Interdependencies between Toll‐like receptors in Leishmania infection

Abstract

Multiple pathogen‐associated molecular patterns (PAMPs) on a pathogen's surface imply their simultaneous recognition by the host cell membrane‐located multiple PAMP‐specific Toll‐like receptors (TLRs). The TLRs on endosomes recognize internalized pathogen‐derived nucleic acids and trigger anti‐pathogen immune responses aimed at eliminating the intracellular pathogen. Whether the TLRs influence each other's expression and effector responses—termed TLR interdependency—remains unknown. Herein, we first probed the existence of TLR interdependencies and next determined how targeting TLR interdependencies might determine the outcome of Leishmania infection. We observed that TLRs selectively altered expression of their own and of other TLRs revealing novel TLR interdependencies. Leishmania major—an intra‐macrophage parasite inflicting the disease cutaneous leishmaniasis in 88 countries—altered this TLR interdependency unfolding a unique immune evasion mechanism. We targeted this TLR interdependency by selective silencing of rationally chosen TLRs and by stimulation with selective TLR ligands working out a novel phase‐specific treatment regimen. Targeting the TLR interdependency elicited a host‐protective anti‐leishmanial immune response and reduced parasite burden. To test whether this observation could be used as a scientific rationale for treating a potentially fatal L. donovani infection, which causes visceral leishmaniasis, we targeted the inter‐TLR dependency adopting the same treatment regimen. We observed reduced splenic Leishman–Donovan units accompanied by host‐protective immune response in susceptible BALB/c mice. The TLR interdependency optimizes TLR‐induced immune response by a novel immunoregulatory framework and scientifically rationalizes targeting TLRs in tandem and in sequence for redirecting immune responses against an intracellular pathogen.

Toll‐Like receptors (TLRs) modulate each other expression—a phenomenon termed as TLR interdependency. Each TLR shows a specific profile of interdependency. TLR interdependency can be targeted to devise immunotherapeutic strategy in various diseases, for example infectious diseases, tumours and autoimmune diseases.

Abbreviations

BPP

pegylated bisacycloxypropylcysteine (BPPcysMPEG)

L. major

Leishmania major

PGN

peptidoglycan

TLR

Toll‐like receptor

INTRODUCTION

The Toll‐like receptor in Drosophila was the first observed to confer protection against fungal infections [1, 2, 3. The observation was followed by the discovery of its homologs as Toll‐like receptors (TLRs) in mouse and human being [4, 5. TLRs recognize the pathogen‐associated molecular patterns (PAMPs) and elicit the innate immune responses eventuating in pathogen elimination. Multiple TLRs cover wide varieties of PAMPs. Mice express twelve TLRs except TLR10, and humans express ten TLRs except TLR11, TLR12 and TLR13. TLR1, TLR2, TLR4, TLR5, TLR6 and TLR11 are present on the cell membrane, whereas TLR3, TLR7, TLR8 and TLR9 are intracellular endosome‐expressed TLRs. TLR1‐TLR2 and TLR2‐TLR6 heterodimers recognize triacylated lipoproteins and diacylated lipoproteins, respectively. TLR2, TLR3, TLR4, TLR5, TLR7‐TLR8, TLR9 and TLR11 bind with peptidoglycan (PGN), ssRNA, lipopolysaccharide (LPS), flagellin, dsRNA, CpG‐ODN and profilin, respectively [6, 7. Macrophages that play host to the protozoan parasite Leishmania express these TLRs, trigger innate anti‐leishmanial responses and serve as antigen‐presenting cells eliciting anti‐leishmanial T‐cell responses.

Transmitted by sand fly, Leishmania major inflicts the disease cutaneous leishmaniasis. Leishmania has a dimorphic life cycle: extracellular, motile, flagellate promastigotes in the vector and intracellular, sessile, aflagellate amastigotes within mammalian macrophages wherein the parasite divides by binary fission [8, 9. Leishmania modulates macrophage functions by regulating the CD40 and TLR responsiveness and CD40‐TLR crosstalk [8]. Leishmania reduced CD40‐induced p38MAPK phosphorylation and TLR9 expression in macrophages [8], whereas TLR9 activation diminished CD40‐induced ERK1/2 phosphorylation [10]. Leishmania‐derived PAMPs such as lipophosphoglycan (LPG) signalled through TLR2 and enhanced parasite survival in macrophages by reducing TLR9 expression in IL‐10‐ and TGF‐β‐dependent manner [11]. TLR2 thus influences the expression and functions of intracellular TLR9 affecting pathogen survival or clearance. The LPG‐expressing but not the LPG‐deficient L. major up‐regulated TLR11 expression; blocking LPG or silencing TLR11 reduced the parasite burden [12]. Therefore, we hypothesized that the pathogens were first recognized by the cell membrane TLRs and that signals from cell membrane‐located TLRs influenced the expression of intracellular TLRs, which recognize the pathogen‐derived nucleic acids after the degradation of the pathogen within host cells. These observations imply that multiple TLRs are likely to be engaged in tandem for pathogen recognition and elimination. Once the pathogens are recognized by the cell surface TLRs, the intracellular signal may modulate the expression of intracellular TLRs to deactivate the anti‐leishmanial response ensuring parasite survival. Merging these two lines of arguments, we propose that an intracellular pathogen both concurrently and sequentially engage multiple TLRs influencing the ensuing immune response through an unknown immunoregulatory framework of TLR interdependency. We also tested whether TLR interdependency could serve as a scientific rationale for formulating a novel anti‐leishmanial therapy.

MATERIALS AND METHODS

Mice, parasite and peritoneal macrophages

BALB/c mice (The Jackson Laboratory, Bar Harbor, ME) were bred and maintained in the Institute's experimental animal facility. Permission from Institutional Animal Care and Use Committee was secured before all animal usage. For L. major (strain MHOM/Su73/5ASKH) and L. donovani (strain MHOM/In83/Ag83) cultures, a complete RPMI‐1640 medium—supplemented with 10% FCS, 50 μM 2‐mercaptoethanol, 100 units/ml penicillin, 100 μg/ml streptomycin, 1 μM sodium pyruvate—was used. The parasites were passaged through BALB/c mice to maintain the virulence.

