- Email: [email protected]
Binding of monomeric and aggregated immunoglobulin to enzymes
Journal of Immunological Methods, 116 (1989) 175-179
175
Elsevier JIM05019
Binding of monomeric and aggregated immunoglobulin to enzymes A source of artefact in antibody assays B.P. O ' H a r a 1, J o s e p h i n e Pyle 1, D. M c C a r t h y 2 a n d J.R. A r c h e r 1 1 ARC Bone and Joint Research Unit, London Hospital Medical College, London, U.K., and 2 School of Biological Sciences, Queen Mary College, London, U.K.
(Received21 August 1986, finn revised version received8 August 1988, accepted 22 August 1988)
Pepsinogen has previously been shown to bind non-specifically to immune complexes and aggregated immunoglobulins. We demonstrate here using a solid-phase immunoassay that immunoglobulins aggregated by heat or glutaraldehyde bind non-specifically to several different enzymes. Some of these, including pepsinogen (marketed as pepsin), hyaluronidase and trypsin, are used in the breakdown of tissues or biochemical preparations during the preparation of antigens. Contamination of impure antigens by enzyme is likely to lead to products which bind non-specifically to immune complexes. This can cause misidentification of complexes as antibodies. We recommend that all tests for specific antibody involving the use of antigens prepared by these or other enzymes should include a control with aggregated immunoglobulin substituted for the test serum. Key words: Enzyme; Autoantibody; ELISA; Immune complex
Introduction
Monomeric immunoglobulins have been shown to bind non-specifically to basic proteins (Poston, 1984) and to pepsinogen (Kirk et al., 1986). In the latter case immunoglobulin aggregates and immune complexes bind much more strongly. We have shown that this can lead to misidentification of immune complexes as autoantibodies if autoantigens which have been prepared by enzymic digestion are used as targets in a solid-phase immunoassay (Kirk et al., 1986).
To determine whether other enzymes, some of them likely to be used in the preparation of autoantigens, can cause similar confusion, we have used an ELISA to test the reactions of several enzymes with aggregated and monomeric immunoglobulin and compared their activity with pepsinogen.
Materials and methods Monomeric IgG
Correspondence to: J.R. Archer, ARC Bone and Joint Research Unit, London Hospital MedicalCollege, 25-29 Ashfield Street, London E1 2AD, U.K.
Human IgG (Sigma, Poole) was dissolved in borate-buffered saline (BBS) to a concentration of 30 m g / m l , clarified at 5000 r e v / m i n for 10 min and the supernatant centrifuged at 25 000 r e v / m i n for 17 h at 4 ° C (MSE 6 × 16 ml S.W.O rotor).
0022-1759/89/$03.50 © 1989 ElsevierSciencePublishers B.V. (Biomedical Division)
176 The topmost 10 ml was collected and stored a t - 2 0 ° C for use as defined monomer.
Aggregated lgG (1) Heat-aggregated IgG was prepared by the protocol of McCarthy et al. (1985a). Aggregated IgG was concentrated by precipitating twice with 5% ( w / v ) polyethylene glycol (PEG) and once with 3.5% PEG. The precipitate was resuspended in borate-buffered saline (0.1 M, p H 7.4) (BBS) and fractionated by sucrose density gradient centrifugation (McCarthy et al., 1981). The fraction containing aggregates of sedimentation coefficient 10-20 S was used. (2) Glutaraldehyde-aggregated IgG (which tends to consist of smaller aggregates) was made by the method of McCarthy et al. (1985b). Aggregates were separated from unaggregated IgG and fractionated as described above. The fraction containing aggregates of sedimentation coefficient 10-20 S was used.
the double reciprocal plot used here seems to linearise adequately the Ig binding data. These yield hyperbolic curves when plotted on an arithmetic basis (as, for example, in Fig. 1). Reciprocal OD values greater than ten represented background and were ignored.
