Professional Documents
Culture Documents
Addison Nature 2012
Addison Nature 2012
Addison Nature 2012
Introduction
Autoimmune Addison disease (AAD) is a rare endo- patient with AAD does not always predict their pheno-
crine condition with a prevalence in white European type, and vice versa.9 In addition, mutations associated
populations of 110–140 cases per million,1–3 making it with AAD show incomplete penetrance, which means
30-fold less prevalent than type 1 diabetes (T1DM) and that an individual who inherits a known disease suscepti-
200-fold rarer than autoimmune thyroid diseases. Like bility allele might not manifest the disease. Furthermore,
other autoimmune diseases, AAD is more frequent in the occurrence of phenocopies means that environ
women (female:male ratio 1.5–3.5:1) and often pres- mental factors could induce an AAD phenotype in an
ents in individuals between the ages of 30 and 50 years, individual without known disease susceptibility alleles
although it can affect individuals at any age.1 that is identical to that of a patient who does carry such
The combination of fatigue, hypotension, weight loss alleles. Finally, variants in any one of several genes could
and skin hyperpigmentation associated with pathological result in the same phenotype (genetic heterogeneity), and
changes in the adrenal glands was first recognized by Dr multiple variants are expected to be necessary to produce
Thomas Addison, a British physician, in 1855.4 Addison’s the phenotype (polygenic inheritance).9
original description referred predominantly to patients This Review article explores the pathogenesis of AAD,
who had tuberculous infiltration of the adrenal glands.5 in particular the potential mechanisms involved in the
However, in the developing world, autoimmune destruc- breakdown of immune tolerance to adrenal antigens.
tion of the adrenal glands is now the most common cause We also outline our current understanding of the genetic
of Addison disease,1,6–8 although many other causes of architecture of this condition derived from studies in
primary adrenal failure are known (Box 1). Regardless humans and in breeds of dogs that can develop Addison
of etiology, Addison disease was invariably fatal until the disease with a similar presentation to humans. Armed
first synthetic cortisol precursors were developed, which with a more detailed understanding of pathophysiology
were introduced as a treatment for this disorder in the and the underlying genetic susceptibility to this rare con-
Institute of Genetic
Medicine, Newcastle 1940s. Such treatment revolutionized the management of dition, it is possible that disease-modifying treatments
University, International AAD, transforming it into a chronic condition. could be feasible in the near future.
Centre for Life, Central
Parkway, Newcastle
Although AAD has long been recognized to be an
upon Tyne NE1 3BZ, UK inherited disorder, investigating the genetic basis of rare, Pathogenesis of AAD
(A. L. Mitchell, complex diseases is challenging and a number of factors In AAD, steroidogenic enzymes in the adrenal cortex,
S. H. S. Pearce).
potentially complicate the process. The genotype of a normally recognized as self-antigens, become the targets
Correspondence to: of a misdirected immunological attack. Circulating anti-
A. L Mitchell
anna.mitchell@ Competing interests bodies directed against steroid 21-hydroxylase can be
ncl.ac.uk The authors declare no competing interests. detected in around 85% of patients who present with
Antigen
presentation
Inflammatory
infiltrate
Local glucocorticoid
concentrations
Figure 1 | A hypothetical model of the pathogenesis of AAD based on the breakdown of glucocorticoid-induced immune
privilege. Gradual immune-mediated destruction of the adrenal gland occurs in patients with AAD. Stage 0 corresponds to
the ‘potential’ phase in the natural history of this disorder. Once tolerance to adrenal antigens is lost (stage 1), increased
antigen presentation leads to increasing inflammatory infiltrates in the gland. Adrenocytes are damaged by this immune
response and, in stage 2, local glucocorticoid production is impaired. Stages 1 and 2 correspond to the ‘subclinical’ phase
of AAD. Eventually, in stage 3, the adrenal cortex is destroyed and steroidogenesis ceases, which corresponds to the phase
of ‘overt’ clinical AAD. Abbreviation: AAD, autoimmune Addison disease.
individuals in the UK and affects males and females AAD, however, the disease in laboratory animals does
equally. 26 However, this syndrome does occur more not progress to impaired steroidogenesis; these models
frequently in certain groups, including Iranian Jewish are, therefore, of limited relevance to the human disease.
(1:9,000),27 Sardinian (1:14,000),28 Finnish (1:25,000)29 Dogs, however, can spontaneously develop Addison
and Norwegian populations (1:90,000). 30 Affected disease and the canine form of the disease is analogous
individuals usually present in childhood with chronic to human AAD in a number of ways. Canine Addison
mucocutaneous candidiasis, adrenocortical failure and disease is predominantly a disease of middle-aged, female
autoimmune hypoparathyroidism. They might also have dogs, and it has a clinical presentation very similar to
hypoplasia of the dental enamel or nail dystrophy,31–33 that in humans.39,40 In affected dogs, common present-
and can go on to develop other autoimmune disorders, ing symptoms include lassitude, vomiting and diarrhea,
such as T1DM and pernicious anemia, at any time of muscle weakness and poor appetite. Blood tests com-
life.31 The molecular basis of APS1 has been reviewed in monly show abnormal electrolyte levels, which indicate
detail in this journal.34 mineralocorticoid as well as glucocorticoid insufficiency.
Autoimmune polyendocrinopathy syndrome type 2 As in humans, an ACTH stimulation test is used to
(APS2) is defined as AAD in conjunction with auto confirm the diagnosis. Canine Addison disease can affect
immune thyroid disease and/or T1DM,31 and is present any breed of dog, and has an incidence ranging from 1.5%
in approximately 50% of individuals with primary adrenal to 9% in susceptible breeds (Portuguese Water Dogs,
failure.35 Other autoimmune conditions might also arise Nova Scotia Duck-tolling Retrievers, Standard Poodles,
in individuals with APS2,31 most commonly pernicious Leonbergers and Bearded Collies).41–44 The notably high
anemia and vitiligo. Like isolated AAD, APS2 is more incidence of the disease in certain breeds is likely to reflect
common in women than men, and the first manifestations founder effects owing to substantial inbreeding of pedi-
often appear in the fourth decade of life.36,37 gree dogs and the frequent use of popular sires (Figure 2).
