Hypoxia‐inducible factor‐1 drives divergent immunomodulatory functions in the pathogenesis of autoimmune diseases

Abstract

Hypoxia‐inducible factor‐1 (HIF‐1) is a heterodimeric (HIF‐1α/ HIF‐1β) transcription factor in which the oxygen‐sensitive HIF‐1α subunit regulates gene transcription to mediate adaptive tissue responses to hypoxia. HIF‐1 is a key mediator in both regulatory and pathogenic immune responses, because ongoing inflammation in localized tissues causes increased oxygen consumption and consequent hypoxia within the inflammatory lesions. In autoimmune diseases, HIF‐1 plays complex and divergent roles within localized inflammatory lesions by orchestrating a critical immune interplay sponsoring the pathogenesis of the disease. In this review, we have summarized the role of HIF‐1 in lymphoid and myeloid immunomodulation in autoimmune diseases. HIF‐1 drives inflammation by controlling the Th17/T_reg/Tr1 balance through the tipping of the differentiation of CD4⁺ T cells in favour of pro‐inflammatory Th17 cells while suppressing the development of anti‐inflammatory T_reg/Tr1 cells. On the other hand, HIF‐1 plays a protective role by facilitating the expression of anti‐inflammatory cytokine IL‐10 in and expansion of CD1d^hiCD5⁺ B cells, known as regulatory B cells or B10 cells. Apart from lymphoid cells, HIF‐1 also controls the activation of macrophages, neutrophils and dendritic cells, thus eventually further influences the activation and development of effector/regulatory T cells by facilitating the creation of a pro/anti‐inflammatory microenvironment within the autoinflammatory lesions. Based on the critical immunomodulatory roles that HIF‐1 plays, this master transcription factor seems to be a potent druggable target for the treatment of autoimmune diseases.

Hypoxia‐inducible factor‐1 is a master transcription factor, which regulates different gene transcriptions in response to hypoxia following an oxygen‐sensitive mechanism. HIF‐1 modulates lymphoid and myeloid immune responses in autoimmune diseases in a divergent manner. HIF‐1 drives autoimmune inflammation by controlling divergent pro/anti‐inflammatory function of different immune cells located in the inflammatory lesions.

Abbreviations

AHR

aryl hydrocarbon receptor

B10

IL‐10‐producing regulatory B cells

C‐MAF

C‐musculoaponeurotic fibrosarcoma oncogene

E3L

ubiquitin ligase 3

EAE

experimental autoimmune encephalomyelitis

eATP

extracellular adenosine triphosphate

FOXP3

forkhead box P3

HIF‐1

hypoxia‐inducible factor‐1

IGM

insulin growth factor

iNOS

inducible nitric oxide synthase

MIP‐β

macrophage inflammatory protein‐β

NF‐κB

nuclear factor kappa‐light‐chain‐enhancer of activated B cells

NO

nitric oxide

P300

P300 transcriptional coactivator protein

PHDs

prolyl hydroxylases

pVHL

von Hippel–Lindau tumour suppressor protein

RNS

reactive nitrogen species

RORγt

retinoic‐acid‐receptor‐related orphan nuclear receptor gamma

ROS

reactive oxygen species

SOCS3

suppressor of cytokine signalling 3

STAT3

signal transducer and activator of transcription 3

STAT4

signal transducer and activator of transcription 4

T_eff

effector T cells

Th1

T‐helper 1 cells

Th17

T‐helper 17 cells

Tr1

type 1 regulatory T cells

T_reg

regulatory T cells

INTRODUCTION

Hypoxia‐inducible factor‐1 (HIF‐1) is a heterodimeric transcription factor consisting of HIF‐1α and HIF‐1β. HIF‐1β is constitutively expressed, whereas HIF‐1α is an oxygen sensor and its availability is inversely regulated by cellular oxygen tension [1]. Under normoxic condition, HIF‐1α is hydroxylated at specific proline residues by oxygen‐dependent prolyl hydroxylases (PHDs). Prolyl hydroxylation signals for ubiquitin proteasome‐mediated degradation of HIF‐α. von Hippel–Lindau tumour suppressor protein (pVHL), an E3 ubiquitin ligase, recognizes and binds to hydroxylated HIF‐1‐α [2]. pVHL, bound with HIF‐1α, acts as the substrate recognition component of E3 ubiquitin ligase complex and possesses ubiquitin ligase activity. This complex leads to the ubiquitination and subsequent proteasomal degradation of HIF‐1α. But under hypoxic condition, activity of PHDs is suppressed due to the reduced oxygen availability, resulting in stabilization and accumulation of HIF‐1α, translocation of HIF‐1α into the nucleus and heterodimerization with HIF‐1β. Heterodimerized HIF‐1 then binds to hypoxia‐responsive elements (HREs) as a transcription factor and activates the transcription of target genes [3] (Figure 1).

FIGURE 1.

