Cholesterol sensing and metabolic adaptation in tissue immunity

Abstract

Cholesterol metabolites, particularly oxidized forms known as oxysterols, play crucial roles in modulating immune and metabolic processes across various tissues. Concentrations of local cholesterol and its metabolites influence tissue-specific immune responses by shaping the metabolic and spatial organization of immune cells in barrier organs like the small intestine (SI) and lungs. We explore recent molecular and cellular evidence supporting the metabolic adaptation of innate and adaptive immune cells in the SI and lung, driven by cholesterol and cholesterol metabolites. Further research should unravel the detailed molecular mechanisms and spatiotemporal adaptations involving cholesterol metabolites in distinct mucosal tissues in homeostasis or infections. We posit that pharmacological interventions targeting the generation or sensing of cholesterol metabolites might be leveraged to enhance long-term immune protection in mucosal tissues or prevent autoinflammatory states.

Cholesterol metabolites as modulators of mammalian immune-metabolic processes in tissues.

Proper architecture and function of barrier organs depend on the correct positioning of cells within the tissue. While non-immune cells are generally seeded in distinct organs during embryogenesis according to their germ layer origin[1], immune cells require discrete steps to acquire the ability for optimal tissue colonization, either after their development or in secondary lymphoid organs (SLOs). Once in the tissue, immune cells are influenced by local cues that tune their effector function and maintenance[2–4]. Cholesterol and cholesterol metabolites have been recognized as central regulators of various aspects of immune responses both in homeostasis and inflammation[5]. In particular, oxidized cholesterol-metabolites (i.e. oxysterols) are versatile molecules that engage distinct intracellular and extracellular pathways based on the position of the hydroxyl group on the four hydrocarbon rings[5, 6].

We postulate that local cholesterol and cholesterol metabolites modulate tissue immune responses by metabolically imprinting immune cells and organizing microanatomical niches that underpin effector functions. These two regulatory modes are likely intertwined, with anatomical zonation providing lipid cues that are sensed by recruited immunocytes.

Metabolic adaptation in the small intestine

The small intestine (SI) is a primary tissue used to study the interaction of sterol metabolites with immune cells because of the central role of intestinal epithelial cells (IECs) in lipid absorption[7]. In mice and humans, IECs express Niemann-Pick C1-Like 1 protein, which mediates the initial step of dietary cholesterol uptake (see Box 1) [8], as well as Aster proteins, which ensure non-vesicular cholesterol movement in enterocytes[9]. Lipids absorbed by IECs are incorporated into chylomicrons to ensure lipid delivery into the lymphatics, and eventually into circulation[10].

Box 1. Molecular mechanisms of cholesterol uptake and signaling.

Major cholesterol uptake systems

Cellular cholesterol concentrations are regulated by a balance of biosynthesis and uptake of exogenous cholesterol [74]. Dietary cholesterol is one exogenous source, and absorption from the intestinal lumen is mediated by apical expression of NPC1L1, the target of ezetimibe, on enterocytes [74]. After entering the enterocyte, cholesterol is esterified by ACAT2 and then packaged into nascent chylomicrons that enter circulation through the lymphatics [74]. Most circulating cholesterol exists as LDL, which is derived from VLDL and ILDL after the removal of triglycerides by lipoprotein lipases [74]. When a cell’s cholesterol requirement exceeds its biosynthetic capacity, activation of SREBP2 results in increased expression of the LDL receptor (LDLR), which drives LDL uptake through clathrin-mediated endocytosis [74]. After endocytic uptake, cholesterol is transported into the cytoplasm via sequential action of the lysosomal transporters NPC1 and NPC2 [74]. Cholesterol can also circulate in the bloodstream in HDL, which dominantly targets the liver for cholesterol export after uptake via SR-B1 in hepatocytes [74]. Macrophages also express SR-B1, though it is less clear whether this promotes HDL uptake or cholesterol efflux [74].

