REVIEW doi: 10.1111/j.1365-3083.2012.02728.

x
..................................................................................................................................................................

Antinuclear Antibodies in Rheumatic Disease: A

Proposal for a Function-Based Classification
D. S. Pisetsky

Medical Research Service, Durham Veterans
Administration Medical Center, Durham, NC,
USA; and Duke University Medical Center,
Department of Medicine, Durham, NC, USA
Received 18 April 2012; Accepted in revised
form 21 May 2012
Correspondence to: Dr D. S. Pisetsky, 151G
Durham VA Medical Center, 508 Fulton St.,
Durham, NC 27705, USA. E-mail: dpiset@
acpub.duke.edu
Abstract
Antinuclear antibodies (ANAs) are a diverse group of autoantibodies that bind
macromolecular components of the cell nucleus. While some ANAs occur in
normal individuals, others are expressed almost exclusively in patients with
rheumatic disease and serve as markers for diagnosis and prognosis. Despite
the clinical associations of ANAs, the relationship of these antibodies to spe-
cific disease manifestations is often unknown because the target antigens are
intracellular molecules that are ubiquitously expressed. In systemic lupus ery-
thematosus, the role of ANAs in disease manifestations is better understood,
especially for antibodies to DNA and related nucleosomal antigens. These anti-
bodies can promote nephritis by the formation of immune complexes that are
deposited in the kidney. In addition, anti-DNA, along with antibodies to
RNA-binding proteins such as anti-Sm, can induce non-specific immune
abnormalities based on the induction of type interferon 1 by plasmacytoid
dendritic cells. Despite ANA expression in rheumatic disease, studies in ani-
mal models of inflammation and tissue injury indicate that antibodies to cer-
tain nuclear molecules such as HMGB1 have protective effects. Together, these
considerations suggest a function-based classification of ANAs based on their
expression in normal and autoimmune individuals as well as their capacity to
induce or attenuate immunological disturbances. This classification provides a
framework to elucidate the serological features of rheumatic disease and
the often uncertain relationship between ANA expression and disease
manifestations.

Antinuclear antibodies (ANAs) are a diverse group of

autoantibodies that target macromolecular components of
the cell nucleus. These antibodies occur commonly in the
sera of patients with autoimmune and rheumatic disease
and bind proteins, nucleic acids and complexes of pro-
teins and nucleic acids. As serological markers, these
antibodies have diagnostic and prognostic significance
and, in some instances, have been directly linked to dis-
ease manifestations. Because of their high association
with various rheumatic diseases, ANAs have been the
focus of intense study. Indeed, there is probably more
information about the generation and properties of ANAs
than any other group of antibodies, normal or abnormal,
in all of medicine. This article will review the expression
of ANAs and their diverse functions.
ANA expression in normal and aberrant immunity
While ANAs can be linked to disease, they, nevertheless,
can be found commonly in the sera of normal individuals
in the absence of obvious signs of rheumatic disease.
Indeed, a recent study indicated that 13.8% of the nor-
mal population was ANA positive using sensitive assays
with the frequency higher in women than men [13].
The basis of this expression is unknown but the wide-
spread expression of putatively abnormal specificities
raises question about the techniques for ANA detection
as well as the potential physiological or pathological
function of these antibodies. Thus, the frequent detection
of ANAs may represent the vagaries of current assays,
which allow detection of low avidity antibodies. Because

 2012 The Author.

Scandinavian Journal of Immunology  2012 Blackwell Publishing Ltd. 223
224 Classification of Antinuclear Antibodies
..................................................................................................................................................................
many nuclear antigens are charged molecules, they could
facilitate cross-reactive binding of normal antibodies
without denoting any aberrant expression [4]. This situa-
tion is especially true for IgM antibodies, so-called natu-
ral autoantibodies, which may have broad reactivity and
mediate clearance of material from infection as well as
debris cellular debris [5, 6].
