Leishmania donovani Utilize Sialic Acids for Binding and Phagocytosis in the Macrophages through Selective Utilization of Siglecs and Impair the Innate Immune Arm

Background Leishmania donovani, belonging to a unicellular protozoan parasite, display the differential level of linkage-specific sialic acids on their surface. Sialic acids binding immunoglobulin-like lectins (siglecs) are a class of membrane-bound receptors present in the haematopoetic cell lineages interact with the linkage-specific sialic acids. Here we aimed to explore the utilization of sialic acids by Leishmania donovani for siglec-mediated binding, phagocytosis, modulation of innate immune response and signaling pathways for establishment of successful infection in the host. Methodology/Principle Findings We have found enhanced binding of high sialic acids containing virulent strains (AG83+Sias) with siglec-1 and siglec-5 present on macrophages compared to sialidase treated AG83+Sias (AG83-Sias) and low sialic acids-containing avirulent strain (UR6) by flow cytometry. This specific receptor-ligand interaction between sialic acids and siglecs were further confirmed by confocal microscopy. Sialic acids-siglec-1-mediated interaction of AG83+Sias with macrophages induced enhanced phagocytosis. Additionally, sialic acids-siglec-5 interaction demonstrated reduced ROS, NO generation and Th2 dominant cytokine response upon infection with AG83+Sias in contrast to AG83-Sias and UR6. Sialic acids-siglecs binding also facilitated multiplication of intracellular amastigotes. Moreover, AG83+Sias induced sialic acids-siglec-5-mediated upregulation of host phosphatase SHP-1. Such sialic acids-siglec interaction was responsible for further downregulation of MAPKs (p38, ERK and JNK) and PI3K/Akt pathways followed by the reduced translocation of p65 subunit of NF-κβ to the nucleus from cytosol in the downstream signaling pathways. This sequence of events was reversed in AG83-Sias and UR6-infected macrophages. Besides, siglec-knockdown macrophages also showed the reversal of AG83+Sias infection-induced effector functions and downstream signaling events. Conclusions/Significances Taken together, this study demonstrated that virulent parasite (AG83+Sias) establish a unique sialic acids-mediated binding and subsequent phagocytosis in the host cell through the selective exploitation of siglec-1. Additionally, sialic acids-siglec-5 interaction altered the downstream signaling pathways which contributed impairment of immune effector functions of macrophages. To the best of our knowledge, this is a comprehensive report describing sialic acids-siglec interactions and their role in facilitating uptake of the virulent parasite within the host.


Introduction
Visceral leishmaniasis (VL) caused by the species Leishmania donovani, is the most severe form of leishmaniasis and is endemic in the region of Indian subcontinent. Due to its digenic nature, parasites multiply as an amastigote form within the phagolysosome of macrophages in the liver and spleen of the host [1]. Thus, entry and survival inside the macrophage is an imperative factor for the fulfillment of pathogenesis.
Sialic acids (Sias) are acidic sugars containing nine carbon backbones predominantly present in the terminal residue of the cell surface as well as secreted glycoproteins and glycolipids [2]. Sias participate in an array of cellular functions such as ligand for sialic acids binding immunoglobulins like lectins (siglecs), shielding element for many receptors, negatively charged barrier for various cells even pathogens, regulation of immune cell activation and leukocyte extravasations [3]. An array of human pathogens such as Haemophilus influenza, Pseudomonas aeruginosa, Escherichia coli, Neisseria meningitides, Campylobacter jejuni and Trypanosoma cruzi either acquire or synthesize Sias for successful infection by dampening the host immune system [4,5].
Siglecs come under the group of immunoglobulin-type (I-type) lectins present mainly on haemotopoetic cell lineages with a vast structural diversity in recognition of their ligands. Fourteen human and nine murine siglecs have been reported so far [6]. Siglec-3/CD33 group is the major group of siglecs with a high degree of interspecies sequence homology. Most of the siglecs contain immunoreceptor tyrosine-based inhibitory motif (ITIM) in the cytoplasmic domain. One of the exceptions is siglec-1 (CD169/sialoadhesin), which has an extended extracellular structure with no ITIM bearing motif in cytosolic region [7]. ITIM motif gets activated upon ligand binding to the siglec and recruits SH2-domain containing protein tyrosine phosphatase 1/2 (SHP-1/2) to carry out various signaling pathways [8].
We have previously reported the presence of sialoglycans, especially 9-O-acetylated form of Sias both on promastigote and amastigote of Leishmania donovani [9][10][11][12][13][14][15][16]. Although the function of 9-O-acetylated Sias in entry and virulence of parasite was reported earlier, no systemic investigation on the role of sialic acids on dampening of host innate immune response by modulation of signaling pathways has been addressed properly. Macrophages are the safe house of this parasite. Sialic acids recognizing receptors (siglec-1 and siglec-5) are present in this immune cell. Hence the detailed study of the involvement of these siglecs in binding, phagocytosis, modulation of SHP-1 and downstream signaling pathways during Leishmania infection became worthwhile.
Our current findings established the involvement of sialic acids of L. donovani in siglec-1 mediated binding and phagocytosis in macrophages. Furthermore, we demonstrated altered downstream signaling pathways leading to suppression of effector functions of innate immune response specifically due to this Sias-siglec-5 interaction. Such in-depth comparative study between high Sias containing virulent strain AG83 (AG83 +Sias ) and sialidase treated AG83 (AG83 -Sias ) along with low Sias containing avirulent strain UR6 conclusively ascertain the important role of sialic acids in siglec-mediated interaction and altered immune regulation for establishment of successful infection.

