Are the C-reactive protein values and erythrocyte sedimentation rate equivalent when estimating the 28-joint disease activity score in rheumatoid arthritis?

UNLABELLED: A formula for calculating disease activity score with 28 joint counts (DAS28) with C-reactive protein (CRP) instead of the erythrocyte sedimentation rate (ESR) has been proposed. OBJECTIVE: Here we analyze the factors that contribute to the differences in the DAS28 when calculated using either the ESR (DAS28-ESR) or the CRP values (DAS28-CRP). METHODS: We analyzed the data from 587 visits made by 220 patients with early arthritis. The age at the onset of the disease was 51+/-16 years old and 76.3% of the patients were women. The disease evolution at the first visit was 5 months and at each visit information related to several variables was collected, including that necessary to calculate the DAS28-ESR and DAS28-CRP. We defined a new variable DIFDAS=DAS28-ESR-DAS28-CRP to analyze which independent variables account for differences between the two indexes. RESULTS: There was a correlation between the two indexes of 0.91 (p<0.0001), although the DAS28-ESR value obtained was higher than that of DAS28-CRP at approximately 90% of the visits. Significantly, the difference between both indexes was higher than 0.6 in 44% of the visits studied. A multivariate analysis showed that female gender and disease duration were associated with the higher values obtained for DAS28-ESR when compared to those of DAS28-CRP. CONCLUSION: Our data show that DAS28-ESR and DAS28-CRP are not fully equivalent, because the former usually produces higher values. This finding is particularly relevant in females and patients with a long disease duration.

Vascular gene transfer driven by endoglin and ICAM-2 endothelial-specific promoters.

The involvement of the vascular endothelium in a large number of diseases supports the importance of vascular-specific gene delivery for their treatment. The hereditary hemorrhagic telangiectasia type 1 is an example of a vascular inherited disease (OMIM 187300). This is an autosomal dominant vascular disorder originated by mutations in the endoglin gene and associated with frequent epistaxis, telangiectases, gastrointestinal bleedings, and arteriovenous malformations in brain, lung and liver. Here, we address for the first time the possibility of using in vivo gene transfer to target endoglin expression to the vasculature. The promoter of the endothelial gene, ICAM-2, was used to generate transgenic animals which demonstrated endothelial expression of endoglin. Next, the promoters of the human endothelial genes, endoglin and ICAM-2, were inserted upstream of the human endoglin cDNA, and the resulting constructs were systemically or locally delivered, demonstrating endoglin expression in the vessel walls of liver, lung and skin. These gene transfer experiments represent an initial step in the treatment of the hereditary hemorrhagic telangiectasia type 1 by gene therapy, and suggest that endoglin and ICAM-2 promoters can be used to deliver other genes to the endothelium specifically.

Expression of normal and truncated forms of human endoglin.

Endoglin is a transmembrane glycoprotein 633 residues in length expressed at the surface of endothelial cells as a disulphide-linked homodimer; the specific cysteine residues involved in endoglin dimerization are unknown. Mutations in the coding region of the endoglin gene are responsible for hereditary haemorrhagic telangiectasia type 1 (HHT1), a dominantly inherited vascular disorder. Many of these mutations, if translated, would lead to truncated forms of the protein. It is therefore of interest to assess the protein expression of different truncated forms of endoglin. Infections in vitro or in vivo with recombinant vaccinia virus, as well as transient transfections with expression vectors, were used to express normal and truncated forms of endoglin. Truncated mutants could be classified into three different groups: (1) those that did not produce stable transcripts; (2) those that produced stable transcripts but did not secrete the protein; and (3) those that secreted a soluble dimeric protein. This is the first time that a recombinant truncated form of endoglin has been found to be expressed in a soluble form. Because a chimaeric construct encoding the N-terminal sequence of platelet/endothelial cell adhesion molecule (PECAM-1) antigen fused to residues Ile281-Ala658 of endoglin also yielded a dimeric surface protein, these results suggest that cysteine residues contained within the fragment Cys330-Cys412 are involved in disulphide bond formation. Infection with vaccinia recombinants encoding an HHT1 mutation did not affect the expression of the normal endoglin, and did not reveal an association of the recombinant soluble form with the transmembrane endoglin, supporting a haploinsufficiency model for HHT1.

Aggravation of adjuvant arthritis by carrageenan.