Another explanation for the high frequency of ANAs
in the general population could relate to the presence of
significant immunoregulatory disturbances in normal
individuals as well as a large pool of at-risk individuals
who are awaiting some event to trigger their disease. As
studies on the genetics of rheumatic disease indicate, the
human population harbours a huge number of gene poly-
morphisms, which appear to contribute to a diathesis for
autoimmunity [7]. These polymorphisms most likely
result from selection in evolution to promote host
defence and lead to accentuated immune responses at the
level of various immune cell populations. When present
in limited numbers or in certain combinations, such
polymorphisms could lead to serological disturbances in
the absence of other disease manifestations. In general,
the specificities of ANAs in normal individuals are
unknown, making it difficult to evaluate the mechanisms
of these responses.
While many serologically positive individuals will
never develop an autoimmune disease, others may be in a
pre-autoimmune state in which ANA positivity is har-
binger of impending disease. As now recognized, ANA
positivity as well as the expression of more specific ANAs
can precede the onset of autoimmunity by many years; in
these patients, an increase in the number of expressed
ANAs can herald the onset of frank disease [8]. Distin-
guishing among patients who are destined for disease and
those in whom ANA positivity is innocuous or incidental
is clinically challenging and can complicate the evalua-
tion of individuals with non-specific or vague symptoms.
A test to identify normal ANAs would be an important
advance and could save considerable worry and work-ups.
An issue further complicating serological assessment
concerns events during infection. Thus, patients with dis-
eases such as tuberculosis, malaria and leprosy may have
an increased expression of ANAs; the specificity of these
autoantibodies is often unknown but appears to differ
from those found in patients with known autoimmune
diseases [9, 10]. With infections from viruses, such as
EpsteinBarr virus, induction of ANAs may occur espe-
cially with patients at risk of systemic lupus erythemato-
sus (SLE), as underlying immune abnormalities may
promote autoreactivity in response to a viral antigen. In
this case, the antibodies induced during virus infection
would correspond to classic ANAs [11]. The development
of SLE in these patients could occur subsequently with
the expression of cross-reactivity to a viral antigen
predicting eventual autoimmunity.
D. S. Pisetsky
These considerations suggest that ANAs are not uni-
form in their properties and can be characterized by their
expression in the normal as well as disease populations.
Those antibodies that occur in the disease setting (except
infection) can be termed pathological because they are
present almost exclusively in patients with autoimmunity
or a pre-autoimmune state. Further distinction comes on
the basis of disease association. Thus, some ANAs are
highly associated with particular diseases (e.g. anti-DNA
and lupus, anti-Scl70 and progressive systemic sclerosis)
so much so that they can serve as markers for diagnosis
or classification. Such antibodies are rarely if ever found
in the normal population although this finding may
depend on the extent of the screening and assay used. In
contrast, some ANAs have a more extensive expression
among various rheumatic diseases without a strict associ-
ation with a particular entity. Thus, antibodies to Ro La
occur commonly in patients with lupus, rheumatoid
arthritis and Sjogrens syndrome. The clinical utility of
these antibodies as markers depends on clinical findings
in patients (e.g. sicca symptoms and findings of dry
eyes).
Another layer of complexity concerns their relationship
between the expression of ANAs and specific disease mani-
festations and hence their value as clinical markers for
prognosis as well as probes for elucidating disease patho-
genesis. While many antibodies have been linked to spe-
cific clinical manifestations, the mechanisms by which
these antibodies promote pathogenesis are obscure. The
target antigens of ANAs are highly conserved and ubiqui-
tously expressed; furthermore, as nuclear components,
these molecules should be shielded from autoantibody
interaction. Understanding how binding of such autoanti-
bodies could contribute to manifestations such as intersti-
tial lung disease or myositis has eluded investigators; it is
formally possible, however, that the autoantigens that are
targeted in the tissue during disease differ from those that
are detected in an ANA assay. In this instance, the ANA
may represent a cross-reaction that, while useful clinically,
does not provide insight at the level of pathogenesis. In
any event, an antibody that can cause disease can be termed
as pathogenic. This determination frequently depends on
vivo models in which a disease finding can be reproduced
by induction of an antibody by immunization or transfer of
sera or a monoclonal antibody preparation.
