Schistosome and intestinal helminth modulation of macrophage immunometabolism

Helminths modulate both whole body, and host macrophage metabolism and polarization. Metabolic changes underlie reprogramming of macrophage inflammatory responses.

Summary

Macrophages are fundamental to sustain physiological equilibrium and to regulate the pathogenesis of parasitic and metabolic processes. The functional heterogeneity and immune responses of macrophages are shaped by cellular metabolism in response to the host’s intrinsic factors, environmental cues and other stimuli during disease. Parasite infections induce a complex cascade of cytokines and metabolites that profoundly remodel the metabolic status of macrophages. In particular, helminths polarize macrophages to an M2 state and induce a metabolic shift towards reliance on oxidative phosphorylation, lipid oxidation and amino acid metabolism. Accumulating data indicate that helminth‐induced activation and metabolic reprogramming of macrophages underlie improvement in overall whole‐body metabolism, denoted by improved insulin sensitivity, body mass in response to high‐fat diet and atherogenic index in mammals. This review aims to highlight the metabolic changes that occur in human and murine‐derived macrophages in response to helminth infections and helminth products, with particular interest in schistosomiasis and soil‐transmitted helminths.

Abbreviations

AKG

α‐ketoglutarate

Arg1

arginase 1

BMDM

bone‐marrow‐derived macrophages

C/EBPβ

CCAAT/enhancer binding protein β

CREB

cAMP response element‐binding protein

FAO

fatty acid oxidation

IL‐4

interleukin‐4

ILC2

innate lymphoid type 2 cells

M2

alternatively activated macrophage

mTORC

mammalian target of rapamycin complex

PPARγ

peroxisome proliferator‐activated receptor γ

SEA

schistosome egg antigens

STAT‐6

signal transducer and activator of transcription 6

STH

soil‐transmitted helminths

TCA

tricarboxylic acid

Th2

T helper type 2

TLR

Toll‐like receptor

TNF‐α

tumor necrosis factor α

Introduction: Perspectives of monocytes and macrophage heterogeneity

Monocytes are part of the mononuclear phagocytic system and are implicated in tissue homeostasis and several infectious and inflammatory processes. ¹ Divergent functional capacities of monocytes are partly explained by the heterogeneity of the cellular subpopulations as evidenced by fate mapping, ² unsupervised multi‐dimensional analysis, ³ transcriptional profiling analysis ⁴ , ⁵ and comprehensive proteomics. ⁶ In humans, monocytes were initially categorized according to their expression of CD14 (glycosylphosphatidylinositol‐anchored receptor) and CD16 (FcγRIII) antigens ⁷ in ‘classical’ (CD14⁺⁺ CD16⁻) and ‘non‐classical’ (CD14⁺ CD16⁺⁺) monocytes. ⁸ Classical monocytes exhibit a more phagocytic and pro‐inflammatory phenotype, whereas non‐classical monocytes are involved in patrolling and viral recognition. ⁹ Subsequent analyses further revealed an ‘intermediate’ subset (CD14⁺ CD16⁺). ¹⁰ , ¹¹ ‘Intermediate’ monocytes are characterized as 6‐sulfo LacNAc‐negative, and express multiple surface expression markers that distinguish them from ‘non‐classical’ monocytes, which has been a topic of extensive revision. ¹² , ¹³ , ¹⁴ Notably, each of the human subsets is not entirely homogeneous and presents rich intercellular variation and heterogeneity. ¹⁵

Monocytes in mice are characterized by their expression of CD115 (macrophage colony‐stimulating factor receptor), CD11b (integrin‐α M chain), and like their counterparts in humans, murine monocytes are also subdivided based on the expression of the surface marker Ly6C. Typically, ‘classical’ monocytes are described as Ly6C⁺⁺ CD43⁻ CCR2⁺ CD62L⁺ CX3CR1^low and ‘non‐classical’ monocytes as Ly6C⁻ CD43⁺ CCR2⁻ CD62L⁻ CX3CR1^hi. ¹⁶ , ¹⁷ As in humans, GFP reporter CX₃CR1 knock‐in mice showed a small population of intermediates monocytes. ¹⁸

Monocytes contribute many niches following the original seeding of tissue‐resident populations during embryonic development. ¹⁹ Relevant to regulation of the whole‐body metabolism, monocytes contribute to the macrophage niche in the self‐renewing arterial tissue, ²⁰ in the expanding population of macrophages in the liver postnatally, in the fully differentiated Kupffer cells post‐depletion following a model of diphtheria toxin Clec4f‐specific depletion, ²¹ and replenish the macrophage pool in the intestine that persists into adulthood. ²² Moreover, during inflammation and in viral, ²³ bacterial ²⁴ and parasitic ²⁵ infections, circulating monocytes aid in the expansion of macrophages in most tissues. In a mouse model of schistosomiasis, host‐protective macrophage proliferation was dependent on recruitment of Ly6C^hi monocytes. ²⁶ , ²⁷ , ²⁸