Thioglycolate (3%, 2 ml) was injected (i.p) into mice for isolation of peritoneal macrophages. 5 days later, mice were killed and peritoneal exudate cells were harvested by 18‐gauge needle in sterile 1x PBS. Macrophages were cultured in RPMI‐1640 supplemented with 10% FCS. After 6 h, non‐adherent cells were washed out and adhered cells were incubated for 66 h at 37° containing 5% CO₂ and humidified atmosphere. Pam₃CSK₄, PGN, PolyI:C, LPS, flagellin, FSL‐1, imiquimod and CpG‐ODN1826 were purchased from InvivoGen (San Diego, CA). Recombinant profilin was from Alexis (San Diego, CA). BPPcysMPEG was synthesized at the Helmholtz Centre for Infection Research (Braunschweig, Germany). The primers for TLRs and cytokines (Table S1) were purchased from Integrated DNA Technologies (San Diego, CA).

Leishmania major infection of BALB/c‐derived macrophages

As previously described [13], macrophages were infected with stationary phase L. major promastigotes at a ratio of 1:10; extracellular parasites were washed out after 6 h. Macrophages were further incubated for 66 h and were left untreated or treated with different doses of TLR ligands for 3, 6 and 12 h before terminating the experiment. The cells were harvested for analysing TLR expression by qPCR.

Leishmania infection and in vivo treatment of mice

For in vivo experiments, 2 × 10⁶ stationary phase L. maj r promastigotes (in 50 μl HBSS) were used to subcutaneously infect BALB/c mice (n = 5) in the hind footpad. Mice were treated with different combinations of TLR ligands post‐infection. In some groups, mice were injected (s.c) with lentivirus (5 × 10⁶ transduction units (TU)) expressing shRNA for the indicated TLRs and control shRNAs followed by L. major infection (2 × 10⁶ parasite/mouse). TLR11 shRNA on 3rd day and TLR2 shRNA on 25th day were injected into the same footpad post‐infection. For phase‐specific combinatorial treatment, some mice were treated with TLR ligands at specific time‐points after infection. Disease severity was assessed by measuring footpad swelling and parasite burden as described earlier [14].

In case of L. donovani infection, BALB/c mice (n = 5) were intravenously infected with 2 × 10⁷ L. donovani two days prior to lentivirally expressed TLR11shRNA or control shRNA treatment followed by treatment with BPPcysMPEG (1 μg/mouse) on 7th and 9th day or CpG (10 μg/mouse) on 11th day that was also followed by TLR2shRNA on 21st day or control shRNA in different combinations or were left untreated. Mice were observed till day 28 and euthanized to study the parasite load and anti‐leishmanial immune response. Parasite load in treated and untreated mice was enumerated following the stamp‐smear method. The splenic smears were fixed with chilled methanol followed by staining with Wright's Giemsa stain. Leishman–Donovan units (LDU) were calculated by multiplying the numbers of parasites/thousand cells with spleen weight.

Reverse transcriptase–PCR

TRI reagent (Sigma‐Aldrich, St. Louis, MO) was used for extraction of total RNA as per the manufacturer's guidelines. For cDNA synthesis, 2 μg RNA from each sample was added with 0·6 μg random primer in a 15 μl volume (Master Mix I) and kept it for 5 min at 65°. Master Mix I was incubated with 10·0 μl of Master Mix II containing 1× RT buffer, 0·1 M DTT, 10 mM dNTPs, 40 U RNase inhibitor and 200 U MMLV (Moloney murine leukaemia virus–reverse transcriptase). cDNA synthesis was carried out at 37° for 1 h followed by heat inactivation of the enzymes at 68°. Using PCR, cDNA from each sample was amplified as described previously [13].

The real‐time PCR was performed in 10 μl reaction mixture containing 10 ng of cDNA as template, 2 ng each of the forward and reverse primer and 5 μl of SYBR Green (Takara Bio. Inc., Shiga, Japan) using thin‐wall 0·2‐ml strip tubes (Axygen, Union City, CA). Quantitative real‐time PCR was performed on the Eppendorf RealPlex4 Mastercycler or Applied Biosystems Step One Plus under the following condition: 95° for 2 min, 40 cycle of 95° for 1 min, 60° for 30 s and 72° for 35 s. Relative quantitation was performed using the comparative threshold (ΔΔCT) method. The mRNA expression levels of the target genes were normalized to GAPDH expression and calculated as relative fold change over the untreated control.

Production of lentiviral particles for TLR1, TLR2, TLR6, TLR11 and control shRNAs

Lentivirus packaging mix, arrest in transfection reagent, TLR2 and TLR11 shRNA (GenBank accession number NM_011905 and NM_205819) in pGIPZ lentivirus vector and TLR1 and TLR6 shRNA (GenBank accession number NM_030682 and NM_011604) in pLKO.1 lentivirus vector were purchased from Open Biosystems (Huntsville, AL). These TLRs shRNA and control shRNA were packaged in trans‐lentiviral packaging system (replication‐incompetent HIV‐1‐based lentivirus; Open Biosystems) as per the manufacturer's guideline. HEK293T cells (5.5 × 10⁶) were seeded in culture plate (100 mm) for overnight. 16 h later, 28·5 μg of trans‐lentiviral packaging mix was added to 9·0 μg TLR shRNA plasmid DNA or 9·0 μg of control shRNA plasmid DNA. Transfection was carried out in a serum‐free media using Arrest‐In transfection reagent (Open Biosystems) at a 1:5 DNA/transfection reagent ratio. After 6 h, medium was replaced with complete growth medium (DMEM supplemented with 10% FCS), and cells were incubated in a CO₂ incubator at 37°. Virus containing supernatant was harvested after 48 h and 72 h, filtered through a 0·45‐mm filter (Millipore) and concentrated 100‐fold.

Lentiviral transduction in macrophages in vitro

BALB/c‐derived macrophages were transduced with TLR shRNA‐Lv or control shRNAs‐Lv for 8 h in RPMI‐1640/0·5% FCS/polybrene (8 μg/ml). Macrophages were washed to remove residual viral particles. After 48 h, uninfected and L. major‐infected cells were treated with TLR ligands for 6 h for assessing the silencing of the target genes by RT‐PCR.

Macrophage‐T‐cell co‐culture

BALB/c‐derived peritoneal macrophages were infected with L. major promastigotes at a ratio of 1:10 for 6 h. The extracellular parasites were washed out. The popliteal lymph node was collected from different groups of mice. Infected macrophages were co‐cultured with lymph node CD4⁺ T cells at 1:3 ratio (macrophage:T cells) for 66 h. Macrophages were washed, fixed with chilled methanol and Giemsa‐stained, followed by parasite counting under a light microscope (E‐600; Nikon, Japan).

