Cytomegalovirus Antigens in IgA Nephropathy: Fact or Artefacts?

Dr. F. Mac-Moune Lai, Department of Morbid Anatomy, Prince of Wales Hospital, Room 34055, Shatin, NT, (Hong Kong) Dear Sir, Glomerular deposition of cytomegalovirus (CMV) antigens and a pathogenetic role of CMV in IgA nephropathy remain to be established. While Gregory et al. [1] reported positive mesangial staining using polyclonal anti-CMV antibodies (Polysciences and Lee Biomolecu-lar) in 31 renal biopsies of patients with IgA nephropathy; Waldo et al. [2] demonstrated that the same antibodies failed to stain after absorption with uninfected fibro-blasts and the antisera reacted with proteins from fibro-blast by western blot analysis. Furthermore, Dueymes et al. [3] showed inconsistent immunofluorescence stain-ings when 3 different anti-CMV monoclonal antibodies (58/2 and 77/8, Institut Pasteur; E13, Biosoft) and polyclonal antisera were used. These observations question the heterogeneity and specificity of anti-CMV antibodies used and undermine the notion of a putative role of CMV in IgA nephropathy. We have also attempted to detect glomerular CMV antigens in biopsy of patients with IgA nephropathy and other mesangial proliferative glomerulonephritis (table 1). Two anti-CMV monoclonal antibodies, one directed against nuclear early antigen, at 1:15 dilution (Du Pont Specialty Diagnostics, Wilmington, Del.), and the other directed against a late nuclear antigen, undiluted (Whit-taker Bioproducts, Walkersville, Md.), were used for indirect immunofluorescence studies. Positive stains were observed in several tissue sections of a patient who died of disseminated CMV infection. None of the specimen from patients with IgA nephropathy and with other mesangial proliferative nephropathies demonstrated glomerular staining for CMV antigens, with either monoclonal antibody. Interpretation of immunofluorescence studies are highly dependent on the specificity of the antibody used [4]. The conflicting findings and disparity of results observed in glomerular staining of CMV antigens clearly reflect the lack of specificity of the antibodies used. We encountered similar problems with use of antibodies against hepatitis B virus antigens [5], but were able to clarify the respective role of hepatitis core, e and surface