Like monocytes, macrophages exhibit marked heterogeneity depending on location and physiological requirements. ²⁹ In their majority, tissue macrophages in the adult at steady‐state originate from the fetal liver and embryonic yolk sac precursors ³⁰ , ³¹ , ³² that give rise to a specialized population that includes osteoclasts (bone), Kupffer cells (liver), perivascular macrophages (liver) and alveolar macrophages (lung) with a versatile transcriptome that bestows on them a distinguishable identity. ³³ Within these diverse populations, macrophages exhibit great plasticity as a result of exposure to various stimuli, signaling molecules, nutrients and metabolites. ³⁴ For example, phenotypical characterization of adipose tissue macrophages differs in lean and fatty adipose tissue in response to, at least in part, the availability of free fatty acids, lipoproteins and carbohydrates in the environment. ³⁵ , ³⁶ In lean adipose tissue, macrophages express CD206 and CD301 (markers of alternative activation), whereas macrophages express pro‐inflammatory inducible nitric oxide synthase and tumor necrosis factor‐α (TNF‐α) in obese adipose tissue. ³⁷ The lung also possesses a heterogeneous macrophage population composed of alveolar macrophages, largely of embryonic origin, located principally in the airways, ³⁸ and interstitial macrophages, located in the lung tissue and shown to be derived from circulating monocytes. ³⁹ The role of localized tissue microenvironmental signals in murine macrophages has been demonstrated in vivo, where alveolar macrophages were less responsive to interleukin‐4 (IL‐4) at steady state, despite expressing normal IL‐4 receptor. However, dysregulated responsiveness to IL‐4 was restored when macrophages were removed from lung tissues, a process that was independent of commensals hosted in the lungs, pointing to an endogenous lung factor that regulates niche‐specific cell characteristics. ⁴⁰ Interestingly, alveolar macrophages displayed reduced respiratory capacity and impaired glycolytic reserve and capacity. ⁴⁰ Altogether, it is likely that with the continuous refinement of the ‘omics’ field more subpopulations of macrophages with defined plasticity and unique functional, transcriptional and metabolic fingerprints will be distinguished, enriching our understanding in how macrophages respond to stress and infections.

Metabolic programming of macrophages in homeostasis and upon activation

Metabolic profile and contributions of macrophages in equilibrium

M0 macrophages are resting or naive macrophages that have not been exposed to external stimulation. Metabolically, M0 macrophages resemble alternatively activated macrophages (M2) as they use an intact tricarboxylic acid (TCA) cycle and mitochondrial oxidative phosphorylation, where nicotinamide adenine dinucleotide and flavine adenine dinucleotide from the TCA cycle are oxidized through a series of electron carriers and form ATP as the result of electron transfer. ⁴¹ In general, in steady‐state, oxidative phosphorylation maintains metabolic homeostasis and an optimal energy state. ⁴² In immunological steady‐state, tissue‐resident macrophages are responsible for maintenance of tissue integrity and stability, pathogen recognition and clearance. ⁴³ , ⁴⁴ , ⁴⁵ Nevertheless, as the term ‘immunometabolism’ evolves, it becomes clearer that many of the functional capacities of immune cells are dependent on the metabolic state of the cell, and that in turn, some of the metabolic changes in the cell depend on external signals from the environment. In this section we explore some of the functions of macrophages associated with homeostatic balance when these cells have not been ‘primed’ or instructed by foreign signals and antigens. The sections hereafter will focus on metabolic and functional changes of macrophages upon recognition and activation by various parasite‐related signals.

As previously observed, macrophage identity can be shaped in response to different cues in compartmentalized tissues. ⁴⁶ For example, the colony‐stimulating factor receptor 1 (CSFR‐1) pathway, via CSF‐1 and/or IL‐34, is required to maintain populations of steady‐state macrophages in liver, kidney, bone marrow and spleen. ⁴⁷ In addition, CCAAT/enhancer binding protein β (C/EBPβ), a bZIP transcription factor, ⁴⁸ is required to maintain homeostatic numbers of functional large peritoneal macrophages and alveolar macrophages, but not splenic, skin or mesenteric macrophages in mice, ⁴⁹ suggesting tissue‐specific heterogeneity in support of macrophage function. Indeed, the complexity of macrophage identity due to the richness of cellular diversity even within the same tissue is reflected in the diversity of transcriptional regulators and tissue‐specific genetic expression. ⁵⁰ , ⁵¹ Just in the kidney, five discrete subpopulations have been identified ⁵² that express high levels of unique transcription factors such as aryl hydrocarbon receptor, nuclear factor of activated T cells 1 and 2, and interferon regulatory factor 9. ⁵¹ In the brain, a variety of transcription factors mediate transcriptional networks during development, steady‐state and disease. ⁵³ Among these, MADS Box transcription enhancer factor 2, polypeptide c, interferon regulatory factor, Smad2/3 and Spalt like transcription factor 1 are key in microglia lineage determination. ⁵³ In the liver, the transcription factor inhibitor of DNA binding 3 is essential for Kupffer cell development ⁵⁴ whereas Nuclear Receptor Subfamily 1 Group H Member 3 maintains Kupffer cell identity. ⁵⁵