Enzymes tested The following enzymes were obtained from Sigma, Poole, Dorset: rabbit aldolase (A1893), chymopapain (C9007), bovine chymotrypsin (C4129), Clostridium collagenase (C0130), pig elastase (E1250), pig fumarase (F6878), bovine hyaluronidase (H3506), egg white lysozyme (L6876), Clostridium neuraminidase (N2876), papain (P4762), pig pepsin (P7012), bovine trypsin (T8253). 'Pepsin' contains a substantial amount of pepsinogen, the source of its main aggregate binding activity. The terms pepsin and pepsinogen are used interchangeably in this paper.
Polyacrylamide gel electrophoresis Enzyme-linked immunosorbent assay (ELISA) The method of Alomari et al. (1983) was used. Falcon Microtest III flexible plates were coated with doubling dilutions of enzyme (initial 100 /~1 at 200/~g/ml in 0.1 M bicarbonate buffer pH 8.3), left for 24 h at 4 ° C, then washed twice in phosphate-buffered saline containing 0.05% Tween 20 (PBS-Tween). Bovine serum albumin (BSA) (190 #1 of 5 m g / m l in PBS-Tween) was added to block non-specific binding and incubated for 30 minutes at 37°C, then re-washed five times. IgG was diluted in PBS-Tween to 20/~g/ml. Aliquots (100 /~1) were added to each well and the plates were incubated at 37 ° C for 2 h. After five washes with PBS-Tween, 100 /d of rabbit anti-human IgG coupled to horseradish peroxidase were added and the plates incubated for a further 2 h at 37°C. Plates were washed five times with PBS-Tween and enzyme activity was measured using a hydrogen peroxide, o-phenylene diamine substrate. Colour intensity was read on a Titertek Multiskan spectrophotometer with a 492 nm filter. An optical density (OD) of 0.4 was taken as the end point and the ratio of pepsinogen concentration to enzyme concentration at that OD was calculated (Fig. 2). ExCept for those enzymes which exhibit a prozone (which give rise to two intersecting lines)
Electrophoresis in 10% polyacrylamide gel and blotting with aggregated immunoglobulins were carried out as described by Kirk et al. (1986).
Results
A number of enzymes bind aggregated immunoglobulins, but with most of them the reaction is weak. Prozones occurred with several enzymes. Three 'typical' titration curves, illustrating a positive, a negative and a prozone, are shown in Fig. 1. Fig. 2 shows results (plotted as reciprocals) from two enzymes which reacted differently with aggregate and monomer, and also gives an example of the calculation of relative binding. Table I shows approximate relative binding of monomeric and both types of aggregated immunoglobulin to the enzymes tested. The figures quoted are intended only as a rough guide to compare activities of enzymes in this particular system. Hyaluronidase and fumarase react equally well whether IgG is aggregated or monomeric. With one exception, the two aggregate results were similar whichever enzyme was used. However, trypsin bound much more strongly to glutaraldehyde than to heat aggregate.
177 0.7-
0.7"
A
B
0.8-
0.6-
0.5"
>, 0 . 5 u> c Q 0.4-
=o ¢c
0.4-
0
.o
o
0
_~ o. 0
0.3.
11.2. 0.1-
0.1.
0
I 5
0
!
110
lU5
I
2O
I
0.5
Enzyme C o n c e n t r a t i o n
(pg/well)
n
1
1.5
Enzyme Concentration (p,g/ml)
Fig. I.A: Binding of heat-aggregated immunoglobulin in ELISA at doubling enzyme dilutions (0-20/zg enzyme). B: As A (0-2 /sg enzyme). 0, pepsinogen (typical Ig aggregate binding); ©, aldolase (showing prozone); l , collagenase (negative).
To obtain an indication of the purity of the enzymes used, samples were reduced, alkylated, denatured and run on a 10% polyacrylamide gel. Staining for protein gave the result in Fig. 3. Most of the enzymes gave only one high molecular weight component. The proteolytic enzymes also
gave several low molecular weight bands which presumably resulted from autolysis. Hyaluronidase and collagenase had large numbers of high molecular weight bands and were clearly impure. The main pepsin bands, as expected, were of higher molecular weight than pepsin (34 500) and presumably represented pepsinogen. We cannot
10.0 -
.-_
7-5
-6
~
o
•"o
o
..-------
o~
o
o •
5.0
m
a.
