Bacteria and yeasts

H. alvei strain PCM 1200 was obtained from the Polish Collection of Microorganisms (PCM) at the Institute of Immunology and Experimental Therapy (Wroclaw, Poland). The bacteria were grown in LB medium, killed with 0.5% phenol, and centrifuged using a CEPA flow laboratory centrifuge [12]. Fresh baker’s yeasts (100 g; Lesaffre Polska S.A., Poland) were suspended in 100 ml of demineralized, Milli-Q (mQ)-purified water, washed a few times with 100 ml mQ water by centrifugation for 10 min at 10000 × g and 4°C, autoclaved, and dried (final weight of dried yeast: 22 g) using an ilShin laboratory freeze dryer (The Netherlands).

Human plasma

Pooled human plasma collected from healthy volunteers was purchased from Biowest SAS (France, cat. no. S4180). Ficolin-3 average concentration was determined as 21.6 μg/ml.

Preparation of LPS and O-specific polysaccharide

LPS 1200 was extracted from bacterial cells of H. alvei PCM 1200 by the hot phenol/water method [29] and purified by ultracentrifugation, as previously described [30]. The yields of LPS preparations were approximately 3.0%. Poly- and oligosaccharides were isolated by mild acidic hydrolysis (1.5% acetic acid) at 100°C for 15 min, and fractionated as previously described [31, 32] using Bio-Gel P-10. The yields of O-PS 1200 of H. alvei PCM 1200 LPS were 17–20%.

O-PS 1200-Sepharose chromatography

O-PS fraction 1 of H. alvei PCM 1200 (93.33 mg) was reduced using the NaBH₄ [33] and purified by size-exclusion chromatography on a column of Bio-Gel P-2 (Bio-Rad, USA), equilibrated with 0.05 M pyridine/acetic acid buffer (pH 5.6). Eluates were monitored using a Knauer differential refractometer (Germany), and the polysaccharide fraction was collected and dried using an ilShin laboratory freeze dryer. The O-PS 1200-Sepharose column was prepared by esterifying reduced O-PS 1200 (76.68 mg) using the 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide method [34] and coupling it to 20 ml CNBr-activated, cross-linked AH (1,6-diaminohexane)-Sepharose CL-4 (Pharmacia Fine Chemicals, USA) [35] via its carboxyl groups (O-PS 1200-Sepharose). A precolumn (20 ml) of CNBr-activated, cross-linked AH-Sepharose CL-4 (Pharmacia Fine Chemicals) was also prepared. The content of free amino groups in AH-Sepharose was determined by reaction with dinitrofluorobenzene, as described elsewhere [36]. Both precolumn and O-PS 1200-Sepharose column were equilibrated with TBS/Ca²⁺ running buffer composed of 10 mM Tris-HCl pH 7.4, 140 mM NaCl, and 5 mM CaCl₂.