In turn, in physiological conditions, macrophages are key to maintaining biological balance. A few examples of this phenomenon are observed in iron metabolism, where macrophages are the main cells that maintain iron balance and can act as ‘ferrostasts’ via regulation of iron levels in different tissue microenvironments. ⁴³ Iron uptake by macrophages is mediated by transferrin receptor, divalent metal transporter 1, low‐density lipoprotein‐related receptor 1, hemoglobin–haptoglobin receptor (CD163) and natural‐resistance‐associated macrophage protein 1, which support iron uptake by erythrophagocytosis. ⁵⁶ , ⁵⁷ Once in the cytoplasm, iron is transported across the phagosomal membrane and ferric reductase and ferroxidases convert it to ferrous and ferric iron that will be exported through ferroportin, dependent on the systemic demand for iron. ⁵⁸ , ⁵⁹

By employing a combination of elegant fate mapping, bulk, and single‐cell RNA‐sequencing analyses, DeSchepper et al. ⁶⁰ characterized a heterogenic population of self‐maintaining gut‐resident macrophages that sustain enteric neurons and submucosal blood‐vessel networks. Upon specific depletion of CX3CR1⁺ gut‐resident macrophages, physiological intestinal function, as well as anion secretion, electrical stimulation‐induced contractions and neuronal activity in the myenteric plexus were dampened in the mouse model when compared with mice not depleted of gut‐resident macrophages. Overall, these findings indicate that in healthy animals, macrophages are essential to maintain intestinal homeostasis and raise a potential role for tissue‐resident gut‐resident macrophages metabolic dysregulation in the development of metabolic disease. ⁶¹

Fundamentals of the metabolic programming of activated macrophages

Metabolic reprogramming of macrophages can be induced not only by the availability of different metabolic substrates like glucose or oxygen, but also by cytokine and antigenic signals. Currently, the nomenclature referring to the activation state of macrophages remains contentious, as many variables (macrophage origin and preparations, markers and stimuli) contribute to macrophage phenotypes. Nevertheless, an attempt to standardize the nomenclature and propose a common framework has been described by Murray et al., as well by Roszer. ⁶² , ⁶³

The current paradigm states that M0 macrophages switch to M1 or classically activated macrophages in response to interferon‐γ and Toll‐like receptor (TLR) agonists that induce a metabolic shift, where M1 macrophages produce ATP through the pentose phosphate pathway, and an increased rate of glycolysis. In this state, M1 macrophages are characterized by expression of inducible nitric oxide synthase, high levels of pro‐inflammatory cytokines like TNF‐α, IL‐1β, IL‐6, IL‐12 and IL‐23. ⁶⁴ Moreover, recent CoMBI‐T pipeline approaches that combine mass spectrometry‐based metabolic analysis and RNA sequencing‐based transcriptional profiling has shown that M1 phenotype associates with a truncated TCA cycle at the isocitrate‐to‐oxoglutarate (AKG) step and an increased ratio of (iso)citrate to AKG. ⁶⁵

