Activation of TLR-pathway to induce host Th1 immune response against visceral leishmaniasis: Involvement of galactosylated-flavonoids

Abstract

Immunotherapeutic strategies against visceral leishmaniasis (VL) are pertinent because of the emergence of resistance against existing chemotherapy, coupled with their toxicity and high costs. Various bioactive components with potential immunomodulatory activity, such as alkaloids, terpenes, saponins, flavonoids obtained primarily from medicinal plants, have been screened against different disease models. Reports suggested that glycans containing terminal β-galactose can skew host immune response towards Th1 by engaging TLRs. In this study, two synthesized terminal galactose-containing flavones, Quercetin 3-d-galactoside (Q-gal) and Kaempferol 3-O-d-galactoside (K-gal), are profiled in terms of inducing host protective Th1 response in both in vitro & in vivo animal models of experimental VL individually against antimony-resistant & antimony-susceptible Leishmania donovani. Further, we explored that both Q-gal and K-gal induce TLR4 mediated Th1 response to encounter VL. Molecular docking analysis also suggested strong interaction with TLR4 for both the galactosides, with a slightly better binding potential towards Q-gal. Treatment with both Q-gal and K-gal showed significant antileishmanial efficacy. Each considerably diminished the liver and splenic parasite burden 60 days after post-infection (>90% in AG83 infected mice and >87% in GE1F8R infected mice) when administered at a 5 mg/kg/day body-weight dose for ten consecutive days. However, the treatments failed to clear the parasites in the TLR4 deficient C3H/HeJ mice. Treatment with these compounds favors the elevation of TLR4 dependent host protective Th1 cytokines and suppression of disease-promoting IL-10. Q-gal and K-gal also triggered sufficient ROS generation in macrophages to kill intracellular parasites directly.

Keywords: Antileishmanial; Galactosylated-Flavonoids; Reactive oxygen species; TLR-4; Th1 response; Visceral leishmaniasis.

Figure 1

Comparison of the antileishmanial activity of different doses of Quercetin 3-d-galactoside (Q-gal) & Kaempferol 3-O-d-galactoside (K-gal) against intra-cellular amastigote forms of L. donovani Sb^S strain AG83 and Sb^R strain GE1F8R in vitro. RAW264.7 cells (10⁶) were adhered for 24 h and subsequently infected with 20 times AG83 and GE1F8R promastigotes for 24 h. They were then treated with different doses of Q-gal and K-gal (as mentioned in the figure) for the next 24 h (A), 36 h (B) and 48 h (C), respectively. Intra-cellular parasite number was measured by Giemsa staining and expressed as parasites/100 mΦ. Results are demonstrative of three independent experiments and data shown are mean ± SD; n = 3. ∗p values versus corresponding infected control; paired two-tailed Student’s t-test. ∗p < 0.0001, ∗∗p < 0.001 and the other values are not significant compared to P < 0.01.

Figure 2

TLR4 and MyD88 are required for Q-gal and K-gal mediated antileishmanial response. A. RAW264.7 cells transfected with siRNAs specific to TLR4, TLR2 and MyD88. Control group of mice was transfected with control siRNA (control). 24 h post-transfection, cells were recuperated and TLR4, TLR2 and MyD88 levels measured by western blots. GAPDH was used as loading controls. Blots are representative of three separate experiments. For the original blots of TLR4, TLR2, MyD88 and GAPDH, refer to supplementary figure number S7A, S7B, S7C & S7D respectively. B. TLR4 transfected, untransfected and TLR2 transfected RAW264.7 cells were infected with LD (APC/parasite 1:20) for 24 h and then treated with 25 μg/mL of Q-gal and K-gal for the next 24 h. Intracellular parasite number was determined by Giemsa staining and expressed as parasites/100 mΦ. The results are demonstrative of three independent experiments and data presented are mean ± SD; n = 3. ∗p values versus corresponding infected control; paired two-tailed Student’s t-test. ∗p < 0.0001, ∗∗p < 0.001 and the other values are not significant compared to P < 0.01. C. RAW264.7 cells transfected with MyD88 and control siRNA (control) and were infected with LD (APC/parasite 1:20) for 6h. RAW264.7 cells were also infected with LD (APC/parasite 1:20) in presence or absence of a TRIF inhibitory peptide, Pepinh-TRIF (100 μM), for six hours. Non-ingested promastigotes were removed by washing, and cells were cultured for another 18 h. Infected APCs were treated with Q-gal and K-gal (25 μg/mL) for twenty-four hours. Intracellular parasites were counted by Giemsa staining. Experiments were performed in triplicate and repeated three times each and one set of demonstrative data is shown. Error bars signify mean ± SD, ∗p < 0.0001 and the other values are not significant compared to p < 0.01; paired two-tailed Student’s t-test.

