Patients with primary biliary cirrhosis make anti-viral and anti-mitochondrial antibodies to mouse mammary tumor virus

Abstract

We wish to comment on the recent publication by Selmi et al.1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar They state they have failed to demonstrate any evidence for our hypothesis that a retrovirus is associated with PBC. However, while they have in fact confirmed many of our observations, differences in findings relate to differences in approach and techniques used. Selmi choose not to study purified human biliary epithelial cells (BEC) or perihepatic lymph nodes or our dynamic co-culture system, where we found robust evidence of the human betaretrovirus (HBRV).2Sadamoto T. Joplin R. Keogh A. Mason A. Carman W. Neuberger J. Expression of pyruvate dehydrogenase complex PDC-E2 on biliary epithelial cells induced by lymph nodes from primary biliary cirrhosis.Lancet. 1998; 352: 1595-1596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar, 3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar, 4Xu L. Sakalian M. Shen Z. Loss G. Neuberger J. Mason A. Cloning the human betaretrovirus proviral genome from patients with primary biliary cirrhosis.Hepatology. 2004; 39: 151-156Crossref PubMed Scopus (88) Google Scholar However, they confirmed our findings that viral proteins cannot be detected in the liver by immunohistochemistry (IHC) and that proviral HBRV DNA is not detected by single round of PCR in hepatic DNA.1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar, 3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar Selmi is incorrect in stating our virologic studies have only been conducted in one center. These observations were made in 3 separate institutions and the PCR, IHC and in vitro co-culture studies were all recapitulated and confirmed in 2 independent centers to generate a large body of in vivo and in vitro data to link the human betaretrovirus (HBRV) with PBC. These studies jointly condensed into 4 publications describe a co-culture model showing that pure viral isolates induce the PBC phenotype in normal human BEC and that combination antiviral treatment, but not monotherapy, coincides with significant biochemical and histological improvement in patients with PBC.2Sadamoto T. Joplin R. Keogh A. Mason A. Carman W. Neuberger J. Expression of pyruvate dehydrogenase complex PDC-E2 on biliary epithelial cells induced by lymph nodes from primary biliary cirrhosis.Lancet. 1998; 352: 1595-1596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar, 3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar, 4Xu L. Sakalian M. Shen Z. Loss G. Neuberger J. Mason A. Cloning the human betaretrovirus proviral genome from patients with primary biliary cirrhosis.Hepatology. 2004; 39: 151-156Crossref PubMed Scopus (88) Google Scholar, 5Mason A.L. Farr G.H. Xu L. Hubscher S.G. Neuberger J.M. Pilot studies of single and combination anti-retroviral therapy in patients with primary biliary cirrhosis.Am J Gastroenterol. 2004; 99: 1-8Google Scholar Selmi is wrong in stating we used serum for IHC studies when it is expressly stated monoclonal antibodies were used in concentrations originally described by Selmi’s group and by Golovkina’s group (cited in reference 2Sadamoto T. Joplin R. Keogh A. Mason A. Carman W. Neuberger J. Expression of pyruvate dehydrogenase complex PDC-E2 on biliary epithelial cells induced by lymph nodes from primary biliary cirrhosis.Lancet. 1998; 352: 1595-1596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). Using double IHC and confocal microscopy of perihepatic lymph nodes from patients with PBC, we found cellular co-localization of viral protein and aberrant distribution of pyruvate dehydrogenase complex E2 (PDC-E2).3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar This is an important finding as it directly links viral infection with the phenotypic marker of PBC. Selmi attempts to discredit this data by suggesting that the concentration of serum used for our study was “greater than was used for the identification of the viral etiology of Kaposi’s sarcoma6Chang Y. Cesarman E. Pessin M.S. Lee F. Culpepper J. Knowles D.M. Moore P.S. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma.Science. 1994; 266: 1865-1869Crossref PubMed Scopus (4900) Google Scholar and prone to provide artifacts.” However, in none of our studies was serum used and no serum or IHC were used for the study Selmi cited6Chang Y. Cesarman E. Pessin M.S. Lee F. Culpepper J. Knowles D.M. Moore P.S. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma.Science. 