Hypoxia induces stabilization, accumulation and nuclear translocation of HIF‐1α and therefore induces the transcription of HIF‐1 target genes. The availability of HIF‐1α in the cytoplasm is controlled by cellular oxygen tension. Under normoxic condition, HIF‐1α undergoes hydroxylation at specific proline residues by PHDs. Prolyl‐hydroxylated HIF‐1α is recognized by pVHL. Binding of hydroxylated HIF‐1α by pVHL leads to the ubiquitination of HIF‐1α. Polyubiquitinated HIF‐1α undergoes proteasomal degradation. Under hypoxic condition, prolyl hydroxylation of HIF‐1α is suppressed, leading to its stabilization, accumulation and nuclear translocation. Within nucleus, HIF‐1α dimerizes with HIF‐1β, binds to HRE, interacts with P300 and induces the transcription of HIF‐1 target genes. HIF‐1, hypoxia‐inducible factor‐1; HRE: hypoxia‐responsive element; P300, P300 transcriptional coactivator protein; PHDs, prolyl hydroxylases; pVHL, von Hippel–Lindau tumour suppressor protein

Hypoxia is defined as a condition with reduced oxygen availability relative to oxygen demand and is contingent upon the pathophysiological circumstance of the afflicted organ or tissue [4]. Hypoxia is intricately related to inflammation, and these two processes mutually contribute to each other and act by reinforcing mechanisms [5, 6, 7]. Hypoxia‐elicited inflammation has been implicated in a wide range of disease conditions [7]. And vice versa, inflammation‐induced hypoxia or ‘inflammatory hypoxia’ reflects the well‐established concept that ongoing inflammation causes increased oxygen consumption, which results in localized tissue hypoxia within inflamed tissues [5, 6]. Multiple mechanisms contribute to the development of hypoxia in inflammatory lesions. One important mechanism is the excessive consumption of available oxygen by the immigrating myeloid cells and activated lymphocytes within the inflammatory lesions [5, 8]. Myeloid cells consume substantial quantities of nutrients, energy and oxygen to accomplish the processes of migration to the sites of inflammation, phagocytosis and release of reactive oxygen species (ROS) [9, 10, 11]. In the case of lymphocytes, clonal expansion in inflammatory sites is an energy‐intensive process, which is predominantly dependent on oxidative phosphorylation during early phases of inflammation [5, 12]. Another important mechanism of elevated oxygen consumption in inflammatory sites is the increased generation of ROS and reactive nitrogen species (RNS). Elevated level of nitric oxide (NO) is generated by enhanced expression of different isoforms of NO synthase in neuron, endothelial cells and immune cells [13]. Increased generation of superoxide (O₂ ⁻) by activated neutrophils and macrophages in inflammatory sites can also result in diminished availability of molecular oxygen [14, 15, 16]. NO and O₂ ⁻ generated in inflammatory lesions react to produce the highly toxic oxidant peroxynitrite (ONOO⁻) [17]. Excessive generation of NO and ONOO⁻ contributes to the random damage of surrounding healthy tissue, leading to the exacerbation of inflammation [13, 17, 18]. Hypoxia developed within the inflammatory lesions, in turn, facilitates the stabilization of HIF‐1α and thus leads to the activation of transcription of target genes [3].

Autoimmune diseases are characterized by tissue destruction caused by misguided autoreactive cell‐mediated or humoral immune responses [19]. Because hypoxia within the inflammatory lesions is a sustained microenvironmental feature in autoimmune diseases, HIF‐1 has been implicated as a pathogenic mediator that enables persistence of inflammation despite the unfavourable hypoxic condition [20, 21, 22]. Indeed, a growing body of evidences revealed that HIF‐1α mediates a critical role in inflammation and pathogenesis of autoimmune diseases, including rheumatoid arthritis, inflammatory bowel disease, psoriasis, type 1 diabetes mellitus, systemic sclerosis and multiple sclerosis [23, 24, 25, 26, 27, 28]. Following inflammation‐elicited hypoxia within the inflammatory lesions, the level of stabilized HIF‐1 is increased, which facilitates the adaptation of the inflamed tissue to the hypoxic microenvironment through the transcriptional activation of target genes [29, 30, 31, 32]. Target genes of HIF‐1 vary widely depending on the cell type, tissue and pathophysiological context. At present, more than 100 direct target genes of HIF‐1 have been identified, which subserve a diverse function in inflammation, angiogenesis, tumorigenesis, energy metabolism, cell survival, differentiation, proliferation and apoptosis [6, 33, 34]. In conjunction with oxygen homeostasis and hypoxia adaptation, HIF‐1 performs an immunomodulatory role, which is critically important for directing the course of progression of autoimmune diseases. Immunomodulatory function of HIF‐1 in autoimmune diseases is very divergent and dependent on immune cell lineages and is intricately connected to other mediators of hypoxic inflammation, for instance, cytokines, hormones such as insulin or insulin‐like growth factors (IGF‐1 and IGF‐2), vasoactive peptides, such as angiotensin II, and availability of NO [5, 35]. This review illustrates the multifaceted role of HIF‐1 in modulating the lineage‐specific immune responses of autoimmune diseases.

