humoral autoimmunity in type 1 diabetes: prediction...

Humoral Autoimmunity in Type 1 Diabetes:Prediction, Significance, and Detection ofDistinct Disease Subtypes

Massimo Pietropaolo1, Roberto Towns1, and George S. Eisenbarth2

1Laboratory of Immunogenetics, The Brehm Center for Diabetes Research, Department of InternalMedicine, University of Michigan Medical School, Ann Arbor, Michigan 48105

2Barbara Davis Center for Childhood Diabetes, University of Colorado Denver, Aurora, Colorado 80045

Correspondence: [email protected]

Type 1 diabetes mellitus (T1D) is an autoimmune disease encompassing the T-cell-mediateddestruction of pancreatic b cells and the production of autoantibodies against islet proteins.In humoral autoimmunity in T1D, the detection of islet autoantibodies and the examinationof their associations with genetic factors and cellular autoimmunity constitute major areasin both basic research and clinical practice. Although insulin is a key autoantigen and maybe primus inter pares in importance among T1D autoantigens, an abundant body of researchhas also revealed other autoantigens associated with the disease process. Solid evidenceindicates that autoantibodies against islet targets serve as key markers to enroll newly diag-nosed T1D patients and their family members in intervention trials aimed at preventing orhalting the disease process. The next challenge is perfecting mechanistic bioassays to be usedas end points for disease amelioration following immunomodulatory therapies aimed atblocking immune-mediated b-cell injury and, in turn, preserving b-cell function in type 1diabetes mellitus.

A famous influential scientist of the past mil-lennium, the Renaissance polymath Leonar-

do da Vinci (1452–1519), wrote: “The suprememisfortune is when theory outstrips perfor-mance.” This is a situation that perhaps sharessome similarities with our knowledge on thepathoetiologyof autoimmune diabetes. The dis-covery of islet autoantigens and the identifica-tion of their immunodominant epitopes hasshifted emphasis from epidemiological to mech-anistic and exploratory intervention studies us-ing these antigens, such as insulin, to prevent

T1D. An incredibly large number of immuno-modulatory strategies were and are currently ap-plied to prevent diabetes in animal models of thedisease, such as the NOD mouse (Shoda et al.2005). Many therapeutic strategies may delay orprevent diabetes in NOD mice, and the mostpromising ones are currently being tested inhumans (Skyler 2011).

Type 1 diabetes mellitus was not always con-sidered the classical autoimmune disease it isnow known to be. For instance, insulin-depen-dent diabetes was known to occur occasionally

Editors: Jeffrey A. Bluestone, Mark A. Atkinson, and Peter Arvan

Additional Perspectives on Type I Diabetes available at www.perspectivesinmedicine.org

Copyright # 2012 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a012831

Cite this article as Cold Spring Harb Perspect Med 2012;2:a012831

1

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org

on March 18, 2019 - Published by Cold Spring Harbor Laboratory Press http://perspectivesinmedicine.cshlp.org/Downloaded from

http://perspectivesinmedicine.cshlp.org/

in the Autoimmune Polyendocrine Syndrome I(APS I), a classic autoimmune syndrome with Tcell and B-cell antibody abnormalities directedat adrenal, parathyroid, gonadal, thyroid, andother tissues. However, diabetes mellitus wasnot a constant, necessary, or sufficient featureof APS I. This condition is now known to becaused by mutations in the autoimmune regu-lator gene (AIRE) (Husebye and Anderson2010). In 1974, Bottazzo et al. (1974) reportedthat sections of human pancreas treated withsera of diabetic patients who also had Addison’sdisease and myxedema showed cytoplasmicfluorescence over islets of Langerhans. This re-sponse was termed cytoplasmic islet cell anti-bodies (ICA). Furthermore, the existence of in-sulin autoantibodies and other autoantibodiesagainst various islet proteins was not uncovereduntil years later. It was in 1983 that insulin au-toantibodies were reported in sera of newly di-agnosed patients with T1D, before any treat-ment with exogenous insulin (Palmer et al.1983). In this finding, improvements of the sen-sitivity of the insulin antibody assay were instru-mental for the determination that about one-half of newly diagnosed patients had autoanti-bodies that bound 125I-labeled insulin.

Following the early discoveries on humoralautoimmunity in T1D, there has been a remark-able expansion in the detection of T1D-associ-ated autoantibodies (Table 1) as well as in thecharacterization of the molecular basis of theantigenicity of their target proteins (Atkinsonand Eisenbarth 2001; Pietropaolo and Eisen-barth 2001). This expansion has led to the un-covering of specific antigenic determinants, thedevelopment of biochemically defined immu-noassays, and also to coordinated efforts tostandardize assays across laboratories (Bonifacioet al. 2010b). However, it should be emphasizedthat T1D is primarily a T-cell-mediated disease.In humans, this conclusion was supported by areport of X-linked agammaglobulinemia inwhom typical T1D developed at the age of 14yr (Martin et al. 2001). This report shows thatT1D can occur in the complete absence of B cellsand autoantibodies. This observation led to theconclusion that B cells are not an essential re-quirement for the development of this disease

and that the principal effector mechanisms aremediated by T cells. Thus, although the presenceof islet autoantibodies may not be a sine quanon feature of autoimmune diabetes, advancesin detection of humoral autoimmunity havehad critical implications in the identificationof at-risk subjects that can become participantsin clinical trials to assess immunomodulatorystrategies to prevent and treat T1D.

ASSAY STANDARDIZATION ANDHARMONIZATION

Currently, the consensus on methodologicalstandardization encompasses assays to detectautoantibodies against four major islet autoan-tigens, namely, insulin, glutamic acid decar-boxylase (GAD), the neuroendocrine antigenICA512/IA-2, I-A2b (phogrin), and the zinctransporter ZnT8. Although there is an overallagreement regarding the methodologies to as-sess humoral autoimmunity in T1D, the abilityto detect T1D-related autoantibodies and to ac-curately measure their titer has also clear orga-nizational implications because of the need tointerpret values across laboratories, popula-tions, and countries and to promote the devel-opment of assay systems that can improve thecomparative assessment of results.

These strategies have included the adoptionof a serum reference standard for GAD and IA-2antibodies by the World Health Organization(WHO) and the creation of the Diabetes Anti-body Standardization Program (DASP), whichwas established in 2000 by the Immunology ofDiabetes Society (IDS) and the Centers for Dis-ease Control and Prevention (CDC) (Verge et al.1998; Bonifacio et al. 2002). It has been DASPactivities that have led to the recognition of in-sulin, GAD, IA-2, and ZnT8 as the main targetsfor the validation and standardization of assays.The need to provide a broader basis for com-parison of results has also evolved into harmo-nization efforts to produce unified protocolsthat can take into account differences in stan-dards such as the WHO standard, relative to theuse of common standards used by the NationalInstitute of diabetes and Digestive and Kidney

M. Pietropaolo et al.

2 Cite this article as Cold Spring Harb Perspect Med 2012;2:a012831

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Table 1. Most characterized islet autoantigens associated with type 1 diabetes

Localization Humoral response Cellular response

Insulina Secretory granulespancreatic b cells

Human thymus and PAEcells (peripheral antigenexpressing cells)

Insulin autoantibodies are foundin virtually 100% of youngchildren (,5 yr of age) beforethe onset of Type 1 diabetes

Correlation with younger age andfast rate of progression toinsulin requirement in first-degree relatives of IDDMpatients

Prophylactic subcutaneousinjection of insulin, oral andintranasal administrationprevents type 1 diabetes inNOD mice

PBLs from humans andNOD mice react withinsulin b-chain

GAD65a andGAD67

Synaptic-like microvesiclesof neuroendocrine cells

Present in testis and ovaryHuman thymus and PAE

cells

A subset of 64-kDaautoantibodies recognize GAD

Autoantibodies to GAD65 arepresent in 70%–80% of pre-diabetic subjects or newlydiagnosed diabetic patients

GAD antibodies are also detectedin patients with stiff mansyndrome, and withautoimmune thyroid disease

Radioimmunoassay of in vitrotranscribed/translated GAD65is useful for large-scalescreening

