Enhanced Translational Activity Is Linked to Lymphatic Endothelial Cell Activation in Cutaneous Leishmaniasis

Abstract

Introduction:

Cutaneous leishmaniasis (CL) is a significant public health problem leading to permanently disfiguring skin lesions caused by Leishmania parasites. Lesion severity stems from an excessive host inflammatory response that prevents healing.

Materials and Methods:

Here, we characterized the transcriptional and translational responses of lymphatic endothelial cells (LECs) during murine CL using single-cell RNA sequencing data combined with flow cytometry and in vivo puromycin incorporation to assess translational activity.

Results:

We identified upregulation of antigen presentation pathways including MHC-I, MHC-II, and immunoproteasome transcripts in dermal LECs from Leishmania major-infected mice compared to naive controls. LECs also exhibited increased expression of guanylate binding proteins and interferon-inducible genes, indicative of immune activation. Moreover, our findings demonstrate that LECs in leishmanial lesions displayed heightened translational activity relative to LECs from uninflamed ears, and LEC translational activity was highest in activated LECs. Furthermore, LEC translational activity exceeded that of other cell types within the lesion microenvironment. Validating the transcriptomic data, LECs in lesions expressed elevated MHC-II and programmed death-ligand 1 (PDL-1), supporting their potential role in antigen presentation. Functional assays using DQ-ovalbumin confirmed that LECs from leishmanial lesions efficiently uptake and process antigens, highlighting their capability as antigen presenting cells in the inflamed dermal microenvironment.

Conclusion:

Overall, our study reveals the activation status of LECs in leishmanial lesions, shedding light on their potential role in shaping local immunity and inflammation in a variety of skin diseases.

Introduction

Cutaneous leishmaniasis (CL) is a significant public health problem caused by Leishmania parasites. Characterized by permanently disfiguring, ulcerative skin lesions, CL is the most common clinical form of leishmaniasis, with ∼1.5 million new cases annually and 12 million people infected worldwide.¹ No CL vaccine exists, and available therapies are often ineffective without multiple rounds of treatment.² Lesion severity is driven by an excessive inflammatory response that prevents healing. Large numbers of circulating immune cells, such as inflammatory monocytes and neutrophils, are recruited to the infection site, promoting host immunopathology and exaggerating inflammation.^3–6 Parasites are skin tropic and largely restricted to the lesion site without extensive dissemination. Specifically, dermal macrophages harbor the parasites. Parasites are controlled by CD4⁺ T cell-produced IFNγ. Notably, even when parasite burdens are controlled, cutaneous lesions often persist, suggesting other factors—such as the host inflammatory response—contribute to disease pathogenesis.^7–12 Due to persistent CL, there is a critical need to develop new host-directed therapeutic strategies that reduce inflammation and promote healing.

Lymphatic remodeling (changes in lymphatic endothelial cells [LECs] or lymphatic vessels [LVs]), including lymphangiogenesis (formation of new LVs) and lymphangiectasia (enhanced LV dilation), occurs during inflammation and dictates lymphatic function. We found that lymphatic remodeling takes place at the dermal lesion after inoculation with Leishmania major (L. major).^13–15 The frequency of LEC proliferation is higher in leishmanial lesions than uninfected skin, suggesting enhanced lymphangiogenesis.¹⁵ Mimicking human leishmanial lesions, the transcription factor hypoxia-inducible factor-1α (HIF-1α), its target vascular endothelial growth factor-A (VEGF-A), and the VEGF-A receptor (VEGFR-2), are elevated in lesions during murine leishmaniasis.^14–18 Furthermore, blocking VEGFR-2 signaling impairs LEC proliferation and increases immunopathology with no effect on parasite load.¹⁵ We identified macrophages as the major cellular source of VEGF-A in lesions and showed that VEGF-A expression requires myeloid aryl hydrocarbon receptor nuclear translocator (ARNT)/HIF-α signaling.¹⁴ Importantly, myeloid ARNT/HIF-α signaling promotes lymphangiogenesis to limit immunopathology, suggesting lymphangiogenesis is required for healing.¹³ Therefore, these data suggest that dermal lymphangiogenesis is required to restrict immunopathology and resolve dermal lesions.