Purification of Ficolin-3 complexes

Pooled, frozen (-20°C) human plasma (500 ml) was completely thawed in a 37°C water bath, then treated with PEG 6000 (Sigma-Aldrich, USA) as described previously [20], with modifications. Briefly, 1 M CaCl₂ was added to the collected supernatant to a final concentration of 20 mM and incubated for 1 h at 37°C with stirring followed by overnight incubation at 4°C with stirring. The formed clot was removed by centrifugation for 30 min at 2000 × g and 4°C, and the supernatant was collected. PEG 6000 in TBS (10 mM Tris-HCl pH 7.4, 140 mM NaCl) was added to the supernatant and the purification process was carried as described previously [20]. The supernatant was collected, divided into 10-ml aliquots, and stored at -20°C for separate use in further purification steps. MBL in stored samples was depleted by first mixing aliquots with 200 mg of lyophilized S. cerevisiae and 50 μl 1 M CaCl₂ (final concentration, 5 mM) and incubating overnight at 4°C with stirring. Then, the suspension was centrifuged for 20 min at 10000 × g and 4°C, after which CaCl₂ and l-fucose (Sigma-Aldrich) were added to the supernatant to a final concentration of 5 and 100 mM, respectively. The preparation was incubated 1 h at 4°C with stirring and centrifuged for 20 min at 10000 × g at 4°C. Sepharose-binding molecules were removed by loading the obtained supernatant (10 ml) onto a precolumn connected to the O-PS 1200-Sepharose (flow rate, 1 ml/min for 30 min) and incubated for 30 min at 4°C. Next, the precolumn was detached from the O-PS 1200-Sepharose column and washed separately with 3 column volumes (CV) of TBS/Ca²⁺/Tween (TBS/Ca²⁺, 0.05% Tween 20), 2 CV of eluting buffer (50 mM Tris-HCl pH 7.8, 1 M NaCl, 0.05% Tween 20, 10 mM EDTA), 2 CV of 3 M KSCN, and at least 3 CV of TBS/Ca²⁺/Tween to complete the KSCN elution. The presence of KSCN was determined by reaction with FeCl₃, which transforms colorless potassium thiocyanate into the deep red iron (III) thiocyanate, Fe(SCN)₃. The O-PS 1200-Sepharose column was washed with at least 2 CV of TBS/Ca²⁺/Tween to remove all non-bound molecules, monitored by measuring absorbance at 190 and 280 nm, and then washed with 2 CV of 200 mM d-GlcNAc in TBS/Ca²⁺ (pH 7.5) to elute remaining ficolin-1 and ficolin-2 [20, 26, 37, 38]. Finally, ficolin-3 complexes were eluted with 1.5 CV of 1 M sodium acetate (pH 7.5), and the column was flushed with 2 CV of eluting buffer and 2 CV of 3 M KSCN, and equilibrated with TBS/Ca²⁺/Tween. All eluates were collected and concentrated (3000 × g, 4°C) to a final volume of ~500 μl using an Amicon Ultra-15 Centrifugal Filter Unit (MWCO, 30 kDa; Millipore, USA). The concentrated ficolin-3 preparation was further fractionated on a TSKgel G3000SW column (21.5 mm ID × 30 cm, 13 μm; Tosoh Bioscience LLC, Germany), equilibrated with PBS (pH 7.5), under the following conditions: injected volume, 1 ml; flow rate, 5 ml/min; fraction volume, 2.5 ml. The eluate was monitored by measuring absorbance at 280 nm. Each fraction was concentrated or pooled and examined for purity, biological activity, and concentration as described below.

Dot-blot detection of MBL, ficolins and immunoglobulins

Concentrated fractions (1 μl) were spotted onto nitrocellulose membranes (GE Healthcare, USA). Membranes were blocked with SuperBlock Blocking Buffer (Thermo Scientific, USA) for 1 h and incubated for 1 h with the following primary antibodies: (i) anti-ficolin-3 mouse monoclonal IgG1 (clone 4H5; Hycult Biotech, The Netherlands), (ii) anti-ficolin-1 goat polyclonal IgG (clone G-13; Santa Cruz Biotechnology, USA), (iii) anti-ficolin-2 goat polyclonal IgG (clone G-12; Santa Cruz Biotechnology) or (iv) anti-MBL mouse monoclonal IgG1 (clone HYB 131–01; BioPorto, Denmark), diluted to a final concentration of 0.5 μg/ml in 10 mM Tris-HCl supplemented with 50 mM CaCl₂ (pH 7.4). Horseradish peroxidase (HRP)-conjugated anti-goat IgG polyclonal rabbit Ab (catalogue number: P0448, Dako, Denmark), anti-mouse IgG monoclonal rabbit Ab (catalogue number: P0260, Dako) and separately used or the mixture of anti-human IgA α chain, IgG γ chain, IgM μ chain polyclonal Ab (the mixture: Dako, catalogue number: P0212; separate Ab: Sigma-Aldrich) were used as secondary antibodies (1 μg/ml, 1 h incubation). Membranes were incubated for 10 min with Immun-Star HRP Chemiluminescent Substrate Kit (Bio-Rad, USA) and visualized using a G:Box Bioimaging System (Syngene, UK). Membranes were washed three times with dot-blot buffer (50 mM Tris-HCl pH 7.4, 0.2 M NaCl) after each step of the procedure. The procedure was carried out at room temperature.

