Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by autoantibodies directed against various nuclear autoantigens, particularly against components of chromatin, such as doublestranded DNA (dsDNA), histones, nucleosomes, and ribonucleoproteins. Lupus nephritis, a severe clinical manifestation of SLE, is generally associated with high titers of antinuclear antibodies, especially antibodies against chromatin, and with immunoglobulin and complement deposits in the glomerulus. Historically, the formation of antibodies against dsDNA has been considered the serologic hallmark of SLE (1), and anti-dsDNA antibodies are found in the majority of SLE patients (2). In recent years, it has become clear that the nucleosome is the driving autoantigen in SLE (3), and some investigators claim that the presence of antinucleosome antibodies could serve as a better marker for SLE than anti-dsDNA antibodies, since the former are present in up to 90% of lupus patients (4,5). In SLE patients and in lupus-prone mice, nucleosomes can be found in the circulation; they are released from apoptotic cells and are present as a result of either disturbed apoptosis or a defective mechanism for clearing of apoptotic cells (3). Chromatin is basically defined as the entire complex of compacted DNA and associated proteins. The nucleosome is the fundamental unit of chromatin. The core nucleosome particle consists of 146 bp of dsDNA, wrapped twice around a histone octamer (2 copies of each of the core histones H2A, H2B, H3, and H4) (6). Antibodies against epitopes on the nucleosome, called nucleosome-specific antibodies, have no or very low reactivity against individual histones or naked dsDNA. The formation of nucleosome-specific antibodies precedes the appearance of anti-dsDNA and antihistone antibodies as a result of epitope spreading. Anti-dsDNA and antihistone antibodies bind to dsDNA and histones, respectively, but also to nucleosomes (7). A serious topic of debate for several decades involves the pathogenicity, and especially the nephritogenicity, of anti-dsDNA antibodies. DNA itself is not immunogenic, and immunization of mice with naked dsDNA leads to rapid removal, mainly through the liver. Nevertheless, anti-dsDNA antibody levels often show a high correlation with disease activity, especially in lupus nephritis, and elucidating their pathogenic properties is thus of great interest. Therefore, the targets of antidsDNA antibodies have been intensively studied to gain clues to mechanisms that could explain the pathogenicity of anti-dsDNA antibodies in SLE. With regard to their role in the development of lupus nephritis, many investigators postulate that antidsDNA antibodies localize in the glomerular basement membrane (GBM) due to direct cross-reactive binding to intrinsic glomerular antigens (mechanism 1), whereas, in our opinion, glomerular localization of anti-dsDNA antibodies is mediated via nucleosomes, which bind to heparan sulfate (HS) in the GBM (mechanism 2). These two hypotheses may not be mutually exclusive, and until recently, there was no unequivocal proof for either of the two proposed mechanisms. In this review, we will discuss the pros and cons of each of these mechanisms, taking into account recent pertinent observations by Supported by the Dutch Kidney Foundation (grant C05.2119), the PhD student program of the Radboud University Nijmegen Medical Centre, the Health and Rehabilitation Organization, Norway (grant 2001/2/0235), and Oslo Sanitetsforening and Skibsreder Tom Wilhelmsens Stiftelse. Casandra C. van Bavel, MSc, Johan van der Vlag, PhD, Jo H. Berden, MD, PhD: Nijmegen Centre for Molecular Sciences, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; Kristin A. Fenton, PhD: University of Tromso, Tromso, Norway; Ole P. Rekvig, MD, PhD: University of Tromso, and University Hospital of North Norway, Tromso, Norway. Address correspondence and reprint requests to Jo H. Berden, MD, PhD, Division of Nephrology (464), Radboud University Nijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail: J.Berden@nier.umcn.nl. Submitted for publication November 8, 2007; accepted in revised form April 14, 2008.