These factors strongly support a genetic etiology for
Animal models of AAD canine Addison disease43 and a number of loci have
Animal studies might provide useful genetic information been identified that confer susceptibility to this disease,
about AAD owing to careful, controlled breeding and mostly through linkage studies. In Portuguese Water
environments.9 Allergic adrenalitis has been experimen- Dogs, loci on chromosomes 12 and 37 have been linked
tally induced in a range of animals and results in lympho- with canine Addison disease and these were identified
cytic infiltration of the adrenal glands and autoantibody as orthologues of human MHC and the cytotoxic T lym-
production,38 as it does in humans. In contrast to human phocyte antigen‑4 (CTLA4) gene regions, respectively.45
Table 1 | Selected case–control, candidate-gene studies in patients with non-APS1 Addison disease
Gene or marker Populations Number of patients Odds ratio P value
MHC class II loci
HLA-DR3–DQ2 Norway37 94 3.6 <10–7
HLA-DR3–DQ2 Finland, Russia, Estonia 131
69 (14 Finnish) 14.5 <0.0001
HLA-DR4–DQ8*0404 Norway37 94 NR <10–5
HLA-DR4–DQ8*0401 <10–4
HLA-DRB1*0301 Norway132 414 22.13 6 × 10–20
HLA-DRB1*0404
Other loci
MICA*5.1 Italy65 28 6.52 0.0015
USA62 46 22.5 <10–5
CYP21Adel Finland61 12 25.0 NR
CYP21 ± L10, R102, A494 Finland64 12 8.9 NR
CLEC16A (rs12917716) Norway71 332 0.72 0.0004
UK71 210 1.06 0.71
Combined71 542 0.81 0.006
CYP27B1 –1260C>A Germany118 124 1.18 0.0062 (genotype);
0.3354 (allele)
UK119 104 1.71 0.003
VDR Fok1 (exon 2) Germany 133
95 NR 0.0351 (genotype);
0.3390 (allele)
FCRL3 (rs7528684) UK104 200 1.61 0.0001
PTPN22 rs2476601 UK 90
104 1.69 0.031
Germany89 121 1.03 0.9878
Norway91 302 1.39 0.016
UK92 251 1.63 0.008
Poland92 87 1.84 0.010
CTLA4 A>G (exon 1) UK81 90 1.64 0.008
CTLA4 JO30G>A UK83 40 1.9 0.02
Norway83 94 1.4 0.04
Combined83 134 1.5 0.03
NLRP1 (rs12150220) Norway107 333 1.25 0.007
Poland108 101 1.5 0.015
PDL1 (rs1411262) UK99 315 1.33 0.032
Norway99 342 1.34 0.026
Combined99 657 1.32 3.03 × 10–3
Abbreviation: NR, not reported.
Autoimmunity is thought to arise as a result of disrup- possibly because some HLA-DR or HLA-DQ poly
tions to both adaptive and innate immune responses. In morphisms encode protein variants that enable self-
patients with AAD, the majority of susceptibility loci peptides to enter the antigen-binding pocket (Figure 4).
identified exert their effects through the adaptive arm of Although an association between AAD and HLA-DR3
the immune system; however, some AAD susceptibility alleles58 has been confirmed by a number of studies,59–61
loci are in genes of the innate immune system (Figure 3). initially, only one study reported an association with
HLA-DR4.58 Subsequent studies then confirmed asso-
Adaptive immunity ciations between AAD and the DRB1*04–DQA1*0301–
The MHC locus is a genomic region on chromosome DQB1*0302 haplotype;37,62 heterozygote carriers of the
6p21 that encodes the human leukocyte antigens DR3-DQ2 and DR4-DQ8 haplotypes were particularly
(HLA) and many other genes with and without known susceptible to this disease. In one study of Norwegian
immunological functions. HLA proteins are expressed patients, the DRB1*0404–DQ8 haplotype was par-
on the surface of antigen-presenting cells and display ticularly strongly associated with AAD, whereas the
peptides (both self and nonself ) to T cells for patho- DRB1*0401–DQ8 haplotype was protective against this
gen surveillance by the immune system. Autoimmune disease.37 The latter finding was replicated in a study
diseases arise when a persistent immune response is from the USA. 63 The association of more than one
aberrantly triggered by self peptides; in autoimmune HLA allele with AAD supports the hypothesis that the
endocrinopathies, this response leads to destruction of disease can be triggered by autoimmune responses to
hormone-producing cells. a range of peptide epitopes of steroid 21-hydroxylase
Variants within the MHC class II genes are strongly and potentially of other enzymes, rather than just one
associated with several autoimmune conditions, specific epitope.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X Y
PTPN22 CTLA4 MICA PD-L1 VDR CLEC16A NLRP1
FCRL3 MHC CYP27B1 CIITA
Figure 3 | Schematic representation of candidate-gene loci that confer susceptibility to autoimmune Addison disease. The
cytogenetic bands of each chromosome are shown and the approximate position of the genes known to confer
susceptibility to autoimmune Addison disease are indicated.
