BALB/c mice (6 weeks old) were purchased from the Jackson Laboratory. All mice were housed at the University of Arizona Animal Care Facilities. All mouse protocols were approved by the Arizona University Institutional Animal Care and Use Committee (IACUC) in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC).

The L. major mutants (Δlpg1- and Δlpg2-) and add-backs (Δlpg1-/+LPG1 and Δlpg2-/+LPG2) were derivatives of WT L. major LV39 clone 5 background. They were made by homologous gene targeting and maintained in selective media as described previously (22, 23, 29, 30). Leishmania major parasites were maintained at 26°C in M199 culture medium (Thermo Fisher Scientific) supplemented with 20% heat-inactivated FBS (Thermo Fisher Scientific)), 20 mM HEPES (Sigma-Aldrich) and 50 ug/ml gentamycin (Thermo Fisher Scientific). Prior to using parasites for infection, metacyclic promastigotes were isolated from stationary-phase cultures using density gradient centrifugation (31, 32). Metacyclic promastigotes were washed three times in cold phosphate-buffered saline (PBS) (Thermo Fisher Scientific) by centrifugation, resuspended in PBS at 2X10⁸/ml and 10 μl containing 2x10⁶ metacyclic promastigotes were injected into the top of the right hind footpad.

Lesion size was measured weekly with Vernier calipers and determined by subtracting the size of the uninfected from that of the infected footpad. Parasite burden in the infected footpad was estimated by limiting dilution analysis as previously described (33).

Blood was drawn by retro-orbital bleeding under isoflurane anesthesia (MWI Animal Health) into tubes containing 3.2% citrate buffer (G-Biosciences) from naïve mice or at days 3, 14 and 42 PI. Platelets were prepared as previously described (34). In brief, blood was diluted with 250 μl of modified Tyrode’s-HEPES buffer (134 mM NaCl, 0.34 mM Na₂HPO₄, 2.9 mM KCl, 12 mM NaHCO₃, 20 mM HEPES, 5 mM glucose, and 1 mM MgCl₂, pH 7.3) and centrifuged at 250 x g for 15 minutes at room temperature. Platelet-rich plasma was removed, and platelets were then isolated by centrifugation at 900 × g for 30 min. Plasma was collected for DKK1 ELISA.

In vitro stimulation of platelets was done using various stock concentrations of soluble leishmania antigen (SLAG) (1:25, 1:50, 1:200, 1:400), Pam2CSK4, Pam3CSK4 (InvivoGen), LPS (Escherichia coli O111:B4-Sigma) and control isotype antibodies (IgG1, IgG2a and IgG2b) (Invitrogen). DKK1 inhibition was done using anti-TLR4, anti-TLR2, anti-TLR1/2, anti-TLR2/6, anti-TLR1/2/6 antibody (Invitrogen), and Go 6976 PKC-alpha inhibitor (Abcam) before incubation for 1 hr at room temperature. Platelets were isolated by centrifugation at 550 x g for 10 minutes, and the supernatant was collected for DKK1 ELISA. For SLAG preparation, L. major parasites were prepared at 5 × 10⁸ parasites/ml in M199 medium (without FCS). SLAG was prepared by repeated five freeze and thaw cycles at -80°C and room temperature, respectively. SLAG supernatant was collected after centrifugation at 1500rpm for 20 minutes and stored at -80°C. For detection of LPG expression, Western Blot of SLAG (7 µg) was resolved by SDS-PAGE electrophoresis using 10% PAGE; molecules were transferred to PVDF membrane using the iBlot 2 semi-dry blotting system (Thermofisher Scientific Inc.). The membrane was blocked with 5% powdered milk in Tris Buffered Saline (TBS) and incubated for 1 hour at room temperature. Blots were probed with WIC 79.3 antibody (1:1000). After washing in TBS supplemented with 0.05% Tween 20 (TBST), the membrane was incubated for 1 h with anti-mouse IgG conjugated with Horseradish peroxidase (1:7500-Invitrogen) and the reaction was visualized using Pico PLUS Chemiluminescent substrate kit (Thermo Fisher Scientific). Signals were acquired using ChemiDoc Imaging System (Bio-Rad).

