Monoclonal antibody 10A5 recognizes an antigen unique to the water-insoluble 25/45 membrane fraction of the rat ocular lens

Screening of hybridoma supernatants by ELISA and immunoblotting

Approximately 1600 hybridoma supernatants were tested by indirect ELISA against the 25/45 fraction and NSMF isolated from 20-day-old rats. Of these, nearly 10% recognized the homologous membrane fraction used for mouse immunization. All but one of the supernatants (MAb 10A5) also reacted with the heterologous membrane fraction. As shown in Figure 1A, immunoblotting revealed that MAb 10A5 recognized a low molecular weight antigen that was unique to the 25/45 fraction, but absent from the water-soluble supernatant fraction (WSF) and the NSMF isolated from 20-day-old rats. The antigen migrated as a series of bands between 10 kD and 15 kD and could be visualized by immunoblotting (Figure 1A), but not by staining with Coomassie Blue R-250 (Figure 1B). In blots of lens fractions prepared from 75-day-old rats (Figure 2A), the antigen was detected in both the NSMF and 25/45 fraction. The multi-band pattern in samples from 75-day-old animals was less distinct than that seen in 20-day-old animals.

Enzymatic deglycosylation of the 25/45 fraction

The 25/45 fraction from 20-day-old rats was evaluated for carbohydrate residues using an enzymatic deglycosylation procedure that removes N-linked and O-linked carbohydrates from glycoproteins (Figure 3). The presence of glycoproteins in the 25/45 fraction was demonstrated by the minor differences observed between untreated (lane 3) and deglycosylated (lane 4) samples stained with Coomassie Blue R-250. These differences were most apparent between 50 and 150 kD. Deglycosylation of the 25/45 fraction did not change the electrophoretic mobility of the antigen reactive with MAb 10A5 or the intensity of its recognition by the antibody (lane 7) compared to the untreated control (lane 6). Deglycosylation of the bovine fetuin positive control (lane 2) resulted in a major shift in its molecular weight compared to the untreated control (lane 1).

Immunoprecipitation of the antigen reactive with MAb 10A5, followed by 1-D and 2-D electrophoresis

The antigen reactive with MAb 10A5 was concentrated by immunoprecipitation from the 25/45 fraction isolated from 20-day-old rats. The immunoprecipitate was subjected to one-dimensional (1-D) (Figure 4A) and two-dimensional (2-D) electrophoresis (Figure 4B) prior to analysis by matrix-assisted laser desorption/ionization-time-of-flight/mass spectrometry (MALDI-ToF/MS). The IgG heavy (IgG_H) and light (IgG_L) chains comprising MAb 10A5 were evident at 55 kD and 25 kD, respectively, in both 1-D and 2-D gels. In the blot of the 1-D gel (Figure 4A), the immunoprecipitated antigen was strongly recognized by MAb 10A5 within the 10–15 kD mass range, but the antigen remained unstained by colloidal Coomassie Blue G-250 in the replicate gel. None of the additional stains, including Alcian blue, silver nitrate, imidazole-zinc sulfate, Stains-all, and Stains-all counterstained by silver nitrate, stained the antigen even after it was concentrated by immunoprecipitation (data not shown). In blots of 2-D gels (Figure 4B), the antigen focused toward the anode and within the expected 10–15 kD mass range. Areas of the 1-D and 2-D gels submitted for MALDI-ToF/MS are marked by boxes in Figure 4A and 4B.

MALDI-ToF/MS

Proteins in gel slices containing the antigen recognized by MAb 10A5 were provisionally identified by MALDI-ToF/MS. The mass lists compiled for the 1-D versus 2-D gel slices were completely unique, with no proteins identified in common. The sole protein identified with confidence in the 1-D gel slice was structural maintenance of chromosomes 1-like 1 (SMC1, gi 13928946), a protein involved in sister chromatid cohesion during the mitotic cell cycle (Zou 2011). An Excel file containing the peptide summary report for the 1-D gel slice has been provided [see Additional file 1]. The sole protein identified with confidence in gel slices excised from duplicate 2-D gels was anionic trypsin-1 precursor (gi 6981420), which may have been a contaminant contributed by the trypsin preparation used in the MALDI-ToF procedure. Details of the peptide summary report for the 2-D gel slices are presented as an additional Excel file [see Additional file 2].