Other genes within the MHC region are also associ- T-cell responses.72 Genomic variants at the CTLA4 locus
ated with AAD; however, as they demonstrate strong have been implicated in the etiology of numerous auto-
linkage disequilibrium with HLA alleles, these genes immune conditions, including autoimmune thyroid
probably do not confer an independent risk of AAD but diseases,73–76 T1DM77,78 and rheumatoid arthritis.79,80
instead act as markers for the HLA alleles carried on the CTLA4 polymorphisms have also been associated with
adjacent chromosomal segment. The CYP21A2 (cyto- AAD in a number of independent cohorts: an A or G
chrome P450, family 21, subfamily A, polyp eptide 2) single nucleotide polymorphism (SNP) in exon 1, 81
gene, which encodes the steroid 21-hydroxylase enzyme, an AT repeat in the 3' untranslated region of exon 3 82
located in the MHC class III region, is one such example. and G or A alleles of the JO30 SNP downstream of this
Polymorphisms in this gene have been studied, but their gene.83 Carriers of the downstream AAD susceptibil-
association with AAD has been widely attributed to ity alleles (3' UTR and intergenic region) are thought
linkage disequilibrium with MHC class II alleles. 61,64 to have decreased levels of a soluble, secreted isoform
One gene that does seem to be independently associ- of CTLA‑4 that is able to engage CD80 and CD86
ated with AAD is the MHC class I polypeptide-related mole cules on antigen-presenting cells. These poly
sequence A (MICA) gene;65 the strength of the associa- morphisms, therefore, might act as a positive regulator
tion with the MICA*5.1 polymorphism is greater than of the adaptive immune response by enabling the activat-
might be expected from linkage disequilibrium with the ing T cell surface molecule, CD28, to access more of its
HLA-DR3–HLA-DR4 haplotype alone.62,66 ligand. Plausible support for this mechanism comes from
The MHC class II transactivator (CIITA) gene on non-obese, diabetic (NOD) mouse models.84,85 However
chromosome 16p13 encodes a protein that is a crucial several studies have found paradoxically increased levels
regulator of MHC class II expression. Its product CIITA of soluble CTLA‑4 in sera from humans with auto
has a clear role in immunity and deleterious mutations immune conditions, suggesting that the complexity of
in this protein result in a severe monogenic primary cell-surface CTLA‑4–CD28 interaction and signaling is
immunodeficiency disease, known as bare lymphocyte incompletely understood.86,87
syndrome. 67 Polymorphisms in the CIITA gene have Like CTLA4, the PTPN22 gene on chromosome
also been linked to the development of systemic lupus 1p13 encodes a negative regulator of T cell signaling,
erythematosus68 and rheumatoid arthritis,69 as well as tyrosine-protein phosphatase non-receptor type 22.
to AAD in Italian and Norwegian populations.70,71 The One particular PTPN22 variant (the T allele of SNP
AAD-associated CIITA variants lie in the promoter and rs2476601, associated with 1858C>T and Arg620Trp
intron 3 regions, but the mechanism by which these poly- substitutions at the DNA and amino acid sequence
morphisms influence susceptibility to AAD remains to level, respectively) has been implicated in susceptibility
be fully elucidated. to rheumatoid arthritis,88 T1DM89 and Graves disease.90
Engagement of co-stimulatory molecules is required This variant has also been implicated in several genetic
to complete the T lymphocyte activation signal and association studies of AAD.90–92 Although LYP has an
enable an immune response to be mounted, following established regulatory role in T-cell activation and sig-
the encounter of an antigenic peptide in the MHC- naling, the mechanisms underlying the association of
binding groove by the T cell receptor complex (Figure 4). the Arg620Trp variant with autoimmunity were initially
The CTLA4 gene on chromosome 2q33 encodes a co- unclear. Some studies demonstrated increased B and
stimulatory molecule that has a role in downregulating T-cell responses in individuals who are homozygous
Vitamin D has a suppressive effect on immunity In this way, we can speculate that progressive and sys-
and autoimmunity; supplementation with vitamin D temic autoreactivity to adrenal antigens might be ini-
prevents the onset of autoimmune diabetes in the tiated and eventually lead to complete adrenal failure.
d iabetes-prone NOD mouse 115 and vitamin D defi- A similar mechanism might contribute to the high
ciency has been associated with human autoimmune rate of adrenal metastases observed in patients with
disease including T1DM.116 Polymorphisms in both the cancer, in that high concentrations of glucocorticoids
vitamin D receptor (VDR) gene on chromosome 12q12 in the adrenal cortex might attenuate the presentation
and the CYP27B1 gene on chromosome 12q13 (which of tumor antigens and, therefore, perturb immune sur-
encodes a cytochrome P450 1α-hydroxylase enzyme veillance. If adrenocortical steroidogenesis does indeed
that catalyses the conversion of 25-hydroxyvitamin D3 protect the healthy adrenal gland from the develop-
to the active 1,25-dihydroxyvitamin D 3) have been ment of autoimmunity, this mechanism might help to
associated with AAD. A small study in Germany found explain why AAD is a rare condition compared to other
evidence of association between VDR genotypes and autoimmune endocrinop athies. A laboratory animal
AAD; however, no individual VDR allele was signifi- model that more closely resembles AAD would greatly
cantly associated with this disease.117 An independent, aid our understanding of disease progression.
larger cohort study is required to confirm VDR as a sus-
ceptibility locus in AAD. By contrast, convincing evi- Conclusions
dence indicates that a promoter polymorphism in the AAD is a condition that results from the interaction
CYP27B1 gene (–1260C>A) is associated with T1DM, of currently unknown environmental factors with
Graves disease, Hashimoto thyroiditis and AAD. This incompletely defined genetic factors in a susceptible
association was initially identified in a German AAD individual. The genetic etiology of AAD has been a
cohort 118 and subsequently confirmed in a UK study.119 subject of research for more than three decades. The
number of susceptibility loci known to contribute to
Hypothetical basis for AAD progression this disease is gradually increasing, but those discov-
As outlined above, there are a series of phases in AAD ered so far make only a modest contribution to disease
development, from steroid 21-hydroxylase antibody susceptibility. Further research is needed for a major
positivity, through compensated, pre-clinical adrenal breakthrough in our understanding of the etiology of
insufficiency (with raised plasma renin and ACTH) AAD. Most genetic studies of AAD have been relatively
to overt, symptomatic adrenal failure. The triggers small and based on candidate-gene selection. Although
or factors that influence progression from one phase this hypothesis-driven approach has yielded some
to the next in AAD are not yet clear. It has been sug- successes, the gene or genes that confer most of the
gested that the zona fasciculata is protected from susceptibility to AAD might never be identified using
lymphoc ytic infiltration for a longer period than the this method.
