Electroelution of antigens immobilized on antibody-linked affinity matrices

ANALYTICALBIOCHEMISTRY

177,314-317

(1989)

Electroelution of Antigens Immobilized Antibody-Linked Affinity Matrices Klaus Schulze-Osthoff, Institute

of Experimental

on

Elmar Michels, Bruno Overwien, and Clemens Sorg

Dermatology,

University

of Miinster,

D-4400 Miinster,

West Germany

ReceivedJune6,1988

An electrophoretic elution procedure for desorption of antigens from antibody-linked gel matrices is described which is performed in a commercially available elution device. The technique presented does not involve denaturing conditions such as chaotropic reagents or low and high pH elution buffers. Comparison of electroelution and conventional elution with acidic buffer reveals that the technique is especially suitable for monoclonal antibodies with high affinity for their ligands. Recovery of antigens after electroelution is significantly higher than by desorption with glycineHCl, pH 2.5. There is no loss of antigens during postelution procedures. Proteins are obtained in small elution volumes and can be desalted without sample transfer. 0 196Q Academic Press, Inc.

arose (4). Most of these methods employed stacking gels or dialysis membranes, which are sometimes difficult to handle or lead to unspecific adsorption of proteins. In this study we compared the use of electrophoretic desorption for the isolation of three different antigens from antibody-linked matrices with conventional acidic buffer elution. Electroelution was performed in an elution-trap device with completely inert membranes as reported by Jacobs and Clad (5), which allowed easy handling, especially of small gel volumes. In this system the elution chamber and membrane trap are constructed as open channels, thus being easily accessible from above, so the elution procedure can be monitored and checked for protein content. MATERIALS Preparation

chromatography is a powerful Immunoaffinity method for rapid and single step purification of proteins. Elution of antigens, however, is a serious problem especially in the case of antibodies with high affinity to their ligands. Desorption of antigen from antibody-linked matrices is traditionally accomplished by extremes of pH or chaotropic ions like NaSCN or MgC&. One disadvantage of using these reagents is that elution might cause denaturation and loss of the biological activity of the antigen. Furthermore, conventional elution techniques result in large elution volumes and concentration and desalting of the sample is often required. These difficulties can be overcome if the ligand is separated electrophoretically from the affinity support, which is a mild method and obviates the need of denaturing conditions. Electrophoretic elution from affinity matrices has already been applied to a variety of interactions: the separation of antibodies from protein A-, Con A-, or antigenlinked Sepharose (1,2), of serum albumin from Cibacron blue (3), and of ferritin from antibody-conjugated Seph-

AND METHODS of Antibodies

Affinity chromatography was performed by using mouse monoclonal antibodies 5F4 (IgM)’ raised against the human angiogenic factor (HAF) (6) and the IgG,,antibody 910D7 directed against the human HLA-DR antigen (7). The elution techniques applied were also evaluated for a rabbit anti-BSA antiserum which was obtained from Miles (Munich, FRG). Serum-free supernatants of the monoclonal antibodies were prepared and purified by hydroxylapatite affinity-HPLC as described (6). Antibodies were desalted against PBS, pH 7.2, on PDlO-columns (Pharmacia, Freiburg, FRG). Preparation

of Affinity

Matrices

Two milliliters of the purified monoclonal antibody solutions containing 16-22 mg protein as determined by the Coomassie blue dye binding assay (8) was incubated ’ Abbreviations used: PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline, pH 7.2; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; IgM, immunoglobulin M; HAF, human angiogenic factor; IgG, immunoglobulin G; BSA, bovine serum albumin. 0003-2697/89

314

Copyright All

rights

0 1989 by Academic

of reproduction

in any

form

$3.00

Press, Inc. reserved.

ELECTROELUTION

OF

with 2 ml of settled Affi-gel 10 (Bio-Rad, Munich, FRG) at 4°C according to the manufacturers’ instructions. BSA-antiserum (5 mg of protein) was conjugated on 600 ~1 of gel. After 5 h the gel matrices were centrifuged and washed twice with 10 ml of PBS. The coupling efficiency of the antibodies was 7580% as determined by measuring unconjugated protein in the supernatant. Excess reactive groups were blocked with 0.1 M ethanolamineHCl for 30 min at room temperature. The gel was equilibrated in PBS and washed with 20 ml of 0.1 M glycineHCl, pH 2.5, containing 0.1 M NaCl and subsequently 20 ml of PBS. Loading

