- Email: [email protected]
Purification of peroxidase-conjugated antibody for enzyme immunoassay by affinity chromatography on concanavalin A
Journal of Immunological Methods, 25 (1979) 171--175
171
© Elsevier/North-Holland Biomedical Press
PURIFICATION OF PEROXIDASE-CONJUGATED ANTIBODY FOR ENZYME IMMUNOASSAY BY AFFINITY CHROMATOGRAPHY ON CONCANAVALIN A
J. ARENDS
Hormone Department, Statens Seruminstitut, Amager Boulevard 80, DK-2300 Copenhagen S, Denmark (Received 26 June 1978, accepted 25 July 1978)
Preparation of peroxidase-conjugated antibody contaminated with very small amounts of unconjugated antibody for highly sensitive enzyme immunoassays is described. The purification process is dependent upon the differential affinity of antibody and peroxidase-conjugated antibody for concanavalin A (Con A). Application of peroxidase and radioactive labelled IgG alone or as a mixture to Con A-Sepharose columns showed that 62--83% of IgG and 0% o f peroxidase could be eluted under starting conditions, while 70--83% of peroxidase and about 6% of IgG could be eluted with appropriate buffer. The contamination of the conjugate with unconjugated antibody is dependent on the ratio of conjugated to unconjugated antibody in the mixture to be purified. Using this procedure on IgG conjugated with peroxidase with a ratio of unconjugated to conjugated IgG of about 2 : 1 a preparation was obtained where the ratio was changed to 0.1 : 1.
INTRODUCTION
Horseradish peroxidase (HRP) conjugated to antibody has proved useful in many enzyme immunoassays. A frequently used method for preparing the enzyme-antibody conjugate is the two-step procedure of Avrameas and Ternynck (1971). The method is simple and the conjugate consists mainly of antibody and enzyme in the ratio 1 : 1, but the bulk of the reaction product is free antibody (Sternberger, 1974). Contamination with unconjugated antibody will seriously reduce the sensitivity of an enzyme immunoassay due to competition with the conjugated antibody for antigenic sites. Preparative separation of conjugated from unconjugated antibody is difficult because there are only small differences in molecular weight and charge. Gel filtration has been used to achieve separation of conjugated and unconjugated antibody. A comparison between different gel filtration media has been made by Boorsma and Streefkerk (1976). Although polyacrylamideagarose gel was found to be superior to dextran gel, the figures they present do not indicate that complete separation was achieved. One of the main problems is that the peaks of the two types of antibody are in close proximity. A fraction has to be chosen in which there is the highest possible yield
172 of conjugated antibody and the least a m o u n t of contamination with unconjugated antibody. The selection of this fraction is more difficult with increasing ratios of unconjugated to conjugated antibody and will preferentially select antibodies of higher molecular weight. IgG and HRP contain different amounts of carbohydrate (2--3% and 18%, respectively). The present communication exploits this difference by using concanavalin A (Con A) for the separation of peroxidase-antibody conjugate from unconjugated antibody. MATERIALS AND METHODS
IgG Two rabbit IgGs of different purity were used. One preparation containing only IgG was obtained from a single antiserum purified by affinity chromatography to specific IgG on solid-phase bound immunogen (mouse IgG) as immunosorbent. Another preparation was a Cohn fraction II (Koch-Light Laboratories) with a stated purity of 1>96% and further purified by gel filtration on a 2.5 cm X 93 cm Sephacryl-S 200 column (Pharmacia). The antibodies were trace labelled using 10 ~g IgG and 90 gCi 12sI and a modification of the m e t h o d of Greenwood et al. (1963). The specific activity was 0.5 ~Ci/mg after dilution with unlabelled IgG.
Horseradish peroxidase (HRP) HRP, Sigma, purity: type VI, RZ: 3.0, of two different batch numbers was used.
Conjugation procedure The two-step glutaraldehyde m e t h o d of Avrameas and Ternynck (1971) for conjugation of HRP and IgG (Cohn fraction II) was followed.
Gel filtration A column (1.6 cm X 94 cm) of Ultrogel polyacrylamide-agarose AcA-44 (LKB products) was prepared according to the manufacturer's instructions. The column was eluted with 0.02 M phosphate buffer, pH 7.4, with 0.15 M sodium chloride at a rate of 6 ml/h and 160 fractions of 1.3 ml were collected.