Protease digestion of the 25/45 fraction from 20-day-old rats

Trypsin or proteinase K digestion of the 25/45 fraction for up to 48 h had no effect on the immunoreactivity or electrophoretic mobility of the 10–15 kD antigen reactive with MAb 10A5 (Figure 5A). In contrast, membrane proteins in the 25/45 fraction were completely destroyed by protease treatment, as demonstrated by Coomassie staining of the digests after SDS-PAGE (Figure 5B). Within 2 h of digestion, the only discernible proteins that remained in the digests were the proteases trypsin and proteinase K, which migrated to 23.8 kD and 28.9 kD, respectively (arrows, Figure 5B).

Folch extraction, thin layer chromatography, and dot blot immunoassay

Folch extraction of the 25/45 fraction isolated from 20-day-old rats yielded three distinct phases: an upper phase, a flocculent interface, and a lower phase. Both the upper phase and interface were reactive with MAb 10A5 in immunoblots (Figure 6A). Thin-layer chromatography of the upper phase in a solvent system comprising propanol/water (7:3 v/v) resolved a single spot that stained with resorcinol (Figure 6B). When eluates from silica gel bands scraped from a duplicate TLC plate were probed with MAb 10A5 by immunoblotting, MAb 10A5 reactivity was present only in the band that corresponded with the ganglioside spot revealed by resorcinol staining (Figure 6C). Resolution of the immunoreactive spot through HPTLC plates in a solvent system comprising chloroform/methanol/0.2% aqueous CaCl₂ (55:45:10 v/v/v) demonstrated that it co-migrated with the GD1a bovine ganglioside standard (Figure 6D). However, neither purified bovine GD1a nor any of the additional bovine brain gangliosides in the standard mixture reacted with MAb 10A5 in a dot blot immunoassay (Figure 7, panel 3). The proteoglycans chondroitin sulfate and heparan sulfate failed to migrate from the application origin on HPTLC plates developed with the chloroform/methanol/0.2% aqueous CaCl₂ (55:45:10 v/v/v) solvent system and stained poorly with resorcinol (Figure 8).

Rat lens membrane preparations

All protocols involving animals were approved by the A.T. Still University Institutional Animal Care Committee and were conducted in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals. The 25/45 fraction and NSMF were isolated from the lenses of 20-day-old and 75-day-old Sprague Dawley rats as previously described (Fleschner 1998). Briefly, a 10% homogenate (wet tissue weight to buffer volume) was prepared by Dounce homogenization of decapsulated lenses in buffer comprising 5 mM Tris, 1 mM EDTA, and 5 mM β-mercaptoethanol, pH 8.0, containing a protease inhibitor cocktail (P8340) (Sigma Chemical Co., St. Louis, MO). The homogenate was centrifuged at 20,000 g for 30 min to obtain the water-insoluble sedimenting membrane fraction and water-soluble supernatant fraction (WSF). The sedimenting membrane fraction was further fractionated by discontinuous sucrose density gradient centrifugation through 25%, 45% and 50% sucrose at 100,000 g for 120 min. The 25/45 fraction was collected from the interface between 25% and 45% sucrose. The NSMF was isolated by adjusting the density of the water-soluble supernatant to 1.22 g/ml with solid KBr, centrifuging the solution at 100,000 g for 16 h, and removing the floating NSMF from the top of the solution. The NSMF was washed twice, dialyzed to reduce KBr concentration, and then concentrated by centrifugation at 68,000 g for 60 min.

Hybridoma production and screening

Groups of four female BALB/c mice were immunized with the 25/45 fraction or NSMF isolated from 20-day-old rats in three intraperitoneal injections over the course of three months. Each injection contained 100 μg of protein in a total volume of 200 μl. For the first immunization, the antigen was emulsified in 50% Freund’s complete adjuvant (MP Biomedicals, Solon, OH). The second and third intraperitoneal injections were given in 50% Freund’s incomplete adjuvant (MP Biomedicals). Titers of sera obtained by tail bleed 11 days after the third intraperitoneal injection were determined by immunoblotting (Towbin et al. 1979) against the homologous antigen used for immunization. Three days prior to hybridoma production, the mouse from each group that had the highest antibody titer received an intravenous booster immunization containing 50 μg of antigen in 50 μl of phosphate-buffered saline (PBS, pH 7.4). Splenocytes from the mice were fused to Sp2/0-Ag14 myeloma cells as described by Van Deusen (1983) and selected in HAT medium prepared from Dulbecco’s modified Eagle’s medium supplemented with 15% horse serum (Sigma-Aldrich, St. Louis, MO).

