Leishmaniasis is the family of diseases caused by infection with protozoan Leishmania parasites. Because pathology depends upon both the species of parasite and the host immune response, leishmaniasis can manifest in three main forms: cutaneous leishmaniasis (CL), mucocutaneous leishmaniasis (MCL), and visceral leishmaniasis (VL). Leishmania parasites are transmitted via sandfly bites and are endemic in more than 90 countries across Africa, Asia, and Latin America resulting in 1-2 million new cases of leishmaniasis each year (1, 2). There is currently no human vaccine and existing treatments against Leishmania parasites are toxic to the host, difficult to administer, require a long duration, and are often ineffective (3). The lack of new treatments or vaccines has made global disease control and elimination efforts challenging, reiterating the importance of understanding the host immune response to identify potential therapeutic targets (4).

Upon a sandfly bite, parasites are taken up by macrophages in the skin. After phagocytosis, parasites reside in phagolysosomes in macrophages and begin multiplying. Controlling parasite burden is dependent on a predominant Th1 immune response where CD4⁺ T cells produce IFNγ which activates macrophages to kill parasites by releasing nitric oxide (NO) and reactive oxygen species (ROS) (5). During CL, neutrophils and inflammatory monocytes are initially recruited to the site of infection (6). Severity of disease is highly dependent upon both parasite burden and the host inflammatory response with excessive inflammation contributing to overall pathology and extending the duration of disease (7). Despite an effective immune response, parasites can persist at low levels in the skin for years even after dermal lesions have resolved (8).

Leishmanial lesions are characterized by hypoxia and the presence of pro-inflammatory cells and cytokines (9–12). During inflammatory hypoxia, transcription factors hypoxia-inducible factor (HIF)-1α and HIF-2α are induced by decreased oxygen availability in tissues (13). HIF-α transcription factors are master regulators of genes involved in metabolism and the cellular response to oxygen deprivation (14, 15). Upon activation, HIF-α subunits bind aryl hydrocarbon receptor nuclear translocator (ARNT; also known as HIF-1β), and ARNT/HIF-α heterodimers translocate to the nucleus where they induce the transcription of HIF-α target genes (16). HIF-α subunits can also be activated by oxygen-independent mechanisms such as TLR ligation, pro-inflammatory cytokines, or ROS stimulation (17, 18). Furthermore, under normoxic conditions LPS induces HIF-1α expression via MyD88/NFκB signaling in macrophages, and mice deficient in HIF-1α are more susceptible to a variety of bacterial and fungal infections (17, 19–22).

During CL, human lesions contain elevated levels of HIF-1α and the HIF-α target, VEGF-A (19, 23, 24). Similarly, HIF-1α and VEGF-A are also elevated in lesions following experimental murine L. major infection (25, 26). Both inflammatory signaling, such as IFNγ production, as well as hypoxia in the skin promote HIF-1α accumulation in L. major-infected macrophages, but which signal occurs first and the relative contributions of each signal to HIF-1α signaling are not known (19, 27, 28). Myeloid-specific HIF-1α^-/- mice infected with L. major exhibit increased lesion sizes and parasite burdens due to impaired expression of NOS2, a HIF-1α-specific target gene (19). These data suggest activated HIF-1α in dermal myeloid cells contributes to parasite control through NO production. Additionally, mice deficient in myeloid ARNT/HIF-α signaling (LysM^CreARNT^f/f; missing both HIF-1α and HIF-2α pathways) infected with L. major exhibit decreased myeloid-derived NOS2 and VEGF-A which impairs lymphangiogenesis at the site of infection, resulting in larger lesion sizes, despite parasites being controlled (26). Altogether, these data suggest myeloid HIF-α signaling plays critical roles in both parasite control and lesion resolution during L. major infection.

