2.1 Preparation of soluble Leishmania antigens

The strain of L. major (MRHO/IR/75/ER) was used to prepare the SLA. The preparation of SLA was done as described by (Scott et al., 1987). In brief, the parasites were collected at stationary phase and treated with 20 mM HEPES, pH 7.5 with 5% w/v sucrose buffer. Then, promastigotes was treated with 1.2 × 10⁹/ml of protease inhibitor cocktail (Sigma) and incubated over ice for 20 min. Promastigotes were lysed via freeze–thaw cycle and then sonicated (10 pulses, 30 sec each, every 2 min). The lysate was centrifuged at high speed centrifugation and the supernatant was dialysed extensively against the HEPES buffer and filter-sterilised. The BCA protein assay kit (Thermo Scientific, USA) was used to measure protein concentration of the SLA.

2.2 Purification of human C1q

C1q was essentially purified using a modified protocol as described earlier (Pondman et al., 2015). 50 ml of freshly frozen human plasma, adjusted to 5 mM EDTA, pH 7.4, was centrifuged for 30 min. at 12000 × g in order to eliminate debris, aggregates, and insoluble lipids. The supernatant was passed three times through a 5-ml column of IgG-Sepharose previously equilibrated with cold PBS-EDTA and extensively washed with the same buffer. The protein was stripped off the affinity column using elution buffer (1 M NaCl, 5 mM EDTA, 50 mM CAPS) at pH 11.2. The eluted fractions were neutralised using 0.5 M NaH₂PO₄. The contamination of IgG was removed by treated with a Hi-Trap protein G column. The flow was rich in C1q that was dialysed against HEPES buffer. The purity of C1q protein was assessed by 12% v/v SDS-PAGE. Before using the C1q in haemolytic assays, the haemolytic activity of C1q was checked through complement reconstitution assays.

2.3 Purification and expression of the human C1q

The recombinant head regions of human C1q, ghA, ghB, and ghC modules were expressed in E. coli BL21 (Kishore et al., 2003; Kojouharova et al., 2004). Plasmids pKBM-B, pKBM-A and pKBM-C containing complementary DNA (cDNA) for ghA, ghB and ghC regions were transformed separately into competent E. coli cells. Cells were grown in Luria-Bertani medium containing (50 μg/mL) antibiotic at 37 °C with shaking, and subsequently induced with 1.5 mM isopropylthio-β-galactoside (IPTG) at OD₆₀₀ of 0.75 for 2 h. The culture was then centrifuged at 4,500 × g for 15 min, the medium discarded. The pellet was suspended in 20 ml of lysis buffer. After incubation for 2 h at 4 °C cell lysate was sonicated 10–15 times, 30 sec each. The sonicate was then centrifuged (15,000 × g for 30 min), and the collected supernatant was diluted 5-fold using column buffer 1 (100 mM NaCl, 0.2% v/v Tween 20, 5 mM EDTA, 20 mM Tris–HCl, pH 8.0 and 5% v/v glycerol) and loaded on a 10-ml amylose resin column re-equilibrated with column buffer 1(New England Biolabs). After extensively washing the amylose resin column with buffer 2 (20 mM Tris–HCl, pH 8.0, 100 mM NaCl, 5 mM EDTA, and 5% v/v glycerol), the bound protein was eluted in 1 ml fractions with appropriate volumes of buffer 2 containing 10 mM maltose. The protein concentration was determined via Nanodrop and purity was examined via SDS-PAGE.

2.4 Expression and purification of substitution mutants of human C1q

The ghA, ghB and ghC derived substitution mutants (Arg^A162Ala/Glu, Arg^B114Gln/Glu, His^B117Asp, Arg^B129Ala/Glu, Lys^B136Glu, Arg^B163Ala/Glu, Tyr^B175Leu, His^C101Ala, Arg^C156Glu, Lys^C170Glu) were expressed as described above (Kishore et al., 2003; Kojouharova et al., 2004). Similar to the wild-type proteins, the substitution mutants were purified using an amylose resin column under gravity. These mutants were investigated based on previous data showing their essential role in C1q binding to wide range of ligands such as IgG, IgM, pentraxin 3 and C-reactive protein (Roumenina et al, 2006; Zlatarova et al., 2006). Contaminating Lipopolysaccharide (endotoxin) was eliminated using Polymyxin B column (Sigma). Limulus amebocyte lysate assay (BioWhittaker) was used to measure the level of contaminating and found to be less than ∼4 pg per μg of the recombinant protein.

