- Email: [email protected]
A simple two-step purification procedure for the iC3b binding collectin conglutinin
Journal of Immunological Methods 362 (2010) 204–208
Contents lists available at ScienceDirect
Journal of Immunological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m
Technical Note
A simple two-step purification procedure for the iC3b binding collectin conglutinin Thomas Krogh-Meibom a, Klaus Lønne Ingvartsen a, Ida Tornoe b, Nades Palaniyar c, Anthony C. Willis d, Uffe Holmskov b,⁎ a b c d
Department of Animal Health and Bioscience, Faculty of Agricultural Sciences, Aarhus University, 8830 Tjele, Denmark Department of Cardiovascular and Renal Research, Institute of Molecular Medicine, University of Southern Denmark, DK-5000 Odense, Denmark Lung Innate Immunity Research Laboratory, Physiology and Experimental Medicine, The Hospital for Sick Children and the University of Toronto, Canada Medical Research Council (MRC) Immunochemistry Unit, Department of Biochemistry, University of Oxford, Oxford OX1 3QU, United Kingdom
a r t i c l e
i n f o
Article history: Received 4 August 2010 Accepted 1 September 2010 Available online 9 September 2010 Keywords: C-type lectin Collectin Conglutinin Purification Complement activation
a b s t r a c t Bovine conglutinin is a serum protein involved in innate immunity. It binds calcium dependently to iC3b, a product of the complement component C3 deposited on cell surfaces, immune complexes or artificial surfaces after complement activation. We here present a simple and efficient two-step procedure for the purification of conglutinin. In the first step, bovine serum is incubated with non-coupled chromatographic TSK beads at 37 °C to allow complement activation and iC3b deposition on the beads and subsequent binding of conglutinin to iC3b. Conglutinin is then eluted from the beads by EDTA. In the second step, conglutinin is separated from iC3b and IgM by ion-exchange chromatography. This purification procedure yielded 81 μg of conglutinin per ml of serum with a recovery of 61.2%. Surface plasmon resonance analysis showed that the purified conglutinin had a high affinity for mannan (Kd = 2.3 − 3.2 nM). SDS-PAGE and time-resolved immunofluorometric assays showed that the conglutinin was not contaminated with other serum collectins such as collectin-43 or mannan-binding lectin. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Bovine conglutinin (BK) is a member of the collectin family of proteins composed of C-type lectin domains connected to collagen-like regions. The abundant form of BK is a dodecamer of 43 kDa monomers arranged in a flexible cross-like structure of four homotrimers. Conglutinin binds microorganisms either directly or via the complement degradation product iC3b through the lectin domains. The binding leads to agglutination of the microorganisms and it Abbreviations: BK, bovine conglutinin; CL-43, collectin-43; EDTA, ethylenediamine tetra-acetic acid; HBS, HEPES buffered saline; MBL, mannan-binding lectin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SPR, surface plasmon resonance; TBS, tris-buffered saline; TRIFMA, time-resolved immunofluorometric assay. ⁎ Corresponding author. Tel.: + 45 65503775; fax: + 45 65503922. E-mail address: [email protected] (U. Holmskov). 0022-1759/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2010.09.010
promotes phagocytosis, and BK thereby plays an important role in the bovine innate immune defence system (for review see Holmskov (2000)). We have previously shown that serum concentrations of BK vary among individual cows, that the levels are genetically determined and that low BK serum concentrations predispose to infection (Holmskov et al., 1998). These results lead to a renewed interest in conglutinin, especially in cattle breeding programs. We have developed a time-resolved immunofluorometric assay for large-scale measurement of serum BK (Krogh-Meibom et al., 2004) and purified BK was needed to establish the assay standard. Also, purified BK is used for the conglutination test to study iC3b coated immune complexes (Tsai et al., 1998) and for purification of immune complexes (Casali and Lambert, 1979). We here describe a simple two-step purification procedure for BK based on iC3b affinity chromatography and ion-exchange chromatography.
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
2. Materials and methods 2.1. Buffers Tris buffered saline (TBS): 10 mM Tris-Base (Tris[hydroxymethyl]aminomethane, Sigma, St. Louis, MO, cat. no. T-1503), 140 mM NaCl, 0.05% emulfogen (polyoxyethylene 10 tridecyl ether, Sigma, cat. no. P-2393), pH 7.4; TBS–Ca: TBS with 5 mM CaCl2; TBS–Ca 1 M NaCl: TBS–Ca containing a total of 1 M NaCl; TBS–EDTA: TBS with 5 mM EDTA; SDS-PAGE sample buffer: 10 mM Tris–HCl, 1 mM EDTA, 10% SDS (Sigma, cat. no. L-4509), 2% Bromophenol Blue; SDS-PAGE running buffer: 14% glycin, 3% Tris-Base, 1% SDS; piperazine loading buffer: 20 mM piperazine (diethylenediamine, Sigma, cat. no. P4, 590-7), 50 mM NaCl, 5 mM EDTA, 0.05% emulfogen, pH 6.2; HEPES buffered saline (HBS): 10 mM HEPES (N-[2hydroxymethyl]piperazine-N′-[2-ethanesulfonic acid], Sigma, cat. no. H-7523), 150 mM NaCl, 0.005% surfactant P-20 (BIAcore, Herts, U.K., cat. no. BR-1000-54), 0.02% NaN3, pH 7.4; HBS–Ca: HBS with 5 mM CaCl2 and HBS–EDTA: HBS with 5 mM EDTA. 2.2. Serum isolation and storage Blood from two cows was randomly collected at the local abattoir and allowed to clot overnight at 4 °C. Serum was isolated by centrifugation at 1000× g for 30 min. Sera from the two cows were pooled and stored in 500 ml aliquots at −20 °C until use. The BK concentration in the pooled serum was 132 μg/ml. The concentrations in adult cows vary from approximately 100 ng/ml to above 1 mg/ml with a geometrical mean of approximately 40 μg/ml (Krogh-Meibom et al., 2004). 2.3. Purification of conglutinin The chromatographic beads (TSK, Toyopearl HW-75F, Sigma, cat. no. 8-07469) were washed extensively with water followed by TBS–Ca buffer on a glass filter. Serum was thawed overnight at room temperature, centrifuged for 30 min at 10,000× g, and filtered through a glass filter. A volume of 100 ml serum was incubated with 60 ml of TSK beads for 2.5 h at 37 °C in a 200 ml glass flask. The TSK beads were gently resuspended every 30 min. The TSK beads were packed under flow on a computer-monitored FPLC system (FPLCdirector Version 1.3, Amersham Biosciences, Uppsala, Sweden) and washed first with TBS–Ca 1 M NaCl until the absorbance at 280 nm was below 0.1, and then equilibrated with TBS–Ca buffer until the absorbance reached the baseline. The column was then eluted with TBS–EDTA, and the fractions containing protein were pooled. The EDTA eluate was diluted 1/4 in piperazine loading buffer pH 6.2 and applied to a 1 ml Mono Q anion-exchange column (Amersham Biosciences, cat. no. 17-0546-01) at a flow rate of 0.5 ml/min. After washing with loading buffer, retained protein was eluted with a linear gradient from 50 to 375 mM NaCl over 30 column volumes in piperazine loading buffer pH 6.2. Samples of the starting materials and from each purification step were collected and analysed by SDS-PAGE and in the BK–TRIFMA-assay as described below.