Ficolin-3, MBL and total protein concentration

Ficolin-3 concentration was measured by ELISA as described previously by Michalski et al. [39], except that microtiter plates were coated with H. alvei PCM 1200 LPS (5 μg/well) instead of a BSA conjugate of its O-PS fraction. Recombinant human ficolin-3 protein (catalogue number: 2367-FC, R&D Systems, USA) was used as a standard. MBL concentration was determined as described previously [40]. Recombinant human MBL protein (catalogue number: 2307-MB/CF, R&D Systems) was used for calibration curve construction. Total protein concentration was determined using the bicinchoninic acid method (Micro BCA Protein Assay kit; Thermo Scientific) [41].

Detection of immunoglobulins, thrombin, collectin-11 and C1q in plasma-derived ficolin-3 complexes

Microtitier plate Maxisorp (Nunc) was coated with H. alvei PCM 1200 LPS (5 μg/well), and then after blocking with 1% BSA in TBS/Ca²⁺, 500 ng of plasma-derived ficolin-3 complexes in MBL-binding buffer (20mM Tris, 1M NaCl, 10mM CaCl₂, pH 7,4) supplemented with 0,1% BSA was added and incubated overnight at 4°C. After washing in TBS/Ca²⁺, antibodies IgA, IgG and IgM were detected by HRP-conjugated primary antibodies (as above); and ficolin-3, thrombin, C1q and CL-11 were detected by following primary Ig: (i) anti-ficolin-3 mouse monoclonal IgG1 (clone 4H5; Hycult Biotech), (ii) anti- thrombin rabbit polyclonal (catalogue number: ab92621, Abcam, UK), (iii) anti-Complement 1q (C1q) goat polyclonal (catalogue number: 234390, Calbiochem, USA), (iv) anti-collectin-11 (CL-11) rabbit polyclonal (catalogue number: HPA035241, Sigma-Aldrich) and HRP-conjugated secondary antibodies: rabbit anti-goat IgG (catalogue number: P0448, Dako) or goat anti-rabbit IgG antibodies (catalogue number: P0448, Dako) or rabbit anti-mouse IgG (catalogue number: P0260, Dako).

MASPs detection

Co-purified MASPs were detected by Western blot analysis. Briefly, plasma-derived ficolin-3 complexes (2 μg) were separated on 10% polyacrylamide gels and transferred onto PVDF membranes (Thermo Scientific). After blocking for 2 h with 5% skim milk, membranes were incubated with goat polyclonal anti-MASP-1 (catalogue number: sc-50841, Santa Cruz Biotechnology), rat monoclonal anti-human MASP-2/MAp-19 (Hycult Biotech, catalogue number: HM2191), or goat polyclonal anti-complement MASP-3 (catalogue number: AF1724-SP, R&D Systems) for 2 h at room temperature. The primary antibodies were diluted to a final concentration of 0.5 μg/ml in buffer containing 10 mM Tris-HCl (pH 7.4) and 50 mM CaCl₂. Next, 1000-fold diluted, HRP-conjugated rabbit anti-goat IgG (catalogue number: P0448, Dako) or anti-rat IgG antibodies (catalogue number: P0450, Dako) were added and the incubation was continued for 1 h at room temperature. After washing, membranes were incubated for 10 min with enhanced chemiluminescence (ECL) substrate (Santa Cruz Biotechnology) and exposed to X-ray film. Membranes were washed six times between each step.