zona glomerulosa by high concentrations of locally Studies of larger cohorts of patients with AAD,
produced cortisol.120 We envisage a more generalized perhaps conducted through international collaboration
role for the high intra-adrenal levels of glucocorticoid, between research institutes, are required. In addition to
in that continuous local steroidogenesis might consti- the candidate-gene approach described above, genome-
tute a milieu that exerts a natural immunosuppressive wide association studies could offer unique insights
effect, protecting the healthy adrenal gland from auto into the genetic etiology of AAD, providing that enough
reactivity. In support of this hypothesis, cortisol levels patients with AAD could be identified to give the study
in the healthy adrenal cortex are in the 20–50 μmol/l sufficient statistical power. Whole-genome sequenc-
range, 121 and these concentrations are sufficient to ing of the DNA of patients with AAD might also yield
inhibit antigen-presenting cell function, 122 and to useful insights into the genetic architecture of AAD,
induce a state of suppressed T lymphoc yte function again as long as the study was adequately powered,
in vitro. 123–127 Moreover, focal areas of lymphocyte although this method could be prohibitively expen-
infiltration have been found at autopsy in the adrenal sive. With an increased understanding of the etiology of
cortex of 7% of individuals who are under the age of AAD and the biological pathways involved, new thera-
50 years, and their prevalence increases with age. 128 As peutic targets and diagnostic tools could be developed
this proportion is far larger than that of people who that will directly benefit individuals with AAD.
develop AAD, we can assume these infiltrates are tran-
sient or of no pathogenic importance in the majority
of individuals. However, it is possible that a key step
in the pathogenesis of AAD could be a local reduction Review criteria
in glucocorticoid concentrations at the site of a focal The PubMed database was searched using the terms
lymphoc ytic infiltrate, which reduces immune toler- “Addison disease”, “autoimmunity”, “adrenal failure”,
ance to adrenal antigens and leads to lymphocyte acti- “adrenal insufficiency” and “immunity”. The search was
vation (Figure 1). Steroidogenesis might then become limited to English-language papers only, and all years
further impaired by augmentation of the immune cell of publication were included. Research papers in the
infiltrate, resulting in a vicious cycle of increasing authors’ personal collections, not available online, were
also used.
adrenocyte damage and decreasing steroid production.
1. Kong, M. F. & Jeffcoate, W. Eighty-six cases of 20. Smans, L. C. & Zelissen, P. M. Partial recovery 38. Drexhage, H. A. Autoimmune endocrinopathies
Addison’s disease. Clin. Endocrinol. (Oxf.) 41, of adrenal function in a patient with (ed. Volpe, R.) 309–336 (Humana Press,
757–761 (1994). autoimmune Addison’s disease. J. Endocrinol. 1999).
2. Laureti, S., Vecchi, L., Santeusanio, F. & Invest. 31, 672–674 (2008). 39. Greco, D. S. Hypoadrenocorticism in small
Falorni, A. Is the prevalence of Addison’s 21. De Bellis, A. et al. Remission of subclinical animals. Clin. Tech. Small Anim. Pract. 22, 32–35
disease underestimated? J. Clin. Endocrinol. adrenocortical failure in subjects with adrenal (2007).
Metab. 84, 1762 (1999). autoantibodies. J. Clin. Endocrinol. Metab. 76, 40. Peterson, M. E., Kintzer, P. P. & Kass, P. H.
3. Løvås, K. & Husebye, E. S. High prevalence and 1002–1007 (1993). Pretreatment clinical and laboratory findings in
increasing incidence of Addison’s disease in 22. De Bellis, A. A. et al. Time course of dogs with hypoadrenocorticism: 225 cases
western Norway. Clin. Endocrinol. (Oxf.) 56, 21-hydroxylase antibodies and long-term (1979–1993). J. Am. Vet. Med. Assoc. 208,
787–791 (2002). remission of subclinical autoimmune 85–91 (1996).
4. Addison, T. On The Constitutional And Local adrenalitis after corticosteroid therapy: 41. Famula, T. R., Belanger, J. M. & Oberbauer, A. M.
Effects Of Disease Of The Supra-Renal Capsules case report. J. Clin. Endocrinol. Metab. 86, Heritability and complex segregation analysis of
(1855). 675–678 (2001). hypoadrenocorticism in the standard poodle.
5. Ten, S., New, M. & Maclaren, N. Clinical review 23. Finnish-German APECED Consortium. An J. Small Anim. Pract 44, 8–12 (2003).
130: Addison’s disease 2001. J. Clin. autoimmune disease, APECED, caused by 42. Oberbauer, A. M. et al. Inheritance of
Endocrinol. Metab. 86, 2909–2922 (2001). mutations in a novel gene featuring two PHD- hypoadrenocorticism in bearded collies. Am. J.
6. Nerup, J. Addison’s disease—a review of some type zinc-finger domains. Nat. Genet. 17, Vet. Res. 63, 643–647 (2002).
clinical, pathological and immunological 399–403 (1997). 43. Pedersen, N. C. A review of immunologic
features. Dan Med. Bull. 21, 201–217 (1974). 24. Nagamine, K. et al. Positional cloning of the diseases of the dog. Vet. Immunol.
7. Söderbergh, A. et al. Adrenal autoantibodies APECED gene. Nat. Genet. 17, 393–398 Immunopathol. 69, 251–342 (1999).
and organ-specific autoimmunity in patients (1997). 44. Smallwood, L. J. & Barsanti, J. A.
with Addison’s disease. Clin. Endocrinol. (Oxf.) 25. Pearce, S. H. et al. A common and recurrent Hypoadrenocorticism in a family of leonbergers.
45, 453–460 (1996). 13‑bp deletion in the autoimmune regulator J. Am. Anim. Hosp. Assoc. 31, 301–305 (1995).
8. Zelissen, P. M., Bast, E. J. & Croughs, R. J. gene in British kindreds with autoimmune 45. Chase, K., Sargan, D., Miller, K., Ostrander, E. A.
Associated autoimmunity in Addison’s disease. polyendocrinopathy type 1. Am. J. Hum. Genet. & Lark, K. G. Understanding the genetics of
J. Autoimmun. 8, 121–130 (1995). 63, 1675–1684 (1998). autoimmune disease: two loci that regulate late
9. Lander, E. S. & Schork, N. J. Genetic dissection 26. Pearce, S. H. & Cheetham, T. D. Autoimmune onset Addison’s disease in Portuguese Water
of complex traits. Science 265, 2037–2048 polyendocrinopathy syndrome type 1: treat with Dogs. Int. J. Immunogenet. 33, 179–184 (2006).
