All the animal experiments were carried out in accordance with the National Regulatory Guidelines issued by Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Ministry of Environment and Forest, Government of India. Use of Syrian Golden hamsters and Balb/c mice were approved by the Institutional Animal Ethics Committee of CSIR-Indian Institute of Chemical Biology, Kolkata, India with license number 147/1999/CPCSEA. Animals were housed under the standard condition such as temperature (25 ± 1°C), relative humidity (55 ± 10%) and 12 h/12 h light/dark cycles and fed with the standard diet.

Fluorescein isothiocyanate (FITC), bovine serum albumin (BSA), 4′, 6-diamidino-2-phenylindole (DAPI), Giemsa stain, and 2′-(4-Methylumbelliferyl)-α-D-N-acetylneuraminic acid (4MU-NeuAc), 4-methylumbelliferone (MU) were from Sigma (St. Louis, MO). Mounting medium was from Amersham Biosciences (Uppsala, Sweden); Maackia amurensis lectin II (MALII) and Sambucu snigra lectin (SNA) were from Vector Labs, and DyNAmo Color Flash SYBR Green qPCR kit was from Thermo Scientific (Rockford, IL). Anti-Neu1, cathepsin A was from Invitrogen (Carlsbad, CA), Anti-TLR4 antibody was from Santa Cruz Biotechnology (MTS510). Anti-Myd88 was from R&D Systems (MN, USA). Anti-phosphotyrosine antibody was from Biolegend (San Diego, CA). All the cytokine ELISA kits were from BD pharmingen, Neu1 plasmid DNA was from Origene (MR1049), Neu1 shRNA was obtained from Sigma (SHCLNG-NM010893), RNeasy Mini Kit was from Qiagen (Limburg, Netherlands); Reverse Transcriptase Kit was from Promega (WI, USA). All other antibodies were from Cell Signaling Technologies (Danvers, MA) unless indicated otherwise.

Promastigotes of an Indian L. donovani (L. d) strain AG83 (MHOM/IN/1983/AG83) were maintained in M-199 medium containing HEPES buffer (20 mM, pH 7.5) supplemented with 10% heat-inactivated fetal calf serum (FCS, v/v), and gentamycin sulfate (200 μg/ml) at 22°C as described before. To ensure the virulence, stationary phase promastigotes (2 × 10⁷ cells/100 μl) were routinely passaged in the golden hamster. After 4–6 weeks, hamsters were sacrificed and spleen tissue was collected in sterile phosphate buffer saline (PBS) to obtain amastigotes as described earlier (46).

Murine macrophage cell line J774A.1 obtained from ATCC was cultured in Iscov's modified Dulbeccos Medium (IMDM) supplemented with 10% heat-inactivated FCS and antibiotic- antimycotic solutions (complete medium) at 37°C with 5% CO₂. To maintain confluency, they were sub cultured every 3–4 days as described previously (47).

Macrophages J774.A1 cells (1 × 10⁶) were infected with stationary phase promastigotes at a multiplicity of infection 1:10 (macrophage: parasite) for 4 h. To remove unbound and/or loosely adhered parasites on the membrane of infected cells, they were repeatedly washed with IMDM. The presence of parasites on the surface of the infected macrophages was monitored under the microscope after each wash. Thereafter, the infection was allowed for additional 8 h.

Cells were extensively washed and resuspended in lectin-binding buffer (20 mM Tris, 0.5 M NaCl, 2.0 mM MnCl₂, 2.0 mM MgCl₂, 2.0 mM CaCl₂) (48) and incubated at 4°C separately with FITC-conjugated MALII (5.0 μg/ml) and FITC-SNA (4.0 μg/ml) which recognizes α2,3 and α2,6-linked sialic acids, respectively for 1 h (49). Thereafter, cells were washed twice with lectin binding buffer (300 μl, 2X) and finally resuspended in the same buffer. FITC positivity was acquired by FACS (FACS Calibur, BD Bioscience) and 10,000 cells were analyzed in CellQuestPro software.

Uninfected or infected J774.A1 cells were similarly processed after staining with anti-Neu1 antibody for 1 h at 4°C. This was followed with Alexa Fluor 488 conjugated secondary antibody for 1 h. Status of cell surface-bound Neu1 was demonstrated through Alexa Fluor 488-positivity acquired by FACS.