205
As a control, a volume of 100 ml serum was heatinactivated for 1 h at 56 °C and used in the purification procedure described above. As further controls, 1 ml TSK was incubated with purified BK (100 μg/ml) for 2.5 h at 37 °C, and 1 ml volumes of TSK that had previously been incubated with serum and eluted with TBS–EDTA were incubated with either purified BK or heat-inactivated serum. The TSK beads were then washed with TBS–Ca 1 M NaCl and with TBS–Ca by repeated suspension, centrifugation and removal of the supernatant. Finally, the beads were resuspended in 1 ml TBS–EDTA (10 mM EDTA) and the supernatants were analysed by SDS-PAGE and by the BK–TRIFMA-assay. 2.4. SDS-PAGE Electrophoresis was performed on 4–20% polyacrylamide gradient gel and samples were reduced by heating at 100 °C for 2 min in 50 mM dithiothreitol diluted in SDS-PAGE sample buffer and alkylated by the addition of iodoacetamide to a concentration of 100 mM. Non-reduced samples were heated in SDS-PAGE sample buffer with 2 mM iodoacetamide. Samples of 10 μl were loaded per lane. Protein bands were detected with Coomassie Brilliant Blue. The molecular mass markers were those of the Mark 12 standard (Invitrogen NOVEX, San Diego, CA, cat. no. LC5677) or Precision Protein Standards (Bio-Rad, cat. no. 161-0362). 2.5. N-terminal amino acid sequencing Samples were reduced and run on a 10% NOVEX Bis-Tris NuPAGE precast gel (Invitrogen NOVEX) at 200 mA per gel in a NOVEX XCell II Mini-cell gel apparatus. The gel was electroblotted to a NOVEX 0.2 μm PVDF membrane in a NOVEX blot module. The membrane was then stained with Coomassie Brilliant Blue. The bands of interest were excised from the PVDF membrane and washed extensively with 10% methanol in water prior to sequencing. They were then sequenced on an Applied Biosystems 494A ‘Procise’ protein sequencer (Applied Biosystems, Warrington, U.K.) using standard sequencing cycles. 2.6. Amino acid analysis Samples were run on an ABI 420A analyser (PE Biosystems, Warrington, U.K.) following hydrolysis for 24 h in 5.7 M hydrochloric acid at 110 °C. Data were integrated using Dionex Chromeleon v6.4 software (Dionex, Macclesfield, U.K.). 2.7. Time-resolved immunofluorometric assay (TRIFMA) for conglutinin A BK–TRIFMA-assay was used to estimate the recovery of BK (Krogh-Meibom et al., 2004). Briefly, wells of microtiter plates (MaxiSorp, Nunc, Kamstrup, Denmark) were coated with anti-BK polyclonal antibodies by overnight incubation. The wells were washed and incubated overnight with dilutions of BK containing samples. The wells were then incubated stepwise with an anti-BK monoclonal antibody and a biotin-labelled subclass specific rabbit anti-mouse antibody, followed by Eu3+ chelates coupled to streptavidin. After
206
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
addition of a solution to release the Eu3+ ions, UV-light induced light emission was measured. 2.8. Surface plasmon resonance analysis (SPR) Immobilization of biotinylated mannan (1 mg/ml) on a streptavidin-coated BIAcore chip (SA-chip, BIAcore) was performed in the biosensor system in 100 mM NaOAc (pH 5.5) buffer at a continuous flow of 5 μl/min for 10 min. Leaving flow cell 1 as blank, the next three flow cells were used for immobilization of yeast mannan (Sigma, M-3640), which was treated with 0.5, 1.5 or 4.5 mM NaIO4 and reacted with biotin-LC-hydrazide (Pierce, Perbio Science, Cheshire, U.K., cat. no. 21340) according to the manufacturer's protocol. Free streptavidin was blocked with biotin, and the chip was normalized with 50% glycerol solution. Purified BK in the range of 0–20 μg/ml in HBS–Ca buffer was injected and allowed to bind with the immobilized mannan. A flow rate of 10 μl/min was maintained for 2 min at 25 °C. The complexes were allowed to disassociate for 2 min, and bound proteins were washed with two pulses of 5 μl
3 mM EDTA, 0.1% SDS solution. The flow cells were reequilibrated after washing with 20 μl HBS–EDTA and 20 μl HBS–Ca buffer. The SPR tracings from one of the flow cells were used for the calculation of disassociation and association rates by BIAcore 2000 software. In order to calculate the affinity of BK to mannan, the molecular weight of the lectin was assumed to be 516 kDa (dodecamer). 3. Results and discussion We have previously purified conglutinin by polyethylene glycol precipitation, affinity chromatography on mannan– Sepharose and an anti-bovine Ig–Sepharose step to remove contaminating anti-carbohydrate antibodies (Holmskov et al., 1993). Other methods employing precipitation, adsorbtion to zymozan, ion-exchange chromatography, enzyme digestion and gel filtration have also been employed (Maire et al., 1981). Common to the known procedures is the fact that contamination with the other known bovine serum collectins mannan-binding lectin (MBL) and collectin (CL)-43 is likely to occur. TSK is a methacrylate-based matrix designed for
Fig. 1. Purification of conglutinin from bovine serum. A: EDTA elution profile of the TSK column incubated with 100 ml of serum. Fractions containing conglutinin are indicated by arrows. B: Elution profile of the conglutinin containing fractions from panel A applied to a Mono Q ion-exchange column and eluted with a linear NaCl gradient. The small vertical bar indicates the fractions used for total amino acid analysis (aaa, fractions 19–20) and the larger bar represent the pool of fractions used to determine the total yield of conglutinin (BK Mono Q pool, fractions 19–21). C: Elution profile of the purified conglutinin (from panel B) on a Superose 6 gel filtration column. The column was calibrated with: a, blue dextran (N 2000 kDa); b, beta-amylase (200 kDa); c, bovine serum albumin (66 kDa); d, cytochrome C (12.4 kDa). D: SDS-PAGE (4–20% gradient gel) of reduced conglutinin from the Superose 6 fractions indicated by a bar in panel C. The molecular weight markers indicated are the Precision Protein Standards.