Figure 3

Effect of Q-gal and K-gal on cytokine response and free radical generation. Cells infected with LD promastigotes (24 h) at 20:1 ratio of parasite-to-cell trailed by treatment with 25 μg/mL of Q-gal and K-gal for 24 h. IL-10 (A), TGF-β (B), IL-12 (C) and TNF-α (D) cytokines levels in culture supernatants were also assessed by ELISA. ROS production was determined by H2DCFDA (E). The results are demonstrative of three independent experiments and data presented mean ± SD; n = 5. ∗p values vs corresponding infected control; paired two-tailed Student’s t-test. ∗p < 0.0001, ∗∗p < 0.001 and the other values are not significant compared to p < 0.01; paired two-tailed Student’s t-test. F. RAW264.7 cells (10⁶) were adhered for 24 h and subsequently infected with 20 times AG83 promastigotes for another 24 h. Infection was then followed by treatment with 25 μg/mL of Q-gal and K-gal for 24 h in the presence or absence NAC (50 μM). Intracellular parasite number was counted by Giemsa staining and expressed as parasites/100 MФ. The results are demonstrative of three independent experiments and data presented mean ± SD; n = 5. ∗p values vs corresponding infected control; paired two-tailed Student’s t-test. ∗p < 0.0001, ∗∗p < 0.001 and the other values are not significant compared to P < 0.01; paired two-tailed Student’s t-test. G. Cells were infected with LD promastigotes (24 h) at 20:1 parasite-to-cell ratio followed by treatment with 25 μg/mL of Q-gal and K-gal for 24 h. Endogenous NO production was determined by Griess reagent in the supernatants.

Figure 4

In vivo efficacy of Q-gal and K-gal against L. donovani infection through TLR4. Q-gal and K-gal were administered at an amount of 5 mg/kg/day orally for 10 successive days starting at 60th day post-infection in BALB/c mice. Animals were sacrificed 20 days after treatment and splenic (A) and hepatic parasite loads (B) were determined for all groups in LDU as described in the materials and method section previously. Parasite load in bone marrow was determined as parasites/100 nucleated cells (C). Data shown as mean ± SD of ten animals/group, and are demonstrative of three independent experiments. ∗p < 0.0001 and the other values are not significant compared to P < 0.01; related with infected control groups at all time points; paired two-tailed Student’s t-test. 5 mg/kg/day of Q-gal and K-gal orally were given to sixty days AG83 and GE1F8R infected TLR4 deficient C3H/HeJ mice for 10 consecutive days. The parasite burdens in liver and spleen (log₁₀ LDU, D) and in bone marrow (parasites/100 nucleated cells, E) of each animals were determined at 20 days post-treatment. The results are demonstrative of three independent experiments and data represent as mean ± SD; n = 5. ∗p < 0.0001 vs corresponding infected group; paired two-tailed Student’s t-test.

Figure 5

Effect of Q-gal and K-gal on Th1 cytokine and ROS production in vivo. 5 mg/kg/day of Q-gal and K-gal were given orally (10 days consecutively) staring on the 60^th day after infection in both BALB/c and C3H/HeJ mice. Animals were sacrificed 20 days after the treatment. Culture supernatants from different groups of experimental mice used to see the expressions of Th1 (A, IFN- γ; B, IL-12) and Th2 (C, IL-10; D, TGF-β) cytokines at protein level were detected by ELISA. The results are demonstrative of three independent experiments and data represents as mean ± SD; ∗p < 0.0001 and the other values are not significant compared to p < 0.01; paired two-tailed Student’s t-test. BALB/c and C3H/HeJ mice were infected with L. donovani AG83 and GE1F8R promastigotes and treated with Q-gal and K-gal as described above. Splenocytes (10⁶ cells) from different infected and treated groups were isolated and incubated with 50 μg/mL CSA in a 5% CO₂ incubator at 37 °C for 72 h. In vivo ROS production was determined by H2DCFDA at 525nm (E). The data shown are representative of three independent experiments.

