Tumour necrosis factor α (TNF-α), interleukin-6 (IL-6) and their soluble receptors (sTNF-α-Rp55 and slL-6R) serum levels in systemic lupus erythematodes

We investigated a possible association between serum concentrations of tumour necrosis factor α (TNF-α), interleukin-6 (IL-6) and their soluble receptors (sTNF-α-Rp55 and sIL-6R) using an enzyme-linked immunosorbent assay (ELISA) in 55 patients with systemic lupus erythematodes (SLE) and 16 healthy controls. We also examined a possible association between the serum levels of these peptides and SLE activity, as well as TNF-α and IL-6 concentrations and the levels of their soluble receptors. The median concentrations of TNF-α, sTNF-α-Rp55 and IL-6 were significantly higher in SLE patients than in normal individuals. In contrast, there was no difference between the serum level of sIL-6R in both groups. We found positive correlations between the serum concentrations of TNF-α and IL-6 as well as their soluble receptors and disease activity. There were also correlations between TNF-α and sTNF-α-Rp55 as well as IL-6 and sIL-6R serum levels in SLE patients but there were no such correlations in the normal control group. In conclusion, an increase in the serum levels of TNF-α, sTNF-α-Rp55 and IL-6 may become useful markers for SLE activity. Patients with SLE have sIL-6R serum concentration similar to that as in normal individuals. However, it correlates with disease activity and the level of IL-6.


Introduction
Systemic lupus erythematodes (SLE) is a disorder of generalized autoimmunity with multisystem organ involvement and autoantibodies against nuclear, cytoplasmic and cell surface antigens. The disease is characterized by 2 B cell activation and autoantibody formation. 1' However, increasing evidence indicates to a critical role of T cells, particularly CD4 + cells, in inducing B cell hyperactivity. 3 To explain the mechanisms responsible for immune dysregulation in SLE, cytokines have become the object of intensive studies. [2][3][4] Cytokines are a group of polypeptides synthesized by the host in response to different injuries. They may affect different cell functions and are involved in immunity and inflammatory response. Among the many cytokines, interleukin-1 (IL-1), interleukin-6 (IL-6), interferon-5, (IFN-,) and tumour necrosis factor-or (TNF-c0 seem to play the most important role in the pathogenesis of inflammatory and autoimmune diseases. [4][5][6] The intracellular signals for the response to cytokines are provided by cell surface receptors. 7'8 Cytokine receptors exist also in soluble forms, apparently deriving from proteolyric cleavage from the cell surface forms. 9 Several preliminary studies show conflicting data on the correlations between some inflammatory cytokines and their soluble receptors' levels in the serum of patients with SLE and disease activity. An initial report suggested that TNF<z was not elevated in the serum of lupus patients except during infection.10 Subsequently, elevated TNF-z levels in SLE patients compared with controls 11 and its correlation with disease activity was reported. Similarly, there have been several reports suggesting that the levels of IL-6 are elevated in active lupus.13-15 However, other reports do not support these observations. 16 The biological role of soluble cytokine recep-(C) 1996 Rapid Science Publishers Mediators of Inflammation Vol 5 1996 435 tors is not fully elucidated. They may modify immunosuppressive agents. Fifty-one patients ligand concentrations by serving as stabilizing were treated with steroids and two of them binding proteins, may downregulate membrane with azatioprine for some time during the receptor number as a mechanism of genesis, course of their disease, but 20 of them had not and may specifically inhibit ligand-receptor asso-been treated for at least 4 weeks before the ciation in the extracellular space. 16 investigation of cytokines.
Two receptors of different molecular struc-We included in the study patients with active ture have been recognized for TNF<z: type I or and inactive disease. Disease activity was scored TNF Rp55, of 55-60 kDa and type II or TNFduring the visit to the outpatient clinic accord-Rp75, with 75-80 kDa molecular mass. 7 Soluble ing to the method proposed by Liang et al. in forms of these two TNF receptors (sTNF-Rs) our modification. 26 Each patient was assessed have been identified in serum and urine. 9 '17 on two separate occasions 2-4 weeks apart. Neutrophils, activated T-cells and monocytes The system Systemic Lupus Activity Measure are the main sources of sTNF<z-Rs) 7 Their (SLAM) includes 24 clinical manifestations and concentration in serum increases significantly in eight laboratory parameters. inflammatory and cancer diseases. 7'1 Both Parameters of immune function were not types of sTNF-(zRs have been reported to be included. Maximal score points in this system is elevated in patients with SLE. [19][20][21] The sTNF-84. We assumed the score of 0-10 points for (-Rs levels correlate with disease activity and inactive disease, and a score of over 10 points the impairment of renal function may contri-for active disease. Our group of patients inbute to a rise of their serum concentrations. 19 eluded 13 inactive patients and 42 active These sTNF-(z-Rs appear to be biologically funcdisease patients ( Table 1). tional because they inhibited TNF-(z action in The control group of 15 healthy volunteers cytotoxicity assays. 21 was also studied. They were 14 women and one The cellular IL-6R complex consists of two man, aged from 25 to 50 years (median 39 different proteins, an 80-kDa ligand binding years).