Because lymphangiogenesis attenuates CL severity, we wanted to further characterize LECs in leishmanial lesions. Murine leishmanial lesions subjected to single-cell RNA sequencing (scRNASeq) revealed LECs from infected ears expressed increased guanylate binding proteins (GBPs) (Gbp2, Gbp4, and Gbp7) and a higher abundance of both MHC-I and MHC-II transcripts (H2-K1, H2-D1, H2-Q7, H2-Aa, and H2-Ab1).¹⁹ Cxcl9 was significantly elevated in LECs following L. major infection. Following Ingenuity Pathway Analysis (IPA) comparing LECs from infected skin and uninfected control skin, we identified predicted upstream regulators of LECs including IFNγ inducing activation, while IL-10 downregulates LEC activation upon infection.¹⁹ Predicted transcriptional regulators activating LECs include IRF3, STAT1, and IRF7, and predicted transcriptional regulators inhibiting LECs include MLXIPL, MYC, SIRT1(sirtuin 1), TRIM24 (tripartite motif-containing 24), and MYCL (MYCL proto-oncogene).¹⁹ Based on elevated transcripts B2m and H2-K1, IPA analysis also revealed the antigen presentation and PD1/PD-L1 pathways are increased with infection in LECs.¹⁹ In contrast, IPA analyses showed LECs from infected mice exhibit decreased EIF2 (eukaryotic translation initiation factor 2), mTOR (mechanistic target of rapamycin kinase), and eIF4/p70S6K signaling, hinting at an important role for oxidative stress at the site of infection.¹⁹ These transcriptomic findings indicate that in addition to their function in draining fluid and cells away from the site of infection, dermal LECs participate in the host immune response during CL.

Here, we validate our scRNASeq findings and show that LECs are activated exhibiting a mixed phenotype, characterized by pro-inflammatory molecules such as GBPs and anti-inflammatory co-stimulatory molecules such as PDL-1, during L. major infection in vivo. Our findings contrast with the widely accepted current view that LECs mainly exhibit an immunosuppressive phenotype.^20–29 We also find LECs exhibit enhanced translational activity that is linked to LEC activation. These results suggest that increases in LEC numbers due to proliferation not only help resolve lesions via their vessel drainage functions, but that LECs could participate in anti-parasitic immunity or restricting parasite dissemination. These findings demonstrate for the first time that LEC activation is a feature of the pathogenesis of CL.

Materials and Methods

Mice

Female C57BL/6NCr mice were purchased from the National Cancer Institute. Mice were housed in the Division of Laboratory Animal Medicine at University of Arkansas for Medical Sciences (UAMS) under pathogen-free conditions. Mice between 6 and 8 weeks of age were used for infections. All procedures were performed in accordance with the guidelines of the UAMS Institutional Animal Care and Use Committee.

Parasites and infections

L. major (WHO/MHOM/IL/80/Friedlin) parasites were maintained in vitro in Schneider’s Drosophila medium (Gibco) supplemented with 20% heat-inactivated FBS (Invitrogen), 2 mM L-glutamine (Sigma), 100 U/mL penicillin, and 100 mg/mL streptomycin (Sigma). Ficoll density gradient separation was used to isolate metacyclic stationary phase promastigotes from 4 to 5 days cultures by (Sigma).³⁰ For dermal ear infections in mice, 2 × 10⁶ promastigote parasites in 10 µL PBS (Gibco) were injected intradermally into the ear. For analyses, ears were excised, dorsal and ventral sheet were separated. Ear sheets were enzymatically digested for 90 minutes at 37°C using 0.25 mg/mL Liberase (Roche) and 10 mg/mL DNase I (Sigma) in incomplete RPMI 1640 (Gibco). After digestion, ears were smashed through a filter to obtain a single-cell suspension.¹⁹

Flow cytometry

To assess cell viability, cells from both infected and contralateral ears were incubated with fixable Aqua dye (Invitrogen) or Zombie (Biolegend) for 10 minutes at room temperature. Cells were treated with FcγR blocking reagent (Bio X Cell) and 0.2% rat IgG for 10 minutes at 4° before staining for the following markers: anti-CD4-PE (GK1.5) and anti-CD11c-PE (N418) were purchased from eBioscience; anti-Ly6C-PerCP-Cy5.5 (clone HK1.4), anti-Ly6G-eFlour 450 (clone 1A8), anti-CD8β-eF780 (H35-17.2), anti-CD31-PECy7 (390), and anti-CD45-AF700 (clone 30-F11), were purchased from Invitrogen; anti-CD64-BV711 (clone X54-5/7.1), anti-CD11b-BV605 (clone M1/70), anti-CD31-AF488 (390), anti-MHCII-APCeF80 (M5/114.15.2), anti-PDL1-BV650 (10F.9G2) and anti-podoplanin-PE/Dazzle 594 (clone 8.1.1) were purchased from BioLegend; anti-CD24-BV560 (M1/69) was purchased from BD Biosciences. Surface staining was performed in Brilliant Violet Buffer (BD Biosciences) or super bright staining buffer (eBiosciences). Intracellular cytokine staining was performed using the Foxp3 intracellular staining kit (Life Technologies) combined with α-CD3 APCeF780 (Invitrogen). Cells were acquired using an LSRII Fortessa (BD Biosciences) or Northern Lights (Cytek) flow cytometers and analyzed using FlowJo software version 10.2 (Tree Star).

