The structure of the octameric histone core of the nucleosome has been determined by x-ray crystaflography to a resolution of 3.1 A. The histone octamer is a tripartite assembly in which a centrally located (H3-H4)2 tetramer is flanked by two H2A-H2B dimers. It has a complex outer surface; depending on the perspective, the structure appears as a wedge or as a flat disk.The disk represents the planar projection of a left-handed proteinaceous superhelix with -28 A pitch. The diameter of the particle is 65 A and the length is 60 A at its maximum and -10 A at its Ium extension; these dimensions are in agreement with those reported earlier by Klug et al.

Su ibity to systec lupus erythematosus has been unequivocally e d to be an inherited trai, but the exact genes and how they confer susceptibility n largely unknown. In this study of (NZB x NZW)F2 inte mice, we used linkage analysis ofmarkers covering >90% ofthe automal genome and idenified eit ptl ld chomosomes 17,(4)(5)(6)(7)18,1,11, pectively) ated with antichromatin autoantibody production, gm nphris, and/or it. Only one locus, the major his patibt complx, was linked to all three traits. Two other loci were ed with both glomerulonephitis and mortalt, whereas the remaining loci were linked to one of the above traits. Two adial loci (Sbwl and -2) that conibut to p m were also iden- Among the murine lupus strains, the (NZB x NZW)F1 (BWF1) hybrid has clinical features most closely resembling human SLE with markedly accelerated disease compared with parental strains and a striking female predilection (2). Conventional genetic studies indicate that each of the parental strains contributes at least one or two genes (3, 4), one of which is linked to the MHC locus, with heterozygosity (H-2dz) conferring maximal susceptibility (5,6).The recent identification of thousands of polymorphic dinucleotide repeats (microsatellites) that can be used to create dense linkage maps between inbred strains (7, 8) has made it feasible to systematically search the entire mouse genome for susceptibility gene loci. We have used this approach to map the genes predisposing to early disease in the BWF1 hybrid and report the identification of several loci predisposing to mortality, glomerulonephritis (GN), antichromatin antibody production, and splenomegaly.MATERIALS AND METHODS Mice. NZB/BlScr, NZW/LacScr, BWF1, and (NZB x NZW)F2 (BWF2) intercross mice were bred and maintained in our animal colony. Female mice from 6 mo of age were examined daily for disease, bled monthly for sera, and sacrificed at either 1 yr of age or earlier if moribund.Phenotyping of Mice. Autopsies and histologic examinations were done as described (9). Severity of GN was graded from 0 to 4+ (9), and mice were considered to have severe GN ifthey were 4+ at 12 mo or .3+ ifmice eitherdied earlier or had anasarca. Survival comparisons and cumulative antichromatin antibody levels were analyzed with the generalized Wilcoxon test. Comparisons of spleen size were done with the Mann-Whitney U test for small sample sizes and the Student's t test for larger samples. ELISA for chromatin was done as described (10). For linkage analysis, BWF2 mice with antichromatin antibody levels of OD > 0.5 by 11 mo were considered positive.Cbromosoma Markers and Genotyping of Mice. Chromosomal markers consisted of simple-sequence-length polymorphisms (SSLPs) identified by PCR (refs. 7,8

To gain insight into the mechanisms of autoantibody induction, sera from 40 patients with systemic lupus erythematosus (SLE) were tested by ELISAs for antibody binding to denatured individual histones, native histone-histone complexes, histone-DNA subnucleosome complexes, three forms of chromatin, and DNA. Whole chromatin was the most reactive substrate, with 88% of the patients positive. By chi-square analysis, only the presence of anti-(H2A-H2B), anti-[(H2A-H2B) -DNA], and antichromatin were correlated with kidney disease measured by proteinuria > 0.5 g/d. SLE patients could be divided into two groups based on their antibody-binding pattern to the above substrates. Antibodies from about half of the patients reacted with chromatin and the (H2A-H2B) -DNA subnucleosome complex but displayed very low or no reactivity with native DNA or the (H3-H4)2-DNA subnucleosome complex. An additional third of the patients had antibody reactivity to chromatin, as well as to both subnucleosome structures and DNA. Strikingly, all sera that bound to any of the components of chromatin also bound to whole chromatin, and adsorption with chromatin removed 85-100% of reactivity to (H2A-H2B)-DNA, (H3-H4)2-DNA, and native DNA. Individual sera often bound to several different epitopes on chromatin, with some epitopes requiring quaternary protein-DNA interactions. These results are consistent with chromatin being a potent immunogenic stimulus in SLE. Taken together with previous studies, we suggest that antibody activity to the (H2A-H2B) -DNA component signals the initial breakdown of immune tolerance whereas responses to (H3-H4)2-DNA and native DNA reflect subsequent global loss of tolerance to chromatin. (J. Clin. Invest. 1994. 94:184-192.)

No abstract