Beginning 10 days after fusion, undiluted hybridoma culture supernates were screened for monoclonal antibodies (MAbs) by indirect enzyme-linked immunosorbent assay (ELISA) (Voller et al. 1976) against microtiter plates coated with 1 μg/well of 25/45 fraction or NSMF from 20-day-old rats. Reactive supernates were detected with 1:2000 alkaline phosphatase-conjugated goat anti-mouse immunoglobulins (IgG, IgM, IgA) (Sigma) and p-nitro phenyl phosphate substrate solution (Pierce Chemical Co., Rockford, IL). Supernatants that were reactive with the membrane fraction used for mouse immunization were subsequently tested by ELISA against the heterologous membrane fraction. From this screening protocol, a single monoclonal antibody was identified that reacted with one membrane fraction but not both: MAb 10A5 recognized an antigen unique to the 25/45 fraction that was absent from the NSMF in 20-day-old animals. The hybridoma line secreting MAb 10A5 was cloned three times by limiting dilution (Campbell 1984) and adapted to growth in HybriMax serum- and protein-free medium (Sigma). The MAb was concentrated and the medium was exchanged for PBS by ultrafiltration through Biomax-30 membranes (Millipore, Billerica, MA). MAb 10A5 was identified as an IgG2b antibody using a Mouse Typer™ Isotyping Kit (Bio-Rad, Hercules, CA).

SDS-PAGE and immunoblotting of lens membrane fractions

Ocular lens fractions were subjected to SDS-PAGE as described by Laemmli (1970) through 18% or 5-20% polyacrylamide gradient resolving gels, and the proteins were visualized by staining with 0.1% w/v Coomassie Blue R-250 in 40% methanol/10% acetic acid. Alternatively, proteins were electrophoretically transferred from the gels to polyvinylidene fluoride (PVDF) membranes and probed by immunoblotting as described by Towbin et al. (1979). Briefly, membranes were blocked by a 1 h incubation in Tris-buffered saline (TBS; 20 mM Tris, 150 mM NaCl, pH 7.5) containing 5% nonfat dry milk (NFDM) before being incubated for 90 min in MAb 10A5 diluted to 1 μg/ml in TBS containing 0.05% Tween 20 (TTBS) and 1% NFDM. The membrane was washed in TTBS and then incubated for 90 min in 1:3000 alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma). After further washes, membranes were developed with nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate substrate (NBT/BCIP) (Bio-Rad).

Enzymatic deglycosylation of the 25/45 fraction

The 25/45 fraction from 20-day-old rats was subjected to enzymatic deglycosylation (GlycoPro™ kit, Prozyme, San Leandro, CA) to remove N-linked (asparagine-linked) and simple O-linked (serine/threonine-linked) carbohydrates from glycoproteins. One hundred μg of 25/45 fraction protein in 30 μl of distilled water were mixed with 10 μl of 5X incubation buffer (0.25 M sodium phosphate, pH 7.0) and 2.5 μl of denaturation buffer (2% SDS, 1% 2-mercaptoethanol). The mixture was heated at 100°C for 5 min. After the mixture had cooled to room temperature, 2.5 μl of 15% NP-40 detergent were added, followed by 1 μl each of N-glycanase, sialidase A, and O-glycanase. Incubation with the enzymes was allowed to proceed for 3 h at 37°C, after which aliquots containing 8 μg of the deglycosylated 25/45 fraction protein were resolved by SDS-PAGE through 5-20% gradient gels. Control lanes were loaded with untreated 25/45 fraction protein (8 μg), and with untreated or deglycosylated bovine fetuin (1 μg/lane). The proteins were transferred to PVDF membranes for staining with Coomassie Blue R-250 or immunoblotting with MAb 10A5 as described previously. The membranes were examined to determine whether deglycosylation of the samples caused a shift in molecular weight or a reduction in reactivity with MAb 10A5 compared to untreated controls.

