Statistics from Altmetric.com
We feel we should comment on a recent article by Iwanaet al 1 on the clinical value of measuring circulating IgA-αl-antitrypsin (IgA-AT) complex concentrations in patients with rheumatoid arthritis (RA) using a prototype ELISA kit. We are concerned about the specificity of the monoclonal antibody used as the capture reagent on their ELISA plates. The authors say that the antibody recognises specific epitopes on the IgA-AT complex. However no direct proof of this is provided here or in previous reports where this particular antibody has been used.2-6 Recently, in response to another study using this assay we provided data to show that the antibody recognises the complement regulatory protein, factor H.7 We have shown that replacing the ‘complex specific’ antibody with other monoclonal antibodies to factor H (OX23 and OX24) in the ELISA essentially makes no difference to measurement of ‘complex’ values. We have also shown that the ‘lgA-AT’ antibody recognises a different epitope on factor H to that recognised by OX23 and OX24, and feel that it would be surprising if monoclonals directed against three different regions on factor H all showed cross reactivity with IgA-AT.
Our studies pose the question as to what is actually being measured in the ‘IgA-AT’ ELISA. The specificity of the coating antibody for factor H might suggest that some form of factor H-lgA complex is being measured. However, a crucial factor in the use of this particular ELISA is the lack of any blocking step (for example, with bovine serum albumin or gelatin) between antibody coating and addition of standards and samples. This may allow IgA (and other serum proteins) to bind non-specifically to free binding sites on the plate. We have run exactly the same ELISA using a blocking step with 1% bovine serum albumin before addition of samples and found that this obliterates most of the binding of the standards and the samples. This suggests that the monoclonal antibody on the plate is largely irrelevant and that most of the IgA detected by the secondary antibody is bound to unblocked sites. Clearly the IgA would have to compete with other serum proteins for these binding sites. Thus, the assay seems to be measuring the ratio of IgA and IgA associated proteins to all other serum proteins. If this is the case then their results are not that suprising as a number of studies have shown IgA values to be increased in RA patients.
We have recently developed a new assay for measuring IgA-AT complexes based on a sandwich ELISA with a monoclonal antibody to α1-antitrypsin as the capture antibody and a secondary antihuman IgA peroxidase antibody for detection of the complexes. Using this assay we have shown that IgA-AT complexes are significantly higher in the serum of RA patients than in those with reactive arthritis.8 In addition we have shown that serum concentrations are higher than synovial fluid concentrations in both RA and ReA, suggesting that such complexes are produced systemically rather than locally within the joint. We were unable to find any association with the concentrations of acute phase reactants and no association with joint inflammation in itself.
IgA-AT complex values may be useful for monitoring the effectiveness of second line drugs because values have been shown to fall during treatment with D-penicilamine, gold, and sulphasalazine.9 10 However these studies used a two dimensional immunoelectrophoresis method unsuitable for screening large numbers of specimens. An ELISA method is clearly more desirable but one needs to be confident that it is only IgA-AT complex values that are being measured. We are doubtful whether this is the case for the assay used by Iwana et al. It would be interesting to use our assay to measure IgA-AT complex values in their RA and osteoarthritis specimens to see if similar correlations were found with the clinical findings.
We appreciate the comments of Dr D L Mattey and colleagues regarding our article.1-1
As the prototype kit used in our study for detecting IgA-α1-antitrypsin (IgA-AT) complex was a generous gift from Professor D R Stanworth, we were not informed about the detailed specificity of the monoclonal antibodies reacting with the specific epitopes on the IgA-AT complex. Therefore, Dr Stanworth is in a better position than ourselves to comment on this issue.
Comments by Professor Stanworth
I should welcome an opportunity to reply to the comments of Mattey and associates as the assay in question was developed in my laboratory in Birmingham.
Since making the assay available to Professor Iwana in the National Medical Centre of Japan, we have been made aware by Dr Mattey that the anticomplex antibody used within the assay may cross react with complement factor H. This, however, does not negate the findings reported by Iwana and his associates as they used a secondary anti-IgA antibody within the assay. This antibody is specific for IgA, and IgA containing complexes, and does not cross react with factor H. Indeed this assay format did not detect factor H. Moreover, the assay was checked to ensure that free IgA was not detected; thus precluding the possibility of non-specific binding to the plate as suggested by Dr Mattey.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.