Leishmania donovani infection suppresses Allograft Inflammatory Factor-1 in monocytes and macrophages to inhibit inflammatory responses

Macrophages and monocytes are important for clearance of Leishmania infections. However, immune evasion tactics employed by the parasite results in suppressed inflammatory responses, marked by deficient macrophage functions and increased accumulation of monocytes. This results in an ineffective ability to clear parasite loads. Allograft Inflammatory Factor-1 (AIF1) is expressed in myeloid cells and serves to promote immune responses. However, AIF1 involvement in monocyte and macrophage functions during parasitic infections has not been explored. This study now shows that Leishmania donovani inhibits AIF1 expression in macrophages to block pro-inflammatory responses. Mice challenged with the parasite had markedly reduced AIF1 expression in splenic macrophages. Follow-up studies using in vitro approaches confirmed that L. donovani infection in macrophages suppresses AIF1 expression, which correlated with reduction in pro-inflammatory cytokine production and increased parasite load. Ectopic overexpression of AIF1 in macrophages provided protection from infection, marked by robust pro-inflammatory cytokine production and efficient pathogen clearance. Further investigations found that inhibiting AIF1 expression in bone marrow cells or monocytes impaired differentiation into functional macrophages. Collectively, results show that AIF1 is a critical regulatory component governing monocyte and macrophage immune functions and that L. donovani infection can suppress the gene as an immune evasion tactic.


Leishmania infection correlates with lowered AIF1 expression in splenic macrophages.
Prior studies have shown that Leishmania infections result in impaired macrophage functions and accumulation of Ly6C + monocytes that failed to transition into terminally differentiated populations 7,8 . To evaluate whether AIF1 expression is altered during infections, wild type mice were in vivo challenged with L. donovani via intravenous administration. After 7 days of infection, spleens were harvested from infected and control groups prior to flow cytometric analyses. Results revealed increased frequency of F4/80 + subsets after infection, which is a marker largely used to identify monocytes and macrophages 25 (Fig. 2A). AIF1 expression was predominately found within F4/80 + CD68 + CD11b neg/lo splenic red pulp macrophages. However, in the L. donovani-infected cohorts, AIF1 expression within this subset was markedly decreased from 7.98% ± 2.1 to 2.30% ± 0.6 ( Fig. 2B,C). Concomitantly, levels of L. donovani within the spleen were measured by qPCR to confirm successful infection (Fig. 2D). As follow-up experiments, relative amount of parasites and AIF1 expression within the splenic macrophages were assessed daily post-infection by flow cytometry. Percentages of CellTracker-labeled L. donovani present within the splenic pre-gated CD68 + F4/80 + macrophages revealed increased loads relative to reduced AIF1 expression in the macrophages compared to the non-infected group controls (Fig. 2E). Notably, the increased replication of labeled parasites, particularly after the 9-day mark, can also contribute to the observed decreased percentage at later time points.
