Association of stress proteins with autoantigens: a possible mechanism for triggering autoimmunity?

Patterns of autoantibody production are diagnostic of many autoimmune disorders; the recent observation of additional autospecificities towards stress-induced proteins may also provide insight into the mechanisms by which such responses arise. Grp78 (also known as BiP) is a target of autoaggressive B and T cell responses in our murine model of anti-Ro (SS-A) autoimmunity and also in rheumatoid arthritis. In this report we demonstrate reciprocal intermolecular spreading occurs between Ro52 and Grp78 in immunized mice, reflecting physiological association of these molecules in vivo. Moreover, we provide direct biochemical evidence that Grp78 associates with the clinically relevant autoantigen, Ro52 (SS-A). Due to the discrete compartmentalization of Ro52 (nucleocytoplasmic) and Grp78 (endoplasmic reticulum; ER) we propose that association of these molecules occurs either in apoptotic cells, where they have been demonstrated indirectly to co-localize in discrete apoptotic bodies, or in B cells themselves where both Ro52 and Grp78 are known to bind to immunoglobulin heavy chains. Tagging of molecules by association with Grp78 may facilitate receptor mediated phagocytotsis of the complex; we show evidence that exogenous Grp78 can associate with cell surface receptors on a subpopulation of murine splenocytes. Given the likelihood that Grp78 will associate with viral glycoproteins in the ER it is possible that it may become a bystander target of the spreading antiviral immune response. Thus, we propose a model whereby immunity elicited towards Grp78 leads to the selection of responses towards the Ro polypeptides and the subsequent cascade of responses observed in human disease.

Cognate T cell help is sufficient to trigger anti-nuclear autoantibodies in naive mice

The mechanisms involved in the initiation of anti-nuclear autoantibodies are unknown. In this study, we show that one factor allowing anti-nuclear autoantibodies to develop is the incomplete nature of immune tolerance to many of these proteins. Immune responses in mice toward the ubiquitous nuclear autoantigen La/SS-B are much weaker than responses to the xenoantigen, human La (hLa; 74% identical). However, in transgenic (Tg) mice expressing hLa, the Ab response to this neo-autoantigen was reduced to a level resembling the weak autoimmune response to mouse LA: Partial tolerance to endogenous La autoantigen was restricted to the T compartment because transfer of CD4(+) T cells specific for one or more hLa determinants into mice bearing the hLa transgene was sufficient to elicit production of anti-hLa autoantibodies. Notably, only hLa- specific T cells from non-Tg mice, and not T cells from hLa Tg mice, induced autoantibody production in hLa Tg mice. These findings confirm partial Th tolerance to endogenous La and indicate the existence in normal animals of autoreactive B cells continuously presenting La nuclear AG: Therefore, the B cell compartment is constitutively set to respond to particular nuclear autoantigens, implicating limiting Th responses as a critical checkpoint in the development of anti-nuclear autoantibodies in normal individuals.

Epitope mimics and determinant spreading: pathways to autoimmunity

Infectious microorganisms have evolved molecules which mimic the host in order to aid in their undetected propagation. In response, mammalian hosts have evolved a highly diverse immune repertoire designed to eradicate rapidly changing pathogens. The generation of diversity in the immune repertoire results in potentially damaging self cross-reactivities which require multiple regulatory controls to keep autoreactive lymphocytes in check. Here, we review how molecular mimicry at the T cell level might be important in the development of systemic autoimmunity.

Molecular chaperones are targets of autoimmunity in Ro(SS-A) immune mice

We have used a murine model of experimental anti-Ro(SS-A) autoimmunity to dissect additional intermolecular interactions between the 52-kD Ro (Ro52) and 60-kD Ro (Ro60) autoantigens and molecular chaperones. Immune responses to members of the heat shock protein hsp70 and hsp90 families were measured by immunoblotting and ELISA in sera from mice immunized and boosted with purified recombinant Ro52, Ro60 and La (SS-B). All Ro52 and Ro60 immune sera immunoblotted the inducible glucose-regulated protein grp78 and hsp70 species but not constitutive hsc70 or hsp90. The kinetics of antibody production and reciprocal affinity purification experiments indicated that the grp78 and hsp70 responses were cross-reactive but distinct from immune responses to the primary Ro52 and Ro60 immunogens and the endoplasmic reticulum (ER)-resident chaperone calreticulin. No responses to molecular chaperones were detected in the La-immunized mice. Control immunizations indicated that the recruited grp78 and hsp70 responses were specific for the Ro proteins and not due to immunization with denatured protein. The rapid spreading of immunity to the inducible grp78 and hsp70 in Ro52- and Ro60-immunized mice suggests that these components may co-localize and physically associate under certain physiological conditions which may promote autoimmunization. The potential importance of the ER-resident chaperones grp78 and calreticulin is further supported by their co-localization with Ro in small apoptotic membrane blebs and the finding of a novel putative grp78 binding motif in the carboxyl-terminal region of Ro52.