(1994). kid gloves. Clin. Endocrinol. (Oxf.) 54, 433–435 46. Hughes, A. M., Jokinen, P., Bannasch, D. L.,
10. Winqvist, O., Karlsson, F. A. & Kampe, O. (2001). Lohi, H. & Oberbauer, A. M. Association of a dog
21-Hydroxylase, a major autoantigen in 27. Zlotogora, J. & Shapiro, M. S. Polyglandular leukocyte antigen class II haplotype with
idiopathic Addison’s disease. Lancet 339, autoimmune syndrome type I among Iranian hypoadrenocorticism in Nova Scotia Duck Tolling
1559–1562 (1992). Jews. J. Med. Genet. 29, 824–826 (1992). Retrievers. Tissue Antigens 75, 684–690 (2010).
11. Nikoshkov, A. et al. A conformation-dependent 28. Rosatelli, M. C. et al. A common mutation in 47. Bowen, D., Schaer, M. & Riley, W. Autoimmune
epitope in Addison’s disease and other Sardinian autoimmune polyendocrinopathy‑ polyglandular syndrome in a dog: a case report.
endocrinological autoimmune diseases maps to candidiasis‑ectodermal dystrophy patients. J. Am. Anim. Hosp. Assoc. 22, 649–654 (1985).
a carboxyl-terminal functional domain of human Hum. Genet. 103, 428–434 (1998). 48. Heggarty, H. Addison’s disease in identical
steroid 21-hydroxylase. J. Immunol. 162, 29. Perheentupa, J. Autoimmune twins. Br. Med. J. 1, 559 (1968).
2422–2426 (1999). polyendocrinopathy‑candidiasis‑ectodermal 49. Simmonds, J. P. & Lister, J. Auto-immune
12. Nikfarjam, L. et al. Mechanism of inhibition of dystrophy. J. Clin. Endocrinol. Metab. 91, Addison’s disease in identical twins. Postgrad.
cytochrome P450 C21 enzyme activity by 2843–2850 (2006). Med. J. 54, 552–554 (1978).
autoantibodies from patients with Addison’s 30. Wolff, A. S. et al. Autoimmune polyendocrine 50. Smith, M. E., Gough, J. & Galpin, O. P. Addison’s
disease. Eur. J. Endocrinol. 152, 95–101 syndrome type 1 in Norway: phenotypic disease in identical twins. Br. Med. J. 2, 1316
(2005). variation, autoantibodies, and novel mutations (1963).
13. Furmaniak, J. et al. Autoimmune Addison’s in the autoimmune regulator gene. J. Clin. 51. Russell, G. A., Coulter, J. B., Isherwood, D. M.,
disease—evidence for a role of steroid Endocrinol. Metab. 92, 595–603 (2007). Diver, M. J. & Smith, D. S. Autoimmune
21-hydroxylase autoantibodies in adrenal 31. Neufeld, M., Maclaren, N. K. & Blizzard, R. M. Addison’s disease and thyrotoxic thyroiditis
insufficiency. J. Clin. Endocrinol. Metab. 79, Two types of autoimmune Addison’s disease presenting as encephalopathy in twins. Arch. Dis.
1517–1521 (1994). associated with different polyglandular Child. 66, 350–352 (1991).
14. Boscaro, M. et al. Hormonal responses during autoimmune (PGA) syndromes. Medicine 52. Fairchild, R. S., Schimke, R. N. & Abdou, N. I.
various phases of autoimmune adrenal failure: (Baltimore) 60, 355–362 (1981). Immunoregulation abnormalities in familial
no evidence for 21-hydroxylase enzyme activity 32. Ahonen, P., Myllarniemi, S., Sipila, I. & Addison’s disease. J. Clin. Endocrinol. Metab. 51,
inhibition in vivo. J. Clin. Endocrinol. Metab. 81, Perheentupa, J. Clinical variation of 1074–1077 (1980).
2801–2804 (1996). autoimmune polyendocrinopathy‑ 53. Hewitt, P. H. Addison’s disease occurring in
15. Krohn, K., Uibo, R., Aavik, E., Peterson, P. & candidiasis‑ectodermal dystrophy (APECED) in a sisters. Br. Med. J. 2, 1530–1531 (1957).
Savilahti, K. Identification by molecular cloning series of 68 patients. N. Engl. J. Med. 322, 54. Vyse, T. J. & Todd, J. A. Genetic analysis of
of an autoantigen associated with Addison’s 1829–1836 (1990). autoimmune disease. Cell 85, 311–318 (1996).
disease as steroid 17 alpha-hydroxylase. Lancet 33. Betterle, C., Greggio, N. A. & Volpato, M. Clinical 55. Risch, N. Assessing the role of HLA-linked and
339, 770–773 (1992). review 93: Autoimmune polyglandular syndrome unlinked determinants of disease. Am. J. Hum.
16. Winqvist, O., Gustafsson, J., Rorsman, F., type 1. J. Clin. Endocrinol. Metab. 83, Genet. 40, 1–14 (1987).
Karlsson, F. A. & Kampe, O. Two different 1049–1055 (1998). 56. Vaidya, B., Kendall-Taylor, P. & Pearce, S. H. The
cytochrome P450 enzymes are the adrenal 34. Akirav, E. M., Ruddle, N. H. & Herold, K. C. The genetics of autoimmune thyroid disease. J. Clin.
antigens in autoimmune polyendocrine role of AIRE in human autoimmune disease. Endocrinol. Metab. 87, 5385–5397 (2002).
syndrome type I and Addison’s disease. J. Clin. Nat. Rev. Endocrinol. 7, 25–33 (2011). 57. Hemminki, K., Li, X., Sundquist, J. &
Invest. 92, 2377–2385 (1993). 35. Betterle, C., Volpato, M., Greggio, A. N. & Sundquist, K. Familial association between
17. Betterle, C. et al. The natural history of adrenal Presotto, F. Type 2 polyglandular autoimmune type 1 diabetes and other autoimmune and
function in autoimmune patients with adrenal disease (Schmidt’s syndrome). J. Pediatr. related diseases. Diabetologia 52, 1820–1828
autoantibodies. J. Endocrinol. 117, 467–475 Endocrinol. Metab. 9 (Suppl. 1), 113–123 (1996). (2009).