Ectopic expression of AIF1 in macrophages restores anti-parasite immunity against Leishmania. As in vivo experiments showed a correlation of L. donovani infection with reduced AIF1 expression, studies next assessed whether infection directly antagonized AIF1 using in vitromodels. To directly address, bone marrow-derived macrophages (BMDM) were employed to study the phenomenon. AIF1 expression was largely restricted to the CD11b + F4/80 + Gr-1 neg CD11c neg BMDM subsets (Fig. 3A). In these committed macrophages, in vitro infection with L. donovani resulted in marked depression of AIF1, as assessed by real time PCR (Fig. 3B). This was in conjunction with reduced levels of IL-6 upon in vitro infection (Fig. 3C). Next, AIF1 was ectopically expressed under a non-repressible promoter in macrophages to assess impact on L. donovani infection (Fig. 3D). Results revealed that overexpression of AIF1 reduced parasite load (Supplemental Fig. 1) and supported increased levels of IL-6 cytokine production (Fig. 3E). Further transcriptomic analyses revealed that over expression of AIF1 inhibited Arg1 gene expression, had no effect on IL-10 and, in turn, resulted in increased iNOS and TNFα levels during L. donovani infection (Fig. 3F). Collectively, these results suggest that AIF1 is responsible for promoting pro-inflammatory responses and that L. donovani can directly antagonize these responses by inhibiting expression of the immunoregulatory gene. AIF1 is important for Ly6C + monocytes conversion into macrophages. Prior studies have shown that L. donovani infection results in accumulation of Ly6C + monocytes in the spleen 7 . Bone marrow infection by the parasite can suppress stromal macrophage functions and alter capacity to regulate hematopoiesis [26][27][28] . Notably, AIF1 is responsible for differentiation of monocytes into dendritic cells 12 . Therefore, to investigate the potential governing role of monocyte differentiation into macrophages, AIF1 was silenced in bone marrow cells prior to M-CSF stimulation in vitro. Transfection with Crispr-Cas9 plasmids carrying gRNA targeting AIF1 (pAIF1) consistently reduced protein expression down to 28% ± 9 compared to scrambled gRNA controls (pControl) www.nature.com/scientificreports/ ( Fig. 4A,B). There was no change in viability or total number of cells upon silencing AIF1. However, flow cytometric analyses revealed an altered distribution of CD11b + F4/80 + and CD11b + F4/80 neg monocyte-macrophage populations within the AIF1 silenced versus control groups (Fig. 4C). Within both populations, silencing of AIF1 resulted in incomplete downregulation of Ly6C expression (Fig. 4D), whereby Ly6C + subsets represent the monocyte pool and Ly6C neg the macrophage groups. Lastly, transcriptomic profiling revealed increased expression of essential monocyte genes C/EBPb and KLF4, with a concomitant decrease in ID2, RelB and IRF4 (Fig. 4E). The expression of the genes PU.1 and Nr4a1 were not affected. Further studies revealed no alteration in p38, ERK1/2, IkB, p65 and p52/p100 phosphorylation signaling cascades (data not shown).
To further evaluate role of AIF1 in monocyte-macrophage lineages, monocytes (CD11b hi CD115 + Ly6C/ G + F4/80 low ) and macrophages (CD11b low CD115 + Ly6C/G neg F4/80 + ) were FACS-sorted from spleen 29,30 (Fig. 5A). Gene expression analyses revealed higher levels of AIF1 in splenic macrophages compared to the monocytes Figure 1. Tissue monocyte-macrophage lineages express AIF1. Spleen sections were prepared from C57BL6 mice (n = 6) to assess AIF1 expression by fluorescence microscopy and flow cytometry. (A) Tissues were cryosectioned at 16 μm and stained with fluorochrome conjugated antibodies to CD11b or F4/80 along with AIF1. DAPI dye was used for nuclear detection. Images were acquired using a standard wild-field fluorescence microscope and presented at 4x magnification. Microscopy data is one representative image of two independent experiments with a total of two mice. Scale bar represents 400 μm. (B) Images were further analyzed in FlowJo software for quantitative analyses. Co-localization of intensity of the Y-channel against the X-channel is plotted to determine co-expression of AIF1 with either CD11b or F4/80. (C) For flow cytometric analyses, spleen tissues were dissociated and stained with fluorochrome conjugated antibodies to CD11b, CD68, F4/80 and CD45 prior to acquisition. Cells were then analyzed for AIF1 and CD68 expression in CD11b neg F4/80 + subsets. Flow cytometric data is representative of four independent experiments. www.nature.com/scientificreports/ ( Fig. 5B). Given moderate levels of AIF1 expression in monocytes, studies next evaluated whether AIF1 expression in monocytes is important for directing differentiation into macrophages ex vivo. CD11b hi CD115 + Ly6C/ G + MHC class II neg monocytes were FACS-sorted from bone marrow prior to Crispr-Cas9-mediated silencing of AIF1 (Fig. 5C). Monocytes were then cultured for 4 days under M-CSF stimulation. On day 2 under M-CSF stimulation, there was a notable twofold increase in Ly6C expression from AIF1-silenced groups compared to controls in the CD11b + F4/80 + population (Fig. 5D). This was also concomitant with higher levels of MHC class II in the CD11b + F4/80 + subsets upon AIF1 silencing; F480 + Ly6C + MHC class II + population has been shown to represent tissue monocytes 31 . By day 4 within the CD11b + F4/80 + population, there was 16.9% ± 2.7 Ly6C + MHC class II neg subsets in AIF1-silenced cohort compared to only 2.79% ± 0.6 in control groups stimulated with M-CSF (Fig. 5E). Furthermore, there was a significant reduction in Ly6C neg MHC class II + subsets in absence of AIF1. Taken together, these studies suggest that in absence of AIF1, subsets of ex vivo-isolated monocyte are restrained from conversion into macrophages under M-CSF stimuli.