Uninfected and infected J774.A1 (2 × 10⁶) cells were resuspended in PBS, containing pepstatin A (1.0 μg/ml), aprotinin (10 μg/ml), and 10 μg/ml leupeptin (Lysis buffer). They were lysed by sonication (Qsonica-LLC, XL-2000 series, Newtown, CT, USA) and centrifuged at 800 × g for 10 min. The supernatant obtained was centrifuged at 1, 000, 00 × g for 30 min to separate cytosolic and membrane fractions (50). Protein was estimated using Bicinchoninic acid assay using bovine serum albumin (BSA) as standard.

Membrane fractions (100 μg/lane) were separated on gradient SDS-PAGE (7–15%) and transferred onto a PVDF membrane for glycoprotein analysis. The membrane was blocked with 4% desialylated BSA in tris buffer saline (TBS) for 1 h and incubated with biotinylated MALII and SNA separately for overnight at 4°C. They were additionally blotted with biotinylated Peanut Agglutinin (PNA). These were subsequently probed with avidin-HRP and signal was detected by West-pico enhanced chemiluminescent substrate (ECL) system (Pierce, Thermo Scientific, USA) as described earlier (41).

For detection of signaling molecules, an equal amount of protein from each sample was separated in SDS-PAGE (10%), transferred as before, blocked with TBS-2%BSA and incubated with specific primary antibody of p-JNK, JNK, p-ERK, ERK, p-p38, p38 (1:1,000 dilutions) overnight at 4°C. The blots were incubated with respective species-specific HRP-conjugated secondary antibody (1:1,000 dilutions) and developed by ECL kit.

For immunoprecipitation experiments, cell lysates (300 μg/lane), cytosolic (300 μg/lane), and membrane (400 μg/lane) fractions were incubated with specific antibodies (1:100) overnight at 4°C. The antibody-bound complex was pulled down with Sepharose-4B-protein-A. This was followed by washing with PBS and the complex was resolved in SDS-PAGE (10%) under non-reducing condition followed by western blotting. The densitometry score for each band was determined by ImageJ software and plotted as band intensities.

Sialidase activity of uninfected and L. donovani.-infected macrophages was determined using equal amount (100 μg) of protein from cytosol and membrane fractions by using a fluorimetric assay with an artificial substrate 4-MU-Neu5Ac (400 μM) as described elsewhere (29, 30). The reaction mixture was incubated at 37°C for 1 h in sodium acetate buffer (50 mM, pH 4.5). The reaction was terminated with 1.0 ml solution containing 0.133 M glycine, 0.06 M NaCl, and 0.083 M Na₂CO₃, pH 10.7. Liberated 4-methylumbelliferone was measured with a spectrofluorometer with excitation at 355 nm and emission at 460 nm. A standard curve of increasing amounts of 4-methylumbelliferone was used to determine the amount of 4-methylumbelliferone liberated from 4-MUNANA. Enzymatic activity was expressed as μM of product per h/mg protein.

Uninfected or infected J774.A1 cells (2 × 10⁴) were adhered on coverslips for 48 h, washed with PBS and fixed with 4% paraformaldehyde for 15 min. Fixed cells were stained with either FITC-MALII or FITC-SNA. Additionally, cells were also permeabilized by 0.5% TritonX-100. Cells before and after permeabilizations were incubated with rabbit anti-Neu1 antibody followed by staining with rabbit Alexafluor 488 antibody for visualizing the membrane and cytosolic Neu1. They were mounted in mounting media containing DAPI to stain nucleus and examined on a Zeiss inverted confocal microscope (51).

J774.A1 (1 × 10⁶ /well) in IMDM at 80% confluence were transfected with Neu1 plasmid DNA or Neu1 shRNA by the transfection reagent, lipofectamine-ltx according to the manufacturer's instructions in a six-well-plate for 6 h. The transfection mixture was replaced with fresh medium. Mock transfection was used as the control in all the experiments. They were infected with stationary phase promastigotes for 4 h.