LYMPHOCYTES

A safe route of inflammatory response of T cells requires the triggering of protective inflammation to eliminate infection, resolution of inflammatory phase after the eradication of invading elements, and containment of dysregulated or overaggressive immune sequelae. However, many mechanisms may contribute to autoimmune pathogenesis, including molecular mimicry, chronic/latent recurrent infection or post‐infectious immune dysregulation. An uncontained self‐aggressive inflammatory response of T cell can result in autoimmune tissue damage. Among different subsets of T cells, Th17 is the most prominent effector T cell in orchestrating T‐cell‐mediated autoimmune diseases, for example multiple sclerosis and experimental autoimmune encephalomyelitis (EAE) (a widely used animal model of multiple sclerosis), collagen‐induced arthritis, inflammatory bowel disease, uveitis and type 1 diabetes [36, 37, 38, 39] On the other hand, proliferation and autoimmune responses of effector T cells, including Th17 cells, are suppressed by anti‐inflammatory regulatory T (T_reg) cells [40, 41, 42]. Therefore, a fine balance is maintained between Th17 and T_reg cells during homeostasis, and interruption of this delicate balance can potentially threaten with the development of autoimmune diseases.

In the context of autoimmunity, non‐pathogenic Th17 cells are induced from naïve T cells by IL‐6 and TGF‐β, whereas the pathogenic Th17 differentiation programme requires IL‐23. Conversely, T_reg cells are induced from naïve T cells by TGF‐β and IL‐2 and undergo clonal expansion upon continued exposure to IL‐2 [40, 41, 43, 44, 45, 46]. The differentiation programmes of both Th17 and T_reg subsets have a common requirement for TGF‐β, which drives enhanced expression of the signature transcription factors RORγt and FOXP3, respectively. In addition to TGF‐β, differentiation of Th17 cells requires an additional signal from IL‐6 pathway. Therefore, differentiation of either Th17 or T_reg cells from naïve T cells becomes dominant depending on the relative abundance of IL‐6 and TGF‐β within the inflammatory lesion [46, 47]. Given that, TGF‐β induces the coexpression of RORγt and Foxp3 in naïve T cells [48, 49]. But, in the presence of high level of TGF‐β and absence of IL‐6, Foxp3 directly interacts, at least in part, with the RORγt and antagonizes its ability to bind the DNA, which eventually directs the differentiation of T cells away from Th17 transcriptional programme and towards the T_reg subtype. Nevertheless, in the presence of IL‐6 and low level of TGF‐β, signal transducer and activator of transcription 3 (STAT3) becomes activated, which allows the transcriptional activity of RORγt by overcoming the inhibition by Foxp3. Therefore, differentiation of Th17 is dependent on a sustained and unopposed transcriptional activity of STAT3 and RORγt [47, 50]. Enhanced activation of STAT3, however, may not be sufficient enough, by itself, to tilt the delicate balance between T_reg and Th17 towards pathogenic Th17. Therefore, a simultaneous downregulation of Foxp3 coupled with sustained expression of RORγt is also critical for pathogenic Th17 differentiation.

In the hypoxic milieu of inflammatory lesion, level of available HIF‐1 within the T cells is increased, which could be contributed both by the enhanced transcription of HIF‐1 through activated STAT3 under inflammatory condition and by the stabilization of HIF‐1 under low oxygen tension [46, 51]. Accumulated HIF‐1 plays a three‐pronged role in the skewed differentiation of Th17 cells in the inflammatory lesions (Figure 2). First, HIF‐1 directly binds to the hypoxia‐responsive element (HRE) at the promoter of RORγt and induces its expression in co‐operation with STAT3. Second, HIF‐1 associates with RORγt at the promoter of Th17 genes and recruits transcriptional coactivator P300 to the transcription complex, resulting in the generation of a permissive chromatin structure. Third, HIF‐1 suppresses the differentiation of T_reg by directly interacting Foxp3 and facilitating its degradation via PHD/VHL/ubiquitin‐mediated proteasomal degradation, the same ubiquitin ligase system responsible for the degradation of HIF‐1α itself. Therefore, HIF‐1 transcriptionally activates the genetic programme of Th17 subset while repressing the T_reg programme in the same time. In the mouse model of EAE, CD4⁺ T‐cell‐specific depletion of HIF‐1 results in a deficiency of IL‐17 production and an increased number of T_reg cells in the central nervous system, which culminates in the attenuation disease [46]. Taken together, HIF‐1 facilitates the transcriptional programme of Th17, pushing the population ratio of Th17/T_reg towards Th17 subset. In addition to the induction of RORγt transcriptional programme, HIF‐1 facilitates the metabolic switch from oxidative phosphorylation to glycolytic pathway as the primary generator of energy by inducing an increased expression of glycolytic enzymes, which reinforces the differentiation of Th17 cells over T_reg cells [52, 53].

FIGURE 2.