Determinant spreading: lessons from animal models and human disease

Spreading of the immune response is a common theme in organ-specific and systemic autoimmune diseases. We evaluated whether some of the mixed antinuclear antibody patterns characteristic of systemic autoimmunity might be the result of determinant spreading from a single initiating event. Immunisation of healthy mice with individual protein components of the La/Ro ribonucleoprotein (RNP) targeted in systemic lupus erythematosus and primary Sjögren's syndrome induced autoantibodies recognising Ro60 (SS-A), Ro52 (SS-A) and La (SS-B) and in some cases the molecular chaperones calreticulin and Grp78. The endogenous antigen(s) driving determinant spreading might be derived from physiological apoptosis which could explain the involvement of some chaperone proteins in the autoimmune response. Diversified anti-La/Ro antibody responses were initiated by challenge with a single subdominant T epitope of La even though some self epitopes of La were efficiently tolerised. The pattern of autoantibody responses in primary Sjögren's syndrome was strongly influenced by HLA class II phenotype which we speculate controls activation of T cells recognising defined peptides from the La/Ro RNP. In this way, HLA class II alleles may be critical in influencing initiation and spreading of systemic autoimmune reactions. Molecular mimicry of such determinants by exogenous agents might readily initiate spreading of an autoimmune response in genetically susceptible hosts.

Spreading of the immune response from 52 kDaRo and 60 kDaRo to calreticulin in experimental autoimmunity

Calreticulin (CR) is widely recognized as a new human autoantigen but there are conflicting data concerning its relationship with the Ro(SS-A) ribonucleoprotein (RNP). Recent evidence suggests that CR binds to 52 kDaRo (Ro52) by a protein/protein interaction and binds to hY RNA and rubella virus RNA. Other studies have shown that initiation of immunity to either Ro52 or 60 kDaRo (Ro60) can lead to reciprocal spreading of autoimmunity to Ro60 or Ro52, respectively, and induce anti-La autoantibodies in some strains of mice. These findings support a physical association of these polypeptides in Ro/La complexes. To test the hypothesis that CR is physically associated with Ro52 and/or Ro60 we examined the sera of Ro52-, Ro60- and La-immunized mice for intermolecular spreading to CR. Immune sera from BALB/c and C3H/HeJ mice immunized with recombinant 6xHis-mouse Ro52, 6xHis-human Ro60 or 6xHis-human La were tested for reactivity by ELISA and immunoblotting with a full-length human CR protein expressed as a soluble maltose binding protein fusion protein (CR-MBP). Five of the six Ro52-immunized C3H/HeJ mice sera and all six Ro60-immunized C3H/HeJ mice sera reacted with the CR-MBP (but not a MBP control) on ELISA. In the BALB/c group, the responder rate was lower with one in six of the Ro52-immunized and one in five of the Ro60-immunized mice spreading to CR. In contrast, none of the BALB/c or C3H/HeJ mice which was immunized with La showed evidence of a recruited anti-CR antibody response. Immunoblotting of the different recombinant proteins with immune sera from the C3H/HeJ mice confirmed the specificity of the initial and recruited antibody responses. The spreading of immunity from Ro52 and Ro60 to CR in Ro-immunized mice suggests that a subpopulation of CR or CR-like molecules must associate under certain circumstances with Ro52 and Ro60 polypeptides in vivo, possibly as Ro/CR complexes concentrated in surface membrane blebs of apoptotic cells. The lack of spreading to CR in La-immunized mice suggests that CR may be associated with a subpopulation of Ro particles from which La has already dissociated.

Monozygotic twins discordant for congenital complete heart block

OBJECTIVE

Isolated congenital complete heart block (CCHB) occurs in 1/20,000 live births. More than 85% of mothers giving birth to affected infants are anti-Ro antibody positive, but only approximately 1% of babies with anti-Ro-positive mothers develop CCHB. We studied 2 sets of monozygotic twins discordant for CCHB.