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
gel filtration. The matrix is synthezised by copolymerisation of glycidyl methacrylate, polyethylene glycol and the crosslinking agent pentaerythritol. It contains both primary and secondary hydroxyl groups that can be activated for immobilization of ligands for affinity chromatography applications (Hermanson et al., 1992). The hydroxyl groups are potential targets for C3b activated via the alternative complement pathway. As opposed to agarose-based matrixes, TSK has a low intrinsic affinity to carbohydrate-binding proteins. TSK coupled with different sugars has previously been used to purify collectins (Hansen et al., 2002). In order to simplify the purification procedure for BK, we used TSK coupled with N-acetyl-D-glucosamine as affinity matrix and non-coupled TSK as precolumn. To our surprise, all BK was bound to the precolumn. BK did not bind to noncoupled TSK if heat-inactivated serum was used, indicating that complement deposition was required for BK binding. BK is stable to heat-inactivation (Kawasaki et al., 1993) but factor H and factor I, responsible for cleaving C3b into iC3b, are inactivated. BK was then purified by a two-step procedure (Fig. 1). First bovine serum was incubated with TSK beads at 37 °C in order to allow iC3b deposition and subsequent binding of BK. After packing and extensive washing of the TSK in a FPLC column BK was eluted from the column with EDTA (Fig. 1A). The EDTA eluate was then separated by ion-exchange chromatography on Mono Q where two major peaks were seen (Fig. 1B). Peak two, which contained BK, was further analysed by Superose 6 gel filtration chromatography (Fig. 1C). In accordance with previous findings by Kawasaki et al. (1993) and Holmskov et al. (1995) the major form of BK followed by the truncated form eluted close to the void volume (Fig. 1C and D). Samples from the two purification steps were analysed by SDS-PAGE (Fig. 2A). Lane 4 shows the EDTA eluate from the TSK column incubated with serum in the reduced state. One major band at approximately 43 kDa is seen, together with two broad diffuse bands of 80 kDa and 30 kDa, a double band of 66 kDa and two faint bands at approximately 49 kDa and 41 kDa. The N-terminal amino acid sequences of the 43 and 49 kDa bands were identical (AEMTTFSQKI) and matched with the N-terminal sequence of the mature BK protein (Lee et al., 1991). The sequence of the 41 kDa band was AGRPGWVGPI, and this sequence corresponds to a form of BK truncated after 54 amino acids (Kawasaki et al., 1993). The post-translational modification that may account for the different sizes (43 and 49 kDa) of the two non-truncated forms of BK is not known, but differential O-glycosylation, which has been reported for the structurally related SP-D molecule, could be an explanation. N-terminal sequencing did not identify the bands with molecular masses of 80 kDa and 30 kDa, but they probably correspond to the heavy and light chains of IgM. The double band around the 66 kDa was identified as the bovine C3 betachain (NPLYSMITPN) (upper band), and as the large iC3b alpha-chain fragment (SDLDDDIIPE) (lower band). When the EDTA eluate was separated by Mono Q ionexchange chromatography, two major peaks occurred. Peak one (Fig. 2A, lanes 8 and 9) contained the bands corresponding to C3b and IgM and two weak bands at 42 kDa and 44 kDa. Peak two (lanes 6 and 7) contained the bands corresponding to BK. The N-terminal sequence of the 42 kDa
207
Fig. 2. Purity and activity of conglutinin (BK). A: SDS-PAGE of reduced (lanes 2, 4, 6, 8) and non-reduced (lanes 3, 5, 7, 9) protein on a 4–20% gradient gel from the BK purification steps and of control fractions. Lanes 2 and 3: pooled EDTA eluate from TSK incubated with heat-inactivated serum. Lanes 4 and 5: pooled EDTA eluate from TSK incubated with serum not heat-inactivated (tubes 8–11, Fig. 1A). Lanes 6 and 7: Mono Q pool of protein eluted late (fractions 19–21) in the NaCl gradient. Lanes 8 and 9: Mono Q pool of protein eluted first (fractions 12–15) in the NaCl gradient. Lanes 1 and 10: the molecular mass markers indicated are those of the Mark 12 standard. B: SPR analysis of BK–mannan interaction. The BK (association) followed by HBS–Ca buffer (disassociation) was flowed on the mannan-coated chip for 2 min each. The BK-specific SPR from different concentrations of the lectin (20, 10, 5, 2.5, 1.25, 0.625, 0.313 and 0.0 μg/ml in that order) was aligned for the affinity measurements (SPR during the wash steps is not shown).