Neu1-transfected and mock-transfected J774A.1 (1 × 10⁶ /well) were infected with stationary phase promastigotes of L. donovani at 1:10 ratio for 4 h in a six well plate. Unbound parasites were washed out and infection was allowed for additional 20 h. Cell-free culture supernatant was collected and assessed for the accumulated cytokines using a sandwich ELISA kit following the manufacturer's protocol (41) and nitrite using Griess reaction (52, 53).

Total RNA from Neu1-transfected and mock-transfected J774A.1 cells was extracted using the RNeasy mini kit following the manufacturer's instruction. First strand cDNA was synthesized by ImPromII-Reverse transcription system according to the manufacturer's protocol. Real-time PCR was performed with specific primers for Th1 (IL-12 and IFNγ), Th2 cytokines (IL-4, IL-10, and TGF-β), and iNOS obtained from Eurofins Genomics India Pvt. Ltd. (Table 1) using a DyNAmo Flash SYBR Green qPCR Kit. Relative amounts of target mRNA were quantitated using the Light Cycler 96 (Roche) software with 18S rRNA as an internal control. Data were expressed as a fold change compared with uninfected control using the comparative cycle threshold (CT) method (54).

Resident peritoneal macrophages from Balb/c mice (8–10 weeks old) were obtained by injecting 5–10 ml chilled PBS supplemented with 3% FCS into the peritoneal cavity (46). The peritoneal exudates cells were pulled out using a syringe, centrifuged (200 × g, 5 min), and washed with PBS. These cells (5 × 10⁵ cells) were grown on 22 mm glass coverslip in complete medium for 20 h. After washing non-adherent cells, adherent cells were transfected followed by infection with stationary phase promastigote. Unbound parasites were washed after 4 h and infection was allowed for further 20 h. After infection, the cells were washed with PBS, methanol fixed and stained with Giemsa. The amastigotes were counted manually under a light microscope (EVOS, Life Technologies) at oil immersion (100x) both in Neu1-transfected and mock-transfected cells (55).

All the data were the mean value from three independent experiments and statistical analysis was performed using Graph Pad Prism 5. Standard error bars represent the standard deviation of the mean (±S.D.) and statistically significant differences were determined using students t-test. *p ≤ 0.05, **p ≤ 0.01, and***p ≤ 0.001 were considered significant.

Altered sialylation is closely associated with different diseases. However, its correlation with impairment of innate immunity was not adequately studied in L. donovani infection. Therefore, initially, we checked the status of sialylation on macrophages during infection by FACS analysis. Accordingly, murine macrophage cells (J774.A1) were infected with stationary phase promastigotes. Infected cells exhibited enhanced α2,6 and α2,3-linked sialic acids as depicted by higher binding with SNA and MALII, respectively. The mean fluorescence intensity (MFI) of SNA bound cells increased significantly (P ≤ 0.001) from 238.7 ± 7.055 to 534.3 ± 6.386 (Figure 1A) and that of MALII (p ≤ 0.05) from 961.3 ± 33.96 to 1163 ± 37.17 (Figure 1B). This was further corroborated by lectin blotting. A few SNA-bound (Figure 1C) and MALII-bound (Figure 1D) membrane sialoglycoproteins showed the enhanced expression on infected cells compared to uninfected macrophages. Ponceau S staining of the same blot was used to show equal loading (Figure 1E). One of the upregulated α2,3 linked sialoglycoprotein on the surface of L. donovani. infected macrophages exhibited an apparent molecular weight of ~100 kDa (Figure 1D) which matched with the size of TLR4 protein as shown by western blotting (Figure 1F). This was further confirmed by confocal microscopy. Infected macrophages showed enhanced binding to FITC-SNA (Figure 1G) and FITC-MALII (Figure 1H) as indicated by more green color on the cell surface compared to uninfected macrophages. Macrophage nucleus was stained with DAPI in blue.

Enhanced cell surface sialylation upon L. donovani infection, prompted us to monitor the status of membrane-bound Neu1 on infected J774.A1 cells by flow cytometry (Figure 2A). Neu1 exhibited a significant (p ≤ 0.001) lower binding on the surface of infected macrophages than in uninfected cells (MFI 536.7 ± 3.180 vs. 750.7 ± 21.46, respectively).