In contrast, M2 (alternatively activated) macrophages display enhanced oxidative phosphorylation, purine synthesis, arginine metabolism and a dependence on glutamine to sustain M2 commitment. ⁶⁵ This metabolic state correlates with an immune phenotype characterized by up‐regulation of arginase‐1 (Arg1), chitinase‐3‐like protein 3, C‐type mannose receptor 1 (CD206), found in resistin‐like α (Retnla) and galactose‐type C‐type lectin (CD301) in mice. ⁶⁶ , ⁶⁷ In humans, polarization of macrophages by the definition of M2 macrophages is less clear, but IL‐4 or IL‐13 has been shown to increase expression of CD209, CD200R, CD1a and CD1b. ⁶⁸ Induction of an alternatively activated state requires oxidative metabolism, as elegant studies by Vats et al. showed that inhibition of mitochondrial respiration and fatty acid oxidation reduced M2‐related markers like Arg1 in a signal transducer and activator of transcription 6 (STAT‐6) and peroxisome proliferator‐activated receptor γ (PPARγ) coactivator 1β‐(PGC‐1β) dependent fashion. ⁶⁹ Moreover, M2 macrophages have been subcategorized into M2a, M2b, M2c and M2d. M2a are induced by IL‐4 and IL‐13, express Arg1, CD206, Retnla (FIZZ1) and transforming growth factor‐β, and are associated with wound healing and both protective and pathogenic roles in asthma, by inducing allergen clearance mediated by MRC1 and eosinophilic inflammation by TGM2, respectively. ⁶³ , ⁷⁰ , ⁷¹ Recently, Wang et al. have shown that when both oxidative phosphorylation and glycolysis are inhibited by 2‐Deoxy‐D‐glucose, macrophages fail to differentiate to M2a. ⁷² Nevertheless, when oxidative phosphorylation remains active, glycolysis is not required to maintain M2a differentiation, underscoring the plasticity of macrophages under different metabolic circumstances. M2b macrophages are stimulated by immune complexes, TLR ligands such as lipopolysaccharide and IL‐1β, secrete IL‐10, CCL1, IL‐6 and TNF‐α, and display angiogenic potential. ⁷³ Recently, M2b polarization induced by Schistosoma japonicum egg antigens was shown to depend on nuclelar factor‐κB signaling via the MyD88/mitogen‐activated protein kinase signaling pathway and TLR2. ⁷⁴ Moreover, M2b macrophages are essential for granuloma regulation during Schistosoma mansoni infection. ⁷⁵ M2c macrophages become activated by glucocorticoids and IL‐10, have high expression levels of CD86, MHCII, mer receptor tyrosine kinase and haptoglobin‐hemoglobin scavenger receptor (CD163) and are implicated in tissue repair via extracellular matrix remodeling. ⁷⁶ M2d macrophages, also identified as tumor‐associated macrophages, can be activated by TLR ligands, A2 adenosine receptor agonists, leukemia inhibitory factor and IL‐6. ⁷⁷ , ⁷⁸ They secrete epidermal growth factor, vascular endothelial growth factor and several cytokine mediators such as IL‐10, transforming growth factor‐β, IL‐12 and TNF‐α, thereby contributing to angiogenesis, tumor growth and immunosuppression in the tumor microenvironment. Overall, recent analysis of the metabolic modulation of each of these subtypes following polarization of human CD14⁺ peripheral blood mononuclear cells suggests a shared increase in fatty acid oxidation (FAO), supported by a decreased carnitine in M2b and up‐regulation of fatty acid‐binding protein 4 and lipoprotein lipase in M2a macrophages. Moreover, metabolite profiling using metabolic pathway topology showed a significant role for amino acid metabolism in each of the subsets, as the β‐alanine pathway was identified as highly impacted following polarization to the M2a, M2b and M2d phenotypes, which is consistent with previous reports that have showed β‐alanine metabolism altered, as a by‐product of increased arginine metabolism in anti‐inflammatory macrophages. ⁷⁹ , ⁸⁰ Importantly, despite many efforts to define macrophage subpopulations in both mice and humans, these classifications are largely limited to in vitro settings and extensive in vivo characterizations are still lacking. ⁶³

Host immunomodulation during helminth infection

Helminths are clinically relevant invertebrates comprising trematodes, also referred to as flukes, that are leaf‐shaped flatworms; cestodes or tapeworms that are elongated flatworms; and nematodes or roundworms that are cylinder‐shaped worms. Rates of infection are typically high in tropical and subtropical areas of the world where helminths are endemic, with 24% of the population, or 1·5 billion people, affected. ⁸¹ Although not typically associated with high mortality, they cause high morbidity, due in part to the parasite’s ability to survive and thrive in the host for decades, leading to chronic disease that often goes unnoticed and untreated. ⁸² , ⁸³ , ⁸⁴ Consequently, helminthiasis has been correlated with long‐lasting pathologies such as anemia, undernutrition, periportal fibrosis and hypertension, and impairment of cognitive development that translates into serious disease burden and high disability‐adjusted life‐years, ⁸⁵ , ⁸⁶ , ⁸⁷ , ⁸⁸ which accounts for the burden of the disease and represents the number of years of life lost through premature mortality or disability. ⁸⁹