Molecular targets of cholesterol that influence cellular signaling

Direct cholesterol-dependent signaling involves detection by sterol sensing domain (SSD)-containing proteins such as SCAP and HMGCR [74]. Binding of cholesterol to SCAP triggers a conformational change that increases association of the SCAP/SREBP2 complex with INSIG1 in the ER, preventing translocation to the Golgi apparatus [74]. Binding of cholesterol to HMGCR induces ER-associated degradation (ERAD), which slows down the rate of cholesterol biosynthesis [16]. The net effect of cholesterol detection by SSDs is to reduce cholesterol uptake/synthesis in states of sterol sufficiency [16]. Of note, inhibition of cholesterol biosynthetic flux via deletion of SCAP results in the spontaneous production of type I interferon from macrophages, suggesting other mechanisms whereby decreased cholesterol concentrations are sensed by immune cells [75].

Cholesterol can also function directly as a negative regulator of TCR signaling [76]. Cryo-EM studies have shown that cholesterol can directly bind to the resting TCR and keep the TCR-CD3 complex in an inactive conformation [76].

Finally, cholesterol can influence cellular signaling indirectly by biophysical modulation of plasma membrane dynamics [77]. Studies have suggested that plasma membrane cholesterol concentrations can influence the signaling threshold of TLRs [77].

Downstream signaling mechanisms that are differentially mediated by cholesterol precursors

The cholesterol biosynthesis pathway is a complex biochemical process that begins with acetyl-CoA and utilizes 22 enzymes to ultimately generate cholesterol [5]. There are many intermediate metabolites within this pathway that influence cellular signaling independent from their roles as cholesterol precursors [78]. The clearest examples are isoprenoids, such as farnesyl-pyrophosphate (FPP) and geranylgeranyl-pyrophosphate (GGPP), which can act as post-translational modifications on proteins that contain a CAAX or CAAL motif in a process called prenylation [78]. Many key signaling molecules, such as Ras, require prenylation for proper membrane targeting and function [78]. While specific immunological roles are emerging, an interesting example of this is the requirement of prenylation for Rho GTPase signaling. Certain bacteria utilize effector toxins to inhibit Rho GTPase prenylation, and loss of GTPase activity is subsequently sensed by the inflammasome protein MEFV [79]; gain-of-function mutations in this protein are responsible for the autoinflammatory disorder Familial Mediterranean Fever (FMF)[80,81]. Also, another genetic autoinflammatory disorder that involves loss of Rho GTPase prenylation is Mevalonate kinase Deficiency, which results in loss of isoprenoid biosynthesis [81].

Cholesterol intermediates can also signal via the nuclear hormone receptors LXR and RORγt, although the precise metabolites that directly bind these receptors remain unclear [82] LXR ligand binding, potentially via desmosterol or 24,25-epoxycholesterol, results in transcriptional activation of the cholesterol transporters ABCA1/ABCG1, which promote cholesterol efflux to HDL [83]. RORγt is the major transcription factor that controls IL-17 production by lymphocytes (Th17, ILC3, and γδ T cells) [84]. While it is difficult to uncouple the roles for cholesterol in driving cellular proliferation versus cytokine production, experiments using reporter cell lines have suggested that cholesterol intermediates may play a role in promoting RORγt activity on the IL17A promoter [85].

We envision at least two non-mutually exclusive strategies that intestinal immune cells could employ to adapt to the lipid-rich environment of the gut. First, immune cells could undergo imprinting in gut-draining SLOs to upregulate metabolic pathways that integrate intestinal lipid cues. The tuning of intracellular metabolism before reaching a tissue destination is conceptually similar to the acquisition of gut homing receptors by lymphocytes -- a process that endows lymphocytes with dedicated integrins and chemokine receptors in an anticipatory fashion, thus enabling their migration from the blood into the intestinal tissue[11, 12]. Second, immune cells in the gut could rapidly integrate fluctuating lipid cues using dedicated sensors to calibrate immune responses to the current lipid environment without pre-existing transcriptional or metabolic imprinting preparing them for tissue environmental cues. In this model, the delineated process would have the advantage of creating adaptation as needed, thus dynamically responding to short-lived alterations of nutrients and commensals.