HIF‐1 modulates the differentiation of T cells towards pathogenic Th17/Th1 or regulatory T_reg/Tr1 cells depending on cytokine milieu within the hypoxic autoinflammatory lesions. The relative abundance of IL‐6, IL‐12, TGF‐β and IL‐27 dictates the differentiation of Th17, Th1, T_reg or Tr1 cells. TGF‐β induces the expression of both FOXP3 and RORγt. In the absence of IL‐6, FOXP3 antagonizes the binding of RORγt to the DNA, leading to the activation of T‐cell transcriptional programme towards Treg cells. But, in the presence of IL‐6 and low level of TGF‐β, IL‐6 pathway‐activated STAT3 induces the expression of RORγt, which forms transcriptional complex with P300, HIF‐1 and STAT3 leading to the expression of IL‐17. In the presence of IL‐27, T cells become IL‐10‐producing Tr1. IL‐27 induces the expression of AHR through the activation of STAT3 pathway. AHR together with C‐MAF induces the expression of IL‐10 and IL‐21. IL‐21 promotes the stability and growth of Tr1 through autocrine signalling. AHR together with STAT3 also induces the expression of ectonucleotidase CD39. CDC9 degrades ATP, a suppressor of Tr1, into AMP on the plasma membrane. In the presence of IL‐12, T cells become IFN‐γ‐producing Th1 cells through the activation of STAT4 pathway. Hypoxia within the inflammatory lesions leads to the accumulation of HIF‐1, also induced by IL‐6‐activated STAT3, which interacts with FOXP3 or AHR and facilitates their degradation via pVHL‐E3L‐mediated proteasomal degradation, suppressing the differentiation of T_reg or Tr1 cells. HIF1 inhibits the expression of STAT4 and therefore suppresses the development of IFN‐γ‐producing Th1 cells. HIF‐1 also inhibits the expression of SOCS3, which forms a feedback inhibitory loop with STAT3 and therefore promotes the differentiation of Th17 cells by restricting the suppression of STAT3 activation. HIF‐1 also induces the expression of glycolytic enzymes, which promotes the differentiation of Th17 cells over T_reg cells by facilitating the metabolic switch from oxidative phosphorylation to glycolytic pathway. HIF‐1 not only enhances but also inhibits the differentiation of Th17 by inducing the expression of iNOS, which suppresses the transcriptional activity of RORγt by NO‐mediated nitration. AHR, arylhydrocarbon receptor; AMP, adenosine monophosphate; ATP, adenosine triphosphate; CD39, cluster of differentiation 39; C‐MAF, c‐musculoaponeurotic fibrosarcoma oncogene; E3L, ubiquitin ligase 3; FOXP3, forkhead box P3; HIF‐1, hypoxia‐inducible factor‐1; IFN‐γ, interferon gamma; IL‐10, interleukin‐10; IL‐12, interleukin‐12; IL‐21, interleukin‐21; IL‐27, interleukin‐27; IL‐6, interleukin‐6; iNOS, inducible nitric oxide synthase; NO, nitric oxide; P300, P300 transcriptional coactivator protein; pVHL, von Hippel–Lindau tumour suppressor protein; RORγt, retinoic‐acid‐receptor‐related orphan nuclear receptor gamma; SOCS3, suppressor of cytokine signalling 3; STAT3, signal transducer and activator of transcription 3; STAT4, signal transducer and activator of transcription 4; TGF‐β, transforming growth factor‐beta; Th1, T‐helper 1 cells; Th17, T‐helper 17 cells; Tr1, type 1 regulatory T cells; T_reg, regulatory T cells

Contrary to the Th17‐skewing function, HIF‐1 may also antagonize the differentiation of Th17 cells. Yang and colleagues reported that T‐cell‐derived inducible nitric oxide synthase (iNOS), an well‐recognized target gene of HIF‐1, suppresses the IL‐17 promoter activation of T cells in vitro through the nitration of tyrosine residues (particularly Tyr346 and Tyr359) in RORᵞt, leading to the impaired differentiation and proliferation of Th17 cells [54, 55]. Moreover, T‐cell‐specific deficiency of iNOS significantly increases the percentage of IL‐17‐producing cells in the colon and exacerbates the disease in murine colitis [54]. Therefore, endogenous NO‐mediated suppression of Th17 cells contradicts the hypoxia‐induced pro‐inflammatory function of HIF‐1 in T cells in mediating the autoimmune diseases. Not only in Th17 cells but also in T_reg cells, the contradictory role of HIF‐1 has also been reported. Clambey and colleagues showed that HIF‐1α, under hypoxia as an environmental cue of inflamed microenvironment, selectively induces the transcription of Foxp3 by directly binding to a single consensus hypoxia‐responsive element (HRE) within the promoter of Foxp3 gene and therefore promotes the proliferation and relative abundance of Treg cells in vitro [56]. Clambey et al. also showed that HIF‐1α is required for optimal function of Treg cells and HIF‐1α‐deficient Treg cells are defective in controlling T‐cell‐mediated colitis. Additional research is needed to understand how these opposing functions of HIF‐1 in T cells are fine‐tuned in the pro‐inflammatory setting of an autoimmune disease.