Discussion
The major aim of these studies was to evaluate the role of AIF1 in combating Leishmania parasitic infections. Prior reports have shown that L. donovani effectively evades immunity, marked by depressed macrophage proinflammatory response and accumulation of poorly functioning inflammatory monocytes within the spleen. In this report, findings reveal that L. donovani infections in mice correlated with a reduction in AIF1 expression in myeloid cells. However, these findings cannot concretely distinguish whether Leishmania infection triggered infiltration of myeloid cells not yet expressing AIF1 or whether the infection does directly result in downregulation of AIF1 expression. This is particularly pertinent given that monocytes express relatively lower levels of AIF1 expression, and levels are upregulated upon conversion into splenic macrophages. Follow-up in vitro studies did reveal that the parasite can directly antagonize AIF1 expression, which was associated with depressed clearance of pathogen load and impaired monocyte-to-macrophage differentiation. Furthermore, overexpression of AIF1 in Leishmania-infected macrophages in vitro retained elevated IL-6 and TNF-α responses, countering the Percentage of AIF1 + in CD11b +/neg F4/80 + / CD68 + macrophages was concomitantly determined in non-infected versus infected groups. Data shown as mean ± SEM representative of 2 independent experiments with 2 mice per group. Statistical significance was determined by unpaired t-test. *p < 0.05, **p < 0.01 and ns = not significant. www.nature.com/scientificreports/ effect of the parasite on inhibition of immune effector responses. The result also led to reduced Arg1 expression concomitant with increased iNOS levels. Although several studies have highlighted the importance of AIF1 in a variety of other diseases, this is the first report to show the importance of the gene in monocyte-macrophage immune responses and presents a model whereby L. donovani can evade host immunity by inhibiting AIF1. Splenic red pulp macrophages, defined as CD11b lo/neg F4/80 hi CD68 + , as well as bone marrow derived CD11b + F4/80 + CD68 + CD11c neg macrophages generated in vitro under M-CSF stimuli, express high levels of AIF1. Care was employed during flow cytometric analyses to compensate for intrinsic differences in autofluorescence among subsets 32 . Infection of mice with L. donovani in vivo or in vitro infection using BMDM approaches had a correlative reduction in frequency of AIF1 + macrophages. To concretely assess the role of AIF1 in directing immune responses, Crispr-Cas9-mediated silencing of AIF1 in macrophages and monocytes corroboratively revealed suppressed immune responses. Notably, silencing of AIF1 expression within bone marrow-derived cells allowed differentiation up through the Ly6C + monocyte stage in vitro under M-CSF stimuli. However, movement beyond Ly6C + monocytes into terminally differentiated macrophages was partially inhibited. This follows suit with literature reports of increased frequency of Ly6C + monocytes subsets upon L. donovani infection 7 . Furthermore, silencing of AIF1 in sorted Ly6C + monocytes resulted in impaired conversion to macrophages, marked by increased frequency of CD11b + F4/80 − subsets. This may suggest that AIF1 is further involved in reinforcing the inflammatory state of monocytes and that loss regresses the committed differentiation state. Future studies will delineate in vivo how loss of AIF1 in the context of steady-state versus inflammatory settings can affect total cell numbers of monocytes versus macrophages, or whether there is solely a disruption in altered frequency distribution (i.e. monocyte-to-macrophage conversion).