Ethics statement
All the animal-related experiments were performed according to the National Regulatory Guidelines issued by Committee for the Purpose of Control And Supervision of Experiments on Animals (CPCSEA), Ministry of Environment and Forest, Govt. of India. Use of Balb/c mice was 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 standard condition such as temperature (25 ± 1°C), relative humidity (55 ± 10%) and 12 hr/12 hr light/dark cycles and fed with standard diet.

Dot blot
Dot blot was performed as described previously with some modifications [22]. Briefly, log phase AG83 +Sias promastigotes (1×10 7 ) were suspended in phosphate buffer saline (PBS) and blotted onto the nitrocellulose membrane. After air drying, blots were blocked in PBS-BSA (3%), overlaid with soluble human siglec-Fc chimeras, washed and incubated with horseradish peroxidase (HRP)-protein-A for 2 hr at 25°C. After washing, blots were developed by diamino benzidine (DAB) in H 2 O 2 solution and quantified by densitometric analysis using Master Totallab Software, version 1.11.
Sialidase treatment and FITC labeling of parasites L. donovani promastigotes grown in FCS-containing medium is designated as parasite +Sias . They were treated with sialidase as described previously with some modifications [9]. Briefly, promastigotes were initially incubated with 9-O-acetyl haemagglutinin esterase of Influenza C virus (100 U/ml, 100 μl/tube) for 1 hr at 37°C to remove O-acetyl group. Subsequently, esterase-treated cells were washed and incubated with Arthrobacter ureafaciens sialidase (100 mU/ ml, 10 mU/100 μl/tube) for 30 min for removal of Sias from the surface. Parasites were washed and designated as parasite -Sias . Parasite +Sias and parasite -Sias were labeled with FITC as stated earlier [21]. Log phase promastigotes were suspended in 0.1% FITC containing carbonate buffer (pH 8.0) and incubated for 1 hr at 22°C [19]. Parasites were washed and suspended in medium supplemented with FCS (2%).

Flow cytometric analysis of phagocytosis
Phagocytosis was performed as stated earlier with modifications [24]. J774A.1 cells (1×10 6 ) were preincubated with anti-murine-siglec-1(anti-Sn) for 1 hr at 4°C in PBS-BSA (2%) or left untreated. These cells were infected with FITC-AG83 +Sias and AG83 -Sias (1: 10 ratio) for 2 hr at 37°C, which is the physiological temperature for phagocytosis. Unbound parasites were removed and the fluorescence intensity coming from extracellular non-phagocytosed FIT-C-AG83 was quenched by trypan blue treatment. Trypan blue quench the fluorescence of FITC by shifting the fluorescence from Fl1 channel to Fl3 channel. So, no signal from the surface of the cell will be detected in Fl1. Only the intracellular FITC signal will be detected in the Fl1. Therefore, visible intracellular fluorescence intensity was only due to phagocytosed parasites. This fluorescence was determined by flow cytometry.

Measurement of extracellular phagocytosis
Siglec1-dependent phagocytosis of AG83 +Sias promastigote was also visualized by confocal microscopy [21]. J774A.1 cells were left untreated or treated with anti-murine-siglec-1, were infected with FITC-AG83 +Sias (sialidase-treated/untreated) for 2 hr at 37°C. Unbound parasites were washed; extracellular fluorescence was quenched by trypan blue and processed similarly as described in the previous section of confocal microscopy.

Quantitation of nitric oxide (NO)
Nitric oxide was measured in cultured supernatant by Griess reaction as stated earlier [25]. Briefly, J774A.1 cells (1×10 6 /well) were infected with AG83 +Sias or AG83 -Sias or UR6 at 1:10 ratio for 4 hr in a six well plate. After the infection, unbound parasites were washed out and the cells were kept in desialylated FCS-containing medium for additional 20 hr. FCS was desialylated by treating with 2M propionic acid at 80°C for 2.5 hr followed by extensive dialysis. Culture supernatant was collected and accessed the accumulated nitrite.