The chronicity of helminth infections is partially explained by the parasite’s ability to manipulate and dampen the host immune responses. In brief, helminths target pattern recognition receptors and can down‐regulate TLRs and other genes associated with inflammatory responses. Hartgers et al. found that down‐regulation of TLR2 was associated with Schistosoma haematobium in school children in Ghana who showed less atopic allergic reactivity. ⁹⁰ Further, helminths also suppress initiating alarmin signals like IL‐33 ⁹¹ , ⁹² and the inflammasome system. ⁹³ , ⁹⁴ Indeed, helminths induce strong type 2 activation and alternative activation of macrophages characterized by the production of IL‐4, IL‐5, IL‐9, IL‐10 and IL‐13. ⁹⁵ , ⁹⁶ Nevertheless, given the complexity of these organisms, the immune response and activation profiles of macrophages are divergent and depend in part on the tissues that individual species migrate through in the host, as well excretory/secretory factors directly secreted by the helminths. Added to this, macrophage responses are dependent on their cellular lineage, ⁹⁷ the susceptibility of the genetic background ⁹⁸ and sex differences of the host. ⁹⁹

Mounting epidemiological evidence shows that helminth infections inversely correlate with many autoimmune, inflammatory and, importantly, metabolic diseases. ¹⁰⁰ , ¹⁰¹ , ¹⁰² , ¹⁰³ , ¹⁰⁴ , ¹⁰⁵ , ¹⁰⁶ This evidence warrants studying these parasitic organisms and their individual roles in host immunomodulation, and the potential of parasite‐derived products in diverse therapeutics. ¹⁰⁷ , ¹⁰⁸ , ¹⁰⁹ , ¹¹⁰ , ¹¹¹

Macrophage immunometabolism in Schistosoma and soil‐transmitted helminths

Metabolic regulation of macrophages in Schistosoma infection

Schistosomiasis is caused by infection with trematode parasites of the genus Schistosoma and affects over 200 million people in 74 countries. ¹¹² Among the species with most clinical relevance in human diseases are S. mansoni, S. haematobium and S. japonicum. ¹¹³ In the host, schistosomes induce a potent T helper type 2 (Th2) ‐biased immune response and polarization of macrophages to a broad M2 phenotype, which is vital for host survival. ¹¹⁴ , ¹¹⁵ Interestingly, multiple epidemiological studies have observed an inverse correlation between both active infection and a previous history of schistosome infection and metabolic diseases such as obesity and diabetes, raising interest in the metabolic reprogramming that schistosomes induce in diverse cells of the immune system. ¹⁰⁵ , ¹¹⁶ , ¹¹⁷ , ¹¹⁸ , ¹¹⁹ As one of the major players in the pathology of disease, macrophages modulate initiation and disease resolution in schistosomiasis. ¹²⁰ In recent years, it has become apparent that metabolic processes can regulate immune responses, and it is likely that appropriate responses to the disease require a partnership between metabolic and immunological pathways. Hence, this section focuses on the metabolic modulation that is induced by schistosome infection and schistosome products on organ‐specific populations of macrophages.

Recently, the kinetics of the expression of genes associated with hepatic lipid metabolism during S. japonicum infection showed that the PPARγ was dynamically regulated in the hepatic compartment following infection, and that its mRNA levels were significantly up‐regulated after 4 weeks of infection. ¹²¹ Previously, PPARγ was identified as a link between macrophage alternative activation and glutamine metabolism. ¹²² PPARγ is necessary for induction of an M2 phenotype and is inducible by IL‐4. ¹²³ , ¹²⁴ Schistosomes have recently been shown to produce the lipid lysophosphatidylcholine, which is able to induce M2 activation in a PPARγ‐dependent manner ¹²⁵ (Fig. 1).

Figure 1.

Schistosome modulation of tissue and macrophage metabolism. Schistosome‐derived products induce a type‐2 response with secretion of interleukin‐4 (IL‐4), IL‐13, IL‐5, IL‐10 cytokines by eosinophils and lymphocytes. Concomitantly, schistosome products are recognized by pattern recognition receptors in macrophages and induce an alternative activated phenotype, dependent on the signal transducer and activator of transcription 6 (STAT‐6), and reprogramming of metabolism‐related genes. Alternative activation by schistosomes up‐regulates fatty acid oxidation and attenuates lipid accumulation. Up‐regulation of AKT and mTORC1 modulates catabolism and glucose metabolism. Additionally, schistosome‐induce macrophage reprogramming contributes to the metabolic regulation of hepatic and adipose tissue and the protection from high‐fat diet induced weight gain, type 2 diabetes and atherosclerosis. AKT, protein kinase B; mTORC1, mammalian target of rapamycin complex 1.