In vivo translation

To determine the translational activity of LECs in vivo, C57BL/6 mice were infected, and translation was measured by puromycin incorporation using flow cytometry as previously described.^31,32 In brief, mice are injected with 21.8 µg/g of puromycin (Sigma P7255) diluted in PBS at 1 hour prior to euthanization. Puromycin was detected by flow cytometry using an anti-puromycin antibody conjugated to A647 (Sigma MABE343-AF647) after intracellular staining with the Foxp3 kit (Life Technologies).

DQ-ovalbumin (DQ-OVA) antigen uptake and presentation

DQ-OVA (Invitrogen D-12053) is labeled with a pH-insensitive, green fluorescent Bodipy FL dye. For ex vivo assays, ears were digested, processed, and counted. A single cell suspension of ear cells in 500 µL of complete RPMI were cultured in 5 mL Falcon polypropylene tubes at 37°. For antigen uptake assays, cells were cultured with DQ-OVA (10 µg/mL) for 10 minutes. For antigen processing assays, cells were cultured with DQ-OVA (10 µg/mL) for 2 hours. DQ-OVA uptake and processing was analyzed by flow cytometry in the Af488/FITC channel. Cells cultured without DQ-OVA for 2 hours provided fluorescence minus one controls. For antigen presentation assays in vivo, DQ-OVA (20 µg/lesion) was intradermally injected into leishmanial lesions in 20 µL of PBS. After 24 hours, ears were harvested, digested, processed, and stained for flow cytometry to examine DQ-OVA median fluorescent intensity (MFI) in individual cell populations.

scRNASeq samples preparation

The scRNASeq samples were prepared and used as a part of a previous study.¹⁹ In brief, the Arkansas Children’s Research Institute Genomics and Bioinformatics Core prepared Next Generation Sequencing (NGS) libraries from fresh single-cell suspensions using the 10× Genomics NextGEM 3′ assay for sequencing on the NextSeq 500 platform using Illumina SBS reagents. Trypan Blue exclusion under 10× magnification determined cell quantity and viability. Library quality was evaluated with the Advanced Analytical Fragment Analyzer (Agilent) and Qubit (Life Technologies) instruments.

scRNASeq data analysis

Data analysis was carried out as a part of a previous study.¹⁹ Briefly, the UAMS Genomics Core generated demultiplexed fastq files which were analyzed using 10× Genomics Cell Ranger alignment and gene counting software, a self-contained scRNASeq pipeline developed by 10× Genomics. The reads were aligned to the mm10 reference transcriptomes using STAR and transcript counts were generated.^33,34 The Seurat R package was used to process the raw counts generated by cellranger count.^35,36 Potential doublets, low quality cells, and cells with a high percentage of mitochondrial genes were filtered out. Cells that have unique feature counts > 75th percentile plus 1.5 times the interquartile range (IQR) or < 25th percentile minus 1.5 time the IQR were filtered. Similarly, cells with mitochondrial counts falling outside the same range for mitochondrial gene percentage were filtered. After filtering, all eight sequencing runs were merged. The counts were normalized using the LogNormalize method which log-transforms the results.¹⁹ Subsequently, the 2000 highest variable features were selected. The data was scaled, and principal component analysis (PCA) was performed. A JackStraw procedure was implemented to define the significant PCA components that have a strong enrichment of low p value features.

A graph-based clustering strategy was used to embed cells in a graph structure.³⁷ Seurat was used to visualize the results in t-distributed stochastic neighbor embedding and Uniform Manifold Approximation and Projection plots.³⁸ Seurat FindNeighbors and FindClusters functions were optimized to label clusters. Seurat FindAllMarkers function finds markers that identify clusters by differential expression, defining positive markers of a single cluster compared to all other cells and comparing those to known markers of expected cell types from previous single-cell transcriptome studies. Cell type determinations were decided by manually reviewing these results, and some clusters were combined if their expression was found to be similar. We specifically examined the Prox-1⁺ LEC cluster. For the heat maps showing individual differentially expressed genes within a specific pathway in the LEC cluster, the expression log ratio is reported.

Statistics

All data were analyzed using GraphPad Prism 9. A Grubbs’ test was used to identify and mathematically remove outlier data points. Statistical significance was calculated using an unpaired two-tailed Student’s t-test for a single comparison between groups. Statistical significance was also calculated using two-way ANOVA with Tukey’s multiple corrections for multiple comparisons between groups and cell types. The test used is listed in the figure legend. For either test, p ≤ 0.05 was considered statistically significant.

Artificial intelligence or large language models statement

During the preparation of the text for this work throughout July 2024, the authors used ChatGPT 40 mini to improve readability and language. After using this tool, the authors confirm that they have reviewed and edited the content as needed, and they take full responsibility for the content of the publication.