(1988). 36. Dittmar, M. & Kahaly, G. J. Polyglandular 58. Maclaren, N. K. & Riley, W. J. Inherited
18. Coco, G. et al. Estimated risk for developing autoimmune syndromes: immunogenetics and susceptibility to autoimmune Addison’s disease
autoimmune Addison’s disease in patients with long-term follow-up. J. Clin. Endocrinol. Metab. is linked to human leukocyte antigens-DR3 and/
adrenal cortex autoantibodies. J. Clin. 88, 2983–2992 (2003). or DR4, except when associated with type I
Endocrinol. Metab. 91, 1637–1645 (2006). 37. Myhre, A. G. et al. Autoimmune adrenocortical autoimmune polyglandular syndrome. J. Clin.
19. Torrejón, S., Webb, S. M., Rodriguez-Espinosa, J., failure in Norway autoantibodies and human Endocrinol. Metab. 62, 455–459 (1986).
Martinez de Osaba, M. J. & Corcoy, R. Long- leukocyte antigen class II associations related 59. Boehm, B. O. et al. The HLA-DQ beta non‑Asp‑57
lasting subclinical Addison’s disease. Exp. Clin. to clinical features. J. Clin. Endocrinol. Metab. allele: a predictor of future insulin-dependent
Endocrinol. Diabetes 115, 530–532 (2007). 87, 618–623 (2002). diabetes mellitus in patients with autoimmune
Addison’s disease. Tissue Antigens 37, 130–132 76. Yanagawa, T., Hidaka, Y., Guimaraes, V., 94. Rieck, M. et al. Genetic variation in PTPN22
(1991). Soliman, M. & DeGroot, L. J. CTLA‑4 gene corresponds to altered function of T and
60. Huang, W. et al. Although DR3-DQB1*0201 may polymorphism associated with Graves’ disease B lymphocytes. J. Immunol. 179, 4704–4710
be associated with multiple component in a Caucasian population. J. Clin. Endocrinol. (2007).
diseases of the autoimmune polyglandular Metab. 80, 41–45 (1995). 95. Arechiga, A. F. et al. Cutting edge: the PTPN22
syndromes, the human leukocyte antigen DR4- 77. Marron, M. P. et al. Insulin-dependent diabetes allelic variant associated with autoimmunity
DQB1*0302 haplotype is implicated only in mellitus (IDDM) is associated with CTLA4 impairs B cell signaling. J. Immunol. 182,
beta-cell autoimmunity. J. Clin. Endocrinol. Metab. polymorphisms in multiple ethnic groups. Hum. 3343–3347 (2009).
81, 2559–2563 (1996). Mol. Genet. 6, 1275–1282 (1997). 96. Zhang, J. et al.The autoimmune disease-
61. Partanen, J., Peterson, P., Westman, P., 78. Nistico, L. et al. The CTLA‑4 gene region of associated PTPN22 variant promotes calpain-
Aranko, S. & Krohn, K. Major histocompatibility chromosome 2q33 is linked to, and associated mediated Lyp/Pep degradation associated with
complex class II and III in Addison’s disease with, type 1 diabetes. Belgian Diabetes Registry. lymphocyte and dendritic cell
MHC alleles do not predict autoantibody Hum. Mol. Genet. 5, 1075–1080 (1996). hyperresponsiveness. Nat. Genet. 43, 902–907
specificity and 21-hydroxylase gene 79. Seidl, C. et al. CTLA4 codon 17 dimorphism in (2011).
polymorphism has no independent role in patients with rheumatoid arthritis. Tissue 97. Brown, J. A. et al. Blockade of programmed
disease susceptibility. Hum. Immunol. 41, Antigens 51, 62–66 (1998). death‑1 ligands on dendritic cells enhances
135–140 (1994). 80. Vaidya, B. et al. An association between the T cell activation and cytokine production.
62. Park, Y. S. et al. Additional association of intra- CTLA4 exon 1 polymorphism and early J. Immunol. 170, 1257–1266 (2003).
MHC genes, MICA and D6S273, with Addison’s rheumatoid arthritis with autoimmune 98. Hayashi, M., Kouki, T., Takasu, N., Sunagawa, S.
disease. Tissue Antigens 60, 155–163 (2002). endocrinopathies. Rheumatology (Oxford) 41, & Komiya, I. Association of an A/C single
63. Yu, L. et al. DRB1*04 and DQ alleles: 180–183 (2002). nucleotide polymorphism in programmed cell
expression of 21-hydroxylase autoantibodies 81. Vaidya, B. et al. Association analysis of the death-ligand 1 gene with Graves’ disease in
and risk of progression to Addison’s disease. cytotoxic T lymphocyte antigen‑4 (CTLA‑4) and Japanese patients. Eur. J. Endocrinol. 158,
J. Clin. Endocrinol. Metab. 84, 328–335 (1999). autoimmune regulator‑1 (AIRE‑1) genes in 817–822 (2008).
64. Peterson, P. et al. Steroid 21-hydroxylase gene sporadic autoimmune Addison’s disease. J. Clin. 99. Mitchell, A. L. et al. Programmed death ligand 1
polymorphism in Addison’s disease patients. Endocrinol. Metab. 85, 688–691 (2000). (PD‑L1) gene variants contribute to autoimmune
Tissue Antigens 46, 63–67 (1995). 82. Kemp, E. H. et al. A cytotoxic T lymphocyte Addison’s disease and Graves’ disease
65. Gambelunghe, G. et al. Microsatellite antigen‑4 (CTLA‑4) gene polymorphism is susceptibility. J. Clin. Endocrinol. Metab. 94,
polymorphism of the MHC class I chain-related associated with autoimmune Addison’s disease 5139–5145 (2009).