2.5 Complement assay for the classical pathway

The modulation of classical pathway activation by SLA was examined by a complement consumption assay (Kishore et al., 1998). Different concentrations with an equal volume of normal human serum (NHS) were incubated with occasional shaking at 37 °C for 1 h. 1:1 diluted in dextrose gelatin veronal buffer composed of (DGVB⁺⁺; 2.5 mM sodium barbital, 71 mM NaCl, 0.15 mM CaCl₂, 0.5 mM MgCl₂, 2.5% w/v glucose, 0.1% w/v gelatin, pH 7.4). The supernatants were tested for their complement activity by measuring their ability to lyse antibody-sensitised sheep erythrocytes (EA). 100 µl of EA cells (1x10⁸ cells/ml in DGVB⁺⁺) were added to serial dilutions (100 µl) of each supernatant in DGVB⁺⁺ and incubated for 1 h at 37 °C. Supernatant was centrifuged at 5000 rpm for 5 min. The released haemoglobin from 100 µl of the supernatant was measured at 405 nm. Total haemolysis was measured by lysing EA with water. Diluted Serum was used without SLA as positive control. In addition, non-sensitised sheep erythrocytes (E) were used as a control for EA lysis at the same dilution of NHS.

2.6 Direct binding ELISA to examine interactions of C1q and recombinant wild type and mutant globular domain proteins with SLA

SLA was diluted in carbonate-bicarbonate buffer (Sigma-Aldrich) in the required dilutions (2.5 µg, 1.25 µg, 0.625 µg and 0.312 µg per well in 100 µl volume) and incubated 12 h at 4 °C. 100 µl of bovine serum albumin (BSA) was used to block microtitre wells at 37 °C for 4 h. The wells were then washed multiple times with PBS containing 0.05% v/v Tween 20 by filling up the wells and discarding the washing buffer. C1q and recombinant globular heads (ghA, ghB, ghC) were concentration-adjusted per well in binding buffer (20 mM Tris pH 8.0, 100 mM NaCl, 10 mM CaCl₂) and incubated for one hour at 37 °C and another one hour at 4 °C. Washing buffer PBS 0.05%Tween 20 was added to the wells and the primary antibody was added to the wells. C1q was probed with anti-human C1q monoclonal antibody (1:5000) while monoclonal anti-MBP antibody (1:7000) (Sigma-Aldrich) was used to probe ghA, ghB, ghC and their respective substitution mutants; the incubation lasted for one hour at 37 °C. After another round of extensive washing, 100 µl of secondary antibody rabbit anti-mouse IgG-horseradish peroxidise conjugate (1:5000) in PBS was added to each microtitre well for one hour at 37 °C. The colour was developed using o-Phenylenediamine dihydrochloride and measured at A₄₅₀ using the BioRad spectrophotometer.

2.7 Secreted cytokine, chemokine and growth factor multiplex assay

Different type of cytokines concentrations such as; (IL-6, IL-8, IL-10, IL12p40, IL12p70, IL-23, IL-27, IL-1α and IL-1β) and chemokines/growth factors concentrations such as; (MIG, I-TAC, MCP-1, G-CSF and M-CSF) were measured by MagPix Milliplex kit (EMD Millipore). THP-1 cells (1 × 10¹⁰ per well) were incubated with SLA (5 µg) with and without purified C1q (5 µg) for 24 h and 48 h, cells were spun down at 300 × g, and supernatants were collected for measuring a wide range of secreted pro-inflammatory and anti-inflammatory cytokines, chemokines, growth factors, and soluble ligands and receptors. The assay was done by adding 25 µl each of assay buffer and the supernatant from the THP-1 cell culture in a 96-well microtiter plate. A mixture of magnetic beads representing individual analytes mixed with dilution buffer was added to each well and incubated overnight at 4 °C. Next morning, the wells were washed with assay buffer and then probed with 25 µl of detection antibodies (1 h at room temperature). This was followed by addition of 25 µl of Streptavidin-Phycoerythrin conjugate (since target primary antibodies were biotinylated) to each well and incubated for 30 min at RT. Then, 150 µl of sheath fluid was added to each well. The Luminex Magpix instrument was used to read the plate wells.

