Subarachnoid hemorrhage and systemic lupus erythematosus.

The frequency, clinical profile, treatment and outcome of subarachnoid hemorrhage (SAH) in patients with systemic lupus erythematosus (SLE) were assessed retrospectively, based on the case records of SLE of the Jichi Medical School Hospital over a 20 year period. Clinically defined SAH was found in 10 (3.9%) out of 258 SLE patients, which represented a frequency higher than previously assumed. Five patients had active SLE and lacked an apparent cause of SAH, other than SLE. A high mortality rate (5/5), no visible aneurysm on angiogram (3/4), and an onset during intractable SLE or after discontinued or no steroid therapy because of medical noncompliance (4/5) were characteristic of patients with active SLE, and thus an earlier successful suppression of SLE, if possible, might have prevented their SAH. In contrast, in the 5 patients with inactive SLE, 2 out of 3 saccular aneurysms were successfully clipped and small bleeding of one patient without aneurysms remitted spontaneously without the need for additional steroid therapy. When one death, which occurred outside of medical care, was excluded, the survival ratio of the hospitalized SAH patients with inactive SLE was significantly better than that with active SLE (3/4 versus 0/5, P=0.0476). In conclusion, the relatively common occurrence of SAH in SLE patients, and a significantly different clinical impact of SAH in respect to active and inactive SLE, were suggested from the results.

In vitro and in vivo inhibition of activation induced T cell apoptosis by bucillamine.

OBJECTIVE: To investigate the mechanism of autoimmune phenomena, occasionally seen in patients with rheumatoid arthritis treated with bucillamine (BUC) and D-penicillamine (D-Pen), by evaluating their effects on apoptosis of T cells induced by T cell receptor activation or dexamethasone. METHODS: In vitro apoptosis was induced in a T cell hybridoma (SSP3.7) and a B cell line (WEHI 231) by activation of respective receptors or dexamethasone, in the presence or absence of BUC or D-Pen. In vivo apoptosis was induced in BALB/c mice by staphylococcal enterotoxin B (SEB), with or without BUC or D-Pen, and thymocytes were examined for it by FACS. RESULTS: Stimulation with anti-CD3 and dexamethasone induced apoptosis in 72% and 71% of SSP3.7 cells, respectively. However, only 16% of SSP3.7 cells became apoptotic by anti-CD3 when BUC was added to the culture media. By contrast, 80% of SSP3.7 cells became apoptotic when stimulated by dexamethasone, even in the presence of BUC. BUC did not affect apoptosis of WEHI 231 cells induced by anti-IgM. Although SA981 (a metabolite of BUC) inhibited apoptosis of SSP3.7 cells induced by anti-CD3, D-Pen did not. BUC, SA981, or D-Pen did not significantly influence the level of interleukin 2 secretion stimulated by anti-CD3. In contrast, both BUC and D-Pen inhibited apoptosis of Vbeta8+ thymocytes induced in vivo by SEB superantigen. Neither BUC nor D-Pen significantly changed the number of CD4+CD8+ thymocytes in BALB/c mice injected with dexamethasone. CONCLUSION: BUC decreased, while D-Pen did not, the apoptosis of T cells stimulated by anti-CD3 in vitro, although they both inhibited the deletion of immature thymocytes reactive with SEB in vivo. This may explain autoimmune phenomena sometimes seen during the treatment of rheumatic patients with these drugs.

A close temporal relationship of liver disease to antiribosomal P0 protein antibodies and central nervous system disease in patients with systemic lupus erythematosus.

OBJECTIVE: To determine whether there is a close temporal relationship of liver disease to serum IgG and/or IgM antiribosomal P0 protein antibodies (anti-P0) and central nervous system (CNS) lupus in patients with systemic lupus erythematosus (SLE). METHODS: The study included 70 patients with active SLE. Of these, 30 had IgG and/or IgM anti-P0 and 14 had CNS lupus other than psychiatric disease (nonpsychiatric CNS lupus). Of these 14 patients, 11 had anti-P0. Laboratory manifestations of liver disease were retrospectively analyzed. RESULTS: Liver disease not attributed to any cause other than SLE (SLE liver disease) was present in 8 of the 11 patients with anti-P0 with nonpsychiatric CNS lupus (72.7%), in none of the 19 patients with anti-P0 without nonpsychiatric CNS lupus (0%), and in one of the 40 patients without anti-P0 (2.5%). The prevalence of SLE liver disease was significantly greater in patients with anti-P0 with nonpsychiatric CNS lupus than in the other 2 groups (p < 0.0001). Mean levels of liver enzymes (lactate dehydrogenase, glutamic oxaloacetic transaminase, glutamate pyruvate transaminase, gammaglutamyl transpeptidase) were significantly higher in patients with anti-P0 with nonpsychiatric CNS lupus than in the other 2 groups. Serial studies in 3 patients showed that the appearance of anti-P0 and liver dysfunction slightly preceded the onset of nonpsychiatric CNS lupus. CONCLUSION: Anti-P0 may be related to the pathogenesis of CNS lupus and SLE liver disease found simultaneously in SLE. The appearance of anti-P0 and liver dysfunction may predict onset of CNS lupus.