HIF-α activation depends on the Leishmania parasite species. In contrast to L. amazonensis and L. donovani parasites, L. major parasites alone do not increase HIF-1α expression or activation under normoxic conditions (11, 19, 27, 29). Additionally, during in vivo L. major infection, both infected and bystander macrophages exhibit HIF-α activation compared to macrophages from naïve skin (28). Based on these findings, we hypothesize that during L. major infection, it is the host’s inflammatory response and microenvironment, rather than the parasite itself, that triggers HIF-α activation. To address this hypothesis, we performed transcriptomic analyses on macrophages from LysM^CreARNT^f/+ or LysM^CreARNT^f/f that either exhibit intact or impaired ARNT/HIF-α signaling in myeloid cells, respectively. LysM^CreARNT^f/+ or LysM^CreARNT^f/f macrophages were infected or not with L. major parasites and then treated or not with LPS and IFNγ to define the importance of ARNT/HIF-α signaling in response to L. major parasites in the presence or absence of a pro-inflammatory milieu. We find infection with L. major parasites induces transcriptional changes in macrophages and some of these early transcriptomic changes are absent in macrophages without HIF-α signaling. This indicates L. major induces some transcriptomic changes that are HIF-α-dependent, and L. major infection is sufficient to induce HIF-α activation in vitro, albeit minimal compared to pro-inflammatory stimuli. We discovered under inflammatory conditions, HIF-α signaling suppresses transcripts and pathways involved in translation such as ribosomal transcripts, EIF2 signaling and the ribosome pathway during infection with L. major. Additionally, we identified top enriched pathways associated with L. major infection during and apart from inflammatory conditions as well as with and without intact HIF-α signaling.

HIF-α signaling is a hallmark of lesions following Leishmania infection, but the specific cell types in lesions undergoing HIF-α activation are not known (11, 12, 27). To identify the cell types in leishmanial lesions that express the transcription factors HIF-1α and HIF-2α as well as their transcriptional target genes, we performed scRNASeq on lesions 4 weeks after dermal L. major inoculation (28). Specifically, single cells from the ears of infected and naive mice were bar-coded and sequenced using the droplet-based 10X Genomics Chromium platform. Unbiased hierarchical clustering using Seurat was performed to identify clusters indicative of individual cell types ( Figure 1A ). Of the 35 distinct cell types, HIF-1α is mainly expressed in keratinocytes, fibroblasts, chondrocytes, and endothelial cells in naïve uninfected skin. In contrast, HIF-1α is predominantly expressed by infiltrating cells including T cells, neutrophils, and monocyte-derived macrophages during L. major infection ( Figure 1B ). Of the 35 distinct cell types, HIF-2α is mainly expressed in fibroblasts, chondrocytes, and endothelial cells in naïve uninfected skin ( Figure 1B ). After infection, HIF-2α retains expression in fibroblasts, chondrocytes, and endothelial cells and is additionally expressed in infiltrating T cells and monocyte-derived macrophages during L. major infection ( Figure 1B ).

Because HIF-α expression does not always correlate to HIF-α activity, we examined HIF-1α and HIF-2α transcriptional target genes as a surrogate for HIF-α activation. Besides Ldha, overall HIF-1α and HIF-2α target genes are expressed at low levels in naïve skin ( Figures 1C, D ). In contrast, HIF-1α-specific target genes including Nos2, Pgk1 and Ldha are dramatically increased upon infection, and these are predominantly expressed in monocyte-derived macrophages ( Figure 1C ). Similarly, the HIF-2α-specific target gene Arg1 is also highly expressed in monocyte-derived macrophages and Arg1 is significantly upregulated during infection ( Figure 1D ). However, another HIF-2α-specific target gene Pou5f1 (protein name Oct4) was only minorly expressed in monocyte-derived macrophages ( Figure 1D ). Altogether these transcriptomic data show that monocyte-derived macrophages exhibit HIF-1α and HIF-2α activation following L. major infection.

Given lesional monocyte-derived macrophages exhibited HIF-1α and HIF-2α activation, we evaluated the host macrophage responses during L. major infection using macrophages derived from monocytes from the bone marrow. To investigate the role of HIF-α signaling, we used macrophages from mice missing both HIF-1α and HIF-2α signaling where ARNT is deleted in myeloid cells and compared those responses to macrophages with intact HIF-1α and HIF-2α signaling (11, 27, 42). To first explore the host macrophage response in cells with intact HIF-1α and HIF-2α signaling, differential expression analysis was conducted on infected LysM^CreARNT^f/+ control macrophages compared to uninfected LysM^CreARNT^f/+ control macrophages referred to as ARNT^f/+ going forward. Macrophages were infected with L. major at an MOI of 5:1. Several differentially expressed genes (DEGs) were upregulated with L. major infection including Gm15564, Gca, Stk35, and Socs1 while only Tcf4 was found to be downregulated with L. major infection ( Figure 2A , Table 1 ).