band from the C3b/IgM peak identified the peptide as the small iC3b alpha-chain fragment that is generated when the 2 kDa C3f fragment is released by factor I to generate
208
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
Table 1 Recoveries of conglutinin. Purification fraction
Heat-inactivation (±)
N
Concentration ⁎ (μg/ml)
Volume (ml)
Recovery from serum ⁎ (%)
Serum EDTA pool BK Mono Q pool Serum EDTA pool
− − − + +
4 4 4 4 4
132 ± 4.00 1090 ± 20.2 2700 ± 54.2 132 ± 1.85 0.13 ± 0.015
100 10 3 100 7.5
100 ± 3.03 82.2 ± 1.53 61.2 ± 1.23 99.8 ± 1.40 0.0073 ± 0.00085
⁎ Mean values ± sem.
iC3b (SEETKENEGF). The faint bands seen in Fig. 2A, lanes 6 and 7 corresponding to molecular masses of 85 kDa and 75 kDa are probably BK dimers. The size of the non-reduced band at 200 kDa (lanes 5 and 9) is consistent with that expected for bovine C3b and the band seen at the top of the gel corresponds to the expected molecular mass of nonreduced IgM. No detectable bands corresponding to CL-43 or MBL were seen in the samples from peak two from the Mono Q column (Fig. 2A, lanes 6 and 7), and no signals were obtained when the same fractions were analysed by CL-43– TRIFMA (data not shown). Heat-inactivated serum did not mediate the binding of BK to TSK (Fig. 2A, lanes 2 and 3) and there was no significant binding of purified BK to TSK that had not been preincubated with serum. Purified BK and BK from heat-inactivated serum bound to serum treated TSK beads after the initial EDTA elution indicating that the beads can be used several times (data not shown). Total amino acid analysis of the BK fractions indicated in Fig. 1B showed a concentration of 3.36 ± 0.055 mg/ml. The recoveries of BK were assessed by the BK–TRIFMA-assay (Table 1) using the total amino acid analysed fractions as standard. The purification procedure used yielded 81 μg BK per ml of bovine serum. The recovery of BK was 61.2%. The lectin activity of the purified BK was evaluated by SPR analysis (Fig. 2B). The association and dissociation rates between purified BK and mannan were in the range of 1.0 × 105 to 4.9 × 105 M/s and 1.1 × 10−3 to 3.3 × 10−3 l/s, respectively. The resulting dissociation constant (Kd) is between 2.3 × 10−9 and 3.2 × 10−9 M, which is comparable to the Kd's of the native forms of other collectins such as MBL, surfactant protein A, surfactant protein D and CL-43 (Holmskov, 2000). 4. Conclusion We have developed a simple two-step procedure for the purification of bovine conglutinin based on the iC3b binding
property of conglutinin and ion-exchange chromatography. This method is easy to perform and gives high yields. The purified BK is free of the other known serum collectins, and is a functional dodecamer. References Casali, P., Lambert, P.H., 1979. Purification of soluble immune complexes from serum using polymethylmetacrylate beads coated with conglutinin or C1q. Application to the analysis of the components of in vitro formed immune complexes and of immune complexes occurring in vivo during leishmaniasis. Clin. Exp. Immunol. 37, 295. Hansen, S., Holm, D., Moeller, V., Vitved, L., Bendixen, C., Reid, K.B., Skjoedt, K., Holmskov, U., 2002. CL-46, a novel collectin highly expressed in bovine thymus and liver. J. Immunol. 169, 5726. Hermanson, G.T., Mallia, A.K., Smith, P.K. (Eds.), 1992. In: Immobilized Affinity Ligand Techniques, 1. Academic Press, pp. 31–34. Holmskov, U.L., 2000. Collectins and collectin receptors in innate immunity. APMIS 108 (Suppl. 100), 1. Holmskov, U., Holt, P., Reid, K.B., Willis, A.C., Teisner, B., Jensenius, J.C., 1993. Purification and characterization of bovine mannan-binding protein. Glycobiology 3, 147. Holmskov, U., Laursen, S.B., Malhotra, R., Wiedemann, H., Timpl, R., Stuart, G.R., Tornoe, I., Madsen, P.S., Reid, K.B., Jensenius, J.C., 1995. Comparative study of the structural and functional properties of a bovine plasma C-type lectin, collectin-43, with other collectins. Biochem. J. 305, 889. Holmskov, U., Jensenius, J.C., Tornoe, I., Lovendahl, P., 1998. The plasma levels of conglutinin are heritable in cattle and low levels predispose to infection. Immunology 93, 431. Kawasaki, N., Yokota, Y., Kawasaki, T., 1993. Differentiation of conglutination activity and sugar-binding activity of conglutinin after removal of NH2terminal 54 amino acid residues by endogenous serine protease(s). Arch. Biochem. Biophys. 305, 533. Krogh-Meibom, T., Holmskov, U., Løvendahl, P., Nielsen, N.I., Ingvartsen, K.L., 2004. A time-resolved immunofluorometic assay for quantification of the bovine collectin conglutinin. J. Immunol. Methods 286 (1–2), 87. Lee, Y.M., Leiby, K.R., Allar, J., Paris, K., Lerch, B., Okarma, T.B., 1991. Primary structure of bovine conglutinin, a member of the C-type animal lectin family. J. Biol. Chem. 266, 2715. Maire, M.A., Barnet, M., Lambert, P.H., 1981. Purification of bovine conglutinin using pepsin digestion. Mol. Immunol. 18, 85. Tsai, J.F., Margolis, H.S., Jeng, J.E., Ho, M.S., Chang, W.Y., Hsieh, M.Y., Lin, Z.Y., Tsai, J.H., 1998. Immunoglobulin- and hepatitis B surface antigen-specific circulating immune complexes in chronic hepatitis B virus infection. Clin. Immunol. Immunopathol. 86, 246.