Con A affinity chromatography Con A-Sepharose (Pharmacia), columns 0.9 cm X 12.6 cm, was equilibrated with a 0.1 M acetate buffer, pH 6.0, with 1 M sodium chloride, 1 mM magnesium chloride, 1 mM calcium chloride and I mM maganese chloride (Con A buffer). The substance to be chromatographed was applied in a volume of 4 ml in Con A buffer. The standard procedure adopted for washing and elution consisted of a wash with Con A buffer until a total of 32--35 ml buffer was collected, the buffer was then changed to Con A buffer with 0.1 M of
173
t h e c a r b o h y d r a t e m e t h y l - a - D - g l u c o p y r a n o s i d e . W h e n t h e c o l u m n was l o a d e d w i t h H R P o r c o n j u g a t e a 3 m l f r a c t i o n was c o l l e c t e d as s o o n as t h e first t r a c e o f b r o w n c o l o u r e d H R P was e l u t e d .
HRP activity Peroxidase substrate used for the enzymatic reaction was 3 . 5 m M 5-amino-2-hydroxybenzoic acid and 2 t~M hydrogen peroxide in 0.02 M phosphate buffer, pH 6.0. The sample volume was 0.1 ml to which was added 1.5 ml peroxidase substrate. The adsorbance was read at 450 nm after 1.5 h at room temperature. RES ULTS
Affinity chromatography on a 8 ml Con A-Sepharose column o f a mixture o f 9.75 mg 12sI.trace labelled IgG and 9.75 mg HRP showed (Fig. 1) that the main part of the trace labelled IgG was not bound on Con A. After elution of the initial peak containing the bulk of the IgG, small and ever decreasing amounts of IgG were slowly washed out. In several experiments with 12sIlabelled IgG alone in amounts of 1--10 mg, the recovery was 62--85% in pool A. There was no difference in the behaviour whether the IgG used was the preparation purified by affinity chromatography or the Cohn fraction II.
~g/TUBE
7000
I !
6000 -
,I
II II ,I ,i 5000 -
~I I|
! t;000 -
3000
-
I I I
I000.
t
~ I
,b
,'s poo .
2'o
15
~o
[,
3'srua~NR.
Fig. 1. Con A-Sepharose chromatograph of a mixture of 9.75 mg trace labelled IgG (solid line) and 9.75 mg HRP (broken line). ~ Indicates change from Con A buffer to Con A buffer with 0.1 M methyl-~-D~lueopyranoside. Fractions o f 1.5 ml are collected.
174
When Con A buffer with 0.1 M methyl-a-D-glucopyranoside was subsequently used, small amounts of IgG were eluted (pool B). The a m o u n t of IgG in pool B usually comprised 4--6% of the a m o u n t in pool A. The behaviour of HRP in the mixture applied to the column (Fig. 1) was easily followed due to its brown colour. The 9.75 mg HRP in 4 ml Con A buffer occupied about 3 ml of the column, and on continued washing with Con A buffer the brown colour expanded slowly down the column. Changing the Con A buffer to Con A buffer with 0.1 M methyl-a-D-glucopyranoside resulted in a rapid desorption of HRP from the column which was visualized as a brown band with increasing colour intensity as it travelled down the column. Pool B on Fig. 1 was easily identified as a sharp band. In several experiments using HRP alone in amounts of 10 mg the recovery in pool B was 70--83%. Affinity chromatography of IgG (Cohn fraction II) conjugated with HRP, with a ratio of unconjugated to conjugated IgG of about 2 : 1 and with a total IgG c o n t e n t of 3190 ~g gave rise to a pool A containing 1260 pg IgG without HRP activity and a pool B with 720 pg IgG having HRP activity. Based on the experiments with IgG alone, it could be calculated that pool B was contaminated with about 63 pg unconjugated IgG, corresponding to a ratio of unconjugated to conjugated IgG of about 0.1 : 1. The conjugate purified on Con A-Sepharose was characterized by gel cpm
ng H R P
-70
2800-
2~00-
A
- 60
:',~
-50
II
2000-
-~0
1600-
,, II pl
1200-
-30
800-
-20
~00-
I0
2'0
~'o
6'0
ao
1oo
do
l;o
t~o rUeE N~.