We next investigated transcriptional changes during L. major infection in macrophages devoid of HIF-α signaling by comparing the transcriptome of infected LysM^CreARNT^f/f macrophages to uninfected LysM^CreARNT^f/f macrophages referred to as ARNT^f/f for the duration of the study. Four genes were differentially expressed, including Mt1, Acod1, Il1β and a predicted gene, gm15564 ( Figure 2B , Table 2 ). Mt1, Acod1, and Gm15564 were also upregulated with infection in HIF-α competent macrophages, indicating these transcriptional changes are independent of HIF-α signaling ( Figure 2A , Table 1 ). Most of the transcriptomic changes seen during L. major infection were ablated in the absence of HIF-α signaling. For instance, 22 DEGs were upregulated in HIF-α competent macrophages with infection, that were not detected during infection in macrophages deficient for HIF-α signaling ( Figure 2A , Table 1 ).

Next, we analyzed enriched pathways during infection with L. major in HIF-α competent macrophages compared to their uninfected counterparts. KEGG analysis revealed several enriched pathways with L. major infection including the ‘PPAR signaling pathway’, ‘Rap1 signaling pathway’, and ‘Chemokine signaling pathway’ ( Figure 2C ). Additionally, the ‘Th17 cell differentiation pathway’ was downregulated in infected macrophages compared to uninfected ARNT^f/+ macrophages ( Figure 2C ). Of note, the ‘HIF-1α signaling pathway’ was upregulated with infection ( Figure 2C ). These data demonstrate that infection with L. major is sufficient to drive transcriptional changes in macrophages and activate HIF-α signaling.

To further characterize the cellular processes most affected by infection in macrophages either with or without HIF-α signaling, we conducted KEGG pathway analyses. The analysis revealed that during infection with L. major, the proteasome pathway is upregulated in the absence of HIF-α signaling ( Figure 2D ). When we investigated enriched pathways in infected macrophages with intact HIF-α, the proteosome pathway was not upregulated suggesting this pathway is normally suppressed by HIF-α during L. major infection. These data together indicate infection alone induces transcriptional changes that are HIF-α-dependent, suggesting infection with L. major parasites is sufficient to activate HIF-α signaling in vitro contrary to other reports (23, 25, 27).

During infection with L. major parasites, a strong Th1 immune response is formed resulting in the release of a multitude of pro-inflammatory mediators. To identify inflammation-related transcriptomic changes, both HIF-α signaling competent and deficient macrophages were infected with L. major and treated with LPS and IFNγ to mimic the in vivo pro-inflammatory environment. We identified 1,076 genes that were differentially expressed when comparing infected ARNT^f/+ macrophages treated with LPS and IFNγ to infected ARNT^f/+ macrophages not treated with LPS and IFNγ ( Figure 3A , Table 3 ). The top upregulated DEGs were Gpr18 and Mmp25 ( Table 3 ). Additionally, there were many immune-related transcripts that were upregulated to a lesser extent including Il12b, Cd40, and Nos2, all of which participate in the immune response to Leishmania parasites ( Figure 3A , Table 3 ). The top downregulated DEGs were Arrdc3, Rasgrp3, and Cdca7l comparing infected ARNT^f/+ macrophages treated or not with LPS and IFNγ ( Table 3 ). Furthermore, we investigated transcriptional changes in infected ARNT^f/f macrophages treated with LPS and IFNγ compared to infected ARNT^f/f not treated with LPS and IFNγ. We identified 1,191 DEGs ( Figure 3B , Table 4 ). The top 25 most upregulated genes in response to pro-inflammatory stimuli were the same for infected macrophages with or without competent HIF-α signaling, indicating these DEGs are independent of HIF-α signaling ( Figures 3A, B , Table 4 ). In contrast, there were differences in the top 25 downregulated DEGs in response to LPS and IFNγ stimulation in infected macrophages that are competent or impaired for HIF-α signaling ( Figure 3B , Table 4 ). The top downregulated DEGs in infected HIF-α deficient macrophages treated with LPS and IFNγ compared to infected HIF-α deficient macrophages not treated with LPS and IFNγ were Mdp1, Arap3, and Prmt3 ( Table 4 ).