METHODS

Monozygosity was determined using placental examination and DNA microsatellite analysis. HLA typing was performed. Autoantibody studies were performed using counterimmunoelectrophoresis, immunoblotting, Ro 52 and Ro 60 enzyme-linked immunosorbent assay (ELISA), and indirect immunofluorescence (IIF) on Ro 60- and Ro 52-transfected HEp-2 cells.

RESULTS

Both sets of twins were monozygotic. They had similar birth weights. Twin 2 in the second set required a pacemaker at age 2 months. Both mothers were positive for anti-Ro 52 and anti-Ro 60 antibody, and neither had anti-La antibody on immunoblot. One set of twins was studied at birth. Similar titers of anti-Ro 52 and anti-Ro 60 antibody were found by IIF and ELISA.

CONCLUSION

There are no previous well-documented reports of monozygotic twins discordant for CCHB. These cases demonstrate that there is still discordance in the development of CCHB despite identical genetics and environmental exposure to anti-Ro antibody.

The immune response to 52-kDa Ro and 60-kDa Ro is linked in experimental autoimmunity

Clustering of autoantibody specificities is a consistent finding in patients with systemic autoimmune diseases. Patients with Sjögren's syndrome frequently have autoantibodies to La, 60-kDa Ro(SS-A) protein (Ro60), and 52-kDa Ro(SS-A) protein (Ro52). In the case of anti-Ro60 and anti-La, there is evidence that these specificities occur together because of the physical association of the Ro60 and La proteins that form a ribonucleoprotein particle (RNP). Thus, the autoantibody response may spread from a single epitope to involve new epitopes located within other components of the RNP. The physical association of Ro52 with the Ro/La RNP has remained controversial, implying that Abs to Ro52 are not a consequence of intermolecular spreading and may be triggered independently of the anti-Ro60 response. To examine this relationship of the immune response to Ro52 and Ro60, mice were immunized with recombinant Ro52, Ro60, or La, and examined for autoantibody production. Immunization with Ro52 resulted in rapid, high titer Ab production to Ro52, followed 7 to 14 days later by lower titer autoantibody production to Ro60. Immunization with Ro60 led to anti-Ro60, which was also followed 7 to 14 days later by a lower titer anti-Ro52 response. Cross-reactivity of affinity-purified Abs from immune mouse sera was not observed. These observations suggest that the autoimmune responses to Ro60 and Ro52 are linked intrinsically, despite previous evidence suggesting they are not associated in vivo. The mechanism of linkage remains unclear, but the data are most consistent with some physical association of Ro52 and Ro60 allowing autoimmunization, presumably as a result of normal cell turnover or specific injury in vivo.

The immune response to 52-kDa Ro and 60-kDa Ro is linked in experimental autoimmunity

Clustering of autoantibody specificities is a consistent finding in patients with systemic autoimmune diseases. Patients with Sjögren's syndrome frequently have autoantibodies to La, 60-kDa Ro(SS-A) protein (Ro60), and 52-kDa Ro(SS-A) protein (Ro52). In the case of anti-Ro60 and anti-La, there is evidence that these specificities occur together because of the physical association of the Ro60 and La proteins that form a ribonucleoprotein particle (RNP). Thus, the autoantibody response may spread from a single epitope to involve new epitopes located within other components of the RNP. The physical association of Ro52 with the Ro/La RNP has remained controversial, implying that Abs to Ro52 are not a consequence of intermolecular spreading and may be triggered independently of the anti-Ro60 response. To examine this relationship of the immune response to Ro52 and Ro60, mice were immunized with recombinant Ro52, Ro60, or La, and examined for autoantibody production. Immunization with Ro52 resulted in rapid, high titer Ab production to Ro52, followed 7 to 14 days later by lower titer autoantibody production to Ro60. Immunization with Ro60 led to anti-Ro60, which was also followed 7 to 14 days later by a lower titer anti-Ro52 response. Cross-reactivity of affinity-purified Abs from immune mouse sera was not observed. These observations suggest that the autoimmune responses to Ro60 and Ro52 are linked intrinsically, despite previous evidence suggesting they are not associated in vivo. The mechanism of linkage remains unclear, but the data are most consistent with some physical association of Ro52 and Ro60 allowing autoimmunization, presumably as a result of normal cell turnover or specific injury in vivo.