of the Affinity

Gel

The 5F4-coupled gel was loaded with 8 ml tritium-labeled supernatant (45 mg protein/ml) of the human melanoma cell line A375/2. The concentrated serum-free supernatant was prepared as described (6). Radiolabeling was performed by adding 10 &i/ml [3H]leucine (120-190 Ci/mmol, Amersham, Braunschweig FRG) to leucine- and serum-free culture medium for 24 h. To the 91OD7-conjugated gel 8 ml tritium-labeled extract (19 mg protein/ml) of Con A-stimulated peripheral blood mononuclear cells was adsorbed. The cells were obtained and cultured as in (9). Extracts were prepared by vortexing the cells with 0.5% sodium desoxycholate, 0.5% Triton X-100, and 1 mM PMSF in 50 mM TrisHCl, pH 8.0, for 30 min at 4°C. Before the 910D7 gel was loaded the extract was centrifuged at 50,OOOg and subjected to chromatography on Con A-Sepharose (Pharmacia). Proteins eluted with 10% methylmannoside in PBS were desalted against PBS on PDlO-columns. The anti-BSA-conjugated affinity gel was incubated with 3ml of ‘*C-methylated BSA (NEN, Dreieich, FRG; 30,000 cpm/ml PBS) containing horse myoglobin (1 mg/ ml) as carrier. Antigen loading of each affinity matrix was carried out for 6 h at 4°C by gentle rotating. After incubation nonspecifically bound material was eliminated by extensive washing steps with the following buffers: (i) 100 ml PBS containing 1 M NaCl, (ii) 50 ml PBS, (iii) 50 ml PBS containing 0.1% octylglycoside, (iv) 10 mM Tris-borate, pH 8.2. Desorption

of Antigens

Acidic buffer elution. Antigens adsorbed to the antibody support were eluted with 10 gel vol of 0.1 M glycineHCl, pH 2.5, containing 0.1 M NaCl at 4°C (6-20 ml eluate). Fractions of 1 ml were collected and immediately neutralized with 15 ~1 1.5 M Tris-HCI, pH 7.0. Electroelution. Electrophoretic desorption was carried out in a membrane trap (Biotrap, Schleicher & Schiill, Dassel, FRG), originally devised for elution of proteins from polyacrylamide gels. The elution chamber

IMMOBILIZED

ANTIGENS

315

is constructed as an open channel, divided into two compartments by three membranes (BTl, BT2, Schleicher & Schull). The migration distance between the two compartments was 2 cm (for a diagram of the apparatus see Ref. (5)). After removal of nonspecifically bound proteins, the affinity gel was transferred into the elution device and placed into a horizontal electrophoresis chamber (28 X 9 X 8 cm) which was filled with 10 mM Trisborate, pH 8.2. In order to reduce the electric current the free buffer surface beside the elution device was blocked with a piece of plastic material. For elution of antigens 240 V was applied at 4°C corresponding to an electric field strength of 9 V/cm and an initial current of 12 mA. Like the pH of the elution buffer, all antigens tested were negatively charged. For monitoring of the elution process the membrane trap was emptied with a pipet at the anode site at different times (1,2,4,6,8,10,12,14 h). The eluate (300-500 ~1) was checked for radioactivity by liquid scintillation counting. Then the trap was refilled and the elution process continued. SDS-PAGE An aliquot of the electroeluates was concentrated in a vacuum centrifuge, diluted with sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 5% mercaptoethanol, 10% glycerol), and heated at 100°C for 3 min. SDS-PAGE was carried out according to Laemmli (10) using 10 and 15% separating gels. Proteins were visualized by fluorography (11). RESULTS

When trying to purify antigens by affinity chromatography on monoclonal antibody-coupled gel matrices only insufficient recovery yields of the antigen were obtained by traditionally used desorption techniques. After elution of the affinity column with low or high pH buffers (0.1 M glycine-HCI, pH 2.5; 0.1% acetic acid; 0.1 M sodium borate, pH 10.0) or chaotropic reagents (MgC&, NaSCN) it was observed that a three- to fivefold amount of radiolabeled antigen was still retained on the gel. This dilemma prompted us to examine electroelution for better recovery yields of the antigens. In order to assess the conditions required for electroelution the membrane trap applied in the study was emptied at different times. The time course of the elution process is shown in Fig. 1. The antigen is almost completely released from the BSA-antiserum-conjugated gel after 4 h. As presented, much longer elution times were necessary for dissociation of antigens from the two monoclonal antibody-linked matrices. The increase of radioactivity eluted from the 5F4- and 910D7affinity gel was nearly linear during the first 8 h. Elution of the antigens was finished after about 12 h.

316

SCHULZE-OSTHOFF

ET

AL. 5F4

>1 .Z .> ; 02 m D al 2 z z P

QlOD7

60 70. 60 50 -

W A ---A

40 30 -

67

-

40

-

BSA-antiserum moAb 910 D7 moAb 5F4

&

20 -

-

After unspecifically bound FIG. 1. Time course of electroelution: material was eliminated antigens adsorbed to the antibody-coupled gels were eluted electrophoretically. The eluate was checked for radioactivity after different times. Total eluted radioactivity was set as

FIG. 3. gated

SDS-PAGE affinity matrices.

28

of electroeluates of the 5F4- and QlOD7-conjuProteins were detected by fluorography.

100%.