MASPs activity

The ability of MASPs in plasma-derived ficolin-3/MASP complexes to cleave C3 and C4 was examined according to the method described by Matsushita and Fujita [42]. Briefly, different concentrations of plasma-derived ficolin-3/MASP complexes were incubated with 2 μg of C3 or C4 (Sigma-Aldrich) for 4 h at 37°C. The reactions were terminated by adding SDS-PAGE sample buffer (62.5 mM Tris-HCl, 2.5% SDS, 5% β-mercaptoethanol, 10% glycerol, 0.002% bromophenol blue) and heating at 99°C for 10 min. Proteins were separated on 10% polyacrylamide gels and stained with Coomassie Brilliant Blue to detect cleavage products. Alternatively, MASP-2 activity, based on C4 activation product deposition, was estimated as described previously using plates coated with LPS 1200, C4 from human plasma (catalogue number: C8195, Sigma) and rabbit anti-human C4c antibodies (catalogue number: F0169, Dako) [12]. Conditions excluding activation of complement classical pathway were applied [43]. MASP-1 activity was measured according to a protocol described by Presanis et al. [44] using the MASP-1 fluorescent substrate, Boc-Val-Pro-Arg-aminomethylcoumarin (VPR-AMC) with slight modifications. Briefly, white microtiter Maxisorp plates (Nunc, Denmark) were coated with H. alvei PCM 1200 LPS (5 μg/well) instead of mannan, which was used in the original protocol for determination of MASP-1 complexed with MBL.

Surface plasmon resonance (SPR)

SPR studies were performed using a Biacore T200 system (GE Healthcare). Recombinant ficolin-3 (R&D Systems) was immobilized on CM5 sensor chip (GE Healthcare Bio-Science AB) using amine coupling chemistry. Recombinant ficolin-3 was injected as a 50 μg/ml solution in 10 mM sodium acetate (pH 4.0) at a flow rate of 5 μl/min to a level of 15,300 resonance units (RU). The reference flow cell was immobilized with 12000 RU of BSA (Sigma-Aldrich). HBS-N buffer (GE Healthcare) supplemented with Ca²⁺ and Mg²⁺ ions (0.01 M HEPES pH 7.4, 0.15 M NaCl, 0.005% v/v surfactant P20, 5 mM MgCl₂, 5 mM CaCl₂) was used as a running buffer and sample buffer. Different concentrations of human IgG, IgA, and IgM (Sigma-Aldrich), recombinant MBL (R&D Systems) and O-PS 1200 were injected over the surface of immobilized recombinant ficolin-3 and BSA (reference flow cell) at a flow rate of 30 μl/min for 150 s or 600 s (O-PS 1200); 0.5% SDS injected for 30 s was used as a regenerator in all SPR experiments. BSA or ethanolamine were used to optimize a reference surfaces for analytes IgM and IgA.

Purification of ficolin-3 complexes from human plasma

Ficolin-3 complexes were isolated from human plasma using the following four-stage procedure: (i) PEG 6000 precipitation, as described by Zacho et al. [20]; (ii) yeast and l-fucose incubation, to remove contaminating MBL; (iii) affinity chromatography based on the identified specific ligand, O-PS of H. alvei 1200 LPS (O-PS 1200); and (iv) size-exclusion high-performance liquid chromatography (SEC-HPLC). Total protein, MBL and ficolin-3 concentration, and the presence of immunoglobulins, ficolin-1, ficolin-2, and MBL contamination (dot-blot) were monitored during the purification procedure (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Ficolin-3 Complexes Recovery and MBL and Total Protein Concentration During the Purification Process.

