacteria, and from Reviewer #2 regarding data visualization and readability.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Mary O'Riordan

Associate Editor

PLOS Pathogens

Raphael Valdivia

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************

Overall, the reviewers concurred that your study focused on an interesting and important question, and were generally positive about the data supporting a role for IRF9 and IFNAR in modulating host metabolism during infection.

While a number of specific suggestions were provided by the reviewers that would improve the manuscript, two major points should be addressed with additional evidence in any future resubmission.

First, two reviewers noted that while your data supported the conclusions that IRF9 and IFNAR regulated host metabolism and also regulated susceptibility to infection, the causal relationship between metabolic changes and susceptibility to infection was not established. The study would greatly benefit from experiments supporting a role for one or more of the identified metabolites in the susceptibility of these mice to Listeria infection. Second, validation of prominent hits from the transcriptomics data would enhance the rigor of the study.

If you choose to resubmit a revised manuscript, please also be attentive to comments from Reviewer #1 regarding concerns about the effects of high gentamicin on viability of intracellular bacteria, and from Reviewer #2 regarding data visualization and readability.

Reviewer's Responses to Questions

Part I - Summary

Please use this section to discuss strengths/weaknesses of study, novelty/significance, general execution and scholarship.

Reviewer #1: Interferon response to infection and immunometabolism are very important areas of research. In this manuscript Demiroz et al. investigate the role of IFN-I in metabolic processes of murine macrophages and liver upon Listeria monocytogenes infection. Using genetically-modified mice and transcriptomic and metabolomic profiling, they uncover a very interesting specificity of IRF9 and IFNAR in regulating host metabolism and susceptibility to infection.

Reviewer #2: In the enclosed manuscript from Demiroz et al the authors investigate the role of type I interferons in modulating metabolism in the context of Listeria monocytogenes infection. Initially they recapitulate much of what has been reported in the literature with respect to loss of interferon signaling, including decreases in liver damage and decreased lymphocyte apoptosis in IFN signaling deficient animals despite minor effects on bacterial burden. Importantly, one novel observation from this work is that IRF9 mutants are even more protected than IFNAR mutants. The authors then go on to use bone marrow derived macrophages to assess the impact of IFNAR signaling on metabolic pathways. They find that ACC phosphorylation is decreased and less FA is imported into the mitochondria in the absence of IFNAR or IRF9. Consistent with this they observe decreased OXPHOS as measured by seahorse, although the contribution of IRF9 was significantly less than that of IFNAR suggesting IRF9 independent pathways downstream of IFNAR. This is confusing as the protection phenotypes presented in Fig. 1 were more prominent in the IRF9-/- than IFNAR-/-. Next the authors attempted to assess whether similar changes take place in vivo during infection and utilized transcriptomics to assess hepatic responses to Listeria in wt, IRF9 and IFNAR deficient mice. Unfortunately the data presented in figure 3 is incredibly confusing as it appears OXPHOS and FA metabolism are even further upregulated in IFNAR and IRF9 which is inconsistent with the metabolism described in figure 2. The authors do make the point that the differential regulated transcripts of IFNAR and IRF9 do not completely overlap again suggesting differential signaling downstream of IFNAR, however it is unclear to me what the data in figure 3 tells us as currently presented. As the data from figure 3 was inconclusive, the authors directly assessed a subset of metabolites from infected livers from the three mice genotypes. Again the data presentation in figure 4 is difficult to assess and as the authors themselves describe them “these data show that L. monocytogenes infection and IFN-I signaling cause metabolic changes in the liver”, a conclusion that does not get at the differences in susceptibility to infection or even really the role of Listeria infection or type I interferons in regulating metabolism. Finally, the data in figure 5 are largely in agreement with the macrophage data from figure 2 demonstrating decreases in ACC phosphorylation and FA transport into the mitochondria, although again these changes do not correlate with the differences in susceptibility reported in figure 1, notably increased resistance of IRF9-/- mice vs IFNAR-/-. In sum, while this manuscript aims to determine how IFN-I regulates metabolism in the context of Listeria infection to discern if this contributes to the differential susceptibility of mice with IFN signaling deficiencies, there is no conclusive data presented either about how IFN alters metabolism or whether or not this contributes to infection. Furthermore, while the observation that IRF9 mice are more protected from infection than IFNAR mice is interesting, there are no data presented to explain this phenomenon.

Reviewer #3: Demiroz and colleagues have tackled an important and outstanding question in Listeria biology and their work implicates a role for IFN signaling in the control of metabolic responses in both infected macrophages and liver tissue from infected animals. They have used a combination of metabolomics, RNA-seq and real time readouts of oxidative phosphorylation using a seahorse analysis to reveal a difference in the metabolism of Irf9 and IFNAR KO animals. This is particularly important since the reason for reduced bacterial load in IFNAR animals remains elusive. I very much enjoyed the paper and have several suggestions to revise it primarily for improved transparency and clarity.

**********

Part II – Major Issues: Key Experiments Required for Acceptance

Please use this section to detail the key new experiments or modifications of existing experiments that should be absolutely required to validate study conclusions.

Generally, there should be no more than 3 such required experiments or major modifications for a "Major Revision" recommendation. If more than 3 experiments are necessary to validate the study conclusions, then you are encouraged to recommend "Reject".

Reviewer #1: 1-Infection route. The puzzling choice of IP injection is all the more surprising since the authors have previously demonstrated that IFN-I have opposite roles in mice infected by the IP route or by the natural, more relevant oral route. Are the different contributions to infection control of IRF9 and IFNAR also seen in mice orally infected with the LO28InlA* strain?

2-Significance of metabolic effects. WT, Ifr9-/- and Ifnar1-/- mice show rather small metabolic differences. There is no evidence that these differences contribute to infection control. The study would greatly benefit from an in vivo experiment showing the role of some of the identified metabolites in the susceptibility of these mice to infection.

3-Gentamicin protection assays. Use of high concentrations of gentamicin such as 50 ug/mL for 1 hour leads to gentamicin internalization and bactericidal effects demonstrated in several infection models. What is the intracellular survival rate of the LO28 strain after 1 hour of infection and treatment with gentamicin 50 ug/mL compared to treatment with gentamicin 20, 10 or 5 ug/mL? Once the optimal concentration of gentamicin is identified, CFU/well (instead of fluorescence) should be determined at 2, 6 and 24 hours (if cell viability is not affected). A time "zero" control should be added to demonstrate that the capacity of internalization is similar in WT, Ifr9-/- and Ifnar1-/- cells.

Reviewer #2: Major Points:

1) The observation that IRF9 mice are even more resistant than IFNAR mice is interesting and novel but there is no work presented to explain this phenomenon. As the transcriptomics data in figure 3 focus only on potential metabolic differences which appear to not explain the phenotype, significant additional studies are necessary to dissect the reason for this phenotype.

2) Although not a major contributor to the story the authors try to tell, the cell intrinsic differences in macrophage control of Listeria infection presented in figure 1H are inconsistent with the existing literature that has found no cell intrinsic difference in IFNAR macrophage control of Listeria infection (ex. Auerbuch et al 2004). Why the authors see something different is not explained and this is important as these same cells are then used for all the metabolic analysis in figure 2 and it is critical to know if there is something different than other papers in the field.

3) The transcriptomics data are presented in a very confusing way and no validation of any of the data were performed. As presented the data is not useful.

4) Figure 4 suffers the same readability issues that as presented it is difficult to glean meaningful information from the data, as the authors themselves seem to concede in stating simply that there are differences in metabolites. This data is at least followed up with analysis of specific metabolites in figure 5.

5) Most importantly, there is no connection with any of the metabolic data presented in the paper and a causal relationship with the differential susceptibility of IFN signaling deficient mice. In the end the manuscript ends up as a list of data with no appreciation for the biological impact of any of it. The authors themselves state “the rather small metabolic effects observed in our macrophage experiments alone are unlikely to make major contributions to the substantial differences between Irf9-/- and Ifnar1-/- mice in the resistance to L. monocytogenes infection”. In the end, this is the crux of the issue, data was presented that there are metabolic differences in the knockout mice but the phenotypes are small and there is no indication that they are biologically relevant without additional experimentation to manipulate the pathways and assess subsequent Listeria infection outcomes.

Reviewer #3: Overall, the use of varied technology and analyses that go from primary cells in culture to in vivo experiments is impressive. Given the varied metabolic contribution of macrophages/circulating monocytes compared with hepatocytes do the authors hypothesize that the same pathways are at play in both conditions? The final conclusions seem to indicate that they suspect the same pathway but it seems like the RNA-seq data would have incorporated a mix of macrophages and hepatocytes whereas the BMDM data would represent metabolic data solely derived from immune cells. When the authors mined their previous BMDM data from Platanitis E, et al., how much does it overlap with the RNA-seq from this analysis? Do they find the same proteins up or down?

Along these lines could they infect animals and separate out immune cells from hepatocytes and analyze levels of p-ACC or GABA to access whether the immune cells and hepatocytes have the same response? They could also add infection of primary hepatocytes or a hepatocyte cell line to sort this out. I’m not sure if there is a way to assess GABA in a less costly experiment than metabolomics, but the overarching question is how general is the response that they describe (same in all infected cells) or could it be different in the macrophages versus hepatocytes? If the authors expect both immune cells and hepatocytes to act in the same manner, this should just be more fully discussed throughout the text and highlighted on the figures (where the source material comes from, while the figure legends and text indicate the source material, it would be better to see it on the figures as well). Adoptive transfer experiments or tissue specific deletion (hepatocyte versus immune CRE) could address this issue elegantly but could be reserved for future work.

The authors spend a fair amount of time in the introduction explaining the role of various metabolic pathways in activated macrophages. Are BMDMs the best model to study this phenomenon or would resident or peritoneal macrophages be a better representation of the monocytes that they are interested in in vivo? How would polarization or pretreatment to activate BMDMs affect or change the phenotype?

The authors should upload their RNA-seq data to a searchable database or upload the log transformed data in the form of an excel table that highlights the individual proteins detected and their values. They have done this with the enrichment analyses but unless I missed it the complete RNA-seq analysis is not part of the supplementary data.

The authors did not show the raw Western blot data for phospho ACC levels. At least a representative blot should be shown, but it would be preferable to see all three in supplementary data in line with increased transparency.

**********

Part III – Minor Issues: Editorial and Data Presentation Modifications

Please use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity.

Reviewer #1: 1-At day 3, injection of 10e6 LO28 led to death of some WT and Ifnar1-/- mice, and thus heterogeneity. It could be worth measuring bacterial survival upon infection with a sublethal inoculum.

2-Many phenotypes are investigated in WT, Ifr9-/- and Ifnar1-/- backgrounds with or without infection. A summary of what is IRF9-dependent, IFNAR-dependent, IRF9- and IFNAR-dependent and IRF9- and IFNAR-independent would be helpful in the discussion. For instance, figure 6 could be completed with an Ifr9-/- cell, liver data and uninfected data.

3-Line 159 It is not clear why the Wang reference is introduced here.

4-Line 396 The mice genotype should be indicated.

5-Fig S1C is not introduced.

6-Some references should be completed (journal, page numbers etc.).

7-Western blots should be shown (Figures 2 and 5)

8-It would help to indicate clearly on figures what is related to uninfected or infected data.

9-The unacceptable quality of the main figures, probably due to file conversion, should be improved. Figures 3 and 4 are partially illegible...

Reviewer #2: (No Response)

Reviewer #3: Discussion points/Figure revision

In their previous work the authors often compared IRF3/7 KO animals to IFNAR. It is beyond the scope of this paper to cross compare those cells/animals but I wonder if they would discuss what they would expect? Would cytosolic DNA/RNA signaling that is IRF9-independent potentially be responsible for this phenotype observed in the IFNAR KO? Along these lines, when they cross compare IFNAR and IRF9 KO mice they have overlapping genes in Venn diagrams but I was not able to find the list of these genes. Are they in a table that I missed?

For the PCA analysis, I think the figure should be revised so one can see the general distribution of the phenotypes and the individual differences in pathways that they wish to highlight could be displayed as bar graphs below. As it stands, it is very difficult to make out the various conditions due to the labeling.

The authors have included diagrams of the metabolic pathways that they discuss in the supplementary figures but it may be worthwhile to incorporate such diagrams into the figures so that readers can cross compare

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at .

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here on PLOS Biology: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see

28 Apr 2021

Submitted filename: Point-by-point reply to PP reviewers.pdf

Click here for additional data file.

7 Jun 2021

Dear Prof. Decker,

We are pleased to inform you that your manuscript 'Listeria monocytogenes infection rewires host metabolism with regulatory input from type I interferons' has been provisionally accepted for publication in PLOS Pathogens.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Mary O'Riordan

Associate Editor

PLOS Pathogens

Raphael Valdivia

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************************************************

Reviewer Comments (if any, and for reference):

Reviewer's Responses to Questions

Part I - Summary

Please use this section to discuss strengths/weaknesses of study, novelty/significance, general execution and scholarship.

Reviewer #2: In the revised manuscript by Demiroz et al the authors aimed to address concerns about the lack of mechanistic understanding behind their interesting observations that IRF9 deficient mice are protected from listeriosis, even more so that the protection previously described for IFNAR deficient mice. While previous rounds of review noted the significant amount of work presented in the manuscript, there were concerns that the data neither clearly indicated how IFN manipulated metabolism nor how this putative manipulated metabolism contributed to the resistance phenotype of the knockout mice. In the revised manuscript, the authors have added data including important validation of transcriptomic data by qPCR for example. In addition the investigators attempted to directly test the hypothesis that regulation of FAO is a mechanism by which IFN contributes to control of Listeria infection by treating mice with etomoxir, however the interpretation of this data is difficult. The investigators see a minor decrease in burden in WT mice but not in IRF9 or IFNAR mice with etomoxir treatment which they interpret as a consistent with their model that IFN regulation of FAO contributes to control of Listeria burden. This data is overinterpreted as all they have shown is that FAO inhibition contributes to infection control but there is no data to conclude that this is the mechanism observed in IFNAR or IRF9 knockout mice. Additionally, the investigators ignore the increase in burden observed in the IFNAR knockout mice that would not be consistent with this model? Furthermore, although they have tried to clarify some of the PCA analysis in the end the metabolomics are incongruent with the transcriptomics. The data are the data so this is fine, however the problem is that none of the newly added data does anything to clarify how IFN regulates metabolism to control infection or why IRF9 mutants would behave differently than IFNAR mutants. As noted by the authors themselves both the metabolic changes and many of the transcriptional changes are very small in magnitude and perhaps this is the reason for significant lack of overlap between the IRF9 and IFNAR data that is surprising and unexpected. I appreciated the authors discussion of why IRF9 might be different than IFNAR, however not following up on these ideas ultimately leads to a paper that is simply an interesting observation that remains unexplained.

Reviewer #3: This paper maps the role of IFNAR and IRF9 on metabolic reprogramming by Listeria monocytogenes and will be a valuable contribution to an interesting and up and coming field.

**********

Part II – Major Issues: Key Experiments Required for Acceptance

Please use this section to detail the key new experiments or modifications of existing experiments that should be absolutely required to validate study conclusions.

Generally, there should be no more than 3 such required experiments or major modifications for a "Major Revision" recommendation. If more than 3 experiments are necessary to validate the study conclusions, then you are encouraged to recommend "Reject".

Reviewer #2: (No Response)

Reviewer #3: I am satisfied with the revised manuscript. The reviewers addressed my concerns and I think the revision is ready for publication.

**********

Part III – Minor Issues: Editorial and Data Presentation Modifications

Please use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity.

Reviewer #2: (No Response)

Reviewer #3: The authors use IFN-I as a plural noun throughout, perhaps it would be better to specifically say Type I interferon. I thought it is more grammatically appropriate to use Type I IFN in the singular; e.g.

Line 102 “IFN-I are integral” – Type I IFN is integral.

I think the acronym is making the grammar complicated.

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

24 Jun 2021

Dear Prof. Decker,

We are delighted to inform you that your manuscript, "Listeria monocytogenes infection rewires host metabolism with regulatory input from type I interferons," has been formally accepted for publication in PLOS Pathogens.

We have now passed your article onto the PLOS Production Department who will complete the rest of the pre-publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Pearls, Reviews, Opinions, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript, if you opted to have an early version of your article, will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064