Distal metabolic imprinting for fate determination and tissue function

One possible fate of CD8⁺ T cells responding to primary infection is to persist as tissue-resident memory cells (TRMs): in contrast to effector memory or central memory cells, TRMs reside in tissues across the body to ensure rapid and long-term protection at the site of re-infection[13].

P14 T cell receptor transgenic mouse CD8⁺ T cells, which are specific for the lymphocytic choriomeningitis virus (LCMV), can be used to track TRM generation and persistency in different organs upon LCMV infection. A combination of CRISPR–Cas9-mediated loss-of-function screening in vivo, metabolomics, and transcriptional analysis of P14 TRMs recently showed that TRMs in the SI have a particularly active mevalonate–cholesterol synthesis pathway upon acute (LCMV Armstrong) infection[14]. This metabolic branch is driven by the Sterol Response Element-Binding Protein 2 (SREBP2) -- a protein that acts both as sensor for cholesterol/cholesterol metabolites and as transcription factor (TF) for the enzymes in the mevalonate–cholesterol synthesis pathway (Box1) [15].

Intracellular concentrations of cholesterol and cholesterol metabolites control SREBP2 activity. Specifically, when cholesterol and oxysterols are abundant, SREBP2 remains in the ER by directly binding to the multi-transmembrane SREBP cleavage-activating protein (SCAP), which is retained in the ER by the ER-resident insulin-induced gene (INSIG). When the intracellular sterol concentration is reduced, a SCAP conformational change allows detachment of SCAP from INSIG, and the SCAP/SREBP2 complex moves to the Golgi where SREBP2 is cleaved. The now activated SREBP2 translocates into the nucleus where it acts as a TF[16].

Deletion of the SREBP2 pathway in CD8⁺ T cells, either using CRISPR-Cas9 or by removing SCAP in T cells (Cd4^Cre Scap^fl/fl mice) globally impaired the acquisition of the TRM phenotype across tissues upon acute LCMV infection, likely due to defects in the T cell effector phase[17]. However, gut TRMs exhibited the most stringent requirements for SREBP2 in establishing metabolic programming, because reduced numbers of SI TRMs were more pronounced compared to TRM numbers in other organs when the SREBP2 pathway was altered. SREBP2-dependent TRM adaptation to the SI environment required the mevalonate–cholesterol synthesis pathway: reduction in cholesterol synthesis, either via a negative feedback mechanism using a high cholesterol diet, statin administration, or deletion of Fdft1 (the gene encoding squalene synthase), specifically reduced SI TRMs compared to TRMs in other organs. Alternative products of the mevalonate–cholesterol synthesis pathway were also essential for SI TRMs, as shown via the targeting of Pdss2 and Hpd, two enzymes involved in the production of important nonsterol isoprenoids such as coenzyme Q (CoQ) [17]. CRIPSR-Cas9 deletion of Hpd specifically impaired SI TRM formation upon acute LCMV infection while P14 TRM in the spleen, liver, or kidney were unaffected. Similarly, Pdss2 overexpression, upregulating CoQ synthesis, restored SI TRMs in mice fed a high-cholesterol diet upon acute LCMV infection, suggesting that non-steroidal products can offset cholesterol inhibitory feedback on cholesterol synthesis to give rise to SI TRMs [17]. Together, these data suggest that while redirecting the mevalonate–cholesterol synthesis pathway toward non-steroidal metabolites benefits TRMs in the spleen, kidney, and liver, SI TRMs exhibit differential metabolic regulation, strongly relying on CoQ-producing enzymes for their adaptations.