The serology of SLE
In contrast to many rheumatic diseases where the role of
ANAs is unknown, SLE illustrates clearly the role of
ANAs in disease pathogenesis as well as the heterogene-
ity of ANA with respect to their function in disease
induction and expression over time. SLE is a proto-
typic autoimmune disease characterized by multisystem
involvement in association with a broad array of ANA
Scandinavian Journal of Immunology, 2012, 76, 223228
D. S. Pisetsky
..................................................................................................................................................................
responses; of these ANA, anti-DNA and anti-Sm anti-
bodies represent criteria for the classification of patients
with lupus [12, 13]. While both are valuable and
informative biomarkers, anti-DNA and anti-Sm show
fundamental differences in specificity and expression.
Antibodies to DNA are directed to nucleic acid determi-
nants, whereas anti-Sm antibodies (and the closely related
anti-RNP antibodies) are directed to the protein compo-
nents of RNAprotein complexes called snRNPs. Most
importantly, anti-DNA antibodies can show marked vari-
ation in expression over time, with levels frequently cor-
related with the activity of nephritis. In contrast, levels
of anti-Sm are frequently static over the course of disease,
showing little variation with therapy, including cytotoxic
agents that can have a powerful impact on B cell popula-
tions [14]. Since anti-Sm antibodies show persistent
expression over time, it has been difficult to link these
antibodies with clinical manifestations, which may come
and go as the disease progresses.
The differences in the expression of antibodies over time
most likely relates to the B cells involved in their produc-
tion, with the persistent expression of antibodies to the
RNA-binding proteins such as Sm, RNP, Ro and La likely
reflecting their production by long-lived plasma cells. In
contrast, anti-DNA antibodies show highly variable levels
of expression and frequently disappear with treatments
including corticosteroids and conventional immunosup-
pressive agents. These findings suggest that anti-DNA
antibodies either arise repetitively de novo with flares or are
produced by memory B-cell populations, which remain
amenable to immunomodulatory agents as well changes in
disease activity. The basis of this difference is unknown
but has important implications for the use of different
ANAs as a biomarker in clinical assessment, especially
novel therapies, as well as a probe for underlying disease
mechanisms [15, 16].
While anti-DNA antibodies are frequently viewed as a
discrete antibody population, they are in fact a subset of
a broader array of antibodies that bind to nucleosomes
that are the essential building block of chromatin. These
antibodies include antibodies to DNA (both single and
double stranded), histones and DNAhistone complexes.
These antibodies frequently show concomitant expression
in a pattern known as linkage. This pattern mostly likely
reflects the role of nucleosomes as the driving antigen in
lupus, with various nucleosomal antigens (e.g. DNA and
histones) representing epitopes of a larger structure [17,
18]. As the antibodies show similar patterns of expres-
sion, it can be difficult to identify the actual contribution
to disease of any specific antibody subpopulation. Thus,
the variation of anti-DNA antibody levels with disease
activity does not necessarily imply that anti-DNA anti-
bodies themselves cause nephritis. Rather, this association
may indicate a role of other members of the family of
anti-nucleosomal antibodies.
 2012 The Authors.
Scandinavian Journal of Immunology  2012 Blackwell Publishing Ltd.
Classification of Antinuclear Antibodies 225
In view of their specificity, anti-DNA antibodies
could promote nephritis by a number of mechanisms,
although immune complex deposition appears to be the
major contributor to immunopathogenesis. Thus, anti-
DNA levels frequently rise in conjunction with falls in
complement levels, indicative of complement consump-
tion. Furthermore, anti-DNA can be isolated in enriched
form from kidneys, while, in mouse models, administra-
tion of anti-DNA antibody preparations, either poly-
clonal or monoclonal, can lead to renal deposition
presumably by forming immune complexes with DNA in
the circulation [13]. This DNA, which most likely exists
in the form of nucleosomes, appears to be the product of
dead and dying cells; the levels of this nucleosomal mate-
rial may be increased in lupus because of either aberrant
cell death or impaired clearance [19]. The use of normal
mice as recipients for transferred anti-DNA may not
reveal with full spectrum of pathogenic antibodies if the
supply of this antigen is inadequate.