1. We have studied the optimum conditions for the induction of adjuvant carrageenan-induced inflammation (ACII) in male Wistar rats with limited susceptibility to adjuvant arthritis (AA). 2. ACII was induced by intradermal injection of Freund's complete adjuvant (CFA), containing 10 mg/ml Mycobacterium tuberculosis, followed by a subplantar inoculation of the nonspecific inflammatory stimulus carrageenan at different times. 3. Data obtained indicate that the arthritis of rats inoculated with CFA is significantly increased by carrageenan, particularly when it is injected 14 days after the adjuvant. Arthritis enhancement was more evident in the joints of the leg that had been previously injected with carrageenan, and remained stable around the peak level for some weeks. The development of joint inflammation was associated histologically with the appearance of inflammatory cells in the synovial membrane of those animals. 4. We found that the injection of carrageenan aggravated the course of AA in general, but very significantly when administered at the moment of the appearance of arthritis (day + 14). This aggravation affected both the intensity of inflammation and the chronicity of the disease.

Therapeutic effect of recombinant vaccinia virus expressing the 60-kd heat-shock protein on adjuvant arthritis.

OBJECTIVE: To analyze the effects of recombinant viruses for 2 heat-shock proteins in the treatment of adjuvant arthritis. METHODS: Virus vaccinia recombinant for mycobacterial heat-shock protein 65 (hsp65-VV) and human hsp60 (hsp60-VV) were administered to rats during different stages of adjuvant arthritis. Arthritis score and immunity to the recombinant virus were analyzed. RESULTS: When delivered at the pre-arthritis stage, both constructs ameliorated arthritis; greater protection was observed with hsp60-VV. A specific T cell response to the recombinant proteins was detected. Furthermore, hsp60-VV displayed a clear therapeutic effect on established arthritis. CONCLUSION: Our results suggest novel avenues of therapeutic intervention in autoimmune arthritis associated with immunity to hsp60.

Modulation of adjuvant arthritis in Lewis rats by recombinant vaccinia virus expressing the human 60-kilodalton heat shock protein.

The immune response to the mycobacterial 65-kDa heat shock protein (hsp65) is considered an important event in the induction of adjuvant arthritis (AA) in rats; this induction probably occurs through a molecular mimicry mechanism involving cross-reactivity against the rat homolog hsp60. To analyze the role of mammalian molecule hsp60 in arthritis, we generated a recombinant vaccinia virus (hsp60-VV) carrying the human hsp60 gene inserted into the thymidine kinase locus under the control of the 7.5k vaccinia virus promoter. Human hsp60 is almost identical to its rat homolog (97.4% linear amino acid homology) and shares about 50% of amino acid positions with Mycobacterium tuberculosis hsp65. The latter supposedly carries a critical epitope for AA induction that is not present in human hsp60. Infections with hsp60-VV of monkey cell cultures led to the expression of the human hsp60 molecule, as evidenced by immunoblotting analysis with specific monoclonal antibodies. Also, Lewis rats infected with hsp60-VV produced specific antibodies, demonstrating the in vivo expression of human hsp60 in the infected animals. Therefore, we used hsp60-VV to analyze whether the delivery of hsp60 could affect the induction of AA in Lewis rats. hsp60-VV clearly reduced and retarded arthritic symptoms when administered to rats at day 7 after AA induction. In contrast, inoculation of rats with a control recombinant vaccinia virus did not affect the course of the disease. The improvement in AA with hsp60-VV administration was associated with a specific immune response, as determined by the presence of antibodies to hsp60 in the sera and the proliferation induced by hsp60 of T cells from popliteal lymph nodes. These results support a critical role for immunity to heat shock proteins in AA. Since the protective construct is virtually identical to rat homolog hsp60, we conclude that immunity directed to conserved areas of this family of proteins is directly involved in the pathogenesis of AA.

Inhibition of T-cell proliferation by rat synoviocytes.

Antigen presentation by synoviocytes to arthritis-related T-cell clones was studied in Lewis rats. Freshly isolated synoviocytes may be able to take up and process mycobacterial antigens but they seem inefficient as antigen presenting cells. Furthermore, synovial cells inhibited the specific proliferative responses of T lymphocytes by a mechanism which apparently was not antigen-specific or mediated by secreted cytokines. Synovial cells isolated from rats at the period of developing adjuvant arthritis showed a lower inhibitory capacity associated with a lower degree of antigen clearance. These results suggest that synoviocytes might inhibit T-cell responses in normal joints and that this negative control diminishes following arthritis induction.

Aggregated human immunoglobulins bind to modified proteins.

Human immunoglobulins treated at 55 degrees C in vitro are able to interact with maleylated bovine serum albumin (mBSA), but not with unmodified BSA. Gel filtration experiments demonstrated that the mBSA binding is associated with a high molecular weight complex of aggregated IgG. This aggregated IgG with binding capacity for mBSA could also be generated in vitro by treatment of human IgG at 37 degrees C or 40 degrees C and by incubation with human neutrophils. Furthermore, IgG aggregates with binding activity for mBSA could be detected in untreated synovial fluids from rheumatoid arthritis patients, indicating that these complexes occur in vivo. The phenomenon of binding to aggregated IgG was extended to other modified proteins such as maleylated human serum albumin (mHSA), acetyl low density lipoprotein (Ac-LDL) and BSA reacted with oxidized linolenic acid. Soluble forms of these modified proteins were able to compete for the interaction between aggregated IgG and surface-bound mBSA. We also found that aggregated IgG enhanced the Ac-LDL-dependent foam cell formation. These findings suggest a role for aggregated IgG in the metabolism of oxidized proteins.