https://doi.org/10.1371/journal.pone.0156691.t001

Ficolin-3 concentration was measured by enzyme-linked immunosorbent assay (ELISA) using H. alvei PCM 1200 LPS as a solid-phase antigen. The plasma (500 ml) was coagulated, after which ficolin-3-containing fraction was precipitated from plasma using a 4–8% PEG cut-off, as described previously [20]. The average recovery of ficolin-3 after precipitation from plasma was 83.3%, whereas MBL contamination was 1.9% (Table 1). At this stage, the total average amount of protein was reduced from 25616.1 mg to 6140.3 mg, of which 9.0 mg was ficolin-3. The resulting preparation was stored in 10-ml aliquots for use in subsequent purification steps. Each 10-ml aliquot was incubated with autoclaved, freeze-dried S. cerevisiae cells and l-fucose solution, which reduced the MBL contamination [20, 45, 46] from 1.9% to 0.0% (Table 1). At this stage trace amount of MBL was observed only by dot-blot. The concentrated supernatant was purified by affinity chromatography on an O-PS 1200-Sepharose column, resulting in the specific capture of ficolin-3. In this stage, a precolumn of activated AH-Sepharose CL-4 was used to eliminate compounds, including traces of MBL that bound non-specifically to Sepharose [47–49]. The ficolin-3-containing fraction was eluted with 1 M sodium acetate, yielding a total recovery of 60.5%. The final fraction obtained from all aliquots contained 6.6 mg (mean average) of ficolin-3 (Table 1). Ficolin-3 complexes were further purified by SEC-HPLC on a TSKgel G3000SW silica-based column (Fig 1), which separated ficolin-3 high-molecular-weight oligomers (>440 kDa, fractions 22–25) from low-molecular-weight oligomers (66–440 kDa, fractions 26–32), and low-molecular-weight compounds and salts (<12 kDa, fractions 48–50). At this stage, the total average amount of protein per 500 ml was still 43.8 mg, of which 2.2 mg was ficolin-3. However it must be noted that two different methods were used for ficolin-3 (LPS 1200-based ELISA, recombinant ficolin-3 was used for calibration curve) and total protein measurements (the bicinchoninic acid method). The final ficolin-3 average recovery calculated for 500 ml of plasma was 19.8%. As an alternative final step, SEC-HPLC on divinylbenzene-based, polystyrene cross-linked column TSK G6000HR (S1 Fig) was used instead of G3000SW to minimize the significant loss of ficolin-3 on silica-based resins [50]. G6000HR-based separation was performed once and obtained results were compared with results for only one set of measured samples (Table 1, values in square brackets). However, this alternative achieved only a slight improvement, producing a final ficolin-3 recovery of 18.3% (in comparison to 13.5% for single measurement of G3000SW eluate), but it allowed increasing the amount of ficolin-3 by 35.6%.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. SEC-HPLC chromatography of plasma-derived ficolin-3 complexes on TSKgel G3000SW.

Presence of ficolin-1, ficolin-2, ficolin-3, MBL and IgA, IgG, IgM (denoted “Ig”) were detected by dot-blot analysis, where a positive result is indicated by “+”. The apparent molecular mass of thyroglobulin (670 kDa), ferritin (440 kDa), BSA (66 kDa), chymotrypsinogen A (25 kDa) and cytochrome C (12 kDa) was determined in a separate chromatographic run; V₀, void volume. Even fraction numbers are indicated below the chromatograph line. Absorbance was monitored at 280 nm.