PBL responses toGAD65 in newlydiagnosed diabeticpatients and in NODmice

ICA512(IA-2)a andphogrin (IA-2b)

ZnT8 (Slc30A8)a

Neurosecretory granules(pancreatic b cells,CNS, pituitary, adrenal)

Human thymus and PAEcells

Zn transporter, a memberof the cation diffusionfacilitator familyshowing abundantexpression in b cells

Expressed also extra-pancreatically

Autoantibodies to ICA512 (IA-2)are present in �60% of pre-diabetics or newly diagnosedIDDM patients

Relationship between 37,000- and40,000-Da tryptic fragmentsand ICA512(IA-2)

Radioimmunoassay of in vitrotranscribed/translatedICA512(IA-2) is useful forlarge-scale screening

Targeted by autoantibodies in60%–80% of newly diagnosedT1D patients and in �26% ofpatients negative for other isletautoantibodies

Relevant polymorphic variantsare Trp-325 and Arg-325

PBL responses in newlydiagnosed diabeticpatients

Autoreactive T cells toZnT8 found inhuman T1D

Continued

Humoral Autoimmunity in Type 1 Diabetes

Cite this article as Cold Spring Harb Perspect Med 2012;2:a012831 3

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



diseases (NIDDK) consortia laboratories (Bo-nifacio et al. 2010b).

A study by Bonifacio et al. (2010a) on theharmonization of GAD and IA-2 autoantibodyassays by NIDDK consortia laboratories, usinglarge-volume calibrators and common proto-cols, indicated that common protocols and useof large-volume working calibrators are effectivemeasures to ensure consistency in autoantibodymeasurements, although the results were com-

parable but not identical. However, other stud-ies indicate that complete harmonization andcross-validation and interpretation of resultswill require additional effort.

Even though there has been considerableprogress in standardizing GAD65 and IA-2autoantibody assays, the insulin autoantibodyassay is least standardized, with most laborato-ries in DASP workshops having less than accept-able sensitivity and/or specificity. This probably

Table 1. Continued

Localization Humoral response Cellular response

Islet cell autoantigen69 kDa (ICA69)

Chromogranin A

Predominantlyneuroendocrine tissues

Human and mousethymus

Neurosecretory granulesNeuroendocrine tissues

Autoantibodies to ICA69 canbe detected in 43% of pre-diabetic subjects byWestern blotting

Circulating ChgA found in NODmice

Association betweenHLA-DR3 and PBLresponses in newlydiagnosed type 1diabetics

Autoreactive T cells toChgA found in NODmice

Carboxypeptidase H Neurosecretory granules Autoantibodies tocarboxypeptidase H foundin �20% of pre-diabetics

Present

Ganglioside GM2-1 Pancreatic islet cells Autoantibodies to GM2-1detected in �80% of pre-diabetic subjects and NODmice

?

Imogen 38 (38 kDa) Mitochondria widelydistributed with variablelevels of expression

Presence of circulating antibodiesto 38-kDa proteins

Possible presence of antibodies toimogen 38

PBLs from newlydiagnosed diabeticsproliferate to imogen38

Glima 38 Amphiphilic N-Aspglycated b-cellmembrane protein thatis expressed in islets andneuronal cell lines

Autoantibodies to Glima 38can be detected in 14%–22.7%of newly diagnosed diabeticsand pre-diabetics. The majorityof these patients are negativefor GAD65 and/or ICA512(IA-2) autoantibodies

?

Peripherin Neuronal cells Autoantibody response againstperipherin in NOD mice and inpatients with type 1 diabetesand other autoimmunedisorders

T-cell responses againstperipherin in NODmice

Heat-shock protein(Hsp60)

Ubiquitously inducible Antibodies to Hsp60 in pre-diabetic NOD mice

Hsp60-reactive T cellscan accelerate diseasein pre-diabetic NODmice

aBiochemical autoantibody assays readily available for large screening programs.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



relates to the very low capacity of T1D autoan-tibodies and a limited conformational epitoperecognized by these antibodies that is obscuredif insulin is bound to solid phase. We have re-cently developed an electrochemiluminescentinsulin autoantibody assay using pro-insulinas target antigen that appears to enhance sensi-tivity and specificity (Yu et al. 2012).

The advent of biochemically defined assayshas permitted the development of chimeric andhybrid antigen constructs that can simultane-ously assess the presence of autoantibodiesagainst more than one antigen. This strategy isillustrated in a recent study with a triple chi-meric islet autoantigen containing key antigenicdeterminants to IA-2 and key variants of the Zntransporter (ZnT8WR) as an accurate and rele-vant T1D antigen (Yu et al. 2010). The use ofthese chimeric assays holds promise to save la-bor and resources to more efficiently screen at-risk populations.

DISCOVERY OF AUTOANTIGENS ANDRELEVANT EPITOPES BASED ONAUTOANTIBODIES, MAJOR T1D-ASSOCIATED ISLET AUTOANTIBODIESCURRENTLY ASSAYED

Initial studies by Atkinson et al. (1993) identi-fied a subset of islet cell antibodies (ICA) asso-ciated with a more clinically significant pancre-atic b-cell injury in a subgroup of first-degreerelatives of T1D probands. This subset of ICAwas termed “non-GAD reactive” because ICAreactivity could only be partially blocked byGAD65 (Atkinson et al. 1993). This observationimplied that multiple islet autoantigens are rec-ognized by T1D-specific humoral responses. Wesubsequently found that a subset of cytoplasmicICA is associated to a more rapid progression toinsulin-requiring diabetes in GAD65 and IA2antibody-positive relatives as compared withrelatives with GAD65 and IA2 antibodies with-out ICA. This ICA reactivity more than likely iscaused by a subset of ICA-recognizing uniden-tified islet autoantigen(s) (Pietropaolo et al.2005).

As autoimmunity in T1D progresses frominitial activation to a chronic state, there is often

a higher number of islet autoantigens reactingwith T cells and autoantibodies. This condi-tion is termed “epitope spreading.” Compellingevidence indicates that islet autoantibody re-sponses against multiple islet autoantigens areassociated with progression to overt disease(Verge et al. 1996). Several additional T1D-re-lated autoantigens have been identified, whichinclude islet cell autoantigen 69 kDa (ICA69),the islet-specific glucose-6-phosphatase catalyt-ic subunit-related protein (IGRP), chromogra-nin A (ChgA), the insulin receptor, heat shockproteins, and the antigens jun-B,16, CD38 (Pie-tropaolo and Eisenbarth 2001), peripherin, andglial fibrillary acidic protein (GFAP), and thelike (Table 1) (Winer et al. 2003).

The existence of IgG immunoglobulins di-rected to epitopes of islet antigens strongly im-plies the influence of T-cell participation in theautoimmune response. Naturally processed epi-topes of islet autoantigens represent the targetsof effector and regulatory T cells in controllingb-cell-specific autoimmune responses (Di Lo-renzo et al. 2007). In particular, naturally pro-cessed HLA class II allele-specific epitopes rec-ognized by CD4þ T cells (corresponding to theintracellular domain of IA2) were identified af-ter native IA2 antigen was delivered to EBV-transformed B cells and peptides were elutedand analyzed by mass spectrometry (Peakmanet al. 1999). Additionally, evidence suggests asynergistic pro-inflammatory role for plasmacy-toid dendritic cells and IA-2 autoantibodies inT1D (Allen et al. 2009). These studies may leadto the identification of novel naturally processedepitopes recognized by CD4þ T cells, which, inturn, may represent potential therapeutic agentsfor T1D, either in native form or as antagonisticaltered peptide ligands.

Insulin Autoantibodies

Insulin is a hormone produced by the pancreaticb cells, which is not only central to regulatingcarbohydrate and fat metabolism, but is alsocentral in its pathological role as a T1D auto-antigen. Insulin is the predominant secretoryproduct of pancreatic b cells whose autoim-mune destruction leads to insulin deficiency



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



and consequent metabolic decompensation ofglucose homeostasis (Nakayama et al. 2005). In-vestigations of the immunologically relevant re-gions of the insulin molecule conducted inNOD mice revealed that the 9–23 amino acidsequence of the insulin B chain (termed B9-23)and the effect of intracellular processing of mol-ecules, such as insulin, within the b cell can leadto formation of immunogenic epitopes (Fig. 1)(Crawford et al. 2011).