2.8 Statistical analysis

For the classical pathway assays, the readings were plotted as percentage of lysis using the following equation

$\left(\frac{\mathit{Cx}-B}{L-B}\right)\ast 100=X$ ; Where Cx = OD₄₁₅ of sample, B = spontaneous lysis, L = maximum lysis. Graphs were generated using GraphPad Prism 6.0; and the statistical analysis was done by using unpaired one-way ANOVA test. Error bars represents SD or SEM (n = 3). The significant values were measured based on p < 0.05. A 2-sided unpaired t-test was used to compare the mean of the cytokine targets between SLA +/- C1q for any significant differences. P values were computed, and graphs were compiled and analysed.

3.1 Soluble Leishmania antigens activate the classical pathway

In order to assess if SLA can activate the classical pathway, we prepared haemolysin-sensitised sheep erythrocytes and assessed only classical pathway activation via determination of CH50 value (Fig. 1A). SLA (5 µg) was coated on microtiter wells and serially double-diluted. NHS was added to the wells for complement consumption for 1 h. The values derived were then plotted on the graph as percentage lysis. EA cells exposed to 1:10 diluted serum incubated with SLA exhibited only ∼22% complement activity, suggesting ∼77% of the total complement activity of NHS, when compared to MBP (Fig. 1B). NHS, untreated with SLA, showed ∼82% lysis at the same dilution and the CH50 value was attained earlier at ∼1:60 dilution of NHS. However, NHS consumed by SLA did not attain a CH50 value, suggesting high level of complement consumption.

Close

Fig. 1A

Haemolytic assay of normal human serum antibodies. 100 µl of human serum was treated with 200 µl of washed sheep red blood cells overnight followed by centrifugation and collection of supernatants. The supernatant serum was incubated at different dilutions (as shown in the graph) in DGVB²⁺ and incubated with 100 µl of sensitised and non-sensitised erythrocytes (EA and E cells respectively) for one hour at 37 °C. After centrifugation, the supernatants were collected and the absorbance was read at OD_415. EA and E with water was used as 100% lysis and the graph was plotted as a percentage of lysis.

Export to PPT

Close

Fig. 1B

Classical pathway haemolytic assay via complement consumption by soluble leishmania antigens (SLA). 5 µg of SLA was coated in microtitre wells using carbonate-bicarbonate buffer overnight. After blocking with 2% w/v BSA in DGVB²⁺ buffer, serially double-diluted normal human serum (NHS) were added to the wells. Recombinant E. coli Maltose-binding protein (MBP) was used as a negative control. Following 1 h incubation at 37 °C, haemolysin-sensitised EA cells (100 μl) were added to each well and incubated for another 1 h. After centrifugation, the supernatants were collected and the absorbance was read at OD_415. NHS diluted serially was used as a negative control of complement consumption. EA and E with water was used as 100% lysis and the graph was plotted as a percentage of lysis. The experiment was carried out in duplicates. SLA (NHS + Ag) consumed nearly 80% of the total complement in the serum. Results are expressed as percentage lysis with reference to a standard curve established using CH50 value from Fig. 1A (calculation formula under the Methods section). The data were expressed as mean of three independent experiments (n = 3) done in triplicates ± SEM. Statistical significance was established using the unpaired one-way ANOVA test (p < 0.01).

Export to PPT

3.2 Full length human C1q binds SLA via its globular domain

Different concentrations of SLA were coated on microtitre wells and allowed to bind C1q, ghA, ghB, and ghC. Recombinant forms of head regions of C1q A, B and C chains (ghA, ghB and ghC, respectively) were expressed and purified (Fig. 2A) to ascertain if C1q binds to SLA via its globular domain or triple-helical collagen domain. C1q and ghB bound SLA dose-dependently (Fig. 3A) and more specifically when compared to ghA, ghC and MBP (Fig. 3B). This suggested that SLA interacted to C1q via the most apically oriented module of the globular domain i.e. ghB. Next, we examined if single amino acid mutations made in the ghA, ghB and ghC modules impacted upon C1q-SLA interaction.

Close

Fig. 2A

SDS-PAGE analysis of purified MBP-ghA, MBP-ghB, MBP-ghC after elution from amylose resin. All three proteins were run simultaneously on a 12% acrylamide SDS-PAGE gel. The three proteins ran as a single thick band at ∼60 kDa. Lane 1: Standard molecular weight markers; Lane 2: MBP-ghA; Lane 3: MBP-ghB; Lane 4: MBP-ghC.