https://doi.org/10.1371/journal.pone.0156691.g001

To identify fractions of interest and potential contaminants (see total protein concentration in Table 1), we examined fractions 20–37 for the presence of ficolin-1, ficolin-2, ficolin-3, MBL, IgA, IgG, and IgM by dot-blot assay (Fig 1). Dot-blot analyses of fractions 21–26 for ficolin-3 (Fig 2A), immunoglobulins (Fig 2B) and MBL (Fig 2C) showed a significant amount of Ig and trace amounts of MBL, a result in agreement with data for pooled fractions 22–25 (Table 1). Since preliminary monitoring of Ig contamination was performed with the mixture of anti-IgA α chain, IgG γ chain, IgM μ chain polyclonal antibodies, the next step involved the use of separate antibodies to determine the class of Ig in the ficolin-3 preparation (Fig 2D–2F). Dot-blot results indicated the presence of each class of antibodies, IgG, IgA and IgM, however unexpectedly stronger reactivity was observed with anti-IgA α chain and -IgM μ chain polyclonal antibodies, suggesting the hindered access of detecting antibodies to IgG epitopes due to the involvement of IgG γ chain in interaction with ficolin-3 reported previously by Panda et al. [27, 28] or differences in detection antibodies affinity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. IgA, IgG, IgM, and MBL tracking in SEC-HPLC fractions.

Dot-blot analyses were performed using (A) anti-ficolin-3 mAb; (B) anti-IgA α chain, IgG γ chain, IgM μ chain polyclonal Ab; and (C) anti-MBL mAb. Fractions 22–25 were eluted from the TSKgel G3000SW column (Fig 1) and concentrated 9-fold. IgA (D), IgG (E), and IgM (F) tracking in ficolin-3 complexes with dot-blotting. Samples: 1. Ficolin-3 preparation, concentrated eluate from G3000SW column (fractions 22–25); 2. 10-fold diluted ficolin-3 preparation; 3. 100-fold diluted ficolin-3 preparation; 4. Human plasma; 5. Recombinant ficolin-3.

https://doi.org/10.1371/journal.pone.0156691.g002

In summary, our purification procedure yielded an MBL- and ficolin-1/2-free preparation of ficolin-3 complexes (average mean 2.2 mg) from 500 ml human plasma with a final recovery of 19.8% (Table 1). SDS-PAGE analyses under non-reducing conditions revealed the presence of high-molecular-weight oligomers (Fig 3, lines 3 and 4) as active complexes with MASPs (Fig 3, lines 5–7). The fractionation under reducing conditions showed a prominent band with a molecular weight of 28 kDa (Fig 3, lane 5) corresponded to the MASP-1 B chain, whereas lighter bands above probably correspond to the unreduced chains. A MASP-2 band (Fig 3, lane 6) was observed at a molecular weight greater than 55 kDa; MASP-3 was also detected (Fig 3, lane 7).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Detection of ficolin-3, MASPs and Ig in plasma-derived ficolin-3 complexes.

Analysis of ficolin-3 preparation (1 μg of final product) fractionated on a 10% polyacrylamide gel under reducing (lanes 1, 2, 5–8) or non-reducing (lanes 3–4) conditions. Coomassie Blue stain (lanes 1, 3) and Western blotting using anti-ficolin-3 (lanes 2, 4), anti-MASP-1 (lane 5), anti-MASP-2 (lane 6), and anti-MASP-3 (lane 7) antibodies. The last lane (8) is the strip 4 re-probed with anti-human IgG, IgA and IgM. MASP bands are marked with asterisks (*). A, b, c–heavy chains of IgM (a), IgA (b), IgG (c); d- ficolin-3 monomer; e- light chains of Ig. The positions of the molecular mass markers are indicated in kDa.

https://doi.org/10.1371/journal.pone.0156691.g003

Biological activity of plasma-derived ficolin-3 complexes

To determine whether plasma-derived ficolin-3 is complexed with active MASPs, we incubated different concentrations of the preparation with C3 or C4 and examined reaction products by SDS-PAGE with Coomassie Blue staining (Fig 4). Both complement components were cleaved in a concentration-dependent manner, yielding C3α’ (Fig 4A, lines 2–4) and C4α’ chains (Fig 4B, lines 2–4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Cleavage of complement components C3 and C4 by plasma-derived ficolin-3/MASP complexes.

