Histones are a set of highly cationic proteins essentially involved in the binding and compaction of DNA in the cell nucleus chromatin (1-3). In mammals, five subclasses of histones exist: H1 (f1), H2a (f2a2), H2b (f2b), H3 (f3), and H4 (f2a1) (4,5), whereby four histones form an octameric core protein aggregate (H2a, H2b, H3, H4) around which a 145-bp DNA molecule is wrapped, forming the core particle followed by a variable (-55 bp) DNA region, complexed with H1 (6). In the cell nucleus, histones and DNA are present in similar quantities .These components of the cell nucleus are two of the most important antigens for autoantibody formation in SLE (7-12), and it has been reported that the titers of both groups of autoantibodies correlate well with disease activity (13-16).A prominent manifestation oforgan involvement in SLE patients is the occurrence of immune complex (IC)' glomerulonephritis . After the original report by Koffler et al. (17), attention has been focussed on the role of DNA-anti-DNA IC for >20 yr. Two proposals have been widely discussed in connection with the glomerular deposition of DNA-anti-DNA complexes: the now unfashionable notion that preformed soluble ICs can locate in the glomerular filter (18); and second, an affinity of DNA for the glomerular basement membrane (GBM) in vivo has been suggested (19-23), which could then initiate in situ IC formation. Extensive studies in animals on the kinetics and clearance of circulating free or immune complexed DNA (24-27) in regard to size and strandedness of DNA, however, revealed rapid nonrenal uptake and clearance. Prolonged DNA persistence is difficult to explain by accepted concepts; new proposals are needed (28). Contradictory findings on the occurrence of free or IC-bound DNA in sera from normal individuals and SLE patients have been reported (for review see reference 29). There are a number of technical problems involved here and it is now accepted that DNA-anti-DNA immune complexes represent only a minor part of the ICs found in lupus patients, and that increased levels ofcirculating DNA are found in a variety of conditions in which lysis ofcells occurs . Preliminary studies by Fournie (29) suggest that nucleosomes, which consist ofDNA This work was supported by the Deutsche Forschungsgemeinschaft, Ba 907/1-1.

These results suggest that increased anti-zymogen antibody titers are the best available marker for streptococcal infection associated with acute glomerulonephritis.

The influence of charge and size on antigen binding to the rat glomerular basement membrane (GBM) was investigated. Chemically cationized ovalbumin, human serum albumin (HSA), human immunoglobulin G (Hu IgG), horse spleen ferritin and human immunoglobulin M (Hu IgM) were injected into rats intravenously. By immunofluorescence significant glomerular binding occurred when the pI exceeded a threshold value of 8.5 to 9.5. At a given pI antigen binding increased with molecular size. Cationized Hu IgM bound only weakly to the glomerular capillary wall, presumably excluded due to size. Subepithelial immune deposits were formed only when antibody was injected subsequently. Detailed electron microscopic studies on in situ formation of immune complexes were performed using cationized horse spleen ferritin. Early on subendothelial deposits were very marked, giving way to subepithelial deposits with time. Under the conditions employed, it appears that deposits can be formed directly at the subepithelial locus but that complexes are also formed subendothelially, dissociating into free molecules or small complexes and then migrating through the lamina densa and reforming.

The maximum tumour thickness is the most important prognostic factor in malignant melanomas of the skin. However, the clinical outcome of thick nodular melanomas remains unpredictable. Therefore, we investigated possible prognostic markers in this subset of melanomas. From a melanoma data base, 12 patients with thick (> 3 mm) stage I nodular melanomas of the skin were identified, who were still without signs of progression after five years of follow-up. These tumours were compared to randomly selected series of 12 cases, who did not survive the first five years after removal of the tumours. We performed immunostaining for the p53-protein and the proliferation associated Ki-67-antigen. For quantification of immunostaining the tumours were entirely scanned. In addition, all tumours were investigated for any differences with conventionally applied prognostic features: the tumour thickness: the level of invasion; the prognostic index (tumour thickness multiplied by mitotic count); and the mean volume-weighted mean nuclear volume. We demonstrated significant differences between survivors and non-survivors exclusively in respect of the staining indices for p53 and Ki-67 (P < 0.03 and 0.02, respectively). With both antibodies the tumours of survivors showed lower counts as compared to non-survivors survivors. However, within both groups we found no significant correlations between the p53- and Ki-67-staining results. We conclude that immunostaining for p53-protein and Ki-67-antigen is helpful to identify individuals with thick nodular melanomas who are at risk of metastatic disease.