A high titer of insulin autoantibodies (IAA)at younger ages is consistent with the conceptthat these patients develop a more aggressive dis-

ease course. In particular, insulin autoantibod-ies levels .2000 nU/mL are almost exclusivelyfound inpatients who progressto T1D before age5 yr, and less than half of individuals developingT1D after age 15 yr carry detectable levels of IAA.

A study from the Finnish Type 1 DiabetesPrediction and Prevention Study, comprising alarge population of 2448 genetically at-risk chil-dren (Kimpimaki et al. 2001), showed that IAAsare usually the first islet autoantibody to appearin the natural history of T1D and a powerfulidentifier of disease progression in children fol-lowed from birth (Steck et al. 2011).

Tregs

CD8+

CD4+

orCD8+

MHC I

Unique processing ofinsulin, ChgA peptides

Trimolecularcomplex

Self-reactive T cell

InsulinChgAICA69IGRP

AIRE

mTEC

eTAC

MHCI or II

Cytokines (IFNγ,IL-1) chemokines,perforin

β Cell

APC

Figure 1. Unique processing of b-cell proteins may lead to antigenicity. The b cell itself not only produces targetantigens, but also it modifies molecules, such as insulin and ChgA, by cleavage at critical sites creating peptidesrecognized by pathogenic T-lymphocytes (Stadinski et al. 2010; Crawford et al. 2011). Processing of moleculessuch as insulin within the b cell generates peptides that are then taken up by antigen-presenting cells either aswhole deadb cells or specifically granules ofb cells for eventual further processing/presentation of islet peptidesto self-reactive T cells (Crawford et al. 2011). The trimolecular complex, involving an (MHC-presenting mol-ecule)/(peptide in appropriate “register”)/(T-cell receptor recognizing both), like a lock and key is an essentialrecognition unit for adaptive organ-specific autoimmunity. The fact that in the thymus the absence of this typeof processing combined with the low affinity of B:9-23 binding to IAg7 in register 3 may explain the escape ofinsulin-specific CD4þ T cells from the mechanisms that usually eliminate self-reactive T cells. Both regulatory(Treg) and effector T-lymphocytes are produced, and their balance is crucial for maintaining tolerance (Blue-stone and Boehmer 2006). Evidence suggests that the inappropriate MHC class I expression detected in pan-creatic islets in T1D may not be solely due to the effect of cytokines such as IFNg (Bottazzo et al. 1985). It isconceivable that this molecule may function as an indicator of cell stress, likely present in residual pancreaticislets from T1D patients, and may be recognized by a subset of T cells in an unusual interaction.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Glutamic Acid Decarboxylase (GAD)Autoantibodies

In an earlier seminal report, incubation of ratislets with radioactively labeled [35S]-methio-nine and subsequent immunoprecipitationof solubilized membranes with serum fromnewly diagnosed patients with T1D or controlsshowed that an antigen with a molecular weightof 64 kDa was precipitated by sera from T1Dpatients (Baekkeskov et al. 1989). The antibod-ies to this 64-kDa antigen were present in �80%of new onset patients and in pre-diabetics be-fore the appearance of clinical disease. The na-ture of the 64-kDa antigen remained unknownuntil the report by Solimena et al. (1988) show-ing autoantibodies to GABA-ergic neuronsand pancreatic b cells in an unusual conditiontermed Stiff Man syndrome. Glutamic aciddecarboxylase is the enzyme that catalyzes theconversion of glutamic acid to gamma aminobutyric acid (GABA), a potent inhibitory neu-rotransmitter. This led Baekkeskov et al. (1990)to rapidly identify GAD as the 64-kDa autoan-tigen in T1D (Karlsen et al. 1991; Atkinson et al.1993). Other molecular-related forms of GAD,such as the 67-kDa isoform, have subsequentlybeen identified (Hagopian et al. 1993). Autoan-tibodies against GAD are a predictor of progres-sion to overt diabetes. When coupled with in-sulin autoantibodies and islet cell antibodies(ICA), their ability to predict the likelihood ofdeveloping T1D in asymptomatic first-degreerelatives of T1D patients is quite high.

IA-2 (ICA512) Autoantibodies

The neuroendocrine antigen IA-2 (ICA512) is amajor autoantigen in T1D (Lan et al. 1994). It isan enzymatically inactive member of the tyro-sine phosphatase family, involved in regulatinginsulin secretion. Assessment of the presence ofIA-2 autoantibodies contributes to the predict-ability of the likelihood of developing T1D. Asshown by Verge et al. (1996) and others (Pietro-paolo et al. 2002; Achenbach et al. 2004), ICA512(IA-2) and its homolog IA-2b (phogrin) areboth neuroendocrine molecules. The deducedICA512(IA-2) cDNA sequence reveals a 979-

amino-acid protein with a single transmem-brane region and with significant homology tothe receptor-type PTP(RT-PTPase). A PTP ho-molog, termed phogrin, was subsequently iden-tified. Subcellular fractionation of insulinomatissue showed that both IA-2 and phogrin hada very similar cellular distribution to that of in-sulin and carboxypeptidase H, and these twomolecules are predominantly localized in the se-cretory granules of neuroendocrine cells (Was-meier and Hutton 1996; Mziaut et al. 2006).

Although the main immune reactive regionof the IA-2 molecule was thought to reside with-in its intracellular domain (amino acids 601–979), the inclusion of this region together withother reactive epitope regions of the molecule(encompassing amino acid residues 256–979)has permitted the development of highly accu-rate constructs to assess the presence of antibod-ies against IA-2 (Kawasaki et al. 1997). However,other research has also suggested that humoralautoimmunity against the intracellular domainof the molecule is related to a high risk of fasterT1D development (Fig. 2) (Morran et al. 2010).

Zinc Transporter Family Member 8 (ZnT8)Autoantibodies

ZnT8 is a member of the cation diffusion facil-itator family, with abundant expression in pan-creatic b cells, although it is also expressed inextra-pancreatic tissues (Chimienti et al. 2004;Wijesekara et al. 2009). In the b cell, it plays animportant physiological role because Zn, whichis highly concentrated in b cells, is needed fornormal insulin storage. b-Cell-specific deletionof ZnT8 in mice results in glucose intolerance,reduced b-cell zinc accumulation, and anoma-lous insulin granules, as well as blunted first-phase glucose-stimulated insulin secretion, re-duced insulin-processing enzyme transcripts,and increased pro-insulin levels (Wijesekaraet al. 2010). Its relevance as an important T1Dautoantigen was first described by Wenzlau et al.(2007), following an evaluation of 68 candidateislet autoantigens compiled from multidimen-sional analyses of microarray mRNA expressionprofiling. The assessment of the zinc transport-er ZnT8 (Slc30A8 encodes ZnT8) indicated that



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



it was targeted by autoantibodies in 60%–80%of new-onset T1D compared with ,2% ofhealthy controls, ,3% type 2 diabetic patients,and in up to 30% of patients with other T1D-associated autoimmune pathologies. Interest-ingly, ZnT8 antibodies were found in 26% ofT1D subjects who had not shown antibody pos-itivity to other commonly measured autoanti-gens such as glutamate decarboxylase (GAD),protein tyrosine phosphatase IA-2 (IA2), insu-lin (I), or in the assay for cytoplasmic islet cellantibodies (ICA). Further research has revealedpolymorphisms in ZnT8 that are relevant to itsrole as a major T1D autoantigen. There are threepolymorphic variants located in the intracellu-lar (carboxyl terminus) domain of the trans-porter protein, namely, Arg-325, Trp-325, andGln-325. Of these variants, Trp-325 (W) andArg-325 (R) have been shown to be the majorautoantigenic polymorphisms in T1D, and useof a construct containing the W and R variants(ZnT8WR) (Wenzlau et al. 2008) has provenits efficacy as a screen for T1D-associated hu-moral autoimmunity. More recently, a chimeric

construct containing amino acid residues 609-979 of the intracellular domain of IA2, linkedto peptides containing both ZnT8 W and Rpolymorphisms, has been successfully devel-oped and tested as a broader and more econom-ical screen to detect patients showing humoralautoimmunity against IA-2 and/or ZnT8 (Yuet al. 2010).