1994; 266: 1865-1869Crossref PubMed Scopus (4900) Google Scholar; the Kaposi’s sarcoma virus was identified using representational difference analysis.6Chang Y. Cesarman E. Pessin M.S. Lee F. Culpepper J. Knowles D.M. Moore P.S. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma.Science. 1994; 266: 1865-1869Crossref PubMed Scopus (4900) Google Scholar Selmi’s comment that previous electron micrograph (EM) studies did not detect viruses fails to take into consideration the point that we focused our search to the target cells of the disease.1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar BEC only represent 2%–4% of cells in the liver and virus was only identified with a frequency of ∼1 particle per 100 cells. Accordingly, it is unlikely that an adequate representative sample of BEC would have been available for study in the small fragments of liver previously analyzed by EM. Consistent with our EM studies, liver from patients with PBC tested universally negative for viral proteins and only 29% had evidence of HBRV RNA by RT-PCR.3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar Using our nested PCR methodology, we were only able to detect HBRV proviral DNA in 2 of 12 PBC patients, which is comparable to the 0 of 6 PBC patients reported by Selmi et al, who used less sensitive (single round) PCR methodologies.1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar Thus, we are in agreement with Selmi’s negative tissue findings but it is unfortunate that they choose not to study lymph nodes with methodologies that might have actually detected HBRV. We developed an in vitro model of PBC using BEC purified from normal (donor) liver, which were incubated for 16 hours with PBC lymph node homogenates. After 5 days with fresh culture media, normal BEC developed the PBC phenotype, with aberrant PDC-E2 distribution. The conditioned media from these studies also propagated the phenotype in successive cultures; an effect that was abrogated by γ-irradiation.3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar Analysis of the conditioned media revealed that the transmissible agent had the hydrodynamic, enzymatic, genomic, and morphological characteristics of a betaretrovirus, genetically indistinguishable from mouse mammary tumor virus (MMTV). Indeed, pure MMTV isolates co-cultured with normal human BEC also resulted in aberrant PDC-E2 expression and viral protein in BEC.3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar, 4Xu L. Sakalian M. Shen Z. Loss G. Neuberger J. Mason A. Cloning the human betaretrovirus proviral genome from patients with primary biliary cirrhosis.Hepatology. 2004; 39: 151-156Crossref PubMed Scopus (88) Google Scholar The hypothesis of Selmi, our findings represent immune complexes adhering to cells was rigorously tested and excluded2Sadamoto T. Joplin R. Keogh A. Mason A. Carman W. Neuberger J. Expression of pyruvate dehydrogenase complex PDC-E2 on biliary epithelial cells induced by lymph nodes from primary biliary cirrhosis.Lancet. 1998; 352: 1595-1596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar; the sort incubation of homogenates and the subsequent 5 day kinetics of PDC-E2 induction, the abrogation of the phenotype by γ-irradiation, the serial passage of a transmissible and the development of the PBC phenotype following incubation of BEC with pure MMTV isolates clearly contradicts the immune complex hypothesis.2Sadamoto T. Joplin R. Keogh A. Mason A. Carman W. Neuberger J. Expression of pyruvate dehydrogenase complex PDC-E2 on biliary epithelial cells induced by lymph nodes from primary biliary cirrhosis.Lancet. 1998; 352: 1595-1596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar We agree with Selmi that PBC patients make antibodies reactive to MMTV viral preparations.1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar We disagree that the only reactivity was from antimitochondrial antibodies (AMA) reacting to mitochondrial contaminants in the viral preparations.7Mason A. Xu L. Guo L. Munoz S. Jaspen J.B. Bryer-Ash M. Coa Y. Sander D.M. Shoenfeld Y. Ahmed A. Van-de-Water J. Gershwin M.E. Garry R. Detection of retroviral antibodies in primary biliary cirrhosis and other idiopathic biliary disorders.Lancet. 1998; 351: 1620-1624Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar, 8Shen Z. Xu L. Neuberger J. Mason A. Betaretroviruses incorporate anti-mitochondrial reactive protein into the mature virion particle [abstract].Gastroenterology. 