In the formation of immune complexes, the interac-
tion of DNA and anti-DNA may occur either in the cir-
culation or in the kidney because nucleosomes have a
predilection for glomerulus; nucleosomal antigens may
also be generated locally in the kidney, perhaps reflecting
local deficiency of DNase [20]. In addition to immune
complexes, anti-DNA antibodies may function to pro-
mote nephritis by direct binding to sites in the glomeru-
lus [21]. Antibodies that cause nephritis (by any
mechanism) can be termed nephritogenic. Table 1 lists
immunochemical properties of antibodies associated with
nephritogenicity.
While a role of anti-DNA in nephritis is clear, dis-
crepancies exist in the clinical setting because many
patients with lupus may lack renal disease despite high
levels of anti-DNA antibodies; a similar situation occurs
in murine model because some strains with high levels of
anti-DNA lack nephritis. These findings are notable and
suggest at least two possible explanations. Thus, while
many anti-DNA antibodies may cause nephritis, nephri-
togenicity may not invariable among antibodies of this
kind but rather depend on fine specificity for DNA anti-
gen, isotype or avidity. Depending on certain immuno-
chemical properties, anti-DNA may fail to assemble
complexes or fix complement to incite nephritis [22].
A second explanation for the failure of anti-DNA to
promote nephritis relates to the requirement of DNA
Table 1 Properties of nephritogenic antibodies.
Isotype
Fine specificity
Avidity
Complement fixation
Glomerular binding
Disease induction in transfer models in mice
226 Classification of Antinuclear Antibodies
..................................................................................................................................................................
antigen to assemble an immune complex. While it is
assumed that sufficient DNA is commonly present in the
blood or tissue to bind anti-DNA, that situation may
not be true. Perhaps immunologically relevant DNA may
exist only some of the time, with extracelluar expression
occurring only during conditions such as trauma or infec-
tion or dramatically impaired clearance such as DNase or
complement deficiency. It is possible that, just as anti-
bodies may be nephritogenic, so too may be DNA. The
features of DNA that may allow assembly into complexes
may relate to the extent of free DNA on relevant anti-
genic structures to allow binding of nephritogenic
antibodies. Features of DNA that influence anti-DNA
binding may relate to strandedness, conformational
mobility and interaction with serum proteins that may
prevent antibody binding [23, 24]. At present, there is
little information on the antigenicity of DNA that is the
blood.
As now recognized, anti-DNA antibodies, like anti-
bodies to anti-RBPs, can cause disease by other mecha-
nisms. Thus, anti-DNA antibodies may cross-react with
non-nuclear antigens such as the NMDA (N-methyl-D-
aspartate) receptor and induce neuropsychiatric distur-
bances such as depression and cognitive impairment by
receptor interaction and excitotoxicity [25, 26]. Anti-
DNA antibodies may also influence the overall state of
vascular and immune system by forming immune com-
plexes that stimulate the production of type 1 interferon
and other cytokines by dendritic cells, especially plasma-
cytoid dendritic cells (PDCs). This stimulation involves
internal nucleic acid receptors, including the toll-like
receptors (TLRs) such as TLR 7 and TLR9 as well as
non-TLR nucleic acid sensors [2729]. In this scenario,
ANAs provide a vehicle to deliver DNA and RNA into
the cytoplasm of cells where they exert immunostimula-
tory activity. In addition to internal nucleic acid recep-
tors, stimulation of PDCs by immune complexes can
involve the Fc receptor as well as receptor for advanced
glycation end-products (RAGE). The role of RAGE
results from the presence of HMGB1 in the complexes
[29]. HMGB1 is an abundant non-histone nuclear protein
that represents a prototype alarmin that can stimulate
immune responses by itself or in conjunction with mole-
cules such as TLR ligands (e.g. LPS) and cytokines such
as IL-1 [3032]. Since HMGB1 can bind DNA, it may
be present in immune complexes depending on their ori-
gin [29]. At present, there is not a term to identify those
ANAs that can form complexes that stimulate interferon
production. Interferogenic is possible but awkward.