Measurement of intracellular parasites by Giemsa staining
J774A.1 cells (2×10 4 ) were grown on glass coverslip and left untreated or blocked with antimurine siglec antibodies (anti-siglec-1,-E,-F) for 1 hr. They were infected with AG83 +Sias or AG83 -Sias or UR6 at 1: 10 ratio for 4 hr. Unbound parasites were removed. Infection was continued for additional 20 hr in desialylated serum-containing IMDM medium. The coverslip was methanol fixed and stained with Giemsa. Intracellular parasites were counted manually in light microscope (Ziess) at oil immersion (100×). The value was represented by amastigote number per 100 macrophages [27].
Genetic expression profiling by reverse transcriptase (RT) and real time PCR J774A.1 cells were infected and stimulated similarly as stated in previous section and total mRNA was isolated by RNeasy Mini Kit as per manufacturer's instructions. RNA (1.0 μg) was reverse transcribed to cDNA by Reverse Transcriptase kit. For RT-PCR analysis, the cDNA was amplified using specific primers targeting genes encoding for the cytokines namely IL-2, IL-4, IL-10, IFNγ, TGFβ, GAPDH and iNOS in a perkin-Elmer DNA thermal cycler. Primer sequences were listed in Table 1. The reaction product was separated in 1% agarose gel stained with ethidium bromide and image captured in BioRad Gel Documentation system [23]. Densitometric score was measured by ImageJ software.
Real time PCR analysis was performed by mixing cDNA with 2× SYBR green master mix on Roche Applied Science light cycler 480.0 instrument using software version 1.5.0. Relative quantification of each target gene were normalized to two housekeeping genes (18s rRNA and α β-actin) and expressed as a fold change compared with uninfected control using the comparative cycle threshold (CT) method [31].
For the immunoprecipitation assay, J774A.1 cells were infected with AG83 +Sias and AG83 -Sias for 2 hr at 37°C. They were processed similarly as stated in the above paragraph. Cell lysate was incubated with anti-SHP-1 antibody (1:100) overnight at 4°C. SHP-1 bound complex was pulled down with Sepharose-4B-protein-A. After washing, the complex was resolved by SDS-PAGE (10%), transferred to nitrocellulose membrane, incubated with anti-SHP-1 and anti-siglec-E antibodies and processed for ECL [32]. Densitometric score was measured by ImageJ software.

Subcellular fractionation
J774A.1 cells (2×10 6 /well) were left uninfected or infected with AG83 +Sias or AG83 -Sias or UR6 for 12 hr in 1:10 ratio in a six well plate. After infection, unbound parasites were removed. Cells were washed and fractionated into cytosol and nuclear portion by NE-PER kit according to manufacturer's protocol. Briefly, cell pellet was suspended in cytosolic extraction reagent, vortexed, spin down and the supernatant was collected as cytosolic fraction. The residual mass was suspended in nuclear extraction reagent, vortexed rigorously, spin down and the supernatant was used as nuclear fraction [32]. Both the fractions were separated in SDS-PAGE gel, transferred and processed for western blot as described in previous section and developed similarly.

Knockdown of siglec-E by siRNA treatment
Short interfering RNA (siRNA) was designed against murine siglec-E and transiently transfected in pre-seeded J774A.1 (1×10 6 ) cells using lipofectamine and plus reagent according to the manufacturer's instructions. Cells were incubated with siRNA for 6 hr. Transfection reagent was removed and further cultured in complete medium for 18 hr. Confirmation of siRNA-mediated knockdown of target gene was evaluated by western blotting [26]. Transfected and untransfected cells (absence of siRNA) were infected with AG83 +Sias . Status of several effector molecules such as ROS, NO, cytokines and various signaling molecules were determined similarly as stated in the previous sections.

Statistical analysis
The data were the mean value derived from at least three independent experiments. Statistical analysis was performed using the two-tailed Student's t test. Error-bars represent mean ± SD from three independent experiments. Significant differences were set at Ã p 0.05, ÃÃ p 0.01, ÃÃÃ p 0.001 and analyzed by GraphPad Prism version 5.01.

Results
Sialic acids containing promastigotes exhibited enhanced siglec-1 and -5-mediated binding with macrophages Although the differential distribution of Sias along with 9-O-acetylated Sias in virulent (AG83 and GE1) and avirulent (UR6) promastigotes of L. donovani were earlier reported [11], the presence of α2-3 and α2-6 linked Sias was further evaluated on the surface of different virulent SAG sensitive (AG83 Dd8 and 2001) and SAG resistant (39 and GE1) typed/untyped strains along with an avirulent UR6 promastigotes. The comparative bindings of FITC-SNA and FITC-MAL were tested in all the strains by flow cytometry. Promastigotes (parasite +Sias ) except the strain 39 showed significantly higher (p 0.05) SNA binding than MAL suggesting the predominance of α2-6 linkage on the surface of L. donovani. Sialidase-treated promastigotes (parasite -Sias ) exhibited minimal binding (S1 Fig).
Additionally, CHO-siglec-5 cells were treated with exogenous sialic acids (1.0 mM) prior to binding with parasite. There was significant (p 0.01) reduction in the binding of AG83 +Sias with pre-treated CHO-siglec-5 with MFI value being 2982 ± 378 compared to unblocked CHO-siglec 5 (4927 ± 422). However exogenous Sias treated CHO-siglec-1 cells showed no significant change in MFI value (8336 ± 680) compared to untreated control (9024 ± 805) under identical condition. This may be due to the elongated structure of siglec-1 that protruded way out from the cell surface and affinity towards linkage-specific sialic acids.
To confirm whether affinity of AG83 +Sias towards siglec-1 and -5 was a generalized phenomenon for L. donovani, this binding was further examined in other L. donovani strains also. Different Sias-siglec interaction played a crucial role in binding of promastigotes with the macrophages As AG83 +Sias promastigotes exhibited preferentially higher binding towards CHO-siglec-1 and CHO-siglec-5, next we examined such interaction using both murine macrophage (J774A.1) and human monocytic cells (THP-1) as they are the natural host of Leishmania parasite. Accordingly, the comparative binding of AG83 +Sias /AG83 -Sias and UR6 with J774A.1/THP-1 cells was evaluated in the light of Sias-siglec interaction.
Additionally, infection of J774A.1 with AG83 -Sias also showed decreased (p 0.001) binding compared to AG83 +Sias . Interestingly binding of UR6 with macrophages remained unaltered in all these conditions. Similar trend of binding was observed when human monocytic cell line THP-1 was used (Fig 2B).
Sias-siglec-mediated interaction of FITC-AG83 +Sias promastigotes with J774A.1 was further visualized by confocal microscopy and shown as a representative image. Higher number of parasites attached to the macrophage surface was observed as indicated by green dots of FITC-AG83 +Sias on the PE-CD14 stained red colored periphery of the J774A.1 cells (Fig 2C). DAPI stained nucleus indicating intactness of the cell. Blocking of siglecs present on macrophages by specific anti-siglecs antibodies or removing sialic acids from parasites by sialidase treatment exhibited a fewer number of FITC-AG83 attached on the macrophage surface. As Sias present on UR6 are generally low, there were no apparent changes in their association with J774A.1 cells before and after inhibition of Sias-siglec interaction ( Fig 2D).
Sias-siglec mediated phagocytosis of AG83 +Sias showed higher MFI value which was significantly reduced (p 0.05 and 0.01) when this interaction was inhibited either by blocking of siglec-1 with specific antibody or removal of sialic acids from AG83 +Sias (AG83 -Sias ) before infection. This value was further reduced (p 0.001) when Sias-siglec-1 dependent phagocytosis was inhibited simultaneously both by blocking siglec-1 on macrophage and removing sialic acids from parasite.
A similar trend was observed in confocal microscopy also in which infected J774A.1 cells showed comparatively higher number of intracellular parasites as indicated by green dots than treated macrophages where Sias-siglec interaction was blocked, further confirming the role of sialic acids and siglec-1 in phagocytosis (Fig 3B).