Decreased tubulin synthesis in synoviocytes from adjuvant-induced arthritic rats.

The first microscopical alterations along adjuvant arthritis induction in rats seem to appear in the synovium. We have studied the protein synthesis pattern of the cells constitutively present in synovial membrane (synoviocytes) and have found an impairment of synthesis of some proteins when synoviocytes are derived from adjuvant arthritic rats. One of these polypeptides was identified as beta tubulin by two-dimensional gel electrophoresis, a membrane transfer assay using a specific monoclonal antibody and peptide mapping. We postulate that a repressed synthesis of tubulin may be an initial step in the triggering of the disease, since the effect was evident at pre-arthritic stages, when infiltration by inflammatory cells had not yet occurred.

Arthritis transferred by cells derived from pre-inflammatory rat synovium.

Different cell populations isolated from rats during the period of latency of adjuvant arthritis were injected into the bloodstream of naive rats to test their ability to transfer articular disorders. Synovium-derived cells (synoviocytes) were able to induce arthritis in 3 out of 4 recipient animals, whereas peripheral blood leukocytes, peritoneal exudate macrophages, lymph node cells, synoviocyte lysates and synoviocytes from control animals were not able to do so. This model of cellular transferred arthritis is associated with antibody titres to hsp65 in rat sera. Our findings suggest a crucial role for synovial cells in the pathogenesis of adjuvant disease, which might be linked to their function as accessory cells.

Antibodies to dietary antigens in rheumatoid arthritis--possible molecular mimicry mechanism.

Antibodies in serum from some patients with rheumatoid arthritis, recognize bovine albumin present in the milk, as determined by immunoprecipitation analysis from 125I-milk extracts. This antigen was also immunoprecipitated from bovine sera. These and ELISA studies showed that BSA is preferentially recognized over other proteins present in the milk. Panel studies demonstrated that although the average reactivity for BSA was high, only one third of the sera tested displayed a reactivity above the mean. The possibility of a molecular mimicry mechanism in RA between this food antigen and other human antigens was investigated. A sequence alignment analysis showed that the residues 141-157 of bovine albumin significantly differed from the corresponding fragment of human albumin, but were highly homologous with human collagen type I, C1q and vitamin D binding protein. In support of the immunogenicity of this fragment, we found that representative RA sera displayed a specific reactivity for a synthetic peptide containing the BSA residues responsible for the homology. Furthermore, most of the epitopes recognized on BSA by the RA sera seem to be conformationally dependent as heat denaturation or reduction followed by alkylation lead to a diminished recognition.

Cellular and humoral reactivity pattern to the mycobacterial heat shock protein HSP65 in adjuvant arthritis susceptible and resistant Wistar rats.

We have analyzed the cellular and humoral immunity to the mycobacterial 65 KDa heat shock protein (hsp65) in a group of Freund's Adjuvant-immunized rats with a limited susceptibility to Adjuvant arthritis. According to the arthritis indices during the period of study (35 days), two different groups of rats could be distinguished; a) autoimmune Adjuvant arthritic rats (AA), and b) Non-arthritic animals (NA), including both rats which did not display any disease symptoms and rats suffering mild transient inflammation. The cellular response to the immunizing agent (Mycobacterium tuberculosis) or the mitogen Concanavalin A was comparable between both groups of rats. However, we detected an impaired cellular response to the individual hsp65 antigen in the animals that did not develop the disease. On the contrary, the level of hsp65-specific antibodies was much higher in NA animals than in AA rats suggesting a protective role for the hsp65 specific antibodies.

Rheumatoid disease resembling lung neoplasia.

We describe a 67-year-old man with severe rheumatoid arthritis of long duration. He developed a peculiar extraarticular rheumatoid complication consisting of localized lung consolidation with a pathological costal fracture, together with an abrupt systemic reaction that occurred in the course of immunosuppressive treatment. The diagnosis first proposed was lung cancer with costal metastases. However exhaustive studies performed in the search of malignancy were systematically negative and pathologic studies finally demonstrated bronchiolitis obliterans with organizing pneumonia (BOOP) as responsible for parenchymal lung consolidation with a rheumatoid nodule eroding bone at the level of the rib fracture. These findings, after long followup of the patient, attest to the rheumatoid origin of his bizarre manifestations and definitely rule out a neoplastic etiology.

Synoviocytes type A bind exogenous antigens recognized by antibodies present in rheumatoid arthritis.

