Anti-mitochondrial antibody immunofluorescent titres correlate with the number and intensity of immunoblot-detected mitochondrial bands in patients with primary biliary cirrhosis

https://doi.org/10.1016/j.cca.2007.01.023Get rights and content

Abstract

Background

Primary biliary cirrhosis (PBC)-specific anti-mitochondrial antibodies (AMA), routinely detected by indirect immunofluoresnce, immunofix various mitochondrial antigens by immunoblot (IB). We have investigated whether there is a relationship between IFL-determined AMA titres and the number, intensity or pattern of IB-bands.

Methods

By computer-assisted imaging technology, an IB analysis using human liver mitochondrial preparation was performed in 100 AMA positive PBC samples, subdivided in 5 groups each consisting of 20 cases with very high, high, intermediate, low or borderline IFL titres and in 19 AMA-negative cases.

Results

Reactivity to at least 1 mitochondrial band (range 1–12) was present in 114 (96%) cases. Reactivity to the major pyruvate dehydrogenase E2-subunit 74-kDa autoantigen positively correlated with AMA IFL titres but was absent in 25% IFL-positive cases. The intensity and number of IB-bands differed significantly among groups (p < 0.001). IFL AMA titres positively correlated with the total number (r = 0.8, p < 0.001) and the intensity of the IB-bands (r = 0.75, p < 0.001).

Conclusions

Our study indicates that IFL-determined AMA relate to the number and intensity of immunofixed bands. Reactivity to PDC-E2 does not correlate with IFL titres in a considerable proportion of patients suggesting that detection methods of AMA need to be based in the entire spectrum of the disease-specific mitochondrial antigens.

Introduction

The serological hallmark of primary biliary cirrhosis (PBC), an immune-mediated chronic cholestatic liver disease, is the presence of disease-specific anti-mitochondrial antibodies (AMA) typically staining the proximal and especially the distal renal tubules in the kidney and the gastric parietal cells in the stomach by indirect immunofluorescence (IFL) on a composite rodent substrate composed of stomach, kidney and liver [1], [2].

The identification of AMA molecular targets as subunits of the 2-oxo-acid dehydrogenase multienzyme complexes, (2-OADC), formerly termed the M2 antigen, has greatly favoured the development of immunoblotting which can simultaneously test for presence and antigen specificity of AMA [2], [3], [4], [5]. Thus, several studies have shown that AMA seropositive PBC patients predominantly immunofix the E2 subunits of pyruvate dehydrogenase complex (PDC), branched-chain 2-oxo-acid dehydrogenase complex (BCOADC) and 2-oxoglutarate dehydrogenase complex (OGDC), and the E3 binding protein (E3BP) of PDC and to a lesser extent the other OADC subunits [1], [2], [3], [6].

Despite the considerable progress on the elucidation of the fine specificity of anti-OADC (AMA) responses in PBC, it remains unclear whether there is a relationship between IFL-determined AMA titres and the number, intensity or pattern of the immunofixed mitochondrial bands [7]. Hence, it is currently unknown whether there is a band or a combination of mitochondrial bands responsible for most of the AMA-specific fluorescent staining [2], [7]. To address this point, we divided a set of 119 PBC patients into six groups according to their IFL titres, classified as very high, high, intermediate, low or borderline positive AMA responses or AMA-negative by IFL PBC cases. We then analysed serum samples from these patient groups using human liver mitochondrial antigenic preparations. We were then able to quantify and compare the intensity of mitochondrial bands amongst various groups using computer-assisted imaging technology.

Section snippets

PBC patients

One hundred and nineteen patients with histologically proven PBC [1], [8], [9], [10], mean age 51.2 ± 12.5 years, 109 female, mean duration of disease 28.9 ± 32.1 months, selected on the basis of their AMA IFL titres, were studied including 20 patients with IFL of ≥ 1/5120 (‘very high AMA positive’ group), 20 with 1/1280 or 1/2560 (‘high AMA positive’ group), 20 with 1/320 or 1/640 (‘intermediate AMA positive’ group), 20 with 1/80 or 1/160 (‘low AMA positive’ group), 20 with 1/40 (‘borderline

Results

Fig. 1 and Table 1 display the results of the immunoblot analysis amongst various groups of PBC patients divided according to their IFL AMA titres. Of the 119 PBC patients tested, 114 (96%) recognised at least 1 mitochondrial band (range 1–12). The most prevalent band is the 74 kDa PDC-E2 antigen recognised by 83/119 (70%) PBC patients, followed by the 51 OGDC-E2/BCOADC-E2 bands (71/119, 60%). Within groups, all 40 patients with ‘very high’ or ‘high’ AMA titres reacted with the 74 kDa PDC-E2

Discussion

In the present study we have analysed by immunoblotting the AMA reactivity of PBC patients subdivided according to their IFL AMA titres and we have found a positive correlation between the intensity of the 74 kDa band corresponding to the major PDC-E2 mitochondrial antigen and IFL AMA titres suggesting that among various OADC antigens, the magnitude of responses directed to PDC-E2 has a predominant effect on the typical staining of AMA on rodent tissue substrate. Thus, only a small fraction of

Acknowledgments

Supported in part by a research grant of the Research Committee of the University of Thessaly, Greece (Code no: 2446); C Liaskos was supported by CE.RE.TE.TH and by a scholarship from the Hellenic Association for the Study of the Liver; DP Bogdanos was supported by Children's Liver Disease Foundation, Birmingham, UK, the PBC Foundation, Edinburgh, Scotland and a CSL Award granted by the Higher Education Funding Council for England.

Cited by (0)

View full text