Sias-mediated interaction modulated the innate effector functions
ROS and NO are two major effector molecules for the macrophage defense arsenal, which encounter the evading pathogens. As Sias-siglec interaction played a major role both in binding and phagocytosis of AG83 +sias in macrophages, we have extended our study to determine its effect in modulation of effector function. Due to the minimal presence of sialic acids on AG83 -Sias or UR6, J774A.1 cells were able to produce significantly higher (p 0.001) amount of ROS compared to infection with AG83 +sias as indicated by higher H 2 DCFDA positivity ( Fig 4A).
Furthermore, during the AG83 +Sias infection, there was a reduction in the NO level in the culture supernatant indicating Sias-mediated interaction in modulation of innate effector function of the macrophages (Fig 4B). In contrast, infection with the AG83 -Sias or UR6, where Sias content is low, exhibited enhanced NO production (p 0.05 & p 0.01) indicating less infection.

Sias-mediated interaction modulated the cytokine level in infected macrophages
Leishmania infection is generally associated with Th2 type of response in vitro. As genes ultimately regulate the expression of proteins through mRNA, so the actual status of any changes of cytokines and NO synthesizing gene iNOS during infection were evaluated at genetic level through reverse transcriptase (Fig 4C) as well as real time PCR (Fig 4D). Due to infection of amounts on J774A.1/THP-1 cells. Subsequently, FITC-AG83/UR6 promastigotes were incubated with untreated and treated J774A.1/THP-1 cells at 1:10 ratio and the binding was measured by gating the macrophages in FITC channel. Data were analyzed by CellQuestPro software. Data is represented as mean ± S.D from three independent experiments. *, p 0.05; **, p 0.01; ***, p 0.001. (C) J774A.1 cells were adhered on the glass coverslip, left untreated or treated with anti-siglec-1,-E,-F antibodies. FITC-AG83 +Sias or sialidase-treated (AG83 -Sias ) were incubated with treated and untreated J774A.1 cells at 4°C. Unbound promastigotes were removed; macrophages were stained with phycoerythrin (PE)-anti-CD14 and mounted with mounting media containing DAPI. Images were captured by Andor Spinning Disc Confocal microscope. Green colour represents the FITC-promastigote while red color indicated the surface of macrophages and DAPI stained nucleus were of blue color. (D) J774A.1 cells were adhered on the glass coverslip and treated in the same manner as in Fig. 2B. Cells were infected with FITC-UR6 promastigotes, stained and visualized by confocal microscopy as described in Fig. 2C.   well) were left uninfected or infected with AG83 +Sias or sialidase-treated (AG83 -Sias ) or UR6 promastigotes in a six well plate at 1:10 ratio for 1 hr. Unbound parasites were removed and intracellular ROS production was measured by H 2 DCFDA staining by FACS. Data was analyzed by CellQuestPro software. (B) J774A.1 cells (1×10 6 / well) were kept uninfected or infected with AG83 +Sias or sialidase-treated (AG83 -Sias ) or UR6 similarly for 4 hr. Unbound parasites were washed and infection was allowed for additional 20 hr. The NO secretion was measured in the culture supernatant by Griess reaction. (C) J774A.1 cells (1×10 6 /well) infected with AG83 +Sias or AG83 -Sias or UR6 for 4 hr as stated in Fig. 4B. Infection was allowed for additional 20 hr in the presence of LPS (2.5 μg/ml) or PMA (25 ng/ml)-ionomycin (1.0 μg/ml) as stimulator for cytokines secretion. The total RNA was isolated by Quigen RNeasy Mini Kit. cDNA was prepared from RNA by reverse transcriptase Kit. Then the specific expression of the Th1 (IL-2, IFNγ) and Th2 (IL-4, IL-10, TGFβ) cytokine genes along with iNOS expression was measured in PCR by using specific primers as listed in Table 1. GAPDH was used as loading control throughout. (D) Total RNA was isolated from infected and uninfected J774A.1 cells and cDNA was prepared similarly as stated in Fig. 4C. The level of mRNA expression of Th1 and Th2 cytokines along with iNOS were quantified by real time PCR analysis. The mRNA level was normalized by 18s rRNA and the fold change was measured using uninfected cells as the control. Comparison was also made of AG83 +Sias infection with AG83 -Sias or UR6 infection. (E) Secreted cytokines were also measured in BMDM cells. BMDMs were isolated from the femur of healthy mice and macrophages were infected and stimulated by LPS or PMAionomycin as stated in Fig. 4C. The secreted cytokines (IL-4, IL-10 and IFNγ) in the culture supernatants were measured by respective ELISA kit as described in material and methods. (F) J774A.1 cells (1×10 4 ) were adhered overnight on glass coverslip. Macrophages were left untreated or