How discrete SREBP2 activity shapes SI TRM generation and maintenance remains to be established. Microenvironments control tissue-specific chromatin accessibility and underpin site- and context-specific TRM regulation[18]. Activated SREBP2 functions as a TF, but aside from its ability to bind SRE sequences in cholesterol-regulating genes, little is known about its transcriptional activity on non-metabolic genes. It is conceivable that SREBP2 in TRMs located in different tissues might interface with largely different chromatin landscapes, thus mediating functionally distinct outcomes. Moreover, because metabolic control of cells is thought to be analog rather than digital, and discrete amounts of SREBP2 can impact lymphocyte fates[19], it is conceivable that signals modulating SREBP2 kinetics and/or persistence might direct functionally-distinct TRM roles. SREBP2 is upregulated in CD4⁺ and CD8⁺ T cells in response to TCR stimulation[17]: thus, the strength of the signal[20], mediated by antigen abundance, specialized antigen-presenting cells[21], and/or intestinal cues that modulate intracellular TCR signaling[22], might underpin specialized tissue TRM generation in SLOs. In tissues, TRMs are not thought to be maintained by engagement of antigen receptors; therefore, SREBP2 regulation might hinge upon environmental signals that tune SREBP2 activity to generate SI TRM. Sterols[23], hypoxia[24], mTORC activation[25] and pH concentrations [26] can impact SREBP2 activation, nuclear translocation and transcriptional activity; also, these are likely different across organs, making them potential candidates in SREBP2-specific processes regulating the maintenance of SI TRMs.

Local response to cholesterol metabolites for mucosal immunity

IgA-producing plasma cells (PCs) are terminally differentiated B cells that home to the gut and secrete IgA against commensals and enteric pathogens[27]. IgA⁺ PCs, in contrast to IgM and IgG PCs, are mainly generated in Peyer’s Patches (PPs)[28]; however, distinct origin niches do not endow these PCs with particularly different transcriptional profiles, other than homing receptors that are required for IgA PC migration into the lamina propria[29, 30].

In the lamina propria, IgA⁺ PCs use the migratory receptor GPR183 to relocalize to sub-anatomical intestinal niches[31]. GPR183 is the only surface oxysterol described so far: a G_αi-coupled GPCR that recognizes the oxysterol 7α,25-HC, and to a lesser extent, 7α,27-HC in mice and humans[32–34]. 7α,25-HC is synthesized from cholesterol through the action of two distinct enzymes: CH25H, that places a hydroxyl group in position 25 of the cholesterol to generate 25-HC; and CYP7B1, which places a hydroxyl group in position 7α of the 25-HC, generating 7α,25-HC[6]. IECs can synthesize 7α,25-HC upon dietary cholesterol uptake and commensal recognition via MyD88 signaling [31]. High amounts of GPR183 ligand, driven by high cholesterol diet, prompts IgA⁺ PCs to relocalize close to intestinal lymphatic capillaries, and reduce the amount of IgA secretion in the mouse intestine. Conversely, reduced GPR183-dependent migration, achieved via dietary intervention (with a cholesterol-free diet), cholesterol uptake inhibition (via ezetimibe treatment), or genetic deletion of GPR183 ligand expression (VIll^Cre Ch25h^fl/fl mice) or sensing (Aicda^Cre Gpr183^fl/fl), allowed IgA⁺ PC relocalization closer to the intestinal epithelial layer (measured via microscopy analysis), and an increased rate of IgA secretion (ELISPOT, ELISA) [31]. The GPR183-IgA secretion axis is dynamic, given that IgA⁺ PCs can move between sub-anatomical compartments cyclically in response to lipid cues, suggesting that antibody responses in the gut can integrate external cues and modulate the magnitude of secretion [31].

GPR183 also controls the tissue positioning of group 3 innate lymphoid cells (ILC3); indeed, mice with ILC3-specific GPR183 deletion (Rorc^Cre Gpr183^fl/fl) failed to form colonic lymphoid tissue [35], and single-nucleotide polymorphisms in GPR183 have been associated with increased risk of inflammatory bowel disease (i.e. ulcerative colitis and Crohn’s disease) in humans[36]. Thus, GPR183 recognition of sterol metabolites is likely to be part of the overarching mechanism that builds intestinal immunity in mice and humans.