Fluorescence‐assisted cell sorting (FACS) analysis

Anti‐CD4PB (558107), anti‐CD3 FITC (555274), anti‐CD25 APC Cy‐7 (557658), anti‐Foxp3 Alexa Fluor 647 (560402) and anti‐IL‐10 PE (554467) were purchased from BD Pharmingen (San Diego, CA). Anti‐CD127 PerCP‐Cy5.5 (121114) was purchased from BioLegend (San Diego, CA).

For multicolour FACS analyses, cells were stained (after blocking with Fc blocking Abs) for surface molecules and TLRs with fluorescently labelled antibodies for 45 min at 4° in dark and washed twice with FACS buffer (1× PBS, 10 mM HEPES buffer and 3% FCS). Intracellular cytokine staining was performed using a Cytofix/Cytoperm Plus Kit with GolgiPlug (555028; BD Pharmingen), as per the manufacturer's instructions. Cells were acquired in the CD3⁺ CD4⁺ CD25⁺ CD127^dim Foxp3⁺ IL‐10⁺ gate for Treg cell analysis by FACSCanto II flow cytometer (BD Biosciences) and analysed using BD FACSDiva software (version 5.2; BD Biosciences). Isotype controls were prepared by staining cells with isotype‐matched fluorochrome‐tagged antibodies and were analysed while applying gates.

Western blotting

Leishmaniamajor‐infected and uninfected macrophages were treated with the indicated doses of TLR ligands for 24 h, washed twice with chilled PBS and lysed with lysis buffer [20 mM Tris‐HCl (pH 7·5), 150 mM NaCl, 1 mM EGTA, 1 mM EDTA, 10% glycerol, 1% NP‐40 and protease inhibitor]. Protein was quantified using Bradford's reagent, equal amount was loaded and run on SDS‐PAGE. The resolved proteins were blotted to PVDF membrane. After blocking with 5% non‐fat dried milk powder in TBST buffer, membranes were incubated with primary antibody at 4° overnight, washed thrice with TBST buffer (2·42 gm Tris base, 8·0 gm NaCl and 0·1% Tween‐20) and incubated with HRP‐conjugated secondary antibody for 45 min at room temperature. Immuno‐reactive bands were visualized by Luminol reagent and blotted on a X‐ray film.

Nitrite generation

Nitrite levels in culture supernatant of TLR ligand‐treated, L. major‐infected and uninfected macrophages were quantified using the Griess reagent as per the manufacturer's protocol. 50 μl cell‐free supernatants were mixed with 50 μl Griess reagent in a 96‐well plate and incubated at room temperature for 5 min. ODs were measured at 540 nm. ODs of the unknown samples were compared with the ODs of the serially diluted NaNO₃ solution, and nitrite (μM) was expressed.

Data processing, analysis and plotting

The dose‐ and time‐dependency data for all the applied ligand types were subject to log2 transformation. We further considered a fold change observed in the range 0·5 to 2 (−1 to 1 in log2 scale) to be in the range of measurement error and reassigned them the value of 0 [log2(1)] indicating no changes in expression due to stimulation. For simplicity of visualization, fold changes in the log2 axis were assigned the values between +5 and −5. Custom‐written MATLAB scripts were used to process, analyse and plot the data in the form of a heatmap. The data were later imported to Cytoscape [15] for plotting the weighed networks of inter‐TLR dependencies (Figure 6).

FIGURE 6.

Alteration of inter‐TLR networks in uninfected and L. major‐infected macrophages. For simplicity of visualization, TLR fold change values from (a–d) uninfected macrophages and (e–h) L. major‐infected macrophages were reassigned with +5 or −5 log scales and plotted using Cytoscape. The negative regulations are shown with red colour and positive regulations with blue colour. Regulations at 3, 6 and 12 h are shown with dotted, dashed and solid lines, respectively

Method for transcription factor assay (TranSignal™ protein/DNA array)

Murine peritoneal macrophages were lysed after incubated with TLR ligands for 45 min. The nuclear fraction was separated from cytoplasmic fraction by using Panomics Nuclear Extraction Kit (Cat No. AY2002) as per the manufacturer's instruction. The concentration of nuclear protein was determined with the bicinchoninic acid protein assay kit (Pierce, Rockford, Michigan, USA). Equal amount of nuclear extracts was incubated with biotin‐labelled DNA binding oligonucleotides to allow formation of protein/DNA complexes according to the protocol suggested by the manufacturer of TranSignal Protein/DNA Spin Combo Array Kit (Cat No. MA1215, Panomics Inc. Fremont, CA). The protein/DNA complexes were separated from the free probes using a spin column separation system (Panomics). The bounded probes were eluted and hybridized to a Protein/DNA Combo Array spotted with 345 different consensus‐binding sequences following a protocol suggested by the manufacturer. Each spot on the array corresponds to a specific transcription factor. Horseradish peroxidase‐labelled streptavidin‐based chemiluminescence signals were detected with ECL substrate using ImageQuant LAS 4000 (GE Healthcare) imaging system. Relative spot intensities were determined using Quantity‐One software (Bio‐Rad). Expression of every transcription factor (stimulated group) was normalized by its respective transcription factor (unstimulated group). Densitometric unit‐based comparison was performed, and any spot with a threefold densitometry unit change was considered as significant.

Statistical analyses

All in vitro studies were done thrice. For in vivo experiments, at least seven mice were used per experimental group. The data shown as mean ± standard error (SE) are from one representative experiment. The significance of the data was calculated using one‐way ANOVA (Holm–Sidak method), and values generated P < 0·05 (*), P < 0·01 (**), P < 0·001 (***) signify the differences between the mean values from the control and the experimental group.

RESULTS

In order to address the interdependencies among the TLRs, we followed the following experimental approaches: firstly, examining the profile of alterations in the expression of all TLRs in response to the treatment of BALB/c‐derived macrophages—uninfected or L. major‐infected—treated with ligands to each TLR; secondly, systematically analysing and simplifying the enormous and hugely complex dose‐dependent and kinetic TLR interdependency; thirdly, influencing each TLR ligand on parasite load in macrophages, iNOS expression and NO production; fourthly, based on the key interdependencies derived from the analyses, validating the observations further by studying the effect of silencing of the target influencing TLRs—alone or in combinations; fifthly, in an L. major infection model, examining the anti‐leishmanial efficacy of targeting the experimentally derived TLR interdependencies in a susceptible BALB/c mice; and finally, testing the same principle in L. donovani infection even without working out the TLR interdependency in this particular infection, validating the strength of this novel principle of anti‐leishmanial therapy.