blocked with anti-siglec antibody cocktail (anti-siglec-1,-E,-F) for 1hr. Untreated and treated cells were infected with AG83 +Sias or AG83 -Sias or UR6 for 4 hr at 1:10 macrophages with AG83 +Sias , reduction of the genetic expression of Th1 cytokine genes (IFNγ and IL-2) and upregulation of Th2 cytokine genes (IL-4, IL-10 and TGFβ) were observed ( Fig  4C). Removal of Sias from AG83 surface reversed the expression pattern of Th1/Th2 in the infected macrophages. This was further confirmed by infecting macrophages with avirulent UR6 strain. As expected, there was a significant reduction in expression of Th2 cytokines and upregulation of Th1 cytokines indicating the host protective immune response. The status of iNOS expression was also reduced upon infection of macrophages with AG83 +Sias compared to AG83 -Sias or UR6 infection.
This was further confirmed by evaluating mRNA transcript of different Th1 and Th2 cytokines by real time PCR. AG83 +Sias infected macrophages showed Th2 bias response, which was reversed to Th1 in AG83 -Sias and UR6-infected cells (Fig 4D). Th2 cytokines (IL-4, IL-10 and TGFβ) were upregulated by~7.2, 6.3 and 8.3 folds respectively in macrophages infected with AG83 +Sias , which were significantly (p 0.05 and p 0.01) down regulated to~4.3, 4 and 6 folds respectively when AG83 -Sias was used. Infection with UR6 showed the further reduction in the Th2 cytokines to~3.5, 2.8 and 3.5 folds respectively. On the other hand, Th1 cytokines (IL-2 and IFNγ) were significantly (p 0.05) enhanced (~3.3-3.7 fold respectively) in AG83 -Sias -infected macrophages and the level was further increased (~5.3 and 4.5 fold) when cells were infected with UR6 compared to AG83 +Sias (~1 and 1.3 fold respectively). Similarly iNOS level was only 1.7 fold increased in AG83 +Sias infected macrophages which was significantly (p 0.05) upregulated to 4.8 fold and 5.2 fold during AG83 -Sias and UR6 infection. 18s rRNA was used as housekeeping gene for normalization. Similar trend was also observed for Th1/Th2 cytokines using β-actin as a normalizing housekeeping gene (S3 Fig). Additionally, the status of a few key cytokines was determined in the culture supernatant. AG83 +Sias -infected macrophages produced significantly (p 0.05) less IFNγ level which was enhanced when AG83 -Sias or UR6 (p 0.01 and p 0.001) were used for infection (Fig 4E). In contrast, the increased levels of two immunosuppressive cytokines (IL-4 and IL-10) were observed during AG83 +Sias infection which was declined significantly (p 0.05) during AG83 -Sias and UR6 infection.

Sias-siglec interaction induced enhanced multiplication of intracellular amastigotes
Sias-siglec interaction played a significant contribution in enhanced phagocytosis and impairment of innate immune function in infected macrophages, thereby creating a perfect environment for multiplication of intracellular amastigotes. We have observed enhanced number of amastigotes in AG83 +Sias infected macrophages, which were significantly reduced (p 0.05) when sialidase-treated AG83 +Sias was used for infection. Additionally, removal of sialic acids and blocking of siglecs together further significantly (p 0.01) reduced the amastigote count confirming the role of Sias-siglec interaction in multiplication of intracellular parasites (Fig 4F). Low Sias containing avirulent strain (UR6) showed minimal level of infection than AG83 +Sias .

Sias-siglec interaction induced enhanced SHP-1 level in the host cell
Next, we addressed the status of downstream molecular events that control the innate immune response for establishment of successful infection. Although siglec-1 and siglec-5 exhibited ratio. Unbound parasite removed and the infection was allowed for 20 hr. Macrophages were fixed in methanol and stained by Giemsa staining. Intracellular amastigotes were counted microscopically. highest binding with AG83 +Sias , only siglec-5 is capable to take part in signaling due to the presence of ITIM motif. Siglec-E, the murine counterpart of siglec-5 was quantified (Fig 5A). Western blot showed upregulation of SHP-1 in AG83 +Sias infected macrophages compared to AG83 -Sias and UR6-infected cells indicating the role of sialic acids in upregulation of this phosphatase. However, as expected we did not find any significant alteration in the siglec-E level in these infected cells. (Fig 5A).
To further confirm that upregulation of SHP-1 is specifically due to Sias-siglec mediated interaction, the association of SHP-1 with siglec-E was tested by immunoprecipitation (Fig 5B). The levels of both SHP-1 and siglec-E were higher in the SHP-1-siglec-E complex during AG83 -+Sias infection compared to AG83 -Sias indicating enhanced association of siglec-E and SHP-1.