For FACS analyses, isolated platelets (10x10⁶/ml) were washed at 550 x g for 10 minutes in the presence of PGE₁ (140 nM; Sigma-Aldrich) and indomethacin (10 μM; Thermo Fisher Scientific). Platelets were resuspended to the required density in modified Tyrode’s-HEPES buffer (Sigma-Aldrich) and rested for 30 minutes at 37°C in the presence of 10 μM indomethacin prior to staining. Staining for P-selectin expression was done using PE-conjugated CD41 (BioLegend) (to determine platelet purity) and APC-conjugated CD62P (Invitrogen; a marker of P- selectin). Stained platelets (1 × 10⁶/ml in Tyrode’s-HEPES buffer; 100 μl) were analyzed using a BD FACSCanto flow cytometer; analysis was done using FlowJo software. A total of 10,000 events per sample were collected. The mean fluorescent intensity of P-selectin was compared among the different groups in the overlay.

Leukocyte-platelet aggregation assessment was performed as described previously with minor modifications (28). Briefly, 100 μl blood was collected via the retro-orbital sinus into tubes containing 3.2% citrate buffer. Peripheral blood (10 µl) was stained with PE-conjugated anti-mouse CD41 antibody and Pacific blue conjugated CD45 antibody (BioLegend) for 10 mins in the dark at room temperature. Fix/red blood cell lysis solution (eBiosciences) was added and incubated for 15 min in the dark at room temperature. Samples were analyzed by flow cytometry within 4 to 6 hours. Live gating was performed on leukocyte-sized events to exclude single platelets. Leukocytes were identified by their forward and side scatter characteristics and CD45 expression. The CD41+ subpopulation identified leukocyte-platelet aggregates.

Enzyme-linked immunosorbent assays were performed using a mouse DKK1 ELISA kit (Thermo Fisher Scientific) to determine the concentration of DKK1 in plasma and cell culture supernatants according to the manufacturer’s protocol.

The in vivo expression of P-selectin and DKK1 production was analyzed using one-way ANOVA with Bonferroni’s post hoc test. Also, in vitro DKK1 production and P-selectin expression following SLAG stimulation was analyzed using one-way ANOVA with Bonferroni’s post hoc test. Comparison of neutralizing antibody inhibition of platelet DKK1 induced by SLAG obtained from WT, addback and mutant parasites was done statistically using Student’s t-test. Likewise, a comparison of PKC-alpha inhibition of DKK1 induced by SLAG obtained from WT, addback and mutant parasites was done statistically using one-way ANOVA with Bonferroni’s post hoc test. In addition, lesion size, parasitic load and percentage of LPA was analyzed using one-way ANOVA with Bonferroni’s post hoc test. Data presented as means ± standard errors were performed using GraphPad Prism software (GraphPad Software, San Diego, CA, USA).

Platelets are rapidly deployed to sites of infection where they can modulate immune/inflammatory processes by secreting cytokines, chemokines, and other inflammatory mediators (35). P-selectin is an adhesion receptor for leukocytes expressed by activated platelets (36). To investigate the possible effects of LPG1- and or LPG2-dependent molecules on platelet activation in vivo, expression of P-selectin was determined from platelets obtained from mice that had been infected by WT, Δlpg1-, Δlpg2^- , Δlpg1-/+LPG1 and Δlpg2 ^-/+LPG2 parasites for varying periods of time. Relative to the WT controls, P-selectin expression was significantly suppressed in platelets obtained from Δlpg1- and Δlpg2- mutant infected mice on days 3, 14 and 42 PI. As expected, mice infected with “add-back’ control parasites (Δlpg1-/+LPG1 and Δlpg2-/+LPG2) manifested restoration of P-selectin expression similar to those infected with WT parasites ( Figures 1A–F ).

We also characterized P-selectin expression in vitro using SLAG-stimulated platelets obtained from naïve mice. Stimulation of platelets obtained from naive mice with SLAG derived from Δlpg1- and Δlpg2- parasites induced poor P-selectin expression. In contrast, platelets stimulated with SLAG derived from WT, Δlpg1-/+LPG1 and Δlpg2-/+LPG2 parasites induced robust P-selectin expression ( Figures 2A, B ). Taken together, these data suggest that the inability of Δlpg1- and Δlpg2- parasites to induce P-selectin expression arises from a lack of LPG1 or LPG2 dependent products. As Δlpg2- parasites lack all phosphoglycan-containing glycans (LPG, PPG and others) while Δlpg1- lacks only LPG, the lack of a significant difference between Δlpg1- and Δlpg2- mutants indicates that platelet activation and P-selectin expression are primarily regulated by LPG.