2"5-
........-...- • - - - - - " - ' - T ' - -
111
205K
f mlt
116 K 97 K
i
°
m I
a
½B 5
10
66 15
1/Enzyme concentration
20
K
15
(pg/ml)
Fig. 2. Demonstration of the effect of aggregation of immunoglobulin on its binding to pepsin(ogen) and trypsin. © pepsinogen with untreated immunoglobulin; • pepsinogen with heat-aggregated immunoglobulin; zx trypsin with untreated immunoglobulin; • trypsin with giutaraldehyde-aggregated immunoglobulin. Relative binding is defined as the ratio of concentration of pepsinogen needed to give a final OD of 0.4 with heat-aggregated immunogiobulin to the enzyme concentration giving the same amount of colour. The ratio is chosen to be in this form so that higher ratios represent higher binding to immunoglobufins and vice versa. For instance the values indicated respectively at A and B are the reciprocal concentrations of trypsin and pepsin required to produce a final OD of 0.4 in the presence of heat-aggregated Ig. Actual concentrations are 5 /~g/ml and 0.09 /.Lg/ml for trypsin and pepsin respectively and the relative binding is therefore 0.09/5 = 0.02.
45K 1
qlm
D
Pe Tr
C T Hy
M CP Pa Co El
Fu AI
29 K
M
Fig. 3. Polyacrylamide gel electrophoresis of enzymes. A1, aldolase; Co, collagenase; CP, chymopapain; CT, chymotrypsin; El, elastase; Fu, fumarase; Hy, hyaluronidase; M, markers; Pa, papain; Pe, pepsin; Tr, trypsin. Protein bands stained with Kenacid blue.
178 TABLE I I M M U N O G L O B U L I N - B I N D I N G ACTIVITY OF ENZYMES IN SOLID-PHASE IMMUNOASSAYS Max. O D
Aldolase Carbonic anhydrase Chymopapain Chymotrypsin Collagenase Fumarase Hyaluronidase Papain Pepsinogen Trypsin
Relative binding
G
H
M
G
H
M
0.72 0.26 0.41 0.34 0.10 0.75 0.68 0.09 0.45 0.45
0.62 0.19 0.23 0.20 0.12 0.60 0.80 > 0.16 0.60 > 0.27
0.38 0.33 0.34 0.14 0.19 0.50 0.60 0.06 0.23 0.10
1 -
0.7 0.7 3 1 0.001 a
0.01 -
1 2 1 0.02
2 3 _
Key: G,
glutaraldehyde aggregate; H, heat aggregate; M, monomer. a Estimate by extrapolation.
explain why chymopapain gave a doublet at an approximately MW of 14 000. The expected values would be MW 34500 and 36 500. After blotting and development with glutaraldehyde-aggregated immunoglobulin only two enzymes, pepsin and aldolase were positive. We assume that binding activity in the others was too weak or, in the case of hyaluronidase, dependent on configuration.
Discussion Immunoglobulin aggregates bind non-specifically to a number of enzymes. We have no explanation for the phenomenon, but we have previously published data consistent with a cause related to an alteration in immunoglobulin configuration rather than a change in size (Alomari et al., 1983). There is no obvious connection between enzyme specificity or p I and strength of reaction. Glutaraldehyde and the active site of trypsin react with the same basic amino acids, suggesting that in this case binding takes place near the active site of the enzyme or that the enzyme still has proteolytic activity which can release some of the bound immunoglobulin. We have previously shown that what we formerly called aggregate binding factor (Alomari et al., 1983) and now know to be pepsinogen (Kirk et al., 1986) binds both to immune complexes in serum from patients with arthritis and to com-
plexes prepared in vitro using BSA and anti-BSA antibodies. It seems probable that this occurs by the same mechanism as the aggregate binding reaction, and that immune complexes are liable to cause anomalous high reactions in certain tests such as the ELISA. Hyaluronidase, pepsin and chymopapain are all enzymes likely to be used in the disruption of tissues for the preparation of pure antigens. Extreme care should be taken to make sure that they are removed before antibody reactions are measured. We recommend that all pure antigens prepared for the assay of antibodies should be checked for their non-specific tendency to bind immunoglobulin aggregates. This should help to avoid confusion of specific antibodies with immune complexes.