For further confirmation, membrane fractions were isolated and checked for level of Neu1 by western blot. Densitometry analysis showed a ~2.5-fold decrease of Neu1 in the membrane fraction of infected cells compared to uninfected cells (Figure 2B). Caveolin was used to show the purity of the plasma membrane. As expected, LPS-stimulated macrophages exhibited a higher amount of Neu1 on the membrane under similar conditions, which was used as a positive control.

In order to check whether altered Neu1 protein level is associated with a change in its enzyme activity, macrophages were infected with L. donovani and the lysates were used as the source of sialidase enzyme. The sialidase activities in uninfected vs. infected macrophages were determined using an exogenous substrate 4-MU-NANA. This assay showed a significant (p ≤ 0.05) reduction in the sialidase activity in the membrane fraction of infected macrophages (Figure 2C).

This finding was further confirmed by confocal microscopy (Figure 2D). Immunostaining with anti-Neu1 antibody followed by Alexa fluor 488 secondary antibodies exhibited significantly (p ≤ 0.01) decreased expression of Neu1 on the cell surface of infected macrophages as indicated by less green color when compared to uninfected macrophages. DAPI stained the intact cell nuclei. The mean intensity graph of green fluorescence was shown in Figure 2E.

We had observed enhanced cell surface sialylation on infected macrophages (Figures 1A,B). As TLR4 is a membrane-bound sialylated glycoprotein, accordingly, we checked the sialylation status of TLR4 in L. donovani-infected macrophages. We immunoprecipitated the cell lysates from both uninfected and infected macrophages with anti-TLR4 antibody and immunoblotted with MALII. An enhanced α2,3-sialyl residues on TLR4 were observed in infected macrophages compared to uninfected cells (Figure 3A). LPS, a well known TLR4 ligand, was used as a positive control. As expected, under similar conditions, LPS-stimulated macrophages exhibited decreased TLR4 sialylation on the membrane in contrast to L. donovani infected cells. Our observations suggest some close relationship between enhanced sialylation of TLR4 and parasite infection.

Our data exhibited reduced membrane-bound Neu1 (Figure 2) and enhanced α2,3-linked sialic acids on TLR4 (Figure 3A) which intrigued us to check the effect of this enhanced sialylation on its association with Neu1 (Figure 3B). Accordingly, immunoprecipitated TLR4 was immunoblotted with anti-Neu1 antibodies. We observed Neu1-TLR4 association was reduced significantly in parasite-infected macrophages than its uninfected counterpart. In contrast, LPS-stimulated macrophages exhibited just opposite effect indicating a role of Neu1 in the enhancement of TLR4 sialylation.

So far we have observed enhanced sialylation of TLR4 possibly due to reduced Neu1-TLR4 association in parasite-infected cells (Figure 3B). This prompted us to investigate whether such hyper sialylation of TLR4 affects its downstream signaling. At the initial stage, TLR4 signaling involves its association with the key adaptor molecule MyD88 for its further activation. To address this, we checked the recruitment of TLR4 with MyD88 (Figure 3C).

Accordingly, immunoprecipitated TLR4 was immunoblotted with anti-MyD88 antibodies. We observed a reduction in TLR4–MyD88 association in infected macrophages indicating some involvement of enhanced sialylation of TLR4 during parasite infection. However, LPS-treated cells showed a better association of TLR4 with MyD88 indicating proper initiation of downstream signaling.

To specifically pinpoint the effect of Neu1 on TLR4 during parasite infection, we overexpressed Neu1 in macrophage by transfecting a plasmid encoding the Neu1 gene followed by infection with parasite (Figure 4A). Enhanced Neu1 in the Neu1-transfected cells ensured its overexpression compared to the mock-transfected cells. Penultimate sugar of terminal sialic acid moiety on sialoglycoconjugate is usually galactose. Accordingly, we checked the binding of a galactose-binding lectin (PNA) with Neu1-transfected parasite-infected cells (Figure 4B). PNA blotting depicted an enhanced binding with exposed galactose units on the Neu1-transfected cells due to the removal of sialic acids by the increased Neu1 compared to mock-transfected cells. To further confirm the overexpression, we measured the sialidase activity of the membrane fractions of Neu1-transfected cells (Figure 4C). Approximately 3.2-fold (p ≤ 0.001) increased enzyme activity was observed in the membrane fraction of Neu1-transfected cells compared to mock transfection during L. donovani infection.