Integrating distal and proximal lipid cues to tune tissue immune populations

We speculate that metabolic changes and migratory response effector functions could still be intertwined as part of a global adaptation to the local lipid niche: maintenance of SI TRM could be underpinned by specialized anatomical zonation that places cells in dedicated niches tuning SREBP2 activity. Similarly, PCs modulate their effector function in response to pro-migratory lipid cues[31], a process that might take place concurrently with the sensing of intermediate cholesterol metabolites present in the sub-anatomical intestinal niche. Notably, SREBP2 has also been shown to play a role in generating PCs during germinal center (GC) reactions in PPs; specifically, deletion of SREBP2 in GC B cells (Aicda^Cre Srebf2^fl/fl mice) [19, 37] led to reduced PC output, possibly initiating a metabolic adaptation required for PC function in the gut.

Moreover, whether the adaptation processes described for TRM and IgA⁺ PCs are conserved in other SI cells remains to be determined. The heterogeneity of intestinal cells includes lymphocytes -- that acquire their effector function in SLOs[38, 39] or in the thymus[4] -- as well as innate immune cells such as dendritic cells and macrophages that differentiate from circulating precursors[40] or are generated from local precursors[41]. Furthermore, certain immune cells reach the intestinal lamina propria after birth[42], and they might respond to lipid cues when influenced by both diet and commensal composition. In contrast, other immune cells such as lymphoid tissue inducer cells, home to the mouse gut in utero during embryogenesis[43], and might be more susceptible to lipid cues of maternal origin. Thus, we argue that metabolic adaptation for these cells is likely to depend on distinct spatiotemporal processes involving different lipid sensors, metabolic pathways, and dedicated transcriptional modules.

Metabolic adaptation in the lung

In contrast to the intestine, the lungs are not a primary site of lipid uptake, and correspondingly, there is much less understood about how immune cells are programmed via the local metabolic milieu in this organ. Nevertheless, there are unique features of respiratory tract biology that are centered around metabolic imprinting by lipids. The alveolar epithelium is coated with a layer of surfactant (90% phospholipid, 10% protein) that prevents airway collapse via surface tension reduction[44]. Alveolar macrophages play a vital role in controlling surfactant homeostasis via direct phagocytosis and crosstalk with type II alveolar epithelial cells that produce surfactant[45]; deficiency in this key cell type, observed in Csf2rb^−/− and Csf2^−/− mice or in humans with autoantibodies against GM-CSF, causes pathological surfactant accumulation[46]. Alveolar macrophage development depends on the cytokine GM-CSF which maintains expression of the nuclear hormone receptor PPARγ[47]. PPARγ is a sensor of polyunsaturated fatty acids and, upon ligand engagement, drives the expression of genes that promote fatty acid oxidation, thereby converting excess lipids into ATP[48]. Therefore, adaptation to the lung environment involves local cytokine cues inducing the expression of a metabolite sensor that allows macrophages to handle the constant phagocytosis of high lipid amounts that are unique to this organ[46].

While alveolar macrophages are chronically adapted to surfactant due to their exceptionally long half-life (>8 months in mice[49]), recruited cells that enter the tissue during inflammation can be acutely reprogrammed by surfactant exposure[50]. During murine infection with Nippostrongylus brasiliensis, surfactant protein A (SP-A) synergizes with IL-4 to drive enhanced proliferation and alternative activation of recruited monocyte-derived macrophages; indeed, Sftpa1^−/− mice show increased parasite burden compared to wildtype, decreased BrdU uptake by macrophages, and reduced macrophage alternative activation markers (RELMα, YM-1)[50].

Alveolar macrophages are the only described TRM population that expresses Ch25h at homeostasis in mice[51]. Ch25h is one of the core genes induced by GM-CSF as part of adaptation to the alveolar compartment; specifically, time-course transcriptional analysis of macrophages transferred into the lungs of Csf2rb^−/− mice showed early upregulation of Ch25h mRNA[52]. To our knowledge, no studies so far have reported a role for CH25H in regulating the alveolar macrophage response to surfactant. An intriguing possibility is that CH25H might be required for macrophages to handle surfactant-derived cholesterol. Surfactant-derived cholesterol could be converted by CH25H to act as a liver X receptor (LXR) ligand to promote cholesterol efflux to high density lipoprotein and/or as a SREBP2 repressor to prevent cholesterol overload. However, this remains to be investigated. While CH25H may play a role in the regulation of homeostatic surfactant metabolism, another likely function is priming the respiratory tract for defense against infection, as outlined below.