Infection with AG83 +Sias inhibited the MAPKs and PI3K/Akt pathways
The correlation between sialic acids-mediated virulence factor in modulation of MAPKs and PI3K/Akt pathways during parasite infection was addressed further. Accordingly, the status of important molecules in these pathways was compared in J774A.1 cells infected with AG83 +Sias , AG83 -Sias or UR6. Upon infection of J774A.1 with AG83 +Sias , phosphorylation of ERK1/2, p38MAPKand JNK were suppressed compared to uninfected control (Fig 5C). In contrast, macrophages infected with AG83 -Sias restored the phosphorylation of these molecules. Similarly, low Sias-containing avirulent stain UR6 also failed to suppress the phosphorylation of the major immune activating MAPKs pathway molecules. However, total level of ERK1/2, p38MAPK and JNK remained constant upon infection.
Macrophages infected with AG83 +Sias also induced the suppression of PI3K, p-PDK1 and p-Akt (Ser 473) in PI3K/Akt pathway (Fig 5D). Infection with AG83 -Sias showed the reversal of phosphorylation of these molecules indicating the role of Sias-mediated interaction by the parasite in deactivation of this pathway. Infection with UR6 also exhibited the similar trend of signaling in this pathway. However, total level of PDK1 and Akt remained constant upon infection.

Sias-mediated interaction inhibited the translocation of a functional subunit of NF-κβ in the nucleus
NF-κβ is a transcription factor translocate to the nucleus for transcription of proinflammatory genes. We have already observed the reduced macrophage effector function during AG83 +Sias infection. Therefore, the involvement of Sias-mediated interaction of Leishmania in the regulation of the translocation of a functional subunit of NF-κβ to the nucleus was assessed further. AG83 +Sias infection reduced the expression of IKKβ and IKKα in the cytoplasmic fraction of macrophages ( Fig 6A). Subsequently, IKKα and IKKβ-mediated phosphorylation status of inhibitory subunit IκBα were also suppressed, indicating the prevention of autodegredation of this protein of NF-κβ complex in the cytosol. So, functional subunit p65 could not set free to translocate into the nucleus and thus accumulated more in the cytosolic fraction (Fig 6A). This reduces the chance of transcription of proinflammatory genes. In contrast, removal of Sias from virulent strain AG83 +Sias induced the expression of IKKβ and IKKα and, followed by IκBα phosphorylation-mediated degradation. This leads to enhanced translocation of p65 into the nucleus (Fig 6B). UR6 infection also showed similar kind of effect like AG83 -Sias .

Knocking down of siglec-E overturned the Leishmania induced regulation of macrophage effector function and signaling pathways
Sialic acids-mediated interaction showed reduced macrophage effector functions and modulated downstream signaling molecules/pathway demanded to pin point the specific well) were uninfected or infected with AG83 +Sias or AG83 -Sias or UR6 at 1:10 ratio for 4 hr. Unbound parasites were removed. The cell lysate was prepared; proteins were quantified, separated in SDS-PAGE. They were transferred to nitrocellulose membrane and incubated overnight with anti-siglec-E and SHP-1 antibodies. α-tubulin was used as a control. The blots were incubated with respective secondary antibody and developed by ECL kit. Densitometric score mentioned below the each band was involvement of siglec (siglec-E) in these events. Accordingly, we transiently knocked down the expression of siglec-E by targeted siRNA in the macrophages. Infection of siglec-E depleted J774A.1 cells with AG83 +Sias showed enhanced (p 0.05) ROS production with MFI value 322 ± 27 compared to untransfected cells (MFI 226 ± 31) (Fig 7A). measured by ImageJ software. (B) Association of siglec-E and SHP-1 was measured by immunoprecipitation assay. J774A.1 cells were infected with AG83 +Sias or AG83 -Sias at 1:10 ratio for 2 hr. The cell lysate was incubated with anti-SHP-1 antibody overnight. The immunocomplex was pulled down by protein A-Sepharose 4B bead and run in SDS-PAGE. The proteins were transferred and processed similarly as in Fig. 5A. IP:Immunoprecipitation; IB: Immunoblot. (C) J774A.1 cells were infected and processed in similar manner as stated in Fig. 5A and incubated with anti-p-ERK-1/2, ERK-1/2, p-p38MAPK, p38MAPK, p-JNK, JNK and α-tubulin antibodies and developed. (D) Similarly, as stated in Fig. 5C, blots were also incubated with anti-PI3K, p-PDK1, PDK1, p-Akt (SER 473), Akt and α-tubulin antibodies and developed.
The status of key Th1 and Th2 cytokines during infection was also checked in siglec-E depleted macrophages. Significant (p 0.05) upregulation of the IFNγ (~3.3 fold) was observed in transfected cells after infection compared to~0.9 fold in untransfected macrophages. In contrast, IL-10 level was significantly (p 0.05) down regulated (~1.8 fold) in siglec-E knockdown cells compared to~3.9 fold in control cells due to the absence of Siassiglec interaction during infection ( Fig 7C).
Next, the level of phosphorylation of key signaling molecules in siglec-E depleted cells was checked to further confirm the importance of siglec in L. donovani infection through Sias-siglec interaction. Siglec-E depleted macrophages showed a complete absence of siglec-E molecule (Fig 7D). There was reduction in the SHP-1 level in these cells (Fig 7D). Other key signaling molecules in MAPK ( Fig 7E) and PI3K/Akt (Fig 7F) pathway such as pERK, pJNK, PI3K and p-Akt were upregulated in AG83 +Sias infected siSiglec-E macrophages. Additionally, status of nuclear translocation of NK-κβ functional subunit p65 was also checked in siglec-E knockdown cells. We observed reduced presence of p65 in the cytosol (Fig 7G) along with enhanced phosphorylation of inhibitory molecule IκBα which leads to the enhance translocation of p65 in the nucleus (Fig 7H) indicating the potential role of siglec in Sias-siglec pathway utilized by Leishmania in modulation of signaling pathways in macrophages.