Fig. 2. AcA-44 gel filtration of Con A-Sepharose purified IgG-HRP conjugate. The solid line represents IgG determined by its radioactivity. The broken line represents HRP activity. The shadowed area indicates the position of pure IgG from a previous calibration.
175 filtration on an AcA-44 column. Fig. 2 shows that IgG eluted as a welldefined peak with a marked left-hand shoulder and a smaller right-hand shoulder. The HRP activity co-eluted with the main peak of IgG. A peak having HRP activity only was eluted in tube 102. DISCUSSION The difference in carbohydrate content of IgG and HRP forms the basis for the different behaviour of the two substances on a Con A-Sepharose column. A high recovery of IgG without HRP contamination is obtained, while the HRP fraction is contaminated with only a small a m o u n t of IgG. The degree of IgG contamination in the HRP fraction is dependent on the a m o u n t of unconjugated IgG present in the mixture to be purified. This suggests the possibility of obtaining an even more purified preparation by repeating the procedure. Gel filtration of HRP-IgG conjugate on Ultrogel polyacrylamide-agarose AcA-44 showed a sharp peak of IgG in tube 72 which co-elutes with HRP activity. This identifies the position of the conjugate. The left-hand shoulder probably represents polymerized conjugate. The smaller right-hand shoulder w i t h o u t concomitant HRP activity is consistent with a contamination of the conjugate with unconjugated IgG of the order of 10%. The HRP activity which appears in tube 102 represents residual, free HRP which has not been removed by the a m m o n i u m sulphate precipitation step in the conjugation procedure. ACKNOWLEDGEMENTS The author thanks Drs. B. Rubin and B. Mansa, Statens Seruminstitut, for providing the affinity chromatography purified IgG and for useful discussions. REFERENCES Avrameas, S. and T. Ternynck, 1971, Immunochemistry 8, 1175. Boorsma, D.M. and I.G. Streefkerk, 1976, J. Histochem. Cytochem. 3,481. Greenwood, F.C., W.M. Hunter and I.S. Glover, 1963, Biochem. J. 89,114. Sternberger, L.A., 1974, Immunocytochemistry (Prentice-Hall, Englewood Cliffs, NJ).
171
© Elsevier/North-Holland Biomedical Press
PURIFICATION OF PEROXIDASE-CONJUGATED ANTIBODY FOR ENZYME IMMUNOASSAY BY AFFINITY CHROMATOGRAPHY ON CONCANAVALIN A
J. ARENDS
Hormone Department, Statens Seruminstitut, Amager Boulevard 80, DK-2300 Copenhagen S, Denmark (Received 26 June 1978, accepted 25 July 1978)
Preparation of peroxidase-conjugated antibody contaminated with very small amounts of unconjugated antibody for highly sensitive enzyme immunoassays is described. The purification process is dependent upon the differential affinity of antibody and peroxidase-conjugated antibody for concanavalin A (Con A). Application of peroxidase and radioactive labelled IgG alone or as a mixture to Con A-Sepharose columns showed that 62--83% of IgG and 0% o f peroxidase could be eluted under starting conditions, while 70--83% of peroxidase and about 6% of IgG could be eluted with appropriate buffer. The contamination of the conjugate with unconjugated antibody is dependent on the ratio of conjugated to unconjugated antibody in the mixture to be purified. Using this procedure on IgG conjugated with peroxidase with a ratio of unconjugated to conjugated IgG of about 2 : 1 a preparation was obtained where the ratio was changed to 0.1 : 1.