Barrier protection against inhaled pathogens

Ch25h expression was initially reported to be interferon-inducible in macrophages because LPS-stimulated Ifnar1^−/− bone marrow-derived macrophages lose upregulation of Ch25h mRNA[53]. Subsequent experiments overexpressing Ch25h cDNA into 293T cells showed that Ch25h, and its product 25-HC, can block infection by vesicular stomatitis virus, demonstrating that synthesis of the oxysterol has antiviral properties. However, it is unclear to what extent this is due to depletion of membrane cholesterol via SREBP2 repression or a result of 25-HC accumulation that directly alters the biophysical properties of the plasma membrane[53–56]. Recent work showed that 25-HC can block SARS-CoV-2 entry into A549 cells by preventing viral fusion; this correlated with depletion of membrane cholesterol in the target cell based on staining intensity with bacterial anthrolysin O as a readout of accessible plasma membrane cholesterol [57, 58]. However, other studies demonstrated that 25-HC enantiomers showed the same antiviral activity in mice, suggesting that the mechanism of action did not require protein binding, such as to INSIG1[59]. It has been difficult to identify in vivo scenarios where Ch25h/25-HC has unique antiviral functionality because interferon-stimulated gene (ISG) redundancy is likely to compensate for lack of Ch25h via other ISGs. Indeed, in contrast to in vitro experiments, CH25H-deficient mice (Ch25h^−/−) exhibited no difference in weight loss or lung viral titers following intranasal SARS-CoV-2 infection, despite high induction of Ch25h in the lungs of SARS-CoV-2-infected mice, or high amounts of 25-HC in the serum of human patients with COVID-19[59, 60].

25-HC, in addition to its antiviral properties, is protective against bacterial cholesterol-dependent cytolysins (CDCs), such as perfringolysin O[61, 62]. Treatment of neutrophils and macrophages in vitro with type I interferon resulted in resistance to CDC-dependent death in a manner that required Ch25h expression, given that the resistance was lost in Ch25h^−/− cells[61]. Of note, these protective effects were phenocopied by addback of 25-HC to macrophages. Subsequent experiments showed that the mechanism involved inhibition of cholesterol biosynthesis given that Scap^−/− and statin-treated macrophages were protected against CDC-dependent cell death, and increased activity of acylcoenzyme A:cholesterol acyltransferase (ACAT); indeed, ACAT inhibitors reversed the protective effect of interferons on macrophage death in response to CDCs. These data suggest a model whereby 25-HC inhibition of SREBP2 and activation of ACAT results in local depletion of ER cholesterol via esterification into lipid droplets[63]. The local ER cholesterol depletion causes a redistribution of the accessible pool of plasma membrane cholesterol to the ER, making it unavailable for CDC binding[61, 62]. Consistent with the idea that the accessible pool of plasma membrane cholesterol is rate-limiting for CDC binding, treatment of macrophages with sphingomyelinase, which results in an increased accessible cholesterol pool[64], reversed the protective effect of 25-HC on macrophages against CDCs.

Synthesizing the above studies, it is likely that alveolar macrophages tonically secrete 25-HC to prevent damage to the respiratory epithelium by inhaled infectious agents[65]. The common mechanism appears to be plasma membrane cholesterol depletion, which prevents viral membrane fusion and impairs the ability of CDCs to bind host cells[57, 61, 62]. Fungi are another common respiratory infectious agent, with estimates that we inhale upwards of 1 billion different fungal spores per day, and little is known about the role(s) of oxysterols in protection against this class of microorganism. It is possible that during fungal infection, 25-HC may play a similar role as during bacterial or viral infections whereby its main target is the host. Alternatively, because fungi are also eukaryotes, 25-HC might bind to fungal homologues of INSIG1 to repress cholesterol biosynthesis in the pathogen, although this remains speculative. Consistent with this notion, homologues of SREBP are essential for virulence in animal models across a variety of fungal pathogens[66].