In comparison with Th17 cells, IFN‐γ‐producing pathogenic Th1 cells are less prominent in driving the pathology of tissue‐specific autoimmune diseases [57]. Shehade et al showed that HIF‐1 has an important role in restricting the function of Th1 cells [58]. Under hypoxia, HIF‐1 inhibits the transcription of STAT4, a key transcription factor mostly activated by pro‐inflammatory cytokine IL‐12 and required for the development and function of Th1 cells. Concurrently, HIF‐1 also inhibits the transcription of suppressor of cytokine signalling 3 (SOCS3), which is a major negative feedback regulator of STAT3 required for the development and function of Th17 cells [58, 59, 60, 61]. Therefore, it is clear that HIF‐1 favours the differentiation and function of pathogenic Th17 cells over Th1 cells. However, this issue requires further investigation to have a better picture of how HIF‐1 restricts the differentiation and function of Th1 cells in a complex blend of cytokines within autoimmune lesions (Figure 2).

In addition to T_reg cells, type 1 regulatory T (Tr1) cells are another important CD4⁺ anti‐inflammatory player in the modulation of autoimmune diseases. Tr1 cells are Foxp3⁻ and play a non‐redundant role in controlling autoimmune inflammation by producing IL‐10 [62]. Decreased numbers and impaired function of Tr1 cells have been documented in a variety of autoimmune diseases [63]. IL‐27 is the key cytokine for Tr1 polarization of naïve CD4⁺ T cells, which is produced by different antigen‐presenting cells including dendritic cells, monocytes and macrophages [64]. Upon interaction with IL‐27R, IL‐27 activates STAT3 through Janus kinase (JAK)/STAT pathway. When activated, STAT3 upregulates the expression of aryl hydrocarbon receptor (AHR) [65, 66]. AHR synergizes with c‐musculoaponeurotic fibrosarcoma oncogene (C‐MAF), another transcription factor induced by IL‐27, to promote the transcription of IL‐10 and IL‐21 [67]. Under the influence of IL‐27, IL‐21 acts as an autocrine growth factor for Tr1 cells [68]. AHR, together with STAT3, also induces the expression of plasma membrane ectonucleotidase CD39. CD39 translocates to the membrane and degrades extracellular ATP (eATP), a danger signal released by inflamed and dying tissue. eATP induces the differentiation of Th17 cells by engaging purinergic P2 receptors, but inhibits the differentiation of Tr1 through P2X7 receptor signalling [66, 69]. Therefore, by hydrolysing eATP, CD39 restricts eATP‐mediated suppression of Tr1 differentiation. HIF‐1α has differing roles in Tr1 differentiation. Stabilized HIF‐1α promotes AHR ubiquitination and its degradation by proteasome. Therefore, under hypoxic condition, HIF‐1α interferes with AHR‐dependent signalling and therefore restricts the differentiation of Tr1 cells and indirectly promotes the differentiation of Th17 cells [66, 70]. On the other hand, HIF‐1α facilitates the metabolic reprogramming of Tr1 cells at the early stage of differentiation by inducing the expression of glycolytic genes, which is similar to that of Th17 cells and different from the oxidative phosphorylation associated with Foxp3⁺ T_reg cells [66]. This HIF‐1α‐induced early‐stage metabolic reprogramming could draw a diverging line between the developmental pathways of naïve T cells towards Tr1 or T_reg cells under the influence of appropriate anti‐inflammatory cytokine signalling (Figure 2).

Most of the current literature regarding the role of HIF‐1 in autoimmune diseases is overwhelmingly focused on CD4⁺ T cells. However, in a recent study, Doedens and colleagues reported that HIF‐1 induces the cytolytic functions of CD8⁺ T cells (e.g. release of cytotoxic granules containing granzymes B and perforin) in the context of persistent viral and tumour antigens, which results in induced apoptosis of host cells [71]. Therefore, the plausibility of a HIF‐1‐induced and CD8⁺ T‐cell‐mediated tissue destruction within the hypoxic microenvironment of pro‐inflammatory lesions in autoimmune diseases cannot be ruled out.

Although numerous studies have delineated the role of HIF‐1 in T cells, the function of HIF‐1 in B cells in relation to autoimmune diseases is yet to be completely understood. However, emerging interest has been focused on the role of HIF‐1 in B cells, with a particular interest in IL‐10‐producing CD1d^hi CD5⁺ B cells (known as regulatory B cells or B10 cells), because of their anti‐inflammatory functions in autoimmune diseases. Recently, Meng and colleagues illustrated the function of HIF‐1 in B cells implicated in the pathogenesis of collagen‐induced arthritis and EAE. In activated B cells, particularly CD1d^hiCD5⁺ B cells, HIF‐1 level increases, which synergizes with STAT3 to induce the expression of anti‐inflammatory cytokine IL‐10 that suppresses pathogenic T cells. In the absence of available HIF‐1, CD1d^hi CD5⁺ B cells become defective in IL‐10 production; hence, they provide an impaired protection against collagen‐induced arthritis and EAE. HIF‐1 also induces the expression of glycolytic genes, which facilitates the metabolic shift from oxidative phosphorylation to glycolysis, a critical factor for activation and proliferation of CD1d^hi CD5⁺ B cells [72]. Therefore, the function of HIF‐1 in B cells seems to be suppressive in the context of autoimmune pathogenesis (Figure 3).