Nonprecipitating anti-La(SS-B) autoantibodies in primary Sjögren's syndrome

Anti-La(SS-B) precipitin-negative sera show a restricted epitope recognition and can be easily overlooked in routine laboratory testing. We have therefore determined the prevalence of nonprecipitating anti-La(SS-B) antibodies in patients with primary Sjögren's syndrome and studied their clinical and immunological associations. Clinical details were obtained from 68 patients with primary Sjögren's syndrome, and serum samples were examined by enzyme-linked immunosorbent assay using purified recombinant La, 60-kDa Ro, and 52-kDa Ro proteins and by counterimmunoelectrophoresis. Thirteen patients (19%) were identified with anti-La antibodies which were nonprecipitating. These patients had similar clinical findings to other groups of patients with Sjogren's syndrome, but had significantly lower rheumatoid factor and serum IgG levels than patients with anti-La precipitins. None of the patients with nonprecipitating anti-La antibodies had previously contained anti-La precipitins in their sera. Furthermore, they tended to have lower levels of antibodies directed against denatured 60-kDa Ro (but not 52-kDa Ro) compared with anti-La precipitin-positive patients. Patients with Sjögren's syndrome associated with nonprecipitating anti-La antibodies represent a stable serological and clinical subset in which there appears to be limited diversification of the autoimmune response to the Ro60 and La proteins of the Ro/La ribonucleoprotein.

Structural differences between the human and mouse 52-kD Ro autoantigens associated with poorly conserved autoantibody activity across species

Anti-nuclear autoantibodies found in human autoimmune diseases frequently cross-react with homologous autoantigens in distant species, supporting the notion that autoantibodies target conserved functional domains. However, the 52-kD Ro(SS-A) protein is an exception, in that human autoantibodies are not known to recognize any equivalent antigen in the cells of rodents and other non-primate species. To understand this lack of cross-reactivity we have isolated cDNAs encoding the mouse 52-kD Ro molecule. The cDNA encoding mouse 52-kD Ro revealed an open reading frame of 470 amino acids, with 70% sequence identity to the human 52-kD Ro antigen. The putative leucine-zipper and zinc-finger motifs present in human Ro52 were conserved in the mouse protein. Recombinant mouse 52-kD Ro protein reacted with human autoantibodies by ELISA and immunoblot, but with approximately 10-fold lower reactivity than recombinant human 52-kD Ro protein under the same conditions. Detection of both human and mouse 52-kD Ro by immunoblot was dependent on antigen concentration which was limiting in the cell equivalents generally used in immunoblot assays. Differential chaotropic disruption of antibody binding suggested a lower avidity of human autoantibody binding to the mouse 52-kD Ro protein compared with the human antigen. Thus the poor reactivity of native mouse 52-kD Ro with human autoantibodies is associated with species divergence diffusely distributed throughout the primary structure of the 52-kD Ro molecule.

Rapid and sensitive detection of anti-Ro (SS-A) antibodies by indirect immunofluorescence of 60kDa Ro HEp-2 transfectants

Anti-Ro (SS-A) antibodies are important diagnostic markers for primary Sjögren's syndrome and systemic lupus erythematosus, but their detection by indirect immunofluorescence (IF) in the diagnostic laboratory is hindered by the low cellular abundance of 60kDa Ro protein (Ro60). The approach we used to overcome this problem was to transfect and over-express the Ro60 gene into HEp-2 cells. In this study we have used a mixture of Ro60 transfectants and untransfected HEp-2 cells (HEp-Ro60) as a substrate for IF-antinuclear antibody (ANA) testing in a hospital laboratory. Screening of 240 routine serum specimens identified 14 Ro transfectant-positive sera which were confirmed by counterimmunoelectrophoresis (CIE); 3 of these sera were ANA-negative on untransfected cells and regular HEp-2. A comparison of HEp-Ro60 and regular HEp-2 showed strong concordance of the different ANA patterns between the 2 substrates. No increase in background staining was observed on the Ro transfectants when reacted with normal human sera. A comparison between HEp-Ro60 and CIE for 53 sera from patients with primary Sjögren's syndrome showed that HEp-Ro60 were a sensitive and specific substrate for detection of anti-Ro antibodies. Masking of positive Ro transfectants was observed rarely in sera containing multiple ANA specificities, but the Ro60 staining on these transfectants were unmasked at higher serum dilutions. We conclude that HEp-Ro60 are a suitable substrate for IF-ANA in the routine laboratory and that they have the additional advantage over regular HEp-2 slides of being able to detect anti-Ro in ANA-negative sera. HEp-RO60 are also a valuable confirmatory test for sera giving equivocal precipitin reactions or ELISA results.