Contents lists available at ScienceDirect
Journal of Immunological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m
Technical Note
A simple two-step purification procedure for the iC3b binding collectin conglutinin Thomas Krogh-Meibom a, Klaus Lønne Ingvartsen a, Ida Tornoe b, Nades Palaniyar c, Anthony C. Willis d, Uffe Holmskov b,⁎ a b c d
Department of Animal Health and Bioscience, Faculty of Agricultural Sciences, Aarhus University, 8830 Tjele, Denmark Department of Cardiovascular and Renal Research, Institute of Molecular Medicine, University of Southern Denmark, DK-5000 Odense, Denmark Lung Innate Immunity Research Laboratory, Physiology and Experimental Medicine, The Hospital for Sick Children and the University of Toronto, Canada Medical Research Council (MRC) Immunochemistry Unit, Department of Biochemistry, University of Oxford, Oxford OX1 3QU, United Kingdom
a r t i c l e
i n f o
Article history: Received 4 August 2010 Accepted 1 September 2010 Available online 9 September 2010 Keywords: C-type lectin Collectin Conglutinin Purification Complement activation
a b s t r a c t Bovine conglutinin is a serum protein involved in innate immunity. It binds calcium dependently to iC3b, a product of the complement component C3 deposited on cell surfaces, immune complexes or artificial surfaces after complement activation. We here present a simple and efficient two-step procedure for the purification of conglutinin. In the first step, bovine serum is incubated with non-coupled chromatographic TSK beads at 37 °C to allow complement activation and iC3b deposition on the beads and subsequent binding of conglutinin to iC3b. Conglutinin is then eluted from the beads by EDTA. In the second step, conglutinin is separated from iC3b and IgM by ion-exchange chromatography. This purification procedure yielded 81 μg of conglutinin per ml of serum with a recovery of 61.2%. Surface plasmon resonance analysis showed that the purified conglutinin had a high affinity for mannan (Kd = 2.3 − 3.2 nM). SDS-PAGE and time-resolved immunofluorometric assays showed that the conglutinin was not contaminated with other serum collectins such as collectin-43 or mannan-binding lectin. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Bovine conglutinin (BK) is a member of the collectin family of proteins composed of C-type lectin domains connected to collagen-like regions. The abundant form of BK is a dodecamer of 43 kDa monomers arranged in a flexible cross-like structure of four homotrimers. Conglutinin binds microorganisms either directly or via the complement degradation product iC3b through the lectin domains. The binding leads to agglutination of the microorganisms and it Abbreviations: BK, bovine conglutinin; CL-43, collectin-43; EDTA, ethylenediamine tetra-acetic acid; HBS, HEPES buffered saline; MBL, mannan-binding lectin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SPR, surface plasmon resonance; TBS, tris-buffered saline; TRIFMA, time-resolved immunofluorometric assay. ⁎ Corresponding author. Tel.: + 45 65503775; fax: + 45 65503922. E-mail address: [email protected] (U. Holmskov). 0022-1759/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2010.09.010
promotes phagocytosis, and BK thereby plays an important role in the bovine innate immune defence system (for review see Holmskov (2000)). We have previously shown that serum concentrations of BK vary among individual cows, that the levels are genetically determined and that low BK serum concentrations predispose to infection (Holmskov et al., 1998). These results lead to a renewed interest in conglutinin, especially in cattle breeding programs. We have developed a time-resolved immunofluorometric assay for large-scale measurement of serum BK (Krogh-Meibom et al., 2004) and purified BK was needed to establish the assay standard. Also, purified BK is used for the conglutination test to study iC3b coated immune complexes (Tsai et al., 1998) and for purification of immune complexes (Casali and Lambert, 1979). We here describe a simple two-step purification procedure for BK based on iC3b affinity chromatography and ion-exchange chromatography.
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
2. Materials and methods 2.1. Buffers Tris buffered saline (TBS): 10 mM Tris-Base (Tris[hydroxymethyl]aminomethane, Sigma, St. Louis, MO, cat. no. T-1503), 140 mM NaCl, 0.05% emulfogen (polyoxyethylene 10 tridecyl ether, Sigma, cat. no. P-2393), pH 7.4; TBS–Ca: TBS with 5 mM CaCl2; TBS–Ca 1 M NaCl: TBS–Ca containing a total of 1 M NaCl; TBS–EDTA: TBS with 5 mM EDTA; SDS-PAGE sample buffer: 10 mM Tris–HCl, 1 mM EDTA, 10% SDS (Sigma, cat. no. L-4509), 2% Bromophenol Blue; SDS-PAGE running buffer: 14% glycin, 3% Tris-Base, 1% SDS; piperazine loading buffer: 20 mM piperazine (diethylenediamine, Sigma, cat. no. P4, 590-7), 50 mM NaCl, 5 mM EDTA, 0.05% emulfogen, pH 6.2; HEPES buffered saline (HBS): 10 mM HEPES (N-[2hydroxymethyl]piperazine-N′-[2-ethanesulfonic acid], Sigma, cat. no. H-7523), 150 mM NaCl, 0.005% surfactant P-20 (BIAcore, Herts, U.K., cat. no. BR-1000-54), 0.02% NaN3, pH 7.4; HBS–Ca: HBS with 5 mM CaCl2 and HBS–EDTA: HBS with 5 mM EDTA. 2.2. Serum isolation and storage Blood from two cows was randomly collected at the local abattoir and allowed to clot overnight at 4 °C. Serum was isolated by centrifugation at 1000× g for 30 min. Sera from the two cows were pooled and stored in 500 ml aliquots at −20 °C until use. The BK concentration in the pooled serum was 132 μg/ml. The concentrations in adult cows vary from approximately 100 ng/ml to above 1 mg/ml with a geometrical mean of approximately 40 μg/ml (Krogh-Meibom et al., 2004). 2.3. Purification of conglutinin The chromatographic beads (TSK, Toyopearl HW-75F, Sigma, cat. no. 8-07469) were washed extensively with water followed by TBS–Ca buffer on a glass filter. Serum was thawed overnight at room temperature, centrifuged for 30 min at 10,000× g, and filtered through a glass filter. A volume of 100 ml serum was incubated with 60 ml of TSK beads for 2.5 h at 37 °C in a 200 ml glass flask. The TSK beads were gently resuspended every 30 min. The TSK beads were packed under flow on a computer-monitored FPLC system (FPLCdirector Version 1.3, Amersham Biosciences, Uppsala, Sweden) and washed first with TBS–Ca 1 M NaCl until the absorbance at 280 nm was below 0.1, and then equilibrated with TBS–Ca buffer until the absorbance reached the baseline. The column was then eluted with TBS–EDTA, and the fractions containing protein were pooled. The EDTA eluate was diluted 1/4 in piperazine loading buffer pH 6.2 and applied to a 1 ml Mono Q anion-exchange column (Amersham Biosciences, cat. no. 17-0546-01) at a flow rate of 0.5 ml/min. After washing with loading buffer, retained protein was eluted with a linear gradient from 50 to 375 mM NaCl over 30 column volumes in piperazine loading buffer pH 6.2. Samples of the starting materials and from each purification step were collected and analysed by SDS-PAGE and in the BK–TRIFMA-assay as described below.