We previously showed that activated platelets release DKK1 following recognition of L. major (28). Furthermore, DKK1 is maintained at a high concentration in L. major infected mice; depletion of platelets resulted in the complete loss of DKK1 (28). To identify the L. major ligand involved in DKK1 production, we first compared the effect of LPG1 and LPG2 gene-dependent molecules in plasma DKK1 production at days 3, 14 and 42 PI. We confirmed a significantly decreased production of plasma DKK1 in Δlpg1- and Δlpg2- mutant infected mice compared to WT controls, but in Δlpg1-/+LPG1 and Δlpg2-/+LPG2 infected mice, DKK1 production was restored ( Figures 3A–F ). Compared to days 3 and 14 PI, there was a reduction in the concentration of DKK1 produced by activated platelets on day 42 PI.

Impaired DKK1 production from the Δlpg1^- and Δlpg2- were further confirmed via an in vitro stimulation of platelets with SLAG obtained from Δlpg1- and Δlpg2- parasites. Consistent with the plasma DKK1, in vitro stimulation of naïve platelets with SLAG obtained from Δlpg1- and Δlpg2- parasites failed to induce significant production of DKK1, in contrast a significant level of DKK1 was released by platelets stimulated with SLAG derived from WT L. major ( Figures 4A, B ) or the genetically reconstituted organisms. Since the levels of DKK1 produced in Δlpg1- and Δlpg2- infected mice are comparable, these data suggest that activated platelets release DKK1, a process which depends on the leishmania-derived LPG virulence factor; PPG does not appear to additionally contribute to platelet activation.

The Δlpg1- and Δlpg2- parasites had a greatly delayed formation of lesions compared with WT controls, whereas lesions produced by the Δlpg1-/+LPG1 and Δlpg2-/+LPG2 parasites showed minimal delay and resembled WT controls ( Figures 5A, B ). In addition, the parasitic burden was significantly decreased in Δlpg1- and Δlpg2- infected mice compared to WT controls, and restoration of the LPG1 and LPG2 genes resulted in parasite survival, which is comparable with the WT controls ( Figures 5C, D ). The pattern of lesion size in the footpads and parasitic burdens of mice infected with genetically deficient Leishmania (LPG1 or LPG2 gene) were comparable ( Figures 5A–D ). These data are consistent with previous publications concerning the virulence of these lpg mutant parasites (25, 29, 37). Repetition of this experiment serves as a confirmation of these results for the mutant lines used in the study.

