RNA-protein complexes as autoantigens : cell and molecular biology of Ro/SSA and disease associations of anti-Ro/SSA antibodies

Abstract

Autoantibodies to the Ro/SSA and La/SSB antigens are present in sera from patients with Sjögren's syndrome (SS) and systemic lupus erythematosus (SLE). The Ro/SSA autoantigen consists of a 52kD and a 60kD protein, complexed with one of four small RNAs. The La protein can also associate with the complex. We have focused mainly on characterisation of Ro 52kD, partly on identification of additional components of the Ro/SSA complex, and partly on the characterisation of anti-Ro and anti-La antibody responses.

To study the intracellular localisation of Ro 52kD, expression plasmids were constructed encoding fusion proteins between Ro 52kD subfragments and green fluorescent protein (GFP) from jelly fish. These experiments demonstrated a cytoplasmic localisation of the Ro 52kD protein and indicated a crucial role for the hydrophilic domain in restricting the Ro 52kD protein to the cytoplasm.

In order to investigate if the putative zinc fingers of Ro 52kD bind Zn2+ various Ro 52kD subclones were expressed as recombinant proteins and assayed for Zn2+ binding in vitro. One fragment containing the first cysteine/histidine cluster at residues 14-54 and other larger overlapping fragments were demonstrated to bind Zn2+. We also demonstrated that the Zn2+ binding domain is a target for conformation-dependent anti-Ro 52kD autoantibodies. Interestingly, antigenicity is dramatically increased by the reducing conditions which also promote Zn2+ binding.

With the aim of searching for additional components of the Ro/SSA complex, a phage display using Ro 60kD and Ro 52kD recombinant fusion proteins as bait proteins was performed on a randomly cloned cDNA library. One candidate gene, phox 40kD, was selected for investigation of the protein-protein interaction with recombinant Ro and La proteins. Further analysis is needed for confirmation of this interaction.

Autoantibodies with Ro 52kD, Ro 60kD and La specificities, respectively, were defined within ANA-negative sera. Samples from ANA-negative but Ro/SSA positive patients undergoing investigation due to suspected connective tissue diseases were analysed by immunoblotting and ELISA using recombinant antigens and synthetic peptides. Autoantibodies to Ro 52kD were present in 65% of selected sera and an autoepitope within the Ro 52kD protein composed of the leucine zipper domain was defined.

IgA autoantibodies to the Ro 52kD, Ro 60kD and La antigens in sera from patients with primary SS and SLE were analysed using a semiquantitative immunoblotting approach. Our data suggest that IgG dominates the autoantibody response followed by IgM > IgA, and 50-80% of the patients with primary SS and SLE had IgA autoantibodies to the three Ro/SSA proteins. The epitope specificity of the IgA autoantibodies was similar to that of IgG and IgM, and the immunodominant epitopes were represented by aa 136-227 for Ro 52 and by aa 181-320 for Ro 60.

In conclusion. the subcellular localisation of Ro 52kD was defined as cytoplasmic and the putative zinc-binding capacity of Ro 52kD demonstrated. IgA autoantibodies to Ro/La were shown to be present in most of Ro/SSA positive sera, and selected ANA-negative sera demonstrated to contain Ro antibodies.