205
As a control, a volume of 100 ml serum was heatinactivated for 1 h at 56 °C and used in the purification procedure described above. As further controls, 1 ml TSK was incubated with purified BK (100 μg/ml) for 2.5 h at 37 °C, and 1 ml volumes of TSK that had previously been incubated with serum and eluted with TBS–EDTA were incubated with either purified BK or heat-inactivated serum. The TSK beads were then washed with TBS–Ca 1 M NaCl and with TBS–Ca by repeated suspension, centrifugation and removal of the supernatant. Finally, the beads were resuspended in 1 ml TBS–EDTA (10 mM EDTA) and the supernatants were analysed by SDS-PAGE and by the BK–TRIFMA-assay. 2.4. SDS-PAGE Electrophoresis was performed on 4–20% polyacrylamide gradient gel and samples were reduced by heating at 100 °C for 2 min in 50 mM dithiothreitol diluted in SDS-PAGE sample buffer and alkylated by the addition of iodoacetamide to a concentration of 100 mM. Non-reduced samples were heated in SDS-PAGE sample buffer with 2 mM iodoacetamide. Samples of 10 μl were loaded per lane. Protein bands were detected with Coomassie Brilliant Blue. The molecular mass markers were those of the Mark 12 standard (Invitrogen NOVEX, San Diego, CA, cat. no. LC5677) or Precision Protein Standards (Bio-Rad, cat. no. 161-0362). 2.5. N-terminal amino acid sequencing Samples were reduced and run on a 10% NOVEX Bis-Tris NuPAGE precast gel (Invitrogen NOVEX) at 200 mA per gel in a NOVEX XCell II Mini-cell gel apparatus. The gel was electroblotted to a NOVEX 0.2 μm PVDF membrane in a NOVEX blot module. The membrane was then stained with Coomassie Brilliant Blue. The bands of interest were excised from the PVDF membrane and washed extensively with 10% methanol in water prior to sequencing. They were then sequenced on an Applied Biosystems 494A ‘Procise’ protein sequencer (Applied Biosystems, Warrington, U.K.) using standard sequencing cycles. 2.6. Amino acid analysis Samples were run on an ABI 420A analyser (PE Biosystems, Warrington, U.K.) following hydrolysis for 24 h in 5.7 M hydrochloric acid at 110 °C. Data were integrated using Dionex Chromeleon v6.4 software (Dionex, Macclesfield, U.K.). 2.7. Time-resolved immunofluorometric assay (TRIFMA) for conglutinin A BK–TRIFMA-assay was used to estimate the recovery of BK (Krogh-Meibom et al., 2004). Briefly, wells of microtiter plates (MaxiSorp, Nunc, Kamstrup, Denmark) were coated with anti-BK polyclonal antibodies by overnight incubation. The wells were washed and incubated overnight with dilutions of BK containing samples. The wells were then incubated stepwise with an anti-BK monoclonal antibody and a biotin-labelled subclass specific rabbit anti-mouse antibody, followed by Eu3+ chelates coupled to streptavidin. After
206
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
addition of a solution to release the Eu3+ ions, UV-light induced light emission was measured. 2.8. Surface plasmon resonance analysis (SPR) Immobilization of biotinylated mannan (1 mg/ml) on a streptavidin-coated BIAcore chip (SA-chip, BIAcore) was performed in the biosensor system in 100 mM NaOAc (pH 5.5) buffer at a continuous flow of 5 μl/min for 10 min. Leaving flow cell 1 as blank, the next three flow cells were used for immobilization of yeast mannan (Sigma, M-3640), which was treated with 0.5, 1.5 or 4.5 mM NaIO4 and reacted with biotin-LC-hydrazide (Pierce, Perbio Science, Cheshire, U.K., cat. no. 21340) according to the manufacturer's protocol. Free streptavidin was blocked with biotin, and the chip was normalized with 50% glycerol solution. Purified BK in the range of 0–20 μg/ml in HBS–Ca buffer was injected and allowed to bind with the immobilized mannan. A flow rate of 10 μl/min was maintained for 2 min at 25 °C. The complexes were allowed to disassociate for 2 min, and bound proteins were washed with two pulses of 5 μl
3 mM EDTA, 0.1% SDS solution. The flow cells were reequilibrated after washing with 20 μl HBS–EDTA and 20 μl HBS–Ca buffer. The SPR tracings from one of the flow cells were used for the calculation of disassociation and association rates by BIAcore 2000 software. In order to calculate the affinity of BK to mannan, the molecular weight of the lectin was assumed to be 516 kDa (dodecamer). 3. Results and discussion We have previously purified conglutinin by polyethylene glycol precipitation, affinity chromatography on mannan– Sepharose and an anti-bovine Ig–Sepharose step to remove contaminating anti-carbohydrate antibodies (Holmskov et al., 1993). Other methods employing precipitation, adsorbtion to zymozan, ion-exchange chromatography, enzyme digestion and gel filtration have also been employed (Maire et al., 1981). Common to the known procedures is the fact that contamination with the other known bovine serum collectins mannan-binding lectin (MBL) and collectin (CL)-43 is likely to occur. TSK is a methacrylate-based matrix designed for
Fig. 1. Purification of conglutinin from bovine serum. A: EDTA elution profile of the TSK column incubated with 100 ml of serum. Fractions containing conglutinin are indicated by arrows. B: Elution profile of the conglutinin containing fractions from panel A applied to a Mono Q ion-exchange column and eluted with a linear NaCl gradient. The small vertical bar indicates the fractions used for total amino acid analysis (aaa, fractions 19–20) and the larger bar represent the pool of fractions used to determine the total yield of conglutinin (BK Mono Q pool, fractions 19–21). C: Elution profile of the purified conglutinin (from panel B) on a Superose 6 gel filtration column. The column was calibrated with: a, blue dextran (N 2000 kDa); b, beta-amylase (200 kDa); c, bovine serum albumin (66 kDa); d, cytochrome C (12.4 kDa). D: SDS-PAGE (4–20% gradient gel) of reduced conglutinin from the Superose 6 fractions indicated by a bar in panel C. The molecular weight markers indicated are the Precision Protein Standards.