Discussion
Tackling of visceral leishmaniasis is still challenging due to inadequacy of effective drug and development of widespread drug resistant strains [33]. Hence study the biology of host-parasite interaction is utmost important for identification of next generation drug target. Due to the unique location, sialic acids possess the strategic advantage to be a prime candidate for molecular interaction with host cells [34].
The most important findings of the present study include the demonstration of the involvement of sialic acids on virulent parasite for their binding with macrophages through siglecs, thereby establishing a new gate for the entry of this parasite. This Sias-siglec-mediated interaction of virulent parasite demonstrated its superior ability to subvert the innate immune response, enhanced multiplication of intracellular form of parasite in the host compared to low Sias-containing avirulent parasite. Current investigation also revealed a mechanistic approach to understand the specific role of Sias-siglec-1 in phagocytosis and involvement of Sias-siglec-E/5 in inducing suppression of MAPKs and PI3K/Akt pathways along with inhibition of translocation of a functional subunit of NF-κβ into the nucleus and thereby impairing the macrophage activation. The role of sialic acids and siglecs is thus established in ascertaining successful infection of Leishmania donovani.
Involvement of many Leishmania surface-associated glycoconjugates like lipophosphoglycan (LPG), glycoprotein 63 (gp63) in pathogenesis have been reported [35][36][37][38][39][40][41]. The presence measured by Griess reaction as stated in Fig 4B. (C) Siglec-E depleted and control J774A.1 cells were infected with AG83 +Sias and the infection was allowed for 24 hr in presence of PMA-ionomycin or LPS as stated in Fig 4C. Total RNA was isolated and processed for quantification of IFNγ and IL-10 genes by real time PCR as stated in Fig 4D. (D-F) Untransfected or siSiglec-E-transfected J774A.1 cells were infected with AG83 +Sias for 4 hr. The cell lysate was prepared and western blotting of key signaling molecules were performed similarly as stated in Fig 5A. (G-H) J774A.1 cells were or transfected with siSiglec-E. Cytosol and nuclear fractions were separated and level of key signaling molecules were determined as stated in Fig 6. doi:10.1371/journal.pntd.0004904.g007 of different derivatives of Sias both on promastigotes and amastigotes of L. donovani was identified in our laboratory [9,10]. Our previous study also revealed the global distribution of Sias on different strains of Leishmania species [42]. However, the acquisition of Sias by L. donovani is still a matter of debate. Presence of any active metabolic enzyme (UDP-GlcNAc 2-epimerase) was not found in the parasite [9]. Whether this enzyme is presented as biologically non-functional remain to be investigated further. Whole genome sequencing of L. donovani revealed the two putative CMP-sialic acid transporters LDBPK_240360 (NC_018251.1) and LDBPK_240350 (NC_018251.1) located at chromosome number 24. This may be further linked with the presence of sialyltransferase activity in this parasite [43]. Based on the comparative study for the presence of linkage-specific sialic acids on different L. donovani strains including drug resistant, drug sensitive, typed, untyped and avirulent strain, a high sialic acids containing virulent strain AG83 was selected for this investigation.
There are many receptors present on macrophages for binding with mannose, fucose, fibronectin, Fcγ, complement etc. and actively take part in host-pathogen interactions [21,44]. Siglecs, a unique class of membrane receptor, function as a negative regulator that prevents the over activation of immune cells. Our detailed investigation using siglec-transfected CHO cells along with macrophages/monocytes using five different strains conclusively confirmed that siglec-1 and -5 are two important candidate surface molecules served as receptors for virulent parasite binding. Specificity of sialic acids-siglec-mediated binding was confirmed by either removal of Sias or blocking siglecs. However, sialidase-treated parasite (AG83 -Sias ) didn't show complete reduction in binding with macrophages which are considered to be the natural host for parasite. This may be due to the presence of number of other receptors available for binding with parasite on the surface of macrophage besides siglecs unlike CHO cells. Thus the presence of linkage-specific Sias on parasite surface and their capability to form the Sias-siglec interaction with macrophages could screen the virulent vs. avirulent parasite. Binding of biotinylated siglec-1,-2,-5 with L. donovani further confirmed our finding [43]. In contrast, other human pathogens such as Pseudomonas aeruginosa, Group B Streptococcus and Campylobacter jejuni exhibited specificity towards siglec-7/9; siglec-5/7/9 and siglec-1/7 respectively. This may also depends on the siglec-specific sialylated ligands present on the pathogen surface and their affinity towards specific siglecs [4,45]. Therefore, siglecs are used by sialylated pathogens for their survival advantages [46].