Immunoprecipitation of the antigen reactive with MAb 10A5

Immunoprecipitation was performed as described by Stuart and Chamberlain (2003). An aliquot containing 150 μg of 25/45 fraction protein from 20-day-old rats was adjusted to a final concentration of 1% Triton X-100 (TX100) in a total volume of 150 μl PBS. The solution was sonicated on ice by three 15-sec bursts, incubated for 60 min in a 37°C water bath with occasional vortexing, and centrifuged for 10 min at 10,000 rpm in a minifuge. The supernatant was incubated for 90 min with 150 μl of Protein G Dynabeads (Invitrogen Corp., Carlsbad, CA) pre-coated with MAb 10A5 (1 μg antibody/μl beads). The beads were washed five times in PBS-0.01% Tween 20, and the immune complexes were eluted into 25 μl of Laemmli denaturing sample buffer (62.5 mM Tris–HCl, pH 6.8, 2% SDS, 5% 2-ME, 20% glycerol, 0.1% bromophenol blue) by incubation for 5 min in a boiling water bath. Eluates containing the immune complexes (12 μl/lane) were subjected to electrophoresis through 18% polyacrylamide gels. The gels were stained with Coomassie Blue R-250 as previously described, or with one of six alternative stains: colloidal Coomassie Blue G-250 (University of Missouri Proteomics Center 2006), Alcian blue (Wall and Gyi 1988), silver nitrate (Terry et al. 2004), imidazole-zinc sulfate (Castellanos-Serra et al. 1999; Hardy et al. 1997), Stains-All (Goldberg and Warner 1997), or a combination of Stains-All and silver nitrate (Goldberg and Warner 1997; Terry et al. 2004). The colloidal Coomassie Blue G-250 staining procedure was used to fix and stain gels in preparation for MALDI-ToF/MS, described below. The gel was washed three times in distilled water, stained overnight at room temperature in 300 ml of Coomassie Brilliant Blue G-250 (0.08% w/v in 20% ethanol, 1.6% phosphoric acid, and 8% ammonium sulfate), and then destained by extensive washing in distilled water.

2-Dimensional (2-D) electrophoresis

The antigen reactive with MAb 10A5 was immunoprecipitated from 450 μg of TX100-solubilized 25/45 fraction using 270 μl of MAb 10A5-coated Protein G Dynabeads. Immune complexes were eluted from the beads into 375 μl of isoelectric focusing (IEF) sample buffer (9 M urea, 4% CHAPS, 3 mM tributylphosphine, 0.04% Bio-Lyte 3/10 ampholytes) by incubation for 1 h in a 37°C water bath with occasional vortexing. The eluate was split into three 125-μl portions, each of which was used to rehydrate a 7-cm ReadyStrip™ IPG strip (Bio-Rad), pI 3–10, under active rehydration conditions. The strips were focused in a programmable Protean IEF cell (Bio-Rad), after which they were equilibrated by successive 10-min incubations in buffer I (6 M urea, 0.375 M Tris–HCl, pH 8.8, 2% SDS, 20% glycerol, 2.2% dithiothreitol) and buffer II (6 M urea, 0.375 M Tris–HCl, pH 8.8, 2% SDS, 20% glycerol, 1.25% [w/v] iodoacetamide). The IPG strips were then subjected to second-dimension SDS-PAGE through 18% polyacrylamide gels. The contents of one gel were transferred to a PVDF membrane and immunblotted with MAb 10A5 to confirm the presence of the antigen. The two replicate gels were fixed and stained with colloidal Coomassie Blue G-250.

MALDI-ToF and MS analysis

For MALDI-ToF/MS, gel slices containing the antigen immunoprecipitated by MAb 10A5 were excised from 1-D and 2-D gels after fixation and staining in colloidal Coomassie Blue G-250. Because the antigen did not take up the dye, it was localized in the gels by comparison to replicate immunoblots probed with MAb 10A5. As an added assurance that gel slices from 2-D gels contained the immunoreactive antigen, ends of the excised pieces were probed with MAb 10A5 by immunoblotting. A 1-mm piece clipped from both ends of each slice were equilibrated in denaturing sample buffer, subjected to SDS-PAGE, transferred to PVDF, and probed with MAb 10A5. Gel slices shown to contain the antigen were submitted to Applied Biomics, Inc. (Hayward, CA) for trypsin digestion and MALDI-ToF/MS on an Applied Biosystems Proteomics analyzer. Mass lists compiled from the mass spectra were searched against the National Center for Biotechnology Information non-redundant (NCBInr) mammalian protein database using GPS Explorer software equipped with the MASCOT search engine.

Protease digestion of the 25/45 fraction

The 25/45 fraction from 20-day-old rats was subjected to in-solution trypsin digestion as described in the product bulletin (Part# 9PIV511) for TPCK-modified sequencing grade trypsin (Promega, Madison, WI). A sample containing 100 μg of 25/45 fraction protein dissolved in 50 μl of protein denaturation buffer (50 mM Tris–HCl, 6 M urea, 4 mM DTT, pH 8.0) was digested at 37°C with 5 μg of trypsin in 300 μl of digestion buffer (50 mM Tris–HCl, 1 mM CaCl₂, pH 7.6). Aliquots containing 20 μg of the 25/45 fraction were removed after 0 h (undigested control containing no enzyme), 2 h, 24 h, and 48 h of incubation. Digestion was stopped by adding 3 μl of 50 mM phenylmethylsulfonyl fluoride (PMSF) to each aliquot and immediately storing the samples at -80°C. Digestion with proteinase K (Fisher Scientific, St. Louis, MO) was performed in similar fashion, except that proteinase K was used at a concentration of 0.1 μg/μl in digestion buffer comprising 50 mM Tris–HCl, 5 mM CaCl₂, pH 7.5, and digestion was stopped by adding 5 μl of 50 mM PMSF to each 20 μg aliquot. All samples were lyophilized to dryness and then reconstituted in 30 μl of Laemmli sample buffer prior to SDS-PAGE through duplicate 18% resolving gels (15 μl/lane). Gels were stained with Coomassie Blue R-250 or immunoblotted with MAb 10A5 as previously described.