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
gel filtration. The matrix is synthezised by copolymerisation of glycidyl methacrylate, polyethylene glycol and the crosslinking agent pentaerythritol. It contains both primary and secondary hydroxyl groups that can be activated for immobilization of ligands for affinity chromatography applications (Hermanson et al., 1992). The hydroxyl groups are potential targets for C3b activated via the alternative complement pathway. As opposed to agarose-based matrixes, TSK has a low intrinsic affinity to carbohydrate-binding proteins. TSK coupled with different sugars has previously been used to purify collectins (Hansen et al., 2002). In order to simplify the purification procedure for BK, we used TSK coupled with N-acetyl-D-glucosamine as affinity matrix and non-coupled TSK as precolumn. To our surprise, all BK was bound to the precolumn. BK did not bind to noncoupled TSK if heat-inactivated serum was used, indicating that complement deposition was required for BK binding. BK is stable to heat-inactivation (Kawasaki et al., 1993) but factor H and factor I, responsible for cleaving C3b into iC3b, are inactivated. BK was then purified by a two-step procedure (Fig. 1). First bovine serum was incubated with TSK beads at 37 °C in order to allow iC3b deposition and subsequent binding of BK. After packing and extensive washing of the TSK in a FPLC column BK was eluted from the column with EDTA (Fig. 1A). The EDTA eluate was then separated by ion-exchange chromatography on Mono Q where two major peaks were seen (Fig. 1B). Peak two, which contained BK, was further analysed by Superose 6 gel filtration chromatography (Fig. 1C). In accordance with previous findings by Kawasaki et al. (1993) and Holmskov et al. (1995) the major form of BK followed by the truncated form eluted close to the void volume (Fig. 1C and D). Samples from the two purification steps were analysed by SDS-PAGE (Fig. 2A). Lane 4 shows the EDTA eluate from the TSK column incubated with serum in the reduced state. One major band at approximately 43 kDa is seen, together with two broad diffuse bands of 80 kDa and 30 kDa, a double band of 66 kDa and two faint bands at approximately 49 kDa and 41 kDa. The N-terminal amino acid sequences of the 43 and 49 kDa bands were identical (AEMTTFSQKI) and matched with the N-terminal sequence of the mature BK protein (Lee et al., 1991). The sequence of the 41 kDa band was AGRPGWVGPI, and this sequence corresponds to a form of BK truncated after 54 amino acids (Kawasaki et al., 1993). The post-translational modification that may account for the different sizes (43 and 49 kDa) of the two non-truncated forms of BK is not known, but differential O-glycosylation, which has been reported for the structurally related SP-D molecule, could be an explanation. N-terminal sequencing did not identify the bands with molecular masses of 80 kDa and 30 kDa, but they probably correspond to the heavy and light chains of IgM. The double band around the 66 kDa was identified as the bovine C3 betachain (NPLYSMITPN) (upper band), and as the large iC3b alpha-chain fragment (SDLDDDIIPE) (lower band). When the EDTA eluate was separated by Mono Q ionexchange chromatography, two major peaks occurred. Peak one (Fig. 2A, lanes 8 and 9) contained the bands corresponding to C3b and IgM and two weak bands at 42 kDa and 44 kDa. Peak two (lanes 6 and 7) contained the bands corresponding to BK. The N-terminal sequence of the 42 kDa
207
Fig. 2. Purity and activity of conglutinin (BK). A: SDS-PAGE of reduced (lanes 2, 4, 6, 8) and non-reduced (lanes 3, 5, 7, 9) protein on a 4–20% gradient gel from the BK purification steps and of control fractions. Lanes 2 and 3: pooled EDTA eluate from TSK incubated with heat-inactivated serum. Lanes 4 and 5: pooled EDTA eluate from TSK incubated with serum not heat-inactivated (tubes 8–11, Fig. 1A). Lanes 6 and 7: Mono Q pool of protein eluted late (fractions 19–21) in the NaCl gradient. Lanes 8 and 9: Mono Q pool of protein eluted first (fractions 12–15) in the NaCl gradient. Lanes 1 and 10: the molecular mass markers indicated are those of the Mark 12 standard. B: SPR analysis of BK–mannan interaction. The BK (association) followed by HBS–Ca buffer (disassociation) was flowed on the mannan-coated chip for 2 min each. The BK-specific SPR from different concentrations of the lectin (20, 10, 5, 2.5, 1.25, 0.625, 0.313 and 0.0 μg/ml in that order) was aligned for the affinity measurements (SPR during the wash steps is not shown).