Consistent with alternative activation, mitochondrial uncoupling protein 2, relevant for lowering mitochondrial membrane potential and maintaining respiration, ¹²⁶ , ¹²⁷ is up‐regulated in the acute stage of infection. ¹²¹ In addition, the modulation of macrophage metabolism does not seem to be limited to active infection as in vitro analysis has shown that Schistosoma egg antigen (SEA), a soluble mixture of highly glycosylated proteins that acts as a potent immunomodulator, ¹²⁸ can increase the functional activity of 5' AMP‐activated protein kinase (AMPK), protein kinase B (AKT) and mammalian target of rapamycin complex 1 (mTORC1), so supporting glucose metabolism in macrophages ¹²⁸ (Fig. 1). Moreover, SEA treatment decreased lipid accumulation in macrophages in a phosphorylated‐AMPK‐dependent manner. ¹²¹ Similarly, Qian et al. found that SEA up‐regulates the expression of genes associated with lipid oxidation in macrophages. ¹²⁹ However, among the essential genes modulated by the parasite antigen they found medium‐chain acyl‐CoA dehydrogenase and Cytochrome P450 protein 4 to be significantly increased. Further, phosphatase and tensin homolog, a tumor suppressor gene involved in enhanced oxidative phosphorylation and decreased glycolysis, ¹³⁰ was down‐modulated in response to exposure to SEA, thereby suggesting that SEA induces reprogramming of glucose and lipid metabolism. These data indicate a role for phosphatase and tensin homologue in the reprogramming of macrophages following parasite antigen exposure. ¹²⁹

In addition to the effects mediated by schistosome egg antigens, the excretory and secretory products of S. mansoni cercariae, which are free‐swimming larvae that are the infective stage of the parasite in mammals, ¹³¹ also drive the production of IL‐10 and in turn rapidly trigger ERK1/2, p38 and cAMP response element‐binding protein (CREB) activation in bone‐marrow‐derived macrophages in a TLR2‐ and TLR4‐dependent fashion. Consequently, CREB activation results in transcriptional regulation of a series of genes involved in metabolic processes, including glycolysis, oxidative phosphorylation and protein ubiquitination. ¹³²

Changes in the catabolic metabolism of the murine host’s spleen have been observed during acute infection with S. mansoni. Several pathways related to the TCA cycle, respiratory transport and amino acid metabolism were differentially expressed in infected mice at the peak of the acute phase as determined by a large‐scale, label‐free shotgun proteomics approach, with many of these alterations attributable to macrophages, highlighting the impact of schistosome infection on the metabolic modulation of organ‐specific macrophage populations. ¹³³

Active infection with S. mansoni also results in metabolic reprogramming of two subpopulations of liver macrophages (perivascular macrophages and Kupffer cells) in a model of atherogenesis‐prone mice (ApoE^−/−) on a high‐fat diet, where infection improves glucose sensitivity, circulating cholesterol levels and body mass index in vivo. Metabolic reprogramming of macrophages coincided with an M2 phenotype (CD301⁺ CD206⁺) and was characterized by regulation of genes involved in amino acid biosynthesis, such as Psat1, as well as genes involved in glycogen synthesis like Gys1. ¹³⁴ These data might suggest that macrophage‐specific metabolic regulation mediated by schistosome infection can affect overall systemic metabolism, underscoring the importance of understanding helminth metabolic regulation of macrophages ¹³⁴ (Fig. 1). Recent data suggest that schistosome‐induced metabolic modulation may be sex dependent. Infection in male ApoE^−/− mice improves glucose tolerance, diet‐induced obesity and triglycerides, but this does not occur in female mice. Interestingly, infection induces an increase in spare respiratory capacity and mitochondrial biogenesis in bone‐marrow‐derived macrophages (BMDM) from infected male mice, but this does not occur in BMDM from infected female mice. Moreover, analysis of the lipidomic profile of BMDM from male mice indicates that infection induces a reduction in cellular cholesterol esters, with a concomitant increase in shuttling of glucose and palmitate to the TCA cycle, ¹³⁵ which supports the idea of an infection‐ and sex‐dependent increase in mitochondrial β‐oxidation. Critically, the presence of these modulations in unstimulated BMDM indicates that metabolic reprogramming occurs in the bone marrow myeloid pool in the context of infection. Collectively, these studies also show that schistosomes induce profound changes in amino acid metabolism. Glutamine accumulation is a feature of M2 macrophages, and glutamine promotes polarization through a glutaminolysis‐derived α‐ketoglutarate‐dependent pathway that also inhibits M1 polarization by suppressing the nuclear factor‐κB pathway. ¹³⁶ To date, in the context of metabolic profiling, females have been studied in relatively few human or murine helminth models, so it is unclear if this is specific to schistosomiasis, but there are known sex‐dependent dimorphisms in the susceptibility and pathology of metabolic disorders. ¹³⁷ , ¹³⁸ , ¹³⁹ Notably, dimorphism in the susceptibility to helminth infections has been described by Wedekind and Jakobsen, who found that males were more likely to be infected and to have higher parasite numbers than females in an experimental model. ¹⁴⁰