Thus, for establishing a foothold for disease, inhibiting macrophage responses, restricting monocyte-tomacrophage differentiation and disrupting circulating inflammatory monocyte antiparasitic functions are necessary for parasite survival. Thus, L. donovani suppression of AIF1 would serve as an effective immune evasion strategy and the premise does support extensive published work 3,7,11,34 . This is particularly relevant as prior studies have shown that suppression of AIF1 in dendritic cells redirects cognate responder naïve T cells towards www.nature.com/scientificreports/ a regulatory phenotype 13,14 . Additional studies have shown that suppression within hematopoietic progenitors or monocytes impairs ability to differentiate into conventional DC and monocyte-deriv 33 ed DC (under GM-CSF stimuli), respectively 12 . Mechanistically, L. donovani infection has been shown to disrupts PKC downstream signaling activities by interfering with binding of Ca 2+ , which were determined to be defects in kinase activation (as opposed to lowering of total levels of PKC) [35][36][37] . This is a particular important and relevant finding, given that AIF1 has been shown to interact directly with PKC to promote both differentiation and effector immune responses in myeloid cells 12 . Taken together, this works supports the notion that Leishmania parasites can effectively evade immune responses by inhibiting expression of AIF1 to antagonize downstream signaling cascades in monocyte-macrophage lineages, which thereby restrains anti-parasitic immunity. Future studies warrant delineating the intracellular signaling mechanisms that allow Leishmania parasites to directly repress AIF1 expression. Knowledge gained will help to develop small molecule inhibitors that can disrupt Leishmania suppression of AIF1 and thereby increase health outcomes of infected individuals.

Materials and methods
Animals. Mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and housed in pathogen-free facilities at Howard University. C57BL/6 (wild type; WT) male and female mice 8-12 weeks of age were used as a source of bone marrow and spleen. WT mice were additionally used for in vivo challenge experiments. All animal procedures were performed in accordance and approved by the Institutional Animal Care and Use Committee through the Office of Regulatory Research and Compliance at Howard University.

Generation of bone marrow-derived macrophages and monocytes. Macrophages and monocytes
were generated from isolated murine bone marrow cells. Briefly, bone marrow cells from mouse tibias and femurs were passed through a 70 μm nylon mesh to remove debris prior to culturing with RPMI (Thermo Fisher; Grand Island NY) supplemented with 10% fetal bovine serum (FBS; Gibco), 100 U/mL penicillin/streptomycin (Gibco) and 20 ng/mL M-CSF (Peprotech; Rochy Hill NJ), or the equivalent of L929 cell line supernatant, for 6 days in culture. Cells were confirmed macrophages by flow cytometric analyses for CD11b + F4/80 + CD115 + Ly6C/ G neg CD11c neg and monocytes for CD11b + CD115 + Ly6C + markers.