Results

To define the transcriptional signature of LECs during CL, we infected C57BL/6 mice in the ear dermis with L. major parasites and performed scRNASeq on leishmanial lesions.¹⁹ Specifically, we analyzed 13,034 total cells from naive ears and 13,524 cells from infected ears.¹⁹ Among the total cells, here we focus on the LEC cluster. IPA analysis revealed the ‘antigen presentation’ and ‘PD1/PD-L1’ pathways are increased with infection in LECs compared to LECs from naïve uninfected, uninflamed skin.¹⁹ In-depth analysis of the LEC transcriptome showed a higher abundance of beta-2 microglobulin and both MHC-I and MHC-II transcripts (H2-K1, H2-D1, H2-Q7, H2-Aa, H2-Ab1, H2-Eb1, H2-Q4, H2-Q6, and H2-T2) in LECs after infection compared to naïve controls (Fig. 1A). Similarly, we also detect a higher abundance of the invariant chain (Cd74) and immunoproteasome transcripts (Psmb8, Psmb9, and Psme2) in LECs with infection compared to controls (Fig. 1A). We discovered LECs from infected ears expressed increased GBPs (Gbp2, Gbp3, Gbp4, Gbp5, Gbp6, Gbp7, Gbp8, and Gbp9) and transcripts of interferon-inducible genes (Cxcl9, Ifi44, Igtp, Iigp1, Irgm1, and Irgm2) (Fig. 1B).¹⁹ Based on elevated Cd274 transcript (Fig. 1B), IPA analysis also revealed the ‘PD1/PD-L1’ pathway is increased with infection in LECs.¹⁹ While we do not find parasites living inside LECs in lesions,¹⁹ LECs are becoming activated by responding to the inflamed tissue microenvironment at the site of infection by increasing their abundance of pro-inflammatory transcripts, potentially associated with anti-parasitic immunity, as well as increasing immunosuppressive molecules such as PDL-1.

FIG. 1.

Single-cell RNA sequencing shows LEC activation in the lesion during L. major infection. C57BL/6 mice were infected with L. major parasites intradermally in the ear. At 4 weeks infected ears and naïve uninfected control ears were digested and subjected to scRNA sequencing as a part of a previous study.¹⁹ Deeper analysis of the LEC cluster showed 114 transcripts with a significant adjusted p value of < 0.05. Within those top transcripts, 47 are upregulated with infection in LECs compared to LECs from naïve controls. The majority of the 47 upregulated transcripts are associated with antigen presentation (A) or have been identified as an IFN-inducible gene (IRG) (B). The color intensity represents the degree of expression. A red-green color scale was used to reflect the standardized gene expression with red representing high expression and green representing low expression. LEC, lymphatic endothelial cells; L. major, Leishmania parasites.

Transcriptomic findings suggest LECs are becoming activated with infection and undergoing translation to participate in the immune response to L. major parasites.¹⁹ For instance, transcripts of GBPs, MHC, and PDL/PDL-1 molecules are elevated in LECs with infection.¹⁹ To determine the translational activity of LECs in vivo after infection, mice were infected, and at 5 weeks p.i. mice were injected with puromycin.^31,32 Puromycin incorporation as assessed by flow cytometry was used to measure the in vivo translation activity of dermal LECs (Fig. 2A). The translational activity of dermal LECs is significantly higher with L. major infection compared to LECs from the contralateral ear of the same mouse (Fig. 2B, C). Interestingly, even though infection enhanced LEC translational activity, the basal level translational activity of LECs is significantly higher than many other cell types in uninflamed skin (Fig. 2C, D). Of all the cell types analyzed in the skin, LECs and macrophages had the highest translational activity both at the site of infection and in the uninflamed contralateral ear (Fig. 2C, D). In addition to LECs, the translational activity of Ly6C⁺ inflammatory monocytes, macrophages, and blood endothelial cells was significantly increased at the site of infection compared to the uninflamed contralateral ear (Fig. 2C). Of note, T cells had low levels of translational activity; in particular, CD8⁺ T cells had low levels of translational activity that further significantly decreased with infection (Fig. 2C). Altogether, these results suggest that dermal LECs exhibit high translational activity that is further elevated at the site of infection.

FIG. 2.