CONNECTING GENETICS WITHAUTOANTIBODIES

Type 1 diabetes is a complex polygenic diseasefor which there is a small number of genes withlarge effects (i.e., HLA) and a large numberof genes with small effects (Todd et al. 2007;Barrett et al. 2009). Risk of T1D progressionis conferred by specific HLA DR/DQ alleles(e.g., DRB1�03-DQB1�0201 [DR3] or DRB1�

04-DQB1�0302 [DR4]) (Eisenbarth 2007). Inaddition, HLA alleles such as DQB1�0602 areassociated with dominant protection from T1Din multiple populations (Pugliese et al. 1999).

1.00

A B

0.75

0.50

0.25

0.000 5 10 15 20 25

Follow-up time (yr)

0 5 10 15 20 25

Follow-up time (yr)

Cum

ulat

ive

prop

ortio

n of

T1D

-fre

e

1.00

0.75

0.50

0.25

0.00

Cum

ulat

ive

prop

ortio

n of

T1D

-fre

e

IA-2FL neg 51 41 27 11 5 1145IA-2FL pos

IA-2FL neg 38 32 22 10 6 14 3 0IA-2FL pos

Figure 2. (A) The rate of progression to T1D development in relatives negative for ICA512bdc carrying GAD65in the absence (dashed line) or presence (solid line) of autoantibodies reacting with IA-2 full-length (IA-2FL)(amino acids 1–979) (log rank, P ¼ 0.016). In this subgroup of relatives, the cumulative risk of developinginsulin-requiring diabetes in IA-2 full-length antibody-positive FDR was strikingly high, being 100% by 11 yr offollow-up in GAD65 antibody positives (log rank, P ¼ 0.026). (B) This effect was also observed in FDR selectedfor being negative for ICA512bdc having at least two islet autoantibodies in the absence (dashed line) or presence(solid line) of autoantibodies reacting with IA-2 full length (IA-2FL) (amino acids 1–979) (log rank, P ¼0.003). In these relatives, the cumulative risk of progressing to overt diabetes was 100% by 10 yr of follow-up (logrank, P ¼ 0.022). (Figure is from Morran et al. 2010; reprinted, with permission, from The Endocrine Society#2010.)



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



It is commonly accepted that HLA typing isnot the optimal primary screening tool for T1D,and it is not sufficient alone to predict the dis-ease onset. However, HLA genotyping in first-degree relatives of T1D probands can be useful.Aly et al. (2006) reported that risk for islet au-toimmunity drastically increased in DR3/4-DQ2/DQ8 siblings who shared both HLA hap-lotypes identical by descent with their diabeticproband sibling (63% by age 7, and 85% by age15) as compared with siblings who did not shareboth HLA haplotypes with their diabetic pro-band siblings (Fig. 3). These data indicate thatHLA genotyping at birth may identify individ-uals at very high risk of developing T1D, beforethe occurrence of clear signs of humoral auto-immunity and eventually overt disease. In addi-tion, we previously identified a phenotype insubjects with high-risk HLADQ haplotypeswho remained seronegative for islet autoanti-bodies even after significant length of follow-up during progression to the insulin-requiringT1D (Pietropaolo et al. 2002). Finally, in a morerecent study, Howson et al. (2011) examined therelationships between the presence of GAD andIA-2 autoantibodies with HLA genes and 3779

single-nucleotide polymorphisms (SNPs) in2531 subjects with childhood-onset T1D. Theresults of this comprehensive assessment indi-cated that GAD autoantibodies were associatedprimarily with HLA-DQB1. For IA-2 autoanti-bodies, the strongest association was with HLA-DRB1. However, there was no association be-tween the presence of antibodies against eitherIA-2 or GAD and the T1D high-risk genotype,HLA-DRB1�03/04, although surprisingly, therewas an association of IA-2 autoantibody posi-tivity with the T1D-protective allele HLA-DQB1�030. These results overall suggest thatthe presence of high-genetic-risk HLA haplo-types and the presence of islet autoantibodiesdo not necessarily follow an obligate correlationin the development of autoimmune T1D.

CLINICAL APPLICATIONS

Relevance as Predictors of Risk for T1D,Role of Age and Specificity

Combining both immunological and metabolicstrategies (e.g., oral glucose tolerance test or the

100A BShare 2

Share 0 or 1

Share 2

Share 0 or 1

Aut

oant

ibod

y po

sitiv

e (%

) 90

80

70

60

50

40

30

20

10

0

100

Dia

betic

(%

)

90

80

70

60

50

40

30

20

10

00.0 2.5 5.0 7.5 10.0 12.5 15.0

Age (yr)

0.0 2.5 5.0 7.5 10.0 12.5 15.0

Age (yr)

Share 2: 29 24 17 11 8 6 3Share 0or 1: 19 17 12 8 3 2 2

Share 2: 29 20 14 7 7 5 2Share 0or 1: 19 15 10 7 3 2 2

Figure 3. Haplotype sharing survival curves. Shown is progression to anti-islet autoimmunity (A) and type 1Adiabetes (B) in DR3/4-DQ8 siblings stratified by the number of HLA haplotypes shared with their probandsiblings. n ¼ 48 for A and B; error bars for all panels represent the SEM. (Figure from Aly et al. 2006; reprinted,with permission, from National Academy of Sciences # 2006.)



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



first-phase [1 þ 3 min] insulin response of anintravenous glucose tolerance) is helpful, thecurrent opinion is that type 1 diabetes progres-sion can be predicted with 80%–100% accuracywithin 5-and 10-yr follow-up, respectively (Xuet al. 2010).

A study on the Diabetes AutoimmunityStudy in the Young (DAISY) cohort showedthat 89% of children who progressed to T1Dhad two or more islet-related autoantibodies(Steck et al. 2011). Importantly, age of diagnosisof diabetes was strongly correlated with age of

appearance of first autoantibody and IAA levels.By life-table analysis (Fig. 4A), children exhibit-ing two or more autoantibodies showed a nearlylinear progression to diabetes (P , 0.0001).Children with persistently positive IAA levelshad a higher progression rate to overt T1D(100% by 5.6 yr) as compared with childrenwith fluctuating IAA levels (63% by the 10-yrfollow-up; P , 0.0001) (Fig. 4B). Finally, in chil-dren enrolled in the DAISY study followed to thedevelopment of diabetes onset, only high IAAtiter correlated with rapid progression to T1D

1 Ab+ IAA pers posFluctuat IAA

P < 0.0001

908070605040302010

0

17.5

15.0

12.5

10.0

7.5

5.0

2.5

0.00.017.515.012.510.07.55.02.50.5

0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

Dia

bete

s on

set a

ge

Dia

bete

s on

set a

ge

Predicted diabetes onset age

P < 0.0001r = 0.83

2.5 5.0 7.5 10.0 12.5 15.0 17.5

P < 0.0001r = 0.79

Predicted diabetes onset age

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5Years since first Ab+Years since first Ab+

With

out d

iabe

tes

(%)

With

out d

iabe

tes

(%)

100100A B

C D

9080706050403020100

0.0 2.5 5.0 7.5 10.0 12.5 15.5

1 Ab+ N = 74 55 26 13 8 3 1 IAA pers pos N = 20 9 2Fluctuat IAA N = 68 61 33 17 7 22 Ab+ N = 47 39 19 10 3 1

3 Ab+ N = 48 33 20 11 6 1

17.5

2 Ab+3 Ab+

P < 0.0001

Diabetes predicted age = –0.6–1.0* log(initial lAA) + 0.9* initial Ab number + 1.1* age first Ab+

Diabetes predicted age =2.6–1.3* log (mean IAA) + 0.8* age first Ab+

Figure 4. (A) Progression to diabetes in children positive for anti-islet autoantibodies (n ¼ 169). There was nosignificant difference in the progression rate between subjects with two or three positive antibodies. (B) Pro-gression to diabetes in children with persistently positive IAA levels and fluctuating IAA levels (n ¼ 88). (IAApers pos) Persistently positive IAA levels; (Fluctuat IAA) fluctuating IAA levels. (C) Predicted age of diagnosis ofdiabetes (initial IAA, GAD, and IA-2 levels; n ¼ 38). Analyses performed in all subjects who had their firstantibody measurement before 1.5 yr and progressed to diabetes. (D) Predicted age of diagnosis of diabetes(mean IAA, GAD, and IA-2 levels; n ¼ 38). Analyses performed in all subjects who had their first antibodymeasurement before 1.5 yr and progressed to diabetes. (Figure from Steck et al. 2011; reprinted, with permission,from the American Diabetes Association # 2011.)