glycoprotein (IL-6R) and a 130-kDa protein (gp 130) involved in cellular signal transduction. 22 Laboratory tests MOllberg et al. have recently shown that two subunits of the IL-6R complex are proteolyti-On the day of blood sampling for TNF-, IL-6 cally cleaved and released from the cell as and their soluble receptors the following laborasoluble receptor proteins. 23 '24 Their concentra-tory parameters were analysed: complete blood tions seem to be elevated during infections, cell count (CBC), erythrocyte sedimentation inflammatory and neoplastic diseases. In conrate (ESR), blood urea nitrogen and creatinine trast to other soluble cytokine receptors, the levels, fibrinogen level, partial thromboplastin slL-6R together with IL-6 acts agonistically on cells that express gp 130. To our best knowledge, the serum level of slL-6R in SLE patients Table 1. Clinical characteristics of patients with systemic has not been investigated so far. lup27us erythematodes (symptoms according to Liang et In the present study we measured the serum a/. concentrations of TNF-(z, and IL-6 and their Symptom soluble receptors (sTNF-(z-R55 and slL-6R) in 55 patients with SLE using ELISA assay. We also Total correlated the serum levels of these proteins Active with disease activity. Serum sampling Venous blood samples for TNFand TNF receptor determinations were collected at the time of clinical assessment into pyrogen-free tubes, and centrifuged within 20 min after being allowed to clot at--4C for 1 h at 2000g for 10 min. The serum obtained was divided into aliquots and stored at -25C until assayed. TNF-(, IL-6 and sTNF-(x-Rp55 slL-6R were assayed by specific commercially available, enzyme-linked (ELISA) assay kits (Quantikine, R&D Systems Inc., USA) in accordance with the manufacturer's protocols.
In each assay, the appropriate recombinant human cytokine was used to generate the standard curve. Sensitivity of the assay for TNFwas 0.5 pg/ml; for sTNF-(x-Rp55, 7.8 pg/ml; for IL-6, 0.3 pg/ml. Serum for IL-6R concentration measurement was diluted 40 times and its level was measured between 7.8 and 500 pg/ml. The concentrations of cytokines and soluble receptors in the samples were determined by interpolation from the standard curve.

Statistical analysis
Differences in parameters between groups were evaluated with Student's t-test. The chi-squared and Fisher's exact tests were used to analyse the relationship between cytokines and soluble receptors serum values and SLE activity.
Internal distribution data were analysed by Student's t-test. Correlations were evaluated using the Spearman rank-sum correlation coefficient and linear regression calculated with the least-squares method.
Results are presented with R 2 coefficients.
Comparisons and correlations were considered significant when p < 0.05.  1). There was also a positive correlation between TNF-(x and sTNF-(x-Rp55 concentrations in the serum of SLE patients (Fig. 2) and a lack of such correlations in normal controls (data not shown). The serum level of IL-6 was higher in SLE  Table 2). However, there was a positive correlation between IL-6 and slL-6R concentrations and SLE activity (Figs 1 and 2). We also found a significant positive correlation between the levels of IL-6 and slL-6R in SLE patients (R 2--0.1954, p<O.O01) and lack of such correlations in the normal group (data not shown). We also analysed the relationship between serum concentrations of TNF-e and IL-6 as well as sTNF-tx-Rp55 and slL-6R (Fig. 2). We observed 4:38 Mediators of Inflammation Vol 5 1996 positive correlations between their parameterg (R 2 0.7008, p < 0.001 and R 2 0.3683, p < 0.001, respectively).