The presence of antibodies in rheumatoid arthritis (RA) patients to antigens on the synoviocyte surface has recently been reported (Scand. J. Immunol. 27, 295, 1988). Here we have further characterized these antigens and found that they are exogenous proteins acquired from the bovine serum used in the culture medium. By immunoprecipitation and ELISA studies, we have identified bovine albumin and transferrin as the antigens recognized by the RA antibodies. These specificities were found not only in the sera but also in the synovial fluid from RA patients. A comparative study with a large panel of RA sera did not show a correlation in the antibody specificities for bovine albumin, bovine transferrin, or the 65-kDa heat shock protein from Mycobacterium bovis. Similar experiments using rabbit and monkey sera as well as human synovial fluid and serum as a source of antigen did not reveal any reactivity with a highly positive RA serum. By sequence alignment, a high degree of homology between residues 142-156 from bovine albumin and residues 65-78 from human pro-collagen alpha 1 (I) was found. The capacity of the synoviocytes to bind exogenous antigens and the presence of antibodies to bovine proteins, normally present in the diet, suggest a role for these type A synoviocytes as well as a possible involvement of food antigens in the pathogenesis of RA.

The calcaneus in ankylosing spondylitis. A radiographic study of 43 patients.

A prospective study analysing the radiological findings of calcaneus bone in ankylosing spondylitis is presented. It includes 43 patients with a relatively low mean age (33.5 years). The morphology of their lesions correlated with disease duration. Abnormal radiological findings were observed in 63% of the patients and in 58% of the radiographs, but in only 3 of 20 normal subjects. Erosive lesions were more frequently found in early stages, whereas sclerotic and proliferative lesions were typical of advanced stages of the disease. Only one patient suffered relevant local pain. No clinico-radiological correlation could be established.

Adjuvant-induced polyarthritis. Synovial cell activation prior to polyarthritis onset.

We studied synoviocytes obtained from rats that had been injected with Freund's complete adjuvant 12 days prior to killing. We found that several activation parameters were affected in these synoviocytes, namely, the concanavalin A protein-binding pattern, the production of superoxide, and the appearance of the p77 polypeptide, which we have previously shown to be associated with the activation caused by the induction of polyarthritis. Our findings suggest that prior to the establishment of the inflammatory process, synoviocytes are in a state of activation, and may be a component of the molecular mechanism during the early stages of disease.

Autoantibodies from rheumatoid arthritis patients recognize antigens on the synoviocyte surface.

We have found autoantibodies in the sera from rheumatoid arthritis (RA) patients which recognize two cell surface antigens of approximately 70 kDa and 28 kDa from synoviocyte extracts as detected by immunoprecipitation analysis. These polypeptides were immunoprecipitated from extracts containing mainly macrophage-like synoviocytes (type A) but not from extracts of homogeneous fibroblast-like synoviocytes (type B). These autoantigens are not selectively expressed by RA synoviocytes, since both RA and non-rheumatoid synovia were reactive for RA sera. From the panel of different RA sera tested, 64% immunoprecipitated the 70 kDa band, and 27% recognized the 28 kDa polypeptide. These differences in the specificity of the sera seemed to be related to the clinical state of the donor. The sera from patients suffering from other autoimmune diseases such as autoimmune thyroiditis and systemic lupus erythematosus (SLE) do not appear to be reactive for these specificities, but sera from patients with Sjögren's syndrome, psoriatic arthritis, and Crohn's disease showed a weak cross-reactivity with the 70 kDa polypeptide. This autoreactivity against synovial cells in RA supports the idea that these cells participate in the initial immune response of the disease.

Biochemical analysis of synoviocytes from normal and arthritic rats. Evidence for an activated state associated with adjuvant polyarthritis.

Adjuvant-induced polyarthritis in rats is a common model system used for the study of the synovitis that occurs in rheumatoid arthritis. Synoviocytes A, the major cell type covering the internal surface of the joint, could be involved in the pathogenesis of rheumatoid arthritis because of their increased proliferation and the intraarticular manifestations of the disease. So far only a few molecular studies have been reported on synoviocytes upon arthritis induction. We report here changes in polypeptides, between control and arthritic synoviocytes, by using two different radiolabeling methods and two-dimensional gel electrophoresis analysis. Major differences were found using metabolic labeling on regions of tropomyosins, cyclin, tubulins and vimentin. In addition, external surface labeling of the cells with lactoperoxidase showed clear differences between control and arthritic synoviocytes in the region of 77-100-kDa proteins. Some of these differences can be reproduced by certain macrophage activators such as phorbol myristate acetate and lipopolysaccharide acting on synoviocytes in vitro and in vivo respectively. These results exclude the possibility that the changes observed were due to a possible infiltration of other cell types in the arthritic synovia and strongly support the existence of an activated state of synoviocytes associated with arthritis induction.