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



onset (Fig. 5) (P , 0.0001). As a matter of fact,this effect was not evident with respect to thepresence of high GAD65 or IA-2/ICA512 auto-antibody titer (Fig. 6). Therefore, insulin auto-antibody levels at the time of diagnosis are in-versely related to the age of the patient, beinghighest in those ,5 yr of age, and hence, IAAsappear to be an early marker of b-cell destruc-tion. The titer of insulin autoantibodies alongwith the insulin secretory response judged bythe first-phase insulin levels at 1 and 3 min afteran intravenous glucose challenge, has also beensuccessfully used to construct mathematicalmodels to predict the likelihood of clinical dia-betes in asymptomatic first-degree relatives ofpatients (Eisenbarth 1986). Investigators fromthe DAISY study reported that five childrenwere found to have persistent IAAs before 1 yrof age, and four of them went on to develop theclinical onset of T1D (all before 3.5 yr of age). Incontrast, children not exhibiting persistent IAAsbefore the age of 1 yr rarely rapidly developed aninsulin requirement. When analyzing only chil-dren followed from birth who progressed todiabetes, the two major predictors of age of dia-betes diagnosis were the age at which auto-antibodies first appeared and the mean level ofinsulin autoantibodies (Fig. 4A,B). These obser-vations emphasize the utilityof IAAs, particular-

ly inyounger populations, and justify the need todesign trials focused on such a young group.However, the success of these strategies dependson the safetyand effectiveness of therapeutic reg-imens. Indeed, this was the strategy in the trial toprevent development of T1D (DPT-1) that suc-cessfully predicted the development of diabetesin first-degree relatives of T1D patients (Skyleret al. 2001).

15 R2 = 0.37 P < 0.0001

10

5

0–3.5 –3.0 –2.5 –2.0 –1.5 –1.0 –0.5 0.0

Mean insulin Abs (log10)

Tim

e to

dia

bete

s fr

om a

ge fi

rst A

b+

Figure 5. Mean log IAA versus time to diabetes pro-gression in first-degree relatives of T1D probands. Inthese children from the DAISY study followed to thedevelopment of diabetes onset, high IAA levels cor-related with rapid progression to T1D onset.

17.5A

B

15.0

12.5

10.0

7.5

5.0

2.5

0.0–3.5 –3.0 –2.5 –2.0 –1.5 –1.0 –0.5 0.0

Mean GAD65 Ab levels (log10)

–4.0 –3.0 –2.0 –1.0 0.0

Mean ICA512 Ab levels (log10)

Tim

e to

dia

bete

s fr

om a

ge fi

rst A

b+

17.5

15.0

12.5

10.0

7.5

5.0

2.5

0.0

Tim

e to

dia

bete

s fr

om a

ge fi

rst A

b+

Figure 6. (A) GAD65, and (B) IA-2/ICA12 antibodytiters in children from the DAISY study followed tothe development of diabetes onset. There is no cor-relation with high GAD65 or IA-2/ICA512 titers andrapid progression to diabetes.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Detection of Novel Forms and Subgroupsof Diabetes, the LADA Example

Although the detection of islet autoantibodies isoverall the major diagnostic hallmark of auto-immune T1D risk assessment and progression,the ability to detect these autoantibodies hasalso allowed the characterization of unique sub-populations of the disease, with evident impli-cations for therapeutic strategies. The specificcharacterization of the diverse subtypes of dia-betes has been a dynamic field of research and anactive area of discussion. Awell-documented ex-ample is the case of patients who are generallyadults, who present a type 2 diabetes phenotypeas well as circulating islet autoantibodies (Zim-met et al. 1994). These characteristics are definedas Latent Autoimmune Diabetes of Adults(LADA) and sometimes are termed type 1.5 di-abetes (Palmer and Hirsch 2003). The immuno-logical diagnosis of LADA relies primarily on thedetection of autoantibodies against GAD65 inthe serum of clinically diagnosed T2D patientswho also show impaired insulin secretion and ahigh frequency of being on insulin treatment.Additionally, it has been previously reportedthat LADApatients possess T cells reactive to isletproteins as is also the case in “canonical” type 1diabetes, in contrast to classic autoantibody-neg-ative T2D patients (Brooks-Worrell et al. 2011).

The presence of GAD65 and/or IA-2 auto-antibodies in T2D patients diagnosed by con-ventional criteria (ADA or WHO) is not un-common among older diabetics, being 5%–10% or higher, especially in those on insulintherapy (Barinas-Mitchell et al. 2004, 2006; Les-lie et al. 2006; Pietropaolo et al. 2007). Thus,there may be as many GAD65 antibody-positiveolder diabetics as there are children affected byT1D. This is not a trivial issue. Moreover, addi-tional immunological markers of autoimmunediabetes might identify even a larger sample ofclinically diagnosed T2DM patients. Given itsrelative simplicity, testing for GAD65 and IA-2autoantibodies should be part of the diagnos-tic assessment for clinically diagnosed T2D be-cause it might predict the rate of progression toinsulin requirement in older populations. Thosefound to be GAD65 antibody-positive are prob-

ably candidates for early insulin therapy or moreaggressive oral therapies to lower their glycohe-moglobin levels. Clinical trials and epidemio-logical observational studies should clearly sep-arate GAD65 antibody-positive older diabeticsfrom those who are GAD65 antibody-negative.

Relationship with Other AutoimmuneDiseases (Altered Antigens)

During immune system development, lympho-cytes that react to self-antigens in the thymusand bone marrow are deleted. However, hostmolecules, in particular proteins and nucleic ac-ids, are constantly being modified in the courseof normal physiological events. A key posttrans-lational modification in autoimmunity appearsto be the citrullination of arginine amino acidresidues, by the enzymatic deimination of argi-nine to citrulline (Eggleton et al. 2008; Doyle andMamula 2012). This reaction is catalyzed by theenzyme peptidyl arginine deiminase (PAD) (So-derlin et al. 2004). In multiple sclerosis and RA,citrullinated isoforms of myelin basic protein(Moscarello et al. 2007) and fibrin (Masson-Bes-siere et al. 2001) have been found in the brain andsynovia, respectively. It must be pointed out thatthe detection of anti-citrullinated protein anti-bodies (ACPA) has proven extremely useful inthe early diagnosis and assessment of prognosisin rheumatoid arthritis (RA) and has also led toinsights into gene environment effects in auto-immune disease (Kastbom et al. 2004).

With regard to T1D, processing of moleculessuch as insulin within the b cell generates pep-tides that are then taken up by APCs either aswhole dead b cells or specifically granules ofb cells for eventual further processing/presen-tation of islet peptides to self-reactive T cells(Fig. 1) (Crawford et al. 2011). Furthermore,Stadinski et al. (2010) have shown that chro-mogranin A (ChgA) is an autoantigen in T1D(Table 1), and that the peptide WE14 from ChgAstimulates diabetogenic CD4þT-cell clones. Thenatural form of the antigen in b-cell extracts isfar more potent than an unmodified syntheticWE14 peptide, suggesting that this peptide maybe posttranslationally modified with a carbonylgroup in murine pancreatic islets.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



It is quite possible that the numberof report-ed altered neo-antigens will increase in T1D,because the attendant hyperglycemic and pro-oxidative metabolic milieu includes abnormalglycosylations and oxidative damage to proteins.The latter are processes that affect autoantigenicproteins in other diseases such as RA and SLE,and thus provide a conceptual relationship ofdiabetes with other autoimmune diseases withrespect to autoantigenesis.