LECs exhibit increased translational activity following L. major infection. C57BL/6 mice were infected with L. major parasites intradermally in the ear. (A) Representative flow cytometry histogram plot showing controls for puromycin incorporation assay using LECs from an infected mouse. Mice were injected or not with puromycin (puro; 21.8 µg/g in PBS). Gated on live, single, CD45⁻CD31⁺podoplanin⁺ LECs. Gray = Fluorescence minus one (FMO) control for mouse injected with puromycin and stained for puromycin. Dotted line = Infected mouse that was not injected with puromycin but stained for puromycin. Solid line = puromycin-injected mouse and stained for puromycin. (B–D) At 3–5 weeks post-infection when mice exhibited lesions, mice were given puromycin i.p. 60 minutes before euthanizing. Infected and contralateral ears were processed, stained for puromycin, and analyzed by flow cytometry. (B) Representative flow cytometry histogram plot showing puromycin MFI after gating on CD45⁻CD31⁺podoplanin⁺ LECs from the infected or contralateral ear. (C) Quantification of puromycin MFI on endothelial and hematopoietic cell populations by flow cytometry including CD45⁺CD3⁺CD4⁺ T cells, CD45⁺CD3⁺CD8⁺ T cells, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6G⁻Ly6C⁺ inflammatory monocytes, CD45⁺CD11b⁺Ly6G⁻CD64⁺ macrophages, CD45⁻CD31⁺podoplanin⁻ BECs, and CD45⁻CD31⁺podoplanin⁺ LECs from infected or contralateral ears. All cells were pre-gated on total, live, singlets. Data shown here are pooled from two experiments with 10 mice per group in total. Data are presented as the mean +SEM. *p < 0.05, **p < 0.01, ***p < 0.001, t-test comparing between infected and contralateral ears. (D) To compare LECs versus other cell types in leishmanial lesions, two-way ANOVA followed by the post hoc Tukey’s test comparing puromycin MFI between all cell types. LEC, lymphatic endothelial cells; L. major, Leishmania parasites; MFI, median fluorescent intensity.

Transcripts of MHC-I, MHC-II, and immunoproteasome molecules are elevated in LECs with infection, suggesting LECs are participating in antigen presentation and processing.¹⁹ To define the expression of MHC-II protein by LECs at the site of L. major infection, mice were infected, and at 5 weeks p.i., MHC-II expression was evaluated on LECs by flow cytometry. We found a significant increase in the percentage of LECs expressing MHC-II in the infected ear compared to the contralateral ear of the same mouse (Fig. 3A, B). In addition to MHC-II, LEC activation has been associated with elevated PDL-1 expression in melanoma, psoriasis, and viral infection.^28,39 Therefore, we evaluated PDL-1 on dermal LECs following L. major inoculation. At 5 weeks p.i., we discovered a significant increase in the percentage of LECs expressing PDL-1 in the infected ear compared to the contralateral ear of the same mouse (Fig. 3C, D). Taken together, these findings demonstrate that dermal LECs increase MHC-II and PDL-1 expression, suggesting LECs are activated following L. major infection.

FIG. 3.

LECs express elevated MHC-II and PDL-1 during L. major infection. C57BL/6 mice were infected with L. major parasites intradermally in the ear. At 3–5 weeks p.i., MHC-II and PDL-1 expression were evaluated on LECs by flow cytometry. (A) Representative flow cytometry contour plots showing the percentage of dermal LECs expressing MHC-II protein after gating on CD45⁻CD31⁺podoplanin⁺ LECs from the infected or contralateral ear. (B) Quantification of the percentage of LECs expressing MHC-II at the site of L. major infection in the skin. (C) Representative flow cytometry contour plots showing the percentage of dermal LECs expressing PDL-1 protein after gating on CD45⁻CD31⁺podoplanin⁺ LECs from the infected or contralateral ear. (D) Quantification of the percentage of LECs expressing PDL-1 at the site of L. major infection in the skin. All cells were pre-gated on total, live, singlets. Data shown here is a representative experiment of 2–3 individual experiments with at least 4 mice per group. Data are presented as the mean +SEM. *p < 0.05, t-test comparing between infected and contralateral ears. LEC, lymphatic endothelial cells; L. major, Leishmania parasites; MFI, median fluorescent intensity; PD-1, programmed death-ligand 1.

To determine if the activated LECs were also the same cells that exhibited increased translational activity, we evaluated activated LECs for translational activity using PDL-1 as a surrogate for LEC activation. To examine the translational activity, we gated on LECs that were positive or negative for PDL-1 (Fig. 4A), and evaluated the puromycin incorporation in these two subpopulations of LECs (PDL-1⁺ and PDL-1⁻ LECs). In a paired analysis, we found PDL-1⁺ LECs possessed a significantly higher puromycin MFI compared to PDL-1⁻ LECs from the same L. major-infected mouse ear (Fig. 4B–C). This finding suggests that increased translational activity is linked to LEC activation following L. major infection.

FIG. 4.