band from the C3b/IgM peak identified the peptide as the small iC3b alpha-chain fragment that is generated when the 2 kDa C3f fragment is released by factor I to generate
208
T. Krogh-Meibom et al. / Journal of Immunological Methods 362 (2010) 204–208
Table 1 Recoveries of conglutinin. Purification fraction
Heat-inactivation (±)
N
Concentration ⁎ (μg/ml)
Volume (ml)
Recovery from serum ⁎ (%)
Serum EDTA pool BK Mono Q pool Serum EDTA pool
− − − + +
4 4 4 4 4
132 ± 4.00 1090 ± 20.2 2700 ± 54.2 132 ± 1.85 0.13 ± 0.015
100 10 3 100 7.5
100 ± 3.03 82.2 ± 1.53 61.2 ± 1.23 99.8 ± 1.40 0.0073 ± 0.00085
⁎ Mean values ± sem.
iC3b (SEETKENEGF). The faint bands seen in Fig. 2A, lanes 6 and 7 corresponding to molecular masses of 85 kDa and 75 kDa are probably BK dimers. The size of the non-reduced band at 200 kDa (lanes 5 and 9) is consistent with that expected for bovine C3b and the band seen at the top of the gel corresponds to the expected molecular mass of nonreduced IgM. No detectable bands corresponding to CL-43 or MBL were seen in the samples from peak two from the Mono Q column (Fig. 2A, lanes 6 and 7), and no signals were obtained when the same fractions were analysed by CL-43– TRIFMA (data not shown). Heat-inactivated serum did not mediate the binding of BK to TSK (Fig. 2A, lanes 2 and 3) and there was no significant binding of purified BK to TSK that had not been preincubated with serum. Purified BK and BK from heat-inactivated serum bound to serum treated TSK beads after the initial EDTA elution indicating that the beads can be used several times (data not shown). Total amino acid analysis of the BK fractions indicated in Fig. 1B showed a concentration of 3.36 ± 0.055 mg/ml. The recoveries of BK were assessed by the BK–TRIFMA-assay (Table 1) using the total amino acid analysed fractions as standard. The purification procedure used yielded 81 μg BK per ml of bovine serum. The recovery of BK was 61.2%. The lectin activity of the purified BK was evaluated by SPR analysis (Fig. 2B). The association and dissociation rates between purified BK and mannan were in the range of 1.0 × 105 to 4.9 × 105 M/s and 1.1 × 10−3 to 3.3 × 10−3 l/s, respectively. The resulting dissociation constant (Kd) is between 2.3 × 10−9 and 3.2 × 10−9 M, which is comparable to the Kd's of the native forms of other collectins such as MBL, surfactant protein A, surfactant protein D and CL-43 (Holmskov, 2000). 4. Conclusion We have developed a simple two-step procedure for the purification of bovine conglutinin based on the iC3b binding
property of conglutinin and ion-exchange chromatography. This method is easy to perform and gives high yields. The purified BK is free of the other known serum collectins, and is a functional dodecamer. References Casali, P., Lambert, P.H., 1979. Purification of soluble immune complexes from serum using polymethylmetacrylate beads coated with conglutinin or C1q. Application to the analysis of the components of in vitro formed immune complexes and of immune complexes occurring in vivo during leishmaniasis. Clin. Exp. Immunol. 37, 295. Hansen, S., Holm, D., Moeller, V., Vitved, L., Bendixen, C., Reid, K.B., Skjoedt, K., Holmskov, U., 2002. CL-46, a novel collectin highly expressed in bovine thymus and liver. J. Immunol. 169, 5726. Hermanson, G.T., Mallia, A.K., Smith, P.K. (Eds.), 1992. In: Immobilized Affinity Ligand Techniques, 1. Academic Press, pp. 31–34. Holmskov, U.L., 2000. Collectins and collectin receptors in innate immunity. APMIS 108 (Suppl. 100), 1. Holmskov, U., Holt, P., Reid, K.B., Willis, A.C., Teisner, B., Jensenius, J.C., 1993. Purification and characterization of bovine mannan-binding protein. Glycobiology 3, 147. Holmskov, U., Laursen, S.B., Malhotra, R., Wiedemann, H., Timpl, R., Stuart, G.R., Tornoe, I., Madsen, P.S., Reid, K.B., Jensenius, J.C., 1995. Comparative study of the structural and functional properties of a bovine plasma C-type lectin, collectin-43, with other collectins. Biochem. J. 305, 889. Holmskov, U., Jensenius, J.C., Tornoe, I., Lovendahl, P., 1998. The plasma levels of conglutinin are heritable in cattle and low levels predispose to infection. Immunology 93, 431. Kawasaki, N., Yokota, Y., Kawasaki, T., 1993. Differentiation of conglutination activity and sugar-binding activity of conglutinin after removal of NH2terminal 54 amino acid residues by endogenous serine protease(s). Arch. Biochem. Biophys. 305, 533. Krogh-Meibom, T., Holmskov, U., Løvendahl, P., Nielsen, N.I., Ingvartsen, K.L., 2004. A time-resolved immunofluorometic assay for quantification of the bovine collectin conglutinin. J. Immunol. Methods 286 (1–2), 87. Lee, Y.M., Leiby, K.R., Allar, J., Paris, K., Lerch, B., Okarma, T.B., 1991. Primary structure of bovine conglutinin, a member of the C-type animal lectin family. J. Biol. Chem. 266, 2715. Maire, M.A., Barnet, M., Lambert, P.H., 1981. Purification of bovine conglutinin using pepsin digestion. Mol. Immunol. 18, 85. Tsai, J.F., Margolis, H.S., Jeng, J.E., Ho, M.S., Chang, W.Y., Hsieh, M.Y., Lin, Z.Y., Tsai, J.H., 1998. Immunoglobulin- and hepatitis B surface antigen-specific circulating immune complexes in chronic hepatitis B virus infection. Clin. Immunol. Immunopathol. 86, 246.