How Do Autoantibody Specificities Relateto T-Cell Response and T-Cell Specificity?

A widening body of both basic and clinical in-vestigations has tried to address the relationshipbetween islet autoantibodies and the activationof T cells by cognate autoantigens in T1D. Anearly report was a study by Keller (1990), whodescribed proliferative responses to human in-sulin in T-lymphocytes during the peri-diabeticperiod, in a significant proportion (67%) ofICA-positive, first-degree relatives of T1D pa-tients who had never been treated with insu-lin, whereas in contrast, none of the controlsshowed the proliferative effect. Interestingly,the T-cell responses were not correlated withlevels of insulin autoantibodies. Similarly, stud-ies addressing other islet autoantigens by Atkin-son et al. (1992) reported that there was a higherlikelihood of a proliferative response to GAD inperipheral-blood mononuclear cells from pa-tients with T1D and in relatives positive forICA, compared with that seen in healthy con-trols and ICA-negative relatives of the patients.Other T1D-associated autoantigens targeted byhumoral autoimmunity have also been shownto stimulate T cells, such as the cases of T-cellproliferation in response to IA-2 (Durinovic-Bello et al. 1996; Roep et al. 1996; Reijonenet al. 2004). More recently, cellular immuno-blot, U.K.-ELISPOT, and T-cell proliferation as-says seemed to distinguish responses from pa-tients with type 1 diabetes and healthy controlsubjects (Herold et al. 2009). Albeit many at-tempts have been made, the demonstration inthe clinical setting of the relationship between Tcell and autoantibody responses is difficult forat least the following reasons: (1) The frequency

of antigen-specific autoreactive CD4þ T cells isvery low in peripheral blood. (2) There might bevariability in their appearance in peripheralblood. (3) The methodologies to detect theseT cells can be laborious (i.e., some methodolo-gies require expansion of CD4þ T cells withantigen for up to 10 d). (4) There is very ahigh degree of variability inherent to T-cell pro-liferation assays.

Does Ig Isotype Matter?

It is well known that pro-inflammatory antibod-ies of the human IgG1 and IgG3 subclass cancause complement activation, attract lympho-cytes to the target organ, and are used as surro-gate markers of disease activity (Gomez et al.2010). In T1D, a study investigating Ig isotypesagainst insulin, assessed insulin autoantibody(IAA) isotypes in children who were at geneticrisk of T1D and encompassed IAA-positive chil-dren who progressed to T1D as well as non-progressor infants (Hoppu et al. 2004a). Inthis study, it was found that children who pro-gressed rapidly to T1D showed strong IgG1 andIgG3 immunoreactivity to insulin, in contrastto a weak or absent IgG3 response, which wasdeemed to confer relative protection. Hoppuet al. (2004b) examined progression to onset,among non-diabetic first-degree relatives ofchildren with T1D and found the GAD65 iso-type response, and concluded that the detectionof isotypes against GAD65 could not reliablydiscriminate progressors from non-progressorsto diabetes onset among antibody-positive sib-lings of children with T1D.

In summary, the characterization of autoan-tibody isotypes against the islet autoantigens inthe estimation of T1D risk has revealed a diver-sity of results that need a body of confirmatoryreports before a consensus can be achieved ontheir applicability for clinical practice.

EVIDENCE FOR INVOLVEMENT OFB-LYMPHOCYTES AND AUTOANTIBODIESIN PATHOGENESIS

An accurate account of the relevance of B cells inautoimmune diabetes was shown earlier in



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



seminal studies using the NOD mouse to gen-erate B-lymphocyte-deficient mice (Hansonet al. 1996) and also through the use of mono-clonal antibodies to deplete the population of Bcells (Noorchashm et al. 1997). In these studies,the inhibition of B cells resulted in the concom-itant inhibition of insulitis. The physiologicalsignificance and potential clinical relevance ofthese early studies have now been validated andunderscored by the use of an anti-CD20 anti-body (rituximab) proved highly promising notonly in the treatment of T1D but also in otherautoimmune disorders that are oftentimes co-incident with T1D, such as autoimmune thy-roiditis and Stiff Person syndrome, in whichhigh-titer GAD autoantibodies are a diagnosticidentifier (Dupond et al. 2010). The use of ri-tuximab in human clinical trials has also stim-ulated the interest in B-cell antigen capture andpresentation in T1D, particularly following theresults of the rituximab trial in newly diagnosedT1D patients showing delay in disease progres-sion, associated with transient B-cell depletion(Pescovitz et al. 2009). These observations inhumans are reinforced by studies in NODmice using specific mAbs for CD20 (Xiu et al.2008) and by in vivo neutralization of the B-lymphocyte stimulator (BLyS/BAFF), whichin both cases prevented autoimmune diabetesprogression l (Zekavat et al. 2008). Interestingly,long-term in vivo BLyS neutralization also ledto reduction of IAA titers. This effect in anti-BLyS-treated NOD mice also correlated with arestoration of Igl clonotype negative selectionat the lymphopenic transitional B-cell com-partment (TR) ! splenic follicle (TR! FO)checkpoint. This is a defect, which has beendelineated as an aberrant characteristic of B-lymphocyte homeostasis in NOD mice (Zeka-vat et al. 2008). These data collectively suggestthat long-term in vivo BLyS neutralization iscapable of correcting a defect in NOD B-celltolerance by increasing the stringency of nega-tive selection at the TR ! FO checkpoint. Tworecent reports have shed additional light on thebroad physiological meaning of B-cell inhibi-tion in autoimmune diabetes by anti-CD20therapy, because the study showed that rituxi-mab differentially inhibited anti-islet autoanti-

bodies in T1D patients, and it blocked IAA for.1 yr, in insulin-treated patients (Yu et al. 2011).Finally, the murine CD20-specific 18B12 anti-body that, like rituximab, depletes the follicular(FO) but not marginal zone subset of B cells(MZB), efficiently inhibited diabetes develop-ment in NOD mice in a likely regulatory T-cell-dependent manner only when treatment wasinitiated before IAA detection. The latter obser-vations suggest that MZBs, which are known tobe potent APCs, may well play a role in the chainof events leading to overt diabetes (Serreze et al.2011).

Potential Role of Humoral Autoimmunityin Non-b-Cell Tissues

Humoral autoimmunity in T1D has beenshown to target pancreatic nervous system tissuestructures, suggesting the possibility that non-b-cell elements can elicit immune responses inT1D (Rabinowe et al. 1989; Winer et al. 2003;Louvet et al. 2009). One of these molecular tar-gets seems to be peripherin (Boitard et al. 1992).This molecule is expressed in multiple endo-crine tissues, including nerve fibers surroundingislets of Langerhans in the pancreas, adrenal me-dulla, nerve fibers in interstitial tissue betweenthyroid follicles, and nerve fibers adjacent toovarian follicles (Chamberlain et al. 2010). Se-rologic responses to peripherin have been foundin autonomic fibers in the pancreas, thyroid,and ovary, supporting clinical observations in-dicating that neuronal elements may be a mo-lecular target for immune-mediated injury inmultiple forms of endocrine autoimmunity, in-cluding T1D (Chamberlain et al. 2010). How-ever, it remains to be established as to whetheror not the presence of peripherin antibodies,along with serologic responses to other putativeneuronal elements, are predictive for the devel-opment of small fiber neuropathy (autonomicand/or somatic) and for the progression toovert diabetes.

Although the presence of these autoantibod-ies reacting with neuronal tissues is thought-provoking and may somehow play a role in thepathophysiology of a subset of diabetic neurop-athy, definite data establishing their role as



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



causal factors remains elusive (Ejskjaer et al.1998). The examination of inflammatory re-sponses in peripheral and autonomic nervousstructures is becoming a growing area of inves-tigation.

CONCLUDING REMARKS

Type 1 diabetes results from autoimmune de-struction/dysfunction of pancreatic b cells. Inphysiological conditions, there is balance be-tween pathogenic T cells that mediate diseasesuch as T cells with marked conservation oftheir TCRs (e.g., insulin), and regulatory cellsthat control autoimmunity (Bluestone et al.2010). In T1D and other autoimmune disor-ders, there is an altered balance between patho-genic and regulatory T cells.