While anti-DNA antibodies can form immune com-
plexes that deposit in the tissue or induce cytokine pro-
duction, the actual specificities of these antibodies that
exert these activities may differ from each other as may
the antigen in the complex. Furthermore, just as in the
case of nephritis, only certain anti-DNA antibodies may
D. S. Pisetsky
have the functional properties to form complexes that
drive cytokine production. In this regard, antibodies to
RBPs appear able to form complexes that stimulate cyto-
kine production although their role in renal disease as
well as ability to fix complement remains unclear. The
static expression of anti-RBPs, in contrast to the recipro-
cal fluctuation of anti-DNA and complement, makes it
difficult to discern the ability of anti-RBPs to fix com-
plement [14]. Because anti-RBPs have not been associated
with nephritis, this issue has not been extensively
explored although data suggest that these antibodies can
form circulating immune complexes [33, 34].
Beneficial autoantibodies?
Antinuclear antibodies are commonly viewed as both
pathological and pathogenic antibodies that can cause
disease manifestations, with studies on their association
with disease conceptualized in terms of the disturbances
that they cause. Data from a number of sources, however,
indicate that certain ANAs may in fact have beneficial
functional properties that should be incorporated into
any model of disease. Two examples are illuminating.
Thus, while antibodies to HMGB1 occur commonly in
the sera of patients with rheumatic disease [35, 36], evi-
dence from animal models indicates that anti-HMGB1
antibodies can strikingly attenuate disease by inhibiting
the pro-inflammatory potential of HMGB1 [37, 38]. The
benefits of anti-HMBG1 antibodies occur with both
monoclonal and polyclonal preparations that have been
elicited by immunization. Interestingly, in patients with
septic shock, the presence of autoantibodies to HMGB1
is associated with increased survival, suggesting that the
induction of autoantibodies in infectious diseases may be
a beneficial response [39]. Another example of this situa-
tion relates to histones because a monoclonal antibody to
H3 can attenuate the deleterious action of histones in
models such as liver disease induced by concanavalin A
or acetaminophen [40].
While protective effects of an ANA might seem sur-
prising, the experience with the treatment of inflamma-
tory disease with anti-HMGB1 antibodies demonstrates
clearly this possibility [37, 38]. In view of these findings,
I would posit that an ANA can function to inhibit as
well as promote disease, with this property resulting
from an antibody interaction that blocks the pathogenic-
ity of the autoantigen. This blockade could result from
binding to certain epitopes that would limit interaction
of the nuclear antigen with a receptor, for example, or
promote clearance. In this regard, epitope selection in
the generation of normal as well as aberrant antibody
responses depends on biochemical determinants of antige-
nicity (i.e. sequence, charge and conformation) rather
than pathogenicity. While formally it is possible that
antigen selection during autoimmunity leads only to the
Scandinavian Journal of Immunology, 2012, 76, 223228
D. S. Pisetsky
..................................................................................................................................................................
expression of pathogenic ANA, the heterogeneity of these
responses would make this situation seem unlikely.
