Article Text

Download PDFPDF

Clinical features of several connective tissue diseases with anti-Golgi antibody
  1. Y YANG,
  2. J FUJITA,
  3. M TOKUDA,
  4. S BANDOH,
  1. M MUROTA,
  2. F WU,
  1. First Department of Internal Medicine
  2. Kagawa Medical University
  3. Kagawa, Japan
  4. Third Department of Internal Medicine
  5. Kagawa Medical University
  6. Kagawa, Japan
  1. Dr J Fujita, First Department of Internal Medicine, Kagawa Medical University, 1750–1, Miki-cho, Kita-gun, Kagawa, 761–0793, Japan jiro{at}

Statistics from

Request Permissions

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.

Rodriguez et al were the first to report autoantibodies directed against the Golgi complex identified in the serum of a patient with Sjögren's syndrome (SS).1 Since then, several isolated reports have described the presence of anti-Golgi antibodies (AGAs) in several connective tissue diseases (CTDs).1-7 In addition, immunoblotting and immunoprecipitation studies have suggested that there are at least 14 different Golgi complex autoantigens, and their molecular masses range from 35 to 260 kDa.5 However, few reports describe the association between the clinical features of CTD and AGAs. In this letter we present a case of rheumatoid arthritis (RA) associated with AGAs and review several reported cases.

The patient was a woman born in 1939 who developed seropositive RA in 1990. She was admitted to our hospital in March 1998 because of high grade fever, cough, sore throat, chest pain, and severe arthralgia. Systemic laboratory examination disclosed no antinuclear antibody, anti-DNA antibody, or anti-Jo-1 antibody. Anti-SS-A antibody was positive. Chest computed tomography showed interstitial pneumonia with fine reticular shadow and honeycombing in both lower lobes. Physical examination showed fine crackles in both lung fields. A laboratory examination showed raised C reactive protein (78 mg/l), serum aspartate aminotransferase (62 U/l), serum alanine aminotransferase (52 U/l), creatine kinase (236 U/l), lactate dehydrogenase (527 U/l), IgA (5.6 g/l), and IgG (26.2 g/l). Mild interstitial lymphocytic infiltration was demonstrated in a muscle biopsy specimen. Therefore, a clinical diagnosis of RA, polymyositis complicated with interstitial pneumonia was made, in addition to a suspicion of SS.

Indirect immunofluorescence stained by the patient's serum (fig 1) showed crescent-shaped cytoplasmic organella that surrounded the nuclear membrane of Hep-2 cells, and these were considered to be Golgi apparatus. Linear-shaped cytoplasmic filaments were also seen and considered to be cytokeratins. Western blotting analysis against lung epithelial cell line (A549) and hepatoma cell line (HLE) showed that proteins of 58 kDa, 54 kDa, and 50 kDa were stained by the patient's serum. The 58 and 50 kDa proteins were considered to be Golgi antigens, and the 54 kDa protein was considered to be cytokeratin 8. Tables 1outlines the clinical features of the 15 patients with AGA reviewed.1-7 The patients comprised 11 women, three men, and in one case the sex of the patient was not given. Western immunoblot analyses disclosed several antigens with molecular weights ranging from 50 to 230 kDa.

Figure 1

Indirect immunofluorescence staining of anti-Golgi antibodies just outside the nuclear membrane of Hep-2 cells with the patient's serum. Crescent-shaped cytoplasmic organella that surround the nuclear membrane of Hep-2 cells are clearly shown, and considered to be Golgi apparatus. Linear-shaped cytoplasmic filaments are also seen and considered to be cytokeratins.

Table 1

Characteristics of patients with connective tissue diseases in whom anti-Golgi antibodies were detected

To date, several possible clinical correlations have been identified in patients with AGA. However, the clinical associations are different; in Blaschek's report, the incidence of an association with SS was shown to be significantly higher than in patients with other CTDs,8 whereas Fritzler's report suggested a strong association with systemic lupus erythematosus (SLE).2 Our present letter also showed that AGA was detected in patients with SS, SLE, and RA. As clinical features, our patient had mild liver dysfunction as indicated by raised liver enzymes and interstitial pneumonia. Fritzler et al have also reported that 5/8 patients with AGA had liver dysfunction.2 In addition, several patients had cardiopulmonary diseases, including pulmonary fibrosis as shown in our case.2 ,3 Because we have shown the existence of antigens in type II epithelial cells (A549) as well as in hepatoma cell lines (HLE), it was speculated that the existence of AGA might be related to liver dysfunction and the onset of interstitial pneumonia. Furthermore, detection of AGA in a patient with RA might also have potential pathogenic implications, because the Golgi apparatus participates in terminal protein glycosylation whereas high levels of agalactosyl IgG occur in RA and correlate with disease activity.9

In conclusion, our case and those of previous reports suggest that although antigens of AGA have diversity and heterogeneity, AGA might be pathogenetically related to some clinical features of CTDs.