FIGURE 3.

HIF‐1 modulates the activation of myeloid and B10 cells and therefore contributes to the pro/anti‐inflammatory proportion of T_eff versus T_reg cells. In M1 macrophages, HIF‐1 induces the expression of pro‐inflammatory cytokines and iNOS. Pro‐inflammatory cytokines induce the differentiation of Teff cells. NO, produced by iNOS, induces the differentiation T_reg cells but inhibits the differentiations and proliferation of T_eff cells. In dendritic cells, HIF‐1 induces the expression of IL‐10 and glycolytic enzymes. IL‐10 induces the development of T_reg cells, and glycolytic enzymes induce the activation and maturation of dendritic cells by shifting the metabolic programme from oxidative phosphorylation to glycolysis. In B10 cells, HIF‐1, in co‐operation with STAT3, enhances the expression of IL‐10, which induces the development of Treg cells. In neutrophils, HIF‐1 induces the expression of pro‐inflammatory cytokines and glycolytic enzymes. Pro‐inflammatory cytokines contribute to the formation of an inflammatory microenvironment within the autoimmune lesions. Pro‐inflammatory cytokines also act on the neutrophils in an autocrine mode and provide anti‐apoptotic survival signal through the activation of NF‐κB and MIP‐β. Glycolytic enzymes, induced by HIF‐1, also contribute to the survival signalling. Longer survival of neutrophils contributes to increased tissue damage in autoimmune diseases. B10, IL‐10‐producing regulatory B cells; HIF‐1, hypoxia‐inducible factor‐1; IL‐10, interleukin‐10; iNOS, inducible nitric oxide synthase; MIP‐β, macrophage inflammatory protein‐β; NF‐κB, nuclear factor kappa‐light‐chain‐enhancer of activated B cells; NO¸nitric oxide; STAT3, signal transducer and activator of transcription 3; T_eff, effector T cells; T_reg, regulatory T cells

MYELOID CELLS

Although lymphocytes play the most prominent role in the pathogenesis of autoimmune diseases, lymphocytes per se cannot account for the pathological course of the disease in its entirety. Rather, myeloid cells also contribute an inevitable role during induction, mediation and resolution of autoimmune pathogenesis. Macrophages, neutrophils and dendritic cells are the most prominent myeloid cells involved in inflammatory processes. Macrophages have a broad immunomodulatory functions and actively participate in the pathogenesis of many autoimmune diseases [73]. In fact, macrophages are among the earliest immune cells that home into inflamed tissues. Macrophages use pattern recognition receptors (PRRs) to recognize tissue damage or abnormalities by a mechanism in which PRR engagement of endogenous danger‐associated molecular patterns (DAMPs) initiates the creation of an inflammatory microenvironment [74]. In a mature microenvironment, macrophages exhibit a spectrum between two polarized phenotypes, which are referred to as pro‐inflammatory (M1) or anti‐inflammatory (M2) macrophages. Depending on the prevailing local stimuli within the microenvironment (e.g. types of PRR and cytokines produced by infiltrating lymphocytes or injured tissues), macrophages become polarized into M1 or M2 subsets. The relative M1/M2 ratio can potentially shape the progression of disease. Once activated by local stimuli, M1 and M2 macrophages release a wide range of immune mediators. M1 macrophages exert pro‐inflammatory cytotoxic effects, whereas M2 macrophages mediate anti‐inflammatory functions, for example wound repair, tissue remodelling and angiogenesis. M1 macrophages produce reactive ROS/RNS, cytokines (tumour necrosis factor‐α (TNF‐α), IL‐1, IL‐6, IL‐12 and IL‐23) and chemokines (CCL‐5, CXCL9, CXCL10 and CXCL5). Conversely, M2 macrophages produce vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), transforming growth factor‐β (TGF‐β), IL‐4, IL‐13 and IL‐10. This complex milieu of different cytokines and chemokines, produced by macrophages, lays the foundation for the formation of an inflammatory lesion and induces the influx and activation of infiltrating lymphocytes and other myeloid cells [73, 74, 75].