Furthermore, we checked the effect of overexpressed Neu1 on the parasite-infected macrophages by confocal microscopy (Figure 4D). A decreased level of MALII-binding suggested decreased α2,3-linked sialic acids on the cell surface of Neu1-transfected parasite-infected cells. All these observations confirmed the overexpression status of Neu1 in parasite-infected cells.

We next checked the sialylation status of TLR4 in Neu1-transfected cells after L. donovani infection. Therefore, we immunoprecipitated the cell lysates from Neu1-transfected cells with anti-TLR4 antibody and immunoblotted with MALII. We observed decreased sialylation of TLR4 in the Neu1-transfected cells compared to mock transfection during infection (Figure 5A). Furthermore, we checked the association between Neu1 and TLR4 in Neu1-transfected cells in infected condition. As expected, we observed an enhanced Neu1-TLR4 association in Neu1 overexpressed condition (Figure 5B). Moreover, infection with L. donovani in Neu1-transfected cells showed enhanced TLR4-MyD88 association (Figure 5C). All these findings suggest that infection after Neu1 overexpression leads to TLR4 activation through its desialylation. This further confirms Neu1 does play an important role in cellular activation which is otherwise hampered during L. donovani infection.

Enhanced Neu1-TLR4 and TLR4-MyD88 associations in Neu1-overexpressed cells further intrigued us to explore the role of sialic acids on TLR4 in TLR4-mediated downstream signaling during parasite infection. We, therefore, checked the status of different signaling molecules of the MAP-kinase pathway in Neu1-transfected cells after L. donovani infection. We observed enhanced phosphorylation of JNK, ERK1/2, and p38 MAPK in Neu1-transfected cells compared to mock-transfected cells (Figure 6A). The band intensities of the blots are shown (Figure 6B).

Nuclear translocation of a transcription factor NFκB leads to transcription of proinflammatory genes. We, therefore, assessed the nuclear localization of the functional subunit of NF-κB namely p65 (Figures 6C,D). The relative band intensities of the blots are shown (Figure 6E). Parasite infection of Neu1-transfected macrophages increased the expression of IKKα and IKKβ in the cytosolic fraction. This enhanced IKKα and IKKβ lead to enhanced phosphorylation of IκBα, causing its degradation and ultimately sets free the functional subunit of NF-κB, p65. This lead to a reduced p-65 in the cytosol (Figure 6C) due to enhanced nuclear translocation (Figure 6D). Thus, we observed a good correlation between Neu1-mediated desialylation of TLR4 and activation of the MAPK pathway in addition to NF-κB nuclear translocation indicating higher cellular activation thereby possible inhibition of parasite infection in presence of enhanced Neu1.

Enhanced nuclear translocation of functional transcription factor (p65 subunit), generally leads to enhanced generation of Th1 cytokines. Therefore, we have assessed both Th1 and Th2 cytokines in Neu1-transfected cells upon L. donovani infection (Figure 6F). We observed an enhanced genetic expression of the Th1 cytokines namely IL-12 (~2.6-fold, p ≤ 0.05) and IFNγ (~1.9-fold, p ≤ 0.05) with down-regulated expression of Th2 cytokines IL-4 (~2.1-fold, p ≤ 0.01), IL-10 (~1.8 fold, p ≤ 0.05), and TGFβ (~3.3-fold, p ≤ 0.01) in Neu1-transfected cells.

Additionally, we also measured the level of a few secreted cytokines in the cell supernatant of Neu1-transfected parasite-infected cells by ELISA (Figure 6G). The level of the secreted Th1 cytokines IFNγ was also significantly (p ≤ 0.001) increased by ~2.6-fold in Neu1-transfected parasite-infected cells. In contrast, there was a significant (p ≤ 0.01), reduction of the Th2 cytokines IL-10(~2.6-fold) and IL-4 (~2.8-fold, p ≤ 0.01) compared to infected mock-transfected cells. This observation well-corroborated with the genetic expression data.