LEC activation is associated with increased translational activity during L. major infection. C57BL/6 mice were infected with L. major parasites intradermally in the ear. At 3–5 weeks p.i., mice were given puromycin, and PDL-1 expression was evaluated on LECs by flow cytometry. (A) Representative flow cytometry contour plots showing subpopulations of dermal LECs that are PDL-1⁺ or PDL-1⁻ after gating on CD45⁻CD31⁺podoplanin⁺ LECs from an infected ear. (B) Representative flow cytometry histogram plot showing puromycin MFI after gating on PDL-1⁺ LECs or PDL-1⁻ LECs from an infected ear. (C) Quantification of paired puromycin MFIs from the same mouse comparing between PDL-1⁺ LECs and PDL-1⁻ LECs in the skin after L. major infection. All cells were pre-gated on total, live, singlets. The data shown here is a representative experiment of two individual experiments with at least five mice per group. **p < 0.01, paired t-test comparing PDL-1⁺ LECs and PDL-1⁻ LECs. LEC, lymphatic endothelial cells; L. major, Leishmania parasites; MFI, median fluorescent intensity; PD-1, programmed death-ligand 1.

Based on the increase in MHC-II⁺ LECs following infection, we determined the capacity of LECs to uptake and process antigens using DQ-OVA. Mice were infected, and at 5 weeks p.i. ear cells (single cell suspension) were placed in culture ex vivo with DQ-OVA, which resists acidic degradation and assesses antigen processing at 2 hours. LECs were identified by gating after flow cytometric analysis. To confirm the technical approach, we first examined CD11b⁺ dendritic cells (DCs), a major professional APC in leishmanial lesions that also take up DQ-OVA.⁴⁰ We detected a population of CD11b⁺ DCs that was positive for DQ-OVA after 2 hours ex vivo, confirming this assay assesses APC antigen uptake and processing using cells from leishmanial lesions ex vivo (Fig. 5A, B). After 2 hours, the percentage of LECs processing DQ-OVA was also significantly higher from the infected ears compared to LECs from the contralateral ears (Fig. 5C, D). Together, these data suggest LECs from the inflamed site of infection are significantly more capable of taking up and processing antigens than LECs from the uninflamed skin.

FIG. 5.

LECs from L. major-infected ears process antigen ex vivo. C57BL/6 mice were infected with L. major parasites intradermally in the ear. At 3–5 weeks p.i., digested ear cells were exposed to DQ-OVA for 2 hours in culture and evaluated by flow cytometry. (A) Representative flow cytometry contour plots showing the percentage of dermal CD11b⁺ DCs processing DQ-OVA after gating on CD45⁺CD11c⁺MHC-II⁺CD11b⁺ DC from the infected or contralateral ear. (B) Quantification of the percentage of dermal CD11b⁺ DCs from the site of L. major infection that are processing DQ-OVA. (C) Representative flow cytometry contour plots showing the percentage of dermal LECs processing DQ-OVA after gating on CD45⁻CD31⁺podoplanin⁺ LECs from the infected or contralateral ear. (D) Quantification of the percentage of dermal LECs from the site of L. major infection that are processing DQ-OVA. All cells were pre-gated on total, live, singlets. The data shown here is a representative experiment of two individual experiments with five mice per group. Data are presented as the mean +SEM. **p < 0.01, t-test comparing between infected and contralateral ears. LEC, lymphatic endothelial cells; L. major, Leishmania parasites; MFI, median fluorescent intensity; PD-1, programmed death-ligand 1; DQ-OVA, DQ-ovalbumin.

The DQ-OVA studies presented in Figure 5 examine LEC antigen processing by skin cells ex vivo. However, the ability of LECs to process and present antigens in vivo during L. major infection is not known. To address this question, we injected DQ-OVA directly into leishmanial lesions to assess which cells, including LECs, could access and process antigens in their physiological location in the inflamed dermal microenvironment. To confirm the technical approach, we first examined CD11b⁺ DCs, a major professional APC in lesions. At 5 weeks p.i., we detected a population of CD11b⁺ DCs that was positive for DQ-OVA in lesions within 24 hours after DQ-OVA injection, confirming this assay assesses APC antigen uptake and processing in vivo in leishmanial lesions (Fig. 6A, B). Similarly, we detected a population of LECs that was positive for DQ-OVA in lesions suggesting LECs can serve as APCs in vivo in leishmanial lesions (Fig. 6C, D). It should be noted that the LECs are negative for CD11b⁺ and CD11c⁺ and the LECs are gated from singlets (excluding doublets), suggesting the DQ-OVA detected on LECs is not due to professional APCs or myeloid cells that can be intimately associated with the lymphatic vasculature.⁴¹ Altogether, these results show that dermal LECs possess antigen presentation machinery and that these cells can uptake and process antigens during L. major infection.

FIG. 6.