Defining inter‐TLR dependency in macrophages and its rewiring by L. major

We infected BALB/c‐derived thioglycolate‐elicited macrophages with L. major. We treated uninfected macrophages (UIM) and infected macrophages (IM) with the indicated doses of the TLR1/2 ligand (Pam₃CSK₄), TLR2 ligand (PGN) and TLR2/6 ligand (FSL) for the indicated periods and examined the expression of all TLRs (Figures 1, 2, 3). After treatment with Pam₃CSK₄, we observed up‐regulation of TLR1 and TLR2 in UIM and IM suggesting positive autoregulatory effects. TLR3 appeared to be non‐significantly reduced in UIM and IM. Pam3CSK4 reduced TLR4 and TLR5 expression. TLR6 and TLR7 showed similar alterations. While TLR8 and TLR9 expression in UIM and IM was reduced, TLR11 expression was found to be up‐regulated in IM as compared to UIM (Figure 1).

FIGURE 1.

Thioglycolate‐elicited BALB/c‐derived macrophages were infected with L. major promastigotes at a ratio of 1:10 for 6 h followed by washing of the extracellular parasites. After incubation for further 63, 60 or 54 h, cells were treated with TLR1/2 ligand (Pam₃CSK₄—10, 50 and 100 ng) or left untreated till 72 h of total incubation. Cells were lysed, and subjected to RNA isolation and cDNA preparation for qPCR analyses. GAPDH was used as internal control to normalize the value of gene of interest. Data shown represent the relative fold change of TLRs by using ΔΔC_T method, compared with untreated controls. Filled and open circles show the expression of TLRs in uninfected and infected macrophages, respectively. Experiments were performed three times, and the data are shown as mean ± SE. *P < 0·05, **P < 0·01 and ***P < 0·001

FIGURE 2.

BALB/c‐derived peritoneal macrophages were infected with L. major promastigotes at a ratio of 1:10 for 6 h followed by washing of the extracellular parasites. After incubation for further 63, 60 and 54 h, cells were treated with different doses of TLR2 ligand PGN (1 μg, 5 μg and 10 μg) or left untreated for indicated time duration. RNA was isolated for the synthesis of cDNA. Real‐time PCR was performed from cDNA templates for the amplification of TLRs. GAPDH was used as a reference control to normalize the value of gene of interest. Data shown represent the relative fold change of TLRs by using ΔΔCT method, compared with untreated controls. Filled and open circles show the expression of TLRs in uninfected and infected macrophages, respectively. Experiments were performed three times, and the data are shown as mean ± SE. *P < 0·05, **P < 0·01 and ***P < 0·001

FIGURE 3.

The elicited BALB/c‐derived macrophages were infected with L. major promastigotes at a ratio of 1:10 for 6 h followed by washing of the extracellular parasites. After incubation for further 63, 60 and 54 h, cells were treated with different doses of TLR2/6 ligand (FSL—10, 50 and 100 ng) or left untreated. RNA was isolated for the synthesis of cDNA. Real‐time PCR was performed from cDNA templates for the amplification of TLRs. GAPDH was used as a reference control to normalize the value of gene of interest. Data shown represent the relative fold change of TLRs by using ΔΔCT method, compared with untreated controls. Filled and open circles show the expression of TLRs in uninfected and infected macrophages, respectively. Experiments were performed three times, and the data are shown as mean ± SE. *P < 0·05, **P < 0·01 and ***P < 0·001

PGN treatment of UIM and IM enhanced the TLR1, TLR2 and TLR6 expression. TLR6 remained unaltered in IM. TLR3 expression was significantly reduced in infected macrophages, but no further changes were observed after the PGN treatment. TLR4 expression remained largely unaffected. TLR5 expression was reduced after the PGN treatment in both UIM and IM. TLR7, TLR8 and TLR9 expression remained unaltered. PGN treatment significantly reduced TLR11 expression (Figure 2).

With FSL treatment, the expression of TLR1 and TLR2 was enhanced in UIM and IM. TLR3 expression remained mostly unaltered with FSL treatment. We observed reduced TLR4 and TLR6 expression in UIM and IM upon treatment with FSL. UIM showed high TLR5 and TLR9 but not TLR7 and TLR11 expression. TLR8 expression was increased (Figure 3). These observations suggest that TLR2 family down‐regulates TLR9 expression but up‐regulates TLR11 expression and plays self‐regulatory roles during L. major infection.

Repeating similar experiments with the ligands for intracellular TLRs—TLR3 (PolyI:C; Figure S1), TLR7/8 (Imiquimod; Figure S2) and TLR9 (CpG; Figure 4)—led to the observations that TLR3 up‐regulated its own expression and also TLR7 and TLR9 expression and that imiquimod up‐regulated TLR1 in UIM and TLR2 expression in both UIM and IM but significantly down‐regulated TLR1 in TLR8 and TLR11 expression in both UIM and IM, in particular, with higher doses. CpG‐treated UIM and IM showed increased expression of TLR5, TLR8 and TLR9 early after CpG treatment but had higher TLR2 and TLR3 expression later (data not shown).

FIGURE 4.

L. major‐infected macrophages were treated with different doses of CpG‐ODN (0·0625, 0·125 and 0·25 μm) for indicated time periods or left untreated. RNA was isolated for the synthesis of cDNA. Real‐time PCR was performed from cDNA templates for the amplification of TLRs. GAPDH was used as a reference control to normalize the value of gene of interest. Data shown represent the relative fold change of TLRs by using ΔΔCT method, compared with untreated controls. Filled and open circles show the expression of TLRs in uninfected and infected macrophages, respectively. Experiments were performed three times, and the data are shown as mean ± SE. *P < 0·05, **P < 0·01 and ***P < 0·001

Continuing to examine the effects of TLR4, TLR5 and TLR11 ligands on the TLRs interdependency, we observed that TLR4 expression was reduced, but TLR11 expression was augmented, in L. major‐infected macrophages (Figures S3 and S4). LPS enhanced TLR1, TLR2 and TLR3 expression in both UIM and IM; TLR6, TLR7, TLR8 and TLR9 expression was augmented in UIM at all time‐points but TLR6, TLR7 and TLR8 and TLR9 only in IM (Figure S3). L. major infection increases TLR5, but the TLR5 ligand flagellin reduced the expression (Figure S4). Flagellin induced high TLR1 and TLR2 in UIM early after the treatment but increased TLR2 expression in IM later. For other TLRs, flagellin did not exert a consistent effect (Figure S4). Although TLR11 expression increased in L. major infection, profilin did not exert any significant up‐regulation of TLR11 (Figure 5). Profilin enhanced TLR2 expression in IM and reduced TLR9 expression late after Leishmania infection (Figure 5).