Fig. S1

Homology modelling and molecular docking of Q-gal and K-gal with TLR4/MD2 (a) Molecular docking predicts possible interaction of the Q-gal and TLR4/MD-2. The heterotetrameric structure of TLR4/MD-2 was obtained from PDB (5IJB) and molecular docking was performed with Q-gal. The model depicts interaction of Q-gal with TLR4/MD-2 (A). Atoms of TLR4 is presented as surface in grey, atoms from MD-2 are represented as surface in olive and Q-gal is represented in ball and stick (A). The ligand binding site is visualized in 3D using PyMOL (B). Interacting residues were labelled and specific H-bonds are presented in ruby. A 2D representation of interactions between ligand and receptor was obtained using BIOVIA Discovery studio. Green spheres depict conventional while orange and pink sphere represent Pi-anion and Pi-alkyl H-bond donor/acceptor (C). (b) Molecular docking predicts possible interaction of the K-gal and TLR4/MD-2. The heterotetrameric structure of TLR4/MD-2 was obtained from PDB (5IJB) and molecular docking was performed with K-gal. The model depicts interaction of K-gal with TLR4/MD-2 (A). Atoms of TLR4 are presented as surface in grey, atoms from MD-2 are represented as surface in olive and K-gal is represented in ball and stick (A). The ligand binding site is visualized in 3D using PyMOL (B). Interacting residues were labelled and specific H-bonds are presented in ruby. A 2D representation of interactions between ligand and receptor was obtained using BIOVIA Discovery studio. Green spheres depict conventional while orange and pink sphere represent Pi-anion and Pi-alkyl H-bond donor/ acceptor (C).

Fig. S2

Effect of Q-gal and K-gal on Th1 cytokine generation in in vitro. IL-10 (A & B), and IL-12 (C & D) cytokines levels were assessed at mRNA level by semi quantitative RT-PCR. β-actin (E, F) was used as loading control. mRNA levels are normalized against β-actin and expressed as fold changes (mentioned in the figure) compared with infected controls. The results are representative of three experiments. For the original gels of IL-10, Il-12 & β-actin in Q-gal treated AG83 & GE1 infected cells, refer to supplementary figure number S8A, S8B, S8C, S8D, S8E & S8F respectively. For the original gels of IL-10, Il-12 & β-actin in K-gal treated AG83 & GE1 infected cells, refer to supplementary figure number S8G, S8H, S8I, S8J, S8K & S8L respectively.

Fig. S3

Effect of Q-gal and K-gal on Th1 cytokine generation in mice model. Inductions of predominant Th1 cytokine in Q-gal and K-gal treated mice were analyzed by semi-quantitative RT- PCR. Fold change in mRNA expression profiles of IL-12 (A, B & G, H) and IL-10 (C, D & I, J) in splenic lymphocytes of L. donovani-infected (60 days) and Q-gal and K-gal treated L. donovani-infected BALB/c and C3H/HeJ mice, respectively. Each gene was normalized against β-actin (E, F & K, L) and fold change was measured to account for differences between samples w.r.t the infected controls (mentioned in the figure). The results are representative of three experiments. For the original gels of IL-10, Il-12 & β-actin in Q-gal treated AG83 & GE1 infected mice, refer to supplementary figure number S9A, S9B, S9C, S9D, S9E & S9F respectively. For the original gels of IL-10, Il-12 & β-actin in K-gal treated AG83 & GE1 infected cells, refer to supplementary figure number S9G, S9H, S9I, S9J, S9K & S9L respectively.

Fig. S4

Schematic diagram of the in vivo infection and treatment in mice.

Fig. S5

Determination of the cytotoxic effect of Q-gal & K-gal on RAW264.7 cell line. Cell proliferation was assessed by MTT assays, on RAW264.7 cell line. Cells were treated with Q-gal (A) or K-gal (B) for 48 h. DMSO treated cells served as vehicle control (denoted as ‘0’) in both the cases. Concentrations of compound used were 12.5, 25, 50 and 100 μg/mL. The results are demonstrative of three independent experiments and data represents as mean ± SD; ∗p < 0.0001 and the other values are not significant compared to p < 0.01; paired two-tailed Student’s t-test.

Fig. S6

Effect of Q-gal and K-gal on expression of the TLR4 and TLR2. RAW264.7 cells infected with AG83 parasites for 24h and treated with Q-gal and K-gal and TLR4 (upper panel) and TLR2 (middle panel) levels measured by western blots. GAPDH was used as loading controls (lower panel. Blots are representative of three separate experiments. For the original blots of TLR4, TLR2 and GAPDH, refer to supplementary figure number S10A, S10B & S10C.

Fig. S7

Originals blots corresponding to Fig2A.

Fig. S8

Originals blots corresponding to Fig. S2.

Fig. S9

Originals gels corresponding to Fig. S3.

Fig. S10

Originals blots corresponding to Fig. S6.