LECs process antigen in vivo during L. major infection. C57BL/6 mice were infected with L. major parasites intradermally in the ear. At 3–5 weeks p.i., DQ-OVA was injected into lesions. After 24 hours, digested ear cells were evaluated by flow cytometry. (A) Representative flow cytometry contour plots showing the percentage of dermal CD11b⁺ DCs processing DQ-OVA after gating on CD45⁺CD11c⁺MHC-II⁺CD11b⁺ DC from the infected or contralateral ear. (B) Quantification of the percentage of dermal CD11b⁺ DCs at the site of L. major infection that are processing DQ-OVA. (C) Representative flow cytometry contour plots showing the percentage of dermal LECs processing DQ-OVA after gating on CD45⁻CD31⁺podoplanin⁺ LECs at the infected or contralateral ear. (D) Quantification of the percentage of dermal LECs at the site of L. major infection that are processing DQ-OVA. All cells were pre-gated on total, live, singlets. The data shown here are a representative experiment of two individual experiments with five mice per group. Data are presented as the mean +SEM. *p < 0.05, t-test comparing between infected and contralateral ears. LEC, lymphatic endothelial cells; L. major, Leishmania parasites; MFI, median fluorescent intensity; PD-1, programmed death-ligand 1; DQ-OVA, DQ-ovalbumin.

Discussion

In CL, lesion severity results from uncontrolled parasite proliferation or an overactive host inflammatory response that impedes the healing process. Factors such as excessive numbers of neutrophils or CD8⁺ T cells, prior or concurrent viral infections, or the microbiome can exacerbate CL immunopathology, thereby increasing lesion size without altering parasite burdens.^42–46 Cumulatively, the body of work from our lab indicates that the lymphatic vasculature also plays a role in immunopathology in CL.^13–15 Specifically, we find lymphangiogenesis mitigates CL severity.^13–15 Building on this foundation, here we find that LECs in leishmanial lesions become activated, increase translational activity, and enhance the ability to present antigens. These findings suggest LECs in leishmanial lesions might participate in the anti-parasitic immune response.

The role of LECs in immunity across inflammatory conditions is an area of active investigation. Currently, the prevailing view is that LECs are immunomodulatory, but most of those studies were performed using tumor models which are characterized by an immunomodulatory microenvironment. Therefore, it remains unclear whether LECs can promote pro-inflammatory immune responses. Our data suggests that in a robust pro-inflammatory microenvironment that is dominated by a Th1 immune response, such as a leishmanial lesion, LECs become activated exhibiting a mixed pro-inflammatory and immunomodulatory phenotype. Because LECs are not infected with parasites,¹⁹ the LEC activation is thought to be a result of the skin inflammatory microenvironment, rather than due to LECs harboring intracellular parasites. In this inflamed microenvironment, macrophages are the predominant cell type infected; however, it is not known how LECs interact with infected macrophages or how much of the inflammatory response is regulated by LECs in lesions. Regardless, we predict our findings are relevant to a variety of settings dominated by skin inflammation, and not restricted to CL.

The dominant perspective of the field is that LECs exert immunomodulatory effects, but the mechanisms of immunomodulation are not completely understood. For instance, LECs can influence immune responses by regulating DC maturation.²² Specifically, LECs can suppress DC activation and function when DCs are matured with TNFα, but not LPS.²² This observation, along with our findings, suggests that LEC-DC interactions at sites of microbial infection differ from those at tumor sites or in general inflammation without microbial involvement. In addition, LECs modulate immune responses through inhibitory co-stimulatory molecules like PDL-1. Because IFNγ upregulates PDL-1 on LECs in culture, it is likely that PDL-1 on LECs in infected skin is induced by IFNγ,²⁸ particularly given IFNγ is highly elevated in leishmanial lesions.^47,48 Alternatively, type I IFNs such as IFNα may be responsible as IFNα also upregulates PDL-1 on LECs in culture,⁴⁹ and IFNα is elevated in leishmanial lesions.⁵⁰ Regardless of the factors influencing PDL-1 expression on LECs, our findings indicate that activated LECs (using PDL-1 positivity as a surrogate for activation) are associated with enhanced translational activity. To our knowledge, this is the first time that LEC translational activity was investigated in vivo. It is noteworthy that LECs were identified as one of the most translationally active cells within leishmanial lesions, even in the absence of inflammation, which were unexpected observations. Future work should focus on whether the enhanced translational activity observed in LECs compared to other cell types is restricted to the skin or if it also occurs in LECs at other barrier or anatomical sites. In addition, future studies will aim to determine if the heightened translational activity in LECs is specific to Leishmania infection or if it is also detected in skin inflammation due to other causes.

Under steady-state, LECs lack MHC-II antigen presentation machinery.⁵¹ However, IFNγ (which is elevated in leishmanial lesions,^52,53) induces MHC-II on LECs, albeit less than other non-professional APCs.^54,55 While our transcriptomic findings suggest at least some MHC-II on LECs is endogenously generated during infection, MHC-II molecules on LECs can also be acquired from hematopoietic cells.⁵⁴ Therefore, it is not clear from our analysis what portion of the LECs expressing MHC-II protein is due to endogenous expression or acquired from immune cells.