Resolution of inflammation

Ch25h is also highly inducible in the airway during inflammatory challenges. Mass spectrometry analyses have shown that 25-HC is the most highly induced oxysterol in the murine airway during LPS challenge and that this induction is lost in Ifnar1^−/− mice, demonstrating a requirement for type I interferons[67]. CH25H-deficient mice (Ch25h^−/−) show a prolonged increase in airway neutrophils after intranasal LPS injection relative to wildtype control mice[68]. This phenotype was also observed in LXR-deficient mice, suggesting that 25-HC can act via LXR to promote the expression of efferocytosis genes such as Axl and Mertk; this is of interest because the true endogenous ligands of LXR remain elusive[68]. Of note, the original reports on 25-HC as an LXR ligand showed marginal activity in reporter assays using luciferase expression driven from an LXR-responsive promoter[69]. Recent experiments utilizing doxycycline-inducible Ch25h expression followed by time-course transcriptional analysis showed that 25-HC production rapidly repressed SREBP2 target genes (Hmgcs1, Hmgcr, Lss) while having little effect on LXR target genes (Abca1, Abcg1); this suggested that 25-HC was not a bona fide LXR ligand[23]. However, while 25-HC might not be an LXR ligand, it is possible that in cell types that express unique enzymes, 25-HC might be further modified into an LXR ligand, which remains to be assessed. Because of the overlapping lung phenotype observed with CH25H- and LXR-deficient mice (e.g., Cyp7b1^−/−), this hypothesis would be particularly interesting to test in the context of alveolar macrophages.

Immune cell recruitment and positioning

Thus far, there is little known about how oxysterols function to organize microanatomical sites within the respiratory tract. However, there is emerging literature suggesting that the 7α,25-HC/GPR183 axis can recruit cells from the blood into the lung[70, 71]. Experiments performed using flow cytometry on GPR183-deficient (Gpr183^−/−) mice and mice treated with a small molecule GPR183 antagonist indicated that this receptor is required for macrophage recruitment to the lungs during influenza virus and SARS-CoV-2 challenge[71]. During Mycobacterium tuberculosis infection in mice, GPR183 promotes eosinophil entry into the lung Gpr183^−/− mice show a defect in lung eosinophil accumulation after intranasal challenge[70]. While deletion of Ch25h also results in a decrease in lung eosinophil entry, the source of the chemotactic lipid remains unclear[70]. In SLOs, laser capture microdissection and RNA hybridization experiments show that Ch25h and Cyp7b1 are expressed by stromal cells[33, 34]. To our knowledge, similar studies have not been performed in the lungs, and it would be important to dissect the relative contribution of stromal versus myeloid expression of Ch25h for generating chemotactic oxysterol gradients. During inflammatory scenarios, Ch25h is highly inducible in the myeloid compartment in mice[68, 72, 73]. This is relevant because induction of this enzyme may result in de novo macrophage-derived gradients of oxysterols that guide recruited cells into emergent niches, or macrophage-derived oxysterols function to quantitatively enhance the recruitment of GPR183⁺ cells from the blood into the lungs. Distinguishing between these possibilities is important because we have a poor understanding of how immune cells position themselves within the lungs after extravasation from the blood to coordinate responses against infection or other exogenous challenges.

Concluding remarks

Several key aspects of tissue adaptation in intestinal immune cells, including both inter and intracellular lipid processes, remain to be defined, and will have both basic and translational implications (see Outstanding questions). Numerous studies have been aimed at dissecting the metabolic requirements for tissue residency, and how sterol metabolism drives cellular localization within discrete intestinal niches. In the lung, there is a rich opportunity to understand how oxysterols may play multi-factorial roles in maintaining barrier integrity, resolving inflammation, recruiting circulating cells, and positioning cells locally. By delineating which factors imprint requirements for sterol metabolism in tissues and identifying the precise molecular targets of oxysterols that control specific biological outcomes, it might be possible to manipulate the phenotypes of barrier tissue immune cells therapeutically during chronic diseases or infection.