Sorting of Ly6C + monocytes and terminally differentiated F4/80 + macrophages. Spleens harvested from wild type mice were dissociated using the GentleMACS dissociator (Miltenyi) prior to red blood cell lysis treatment and staining with antibodies to MHC class II, Ly6C/G, CD115 (M-CSF receptor), and CD11b. All antibodies were purchased from BioLegend. Cells were then sorted for CD11b hi Ly6C/G + CD115 + MHC class II neg using the BD FACSJazz flow cytometric sorter (BD Biosciences). For macrophage isolation, spleens were dissociated using GentleMACS prior to red blood cell lysis treatment and staining with antibodies to CD11b, Ly6C/G, F4/80 and MHC class II. Cells were then sorted for CD11b lo , F4/80 + , Ly6C/G neg , MHC class II + . obtained from ATCC. For growth and expansion, parasites were inoculated into Novy, Mac Neal and Nicole (NNN) and Schneider's Insect medium (Thermo Fisher) supplemented with 10% heat-inactivated fetal bovine serum and 1% penicillin/streptomycin. Parasites were kept in frozen stocks after only one passage in culture to limit expansion of mutant strains. Leishmania isolates were expanded in supplemented Schneider's medium at 24 °C. Promastigotes were examined daily using light microscopy to determine growth curves. For in vivoinfections, 10-week old mice were infected by intravenous injection of 10 6 stationary-phase parasites in 200 μl of warm PBS 38 . In some studies, Leishmania parasites were pre-labeled with CellTracker staining dye (Thermo Fisher). Intracellular parasite load within cells was determined by pre-gating on target macrophage populations prior to evaluating percentage of CellTracker + Leishmania parasites present by flow cytometry. The approach follows published methods by Silva et al. 1 . Dilution of the dye due to parasite in vivo replication allows for up to 2 weeks for tracking. Control mice received PBS only. Infection was allowed to proceed for 7-14 days. Blood and tissues of the spleen, liver and lungs were harvested after mice were sacrificed; sera were prepared from whole blood. For in vitro infections, bone marrow cells, monocytes or macrophages were infected at a 5:1 ratio of promastigotes to cells prior to incubation for 48-72 h. Parasite load was quantified by preparing DNA from tissues or culture samples prior to performing real-time PCR using a modified approach by Wilson et al 39 .
Antibodies and live/dead staining dye. AIF1  Cryosectioning and fluorescence microscopy. Spleens were fixed and cryosectioned prior to transferring onto L-poly Lysine-coated glass slides. 0.3% Triton-X solution was used to permeabilize the sections followed by blocking with 0.2% BSA. Sections were stained with CD11b, CD64, F4/80, AIF1 or IgG isotype control antibodies. DAPI (Thermo Fisher) was used as a nuclear staining dye. Slides were imaged using the FSX100 fluorescence microscope (Olympus, Waltham MA). Acquired images were then analyzed using ImageJ (Rasband, W.S., ImageJ, U.S. National Institutes of Health) and FlowJo (Flow Jo LLC; Ashland, OR).

Gene expression profiling by quantitative PCR (qPCR).
To evaluate gene expression, cells were harvested and resuspended in Trizol (Thermo Fisher) prior to total RNA extraction. Total RNA was reverse transcribed into single-stranded cDNA using the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher). Flow cytometry. For in vitro cell culture experiments, cells are isolated from plates using physical scraping approaches. Tissue isolation procedures employed the GentleMACS dissociator (Miltenyi) in PBS; no trypsin or collagenase treatments were used. Single suspension of cells were washed with PBS supplemented with 1 mM EDTA and stained with fluorochrome-labeled antigen-specific antibodies or respective isotype controls. For intracellular antibody labeling, cells were fixed in 3% PFA prior to permeabilizing with 0.2% saponin in PBS and staining with antibodies. Cells were then acquired on a BD FACSVerse flow cytometric analyzer (BD Biosciences) and data analyzed using FlowJo (FlowJo LLC; Ashland, OR). All gates for dot plots and histograms were established using appropriate isotype controls.

Western blot analysis.
Lysates were prepared using NP-40 lysis buffer supplemented with a proteasephosphatase inhibitor cocktail (IBI Scientific). Lysates were ran on a 10% SDS-PAGE gel prior to transferring to nitrocellulose blots using semi-dry transfer system (PowerBlotter; Thermo Fisher). Membranes were stained with primary antibody prior to wash and secondary antibody labeling. Protein bands within blots were then www.nature.com/scientificreports/