A functional analysis was performed to identify pathways associated with pro-inflammatory stimulus administration in infected macrophages with or without HIF-α signaling compared to their infected macrophage counterparts with no stimulus. The KEGG analysis revealed pro-inflammatory stimuli upregulated pathways such as ‘TNF signaling receptor’, ‘IL-17 signaling pathway’, and several other inflammatory pathways ( Figure 3C ). In addition, these infected macrophages downregulated the ‘lysosome’ and ‘cGMP-PKG pathway’ in response to LPS and IFNγ administration ( Figure 3C ). Next, we compared infected ARNT^f/f macrophages treated or not with LPS and IFNγ by KEGG analysis. The results revealed similar upregulated pathways as the pro-inflammatory treated and infected ARNT^f/+ macrophages including ‘cytokine-cytokine receptor interaction’ and ‘TNF signaling pathway’ ( Figure 3D ). Interestingly, there were no significantly downregulated pathways in infected pro-inflammatory stimulated macrophages deficient for HIF-α signaling compared to infected macrophages also deficient for HIF-α signaling. In contrast to the HIF-α competent macrophages, during HIF-α deficiency, the ‘lysosome’ and ‘cGMP-Pk3 signaling pathways’ were upregulated in infected macrophages treated with LPS and IFNγ ( Figure 3D ). These data suggest in a pro-inflammatory environment, HIF-α suppresses pathways related to L. major infection including those involved in production of the phagolysosome and second messenger signaling.

After investigating changes involved with L. major infection alone and with pro-inflammatory stimulus in macrophages with and without competent HIF-α signaling, we directly compared the gene expression profiles of macrophages without HIF-α signaling to macrophages with HIF-α signaling under each condition. First, we compared ARNT^f/f to ARNT^f/+ under basal conditions. We identified two upregulated DEGs in the ARNT^f/f macrophages compared to ARNT^f/+ macrophages including, Isg20 and Spp1 suggesting HIF-α inhibits these genes during homeostasis ( Figure 4A , Table 5 ). Next, we analyzed differences between ARNT^f/f to ARNT^f/+ during infection with L. major parasites. When comparing macrophages without or with HIF-α signaling during L. major infection, we found several downregulated DEGs in macrophages with impaired HIF-α signaling which suggests under normal conditions these DEGs are mediated by HIF-α ( Figure 4B , Table 6 ). These DEGs include Il1β, Ccl5, Mcoln2, Mevf, and Socs1 ( Figure 4B , Table 6 ). In line with these results, Socs1, Mcoln2, Mevf were upregulated during infection with L. major parasites in HIF-α competent macrophages compared to uninfected HIF-α competent macrophages further suggesting these specific genes are dependent on HIF-α during infection with L. major ( Figure 2A ). By KEGG analysis, we found under basal conditions macrophages without HIF-α signaling upregulate the ‘ribosome’ and ‘DNA replication pathways’ compared to macrophages with HIF-α signaling ( Figure 4C ). This finding suggests that HIF-α restricts cell processes in the absence of infection in steady state. Furthermore, when we analyzed enriched pathways in infected macrophages without HIF-α signaling compared to infected macrophages with HIF-α signaling, we found there were minimal significantly enriched pathways ( Figure 4D ). Together, this dictates there are HIF-α-dependent transcriptomic changes during homeostasis and in response to L. major infection supporting previous data depicting infection activates HIF-α.

Finally, to further characterize the role of HIF-α signaling in infected macrophages under pro-inflammatory conditions, we compared the gene expression profile of infected macrophages deficient for HIF-α signaling stimulated with LPS/IFNγ to infected macrophages with intact HIF-α signaling under the same conditions. There were 102 DEGs between infected and stimulated macrophage without and with HIF-α signaling, 63 being upregulated and 39 downregulated ( Figure 5A , Table 7 ). Of note, the top upregulated DEGs included Slc26a11, Agap1, and Cxcr4 and the top downregulated DEGs contained Mmgt1, Arhgap4, and Mdn1 ( Figure 5A , Table 7 ). We predict the upregulated genes are inhibited by HIF-α signaling (Slc26a11, Agap1, and Cxcr4) while the downregulated DEGs are mediated by HIF-α signaling (Mmgt1, Arhgap4, and Mdn1). Interestingly, Cxcr4 expression is downregulated during infection and pro-inflammatory stimulation in HIF-α competent macrophages compared to HIF-α competent infected macrophages ( Table 3 ). This indicates HIF-α inhibits Cxcr4 during L. major infection under inflammatory conditions. To identify pathways enriched for our DEGs, we performed a KEGG analysis. The KEGG pathway analysis revealed the DEGs clustered into pathways related to translation and protein production (‘Ribosomes’ and ‘Protein export’) ( Figure 5B ). Upregulation of the ‘ribosome’ and ‘protein export’ pathways in macrophages without HIF-α signaling suggests that these pathways are suppressed by HIF-α. To further conduct gene set enrichment analysis we utilized the molecular signature database (MSigDB). The MSigDB analysis revealed that the ‘interferon gamma response pathway’ was significantly upregulated in stimulated and infected macrophages without HIF-α signaling indicating this pathway is inhibited by HIF-α signaling ( Figure 5C ). Additionally, the ‘oxidative phosphorylation pathway’ was found to be upregulated in HIF-α deficient macrophages indicating HIF-α signaling suppresses this pathway in infected macrophages in response to pro-inflammatory stimuli ( Figure 5C ). This suggests that HIF-α deficient macrophages are shunted towards a predominant oxidative phosphorylation profile rather than a dominant metabolic glycolytic profile.