Export to PPT

3.3 SLA-C1q binding involves charge-charge interaction

The nature of interaction between SLA, and recombinant ghB and its single residue mutants (ghB-R114Q, ghB-R129A, ghB-R163E, ghB-R114E, ghB-R129E, ghB-H117D, ghB-T175L, ghB-L163G) were examined using solid-phase binding assays (Fig. 2B, C, D). At 10 µg concentration of SLA, the substitution of ghB mutant Arg¹²⁹ to alanine and glutamate resulted in 2% and 13% reduction in binding, respectively. Arg¹¹⁴ substitution of ghB led to 39% decrease of binding when substituted with both glutamic acid and glutamine. A similar decrease of binding was observed in the Arg163Glu substitution (31%). However, substitution mutants His117 to aspartic acid, Thr175 to leucine and Leu136 to glycine had more significant effects on SLA-ghB binding i.e., 49%, 61% and 46%, respectively. At 5 µg concentration of SLA, the substitutional ghB mutant Arg¹²⁹ to alanine resulted in a reduction of 34%. However, at this particular concentration, Arg¹²⁹ substitution to a negatively charged glutamic acid leads to an increase of 6% compared to wild type ghB. Arg114 substitution of ghB showed a minor reduction of binding i.e. 13% for glutamine substitution and 24% for glutamic acid substitution. However, substituting Arg¹⁶³ with glutamic acid exhibited a decreased binding of 62% to SLA. His117 to aspartic acid, Thr175 to leucine and Leu136 to glycine had more significant effects on SLA-ghB binding at 5 µg i.e., 52%, 60% and 46%, respectively. On an average, Arg substitutions resulted in less reduced binding i.e. 29% reduced binding in R114Q, 33% reduced binding in R114E, 25% less binding in R129A and 13% reduced binding in R129E to SLA compared to His and Leu substitutions leading to 50% less binding and Thr substitution leading to 61% less binding. However, one arginine substitution (R163E) showed a reduced binding of 48% to SLA (Table 1).

Close

Fig. 2B

SDS-PAGE analysis of purified ghA mutants. The ghA mutants expressed as MBP-fusion proteins were purified over amylose resin matrix, eluted with 10 mM maltose, and run on a 12% vol/vol acrylamide gel. Purified ghA mutants: lane 1: molecular marker; lane 2: ghA-R162A; lane 3: ghA-R162E.

Export to PPT

Close

Fig. 2C

SDS-PAGE analysis of purified ghB mutants. The ghB mutants expressed as MBP-fusion proteins were purified on amylose resin column and eluted with 10 mM maltose. The purified recombinant proteins were run on a 12% vol/vol acrylamide gel, as described under Methods section. Lane 1: Molecular marker; Lane 2: ghB-R163A; Lane 3: ghB-R163E; Lane 4: ghB-R114Q; Lane 5: ghB-R114E; Lane 6: ghB-R129E; Lane 7: ghB-R129A; Lane 8: ghB-T175L; Lane 9: ghB-L136G; Lane 10: ghB-H117D.

Export to PPT

Close

Fig. 2D

SDS-PAGE analysis of purified ghC mutants. The ghC mutants expressed as MBP-fusion proteins were purified using an amylose resin matrix and eluted with 10 mM; each purified mutant protein was separately run on a 12% vol/vol acrylamide gel. Lane 1: Molecular marker; Lane 2: ghC-H101A; Lane 3: ghC-R156E; Lane 4: ghC-L170E; Lane 5: ghC-R156Q.