Folch extraction of the 25/45 fraction and thin-layer chromatography

A sample containing 400 μg of 25/45 fraction protein from 20-day-old rats was subjected to the Folch lipid extraction procedure (Folch et al. 1957). The upper phase, interface, and lower phase were collected into separate glass tubes, lyophilized to dryness, and rehydrated in 60 μl of PBS. Fifteen μl from each sample were mixed with 5 μl of 4X Laemmli sample buffer and incubated in a boiling water bath for 5 min. Each 20-μl sample was then subjected to SDS-PAGE and immunoblotting to determine which fraction(s) contained the antigen recognized by MAb 10A5.

The upper phase obtained by Folch extraction from 200 μg of 25/45 fraction protein was analyzed by thin-layer chromatography (TLC) on duplicate 20 × 20 cm, 250 micron Uniplate™ Silica Gel H, binder-free plates (Analtech, Newark, DE) with a propanol/water (7:3 v/v) solvent system. One TLC plate was stained with resorcinol (Findlay and Evans 1990), while the duplicate plate was used to recover gangliosides for immunoblotting. For ganglioside recovery, consecutive 2 × 2 cm silica gel bands were scraped from lanes containing the upper phase and eluted into 1 ml of chloroform/methanol (1:1 v/v). Volumes were reduced under a nitrogen stream, and samples were solubilized in 40 μl of Laemmli sample buffer prior to SDS-PAGE (20 μl/lane) and immunoblotting with MAb 10A5.

The upper phase of the 25/45 fraction Folch extract was also subjected to high performance TLC on 10 × 10 cm HPTLC Silica Gel 60 plates (Fisher Scientific) (Schlosshauer et al. 1988). Gangliosides extracted from the equivalent of 20, 9.7, and 6.4 μg of 25/45 fraction protein were spotted onto the HPTLC plates in 1 μl of chloroform/methanol (2:1 v/v) along with bovine mixed ganglioside standards (Matreya, Pleasant Gap, PA) in adjacent lanes. Additional samples applied to the HPTLC Silica Gel 60 plates included the proteoglycans chondroitin sulfate and heparan sulfate, each applied at 10 μg. Plates were developed in a solvent system comprising chloroform/methanol/0.2% aqueous CaCl₂ (55:45:10 v/v/v) and stained with resorcinol.

Ganglioside dot blot immunoassay

MAb 10A5 was tested for reactivity against bovine gangliosides by a dot blot immunoassay adapted from the method of Chabraoui et al. (1993). A PVDF membrane was wetted in methanol, soaked for 5 min in PBS, and then inserted while still wet into a dot blot apparatus (Bio-Rad). Membrane spots were dried by vacuum pressure and then coated with antigens solubilized in methanol (1.5 μl/spot). Antigens included bovine mixed gangliosides (5 μg), purified bovine GD1a (1 μg, Matreya), the Folch upper phase extracted from 2 μg of 25/45 fraction protein, and 50 ng of the irrelevant bacterial protein RadA-6xHis (Richardson et al. 2012), the latter serving as a negative control. After a 90-min incubation at room temperature to facilitate antigen adsorption, the membrane was removed from the apparatus, rinsed in PBS, and blocked for 60 min in PBS containing 5% bovine serum albumin (BSA). Replicate strips cut from the membrane were incubated for 60 min in MAb 10A5 or an isotype-matched negative control antibody [MAb 2A2 specific for RadA-6×His (Richardson et al. 2012)], each diluted to 2 μg/ml in PBS-1% BSA. After several washes in PBS, the strips were incubated for 60 min in 1:2000 goat anti-mouse IgG-alkaline phosphatase, washed again, and then developed in NBT/BCIP substrate. To demonstrate that gangliosides remained bound to the PVDF membrane throughout the dot blot protocol, one of the replicate strips was stained with Coomassie Blue R-250 to enable visualization of white ganglioside spots against a dark blue background.