INTRODUCTION
Horseradish peroxidase (HRP) conjugated to antibody has proved useful in many enzyme immunoassays. A frequently used method for preparing the enzyme-antibody conjugate is the two-step procedure of Avrameas and Ternynck (1971). The method is simple and the conjugate consists mainly of antibody and enzyme in the ratio 1 : 1, but the bulk of the reaction product is free antibody (Sternberger, 1974). Contamination with unconjugated antibody will seriously reduce the sensitivity of an enzyme immunoassay due to competition with the conjugated antibody for antigenic sites. Preparative separation of conjugated from unconjugated antibody is difficult because there are only small differences in molecular weight and charge. Gel filtration has been used to achieve separation of conjugated and unconjugated antibody. A comparison between different gel filtration media has been made by Boorsma and Streefkerk (1976). Although polyacrylamideagarose gel was found to be superior to dextran gel, the figures they present do not indicate that complete separation was achieved. One of the main problems is that the peaks of the two types of antibody are in close proximity. A fraction has to be chosen in which there is the highest possible yield
172 of conjugated antibody and the least a m o u n t of contamination with unconjugated antibody. The selection of this fraction is more difficult with increasing ratios of unconjugated to conjugated antibody and will preferentially select antibodies of higher molecular weight. IgG and HRP contain different amounts of carbohydrate (2--3% and 18%, respectively). The present communication exploits this difference by using concanavalin A (Con A) for the separation of peroxidase-antibody conjugate from unconjugated antibody. MATERIALS AND METHODS
IgG Two rabbit IgGs of different purity were used. One preparation containing only IgG was obtained from a single antiserum purified by affinity chromatography to specific IgG on solid-phase bound immunogen (mouse IgG) as immunosorbent. Another preparation was a Cohn fraction II (Koch-Light Laboratories) with a stated purity of 1>96% and further purified by gel filtration on a 2.5 cm X 93 cm Sephacryl-S 200 column (Pharmacia). The antibodies were trace labelled using 10 ~g IgG and 90 gCi 12sI and a modification of the m e t h o d of Greenwood et al. (1963). The specific activity was 0.5 ~Ci/mg after dilution with unlabelled IgG.
Horseradish peroxidase (HRP) HRP, Sigma, purity: type VI, RZ: 3.0, of two different batch numbers was used.
Conjugation procedure The two-step glutaraldehyde m e t h o d of Avrameas and Ternynck (1971) for conjugation of HRP and IgG (Cohn fraction II) was followed.
Gel filtration A column (1.6 cm X 94 cm) of Ultrogel polyacrylamide-agarose AcA-44 (LKB products) was prepared according to the manufacturer's instructions. The column was eluted with 0.02 M phosphate buffer, pH 7.4, with 0.15 M sodium chloride at a rate of 6 ml/h and 160 fractions of 1.3 ml were collected.
Con A affinity chromatography Con A-Sepharose (Pharmacia), columns 0.9 cm X 12.6 cm, was equilibrated with a 0.1 M acetate buffer, pH 6.0, with 1 M sodium chloride, 1 mM magnesium chloride, 1 mM calcium chloride and I mM maganese chloride (Con A buffer). The substance to be chromatographed was applied in a volume of 4 ml in Con A buffer. The standard procedure adopted for washing and elution consisted of a wash with Con A buffer until a total of 32--35 ml buffer was collected, the buffer was then changed to Con A buffer with 0.1 M of
173
t h e c a r b o h y d r a t e m e t h y l - a - D - g l u c o p y r a n o s i d e . W h e n t h e c o l u m n was l o a d e d w i t h H R P o r c o n j u g a t e a 3 m l f r a c t i o n was c o l l e c t e d as s o o n as t h e first t r a c e o f b r o w n c o l o u r e d H R P was e l u t e d .
HRP activity Peroxidase substrate used for the enzymatic reaction was 3 . 5 m M 5-amino-2-hydroxybenzoic acid and 2 t~M hydrogen peroxide in 0.02 M phosphate buffer, pH 6.0. The sample volume was 0.1 ml to which was added 1.5 ml peroxidase substrate. The adsorbance was read at 450 nm after 1.5 h at room temperature. RES ULTS
Affinity chromatography on a 8 ml Con A-Sepharose column o f a mixture o f 9.75 mg 12sI.trace labelled IgG and 9.75 mg HRP showed (Fig. 1) that the main part of the trace labelled IgG was not bound on Con A. After elution of the initial peak containing the bulk of the IgG, small and ever decreasing amounts of IgG were slowly washed out. In several experiments with 12sIlabelled IgG alone in amounts of 1--10 mg, the recovery was 62--85% in pool A. There was no difference in the behaviour whether the IgG used was the preparation purified by affinity chromatography or the Cohn fraction II.
~g/TUBE
7000
I !
6000 -
,I
II II ,I ,i 5000 -
~I I|
! t;000 -
3000
-
I I I
I000.
t
~ I
,b
,'s poo .