FIGURE 5.

BALB/c‐derived, thioglycolate‐elicited, peritoneal macrophages were plated and incubated for 24‐h resting. The cells were infected with stationary phase L. major promastigotes at a ratio of 1:10 for 6 h, and extracellular parasites were washed. Cells were further incubated for 48 h and treated with TLR11 ligand (profilin—125, 250 and 500 ng) or left untreated for indicated time duration. Cells were lysed for total RNA isolation, and cDNA was prepared for real‐time PCR analysis for TLR expression. Filled and open circles show the expression of TLRs in uninfected and infected macrophages, respectively. Experiments were performed three times, and the data are shown as mean ± SE. *P < 0·05, **P < .01 and ***P < .001

As comprehensible from the above results, the dose‐ and time‐dependencies of one TLR affecting the expression of other TLRs were hugely complex. Therefore, these data were subjected to normalization and logarithmic transformation for systematic and quantitative assessment of plausible inter‐TLR dependencies. The analyses highlighted the key selective interdependencies between the TLRs and rewiring of this interdependency by L. major (Figure S5).

Mapping the profiles of TLR interdependencies in macrophages

Detailed analyses of the inter‐TLR dependencies were categorized into four clusters: cell membrane‐located TLRs modulating the expression of (1) other cell membrane TLRs and (2) endosomal TLRs, and endosomal TLRs altering the expression of (3) cell membrane TLRs and (4) intracellular TLRs in uninfected (Figure 6a–d) and in L. major‐infected macrophages (Figure 6e–h). We observed TLR2 up‐regulation but TLR4 down‐regulation by most of the TLRs in both uninfected (Figure 6a,c) and L. major‐infected (Figure 6e,g) macrophages identifying two most common regulatory targets. In uninfected macrophages, TLR1, TLR2, TLR3, TLR6 and TLR7/8 expressions were up‐regulated, whereas TLR4, TLR5, TLR9 and TLR11 expressions were down‐regulated by their own ligands (Figure 6a–d). By contrast, in L. major‐infected macrophages, the ligand‐induced up‐regulated expression of TLR6 and TLR7/8 or down‐regulation of TLR4, TLR9 and TLR11 expression was not observed (Figure 6e–h). In uninfected macrophages, TLR1/2 ligand Pam₃CSK₄ reduced TLR6 expression, whereas the TLR2/6 ligand FSL increased the TLR1 and TLR2 expression (Figure 6a) suggesting a smaller, potentially self‐regulatory module within the inter‐TLR network. TLR4 reduced TLR1 expression in uninfected macrophages (Figure 6a). This effect implied a positive regulatory loop for inflammation, because both TLR4 and TLR2‐TLR6 are pro‐inflammatory and inhibition of TLR1 by TLR4 withdraws the TLR1’s inhibitory effect on TLR6 [13]; TLR5 inhibited TLR6 but up‐regulated TLR11 expression (Figure 6a). TLR2/6 and TLR11 reduced TLR9 expression, whereas TLR9 reduced TLR6, but enhanced TLR2, expression (Figure 6b). In L. major‐infected macrophages, TLR4 and TLR11 up‐regulated TLR7 expression (Figure 6f), but TLR7 down‐regulated TLR11 (Figure 6g) suggesting a negative feedback loop between TLR7 and TLR11. In both uninfected and L. major‐infected macrophages, TLR3 increased the expression of TLR7 (Figure 6d), whereas TLR7 reduced, but TLR9 enhanced, TLR8 expression only in L. major‐infected macrophages (Figure 6h). In uninfected macrophages, TLR4 up‐regulated TLR3 expression (Figure 6b), whereas TLR3 enhanced, but TLR7 reduced, TLR11 expression (Figure 6c) implying differential regulation of co‐infection with the TLR11 ligand profilin‐expressing uropathogenic bacteria in double‐stranded RNA (a TLR3 ligand) virus‐infected or single‐stranded RNA (a TLR7/8 ligand) virus‐infected cells. In the uninfected macrophages, TLR3 up‐regulated TLR7 and TLR9 expression (Figure 6d) and TLR9 up‐regulated TLR3 expression suggesting a positive feedback loop between TLR3 and TLR9 (Figure 6d). These observations suggest that both positive and negative autoregulatory motifs are present in the inter‐TLR network and that compared with the inter‐TLR dependencies in uninfected macrophages, there was substantial rewiring of TLRs resulting in differences in the TLR interdependencies in L. major‐infected macrophages. For example, the autoenhanced TLR6 and TLR7/8 expression or TLR3‐induced TLR9 up‐regulation in uninfected macrophages was not detectable in the L. major‐infected macrophages (Figure 6e,h). These findings suggest Leishmania altered TLR interdependencies that may (1) influence TLR‐dependent co‐infection profiles, (2) provide a mechanism to shift different TLR signalling through MyD88‐dependent or MyD88‐independent pathways or (3) serve as a scientific rationale for multiple TLR‐targeted therapy.

TLRs differentially modulate the iNOS expression and parasite survival

As iNOS is associated with leishmanicidal functions in macrophages [16], we examined the expression of iNOS in infected and uninfected macrophages after TLR ligand treatment. In another experiment, infected and uninfected cells were treated with TLR ligands for 60 h and used for Giemsa staining for enumerating parasite count. We observed that different TLR ligands individually induced different levels of iNOS (Figure 7a,b) and nitrites (Figure 7c) controlling amastigote counts in macrophages (Figure 7d). TLR9 stimulation resulted in high iNOS expression and fewer amastigotes, whereas TLR2 stimulation resulted in less iNOS expression and very high parasite load. TLR4 induced high nitrite generation in infected and uninfected macrophages (Figure 7c). These observations indicated that single TLR targeting might not achieve significant anti‐leishmanial effects and that targeting inter‐TLR dependencies was imperative.

FIGURE 7.