Nitric oxide is an important effector molecule involved in host defense against the parasite. Therefore, we checked the genetic expression of iNOS gene. We observed a significantly (p ≤ 0.05) enhanced iNOS expression of 2.3-fold in Neu1-transfected cells after parasite infection (Figure 6F). Additionally, secreted nitric oxide (NO) level in the supernatant was determined by Griess assay (Figure 6H). Approximately, ~4-fold enhancements in the NO level (p ≤ 0.001) were found in parasite-infected Neu1-transfected cells compared to its mock-transfected counterpart.

So far we have demonstrated that Neu1-transfection induced enhanced TLR4-mediated activation of several signaling molecules in the MAPK pathway which leads to enhanced Th1 cytokines along with iNOS and NO suggesting increased effector immune response in these parasite-infected cells. Therefore, the next obvious step was to check the actual status of the parasite burden in these Neu1-transfected cells.

Accordingly, resident peritoneal macrophages cells were transfected with Neu1 plasmid followed by L. donovani infection and the parasite load was determined (Figure 7A). We observed a decreased number of intracellular amastigote in the Neu1-transfected cells after staining with Giemsa (purple dots indicated by black arrow) than mock-transfection. Corroborating with this data, there was also a significant (p ≤ 0.001) reduction in the number of amastigotes per 100 macrophages in Neu1-transfected cells (Figure 7B). This was further reflected by a significant (p ≤ 0.001) reduction in the percentage of infected macrophages in the Neu1-transfected cells compared to mock counterpart suggesting effective activation of the immune response leading to parasite killing due to overexpression of Neu1 (Figure 7C).

So far we have demonstrated that Neu1-overexpression resulted in activation of the innate immune response in the parasite-infected cells. This led us to further investigate the effect of Neu1 silencing on TLR4 during this parasite infection. Accordingly, the cells were transfected with Neu1 shRNA and thereafter infected with the parasites. We observed reduced Neu1 protein level expression in these transfected cells compared to the mock thus confirming Neu1 silencing (Figure 8A). The Neu1-silenced parasite-infected cells exhibited an enhanced MALII binding (p ≤ 0.05) when compared to the mock-transfected infected macrophage by confocal microscopy (Figure 8B). The fluorescence intensity graph is shown.

As expected, we observed a reduced Neu1-TLR4 association in the Neu1-silenced parasite-infected macrophages compared to mock-infected cells (Figure 8C). Earlier, we have observed decreased TLR4-MyD88 association in infected cells (Figure 3C), here we were unable to observe any further detectable reduction in association of TLR4 with MyD88 after Neu1 silencing (Figure 8D). This observation indicates reduced Neu1 on the parasite-infected cells is indeed responsible for impaired TLR4 activation.

Neu1 is usually complexed with β-galactosidase and the serine carboxypeptidase protective protein/cathepsin A. Cathepsin A is essential for the catalytic activity of Neu1. So far we have observed a decrease in Neu1 activity in the membrane fraction of infected macrophages with hypersialylation in TLR4. Therefore, we checked the status of cathepsin A in L. donovani infected macrophages. An enhanced level of cathepsin A was demonstrated during this infection (Figure 9A). Furthermore, immunoblotting of cytosolic Neu1 with cathepsin A exhibited an increased association in the cytosolic fraction (Figure 9B) which was reversed when membrane fraction was used (Figure 9C) suggesting less available Neu1-cathepsin A association on the membrane of L. donovani infected macrophages.

It has been reported that in activated lymphocytes Neu1 phosphorylation leads to its membrane translocation. To address if such phosphorylation plays any role in surface translocation of Neu1 during infection, we immunoprecipitated Neu1 from the cell lysate of uninfected and infected cells with anti-Neu1 antibody and immunoblotted with anti-phosphotyrosine antibody. We observed a decreased level of Neu1-phosphorylation in cell lysate during infection (Figure 9D). A similar trend was observed both in the cytosol (Figure 9E) and membrane (Figure 9F) fractions. This was corroborated by the fluorescence microscopy (Figure 9G). The mean intensity graph of intracellular Neu1 is also shown (Figure 9H). Western blot analysis further revealed more cytosolic Neu1 during infection (Figure 9I). Thus, this reduced tyrosine phosphorylation may probably be one of the reasons for preventing the translocation of Neu1 to the cell surface of L. donovani infected macrophages.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.