Association of interleukin 6 release from endothelial cells and pulmonary hypertension in SLE.

OBJECTIVE: To investigate how sera from 37 patients with systemic lupus erythematosus (SLE) stimulate interleukin (IL) 6 release from IL-1beta pretreated endothelial cells and compare these effects to those of sera from 16 normal controls. METHODS: Endothelial cells pretreated 18 h with IL-1beta (5 U/ml) were incubated 2 h with sera diluted 10-fold with phosphate buffered saline (PBS). IL-6 concentrations in endothelial culture supernatants collected after incubation were measured by ELISA. RESULTS: Compared with PBS, sera from controls and 24 patients with SLE suppressed IL-6 release from IL-1beta pretreated cells. However, sera from 13 patients with SLE augmented IL-6 release. Of note, sera from 5 patients with pulmonary hypertension induced the highest level of IL-6 release. IgG from control sera suppressed IL-6 release, whereas F(ab')2 did not. Both IgG and F(ab')2 from the sera of patients with SLE with pulmonary hypertension augmented IL-6 release from IL-1beta pretreated cells. CONCLUSION: IgG antiendothelial cell antibodies from patients with SLE may be associated with the pathogenesis of SLE and pulmonary hypertension.

Preferential binding with Escherichia coli hsp60 of antibodies prevalent in sera from patients with rheumatoid arthritis.

One hundred thirty-two patients with various connective tissue disorders, including 60 with rheumatoid arthritis (RA), had antibodies against human as well as Escherichia coli hsp60 in titers significantly higher than those of normal controls. There was a correlation between titers of antibody to human hsp60 and those to E. coli hsp60. Levels of antibodies against human and E. coli hsp60 were lower in joint fluids than in sera, indicating little production of antibodies in the joint. Antibodies affinity-purified with E. coli hsp60 bound strongly with the homologous hsp60, but weakly with human hsp60. However, antibodies affinity-purified with human hsp60 bound comparably with both E. coli hsp60 and human hsp60. Antibodies affinity-purified with Mycobacterium tuberculosis hsp65 bound to human hsp60 with a reactivity similar to the reactivity of those affinity-purified with human hsp60. The reactivity to the three hsp60 species was lost when sera were absorbed with E. coli hsp60, while the reactivity to E. coli hsp60 remained after extensive absorption with M. tuberculosis hsp65 or human hsp60. These results indicate that anti-hsp60 antibodies in patients with RA and other connective tissue disorders are raised by infection with intestinal microorganisms such as E. coli. They may represent another example of autoimmune responses triggered by antigenic mimicry of host proteins to microbes and suggest that the reactivity of antibodies from RA patients with M. tuberculosis hsp65 might have been a cross-reaction with the E. coli homologue.

Human monocyte-endothelial cell interaction induces synthesis of granulocyte-macrophage colony-stimulating factor.

BACKGROUND: Adhesion of monocytes to the endothelium is an initial step in the early stages of atherosclerosis and inflammation. Granulocyte-macrophage colony-stimulating factor (GM-CSF) stimulates a range of functional activities of monocytes, including regulation of monocyte adhesion and induction of cytokine production. We investigated in this study whether CM-CSF synthesis was induced by the direct cell-to-cell interaction between human monocytes and human umbilical vein endothelial cells (ECs). METHODS AND RESULTS: The expressions of GM-CSF mRNA and protein were analyzed by Northern blotting and ELISA, respectively. Coculture of monocytes and ECs induced the high levels of GM-CSF mRNA expression, whereas culture of ECs or monocytes alone or coculture of neutrophils with ECs induced no GM-CSF mRNA expression. A large amount of GM-CSF was secreted into the supernatant upon coculture of monocytes with ECs. The supernatant from the coculture markedly stimulated 02- release in neutrophils, and this effect was significantly inhibited by anti-GM-CSF antibody (Ab). Immunohistochemistry and in situ hybridization revealed that GM-CSF protein and mRNA were clearly detectable in both ECs and monocytes adhered to ECs but not in nonadherent monocytes. The GM-CSF production by the coculture was markedly inhibited by genistein and partially inhibited by Abs against interleukin-1 and tumor necrosis factor-alpha. CONCLUSIONS: The present results indicate that GM-CSF is produced by direct interaction between monocytes and ECs and suggest that GM-CSF produced locally by monocyte-EC adhesive interaction plays an important role in the pathogenesis of atherosclerosis and inflammation by modulating monocyte/macrophage functions in vivo.