References Alomari, W.R.S., Archer, J.R., Brocklehurst, R. and Currey, H.L.F. (1983) Binding of immunoglobulins and immune complexes to cartilage derived extracts. Clin. Exp. Immunol. 59, 716. Kirk, A.P., O'Hara, B.P., Mageed, R.A.K., McMahon, M.S., McCarthy, D., Menashi, S., Archer, J.R. and Currey, H.L.F. (1986) Pepsinogen - an immunoglobulin binding artefact in 'collagen' preparations. Clin. Exp. Immunol. 65, 671. McCarthy, D., Goddard, D.H., Holborow, E.J., Horsfall, A.C., Mumford, P.A. and Maini, R.N. (1981) The effect of lgG aggregate size and concentration on reactivity in immune complex assays. J. Immunol. Methods 47, 349.
179 McCarthy, D.A., Field, M., Mumford, P., Moore, S.R., Holborow, E.J. and Maini, R.N. (1985a) Soluble IgG aggregates produced by heating remain stable on freeze-drying. J. Immunol. Methods 82, 155. McCarthy, D.A., Field, M., Mumford, P., Pell, B.K., Holborow,
E.J. and Maim, R.N. (1985b) The production of small IgG aggregates by glutaraldehyde cross-linking. J. Immunol. Methods 82, 349. Poston, R.N. (1984) Basic proteins bind immunoglobulin G: a mechanism for demyelinating disease? Lancet i, 1268.
175
Elsevier JIM05019
Binding of monomeric and aggregated immunoglobulin to enzymes A source of artefact in antibody assays B.P. O ' H a r a 1, J o s e p h i n e Pyle 1, D. M c C a r t h y 2 a n d J.R. A r c h e r 1 1 ARC Bone and Joint Research Unit, London Hospital Medical College, London, U.K., and 2 School of Biological Sciences, Queen Mary College, London, U.K.
(Received21 August 1986, finn revised version received8 August 1988, accepted 22 August 1988)
Pepsinogen has previously been shown to bind non-specifically to immune complexes and aggregated immunoglobulins. We demonstrate here using a solid-phase immunoassay that immunoglobulins aggregated by heat or glutaraldehyde bind non-specifically to several different enzymes. Some of these, including pepsinogen (marketed as pepsin), hyaluronidase and trypsin, are used in the breakdown of tissues or biochemical preparations during the preparation of antigens. Contamination of impure antigens by enzyme is likely to lead to products which bind non-specifically to immune complexes. This can cause misidentification of complexes as antibodies. We recommend that all tests for specific antibody involving the use of antigens prepared by these or other enzymes should include a control with aggregated immunoglobulin substituted for the test serum. Key words: Enzyme; Autoantibody; ELISA; Immune complex
Introduction
Monomeric immunoglobulins have been shown to bind non-specifically to basic proteins (Poston, 1984) and to pepsinogen (Kirk et al., 1986). In the latter case immunoglobulin aggregates and immune complexes bind much more strongly. We have shown that this can lead to misidentification of immune complexes as autoantibodies if autoantigens which have been prepared by enzymic digestion are used as targets in a solid-phase immunoassay (Kirk et al., 1986).
To determine whether other enzymes, some of them likely to be used in the preparation of autoantigens, can cause similar confusion, we have used an ELISA to test the reactions of several enzymes with aggregated and monomeric immunoglobulin and compared their activity with pepsinogen.