TLRs control the Leishmania infection by enhancing iNOS expression. (a) Mouse macrophages were isolated from BALB/c mice and 60 h L. major‐infected macrophages or uninfected macrophages treated with different doses of TLR ligands, as previously described, for another 8 h. The line graphs show the relative fold changes in iNOS expression. GAPDH was used as internal control for normalization, and fold changes were calculated by ΔΔCT method. (b) After collecting the cell‐free supernatants, cells were washed with PBS, lysed and subjected to Western blotting analysis of the expression of inducible nitric oxide synthase (iNOS). (c) In another experiment, infected and uninfected cells were treated with the selected dose of TLR ligands, as described earlier. The cell‐free supernatants were collected after 24 h of TLR treatment, and nitrite generation (μM) was measured using Griess reagent and shown as bar graphs. (d) Similar experiment was conducted by plating macrophages in chamber slides for checking the parasite count. Macrophages were infected with L. major promastigotes at a ratio of 1:10 for 6 h. Extracellular parasites were washed out and after another 12 h of infection; cells were treated with indicated doses of TLR ligands (stimulation of TLR ligands was used in per ml culture media) for more 60 h. Cells were washed with 1xPBS, fixed with chilled methanol for 4 min and stained with Giemsa for amastigote enumeration

TLR2, TLR6, TLR9 AND TLR11 as a key module of inter‐TLR dependency

These inter‐TLR dependencies reveal two important hidden profiles: (1) TLR2 and TLR11 up‐regulated each other's expression implying a feed‐forward motif; and (2) both TLR2 and TLR11 down‐regulated TLR6 and TLR9 expression. Therefore, we examined the effects of TLR1, TLR2 and TLR6 silencing on TLR9 and TLR11 expression. We observed that TLR1 or TLR2 silencing enhanced TLR9, but reduced TLR11, expression, whereas TLR6 silencing had opposite effects (Figure 8a). By contrast, TLR11 silencing prevented the profilin‐enhanced TLR2 up‐regulation but partly restored profilin‐inhibited TLR9 expression (Figure 8b). TLR1 and TLR6 silencing failed to restore the PGN‐inhibited TLR9 expression, suggesting that TLR2‐PGN interaction down‐regulated TLR9 expression (Figure 8c). Examining the combined effect of these four TLRs on TLR9 expression, we observed that subthreshold doses of PGN and Pam3CSK4—both TLR2 ligands—failed to reduce TLR9 expression, but simultaneous addition of LPS augmented TLR9 expression, which was significantly down‐regulated by profilin (Figure 8d). Functionally, TLR2 silencing enhanced CpG‐induced IL‐12 production (Figure 8e), whereas TLR6 silencing enhanced IL‐10 production from the FSL‐treated macrophages (Figure 8f). These observations confirm the following: TLR1/2 up‐regulates TLR11 but down‐regulates TLR9; TLR2/6 down‐regulates TLR11 but up‐regulates TLR9; TLR11 down‐regulates TLR9, but up‐regulates TLR2, expression; and TLR9 expression is differentially regulated by combination of TLRs. Such inter‐TLR interactions selectively modulated IL‐10 and IL‐12 production (Figure 8e,f). These observations imply that specific inter‐TLR dependencies in macrophages play significant roles in intra‐macrophage infections such as HIV, Salmonella, Mycobacterium, Listeria, Toxoplasma and Leishmania [17, 18, 19, 20, 21.

FIGURE 8.

TLR2, TLR6, TLR9 and TLR11 are key modules of inter‐TLR dependency. (a) BALB/c macrophages were transduced with TLR1 shRNA‐Lv (left), TLR2 shRNA‐Lv (middle) and TLR6 shRNA‐Lv (right) (2 lentiviral particles/cells) or with control lentiviral particles for 48 h. TLR1 shRNA‐Lv and TLR2 shRNA‐Lv‐transduced or untransduced macrophages were stimulated with Pam₃CSK₄ (50 ng/ml), and TLR6 shRNA‐Lv‐transduced or untransduced macrophages were stimulated with FSL (50 ng/ml) for another 6 h and assessed for TLR9 and TLR11 expression. (b) Macrophages were transduced with TLR11 shRNA‐Lv (two lentiviral particles/cell) or with control lentiviral particles for 48 h. Lentivirally transduced or untransduced macrophages were stimulated with profilin (250 ng/ml) for another 6 h and examined for TLR2 and TLR9 expression. (c) BALB/c macrophages were transduced with TLR1 shRNA‐Lv and TLR6 shRNA‐Lv alone and with combination (two lentiviral particles/cells) or with control lentiviral particles for 48 h. Lentivirally transduced or untransduced macrophages were stimulated with PGN (5 μg/ml) for another 6 h and assessed for TLR9 expression. (d) Mouse macrophages were treated with the indicated doses of Pam₃CSK₄ (1 ng), PGN (0·2 μg), LPS (5 ng), FSL (2·5 ng) and profilin (20 ng) for 6 h and assessed for TLR9 expression. (e) Macrophages were transduced with TLR1 shRNA‐Lv (left) and TLR2 shRNA‐Lv (right) (two lentiviral particles per cells) or with control lentiviral particles for 48 h. Transduced or untransduced macrophages were stimulated with CpG (0·125 μm) for another 6 h and examined for IL‐12 expression. (f) Macrophages were transduced with TLR6 shRNA‐Lv for 48 h. Transduced or untransduced macrophages were treated with FSL (50 ng/ml) for another 6 h and assessed IL‐10 production. RNA was isolated for the synthesis of cDNA. Real‐time PCR was performed from cDNA templates for the amplification of TLRs, IL‐10 and IL‐12 by using their specific forward and reverse primers. GAPDH was used as a control reference to normalize the value of gene of interest. Data shown represent the relative fold change in TLRs by using ΔΔCT method compared with untreated controls. Experiments were performed three times, and the data are shown as mean ± SEM

Because IL‐10 and IL‐12 are key downstream effectors of TLRs, we also checked the production of these two cytokines from macrophages pre‐transduced with TLR1shRNA or TLR2shRNA or TLR6shRNA followed by PGN treatment. We observed that PGN enhanced IL‐10 but reduced IL‐12 production from macrophages pre‐transduced with TLR1 shRNA‐Lv and TLR2 shRNA‐Lv (Figure S6a,b) and PGN reduced IL‐10, but not IL‐12, production from macrophages pre‐transduced with TLR6 shRNA‐Lv (Figure S6c).