Hypoxia is not a key feature of the inflammatory lesions at the initial stage of their formation. Rather, hypoxia becomes more prominent during the progression of inflammation as the inflammatory sites become metabolically and immunologically more active by the infiltrating myeloid and lymphoid cells. Therefore, the role of HIF‐1 in myeloid cells comes to the prominence at the mature stage of the inflammatory lesions [73, 74, 75]. In contrast to M2 macrophages, HIF‐1 is mainly expressed in M1 macrophages. HIF‐1 induces the expression of pro‐inflammatory cytokines and chemokines in M1 macrophages, which reinforces the pro‐inflammatory nature of the microenvironment [76]. In its mature stage, autoimmune lesions become more populous with pro‐inflammatory infiltrates, which exacerbate the local tissue damage. In addition to cytokines and chemokines, iNOS is one of the most important transcriptional targets of HIF‐1, which is primarily expressed in M1 macrophages [77]. The iNOS enzyme generates nitric oxide (NO) using l‐arginine as substrate. Arginase 1, which is abundantly expressed in M2 macrophages, also uses l‐arginine as substrate to generate ornithine and urea. Expression of arginase 1 is induced by HIF‐2, an isoform of HIF‐1, in M2 macrophages. Because l‐arginine is used as a common substrate by both iNOS and arginase 1, arginase 1 in M2 macrophages indirectly regulates NO production by iNOS in M1 macrophages by limiting the availability of l‐arginine [76]. NO induces the differentiation of NO‐T_reg cells (CD4⁺ CD25⁺ Foxp3⁻), which suppresses the proliferation of CD4⁺CD25⁻ effector T cells in colitis, collagen‐induced arthritis and EAE [78, 79]. In the same time, NO inhibits the differentiation, proliferation and function of Th17 cells through the nitrosylation of aryl hydrocarbon receptor and RORᵞt, and therefore suppresses the diseases in EAE and colitis [54, 80, 81, 82]. Therefore, HIF‐1 in macrophages performs both pro‐inflammatory and immunosuppressive functions in autoimmune lesions (Figure 3). In addition to M1 macrophages, myeloid‐derived suppressor cells (MDSCs) are also well known for their anti‐inflammatory role through the upregulated expression of iNOS. In chronic inflammatory states (e.g. cancer and autoimmune diseases), impaired differentiation of myeloid progenitors into mature myeloid cells leads to the expansion and accumulation of MDSCs [83] MDSCs comprise two subsets including monocytic (M‐MDSCs) and granulocytic/polymorphonuclear (PMN‐MDSCs). Unlike PMN‐MDSCs, M‐MDSCs predominantly express iNOS [83]. Notably, HIF‐1 promotes accumulation of MDSCs in tumour microenvironment, which is hypoxic in nature [84]. Therefore, HIF‐1 may also drive accumulation of MDSCs and therefore play an anti‐inflammatory role in autoimmune lesions; however, additional research is needed to assess this possibility.

Neutrophils are typically the earliest responders to infectious or sterile inflammation (e.g. ischaemia/reperfusion or trauma) [82, 85]. Neutrophils are also known to participate in the pathology of autoimmune diseases of the CNS (e.g. MS, EAE and neuromyelitis optica spectrum disorder). They are prominently present in the autoimmune lesions of the CNS [86, 87, 88, 89, 90]. Neutrophils contribute to chronic inflammatory diseases of CNS by the secretion of pro‐inflammatory cytokines and chemokines, promoting the maturation of antigen‐presenting cells and disruption of the blood–spinal cord barrier [86, 90, 91, 92, 93, 94, 95]. A major regulator of neutrophil function is their short life span through constitutive apoptosis, which is vital for the limitation of tissue damage [96]. Chilvers and colleagues reported that peripheral blood neutrophils become resistant to apoptosis during hypoxia in vitro, which reveals a potential role of HIF‐1 in preserving neutrophil survival and activity in inflammatory environments [97]. Indeed, HIF‐1α‐deficient murine neutrophils displayed a remarkable reduction of cell survival in anoxia, which supported the role of HIF‐1 in the modification of intrinsic apoptotic thresholds in neutrophils. HIF‐1 regulates neutrophil apoptosis by controlling the activation of transcription factor NF‐κB pathway and secretion of macrophage inflammatory protein‐1β, both of which are critically involved in enhancing neutrophil survival, in hypoxic condition through the enhanced expression of an array of extracellular pro‐inflammatory cytokines, for example TNF‐α, IL‐1β and IL‐6 [98, 99, 100]. HIF‐1 also controls neutrophil survival and function through the regulation of metabolic switch to glycolysis, the major ATP‐generating pathway in myeloid cells, through an enhanced expression of glucose transporter 1 and glycolytic enzymes [101]. Taken together, the involvement of neutrophils in the pathology of autoimmune diseases of CNS and the role of HIF‐1 in neutrophil survival and function in response to hypoxia provide evidence that HIF‐1 facilitates the progression of autoimmune diseases by promoting an ameliorated apoptosis and functional persistence of neutrophils (Figure 3).