Among S. mansoni‐specific molecules, ω1 plays a dominant role in type 2 activation ¹⁴¹ , ¹⁴² and in the maintenance of metabolic homeostasis. ¹⁴³ While ω1 does not directly activate macrophages, a mechanism dependent on its T2 RNase activity induces binding to CD206, so triggering other innate cells to secrete IL‐4, IL‐5 and IL‐13 to polarize macrophages to an alternative activation state. ¹⁴³

Interleukin‐4‐induced alternative activation has previously been shown to be dependent on both FAO via cell intrinsic lysosomal lipolysis, and mTORC2‐dependent up‐regulation of glycolysis. ¹⁴⁴ , ¹⁴⁵ It has recently been demonstrated that mitochondrial FAO is higher in human adipose‐associated macrophages than in adipocytes, and that enhancing mitochondrial FAO in a model of type 2 diabetes reduces triglyceride content and inflammation. ¹⁴⁶ These data support the idea that during schistosome infection, reprogramming of macrophages could underlie whole‐body metabolic changes. Although it is not clear how schistosome products induce metabolic modulation in human macrophages, new more sensitive tools to study cellular metabolism will allow an in‐depth analysis to help dissect the contributions of these complex metabolic pathways to the biology of infection in humans. Many questions remain regarding the evolutionary advantages that these switches provide to helminths in general and schistosomes in particular.

Effect of soil‐transmitted helminths on metabolism

Soil‐transmitted helminths (STH) are also known for their immunomodulatory properties. Infections with Ascaris lumbricoides, Ancylostoma duodenale, Necator americanus, Strongyloides stercoralis and Trichuris sp. (collectively known as soil‐transmitted helminths) elicit a potent Th2 response and correlate with enhanced insulin sensitivity and improved metabolic index in both human populations and mouse models. ¹⁴⁷ , ¹⁴⁸ , ¹⁴⁹ The effects of STH on metabolic disease are, to some extent, explained by regulation of metabolic homeostasis by IL‐4‐competent eosinophils in adipose tissue ¹⁵⁰ and reduced Th1 inflammation, ¹⁵¹ , ¹⁵² which is partially due to regulation of fatty acid metabolism by STH. ¹⁵³ For example, infection with Strongyloides venezuelensis, the murine analogue of S. stercoralis, lowers palmitic acid in the intestine of infected mice on a high‐fat diet, ¹⁵³ thereby counteracting the TLR4‐dependent indirect induction of inflammation by palmitic acid. ¹⁵⁴ , ¹⁵⁵ Moreover, human infection with S. stercoralis is associated with lower circulating Th1 cytokines, insulin, glucagon, and the hormones adiponectin and adipsin (required for fatty acid breakdown and regulation of glucose and β‐cell functions), all of which showed a reversal following antihelminthic treatment. ¹⁵⁶ Similar to schistosomes, most of the STH produce antigens that induce an alternatively activated phenotype in macrophages, ¹⁵⁷ which is necessary to contribute to the expulsion of the intestinal parasites and wound repair. ¹⁵⁶ , ¹⁵⁸ , ¹⁵⁹ In the case of Trichuris suis, excretory/secretory products by themselves induce Arg1 and nitric oxide synthase 2 expression in BMDM, which highlights the phenotypic heterogeneity of macrophages in vivo and probably reflects an immunosuppressive state rather than an M2 phenotype. Additionally, T. suis recombinant proteins such as nucleoside diphosphate kinase and triosephosphate isomerase inhibit TNF‐α expression and induce phosphorylation of STAT‐3 and expression of transcriptions factors that drive anti‐inflammatory processes (such as C/EBPβ and nuclear factor IL‐3‐regulated 3) in BMDM. ¹⁶⁰ Like schistosomiasis, changes in the activation phenotype of macrophages in response to STH also correlates with regulation of metabolic disease. ¹⁶¹ For instance, infection with the murine intestinal parasite Heligmosomoides polygyrus bakeri polarized macrophages to an M2 phenotype, that when transferred to uninfected recipients receiving a high‐fat diet, induced tissue browning and uncoupling protein 1 protein levels in adipose tissue, which was accompanied by ameliorated levels of glucose, fat, cholesterol level, leptin and TNF‐α. ¹⁶¹ The maintenance of the M2 phenotype requires populations of both eosinophils and innate lymphoid type 2 cells (ILC2) that produce IL‐4 and IL‐13 and are localized in perigonadal and other areas of visceral adipose tissue, as depletion of both cell populations is linked to reduced Arg1 expression in murine adipose alternative activated macrophages ¹⁶² (Fig. 2). Using the murine Nippostrongylus brasiliensis model of STH, Yang et al. have recently shown that infection reverses high‐fat‐diet‐induced adipose mass gains, hepatic steatosis and liver triglycerides, while restoring glucose homeostasis. These protections correlate with the accumulation of M2 macrophages in the epididymal fat, a process that is only partially dependent on IL‐13 and STAT6, ¹⁶³ suggesting that metabolic modulation is not simply driven by alternative activation, but may also be driven directly by parasite antigens. Expansion of IL‐4‐ and IL‐13‐expressing ILC2 following helminth infection ¹⁶⁴ highlights the ability of these parasites to orchestrate a systemic response that shapes diverse tissue compartments as well as key cellular players, which results in systemic metabolic regulation (Fig. 2). As our knowledge of immunometabolism and its regulation by intestinal parasites grows, it is likely that processes that influence systemic metabolism will come to light that are distinct from what has been detailed in response to schistosome infection. In particular, data revealing changes in the microbiota as a result of intestinal helminth infections and disruption in the metabolome of the intestine ¹⁶⁵ , ¹⁶⁶ , ¹⁶⁷ that results in improved insulin sensitivity ¹⁵³ add another layer of complexity to the understanding of the impact of STH on the immunometabolism of diverse organs and immune cells.