Discussion
The objective of our study was to evaluate the concentrations of two cytokines, TNF-tx and IL-6, whose role is inflammatory and immunologic processes is fairly well known, as well as their soluble receptors, sTNF<z-Rp55 and IL-6R, in patients with SLE. In the study, the correlation between concentrations of these peptides and the activity of SLE as well as the incidence rate of particular clinical and laboratory symptoms of the disease was analysed. We have demonstrated that the concentration of both TNF-cz and IL-6 in the serum of SLE patients is higher than in healthy individuals, and that it displays a  [11][12][13][14][15] However, in contrast tients despite the fact that in these cells the with our studies, Gordon and Emery did not mRNA is higher for TNF-cz than in healthy 27 observe any correlation between TNF-ct concenindividuals. Moreover, monocytes in SLE patration and SLE activity, 5 and Metsariune et al. tients can spontaneously secrete TNFin in did not find IL-6 concentrations to be higher in vitro culture, with immunologic complexes controls. The being a strong stimulator of its production by SLE patients than in healthy I6 above differences may arise due to a different these cells. 28 In SLE, a higher TNF-ct concentrasensitivity and specificity of the methods aption was also found in the disease affected plied to assay cytokines (ELISA and bioassay), tissues compared with healthy tissues. It should and to the number of patients under investiga-also be added that excessive TNF-ct production tion. may be conditioned genetically. It may be It seems justified to assume that both TNF-ct associated with the presence of TNF-tx gene, and IL-6 play a significant role in the pathogen-and this presence can in consequence condition esis of SLE. In vitro studies show a disturbed the development of autoimmune diseases, inproduction of TNF-ct in SLE. Malave et al. cluding SLE. The above observations may ac-count for a higher TNF-c concentration in the serum of patients with an active form of SLE, as well as point to a significant role of this cytokine in the pathogenesis of the disease. In SLE, anomalies were also observed in the production process of IL-6 and the response of target cells to its activity. Lymphocytes T and B, as well as monocytes of SLE patients produce IL-6 in greater quantities than their equivalents in healthy individuals; this intensification in the production may be induced by immunologic complexes.13 Another observation was that lymphocytes B obtained from SLE patients were more sensitive to the effect of IL-6 in comparison with lymphocytes from healthy controls; thus, this cytokine may be for them an autocrinic growth factor. 29 Of significance may be the observation that in SLE patients there is no evidence of subpopulation of cells specifically inhibiting IL-6 expression, and that the majority of lymphocytes B exhibit spontaneous expression of the receptors for this cytokine.
In our studies, the sTNF-x-Rp55 concentration in serum was higher in SLE patients than in healthy controls and correlated positively with the activity of the disease, which is in agreement with observations by other authors. [19][20][21]31 However, we did not observe differences in the sIL-6R concentration between SLE patients and healthy controls, although the concentration of this cytokine corrrelated positively with the disease activity.
The studies conducted demonstrate that sTNF<z-Rp55, along with TNF<z and IL-6, can be regarded as a marker of SLE activity, whereas sIL-6 cannot serve as such. Another marker of the disease activity is soluble receptor IL-2 (slL-2R, CD25), the concentration of which, when monitored, can warn of an exacerbation in the course of SLE. 32 Soluble receptors for TNF-cz, apart from sIL-2R, neopterins, and intracellular adhesive molecule 1 (IL AM-I), are considered the best markers of immunologic activity. 17 Determination of the receptors investigated even has a certain advantage over TNF-x concentration measurements, considering that this cytokine is quickly eliminated from the circulation and can be thus difficult or impossible to determine by some methods, especially with bioassays. However, in our study we used the ELISA method by which it is possible to determine the fraction of TNF-x that is associated with soluble receptors. 33 The biological and pathophysiological role of soluble receptors for cytokines is as yet little known. If they are present in serum at a high concentration, they can inhibit the activity of cytokines. 9  It is not fully explained yet if the increase in sTNF--R concentration, observed in our SLE patients, is sufficient to alter the biological activity of TNF<z. In individuals exposed to the endotoxin activity, such effect can be then achieved at the sTNF-c-R concentration of 5 ng/ml, which value is considerably higher than in our patients. 34 At lower concentrations, sTNF-x-Rs function as carriers for their ligand and protect it from proteolysis, constituting a reservoir of active or potentially active cytokine. 35 It has been demonstrated that both TNF-z and its production stimulating factors cause at the same time a split-off of soluble receptors, 17 an evidence of which can be the positive correlation between TNF-cz and sTNF-Rs observed in our patients. Therefore, the concentration of these receptors in serum is regarded as an indicator of the activity of the whole TNFsystem, and when increased in the result of the system stimulation, it persists longer then high TNF-cz concentration. 17 Our studies indicate that determination of sIL-6R concentration, as opposed to sTNF-cz-Rs, is of less importance in SLE. The concentration of the former receptor did not significantly differ in SLE patients and in healthy controls. We have demonstrated, though, a positive correlation between the concentration of slLo6R and of IL-6, as well as between slL-6R concentration and disease activity. The results obtained are difficult to interpret since in pathological conditions characterized by high IL-6R concentration high slL-6R concentration was also observed. 9'36 The pathophysiological role of sIL-6R has not yet been fully determined. Nevertheless, IL-6 is known to bind in physiological conditions with the chain of this receptor, and the complex formed binds them with the endothelial chain fl (gpl30) of the cellular receptor, and only then transduction of the signal in the cell occurs. 37 In in vitro studies on myeloma cell lines it has been demonstrated that sensitivity of these cells on IL-5 activity in the presence of sIL-6R increases ten-fold. 38 We have shown that the concentration of TNF-z, IL-6 and sTNF<z-Rp55 was higher in SLE patients than in healthy individuals, and that these peptides correlate positively with the activity of the disease. Although sIL-6R concentration in SLE patients did not significantly differ, the level of this receptor indicated a positive correlation with the disease activity and IL-6 concentrations.