C3 (A) or C4 (B) was incubated with different concentrations of ficolin-3 preparation: 5 μg (lane 2), 2.5 μg (lane 3), 1.25 μg (lane 4), and control sample without ficolin-3 (lane 1). The cleavage of C3 and C4 was examined by SDS–PAGE and Coomassie Blue staining. M, molecular mass markers. Description of cleavage products is underlined.

https://doi.org/10.1371/journal.pone.0156691.g004

The activity of ficolin-3/MASP-1 complexes was examined using the synthetic substrate, Val-Pro-Arg-aminomethylcoumarin (VPR-AMC) and LPS 1200 as a solid-phase ligand for ficolin-3, and was expressed as the fluorescence of cleaved VPR-AMC (Fig 5A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Concentration-dependent activity of plasma-derived ficolin-3/MASP-1 and ficolin-3/MASP-2 complexes to activate LP.

(A) MASP-1 activity is expressed as the fluorescence of a product emerging due to the capacity of MASP-1 to cleave VPR-AMC substrate. (B) MASP-2 activity was determined by detection of C4 activation products deposition assay by specific antibodies. Conditions excluding activation of complement classical pathway were applied (Petersen et al., 2001).

https://doi.org/10.1371/journal.pone.0156691.g005

Ficolin-3/MASP-2 activity was demonstrated using a complement C4 activation products deposition assay (Fig 5B), which showed that plasma-derived ficolin-3 complexes exhibited concentration-dependent activity to cleave C4. C4 degradation products were detected by polyclonal primary rabbit anti-human C4c antibodies and secondary goat anti-rabbit IgG antibodies conjugated with HRP. Similarly, LPS 1200 was used as a solid-phase ligand for ficolin-3.

Interactions between recombinant ficolin-3 and immunoglobulins

In-process monitoring of ficolin-3 purification demonstrated the difficulty in eliminating contaminating IgA, IgG, and IgM from plasma-derived ficolin-3 preparations (Figs 1 and 2B, Table 1). Since Panda et. al previously showed that natural IgG and ficolin-3 interact [27, 28], we investigated binding between recombinant ficolin-3 and human IgG (Fig 6), IgA and IgM (S2 Fig) using surface plasmon resonance (SPR). Additionally, recombinant MBL was also analysed, since it was also observed as the contamination during preliminary steps of the purification process. Ficolin-3 (ligand) was immobilized on a CM5 chip, and different concentrations of Ig in buffer supplemented with Ca²⁺ and Mg²⁺ ions were injected over the chip surface. Human IgG exhibited concentration-dependent binding to ficolin-3 (Fig 6).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. SPR analysis of IgG binding to recombinant ficolin-3.

Ficolin-3 was immobilized on a CM5 chip. Ficolin-3 concentrations are indicated adjacent to sensorgrams. RU, resonance units.

https://doi.org/10.1371/journal.pone.0156691.g006

In contrast, no binding was observed between recombinant ficolin-3 and recombinant human MBL, used as an analyte and negative control (S2 Fig). The moderate binding was observed for IgM for reference surface (reference flow cell) with immobilized ethanolamine. No interpretable data were obtained for IgA, due to the difficulties with reference cell optimization. Both ethanolamine and BSA showed strong binding with IgA. Moreover BSA used as a reference surface showed strong interactions with IgM. Reference cell optimization experiences with BSA may indicate non-specific interactions between BSA and IgA and IgM. The specificity of the interaction between ficolin-3 and IgG was further examined using a competitive binding assay in which 670 nM O-PS 1200 was first injected over ficolin-3, followed by addition of human IgG. These assays revealed simultaneous binding of both analytes to ficolin-3 (Fig 7), suggesting the presence of different binding sites for each analyte.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. SPR-based competition assay of recombinant ficolin-3 interaction with O-PS 1200 and IgG.

Two-step SPR analysis of the interaction of immobilized recombinant ficolin-3 with 670 μM O-PS 1200 and 6.67 μM human IgG. No regeneration step was included between O-PS and IgG. Sensor chip: CM5. RU, resonance units.

https://doi.org/10.1371/journal.pone.0156691.g007