Targeting TLR interdependency reduces parasite load in susceptible BALB/c mice

Guided by these TLR interdependency profiles in uninfected and Leishmania‐infected macrophages, we first silenced TLR11 on day 3 post‐infection to interrupt the TLR2‐TLR11 feed‐forward motif keeping TLR2 intact so that in the presence of TLR2‐TLR6 ligand—administered on 7th and 9th day post‐infection—TLR2‐TLR6 heterodimer could provide the host‐protective signals in BALB/c mice. To enhance the host‐protective effect, TLR6 ligand and TLR9 ligand were administered together on day 11. TLR2 was silenced by a lentivirally expressed TLR2shRNA on day 25 post‐infection. BALB/c mice were killed on day 35 post‐infection to assess anti‐leishmanial effect of the treatments. We observed that this rationalized multi‐TLR‐targeted immunotherapy significantly reduced footpad thickness and parasite load (Figure 9a,b). The CD4⁺ T cells were purified and co‐cultured with the 36 h L. major‐infected macrophages for another 36 h followed by enumeration of amastigotes. We observed that the CD4⁺ T cells from the untreated or control shRNA‐treated L. major‐infected mice enhanced parasite load, but the CD4⁺ T cells from the multiple TLR‐targeted BALB/c mice reduced the infection in these macrophages (Figure 9c), indicating that the multi‐TLR targeting rendered disease‐promoting T‐cell response to host‐protective T‐cell response. The reduced parasite load was accompanied by reduced Treg cell numbers (Figure 9d), low IL‐4 but high IFN‐γ response (Figure 9e).

FIGURE 9.

TLR1, TLR2, TLR6, TLR9 and TLR11 rewiring reduces the parasite load in vivo. TLR11 was silenced on 3rd day after 2 × 10⁶ L. major infection of BALB/c mice. The mice were subcutaneously treated with TLR2/6 ligand alone on 7th and 9th day after the infection, followed by TLR2/6 ligand BPPcysMPEG (1 μg/mouse) and TLR9 ligand CpG (10 μg/mouse) on 11th day of infection. TLR2 was silenced on 25th day after infection. (a) Disease progression was scored weekly by evaluating the net footpad swelling (the thickness of the uninfected left footpad subtracted from the infected right footpad) by digital micrometer for 5 weeks. (b) 5 weeks after infection, indicated group of mice were killed, and parasites were enumerated from the popliteal lymph node cells. Silencing of TLR2 and TLR11 in footpad was confirmed by performing RT‐PCR (Inset). (c) BALB/c‐derived peritoneal macrophages were infected with L. major for 72 h. After washing out the extracellular parasites, macrophages were co‐cultured with the CD4⁺ T cells isolated from the lymph nodes of indicated group of BALB/c mice at 1:3 ratio (MФ:T cells). After 72‐h infection, macrophages were washed, fixed, stained with Giemsa stain, and evaluated for the amastigotes/ 100 macrophages. (d) BPPcysMPEG or CpG or TLR11 shRNA or TLR2 shRNA ‐treated or untreated L. major‐infected BALB/c mice were killed on 35th day. Lymph node T cells analysed for the Treg cell expansion by gating FoxP3⁺ IL‐10⁺ cells on CD3⁺ CD4⁺ CD25⁺ CD127^low cells by FACS. (e) Uninfected and L. major‐infected mice treated with different combination of TLRs were killed, their spleens were crushed, RBCs were lysed with Gey's solution, and the lymph node cells were lysed to extract RNA for real‐time PCR. Real‐time PCR was performed from cDNA templates for the amplification of IFN‐γ and IL‐4 by using their specific forward and reverse primers. GAPDH was used as a reference control to normalize the value of gene of interest. Data shown represent the relative fold change in IFN‐γ and IL‐4 by using ΔΔCT method compared with uninfected controls. (f) TLR1 and TLR11 were silenced on day 3 after 2 × 10⁶ L. major infection of BALB/c mice. The mice were treated with FSL (10 μg/mouse), a TLR2/6 ligand, and CpG (10 μg/mouse), a TLR9 ligand on 7th, 9th and 11th day after the infection. Disease progression was scored weekly by evaluating the net footpad swelling (the thickness of the uninfected left footpad subtracted from the infected right footpad) by digital micrometer for 5 weeks. (g) Mice were killed, and parasites were enumerated from the popliteal lymph node cells, as indicated. (h) TLR1 silencing in footpad was validated by RT‐PCR. (i) Lymph nodes were isolated from the above groups. Stamp smears of the lymph nodes from the group A (left) and the group F (right) are shown here as representative of the effects of targeting TLR interdependencies

Phase‐specific TLR interdependency targeting reinforces host‐protective anti‐leishmanial response

Because TLR1 and TLR11 expressions were up‐regulated whereas TLR6 and TLR9 expressions were down‐regulated in L. major‐infected macrophages [10, 11, 12, 13, we designed a new phase‐specific therapeutic strategy. We first silenced TLR1 and TLR11 on day 3 post‐infection, followed by TLR2‐TLR6 ligand and TLR9 ligand administration on 7th, 9th and 11th day. We assessed parasite burden on 35th day post‐infection. We observed that this multi‐TLR targeting strategy provided significant protection to L. major‐infected mice, as assessed by reduced footpad thickness (Figure 9f) and parasite burden (Figure 9g). TLR1 silencing was verified (Figure 9h). Representative parasite load in the lymph nodes from control‐infected mice and the optimized TLR interdependency‐targeted mice was examined (Figure 9i): the lymph nodes from the targeted mice had very little detectable parasites.

Although such TLR interdependency was not worked out in L. donovani infection that inflicts potentially fatal visceral disease, this same combination of interdependency targets (Figure 10a) significantly lessened splenic parasite burden (Figure 10b), host‐protective cytokine profile (low IL‐4 and IL‐10 and high IL‐12 and IFN‐γ; Figure 10c) and less Treg cells (Figure 10d). These results established the inter‐TLR regulatory relationship and its successful use in devising an effective anti‐leishmanial therapeutic strategy. Although discovered in the context of Leishmania infection, the same TLR interdependency principle might be tested in other infectious diseases, especially in immunosuppressive infections. In those infections that induce hugely inflammatory cytokine storm, as observed in SARS‐CoV‐2 infection, this TLR interdependency profile may be reversed. If tested and turned out to be so, a reverse targeting, that is silencing TLR6 and TLR9 but treatment with the ligands for TLR1/2 and TLR2/2, might prove host‐protective.

FIGURE 10.