Materials and methods Monomeric IgG
Correspondence to: J.R. Archer, ARC Bone and Joint Research Unit, London Hospital MedicalCollege, 25-29 Ashfield Street, London E1 2AD, U.K.
Human IgG (Sigma, Poole) was dissolved in borate-buffered saline (BBS) to a concentration of 30 m g / m l , clarified at 5000 r e v / m i n for 10 min and the supernatant centrifuged at 25 000 r e v / m i n for 17 h at 4 ° C (MSE 6 × 16 ml S.W.O rotor).
0022-1759/89/$03.50 © 1989 ElsevierSciencePublishers B.V. (Biomedical Division)
176 The topmost 10 ml was collected and stored a t - 2 0 ° C for use as defined monomer.
Aggregated lgG (1) Heat-aggregated IgG was prepared by the protocol of McCarthy et al. (1985a). Aggregated IgG was concentrated by precipitating twice with 5% ( w / v ) polyethylene glycol (PEG) and once with 3.5% PEG. The precipitate was resuspended in borate-buffered saline (0.1 M, p H 7.4) (BBS) and fractionated by sucrose density gradient centrifugation (McCarthy et al., 1981). The fraction containing aggregates of sedimentation coefficient 10-20 S was used. (2) Glutaraldehyde-aggregated IgG (which tends to consist of smaller aggregates) was made by the method of McCarthy et al. (1985b). Aggregates were separated from unaggregated IgG and fractionated as described above. The fraction containing aggregates of sedimentation coefficient 10-20 S was used.
the double reciprocal plot used here seems to linearise adequately the Ig binding data. These yield hyperbolic curves when plotted on an arithmetic basis (as, for example, in Fig. 1). Reciprocal OD values greater than ten represented background and were ignored.
Enzymes tested The following enzymes were obtained from Sigma, Poole, Dorset: rabbit aldolase (A1893), chymopapain (C9007), bovine chymotrypsin (C4129), Clostridium collagenase (C0130), pig elastase (E1250), pig fumarase (F6878), bovine hyaluronidase (H3506), egg white lysozyme (L6876), Clostridium neuraminidase (N2876), papain (P4762), pig pepsin (P7012), bovine trypsin (T8253). 'Pepsin' contains a substantial amount of pepsinogen, the source of its main aggregate binding activity. The terms pepsin and pepsinogen are used interchangeably in this paper.
Polyacrylamide gel electrophoresis Enzyme-linked immunosorbent assay (ELISA) The method of Alomari et al. (1983) was used. Falcon Microtest III flexible plates were coated with doubling dilutions of enzyme (initial 100 /~1 at 200/~g/ml in 0.1 M bicarbonate buffer pH 8.3), left for 24 h at 4 ° C, then washed twice in phosphate-buffered saline containing 0.05% Tween 20 (PBS-Tween). Bovine serum albumin (BSA) (190 #1 of 5 m g / m l in PBS-Tween) was added to block non-specific binding and incubated for 30 minutes at 37°C, then re-washed five times. IgG was diluted in PBS-Tween to 20/~g/ml. Aliquots (100 /~1) were added to each well and the plates were incubated at 37 ° C for 2 h. After five washes with PBS-Tween, 100 /d of rabbit anti-human IgG coupled to horseradish peroxidase were added and the plates incubated for a further 2 h at 37°C. Plates were washed five times with PBS-Tween and enzyme activity was measured using a hydrogen peroxide, o-phenylene diamine substrate. Colour intensity was read on a Titertek Multiskan spectrophotometer with a 492 nm filter. An optical density (OD) of 0.4 was taken as the end point and the ratio of pepsinogen concentration to enzyme concentration at that OD was calculated (Fig. 2). ExCept for those enzymes which exhibit a prozone (which give rise to two intersecting lines)
Electrophoresis in 10% polyacrylamide gel and blotting with aggregated immunoglobulins were carried out as described by Kirk et al. (1986).