The electrophoretic desorption technique of the antigens was compared with acidic buffer elution in respect to recovery yield as measured by the radioactivity of the eluates. The antibody-linked matrices were eluted with glycine-HCl, pH 2.5, or electrophoretically for 14 h. As shown in Fig. 2, the yield of antigen was higher after electroelution of all examined affinity gels. From the antiserum-coupled matrix the recovery of BSA was almost similar in both eluates; 9% more radioactivity was obtained in the electroeluate, corresponding to an overall recovery of about 75 %. Strong differences between the two elution techniques could be demonstrated for affinity chromatography with the monoclonal antibodies employed. The glycine-HCl eluate of the 5F4-column accounted for only 36% radioactivity of the electroeluate. A SDS-PAGE of the electroeluate (Fig. 3) showed two polypeptides of 67 and 40 kDa from which the 67-kDa protein represents the human angiogenic factor (6). Electrophoretic desorption of the 91OD7-conjugated gel led to a 4.2 times higher amount of antigen than acidic buffer elution. The elec-

5F4

FIG. 2.

Comparison of electroelution (glycine-HCl, pH 2.5).

910D7

anti-EEA

(14 h) and acidic

buffer

elution

trophoretic pattern of the electroeluate (Fig. 3) reveals that monoclonal antibody 910D7 reacted with two components with an apparent molecular mass of 28 and 34 kDa, which are identical to the LY-and P-chain of the HLA-DR antigen (7). During the electroelution process no warming up of the buffer was observed. Furthermore, the applied conditions were not detrimental for antigenic activity, as measured by the dot blot technique. DISCUSSION

Affinity chromatography utilizing specific antibodies has been proven very effective for protein purification. Many proteins have been isolated with great success, mostly by employing polyclonal antisera (for review see (12)). Probably due to the diversity of antigenic epitopes and affinity of the polyclonal antibodies no strong and uniform linkage is generated. Therefore, retained antigens can be easily dissociated from affinity columns. However, several approaches to purify antigens by affinity chromatography with monoclonal antibodies were found to be insufficient. All attempts to desorb the antigen from the antibody by conventional elution with chaotropic reagents high and low pH buffers or organic solvents failed. Also, modification of the elution temperature did not improve the recovery. Because of the strong affinity of the monoclonal antibodies antigens remained associated with the antibody-conjugated gel matrix. To overcome the difficulties we investigated the use of electroelution for desorption of proteins adsorbed on monoclonal antibody-linked matrices. The procedure was carried out in a commercial elution device with inert membranes, which retain macromolecules greater than 5 kDa. The study reveals that electrophoretic desorption afforded a significantly higher yield of antigens than

ELECTROELUTION

OF

acidic buffer elution. Differences in recovery were considerable for affinity chromatography with monoclonal antibodies, while nearly similar results were obtained in the case of the anti-BSA antiserum. Electroeluates were obtained in small volumes. Therefore, the need of concentration and desalting of the sample which is often required in traditional affinity chromatography is obviated. In addition no loss of antigenic or biological activity was observed during the elution. The mild conditions of electroelution and the omission of postelution procedures suggest a general application of electroelution for other kinds of affinity chromatography. The technique presented seems to be a suitable method, especially for purification of proteins which are sensitive to extremes of PI-I. REFERENCES 1. Morgan, Immunol.

M. R. A., Johnson, P. M., Methods 23,381-387.

and Dean,

P. D. G. (1978)

J.

IMMOBILIZED

317

ANTIGENS

2. Grenot, C., and Cuilleron, C.-Y. (1977) Biochem. Biophys. Res. C0mmu.n. 79,274-279. 3. Brown, P. J., Leyland, M. J., Keenan, J. P., and Dean, P. D. G. (1977) FEBS L&t. 83,256-259. 4. Morgan, M. R. A., Slater, N. A., and Dean, P. D. G. (1979) Anal. Biochem. 92,144-146. 5. Jacobs, E., and Clad, A. (1986) Anal. Biochem. 164,583-589. 6. Schulze Osthoff, K., Friihbeis, B., Overwien, B., Hilbig, B., and Sorg, C. (1987) Biochem. Biophys. Res. Commun. 146,945-952. 7. Sorg, C., Brocker, E.-B., Zwadlo, G., Redmann, K., Feige, U., Ax, W., and Feller, A. C. (1985) Transplantation 39,90-92. 8. Read, S. M., and Northcote, 64. 9. Burmeister, G., Tarcsay, 171,461-474. 10. Laemmli, 11. Bonner, 87.

U. K. (1970)

D. H. (1981) L., and Sorg,

Nature

W. M., and Laskey,

(Londonj R. A. (1974)

Anal.

Biochem.

C. (1986)

116,53-

Immurwbiology

227,680-685. Eur.

12. Kristiansen, T. (1978) in Affinity Chromatography tenhot, O., Koller, F., Kraft, D., and Scheiner, 206, Pergamon, New York.

J. Biochem.

46,83-

(Hoffman-OsO., Eds.), pp. 191-