After binding with the host cells, Leishmania promastigote initially phagocytosed by macrophages into the phagosome and subsequently fused with lysosome to form a phagolysosomal vacuole [47]. Thus phagocytosis is a prerequisite phenomenon in the life cycle of this pathogen. Leishmania undergo receptor-mediated phagocytosis and there are many candidate receptors that take part in this event [44]. Due to the elongated structure of siglec-1, it is extended way out from the cell surface making it a prime target for receptor-mediated phagocytosis by pathogens [7]. We have established siglec-1 as a new molecule that takes part in phagocytosis of AG83 +Sias during infection. Role of Sias-siglec-1 mediated phagocytosis was further confirmed as blocking of siglec-1 by antibody or removing Sias or combination treatment reduced the phagocytosed parasite inside the macrophages. Siglec-1 dependent phagocytosis was also reported in Campylobactor jejuni infection to the macrophages [48]. Impairment of innate immune system is an indispensable event for L. donovani to establish a successful infection. The immunosuppressive effect associated with leishmaniasis is exerted by the alteration of Th1/Th2 cytokine balance due to deactivation of macrophages induced effector molecules (ROS and NO) [49]. They were impaired during Sias-siglec mediated interaction of AG83 +Sias infection, indicating their direct involvement in immunosuppression. IFNγ and IL-4 are the key players involved both in VL and cutaneous leishmaniasis. However, IL-10 emerged as a most potent immunosuppressive cytokine in VL pathogenesis [50]. As cytokine secretion is more prolific in the primary macrophages, the secreted cytokines were measured in BMDM also. Sias-siglec mediated interaction of AG83 +Sias showed Th2 dominant (IL-10, IL-4 and TGFβ) cytokine response. All these events resulted in the enhanced intracellular amastigote replication. Thus impairment of this specific Sias-siglec interaction either by removal of Sias from parasite or silencing siglec from macrophages resulted in the reversal of immune response leading to reduction of parasite burden.
L. donovani engages a negative regulator SHP-1 during their infection to the macrophages for inactivation of the defense mechanism [51]. It is important to mention here that specific receptor present on macrophages as well as the molecular determinant on parasite which are directly involved in SHP-1 induction was not well documented. Present study conclusively demonstrated Sias-siglec-mediated interaction plays an important role in the upregulation of SHP-1 molecule and association between SHP-1 and siglec-E during infection. The specificity of the involvement of this receptor-ligand interaction for SHP-1 upregulation was further confirmed by removal of Sias from parasite or depleting siglec-E from macrophages.
SHPs act as negative regulator of NO production and suppressing MAPKs and Akt signaling pathways [52]. An array of molecular determinants present on Leishmania is known to deactivate these pathways [53,54]. Here we have established an alternative pathway (Siassiglec-mediated interaction) utilized by the virulent parasite (AG83 +Sias ) for deactivation of MAPK and PI3K/Akt pathway in the infected macrophages. Thus, Sias-siglec interaction could be considered as another new avenue of host pathogen interaction that can suppress the downstream signaling leading to deactivation of macrophages.
MAPKs and Akt pathways transmit the signal to the nucleus through a well known transcription factor NF-κβ, which control an array of proinflammatory genes [55]. We have demonstrated inhibition of nuclear translocation of a functional component (p65 subunit) of NFκβ in AG83 +Sias -infected macrophage through Sias-siglec-mediated interaction. This was correlated with the reduced ROS, NO and Th1 cytokines production. Removal of Sias or depletion of siglec showed reversal of this effect, indicating Sias-siglec interaction is an important determinant for immune suppression. A recent study on Group B Streptococcus also exhibited similar kind of distortion of innate immune response due to the sialic acids-mediated recruitment of siglec-E [24].
Our study conclusively ascertains that the Sias-siglec mediated interactions play a significant role in the binding and phagocytosis of virulent parasites with macrophages through siglec-1. This important molecular determinant (Sias) is essentially responsible for subversion of innate immune function by modulation of downstream signaling pathways favoring the multiplication of intracellular parasite in the hostile environment of host through siglec-E/5 (S4 Fig). Taken together, this in-depth study paved a novel horizon of host pathogen interaction, utilized by sialylated pathogens for their own survival advantages. Such finding may also help to design newer drug target. However, this needs further in-depth investigation.
Supporting Information S1 Fig. Presence of α2-3 and α2-6 linked Sias on different L. donovani strains. Six different L. donovani strains as mentioned in materials and methods, were grown in 10% FCS containing medium was designated as parasite +sias and sialidase (along with esterase) treatment of this parasite +Sias was designated as parasite -Sias . Parasite +Sias and parasite -Sias were incubated with FITC-SNA (specific for α2-6 linked Sias) and MAA (specific for α2-3 linked Sias) and the binding was measured by flow cytometry. (TIF)