Results
A number of enzymes bind aggregated immunoglobulins, but with most of them the reaction is weak. Prozones occurred with several enzymes. Three 'typical' titration curves, illustrating a positive, a negative and a prozone, are shown in Fig. 1. Fig. 2 shows results (plotted as reciprocals) from two enzymes which reacted differently with aggregate and monomer, and also gives an example of the calculation of relative binding. Table I shows approximate relative binding of monomeric and both types of aggregated immunoglobulin to the enzymes tested. The figures quoted are intended only as a rough guide to compare activities of enzymes in this particular system. Hyaluronidase and fumarase react equally well whether IgG is aggregated or monomeric. With one exception, the two aggregate results were similar whichever enzyme was used. However, trypsin bound much more strongly to glutaraldehyde than to heat aggregate.
177 0.7-
0.7"
A
B
0.8-
0.6-
0.5"
>, 0 . 5 u> c Q 0.4-
=o ¢c
0.4-
0
.o
o
0
_~ o. 0
0.3.
11.2. 0.1-
0.1.
0
I 5
0
!
110
lU5
I
2O
I
0.5
Enzyme C o n c e n t r a t i o n
(pg/well)
n
1
1.5
Enzyme Concentration (p,g/ml)
Fig. I.A: Binding of heat-aggregated immunoglobulin in ELISA at doubling enzyme dilutions (0-20/zg enzyme). B: As A (0-2 /sg enzyme). 0, pepsinogen (typical Ig aggregate binding); ©, aldolase (showing prozone); l , collagenase (negative).
To obtain an indication of the purity of the enzymes used, samples were reduced, alkylated, denatured and run on a 10% polyacrylamide gel. Staining for protein gave the result in Fig. 3. Most of the enzymes gave only one high molecular weight component. The proteolytic enzymes also
gave several low molecular weight bands which presumably resulted from autolysis. Hyaluronidase and collagenase had large numbers of high molecular weight bands and were clearly impure. The main pepsin bands, as expected, were of higher molecular weight than pepsin (34 500) and presumably represented pepsinogen. We cannot
10.0 -
.-_
7-5
-6
~
o
•"o
o
..-------
o~
o
o •
5.0
m
a.
2"5-
........-...- • - - - - - " - ' - T ' - -
111
205K
f mlt
116 K 97 K
i
°
m I
a
½B 5
10
66 15
1/Enzyme concentration
20
K
15
(pg/ml)
Fig. 2. Demonstration of the effect of aggregation of immunoglobulin on its binding to pepsin(ogen) and trypsin. © pepsinogen with untreated immunoglobulin; • pepsinogen with heat-aggregated immunoglobulin; zx trypsin with untreated immunoglobulin; • trypsin with giutaraldehyde-aggregated immunoglobulin. Relative binding is defined as the ratio of concentration of pepsinogen needed to give a final OD of 0.4 with heat-aggregated immunogiobulin to the enzyme concentration giving the same amount of colour. The ratio is chosen to be in this form so that higher ratios represent higher binding to immunoglobufins and vice versa. For instance the values indicated respectively at A and B are the reciprocal concentrations of trypsin and pepsin required to produce a final OD of 0.4 in the presence of heat-aggregated Ig. Actual concentrations are 5 /~g/ml and 0.09 /.Lg/ml for trypsin and pepsin respectively and the relative binding is therefore 0.09/5 = 0.02.
45K 1
qlm
D
Pe Tr
C T Hy
M CP Pa Co El
Fu AI
29 K
M
Fig. 3. Polyacrylamide gel electrophoresis of enzymes. A1, aldolase; Co, collagenase; CP, chymopapain; CT, chymotrypsin; El, elastase; Fu, fumarase; Hy, hyaluronidase; M, markers; Pa, papain; Pe, pepsin; Tr, trypsin. Protein bands stained with Kenacid blue.