Figure 2.

Soil‐transmitted helminths (STH) and regulation of metabolic homeostasis. Intestinal helminths such as roundworms, whipworms and hookworm lodge in the colon and cecum, elicit a strong type 2 response, and reduce T helper type 1 inflammation by lowering fatty acids in the intestine of infected mammals. Moreover, they induce polarization of macrophages to an alternative activated phenotype, characterized by Arg‐1 and Nos‐2 up‐regulation, downmodulation of pro‐inflammatory factors like IL‐6, TNF‐α, STAT‐3, C/EBPβ, and NFIL3 and secretion of IL‐4, IL‐13 and IL‐5 in macrophages, eosinophils and innate lymphoid cells 2 (ILC2). Infection by STH contributes to changes in the adipose compartment and adipose tissue browning, mediated in part by UCP‐1, which correlates to reduced fat and cholesterol levels and overall improved metabolic index in humans. Arg‐1, arginase‐1; IL‐6, interleukin‐6; Nos‐2, Nitric oxide synthase 2; STAT‐3, signal transducer and activator 3; C/EBPβ, CCAAT‐enhancer‐binding protein; NFIL3, Nuclear Factor, IL‐3 Regulated; TNF‐α, tumor necrosis factor α.

Concluding remarks

Over the past decade, a combination of cross‐sectional studies in populations endemic for parasitic helminths, and animal models have demonstrated that both active infection and treatment with helminth antigens such as SEA ¹¹⁹ can improve parameters of whole‐body glucose homeostasis, obesity and cholesterol metabolism. Multiple mechanisms have been proposed, but most revolve around the generation of type 2 immunity in innate lymphocytes (macrophages ¹³⁷ , ILC2 ¹¹⁹ , ¹⁴³ , eosinophils ¹²² ) recruited to metabolic tissues such as the liver and white adipose tissue (Figs 1 and 2). It is most probable that all of these cell types work in concert with the microbiota to orchestrate metabolic homeostasis in the context of helminth infection, as well as during ideal lean conditions. In the case of macrophages, there is compelling data that helminth‐polarized macrophages alone are able to modulate whole‐body metabolism, and that helminth antigens are able to modulate metabolism via IL‐33 induction ¹⁴³ and independently of IL‐13 and STAT‐6. ¹⁶³ However, the relative contributions of cytokines versus helminth antigens and the mechanistic role of each innate cell type in regulating glucose and lipid metabolism remain to be fully elucidated (see Outstanding Questions). As such, it is critically important to identify both new molecular pathways modulated by helminths, as well as novel modulatory helminth antigens capable of inducing whole‐body metabolic reprogramming.

Outstanding Questions in the Field.

• What is the relative contribution of macrophages to the metabolic profile of metabolic organs such as liver, skeletal muscle, adipose tissue, pancreas? Are the molecular mechanisms driving cellular metabolism the same in each compartment?

• What is the dominant driver of re‐programming of macrophage and other innate cells metabolism: cytokines or helminth antigens?

• What role does biological sex play in metabolic re‐programming by helminths? Are there sex‐dependent effects in other innate cells such as ILC2 and eosinophils during schistosomiasis? Is there sexual dimorphism in the metabolic protection induced by various STH?

• How long lived is myeloid and whole‐body metabolic re‐programming in the absence of ongoing infection?

• Does the type‐2 induced metabolic re‐programming of murine innate lymphocytes recapitulate in humans? Is reprogramming tissue specific, or does it extend to monocytes and myeloid progenitors?

Disclosures

The authors have no competing interests. Diane Cortes‐Selva is currently and employee of Janssen Biotherapeutics.