Together, these considerations suggest that an ANA
may not only be non-pathogenic but may actually be
protective. This situation could explain the discrepancy
between serology and clinical events, suggesting that
some serologically positive, clinical negative patients
would actually be protected from renal disease, for exam-
ple, rather than simply spared from this serious compli-
cation. Beyond the examples given, the protective
properties of ANAs have not been explored because the
models used to evaluate pathogenicity only investigate,
for example, whether a monoclonal anti-DNA antibody
causes nephritis. Such models do not assess whether an
antibody that fails to induce nephritis (i.e. a non-nephri-
togenic antibody) could protect against disease when
administered in the presence of a bona fide nephritogenic
specificity. Mixing experiments both in vivo and in vitro
could address these possibilities by testing combinations
of antibodies to see if one can block the activity of
another. This approach could lead to the discovery of
novel biological agents, although it would be performed
optimally with affinity-purified or monoclonal products.
Identification of protective specificities could also lead
to the development of new ANA assays if, for example,
protection could be related to certain immunochemical
properties of antibodies such as avidity, isotype or epitope
specificity. In this regard, determining a larger spectrum of
ANAs in patients with autoimmunity may also be infor-
mative if antibodies to certain nuclear molecules (e.g.
HMGB1) could be shown convincingly to be protective on
the basis of in vivo or in vitro model experiments as well as
clinical studies relating the presence of antibodies to either
the frequency or the severity of clinical manifestations,
assessing both positive and negative effects.
Conclusions
On the basis of this discussion, I would like to posit a
function-based classification of ANA that extends beyond
the current model. This scheme, which is speculative and
based on limited data at this time, suggests the existence
of several distinct types of ANA:
Non-pathological: These antibodies occur in otherwise
normal individual and have no clinical significance with
respect to an association with disease manifestations. Such
ANAs can also be called benign, innocuous or indifferent.
Pathological: These ANAs occur only, or almost exclu-
sively, in individuals with an autoimmune disease or a
pre-autoimmune state in which clinical disease manifesta-
tions have not yet developed. Pathological antibodies
may or may not be pathogenic.
Pathogenic: These ANAs can cause disease manifesta-
tions by any mechanism including immune complex
deposition, cytokine stimulation or receptor binding.
 2012 The Authors.
Scandinavian Journal of Immunology  2012 Blackwell Publishing Ltd.
Classification of Antinuclear Antibodies 227
Nephritogenic: These pathogenic ANA cause nephritis.
Interferogenic: These pathogenic ANAs induce cytokine
production via stimulation of interferon production of
PDCs by immune complexes. While nephritogenic
antibodies can be interferogenic and vice versa, these
antibodies could be separate populations.
Protective: These antibodies prevent disease by inhibit-
ing the immunological activity of nuclear antigens, pro-
moting their clearance in a non-phlogistic way or
blocking the formation of pathogenic immune complexes.
Because these antibodies occur only in disease settings,
they are pathologic but not pathogenic. ANAs arising
during infection may have this property.
This classification scheme suggests that the serology of
the rheumatic disease may be even more complicated
than previously thought, with clinical heterogeneity of
disease more than matched by the serologic heterogene-
ity. Nevertheless, by raising the possibility that some
ANAs can be protective, this scheme provides a new
strategy to elucidate the relationship between serology
and disease and hopefully yield to the development of a
new class of biological agents in which autoreactivity can
be turned into a therapeutic direction.
Acknowledgment
This work was supported by a VA Merit Review, Alli-
ance for Lupus Research, and NIH (AI093960) grants.
References
1 Wandstrat AE, Carr-Johnson F, Branch V et al. Autoantibody pro-
filing to identify individuals at risk for systemic lupus erythemato-
sus. J Autoimmun 2006;27:15360.
2 Li QZ, Karp DR, Quan J et al. Risk factors for ANA positivity in
healthy persons. Arthritis Res Ther 2011;13:R38.
3 Satoh M, Chan EK, Ho LA et al. Prevalence and sociodemographic
correlates of antinuclear antibodies in the United States. Arthritis
Rheum 2012;Doi: 10.1002/art.34380. [Epub ahead of print].
4 Pisetsky DS. Antinuclear antibodies in healthy people: the tip of
autoimmunitys iceberg? Arthritis Res Ther 2011;13:109.