Furthermore, many ribosomal related transcripts were enriched in the HIF-α deficient infected macrophages treated with pro-inflammatory stimuli such as Rpl4, Rpl7a, Rpl12, Rpl23, Rpl38, Rpl39, and Rps21. To further investigate these altered ribosomal transcripts, we conducted an ingenuity pathway analysis (IPA) comparing infected macrophages treated with LPS and IFNγ with or without intact HIF-α signaling ( Figures 5D, E ). We pinpointed ‘EIF2 signaling’ as the top upregulated pathway in macrophages without HIF-α signaling, proposing that in a scenario with both infection and inflammatory stimuli, HIF-α inhibits EIF2 signaling. This data is consistent with the KEGG pathway analysis indicating HIF-α signaling suppresses protein translation during inflammatory conditions. Of note, several other enriched pathways were identified by the IPA including ‘RhoA signaling’ and ‘Ephrin B signaling’ suggesting these pathways are inhibited by HIF-α ( Figures 5D, E ). Finally, ‘Ephrin Receptor Signaling’ and ‘Leukocyte Extravasation’ were downregulated in infected macrophages stimulated with LPS and IFNγ without HIF-α, again suggesting this pathway is mediated by HIF-α ( Figures 5D, E ).

To validate the transcriptomic findings, we employed quantitative PCR (qPCR). To directly investigate HIF-α dependent transcriptomic changes during L. major infection, we designed an assay to selectively stabilize HIF-α by utilizing dimethyloxallyl glycine (DMOG), a prolyl hydroxylase inhibitor that prevents HIF-α from being targeted for degradation by the proteosome (43). Briefly, macrophages were derived from C57BL/6 mice and 1) cultured in media, 2) infected with L. major, 3) treated with DMOG, or 4) infected and treated with DMOG. We analyzed the relative expression of ribosomal transcripts upregulated in response to HIF-α deletion in our transcriptomic data, suggesting DMOG administration should decrease the relative expression of these transcripts. These selected transcripts were contained within the EIF2 signaling pathway which was the top hit of differentially regulated pathways during L. major infection and pro-inflammatory stimulus administration, suggesting HIF-α suppresses this pathway. In confirmation, we found the expression of Rpl4 was significantly decreased in L. major-infected macrophages treated with DMOG compared to infected macrophages without DMOG suggesting HIF-α stabilization results in downregulation of Rpl4, consistent with our transcriptomic data ( Figure 6A ). Additionally, the expression of two other ribosomal transcripts, Rpl12 and Rpl23, were additionally decreased in infected macrophages treated with DMOG compared to infected macrophages treated with DMOG ( Figure 6A ).

To further validate our transcriptomic findings and investigate the functional impact of HIF-α stabilization during L. major infection, we designed an in vitro experiment to assess translational activity with or without HIF-α stabilization. We used puromycin (puro), a tyrosyl-tRNA mimic that inhibits translation and labels active ribosomes, to determine if HIF-α suppresses translation as suggested by our IPA analysis ( Figure 5 and Figure 6 ). HIF-α stabilization was achieved using DMOG. Macrophages were derived from C57BL/6 mice and cultured in four conditions: media alone, L. major parasites, DMOG alone, or both L. major and DMOG. Previously, we showed that lesional macrophages exhibit the highest puro signal during L. major infection compared to other cell types within the lesion, demonstrating lesional macrophages exhibit high translational activity in vivo (206). In line with this, macrophages cultured with media, L. major, or DMOG alone had 90-95% of cells positive for puro ( Figures 6B, C ). However, when macrophages were treated with both L. major and DMOG, the percentage of puro⁺ macrophages significantly decreased ( Figures 6B, C ). These results support that HIF-α stabilization during infection inhibits translation. Notably, this effect was specific to L. major infection, as macrophages treated with DMOG alone showed similar puro levels to those cultured with media or L. major alone ( Figures 6B, C ). Overall, these findings indicate that HIF-α suppresses translation during L. major infection, but this effect requires a pro-inflammatory environment potentially to allow for maximal HIF-α stabilization.