Outstanding Questions.

How are cholesterol metabolites transported inside cells?

Oxysterols are generated by enzymes residing in either the endoplasmic reticulum (ER) or the mitochondria, while their sensors can be found in the ER, cytoplasm, or nucleus. The distinct locations of ligands and receptors inside cells suggest that active systems must exist to transport cholesterol metabolites across compartments, but little is known about the nature, cell- and tissue-specificity, as well as the regulation of the molecules governing intracellular cholesterol metabolite function

Can targeting metabolic branches enhance tissue-specific anti-tumor responses?

CAR-T cells need to reach specific tissues to eradicate organ-specific tumors. Metabolic reprogramming of CAR-T cells might ensure a long-term, specific anti-tumor response by increasing their migration and/or maintenance in the tissue.

Are lipids involved in metabolic immune adaptation in the large intestine?

TRM and IgA⁺ plasma cells (PCs) exist in the small intestine and large intestine. However, the cholesterol/oxysterol/SREBP2 axis appears to be active only in the small intestine. Short-chain fatty acids (SCFAs), which are produced by anaerobic intestinal microbiota in the colon during dietary fiber fermentation, are possible candidates for metabolic adaptation in the colon.

Does local immune cell metabolism in the gut influence lipid absorption and/or systemic lipid metabolism?

Induction of nutrient transporters of carbohydrate digestion in the gut are controlled by lymphocytes in the intestine; thus, a similar crosstalk between epithelial cells and immune cells in the regulation of lipid absorption might exist.

Tonic production of 25-HC occurs in the lungs and plays a role in host protection against inhaled insults. Is there a common mechanism of action, or does cholesterol reprogramming have different functions against unique classes of microorganisms (virus, bacteria, fungi, parasites)?

Ch25h expression increases in the lung during inflammation in macrophages, although there is homeostatic expression in the stroma. Does induction of this enzyme generate de novo microanatomical niches that organize immune cells against infection?

Key Figure, Figure 1. Coordination of barrier immunity by the sterol microenvironment.

(A) The alveolar space is filled with lipid-rich surfactant that is processed by alveolar macrophages (AMs). Constitutive expression of 25-hydroxycholesterol (25-HC) by AMs may drive three key functions: 1) regulation of macrophage surfactant uptake/metabolism, 2) the formation of chemotactic oxysterol gradients that recruit circulating immune cells into the lung, and 3) protection of epithelial cells against viral infection and bacterial cholesterol-dependent cytolysins via downregulation of host cell plasma membrane cholesterol concentrations. (B) The intestine is the primary site of cholesterol uptake. In settings of high dietary cholesterol, there is increased synthesis of 7α,25-hydroxycholesterol (7α,25-HC) from the lacteal, which re-positions plasma cells away from the epithelium and reduces IgA secretion. Tissue-resident memory (TRM) lymphocytes in the intestine have high SREBP2 activity, and deletion of this pathway reduces TRM numbers. The SREBP2 requirement may be due to differential TCR signal strength upon antigen encounter in the lymph nodes or a result of local tissue cues that drive SREBP2 expression (see main text for references).

Highlights.

• Upon entry into mammalian barrier tissues such as the intestine and the lungs, immune cells are metabolically reprogrammed by the local milieu.

• Increased cholesterol metabolism is a feature of tissue resident lymphocytes in the intestinal lamina propria.

• Chemotactic oxysterol gradients position immune cells within microanatomical niches in mucosal tissues, and additionally drive immune cell recruitment into the respiratory tract.

• Recent work provides increasing evidence of the importance of cholesterol metabolism in regulating homeostatic immune function and barrier defense during infection.

Significance.

Immune cell function is highly regulated by cell-autonomous cholesterol metabolism; in turn, secretion of cholesterol-derived metabolites by stromal and immune cells is crucially involved in controlling mucosal immunity via cell positioning and metabolic reprogramming. An in-depth understanding of the complex crosstalk between sterols and immune cells might facilitate the development of targeted candidate therapeutics to treat both chronic and infectious diseases.