Autoantibodies are some of the most po-tent risk determinants for autoimmune diseasessuch as T1D with an estimated relative risk ex-ceeding 100%. The archetypical model for theapplication of autoantibodies in disease predic-tion is T1D. Humoral bioassays have achieved ahigh degree of accuracy and sophistication toallow the design of algorithms for inclusion ofindividuals at risk of developing T1D in clinicaltrials around the world. In particular, seminalstudies have suggested that using a combinationof humoral immunological markers gives ahigher predictive value for T1D progression,and greater sensitivity, without significant lossof specificity. There is a growing effort and alarge opportunity for exploring novel strategiesalone or in combination with immunomodula-tion with the ultimate goal to find the cure forT1D. Development of therapies targeting spe-cific B-lymphocytes is underway in animalmodels of autoimmune diabetes and eventuallymay be examined in humans. Even though thepossible pathogenic role of islet-associated au-toantibodies has not been elucidated, the list ofuseful biomarkers such as autoantigen-specificCD4þ and CD8þ effectors, memory and regu-latory T cells related to disease pathogenesis(Khadra et al. 2011), and their relationshipwith the relevance of cognate autoantibodieswill continue to expand.

ACKNOWLEDGMENTS

This work is supported by the National Insti-tutes of Health grants R01 DK53456, DK56200,NIDDK PA-04-081 (to M.P.) and NationalInstitutes of Health grants R37-DK32493, DK320083, DK050979, DK57516, AI050864, andN01-AI15416 (to G.E.). We greatly acknowledgethe Brehm Coalition for its support.

REFERENCES

Achenbach P, Warncke K, Reiter J, Naserke HE, Williams AJ,Bingley PJ, Bonifacio E, Ziegler AG. 2004. Stratificationof type 1 diabetes risk on the basis of islet autoantibodycharacteristics. Diabetes 53: 384–392.

Allen JS, Pang K, Skowera A, Ellis R, Rackham C, Lozanoska-Ochser B, Tree T, Leslie RD, Tremble JM, Dayan CM, et al.2009. Plasmacytoid dendritic cells are proportionally ex-panded at diagnosis of type 1 diabetes and enhance isletautoantigen presentation to T-cells through immunecomplex capture. Diabetes 58: 138–145.

Aly TA, Ide A, Jahromi MM, Barker JM, Fernando MS, BabuSR, Yu L, Miao D, Erlich HA, Fain PR, et al. 2006. Ex-treme genetic risk for type 1A diabetes. Proc Natl Acad Sci103: 14074–14079.

Atkinson MA, Eisenbarth GS. 2001. Type 1 diabetes: Newperspectives on disease pathogenesis and treatment. Lan-cet 358: 221–229.

Atkinson MA, Kaufman DL, Campbell L, Gibbs KA, ShahSC, Bu D-F, Erlander MG, Tobin AJ, Maclaren NK. 1992.Response of peripheral-blood mononuclear cells to glu-tamate decarboxylase in insulin-dependent diabetes.Lancet 339: 458–459.

Atkinson MA, Kaufman DL, Newman D, Tobin AJ, Macla-ren NK. 1993. Islet cell cytoplasmic autoantibody reac-tivity to glutamate decarboxylase in insulin-dependentdiabetes. J Clin Invest 91: 350–356.

Baekkeskov S, Warnock G, Christie M, Rajotte RV, LarsenPM, Fey S. 1989. Revelation of specificity of 64K autoan-tibodies in IDDM serums by high-resolution 2-D gelelectrophoresis: Unambiguous identification of 64K tar-get antigen. Diabetes 38: 1133–1141.

Baekkeskov S, Aanstoot H, Christgau S, Reetz A, SolimenaMS, Cascalho M, Folli F, Richter-Olsen H, DeCamilli P.1990. Identification of the 64K autoantigen in insulindependent diabetes as the GABA-synthesizing enzymeglutamic acid decarboxylase. Nature 347: 151–156.

Barinas-Mitchell E, Pietropaolo S, Zhang YJ, Henderson T,Trucco M, Kuller LH, Pietropaolo M. 2004. Islet cell au-toimmunity in a triethnic adult population of the Na-tional Health and Nutrition Examination Survey(NHANES) III. Diabetes 53: 1293–1302.

Barinas-Mitchell E, Kuller LH, Pietropaolo S, Zhang YJ,Henderson T, Pietropaolo M. 2006. The prevalence ofthe 65-kilodalton isoform of glutamic acid decarboxylaseautoantibodies by glucose tolerance status in elderly pa-tients from the cardiovascular health study. J Clin Endo-crinol Metab 91: 2871–2877.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Barrett JC, Clayton DG, Concannon P, Akolkar B, CooperJD, Erlich HA, Julier C, Morahan G, Nerup J, Nierras C,et al. 2009. Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes.Nat Genet 41: 703–707.

Bluestone JA, Boehmer H. 2006. Regulatory T cells. SeminImmunol 18: 77.

Bluestone JA, Herold K, Eisenbarth G. 2010. Genetics, path-ogenesis and clinical interventions in type 1 diabetes.Nature 464: 1293–1300.

Boitard C, Villa MC, Becourt C, Gia HP, Huc C, Sempe P,Portier MM, Bach JF. 1992. Peripherin: An islet antigenthat is cross-reactive with nonobese diabetic mouse classII gene products. Proc Natl Acad Sci 89: 172–176.

Bonifacio E, Atkinson MA, Eisenbarth GS, Serreze DV, KayTW, Lee-Chan E, Singh B. 2002. International workshopon lessons from animal models for human type 1 diabe-tes: Analyzing target autoantigens of humoral immunityin nonobese diabetic mice. Ann NY Acad Sci 958: 1–2.

Bonifacio E, Yu L, Williams AK, Eisenbarth GS, Bingley PJ,Marcovina SM, Adler K, Ziegler AG, Mueller PW, SchatzDA, et al. 2010a. Harmonization of glutamic acid decar-boxylase and islet antigen-2 autoantibody assays for na-tional institute of diabetes and digestive and kidney dis-eases consortia. J Clin Endocrinol Metab 95: 3360–3367.

Bonifacio E, Yu L, Williams AK, Eisenbarth GS, Bingley PJ,Marcovina SM, Adler K, Ziegler AG, Mueller PW, SchatzDA, et al. 2010b. Harmonization of glutamic acid decar-boxylase and islet antigen-2 autoantibody assays for na-tional institute of diabetes and digestive and kidney dis-eases consortia. J Clin Endocrinol Metab 95: 3360–3367.

Bottazzo GF, Florin-Christensen A, Doniach D. 1974. Islet-cell antibodies in diabetes mellitus with autoimmunepolyendocrine deficiencies. Lancet 2: 1279–1282.

Bottazzo GF, Dean BM, McNally JM, Mackay EH, Swift PGF,Gamble DR. 1985. In situ characterization of autoim-mune phenomena and expression of HLA molecules inthe pancreas in diabetic insulitis. N Engl J Med 313:353–360.

Brooks-Worrell BM, Reichow JL, Goel A, Ismail H, PalmerJP. 2011. Identification of autoantibody-negative autoim-mune type 2 diabetic patients. Diabetes Care 34: 168–173.

Chamberlain JL, Pittock AJ, Oprescu AM, Dege C, Apiwat-tanakul M, Kryzer TJ, Lennon VA. 2010. Peripherin-IgGassociation with neurologic and endocrine autoimmuni-ty. J Autoimmun 34: 469–477.

Chimienti F, Devergnas S, Favier A, Seve M. 2004. Identifi-cation and cloning of a b-cell-specific zinc transporter,ZnT-8, localized into insulin secretory granules. Diabetes53: 2330–2337.

Crawford F, Stadinski B, Jin N, Michels A, Nakayama M,Pratt P, Marrack P, Eisenbarth G, Kappler JW. 2011. Spe-cificity and detection of insulin-reactive CD4þ T cells intype 1 diabetes in the nonobese diabetic (NOD) mouse.Proc Natl Acad Sci 108: 16729–16734.