2004; 127: A671Google Scholar In previous retrovirus Western blots studies both our New Orleans’ laboratory and Gershwin’s laboratory were unable to detect AMA reactivity using PBC patient serum or monoclonal AMA to HIV or a retrovirus isolated from patients with Sjogren’s syndrome.7Mason A. Xu L. Guo L. Munoz S. Jaspen J.B. Bryer-Ash M. Coa Y. Sander D.M. Shoenfeld Y. Ahmed A. Van-de-Water J. Gershwin M.E. Garry R. Detection of retroviral antibodies in primary biliary cirrhosis and other idiopathic biliary disorders.Lancet. 1998; 351: 1620-1624Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar Since the latter viral preparations were purified much the same as those used for Selmi’s study, it is unknown why the MMTV preps should be contaminated with more PDC-E2. In our recent serologic studies, we have not only shown that AMA reacts with purified extracellular MMTV preparations derived from weanling pup milk but we have also found that AMA can immunoprecipitate MMTV, suggesting that the viral particle incorporates an AMA reactive protein.8Shen Z. Xu L. Neuberger J. Mason A. Betaretroviruses incorporate anti-mitochondrial reactive protein into the mature virion particle [abstract].Gastroenterology. 2004; 127: A671Google Scholar Moreover, we have found that by blocking PBC patient serum with 50 μg porcine PDC-E2 (Sigma, MO) abrogates serum reactivity to murine PDC-E2 and PDC-E3 binding protein but not other well characterized MMTV proteins, such as gp52 Env, p27 Gag and p15 Gag proteins (Figure 1).8Shen Z. Xu L. Neuberger J. Mason A. Betaretroviruses incorporate anti-mitochondrial reactive protein into the mature virion particle [abstract].Gastroenterology. 2004; 127: A671Google Scholar The statement that “contamination by viral particles had not been properly ruled out1Selmi C. Ross S.R. Ansari A.A. Invernezzi P. Podda M. Coppel R.L. Gershwin M.E. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biloiary cirrhosis.Gastroenterology. 2004; 127: 493-501Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar” is confusing as Selmi et al suggest that MMTV cannot infect human cells. Are they suggesting that MMTV (selectively) infected human PBC BEC cultures in Birmingham (UK)? None of the laboratories involved in these studies had any MMTV products prior to the original cloning of HBRV from the PBC cDNA library. RNA virus particles should not cause PCR artifacts, all steps were taken to avoid PCR contamination and there was no evidence of contamination, especially with either the original cloning of the HBRV from the PBC BEC library or the identification of HBRV in sucrose gradients of the pooled conditioned media from the co-culture studies. In both cases, the cloned products provided sequence variability inconsistent with a point source DNA contamination and in each study the negative controls were completely negative.3Xu L. Shen Z. Guo L. Fodera B. Keogh A. Joplin R. O’Donnell B. Aitken J. Carman W. Neuberger J. Mason A. Does a betaretrovirus infection trigger primary biliary cirrhosis?.Proc Natl Acad Sci. 2003; 100: 8454-8459Crossref PubMed Scopus (197) Google Scholar We agree with Selmi’s, statement that other research groups should attempt further independent studies. We hope that such studies focus on the samples, such as lymph nodes and BEC, where robust evidence of HBRV infection has been identified. Lack of immunological or molecular evidence for a role of mouse mammary tumor retrovirus in primary biliary cirrhosisGastroenterologyVol. 127Issue 2Preview Recent observations, including a pilot clinical trial, have suggested that a human mouse mammary tumor virus (MMTV) causes primary biliary cirrhosis (PBC). We attempted to confirm such data. We obtained sera from 101 patients (53 with PBC and 48 controls), fixed liver sections from 10 patients (8 PBC and 2 controls), fresh liver specimens (6 PBC and 6 controls), and fresh peripheral blood lymphocytes (PBLs) (10 PBC and 10 controls). We studied sera for reactivities against 3 different strains of MMTV virions, MMTV(C3H), MMTV(FM), and MMTV(LA), including goat polyclonal antibodies against MMTV virions, gp52, and p27 as positive controls. Full-Text PDF ReplyGastroenterologyVol. 127Issue 6PreviewWe thank Dr. Mason for his comments,1 but we submit that our data is extensive and fails to confirm the PBC betaretroviral story.2 Indeed there are several other failures in attempting to link endogenous retroviruses with human autoimmunity.3,4 For example, we did not find PCR evidence of MMTV in liver or PBMC from patients with PBC,1 while Dr. Mason’s data indicated that in PBC 14/49 liver samples and 2/12 PBMC were positive for the retrovirus.2 We used different conditions to avoid falsely positive results and also included rigorous negative and positive controls. Full-Text PDF