5 Kaveri SV, Silverman GJ, Bayry J. Natural IgM in immune equi-
librium and harnessing their therapeutic potential. J Immunol
2012;188:93945.
6 Notley CA, Brown MA, Wright GP, Ehrenstein MR. Natural IgM
is required for suppression of inflammatory arthritis by apoptotic
cells. J Immunol 2011;186:496772.
7 Cho JH, Gregersen PK. Genomics and the multifactorial nature
of human autoimmune disease. N Engl J Med 2011;365:1612
23.
8 Arbuckle MR, McClain MT, Rubertone MV et al. Development of
autoantibodies before the clinical onset of systemic lupus erythemat-
osus. N Engl J Med 2003;349:152633.
9 Bonfa E, Llovet R, Scheinberg M, De Souza JM. Comparison
between autoantibodies in malaria and leprosy with lupus. Clin Exp
Immunol 1987;70:52937.
10 Adebajo AO, Charles P, Maini RN, Hazleman BL. Autoantibodies
in malaria, tuberculosis and hepatitis B in a West African popula-
tion. Clin Exp Immunol 1993;92:736.
228 Classification of Antinuclear Antibodies
..................................................................................................................................................................
11 Harley JB, James JA. Epstein-Barr virus infection induces lupus au-
toimmunity. Bull NYU Hosp Jt Dis 2006;64:4551.
12 Hahn BH. Antibodies to DNA. N Engl J Med 1998;338:135968.
13 Ardoin SP, Pisetsky DS. Developments in the scientific understand-
ing of lupus. Arthritis Res Ther 2008;10:218.
14 McCarty GA, Rice JR, Bembe ML, Pisetsky DS. Independent
expression of autoantibodies in systemic lupus erythematosus.
J Rheumatol 1982;9:6915.
15 Hiepe F, Domer T, Hauser AE, Hoyer BF, Mei H, Radbruch A.
Long-lived autoreactive plasma cells drive persistent autoimmune
inflammation. Nat Rev Rheumatol 2011;7:1708.
16 Pisetsky DS, Grammer AC, Ning TC, Lipsky PE. Are autoantibod-
ies the targets of B-cell-directed therapy? Nat Rev Rheumatol
2011;7:5516.
17 Kalaaji M, Fenton KA, Mortensen ES et al. Glomerular apoptotic
nucleosomes are central target structures for nephritogenic antibod-
ies in human SLE nephritis. Kidney Int 2007;71:66472.
18 Mortensen ES, Rekvig OP. Nephritogenic potential of anti-DNA
antibodies against necrotic nucleosomes. J Am Soc Nephrol 2009;
20:696704.
19 Kruse K, Janko C, Urbonaviciute V et al. Inefficient clearance of
dying cells in patients with SLE: anti-dsDNA autoantibodies,
MFG-E8, HMGB-1 and other players. Apoptosis 2010;15:1098113.
20 Seredkina N, Rekvig OP. Acquired loss of renal nuclease activity is
restricted to DNaseI and is an organ-selective feature in murine
lupus nephritis. Am J Pathol 2011;179:11208.
21 Krishnan MR, Wang C, Marion TN. Anti-DNA autoantibodies ini-
tiate experimental lupus nephritis by binding directly to the glo-
merular basement membrane in mice. Kidney Int 2012;Doi:
10.1038/ki.2011.484. [Epub ahead of print].
22 Lefkowith JB, Gilkeson GS. Nephritogenic autoantibodies in lupus:
current concepts and continuing controversies. Arthritis Rheum
1996;39:894903.
23 Pisetsky DS, Ullal AJ. The blood nucleome in the pathogenesis of
SLE. Autoimmun Rev 2010;10:357.
24 Beyer C, Stearns NA, Giessl A, Distler JH, Schett G, Pisetsky D.
The extracellular release of DNA and HMGB1 from Jurkat T cells
during in vitro necrotic cell death. Innate Immun 2012; Doi:
10.1177/1753425912437981. [Epub ahead of print].