(a) Phase‐specific TLR treatment in L. donovani‐infected BALB/c mice. BALB/c mice were infected with 2 × 10⁷ L. donovani 2 days prior to lentivirally expressed TLR11shRNA or control shRNA treatment followed by treatment with BPPcysMPEG (1 μg/mouse) on 7th and 9th day or CpG (10 μg/mouse) on 11th day; these treatments were followed by TLR2shRNA on 21st day or control shRNA in different combinations or were left untreated. Mice were euthanized on the 28th day. (b) Spleens were isolated from above groups. Giemsa‐stained stamp smears were used for splenic parasite load assessment. The graph represents the parasite load as Leishman–Donovan unit (LDU) in spleens of the control and treated mice. (c) Naive and infected mice treated with different combinations of TLRs were killed, their spleens were crushed and RBCs were lysed with Gey's solution, and the splenocytes were lysed to extract RNA for real‐time PCR. Real‐time PCR was performed from cDNA templates for the amplification of IL‐12, IFN‐γ, IL‐10 and IL‐4 by using their specific forward and reverse primers. GAPDH was used as a reference control to normalize the value of gene of interest. Data shown represent the relative fold change in TLRs by using ΔΔCT method compared with uninfected controls. Inhibition of TLR2‐ and TLR11‐specific shRNA was checked by RT‐PCR by transducing the mouse macrophages with their respective TLR‐Lv or control lentiviral particles for 48 h, as shown in Figure 8 and Figure 9. (d) The collected splenocytes were stained for T‐regulatory cell‐specific fluorochrome‐conjugated antibodies and analysed by flow cytometry. T cells were analysed for the FoxP3⁺ IL10⁺ cells by serial gating of CD3⁺ CD4⁺ CD25⁺ CD127^low cells. The data shown are representative of one of the triplicate experiment. Experiments were performed three times, and the data are shown as mean ± SEM

DISCUSSION

With millions of people affected in 88 countries, extremely restricted choice for chemotherapy and unavailability of a vaccine for human use, leishmaniases are major neglected tropical diseases. The cutaneous form of the disease is caused by L. major that infects macrophages, alters macrophage expression of TLRs and causes immunomodulation to prevent its elimination [22, 23, 24. However, the reported studies targeted single TLR at a time to understand the role of each TLR in Leishmania infection [25, 26, 27. Such studies precluded any possible inter‐TLR interactions and thus underestimated the importance of expression of multiple TLRs on macrophages that play host to the parasite. We first hypothesized that multiple molecular patterns on pathogens’ surface resulted in possible interactions with multiple cell membrane‐located TLRs. Secondly, these interactions eventuated in the modulation of the expression of intracellular TLRs, as these TLRs would recognize the nucleic acids released from the intracellularly degraded pathogens. These assumptions led to the third proposition that the TLRs influenced each other's modulation resulting in profiles of interdependency that collectively regulated the anti‐pathogen immune response. Thus, once the TLR interdependency profiles in an infection were worked out, these would be targeted to devise an anti‐pathogen immunotherapy. Our findings in this report uncover the functional interactions between TLRs in macrophages and its rewiring in L. major‐infected macrophages. Thus, this novel TLR interdependency becomes an interface where the pathogen and the host interact to ensure pathogen survival or elimination.

In the first part of our study, we therefore treated macrophages with different doses of each TLR ligand for different periods of time, followed by assessment of the expression of all TLRs. These experiments suggested the following facts: (1) the ability to autoregulate its own expression differed among TLRs—not all TLRs autoregulated their expressions in response to their respective ligands; (2) each TLR prioritized regulation of other TLRs; (3) the inter‐TLR regulation plausibly had both feed‐forward and negative feedback motifs; and (4) each TLR’s expression is regulated by multiple TLRs. Once the TLR interdependency profiles were known, the effect of L. major infection on the interdependency was tested. It was observed that Leishmania rewired the connections between the TLRs. Analyses of these alterations suggested how Leishmania might target the TLR interdependency profile to ensure its survival (Figure S7). Finally, extracting the key module of these TLRs interacting with each other, we devised a TLR interdependency‐targeted anti‐leishmanial therapy. In BALB/c mice, this sequential targeting of multiple TLRs drastically reduced the parasite load.

Besides uncovering the TLR interdependency, the results projected several significant perspectives that can be investigated through a series of independent experiments. Firstly, the ten TLRs present in humans can influence each other's expression and functions in enormous combinations, each of which can be specific for one infection. Each combination thus represents one inter‐TLR dependency profile. The specificity of the profile may depend on the unique transcription factor(s) activated by each TLR (Figure S8). Secondly, as each TLR’s activation can result in a specific cytokine induction profile, the combination may lead to a very specific cytokine profile characteristic of the infectious disease. Thirdly, as cytokines are key mediators of inflammation or suppressive immune response, these cytokine profiles may in fact determine the outcome of the infection. Inter‐TLR dependency can thus be targeted for the desired outcome of the infection, as shown by us in this report. Fourthly, as each pathogen may have their specific patterns of PAMP expression, the initial interaction with multiple TLRs alters the expression of cell surface TLRs, as shown by us in this study. For example, Leishmania infection enhances the expression of TLR11. This TLR11 may enhance interaction with profilin, a ligand found on uropathogenic bacteria. TLR11, along with TLR5, may also determine the outcome of Salmonella infection [19]. So, this inter‐TLR dependency may serve as the basis for co‐infection profile [28, 29. Finally, as TLRs are thus suggested to contribute to specific infection profile and as TLRs influence the adaptive immune response, it is possible that the nature of infection may be associated with the propensity of triggering a specific autoimmune disease [30].

Our studies therefore map these quantitative inter‐TLR dependencies as a novel immunoregulatory framework for the first time. Detailing the molecular mechanism of this phenomenon deserves multiple independent studies and thus identifies the broader scope for future research. We identify the TLRs with major regulatory connections and devise a multiple TLR targeting strategy to redirect immune response for the benefit of the host, for example as a novel immunotherapy in not only infectious diseases but also non‐infectious diseases such as autoimmune diseases and cancer. On the other hand, the results uncover an intricate immune evasion mechanism adopted by Leishmania. The study thus delivers a novel scientific rationale and a technical platform for devising similar immunotherapeutic strategies against other diseases.

CONFLICT OF INTERESTS

Only CAG and TE have been named as co‐inventors of BPPcysMPEG. All other authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

DS and AP performed the in vitro and in vivo experiments, analysed and arranged data. US designed and implemented the data processing and analysis pipeline including in silico analysis. SPP and HSC designed experiments and made lentiviral particles including titre assessment. PC, NB and SKG contributed in performing animal experiments and parasite counting. BS and SB conceptualized the study, designed the experiments and wrote the grant proposal. BS, AS, RS and SB prepared manuscript. CAG and TE provided BPPcysMPEG.

Supporting information

Figures S1‐S8

Table S1