Myeloid‐derived dendritic cells are centrally involved antigen in presentation and activation of T cells. Tissue‐resident immature dendritic cells become activated and mature upon detection and taking up of antigen patterns in a pro‐inflammatory microenvironment. Mature dendritic cells are characterized by increased antigen presentation, cytokine production, expression of maturation markers and morphologic changes. Upon maturation, dendritic cells migrate to the lymph nodes, and, depending on their surface receptor expression and cytokine profile, they induce the differentiation of subsets of T cell [102, 103]. Under hypoxic condition in inflamed tissue, surface receptor and cytokine profile expressed by dendritic cells may change, which could potentially impact the differentiation of T cells. Hammami et al. [104] showed that HIF‐1α inhibits IL‐12 production in DCs and therefore limits the development of Th1 cells. On the other hand, hypoxia‐activated HIF‐1 in dendritic cells induces the expression of IL‐10 and therefore promotes the differentiation of T_reg cells, which is protective in colitis [105]. Like other myeloid cells, the major source of energy generation in mature dendritic cells is glycolysis. In the hypoxic environment of a pro‐inflammatory niche, HIF‐1 in mature dendritic cells mediates a striking increase in glycolysis as compared to basal level, leading to the development of a phenotype, which is optimal for rapid migration and antigen presentation to T cells [106] (Figure 3).

HIF‐1 IS A POTENTIAL THERAPEUTIC TARGET FOR AUTOIMMUNE DISEASES

Critical role of HIF‐1 in immunomodulation of autoimmune diseases has drawn attention towards it as a promising therapeutic target. An emerging literature has revealed the efficacy of HIF‐1‐targeted drugs in inhibiting cancer development, progression and metastasis. HIF‐1‐targeted drugs can be categorized as inhibitors of HIF‐1‐DNA interaction, inhibitors of transcription/translation of HIF‐1 mRNA, inhibitors of HIF‐1‐mediated transcriptional activity or activators of prolyl hydroxylase [107, 108]. Epigallocatechin‐3‐gallate (EGCG) attenuates arthritis in mice by inhibiting STAT3 and HIF‐1α and therefore increasing the relative proportion of T_reg over Th17 cells [109]. Under hypoxic condition, NO competes with O₂ for binding to mitochondrial cytochrome c oxidase, resulting in the cessation of mitochondrial respiration with the consequence of an increased level of non‐respiratory O₂ in the cell. O₂ that is not reduced to H₂O in oxidative phosphorylation is redistributed and becomes available for prolyl hydroxylation of HIF‐1α leading to its increased proteasomal degradation [110, 111]. Therefore, an exogenous donor of NO such as S‐nitrosoglutathione (GSNO), S‐nitroso‐N‐acetylpenicillamine (SNAP) or diethylenetriamine/nitric oxide adduct (NOC‐18) could be a promising therapeutic option for HIF‐1‐targeted treatment of autoimmune diseases. Antisense oligonucleotide targeting HIF‐1α (EZN‐2698) inhibits tumour growth in xenograft models of human prostate cancer [112]. Topotecan, an FDA‐approved drug currently being used for the chemotherapy of lung cancer and ovarian cancer, inhibits HIF‐1α translation [113]. Echinomycin selectively inhibits the binding of HIF‐1 to DNA showing promise for cancer treatment [114]. Bortezomib (PS‐341), an FDA‐approved drug for treatment of patients with multiple myeloma and patients with mantle cell lymphoma, represses transcriptional activity of HIF‐1α [115]. Although mostly studied for cancer treatment, findings from these studies suggest that HIF‐1 could be a potential target for designing drugs for autoimmune diseases. Therefore, based on the impact of these drugs in cancer, it would be worthwhile to determine whether these inhibitors of HIF‐1 could alter the immune interplay and therefore ameliorate autoimmune diseases in animal models and patients as well. On top of that, considering the divergent roles of HIF‐1 in different immune populations, possibility of potential adverse effects or compromised therapeutic output cannot be ruled out in a HIF‐1‐targeted treatment. Therefore, a lineage‐specific targeting of HIF‐1 might be more effective than the global‐targeting of HIF‐1.

CONCLUSION

Although the exact aetiology and pathogenesis of autoimmune diseases are complex and incompletely understood, a review of the literature reveals abundant evidence that HIF‐1 is pivotal in the modulation of immune mechanisms underlying these diseases. It is well established that inflammation and hypoxia are mutually intertwined. HIF‐1 accumulates under hypoxic condition within the inflammatory lesions and diversely modulates the function of different lymphoid and myeloid cells, which can potentially impact the pathogenesis of autoimmune diseases in a divergent manner. For instance, HIF‐1‐mediated polarization of the balance of Th17/Treg in favour of T_reg cells, metabolic switching of lymphocytes towards glycolysis, activation of pro‐inflammatory M1 macrophages, maturation of dendritic cells and reduced apoptosis of neutrophils promote the progression of autoimmune diseases. Conversely, HIF‐1‐induced expansion of CD1d^hiCD5⁺B cells, overexpression of iNOS by M1 macrophages and increased production of IL‐10 by B cells and dendritic cells play anti‐inflammatory roles in autoimmune diseases. Therefore, it deserves further comprehensive investigation to determine how to fine‐tune these contradictory immunomodulatory functions of HIF‐1 in autoimmune diseases by using available therapeutic interventions.

CONFLICT OF INTEREST

The authors have no conflict of interest to declare.