(MIC‑A and MIC‑B) genes marks the risk for in English patients. Clin. Endocrinol. (Oxf.) 49, 100. Capon, F. et al. Fine mapping of the PSORS4
autoimmune Addison’s disease. J. Clin. 609–613 (1998). psoriasis susceptibility region on
Endocrinol. Metab. 84, 3701–3707 (1999). 83. Blomhoff, A. et al. Polymorphisms in the chromosome 1q21. J. Invest. Dermatol. 116,
66. Triolo, T. M. et al. Homozygosity of the cytotoxic T lymphocyte antigen‑4 gene region 728–730 (2001).
polymorphism MICA5.1 identifies extreme risk confer susceptibility to Addison’s disease. J. Clin. 101. Kochi, Y. et al. A functional variant in FCRL3,
of progression to overt adrenal insufficiency Endocrinol. Metab. 89, 3474–3476 (2004). encoding Fc receptor-like 3, is associated with
among 21-hydroxylase antibody-positive patients 84. Gerold, K. D. et al. The soluble ctla‑4 splice rheumatoid arthritis and several
with type 1 diabetes. J. Clin. Endocrinol. Metab. variant protects from type 1 diabetes and autoimmunities. Nat. Genet. 37, 478–485
94, 4517–4523 (2009). potentiates regulatory T‑cell function. Diabetes (2005).
67. Dziembowska, M. et al. Three novel mutations of 60, 1955–1963 (2011). 102. Kyogoku, C. et al. Fcgamma receptor gene
the CIITA gene in MHC class II‑deficient patients 85. Ueda, H. et al. Association of the T‑cell regulatory polymorphisms in Japanese patients with
with a severe immunodeficiency. gene CTLA4 with susceptibility to autoimmune systemic lupus erythematosus: contribution of
Immunogenetics 53, 821–829 (2002). disease. Nature 423, 506–511 (2003). FCGR2B to genetic susceptibility. Arthritis
68. Koizumi, K. et al. Single nucleotide 86. Daroszewski, J. et al. Soluble CTLA‑4 receptor an Rheum. 46, 1242–1254 (2002).
polymorphisms in the gene encoding the major immunological marker of Graves’ disease and 103. Ehrhardt, G. R. et al. Fc receptor-like proteins
histocompatibility complex class II transactivator severity of ophthalmopathy is associated with (FCRL): immunomodulators of B cell function.
(CIITA) in systemic lupus erythematosus. Ann. CTLA‑4 Jo31 and CT60 gene polymorphisms. Adv. Exp. Med. Biol. 596, 155–162 (2007).
Rheum. Dis. 64, 947–950 (2005). Eur. J. Endocrinol. 161, 787–793 (2009). 104. Owen, C. J. et al. Analysis of the Fc
69. Eyre, S. et al. Investigation of the MHC2TA gene, 87. Oaks, M. K. & Hallett, K. M. Cutting edge: receptor‑like‑3 (FCRL3) locus in Caucasians with
associated with rheumatoid arthritis in a a soluble form of CTLA‑4 in patients with autoimmune disorders suggests a complex
Swedish population, in a UK rheumatoid arthritis autoimmune thyroid disease. J. Immunol. 164, pattern of disease association. J. Clin.
cohort. Arthritis Rheum. 54, 3417–3422 5015–5018 (2000). Endocrinol. Metab. 92, 1106–1111 (2007).
(2006). 88. Begovich, A. B. et al. A missense single- 105. Cummings, J. R. et al. The genetics of NOD-like
70. Ghaderi, M. et al. MHC2TA single nucleotide nucleotide polymorphism in a gene encoding a receptors in Crohn’s disease. Tissue Antigens
polymorphism and genetic risk for autoimmune protein tyrosine phosphatase (PTPN22) is 76, 48–56 (2010).
adrenal insufficiency. J. Clin. Endocrinol. Metab. associated with rheumatoid arthritis. Am. J. 106. Jin, Y., Birlea, S. A., Fain, P. R. & Spritz, R. A.
91, 4107–4111 (2006). Hum. Genet. 75, 330–337 (2004). Genetic variations in NALP1 are associated with
71. Skinningsrud, B. et al. Polymorphisms in 89. Kahles, H. et al. Sex-specific association of generalized vitiligo in a Romanian population.
CLEC16A and CIITA at 16p13 are associated PTPN22 1858T with type 1 diabetes but not with J. Invest. Dermatol. 127, 2558–2562 (2007).
with primary adrenal insufficiency. J. Clin. Hashimoto’s thyroiditis or Addison’s disease in 107. Magitta, N. F. et al. A coding polymorphism in
Endocrinol. Metab. 93, 3310–3317 (2008). the German population. Eur. J. Endocrinol. 153, NALP1 confers risk for autoimmune Addison’s
72. Brunet, J. F. et al. A new member of the 895–899 (2005). disease and type 1 diabetes. Genes Immun. 10,
immunoglobulin superfamily-‑CTLA‑4. Nature 90. Velaga, M. R. et al. The codon 620 tryptophan 120–124 (2009).
328, 267–270 (1987). allele of the lymphoid tyrosine phosphatase 108. Zurawek, M. et al. A coding variant in NLRP1 is
73. Akamizu, T. et al. Association of autoimmune (LYP) gene is a major determinant of Graves’ associated with autoimmune Addison’s disease.
thyroid disease with microsatellite markers for disease. J. Clin. Endocrinol. Metab. 89, Hum. Immunol. 71, 530–534 (2010).
the thyrotropin receptor gene and CTLA‑4 in 5862–5865 (2004). 109. Hakonarson, H. et al. A genome-wide association
Japanese patients. Thyroid 10, 851–858 91. Skinningsrud, B. et al. Mutation screening of study identifies KIAA0350 as a type 1 diabetes
(2000). PTPN22: association of the 1858T-allele with gene. Nature 448, 591–594 (2007).
74. Kotsa, K., Watson, P. F. & Weetman, A. P. Addison’s disease. Eur. J. Hum. Genet. 16, 110. Zoledziewska, M. et al. Variation within the
A CTLA‑4 gene polymorphism is associated with 977–982 (2008). CLEC16A gene shows consistent disease
both Graves disease and autoimmune 92. Roycroft, M. et al. The tryptophan 620 allele of association with both multiple sclerosis and
hypothyroidism. Clin. Endocrinol. (Oxf.) 46, the lymphoid tyrosine phosphatase (PTPN22) type 1 diabetes in Sardinia. Genes Immun. 10,
551–554 (1997). gene predisposes to autoimmune Addison’s 15–17 (2009).
75. Sale, M. M. et al. Association of autoimmune disease. Clin. Endocrinol. (Oxf.) 70, 358–362 111. Dubois, P. C. et al. Multiple common variants for
thyroid disease with a microsatellite marker for (2009). celiac disease influencing immune gene
the thyrotropin receptor gene and CTLA‑4 in a 93. Vang, T. et al. Autoimmune-associated lymphoid expression. Nat. Genet. 42, 295–302 (2010).
Japanese population. Proc. Assoc. Am. Physicians tyrosine phosphatase is a gain‑of‑function 112. Marquez, A. et al. Specific association of a
109, 453–461 (1997). variant. Nat. Genet. 37, 1317–1319 (2005). CLEC16A/KIAA0350 polymorphism with
NOD2/CARD15– Crohn’s disease patients. Eur. J. applicability in diagnosis and disease prediction. 129. Vergeer, M. et al. Genetic variant of the
Hum. Genet. 17, 1304–1308 (2009). Endocr. Rev. 23, 327–364 (2002). scavenger receptor BI in humans. N. Engl. J. Med.
113. Skinningsrud, B. et al. A CLEC16A variant 121. Dickerman, Z., Grant, D. R., Faiman, C. & 364, 136–145 (2011).
confers risk for juvenile idiopathic arthritis and Winter, J. S. Intraadrenal steroid concentrations 130. Metherell, L. et al. Mutations in GPX1 and NNT,
anti-CCP negative rheumatoid arthritis. Ann. in man: zonal differences and developmental endcoding antioxidant defense genes, cause
Rheum. Dis. 69, 1471–1474 (2009). changes. J. Clin. Endocrinol. Metab. 59, familial glucocorticoid deficiency [Abstract].
114. Todd, J. A. et al. Robust associations of four new 1031–1036 (1984). Endocr. Rev. 32 (meeting abstracts), OR22–25
chromosome regions from genome-wide 122. Rozkova, D., Horvath, R., Bartunkova, J. & (2011).
analyses of type 1 diabetes. Nat. Genet. 39, Spisek, R. Glucocorticoids severely impair 131. Gombos, Z. et al. Analysis of extended human
857–864 (2007). differentiation and antigen presenting function leukocyte antigen haplotype association with
115. Mathieu, C., Waer, M., Laureys, J., Rutgeerts, O. of dendritic cells despite upregulation of Toll-like Addison’s disease in three populations. Eur. J.
& Bouillon, R. Prevention of autoimmune receptors. Clin. Immunol. 120, 260–271 (2006). Endocrinol. 157, 757–761 (2007).
diabetes in NOD mice by 1,25 123. Franchimont, D. et al. Inhibition of Th1 immune 132. Skinningsrud, B. et al. Multiple loci in the HLA
dihydroxyvitamin D3. Diabetologia 37, 552–558 response by glucocorticoids: dexamethasone complex are associated with Addison’s disease.
(1994). selectively inhibits IL‑12‑induced Stat4 J. Clin. Endocrinol. Metab. 96, E1703–E1708
116. Zella, J. B. & DeLuca, H. F. Vitamin D and phosphorylation in T lymphocytes. J. Immunol. (2011).
autoimmune diabetes. J. Cell Biochem. 88, 164, 1768–1774 (2000). 133. Pani, M. A., Seissler, J., Usadel, K. H. &
216–222 (2003). 124. Abe, M. & Thomson, A. W. Dexamethasone Badenhoop, K. Vitamin D receptor genotype is
117. Pani, M. A. et al. A polymorphism within the preferentially suppresses plasmacytoid dendritic associated with Addison’s disease. Eur. J.
vitamin D‑binding protein gene is associated cell differentiation and enhances their apoptotic Endocrinol. 147, 635–640 (2002).
with Graves’ disease but not with Hashimoto’s death. Clin. Immunol. 118, 300–306 (2006).
thyroiditis. J. Clin. Endocrinol. Metab. 87, 125. Cidlowski, J. A. et al. The biochemistry and Acknowledgments
2564–2567 (2002). molecular biology of glucocorticoid-induced A. L. Mitchell is in receipt of a clinical research
118. Lopez, E. R. et al. A promoter polymorphism of apoptosis in the immune system. Recent Prog. training fellowship from the Medical Research
the CYP27B1 gene is associated with Addison’s Horm. Res. 51, 457–490; discussion 490–491 Council, UK. S. H. S Pearce’s research work is
disease, Hashimoto’s thyroiditis, Graves’ (1996). supported by European Union Framework 7 grant
disease and type 1 diabetes mellitus in 126. Moser, M. et al. Glucocorticoids down-regulate 201167 to the Euradrenal Consortium and grants
Germans. Eur. J. Endocrinol. 151, 193–197 dendritic cell function in vitro and in vivo. Eur. J. G07017632 & G0900001 from the Medical
(2004). Immunol. 25, 2818–2824 (1995). Research Council, UK. A. L. Mitchell and
119. Jennings, C. E., Owen, C. J., Wilson, V. & 127. Zubiaga, A. M., Munoz, E. & Huber, B. T. IL‑4 and S. H. S. Pearce thank their collaborator, Professor
Pearce, S. H. A haplotype of the CYP27B1 IL‑2 selectively rescue Th cell subsets from Eystein Husebye, Dr Martina Ericksen and the group
promoter is associated with autoimmune glucocorticoid-induced apoptosis. J. Immunol. at Haukeland University Hospital, Bergen, Norway, for
Addison’s disease but not with Graves’ disease 149, 107–112 (1992). contributing family data on Addison disease gathered
in a UK population. J. Mol. Endocrinol. 34, 128. Hayashi, Y., Hiyoshi, T., Takemura, T., from the Norwegian registry of organ-specific
859–863 (2005). Kurashima, C. & Hirokawa, K. Focal lymphocytic autoimmune diseases for this Review.
120. Betterle, C., Dal Pra, C., Mantero, F. & infiltration in the adrenal cortex of the elderly:
Zanchetta, R. Autoimmune adrenal insufficiency immunohistological analysis of infiltrating Author contributions
and autoimmune polyendocrine syndromes: lymphocytes. Clin. Exp. Immunol. 77, 101–105 Both authors contributed equally to all aspects of the
autoantibodies, autoantigens, and their (1989). article.