Our findings are consistent with a previous report showing that LECs have antigen presentation machinery and immunoproteasomes during inflammation, though their exact role in immunity is not completely clear.⁵⁵

Future work will explore whether LEC antigen presentation boosts the anti-parasitic pro-inflammatory response, or if LEC antigen presentation functions similarly to other skin conditions such as melanoma or viral infection where LECs are predominantly immunomodulatory. In these instances, LECs limit anti-viral immunity and prevent immunopathology.³⁹ In contrast to melanoma and viral infection, we hypothesize that the LECs are taking on a more pro-inflammatory phenotype based on elevated MHC and GBP molecule expression from this work combined with our scRNASeq transcriptomic analyses of LECs after L. major infection. However, the elevated PDL-1 on LECs in CL also hints LECs may be involved in some immunomodulation of the T cell response during L. major infection. Given this discrepancy, a future objective of the lab is to determine whether LECs play a role in promoting anti-parasitic pro-inflammatory immune responses, or if they remain primarily immunomodulatory, as they have been in other conditions. This clarification will be crucial in understanding the role of LECs in immune responses.

Limitations

Here, we compared the transcriptomic profiles of LECs from leishmanial lesions to uninfected uninflamed ears. To the best of our knowledge, this is the first comparison of this cell type following Leishmania infection in vivo. While these findings potentially shed light on LEC activation, there are limitations to these results. For instance, this work was only performed in a resistant C57BL/6 mouse model, and it is not obvious if the results will be replicated in a susceptible BALB/c mouse infected with L. major parasites. Dermal LECs were also only assessed during a specific phase of L. major infection from 3 to 5 weeks p.i. when C57BL/6 mice exhibit lesions. However, over the course of infection, parasite numbers and lesion severity change with time generating a dynamic tissue microenvironment. In resistant C57BL/6 mice, parasites are controlled, and lesions change over time with the inflammatory response lessening as the wound healing response takes over. As the lesion changes over time, LECs also respond to soluble mediators in the tissue microenvironment such as proinflammatory cytokines and growth factors as well as hypoxia. As a result, LECs in the tissue may exhibit variable activation states at different time points following infection depending on factors in the tissue microenvironment. Because lesions were mainly analyzed at 3–5 weeks p.i., it will be necessary to assess other time points following infection when LECs may exhibit altered activation states such as 1–2 weeks p.i. when the IFNγ Th1 response is dominant but lesions are not present, 8–10 week p.i. when lesions are healing and at 12 weeks p.i. when lesions have resolved.

Authors’ Contributions

L.F.: Conceptualization, investigation, formal analysis, validation, visualization, review, and editing. H.R. and A.B.: Investigation, formal analysis, validation, review, and editing. G.V.: Investigation, formal analysis, review, and editing. J.T.B.: Formal analysis, visualization, review, and editing. A.W.: investigation. S.D.B.: Formal analysis, supervision, visualization, review, and editing. T.W.: Conceptualization, funding acquisition, investigation, formal analysis, supervision, validation, visualization, writing original draft, and review and editing. This article was previously published on the BioRxiv preprint server (https://www.biorxiv.org/content/10.1101/2024.08.07.605632v1, doi: https://doi.org/10.1101/2024.08.07.605632).

Author Disclosure Statement

The authors have declared that no competing interests exist.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funders had no role in study design, data analysis, decision to publish, or preparation of the article.

Funding Information

This work was supported by the Arkansas IDeA Network for Biomedical Research Excellence (INBRE) (funded by the National Institutes of Health (NIH) National Institute of General Medical Sciences (NIGMS) Centers of Biomedical Research Excellence Grant P20-GM103429) and the Center for Microbial Pathogenesis and Host Inflammatory Responses (funded by NIH NIGMS Centers of Biomedical Research Excellence Grant P20-GM103625). This publication was also supported in part by funds provided by the National Center for Advancing Translational Sciences of the NIH under awards TL1 TR003109 and UL1 TR003107 for the Systems Pharmacology and Therapeutics NIH T32 training grant GM106999 to Lucy Fry. This work was also supported by the Arkansas INBRE PRO Program summer internship (funded by NIH NIGMS Grant P20-GM103429) to Alexx Weaver. This study was supported by the Arkansas Children’s Research Institute, the Arkansas Biosciences Institute, and the Center for Translational Pediatric Research funded under the National Institutes of Health National Institute of General Medical Sciences (NIH/NIGMS) grant P20-GM121293 and the National Science Foundation Award No. OIA-1946391.