Di Lorenzo TP, Peakman M, Roep BO. 2007. Translationalmini-review series on type 1 diabetes: Systematic analysisof T cell epitopes in autoimmune diabetes. Clin Exp Im-munol 148: 1–16.

Doyle HA, Mamula MJ. 2012. Autoantigenesis: The evolu-tion of protein modifications in autoimmune disease.Curr Opin Immunol 24: 112–118.

Dupond JL, Essalmi L, Gil H, Meaux-Ruault N, Hafsaoui C.2010. Rituximab treatment of Stiff-Person syndrome in apatient with thymoma, diabetes mellitus and autoim-mune thyroiditis. J Clin Neurosci 17: 389–391.

Durinovic-Bello I, Hummel M, Ziegler AG. 1996. Cellularimmune response to diverse islet cell antigens in IDDM.Diabetes 45: 795–800.

Eggleton P, Haigh R, Winyard PG. 2008. Consequence ofneo-antigenicity of the “altered self.” Rheumatology (Ox-ford) 47: 567–571.

Eisenbarth GS. 1986. Type I diabetes mellitus: A chronicautoimmune disease. N Engl J Med 314: 1360–1368.

Eisenbarth GS. 2007. Update in type 1 diabetes. J Clin En-docrinol Metab 92: 2403–2407.

Ejskjaer NT, Zanone MM, Peakman M. 1998. Autoimmu-nity in diabetic autonomic neuropathy: Does the im-mune system get on your nerves? Diabet Med 15: 723–729.

Gomez AM, Van Den Broeck J, Vrolix K, Janssen SP, Lem-mens MA, Van Der Esch E, Duimel H, Frederik P, Mole-naar PC, Martinez-Martinez P, et al. 2010. Antibody ef-fector mechanisms in myasthenia gravis-pathogenesis atthe neuromuscular junction. Autoimmunity 43: 353–370.

Hagopian WA, Michelsen B, Karlsen AE, Larsen F, Moody A,Grubin CE, Rowe R, Petersen J, McEvoy R, Lernmark A.1993. Autoantibodies in IDDM primarily recognize the65,000-M(r) rather than the 67,000-M(r) isoform of glu-tamic acid decarboxylase. Diabetes 42: 631–636.

Hanson MS, Cetkovic-Cvrlje M, Ramiya VK, Atkinson MA,Maclaren NK, Singh B, Elliott JF, Serreze DV, Leiter EH.1996. Quantitative thresholds of MHC class II I-E ex-pressed on hemopoietically derived antigen-presentingcells in transgenic NOD/Lt mice determine level of dia-betes resistance and indicate mechanism of protection.J Immunol 157: 1279–1287.

Herold KC, Brooks-Worrell B, Palmer J, Dosch HM, Peak-man M, Gottlieb P, Reijonen H, Arif S, Spain LM,Thompson C, et al. 2009. Validity and reproducibilityof measurement of islet autoreactivity by T-cell assaysin subjects with early type 1 diabetes. Diabetes 58:2588–2595.

Hoppu S, Ronkainen MS, Kimpimaki T, Simell S, KorhonenS, Ilonen J, Simell O, Knip M. 2004a. Insulin autoanti-body isotypes during the prediabetic process in youngchildren with increased genetic risk of type 1 diabetes.Pediatr Res 55: 236–242.

Hoppu S, Ronkainen MS, Kulmala P, Akerblom HK, KnipM. 2004b. GAD65 antibody isotypes and epitope recog-nition during the prediabetic process in siblings of chil-dren with type I diabetes. Clin Exp Immunol 136:120–128.

Howson JM, Stevens H, Smyth DJ, Walker NM, ChandlerKA, Bingley PJ, Todd JA. 2011. Evidence that HLA class Iand II associations with type 1 diabetes, autoantibodiesto GAD and autoantibodies to IA-2, are distinct. Diabetes60: 2635–2644.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Husebye ES, Anderson MS. 2010. Autoimmune polyendo-crine syndromes: Clues to type 1 diabetes pathogenesis.Immunity 32: 479–487.

Karlsen AE, Hagopian WA, Grubin CE, Dube S, DistecheCM, Adler DA, Barmeier H, Mathewes S, Grant FJ, FosterD, et al. 1991. Cloning and primary structure of a humanisoform of glutamic acid decarboxylase from chromo-some 10. Proc Natl Acad Sci 88: 8337–8341.

Kastbom A, Strandberg G, Lindroos A, Skogh T. 2004. Anti-CCP antibody test predicts the disease course during 3years in early rheumatoid arthritis (the Swedish TIRAproject). Ann Rheum Dis 63: 1085–1089.

Kawasaki E, Yu L, Gianani R, Verge CF, Babu S, Bonifacio E,Eisenbarth GS. 1997. Evaluation of islet cell antigen(ICA) 512/IA-2 autoantibody radioassay using overlap-ping ICA512/IA-2 constructs. J Clin Endocrinol Metab82: 375–380.

Keller RJ. 1990. Cellular immunity to human insulin inindividuals at high risk for the development of type Idiabetes mellitus. J Autoimmun 3: 321–327.

Khadra A, Pietropaolo M, Nepom GT, Sherman A. 2011.Investigating the role of T-cell avidity and killing efficacyin relation to type 1 diabetes prediction. PLoS ONE 6:e14796.

Kimpimaki T, Kupila A, Hamalainen AM, Kukko M, Kul-mala P, Savola K, Simell T, Keskinen P, Ilonen J, Simell O,et al. 2001. The first signs of b-cell autoimmunity appearin infancy in genetically susceptible children from thegeneral population: The Finnish Type 1 Diabetes Predic-tion and Prevention Study. J Clin Endocrinol Metab 86:4782–4788.

Lan MS, Lu J, Goto Y, Notkins AL. 1994. Molecular cloningand identification of a receptor-type protein tyrosinephosphatase, IA-2, from human insulinoma. DNA CellBiol 13: 505–514.

Leslie RD, Williams R, Pozzilli O. 2006. Clinical review: Type1 diabetes and latent autoimmune diabetes in adults: Oneend of the rainbow. J Clin Endocrinol Metab 91: 1654–1659.

Louvet C, Kabre BG, Davini DW, Martinier N, Su MA, De-voss JJ, Rosenthal WL, Anderson MS, Bour-Jordan H,Bluestone JA. 2009. A novel myelin P0-specific T cellreceptor transgenic mouse develops a fulminant autoim-mune peripheral neuropathy. J Exp Med 206: 507–514.

Martin S, Wolf-Eichbaum D, Duinkerken G, ScherbaumWA, Kolb H, Noordzij JG, Roep BO. 2001. Developmentof type 1 diabetes despite severe hereditary B-lymphocytedeficiency. N Engl J Med 345: 1036–1040.

Masson-Bessiere C, Sebbag M, Girbal-Neuhauser E, No-gueira L, Vincent C, Senshu T, Serre G. 2001. The majorsynovial targets of the rheumatoid arthritis-specific anti-filaggrin autoantibodies are deiminated forms of the a-and b-chains of fibrin 1. J Immunol 166: 4177–4184.

Morran MP, Casu A, Arena VC, Pietropaolo S, Zhang YJ,Satin LS, Nelson P, Omenn GS, Trucco M, Becker DJ, et al.2010. Humoral autoimmunity against the extracellulardomain of the neuroendocrine autoantigen IA-2 height-ens the risk of type 1 diabetes. Endocrinology 151: 2528–2537.

Moscarello MA, Mastronardi FG, Wood DD. 2007. The roleof citrullinated proteins suggests a novel mechanism in

the pathogenesis of multiple sclerosis. Neurochem Res 32:251–256.

Mziaut H, Trajkovski M, Kersting S, Ehninger A, AltkrugerA, Lemaitre RP, Schmidt D, Saeger HD, Lee MS, DrechselDN, et al. 2006. Synergy of glucose and growth hormonesignalling in islet cells through ICA512 and STAT5. NatCell Biol 8: 435–445.

Nakayama M, Abiru N, