Loss of stearoyl-CoA desaturase 2 disrupts inflammatory response in macrophages

ABSTRACT Macrophages are innate immune cells that patrol tissues and are the first responders to detect infection. They orchestrate the host immune response in eliminating invading pathogens and the subsequent transition from inflammation to tissue repair. Macrophage dysfunction contributes to age-related pathologies, including low-grade inflammation in advanced age that is termed “inflammaging.” Our laboratory has previously identified that macrophage expression of a fatty acid desaturase, stearoyl-CoA desaturase 2 (SCD2), declines with age. Herein, we delineate the precise cellular effects of SCD2 deficiency in murine macrophages. We found that deletion of Scd2 from macrophages dysregulated basal and bacterial lipopolysaccharide (LPS)-stimulated transcription of numerous inflammation-associated genes. Specifically, deletion of Scd2 from macrophages decreased basal and LPS-induced expression of Il1b transcript that corresponded to decreased production of precursor IL1B protein and release of mature IL1B. Furthermore, we identified disruptions in autophagy and depletion of unsaturated cardiolipins in SCD2-deficient macrophages. To assess the functional relevance of SCD2 in the macrophage response to infection, we challenged SCD2-deficient macrophages with uropathogenic Escherichia coli and found that there was impaired clearance of intracellular bacteria. This increased burden of intracellular bacteria was accompanied by increased release of pro-inflammatory cytokines IL6 and TNF but decreased IL1B. Taken together, these results indicate that macrophage expression of Scd2 is necessary for maintaining the macrophage response to inflammatory stimuli. This link between fatty acid metabolism and fundamental macrophage effector functions may potentially be relevant to diverse age-related pathologies. IMPORTANCE Macrophages are immune cells that respond to infection, but their dysfunction is implicated in many age-related diseases. Recent evidence showed that macrophage expression of a fatty acid enzyme, stearoyl-CoA desaturase 2, declines in aged organisms. In this work, we characterize the effects when stearoyl-CoA desaturase 2 is deficient in macrophages. We identify aspects of the macrophage inflammatory response to infection that may be affected when expression of a key fatty acid enzyme is decreased, and these findings may provide cellular insight into how macrophages contribute to age-related diseases.

inflammation that eliminates invading pathogens to tissue repair and wound healing (1). In addition to their critical role in the host response to infection, macrophages have been shown to contribute to inflammation that occurs with aging (2) and may have widespread contributions to cardiovascular disease, pathologies of the genitourinary system (3), eye diseases (4), and cancers (5).
In our previous studies, we have demonstrated that macrophage dysfunction in aging is coordinated in part by perturbations in lipid homeostasis (6)(7)(8). Specifically, we have identified age-associated downregulation of stearoyl-CoA desaturase 2 (SCD2) (7), an enzyme that converts the long-chain fatty acids 16:0 and 18:0 to the monounsaturated fatty acids 16:1 and 18:1, respectively. Several recent studies have shown that SCD fatty acid desaturation regulates macrophage-mediated inflammation (9,10). Macrophage activation leads to altered abundance of monounsaturated fatty acids including those produced by SCD2, and deletion of SCD1/2 prolonged upregulation of cytokines elicited by inflammatory stimuli (9). In the demyelinating disease multiple sclerosis, microglia (i.e., resident macrophages of the central nervous system) upregulate SCD1 in response to accumulation of phagocytosed myelin, which ultimately increased inflammation preventing resolution of lesions (10). Our laboratory has shown that macrophage-spe cific deletion of Scd2 increases pathologic angiogenesis in the eye in a mouse model of the blinding disease age-related macular degeneration (7).
However, a cell-intrinsic, functional role for SCD2 in macrophages during the inflammatory response to infection has not been explored. In this work, we investiga ted the role of Scd2 in macrophages by studying Scd2-deficient macrophages deleted via lysozyme M-Cre recombinase. We show that SCD2-deficient macrophages exhibit dysregulation of several inflammatory genes at baseline and in response to lipopolysac charide (LPS) stimulation. Though SCD2 was not required for general NFkB activation, it was necessary for IL1B activation following LPS stimulation. Specifically, we identified that there was a deficit in signal 1 that is responsible for induction of Il1b mRNA that is translated to precursor IL1B, whereas SCD2 was not necessary for signal 2 that cleaves pro-IL1B and releases it. Furthermore, we found that there were disruptions to autoph agy as well as dramatically decreased amounts of unsaturated cardiolipins (CL). Finally, we tested the relevance of these findings to the macrophage response to microbes by challenging SCD2-deficient macrophages with a uropathogenic strain of Escherichia coli (UPEC). Loss of SCD2 was associated with impaired clearance of intracellular UPEC as well as a concomitant increase in the release of IL6 and TNF but decreased IL1B release. Therefore, we identify a novel role for SCD2-mediated fatty acid desaturation in macrophage function during infection response.

SCD2 deficiency dysregulates inflammatory gene expression in macrophages
In order to first confirm Scd2 deletion in macrophages, we prepared bone marrowderived macrophages (BMDMs) from wild-type mice and from mice with conditional deletion of Scd2 in myeloid cells driven by the lysozyme M-Cre recombinase (Scd2 -m/-m ). We found that in macrophages prepared from Scd2 -m/-m mice, Scd2 transcript was ~20% that of macrophages prepared from wild-type mice (Fig. 1A). We also measured expression of Scd1 and Scd4, the other two SCD isoforms with detectable expression in macrophages. There was no compensatory upregulation by either Scd1 or Scd4 in Scd2 -m/-m macrophages, though Scd4 was downregulated by 40% (Fig. 1A). It has been previously reported that Scd2 is expressed at a much higher level than Scd4 in BMDMs with the latter not being detected in RNA sequencing data (9). Together, these data indicate that Scd2 deletion from macrophages does not lead to compensatory upregula tion of other SCD isoforms at the mRNA level.
To broadly assess the transcriptional changes that occur upon deletion of Scd2, we performed RNA sequencing. We obtained an average of 35.4 million reads per samples (n = 3 wild type, n = 3 Scd2 -m/-m ). Of these, 35.2 million reads (99.6%) were mapped to the mouse genome. There were 63 differentially expressed genes with false discovery rate (FDR) <0.05 and |log 2 FC| >2 (Fig. 1B). To determine whether or not these transcriptional changes could be indicative of specific cellular pathways, we performed pathway analysis using Gene Set Enrichment Analysis using the hallmark gene sets database (11,12). There were 22 pathways with FDR below the recommended cutoff of 0.25 (Fig. 1C). Ten pathways were upregulated in Scd2 -m/-m macrophages and the remaining 12 were downregulated. There was mixed dysregulation of pro-inflammatory and anti-inflammatory pathways, including transforming growth factor beta (TGFB) signaling, interferon alpha response, inflammatory response, TNFA signaling via NFkB, IL6-JAK-STAT3 signal ing, allograft rejection, IL2-STAT5 signaling, interferon gamma response, and comple ment (Fig. 1C). Notably, pathway analysis also identified fatty acid metabolism as a perturbed pathway in SCD2-deficient macrophages (Fig. 1C), consistent with SCD2's role as a fatty acid desaturase.
We then compared mRNA expression of 20 macrophage activation markers. In naïve macrophages, deletion of Scd2 caused dysregulation of 14 of the 20 genes assessed (Fig. 1D). Of these, eight were upregulated (Ccl2, Fasl, Il6, Ptgs2, Tnf, Arg1, Fgf2, and Vcam1), and six were downregulated (Fas, Il1b, Mmp9, Nos2, Socs3, and Vegfa). Some pro-inflammatory genes were upregulated (e.g., Il6 and Tnf), whereas others were downregulated (e.g., Il1b). Similarly, some anti-inflammatory markers were upregulated (e.g., Arg1), whereas others were downregulated (e.g., Vegfa). To assess macrophage response to inflammatory stimuli, we also compared the fold change of these 20 genes after activation of macrophages with the Gram-negative bacterial membrane component LPS (100 ng/mL) for 24 h. All genes assessed were responsive to LPS with the same direction of change in both wild-type and Scd2 -m/-m macrophages (Fig. 1E). However, there were six genes with blunted responses to LPS (Il1b, Socs3, Il12a, Tgfb1, Il10, and Vcam1) and five genes with an exaggerated response to LPS (Nos2, Il6, Ccl2, Socs1, and Fas). Taken together, these gene expression data suggest that the deletion of Scd2 from macrophages perturbs inflammatory gene expression at baseline and in response to inflammatory stimuli.

SCD2 is dispensable for LPS-induced NFkB phosphorylation and nuclear translocation
Because inflammatory stimuli like LPS are known to activate macrophages via NFkB signaling, we assessed for LPS-elicited phosphorylation and nuclear translocation of the NFkB subunit P65. After LPS treatment, there was substantial translocation of P65 from the cytoplasm into the nucleus that peaked at 1 h ( Fig. 2A), along with a concomitant and slight increase in P65 phosphorylation (Fig. 2B). To quantify these changes, we performed western blot of whole macrophage lysates. Total cellular P65 abundance was stable up to 24 h after LPS treatment ( Fig. 2C and D). P65 phosphorylation peaked at 1 h after LPS stimulation and decreased thereafter and was similar between wild-type and SCD2-deficient macrophages ( Fig. 2E and F). These data indicate that SCD2 is dispensa ble for NFkB activation.

SCD2 deletion from macrophages impairs IL1B activation
It is known that inflammation is regulated by autophagy (13,14). We found there was general dysregulation of pro-inflammatory markers Il6, Tnf, and Il1b in Scd2 -m/-m macro phages (Fig. 1D). As such, we tested whether there were any changes in autophagic clearance. In transmission electron micrographs, we noted that there appeared to be and Hprt and then expressed as the log 2 of the fold change relative to untreated cells. Each bar indicates the mean ± SEM of n = 7-8 wells from two independent experiments. Statistical significance was assessed using the Mann-Whitney test. Statistical significance is indicated with asterisks as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Research Article mBio
increased accumulation of autophagosomes as identified by cytoplasmic doublemembraned vesicles (Fig. 3A). Next, we stained for the autophagy substrate marker, P62, and, consistent with electron microscopic evaluation, there was increased P62 To quantify these changes, we immunoblotted for P62 ( Fig. 3C and D). There was a statistically significant increase in P62 abundance in Scd2 -m/-m macrophages (Fig. 3D). Therefore, loss of SCD2 appears to disrupt autophagic flux in macrophages via P62. Altered autophagic flux can lead to decreased IL1B signaling by affecting P62 degradation (15)(16)(17); thus, we next examined whether IL1B secretion was affected in Scd2 -m/-m macrophages. IL1B is a pro-inflammatory cytokine whose release by macro phages is activated by two sequential signals. After signal 1, Il1b is transcribed and translated into pro-IL1B protein (31 kDa). After signal 2, the precursor IL1B protein is cleaved into mature IL1B protein (17 kDa) by caspase 1 and then released nonspecifically through gasdermin pores during pyroptosis. LPS activates the first signal via NFkB. Though NFkB nuclear translocation and phosphorylation were unaffected in SCD2deficient macrophages, Il1b mRNA was downregulated at baseline ( Fig. 1B and D) and its activation was blunted after LPS stimulation (Fig. 1E). Therefore, we assessed IL1B protein by western blotting and found that in SCD2-deficient macrophages, there was appropri ate induction of pro-IL1B up until 6 h after LPS stimulation ( Fig. 4A and B). However, at the later 24-h timepoint, there was significantly less pro-IL1B protein ( Fig. 4A and B).
It is possible that decreased amounts of pro-IL1B are due to decreased translation or increased cleavage and/or release because of inappropriate constitutive signal 2 activity in SCD2-deficient macrophages. There was very little, if any, mature IL1B detected (Fig.  4A), suggesting that pro-IL1B is not constitutively cleaved. Nonetheless, we assessed for pro-caspase 1 and active caspase 1 (48 and 10 kDa, respectively), the latter of which cleaves pro-IL1B to mature IL1B. Consistent with lack of mature IL1B, there was very little, if any, active caspase 1 under the conditions tested (Fig. 4C). Baseline levels of procaspase 1 were similar in both wild-type and SCD2-deficient macrophages, and LPS treatment for 24 h induced a similar 1.7-fold increase in pro-caspase 1 for both geno types ( Fig. 4C and D). Finally, we assessed for precursor and cleaved gasdermin D (53 and 30 kDa, respectively), the latter of which forms pores on the plasma membrane that ultimately leads to pyroptosis and IL1B release. There was very little, if any, cleaved gasdermin D detected (Fig. 4E). Baseline levels of precursor gasdermin D were similar in both wild-type and SCD2-deficient macrophages, and LPS treatment induced a similar 1.6-to 1.7-fold increase in precursor gasdermin D for both genotypes ( Fig. 4E and F).
To assess whether deficient induction of pro-IL1B leads to decreased release of mature IL1B to the extracellular space, we first activated signal 1 with LPS (100 ng/mL) followed by activation of signal 2 with ATP (2.5 mM). We performed this experiment with short (2 h) or long (12 h) LPS stimulation, both followed by the same 1 h of ATP stimula tion ( Fig. 4G and H). We collected media and measured released IL1B. Consistent with immunoblotting experiments, there was no difference in IL1B release from Scd2 -m/-m macrophages that were stimulated acutely with LPS for 2 h (Fig. 4G). However, when Scd2 -m/-m macrophages were stimulated with LPS for longer (12 h), there was a statistically significant decrease in the amount of IL1B released by SCD2-deficient macrophages (Fig.  4H). Therefore, deletion of SCD2 from macrophages impairs activation and release of IL1B, especially in the later phase of inflammatory activation. The delay in IL1B reduction after autophagy disruption may represent the time lag necessary for Il1b transcription and translation to protein and its cleavage, cytoplasmic transport, and extracellular release.

Loss of macrophage SCD2 depletes unsaturated cardiolipins
Because SCD2 is a fatty acid desaturase, we tested whether its deletion affected its substrates or products or other lipid species. Using liquid chromatography-mass spectrometry, we measured 11 classes of lipids: phosphatidylcholine (PC), phosphatidyl serine (PS), phosphatidylinositol (PI), phosphatidylglycerol (PG), phosphatidylethanola mine (PE), phosphatidic acid (PA), free fatty acids (FFA), CL, carnitine and acylcarnitines (AC), acyl CoAs (ACoA), and triacylglycerols (TAG). Of the 139 lipid species assessed, 56 had statistically significant changes between wild-type and Scd2 -m/-m macrophages with fold changes ranging from 0.14× to 2.96× (Fig. 5). For 7 of the 11 lipid classes that we assessed, there were mild changes with statistically significant differences in 1/2 or less of the lipid species detected within each class (Fig. 5A through K). For example, among the On the other hand, PS, PA, CL, and TAG had statistically significant differences in >1/2 of the individual species within each class. Individual species of PS, PA, and TAG with statistically significant changes were all increased in abundance, and all of these species contained monounsaturated or polyunsaturated acyl chains (Fig. 5B, F and K). On the other hand, all five unsaturated cardiolipins detected were decreased 3.3-to 7.1-fold compared to wild-type macrophages and were the only lipid species with more than threefold change out of the 139 lipid species assessed (Fig.  5H). Notably, cardiolipins have previously been shown to play a direct role in the priming and activation of the nucleotide-binding oligomerization domain-like receptors family pyrin domain containing 3 (NLRP3) inflammasome responsible for IL1B release (18,19).

SCD2 deficiency perturbs macrophage response to uropathogenic E. coli
We have found that in macrophages, SCD2 deletion leads to dysfunction in inflammatory gene expression and response to LPS. In order to test the relevance of these impairments during the response to pathogens, we infected macrophages in vitro with the UTI89 strain of uropathogenic E. coli. We first assessed for intracellular bacterial burden in macrophages at 2, 8, and 24 h post-infection. We found that there was no difference in intracellular bacterial burden at 2 h post-infection, indicating that uptake or entry of bacteria into macrophages is not affected by SCD2 deficiency (Fig. 6A). However, at 8 and 24 h post-infection, there was significant increase in intracellular bacteria, indicating that when Scd2 is deleted from macrophages, there is impaired clearance of intracellular E. coli (Fig. 6A).
Because there was dysregulation of inflammatory gene expression response to LPS ( Fig. 1D and E), we measured IL1B, IL6, and TNF released by Scd2 -m/-m macrophages challenged with UTI89. Consistent with impaired IL1B activation after LPS stimulation (Fig. 4A, B, G and H), Scd2 -m/-m macrophages challenged with UTI89 released significantly less IL1B to the extracellular space (Fig. 6B). However, there was more IL6 and TNF released ( Fig. 6C and D). Taken together, these data indicate that loss of Scd2 from macrophages perturbs their response to UPEC infection.

DISCUSSION
In this study, we delineate the cellular effects when Scd2 is deleted from murine macro phages. First, we found that there were perturbations to inflammatory gene expression and response, providing independent validation of previous reports that SCD-deficient macrophages exhibit dysregulated inflammatory gene expression (7,9,10). We also found that although SCD2 was dispensable for general NFkB activation, loss of SCD2 impaired the activation of IL1B in response to inflammatory stimuli. We specifically identified that SCD2 is uniquely required for signal 1 activation of Il1b transcription and translation into precursor IL1B protein. As such, when SCD2 is absent, activation and release of IL1B are blunted upon macrophage stimulation with LPS or challenge with uropathogenic E. coli. Furthermore, we found that macrophage SCD2 deficiency led to disruption of autophagy with accumulation of intracellular autophagosomes. This finding is not surprising given previous reports that SCD deficiency in pancreatic β cells impairs autophagosome fusion with lysosomes (20). In macrophages, autophagic flux and inflammation are inversely related; thus, disruptions to autophagic clearance in Scd2m/-m macrophages could lead to downstream activation of inflammatory gene expression such as those we found for Il6, Tnf, and Ccl2. . Each bar indicates the mean ± SEM for n = 3 wells/group. Statistical significance was assessed using two-way ANOVA with Šídák's multiple comparisons test. Statistical significance is indicated with asterisks as follows: ***, P < 0.001; ****, P < 0.0001.

Research Article mBio
When we assessed for changes in lipids in Scd2 -m/-m macrophages, we found that there were no changes in either of SCD2's two substrates, FFA(16:0) and FFA(18:0). In addition, the products of SCD2 catalysis, FFA(16:1) and FFA(18:1), were either unchanged or mildly increased 1.4-fold, respectively. Though we found there was no transcriptional compen sation by other Scd isoforms, there could still be post-transcriptional changes that compensate for loss of Scd2, which is the most highly expressed isoform in murine macrophages (9). An alternative hypothesis is that when biosynthesis of monounsatura ted fatty acids stalls due to loss of SCD2, utilization of monounsaturated FFA by some downstream pathways is restricted. For example, we found that loss of SCD2 dramatically depletes unsaturated CL, which are primarily localized to the mitochondrial inner membrane. This de-prioritization of CL synthesis may enable maintenance of steadystate pools of unsaturated FFA along with phospholipids that are major components of Research Article mBio the plasma membrane whose composition and organization critically regulate macro phage function (21). Of pertinence to the deficient priming of IL1B in Scd2 -m/-m macro phages, previous studies have established a direct role for CL in the activation of the NLRP3 inflammasome that activates and releases IL1B (18,19). Our data suggest that CL, particularly those containing unsaturated fatty acids, may also play important role in the production of pro-IL1B following LPS priming. These findings provide a link between fatty acid desaturation by Scd2 and funda mental macrophage effector functions. We previously reported that macrophage Scd2 expression declines in aging and that this could promote pathologic ocular angiogenesis (7). However, age-related loss of macrophage SCD2 could be of potential significance to other pathologies of aging. For example, urinary tract infections (UTIs) are common bacterial infections that affect more than 50% of women in their lifetimes and are most often caused by UPEC. The burden of this disease is exacerbated in elderly women with over 10% reporting having a UTI within the past year (22) such that this infection is the leading cause of bacteremia in the elderly (23). This is further highlighted by UTI recurrence that cannot be cured by antibiotic therapy, which is seen in over 50% of elderly patients (24). In fact, women aged 55-64 suffer from the highest prevalence of recurrent UTIs (i.e., ≥3 UTIs/year) (25). Although reduced estrogen (26) and incontinence contribute to increased susceptibility, there may be other risk factors involved such as age-related dysfunction in immune response (3,27). In this study, we tested the relevance of macrophage SCD2, whose expression declines in aging, in macrophage response to the uropathogenic E. coli strain UTI89. SCD2-deficient macrophages were less able to clear internalized uropathogenic E. coli, and there was impaired IL1B release but increased IL6 and TNF release. Therefore, SCD-mediated fatty acid desaturation could be a key molecular program whose downregulation in aging could perturb macrophage effector functions such as those essential for the immune response to UPEC (28), thus contributing to increased risk for recurrent UTIs in elderly women. If found to be relevant to human disease, SCD-mediated fatty acid desaturation could be a key metabolic step that maintains macrophage inflammatory responses important for many diverse pathologies of aging.

Mice
We obtained mice with floxed Scd2 alleles (Scd2 f/f ) (29,30) from Hide Tsukamoto (Keck School of Medicine of the University of Southern California). We crossed these mice with mice carrying the lysozyme M-Cre transgene (31) to generate mice with myelo monocyte-specific deletion of Scd2 (Scd2 -m/-m ). We used male and female mice aged

Bone marrow-derived macrophages
To isolate bone marrow cells, we sacrificed mice by CO 2 euthanasia, dissected out both tibias and femurs, and passed 8 mL of ice-cold Dulbecco's Modified Eagle Medium (DMEM) through each bone's medullary cavity. We filtered the bone marrow aspirates through a 40-µm nylon mesh strainer, pelleted the cells, and re-suspended the pel let in macrophage differentiation media consisting of DMEM with 10% fetal bovine serum (FBS), 10% media conditioned in CMG cells (32), 1% penicillin/streptomycin, and 1% L-glutamine. We replenished differentiation media 3 days after initial plating and re-plated macrophages for experimentation 5-6 days after initial plating. After differentiation, we used mature macrophage media consisting of DMEM with 10% FBS, 2% CMG-conditioned media, 1% penicillin/streptomycin, and 1% L-glutamine.

RNA isolation
We washed cells twice with phosphate buffered saline (PBS), then used TRIzol (Thermo Fisher Scientific, Waltham, Massachusetts, USA) and RNeasy Plus Mini Kit (Qiagen, Hilden, Germany) to extract total RNA following the manufacturer's recommended protocol.

RNA sequencing and analysis
We quantified the quality and quantity of the RNA samples with an Agilent Bioanalyzer or 4200 Tapestation. All RNA samples (n = 6) were of high quality with RNA integrity numbers ≥9.3. We enriched our samples for mRNA by performing polyA selection according to the manufacturer's recommended protocol. Then, we prepared sequencing libraries with standard protocols and sequenced the samples on an Illumina NovaSeq 6000 at the McDonnell Genome Institute at Washington University School of Medicine in St. Louis. RNA sequencing reads were aligned to the mouse genome (Ensembl release 76 primary assembly) with Spliced Transcripts Alignment to a reference (33). Finally, we performed standard analysis of gene-level features using EdgeR and Limma (34,35). To identify dysregulation of pathways, we performed Gene Set Enrichment Analysis (11,12).

Quantitative polymerase chain reaction (qPCR)
We reverse transcribed total RNA to cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Waltham, Massachusetts, USA). Target gene expression was normalized to geometric mean of the following housekeeping genes: 18S, Actb, Gapdh, and Hprt. We calculated gene expression differences using the ΔΔCt method.

Enzyme-linked immunosorbent assay
To measure secreted cytokines, we collected media from BMDMs and measured mouse IL6, IL1B, or TNF using Quantikine Colorimetric Sandwich ELISA Kits (R&D Systems, Minneapolis, Minnesota, USA) following the manufacturer's instructions.

Immunocytochemistry
We plated cells in chamber slides. After incubation with any cell culture reagents, we washed cells with pre-warmed PBS and fixed cells in 4% paraformaldehyde or 10% neutral-buffered formalin for 15 min at room temperature. Then, we blocked and permeabilized cells with PBS containing 5% serum from the secondary antibody host and 0.3% Triton X-100. We incubated cells with primary antibody overnight at 4°C. We performed secondary antibody staining for 1 h at room temperature. The following primary antibodies were used: anti-P65 (8242 from Cell Signaling Technology, Danvers, Massachusetts, USA), anti-phosphorylated P65 (Cell Signaling Technology 3033), and anti-P62 (GP62-C from Progen, Heidelberg, Germany). We imaged cells using a Zeiss LSM800 confocal microscope.

Immunoblotting
We loaded cell lysate containing ~15 µg protein into individual lanes, separated proteins using SDS-polyacrylamide gel electrophoresis, and then transferred proteins to a nitrocellulose membrane (0.22 µm pore size). We blocked membranes with 5% nonfat dry milk or bovine serum albumin (BSA) in PBS. We incubated membranes with primary antibodies overnight at 4°C, followed by secondary antibodies for 1 h at room temper ature. We used the following primary antibodies: anti-P65 (Cell Signaling Technology 8242), anti-phosphorylated P65 (Cell Signaling Technology 3033), anti-IL1B (ab9722 from Abcam, Cambridge, United Kingdom), anti-caspase 1 (Cell Signaling Technology 83383), anti-gasdermin D (Cell Signaling Technology 39754), anti-P62 (Progen GP62-C), and anti-beta actin (A5316 from Sigma, St. Louis, Missouri, USA). We visualized blots using the dual-channel Odyssey CLx Imaging System and quantified protein bands of interest using Image Studio. We normalized signal for proteins of interest against β-actin as a loading control.

Transmission electron microscopy
We fixed macrophages with 2% paraformaldehyde and 2.5% glutaraldehyde in 100 mM sodium cacodylate buffer (pH 7.2) for 1 h at room temperature. We washed samples with sodium cacodylate buffer and performed a post-fix in 1% osmium tetroxide for 1 h. We then rinsed samples with distilled water prior to en bloc staining with 1% aqueous uranyl acetate for 1 h. Following several rinses in distilled water, we dehydrated samples in a graded series of ethanol and embedded samples in Eponate 12 resin. We cut 95 nm sections with a Leica Ultracut UCT ultramicrotome, stained them with uranyl acetate and lead citrate, and imaged them on a JEOL 1200 EX transmission electron microscope equipped with an AMT 8.0 megapixel digital camera and AMT Image Capture Engine V602 software.

Lipidomics
We assessed by liquid chromatography-mass spectrometry total cellular abundances of PC, PA, PE, PS, PI, PG, CL, TAG, FFA, carnitine, AC, and ACoA. We precipitated protein to extract PC, PA, PE, PS, PI, PA, PG, carnitine, AC, and ACoA. We extracted CL, FFA, and TAG using the Blyth-Dyer method. We added internal standards to samples prior to extrac tion. These were PC We analyzed PC, PE, PS, PI, PG, TAG, carnitine, and AC with a Shimadzu 20A HPLC system coupled to an API4000 mass spectrometer operated in positive multiple reaction monitoring (MRM) mode. We analyzed PA and CL in positive and negative MRM modes, respectively, on 4000QTRAP mass spectrometer coupled to a Shimadzu 20A HPLC system. We analyzed ACoA with a Shimadzu 20A HPLC system coupled to a 6500QTRAP + mass spectrometer operated in positive MRM mode. We processed data using Analyst 1.6.3.
To monitor instrument performance, we created a quality control sample by pooling aliquots of the study samples. This quality control sample was injected between every 10 samples. We report only the lipid species with coefficient variance less than 15% in the quality control sample. We calculated the ratio of analytes to their corresponding internal standards and reported relative quantifications comparing ratios in wild-type samples to those in Scd2 -m/-m samples.
To assess for statistical significance comparing wild-type (n = 3) and Scd2 -m/-m (n = 3) relative quantifications, we performed hypothesis testing using bootstrapping. Briefly, we constructed a null population (n = 99,999) of the differences between two sample means (n = 3 each). Both samples were created by random sampling with replacement from the pooled data. Finally, we calculated a two-sided P-value based on the percentile of the actual difference compared to the null population.

Infection of BMDMs with UTI89
We infected BMDMs with the uropathogenic E. coli strain UTI89 for 2 h and measured intracellular CFU at 2, 8, and 24 h after the start of infection (15). We grew UTI89 statically in Luria-Bertani broth for 17 h at 37°C. We challenged BMDMs with UTI89 at an MOI of 1.0 for 2 h. After the 2-h infection period, we removed UTI89-containing media, washed BMDMs with PBS, and applied gentamicin to kill extracellular bacteria. We applied gentamicin at 100 µg/mL for either 15 min (for the 2-h timepoint) or 1 h (for the 8and 24-h timepoints); after which, we switched to a lower 10 µg/mL gentamicin dose.
After gentamicin treatment, we prepared cell lysates and then plated serial dilutions for CFU analysis.

Statistics
We describe the statistical analyses used for bulk RNA sequencing and lipidomics in prior sections. For all other data, we performed statistical analyses using GraphPad Prism 9. We first assessed the normality of our data graphically and by using a Kolmo gorov-Smirnov test. When comparing a single variable between two different groups, we used two-tailed t tests or two-tailed Mann-Whitney tests. For comparisons of a single variable among >2 groups, we used one-way or two-way analysis of variance with Šídák's multiple comparisons test. A P-value <0.05 was considered statistically significant.