HIF-α activation is a hallmark of both CL and VL occurring in response to tissue hypoxia, TLR activation, ROS and cytokines like TNFα and IL-1β, all of which are present during Leishmania infection (9, 12, 29, 44–46). However, the direct contribution of the parasite versus the host response/microenvironment to HIF-α activation is not clear. Leishmania parasites can directly activate HIF-1α in macrophages, but the direct activation of macrophage HIF-1α is context dependent with the parasite species playing a major role. For instance, L. amazonensis parasites, which cause CL in South America, directly induce the expression of HIF-1α in human and mouse macrophages in vitro under normoxic conditions (10, 47). HIF-1α is also present in L. amazonensis-infected skin (10). While L. amazonensis parasites can drive HIF-1α expression on their own, HIF-1α also promotes L. amazonensis killing by macrophages under hypoxic conditions (47). Similar to L. amazonensis, L. donovani parasites, which cause VL in Africa and Asia, directly activate HIF-1α in macrophages in vitro under normoxic conditions (48, 49). L. donovani parasites increase HIF-1α expression, nuclear translocation and activity in a variety of macrophages including J774 cells, peritoneal macrophages and splenic-derived macrophages from BALB/c mice (48). To stabilize HIF-1α, L. donovani parasites use an array of mechanisms including depleting host iron pools to modulate prolyl hydroxylase activity and inducing microRNAs to limit NF-κB activation which establishes a suitable environment for parasite survival (48, 49). In vitro, HIF-1α blockade inhibits L. donovani intracellular growth and HIF-1α stabilization promotes L. donovani growth inside macrophages (48, 49). However, myeloid-specific HIF-1α^-/- mice infected with L. donovani and humans with a loss-of-function HIF1A gene polymorphism are more susceptible to infection (50). The role of HIF-2α in L. amazonensis and L. donovani infection has not been investigated.

Although L. amazonensis and L. donovani parasites can activate HIF-α directly, previous work shows that L. major parasites do not increase HIF-1α expression or activation under normoxic conditions in macrophages (11, 23, 27, 29). For example, HIF-1α and HIF-2α as well as HIF-1α-specific and HIF-2α-specific target genes are increased at the site of murine L. major infection, but in vitro infection of macrophages with L. major does not induce HIF-1α expression (11, 25). Rather L. major parasites require additional inflammatory signals such as LPS and/or IFNγ to induce HIF-1α accumulation and subsequent HIF-1α target expression like NOS2 and VEGF-A in macrophages (23, 27, 29). While HIF-α stabilization promotes L. donovani survival in macrophages, previous work has shown HIF-α stabilization does not impact L. major parasite growth in macrophages and may be why L. major parasites alone do not induce significant HIF-α protein accumulation (11, 27, 29). However, previous work from our laboratory found that macrophages derived from mice deficient in HIF-α signaling possess higher parasite burdens at 2 and 72 hours post-infection compared to macrophages derived from HIF-α competent mice (27). In support of this data, through pathway analysis, we have shown that in vitro, the HIF-1α signaling pathway is enriched during infection with L. major and many initial transcriptomic changes are HIF-α-dependent suggesting infection with L. major initiates the HIF-α transcriptional program ( Figure 2 ). This is consistent with an additional study investigating initial transcriptomic changes after in vitro L. major infection, reporting HIF-1α signaling is enriched in murine macrophages at 4 hours post-infection (51). Despite these transcriptomic indications, it is possible pro-inflammatory stimuli are required for optimal HIF-α activation and subsequent target gene activation. It is important to note that in the above-mentioned study, L. major infection was not associated with changes specifically in HIF-1α accumulation. In the present study we have investigated changes in the absence of both HIF-1α and HIF-2α signaling which could account for the discrepancies.