Cytoplasmic accumulation of the 52 kDa Ro/SS-A nuclear autoantigen in transfected cell lines

The 60 kDa Ro/SS-A (Ro60) autoantigen is thought to reside predominantly in the nucleus; however, the intracellular localization of 52 kDa Ro/SS-A (Ro52) in normal cells is controversial, probably due to its low abundance. Therefore, we studied the intracellular expression and localization of the human Ro52 following transfection of a human Ro52 cDNA into cultured cell lines. Immunofluorescence staining of human (HEp-2) and mouse (LTA-5) cell transfectants with affinity-purified anti-Ro52 antibodies revealed that Ro52 antigen was most abundant in the cytoplasm and present to a lesser extent in the nucleus. This relative localization was supported by a preponderance of the Ro52 antigen in the cytoplasmic rather than nuclear fraction of enucleated cell lines detected by immunoblotting. In contrast to the Ro52 autoantigen, Ro60 and La autoantigens were mainly expressed in the nucleus of transfected cells under similar circumstances, indicating distinct localization of the intracellular pools of these autoantigens. The findings indicate that the Ro52 autoantigen lacks intrinsic signals required for nuclear localization and suggest that a significant pool of this autoantigen resides in the cytoplasm. Ro52 may therefore rely upon an association with other molecules for any specific nuclear transport.

Transfection and overexpression of the human 60-kDa Ro/SS-A autoantigen in HEp-2 cells

A full-length cDNA encoding human 60-kDa Ro/SS-A protein was transfected into and overexpressed in the human cell line HEp-2, with the aim of developing an improved reagent for indirect immunofluorescence (IF) detection of anti-Ro autoantibodies. Stable transfectants were analyzed by IF using a panel of 20 precipitin-positive anti-Ro human sera. Transfectants showed bright finely speckled nuclear and nucleolar staining. No surface membrane expression was detected despite marked overexpression of 60-kDa Ro. All human anti-Ro sera reacted with the transfectants with titers ranging from 1:320 to 1:40,960. The same sera tested on untransfected cells showed titers from negative (3 sera) to 1:160. These transfectants dramatically increase the sensitivity of IF testing (mean increase in titer of 41-fold) and allow detection of specific anti-Ro antibodies in samples negative or equivocal on untransfected cells. The staining patterns of antisera of other important specificities remained unaltered. In a study of sera from 22 patients with systemic lupus erythematosus, anti-60-kDa Ro autoantibodies were detected in all sera when tested on 60-kDa Ro-transfected HEp-2 cells; however, in 12 of 22 sera autoantibodies were undetectable by recombinant 60-kDa Ro ELISA. The 60-kDa Ro transfectants are a simple and sensitive method for detection of anti-Ro antibodies in patients with systemic rheumatic diseases.

Expression and functional conservation of the human La (SS-B) nuclear autoantigen in murine cell lines

We have studied the cellular expression and functional conservation of the human La autoantigen across species by introducing genes encoding human La antigen into cultured murine cell lines. In transfected murine fibroblasts and lymphoid cell lines human La was expressed as a predominantly nuclear antigen, with a typical pattern of nuclear speckling. Radiolabeled human La was of the predicted molecular mass (48 kDa) when expressed in murine cells and was associated intracellularly with murine 60 kDa Ro antigen. Under certain conditions of cell lysis the association between human La and murine Ro was disrupted, whereas human La and human Ro complexes remained intact, suggesting the possibility of species-specific protein interaction between La/Ro antigens. Expression of human La from the genomic gene construct was associated in some transfectants with the presence of additional high molecular weight bands reactive with anti-La monoclonal antibodies. Cell surface La determinants were not detected on any of the transfected cells. Human La protein associated with many of the same radiolabeled RNAs as the murine La antigen in transfected murine cells, indicating intact RNA-binding function of La protein across species. The findings indicate that despite 23% non-identity within the primary structure of human and murine La, the human molecule is functionally highly conserved across species.