178 TABLE I I M M U N O G L O B U L I N - B I N D I N G ACTIVITY OF ENZYMES IN SOLID-PHASE IMMUNOASSAYS Max. O D
Aldolase Carbonic anhydrase Chymopapain Chymotrypsin Collagenase Fumarase Hyaluronidase Papain Pepsinogen Trypsin
Relative binding
G
H
M
G
H
M
0.72 0.26 0.41 0.34 0.10 0.75 0.68 0.09 0.45 0.45
0.62 0.19 0.23 0.20 0.12 0.60 0.80 > 0.16 0.60 > 0.27
0.38 0.33 0.34 0.14 0.19 0.50 0.60 0.06 0.23 0.10
1 -
0.7 0.7 3 1 0.001 a
0.01 -
1 2 1 0.02
2 3 _
Key: G,
glutaraldehyde aggregate; H, heat aggregate; M, monomer. a Estimate by extrapolation.
explain why chymopapain gave a doublet at an approximately MW of 14 000. The expected values would be MW 34500 and 36 500. After blotting and development with glutaraldehyde-aggregated immunoglobulin only two enzymes, pepsin and aldolase were positive. We assume that binding activity in the others was too weak or, in the case of hyaluronidase, dependent on configuration.
Discussion Immunoglobulin aggregates bind non-specifically to a number of enzymes. We have no explanation for the phenomenon, but we have previously published data consistent with a cause related to an alteration in immunoglobulin configuration rather than a change in size (Alomari et al., 1983). There is no obvious connection between enzyme specificity or p I and strength of reaction. Glutaraldehyde and the active site of trypsin react with the same basic amino acids, suggesting that in this case binding takes place near the active site of the enzyme or that the enzyme still has proteolytic activity which can release some of the bound immunoglobulin. We have previously shown that what we formerly called aggregate binding factor (Alomari et al., 1983) and now know to be pepsinogen (Kirk et al., 1986) binds both to immune complexes in serum from patients with arthritis and to com-
plexes prepared in vitro using BSA and anti-BSA antibodies. It seems probable that this occurs by the same mechanism as the aggregate binding reaction, and that immune complexes are liable to cause anomalous high reactions in certain tests such as the ELISA. Hyaluronidase, pepsin and chymopapain are all enzymes likely to be used in the disruption of tissues for the preparation of pure antigens. Extreme care should be taken to make sure that they are removed before antibody reactions are measured. We recommend that all pure antigens prepared for the assay of antibodies should be checked for their non-specific tendency to bind immunoglobulin aggregates. This should help to avoid confusion of specific antibodies with immune complexes.
References Alomari, W.R.S., Archer, J.R., Brocklehurst, R. and Currey, H.L.F. (1983) Binding of immunoglobulins and immune complexes to cartilage derived extracts. Clin. Exp. Immunol. 59, 716. Kirk, A.P., O'Hara, B.P., Mageed, R.A.K., McMahon, M.S., McCarthy, D., Menashi, S., Archer, J.R. and Currey, H.L.F. (1986) Pepsinogen - an immunoglobulin binding artefact in 'collagen' preparations. Clin. Exp. Immunol. 65, 671. McCarthy, D., Goddard, D.H., Holborow, E.J., Horsfall, A.C., Mumford, P.A. and Maini, R.N. (1981) The effect of lgG aggregate size and concentration on reactivity in immune complex assays. J. Immunol. Methods 47, 349.
179 McCarthy, D.A., Field, M., Mumford, P., Moore, S.R., Holborow, E.J. and Maini, R.N. (1985a) Soluble IgG aggregates produced by heating remain stable on freeze-drying. J. Immunol. Methods 82, 155. McCarthy, D.A., Field, M., Mumford, P., Pell, B.K., Holborow,
E.J. and Maim, R.N. (1985b) The production of small IgG aggregates by glutaraldehyde cross-linking. J. Immunol. Methods 82, 349. Poston, R.N. (1984) Basic proteins bind immunoglobulin G: a mechanism for demyelinating disease? Lancet i, 1268.