Among the transcripts involved in the subtle HIF-α program activated during infection with L. major were Socs1 and Mevf ( Figure 2A ). Here, we show that during infection these transcripts are upregulated in HIF-α competent macrophages ( Figure 2A ). Additionally, when we compared infection in HIF-α deficient macrophages compared to HIF-α competent infected macrophages, Socs1 and Mevf were strongly downregulated suggesting that HIF-α mediates the expression of Socs1 and Mevf during L. major infection ( Figure 4B ). Interestingly, Socs1 is involved in immune regulation suggesting that mediation of this transcript by HIF-α is another mechanism to limit excess energetic use during conditions of low oxygen availability.

HIF-α activation occurs in a wide variety of circumstances playing a central role in tissue adaptation to low oxygen tensions (52, 53). Namely, hypoxia can be a characteristic of both tissue injury and subsequent inflammation, where infiltrating cells increase the demand for nutrients and oxygen, further depleting the tissue stores (54, 55). Protein translation is an energetically demanding process and during hypoxia, inhibition of translation supports energy homeostasis and possibly promotes survival when energy stores are insufficient (56, 57). Therefore, translation during hypoxic conditions becomes selective; coordinating adaptation to promote cell survival under low oxygen and energy conditions (58). Specifically, hypoxic conditions have been shown to stifle protein translation through downregulation of EIF2α signaling which we have shown is directly suppressed by HIF-α signaling in macrophages during inflammatory conditions ( Figure 5 ) (59). Phosphorylation of eIF2a is necessary for mRNA translation inhibition during hypoxia and may be coordinated by HIF-α based on the current findings (57, 60).

In addition to acclimating tissue to low oxygen availability, HIF-α is also a master regulator of macrophage inflammatory and innate immune function (15, 61, 62). Inhibition of protein translation coupled with a shift in metabolism to glycolysis during hypoxia are both mechanisms to conserve energy directly manipulated by HIF-α (63–65). Previous reports have demonstrated that HIF-α is capable of shunting macrophages towards a M1 dominant phenotype by targeting glucose metabolism (66, 67). Elevated glucose metabolism coupled with HIF-α-induced ATP production are two major cellular mechanisms of overcoming low oxygen tension. As a result, macrophage-specific deletion of HIF-1α leads to impaired macrophage responses including lower glycolytic rates, lower energy generation, and impaired motility (68–70). Here we have shown that macrophages deficient for HIF-α signaling are predisposed to a dominant oxidative phosphorylation profile in comparison to HIF-α competent macrophages ( Figure 5 ). Our study confirms that HIF-α reprograms macrophages during L. major infection to cope with the energetic demand. We have also shown through IPA analysis that pathways associated with macrophage motility are dysregulated during genetic deletion of HIF-α signaling including RhoA signaling and leukocyte extravasation consistent with what is reported in the literature (71–73). Although we investigated the impact of HIF-α signaling in resting M0 macrophages and M1 polarized macrophages (through LPS/IFNγ administration), a limitation of our study is that we have not considered the importance of HIF-α signaling in M2 polarized macrophages. Because M2 macrophages serve as a permissive niche during Leishmania infection, future work will investigate the role of HIF-α signaling in M2 macrophages during L. major infection (74, 75).

In summary, we showed L. major infection elicits a subtle macrophage HIF-α program, but major transcriptomic changes dependent on HIF-α are only present in a pro-inflammatory environment. This supports our hypothesis that during in vivo L. major infection, HIF-α stabilization is dependent on the pro-inflammatory milieu and not L. major directly, which is in contrast to L. donovani infection where the parasite alone can stabilize HIF-α (48). Additionally, we have evidence suggesting HIF-α suppresses protein translation in response to L. major infection and pro-inflammatory stimulus. However, a limitation of our study is that we have not determined if HIF-α suppresses protein translation during infection of primary macrophages, human macrophages, or following in vivo infection with L. major. So, future work will assess the extent to which protein translation occurs in a HIF-α dependent manner, and if this is unique to L. major or if it is conserved in other skin infections and diseases. We hypothesize suppression of translation is a mechanism of cellular adaptation to the pro-inflammatory response and subsequent hypoxic conditions from infiltrating cells and their high energetic demand during infection. A complete understanding of HIF-α during inflammation is vital in developing targeted therapeutics not only for CL, but also for other inflammatory skin diseases psoriasis (76). These results are also broadly relevant to diseases where HIF-α is highly expressed such as metabolic disorders including obesity and diabetes and inflammatory conditions like rheumatoid arthritis (77–79).