2'o
15
~o
[,
3'srua~NR.
Fig. 1. Con A-Sepharose chromatograph of a mixture of 9.75 mg trace labelled IgG (solid line) and 9.75 mg HRP (broken line). ~ Indicates change from Con A buffer to Con A buffer with 0.1 M methyl-~-D~lueopyranoside. Fractions o f 1.5 ml are collected.
174
When Con A buffer with 0.1 M methyl-a-D-glucopyranoside was subsequently used, small amounts of IgG were eluted (pool B). The a m o u n t of IgG in pool B usually comprised 4--6% of the a m o u n t in pool A. The behaviour of HRP in the mixture applied to the column (Fig. 1) was easily followed due to its brown colour. The 9.75 mg HRP in 4 ml Con A buffer occupied about 3 ml of the column, and on continued washing with Con A buffer the brown colour expanded slowly down the column. Changing the Con A buffer to Con A buffer with 0.1 M methyl-a-D-glucopyranoside resulted in a rapid desorption of HRP from the column which was visualized as a brown band with increasing colour intensity as it travelled down the column. Pool B on Fig. 1 was easily identified as a sharp band. In several experiments using HRP alone in amounts of 10 mg the recovery in pool B was 70--83%. Affinity chromatography of IgG (Cohn fraction II) conjugated with HRP, with a ratio of unconjugated to conjugated IgG of about 2 : 1 and with a total IgG c o n t e n t of 3190 ~g gave rise to a pool A containing 1260 pg IgG without HRP activity and a pool B with 720 pg IgG having HRP activity. Based on the experiments with IgG alone, it could be calculated that pool B was contaminated with about 63 pg unconjugated IgG, corresponding to a ratio of unconjugated to conjugated IgG of about 0.1 : 1. The conjugate purified on Con A-Sepharose was characterized by gel cpm
ng H R P
-70
2800-
2~00-
A
- 60
:',~
-50
II
2000-
-~0
1600-
,, II pl
1200-
-30
800-
-20
~00-
I0
2'0
~'o
6'0
ao
1oo
do
l;o
t~o rUeE N~.
Fig. 2. AcA-44 gel filtration of Con A-Sepharose purified IgG-HRP conjugate. The solid line represents IgG determined by its radioactivity. The broken line represents HRP activity. The shadowed area indicates the position of pure IgG from a previous calibration.
175 filtration on an AcA-44 column. Fig. 2 shows that IgG eluted as a welldefined peak with a marked left-hand shoulder and a smaller right-hand shoulder. The HRP activity co-eluted with the main peak of IgG. A peak having HRP activity only was eluted in tube 102. DISCUSSION The difference in carbohydrate content of IgG and HRP forms the basis for the different behaviour of the two substances on a Con A-Sepharose column. A high recovery of IgG without HRP contamination is obtained, while the HRP fraction is contaminated with only a small a m o u n t of IgG. The degree of IgG contamination in the HRP fraction is dependent on the a m o u n t of unconjugated IgG present in the mixture to be purified. This suggests the possibility of obtaining an even more purified preparation by repeating the procedure. Gel filtration of HRP-IgG conjugate on Ultrogel polyacrylamide-agarose AcA-44 showed a sharp peak of IgG in tube 72 which co-elutes with HRP activity. This identifies the position of the conjugate. The left-hand shoulder probably represents polymerized conjugate. The smaller right-hand shoulder w i t h o u t concomitant HRP activity is consistent with a contamination of the conjugate with unconjugated IgG of the order of 10%. The HRP activity which appears in tube 102 represents residual, free HRP which has not been removed by the a m m o n i u m sulphate precipitation step in the conjugation procedure. ACKNOWLEDGEMENTS The author thanks Drs. B. Rubin and B. Mansa, Statens Seruminstitut, for providing the affinity chromatography purified IgG and for useful discussions. REFERENCES Avrameas, S. and T. Ternynck, 1971, Immunochemistry 8, 1175. Boorsma, D.M. and I.G. Streefkerk, 1976, J. Histochem. Cytochem. 3,481. Greenwood, F.C., W.M. Hunter and I.S. Glover, 1963, Biochem. J. 89,114. Sternberger, L.A., 1974, Immunocytochemistry (Prentice-Hall, Englewood Cliffs, NJ).