Export to PPT

3.4 C1q modulates SLA-induced immune response by macrophages

To further understand the immune response of THP-1 cells, when challenged with SLA, with and without C1q, we cultured THP-1 cells in serum for 24 h and 48 h, collected their supernatants, and subjected them to analyte measurement of cytokines (Fig. 4), chemokines (Fig. 5), growth factors (Fig. 6) and soluble ligands (Fig. 7) using Multiplex array. SLA on its own induced in THP-1 cells secretion of IL-1α, IL-6, IL-1 α, IL-12, IL-17A, and IL-15 (Fig. 4). Similarly, chemokines and growth factors and soluble ligands were induced in THP-1 cells and secreted (Figs. 5–7). These data suggested a pro-inflammatory effect of SLA on THP-1 cells. Using 48 h as a reference point, we found that C1q seemed to upregulate IL-1 α, while it downregulated in THP-1 cells IL-6, IL-1 α, IL-12, IL-17A, and IL-15; the suppressive effects of C1q on IL-1 α, IL-12 and IL-17A quite considerable (Fig. 4). On the chemokine side, MIP-1a, was upregulated by C1q + SLA combination, while IL-8, IP-10, GRO and Eotaxin were downregulated; the suppression of IP-10 in the presence of C1q was very dramatic (Fig. 5). Interestingly, almost all growth factors (FGF, VEGF, EGF, GM-CSF and G-CSF) were downregulated by the presence of C1q (Fig. 6). With respect to other soluble ligands, such as IFN-g, IFN-a, Flt-3, IL-1RA and sCD40L, C1q was able to suppress their levels triggered by SLA consistently (Fig. 7). These results suggest that C1q can considerably influence pro-inflammatory soluble factors secreted by THP-1 cells, with the exception of IL-1 α (Fig. 4) and MIP-1 α (Fig. 5). Of note is the suppression of IL-12 (Fig. 4) that appears to suggest that C1q may be altering negatively to Th1 response that is required for parasite clearance as well as vaccine outcome using SLA.

Close

Fig. 3

Binding of human C1q (A), ghA, ghB and ghC (B) to solid phase SLA. Various concentrations of SLA were coated on microtitre wells and a fixed concentration (5 µg) of C1q, ghA, ghB and ghC were added. The interactions were detected by probing with a primary antibody, rabbit anti-human C1q antibody for C1q and rabbit anti-MBP antibody for globular head regions, followed by secondary antibody IgG conjugated with HRP. The colour was developed using o-Phenylenediamine dihydrochloride and A415 values were measured. All experiments were done in triplicates with MBP used as a negative control and C1q used as a positive control. The data were expressed as mean of three independent experiments (n = 3) done in triplicates ± SEM. Statistical significance was established using the unpaired one-way ANOVA test (p < 0.05).

Export to PPT

Close

Fig. 4

Multiplex analysis of cytokine analytes at 24 and 48 h using supernatant from the SLA challenged THP-1 cells with (SLA + C1q) and without C1q (SLA). Error bars represent ± standard deviation. A 2-side unpaired t-test (with Welch’s correction) was performed on the data to determine significant differences in protein levels between each group of treated and untreated supernatant. All these comparisons were significant (p < 0.05). IL, interleukin.

Export to PPT

Close

Fig. 5

Multiplex analysis of chemokine analytes at 24 and 48 h using supernatant from the SLA challenged THP-1 cells with and without C1q. Error bars represent ± standard deviation. A 2-side unpaired t-test (with Welch’s correction) was performed on the data to determine significant differences in protein levels between each group of treated and untreated supernatant. All these comparisons were significant (p < 0.05). IL, interleukin; IP, interferon-gamma induced protein; MCP, monocyte chemoattractant protein; MDC, macrophage derived chemokine; GRO, growth-related oncogene.

Export to PPT

Close

Fig. 6

Multiplex analysis of growth factor analytes at 24 and 48 h using supernatant from the SLA challenged THP-1 cells with and without C1q. Error bars represent ± standard deviation. A 2-side unpaired t-test (with Welch’s correction) was performed on the data to determine significant differences in protein levels between each group of treated and untreated supernatant. All these comparisons were significant (p < 0.05). FGF, fibroblast growth factor; VEGF, vascular endothelial growth factor; EGF, epidermal growth factor; GM-CSF, granulocyte-monocyte-colony stimulating factor; G-CSF, granulocyte-colony stimulating factor.

Export to PPT

Close

Fig. 7

Multiplex analysis of soluble immune ligand analytes at 24 and 48 h using supernatant from the SLA challenged THP-1 cells with and without C1q. Error bars represent ± standard deviation. A 2-side unpaired t-test (with Welch’s correction) was performed on the data to determine significant differences in protein levels between each group of treated and untreated supernatant. All these comparisons were significant (p < 0.05). IFN, interferon; Flt-3L, Fms-related tyrosine kinase 3 ligand.

Export to PPT