25 DeGiorgio LA, Konstantinov KN, Lee SC, hardin JA, Voipe BT,
Diamond B. A subset of lupus anti-DNA antibodies cross-reacts
with the NR2 glutamate receptor in systemic lupus erythematosus.
Nat Med 2001;11:118993.
26 Faust TW, Chang EH, Kowal C et al. Neurotoxic lupus autoanti-
bodies alter brain function through two distinct mechanisms. Proc
Natl Acad Sci USA 2010;1074:1856974.
27 Vallin H, Perers A, Alm GV, Ronnblom L. Anti-double-stranded
DNA antibodies and immunostimulatory plasmid DNA in combi-
D. S. Pisetsky
nation mimc the endogenous IFN-a inducer in systemic lupus
erythematosus. J Immunol 1999;163:630613.
28 Boule MW, Broughton C, Mackay F, Akira S, Marshak-Rothstein
A, Rifkin IR. Toll-like receptor 9-dependent and independent
dendritic cell activation by chromatin-immunoglobulin G
complexes. J Exp Med 2004;199:163140.
29 Tian J, Avalos AM, Mao S-Y et al. Toll-like receptor 9-dependent
activation by DNA-containing immune complexes is mediated by
HMGB1 and RAGE. Nat Immunol 2007;8:48796.
30 Wahamaa H, Schierbeck H, Hreggvidsdottir HS et al. High mobil-
ity group box protein 1 in complex with lipopolysaccharide or IL-1
promotes an increased inflammatory phenotype in synovial fibro-
blasts. Arthritis Res Ther 2001;13:R136.
31 Harris HE, Andersson U, Pisetsky DS. HMGB1: a multifunctional
alarmin driving autoimmune and inflammatory disease. Nat Rev
Rheumatol 2012;8:195202.
32 Yang H, Lundback P, Ottosson L et al. Redox modification of cyste-
ine residues regulates the cytokine activity of high mobility group
box-1 (HMGB1). Mol Med 2012;18:2509.
33 Skinner RP, Maddison PJ. Analysis of polyethylene glycol precipi-
tates from SLE era: antibody enrichment in association with disease
activity. Clin Exp Rheumatol 1990;8:55360.
34 Ahlin E, Mathsson L, Eloranta ML et al. Autoantibodies associated
with RNA are more enriched than anti-dsDNA antibodies in circu-
lating immune complexes in SLE. Lupus 2012;21:58695.
35 Wittemann B, Neuer G, Michels H, Truckenbrodt H, Bautz FA.
Autoantibodies to nonhistone chromosomal proteins HMG-1 and
HMG-2 in sera of patients with juvenile rheumatoid arthritis.
Arthritis Rheum 1990;33:137883.
36 Abdulahad DA, Westra J, Blizet J et al. High mobility group box
1 (HMGB1) and anti-HMGB1antibodies and their relation to dis-
ease characteristics in systemic lupus erythematosus. Arthritis Res
Ther 2011;13:R71.
37 Yang H, Ochani M, Li J et al. Reversing established sepsis with
antagonists of endogenous high-mobility group box 1. Proc Natl
Acad Sci USA 2004;101:296301.
38 Schierbeck H, Lundback P, Palmblad K et al. Monoclonal anti-
HMGB1 (high mobility group box chromosomal protein 1) anti-
body protection in two experimental arthritis models. Mol Med
2011;17:103944.
39 Barnay-Verdier S, Fattoum L, Borde C, Kaveri S, Gibot S, Marechal
V. Emergence of autoantibodies to HMGB1 is associated with sur-
vival in patients with septic shock. Intensive Care Med 2011;37:957
62.
40 Xu J, Zhang X, Monestier M, Esmon NL, Esmon CT. Extracellular
histones are mediators of death through TLR2 and TLR4 in mouse
fatal liver injury. J Immunol 2011;187:262631.
Scandinavian Journal of Immunology, 2012, 76, 223228