Platelets express functional TLR2/4 and MyD88, which participate in platelet responsiveness to infection (38). We previously showed that activated platelets release DKK1 following recognition of L. major and PKC-alpha inhibitor or neutralization of TLR2 blocks secretion of DKK1 from SLAG-stimulated human platelets (28). PKC-alpha inhibitor is known to block the TLR2/MyD88 signalling pathway (39). An important feature of TLR2 is the ability to form heterodimers with its co-receptors (TLR1 and TLR6). Therefore, to confirm that TLR1/2/6 recognizes Leishmania-derived LPG and initiates the production of DKK1, neutralizing antibodies were used to inhibit the function of these specific TLRs in the presence of SLAG. Exposure of Δlpg1- and Δlpg2- SLAG-stimulated platelets to blocking anti-TLR1/2/6 antibodies showed no significant effect, but significant inhibition of DKK1 production by anti TLR1/2/6 antibodies was observed in WT, Δlpg1-/+LPG1 and Δlpg2-/+LPG2 SLAG treated platelets ( Figures 6A, B ). To determine the level of DKK1 inhibition by the neutralizing antibodies, platelets were treated with neutralizing antibodies (anti-TLR2, TLR1/2 and TLR2/6) or control isotype antibodies (IgG2a and IgG1) in the presence of WT-derived SLAG. Results showed that neutralization of TLR1/2/6 with an anti-TLR1/2/6 mAb markedly reduced SLAG-induced DKK1 production in platelets compared with control isotype antibodies. Relative to DKK1 inhibition by anti-TLR2 blocking antibody, addition of anti-TLR1 (TLR1/2) antibody significantly deceased DKK1 production, while addition of anti-TLR6 antibody had no significant effect ( Figure 6C ). This suggests that only TLR1/2 played a primary role in LPG induced DKK1 production. Previous studies reported that inhibition of TLR2 and 4 attenuates inflammatory response and parasite burden in cutaneous leishmaniasis (40). Therefore, to determine whether TLR4 contributes in DKK1 production after LPG recognition by platelets, platelets were treated with anti-TLR4 neutralizing antibody or control isotype antibody (IgG2b) in the presence of WT-derived SLAG. Results showed that neutralization of TLR4, with an anti-TLR4 mAb failed to reduce SLAG-induced DKK1 production. Also, there was no change in DKK1 production in platelets treated with the control isotype antibody. This suggests that TLR4 played no role in LPG induced DKK1 production ( Figure 6C ). To confirm the specificity of this data, we assessed the possibility of other PAMPs (not directly related to LPG) to induce DKK1 release. Platelets treated with LPS (TLR4) and Pam2CSK4 (TLR2/6) failed to elicit DKK1 production. However, Pam3CSK4 (TLR1/2) slightly induced DKK1. This suggests that LPS and Pam2CSK4 lack the ability to induce DKK1 production ( Figure S3 ). To verify that SLAG preparations of the mutant lines lack LPG, Western blot assay of SLAG preparations confirms the absence of LPG in Δlpg1- derived SLAG and the presence of LPG in the SLAG of the WT and lpg1 add-back lines ( Figure S1 ).

In addition, when platelets were incubated with SLAG obtained from WT, Δlpg1-/+LPG1 and Δlpg2-/+LPG2 parasites in the presence of varying concentrations of PKC-alpha inhibitor, a significant reduction in secretion of DKK1 was observed at higher dose concentrations of PKC-alpha inhibitor. In contrast, DKK1 production from Δlpg1- and Δlpg2- SLAG-stimulated platelets exhibited some degree of inhibition by the PKC-alpha inhibitor at the higher concentrations ( Figures 6D, E ), but this was confirmed to be background platelet DKK1 that is being inhibited, as the non-stimulated platelets show a comparable level of inhibition ( Figure S2 ). This suggests that the recognition of leishmania-derived LPG and induction of DKK1 from activated platelets may occur via the TLR1/2-MyD88 pathway.

Platelet-leukocyte interactions are indispensable events in hemostasis and inflammation (41, 42). Under inflammatory conditions, platelets interact with leukocytes, thus promoting their recruitment by the formation of platelet leukocyte aggregates via the interaction of PSGL-1 and P-selectin (28, 43). The consequence of this interaction enables leukocytes to fulfill their multiple cell-intrinsic functions and immunological tasks. We showed previously that pre-treatment of mice with DKK1 inhibitor 24 hours prior to infection with L. major reduced the elevation of LPA formation at 4 h post-infection (28). Furthermore, repeated administration of DKK1 inhibitor led to significant reduction in cellular recruitment at the site of infection and to the draining lymph node. These findings suggest that leukocyte platelet aggregation and infiltration of leukocytes to the infection site is driven by DKK1 production.

Given that DKK1 production in mice infected with Δlpg1- and Δlpg2- parasites was impaired, we considered the possibility that the loss of LPG1 and LPG2 gene-dependent molecules might decrease LPA formation in blood obtained from Δlpg1- and Δlpg2- mutant infected mice ( Figures 7 ). Relative to the WT-infected mice, LPA circulating in the blood was comparable to those observed in Δlpg1-/+LPG1 and Δlpg2-/+LPG2 infected mice, but this aggregate was significantly impaired in Δlpg1- and Δlpg2- infected groups at day 3 PI. In addition, the LPA level formed in response to infection with Δlpg1- parasites is comparable to those formed in response to infection with Δlpg2- parasites ( Figures 7A–E ). This suggests that LPG-activated platelets enhance P-selectin expression, resulting in platelet leukocyte aggregation.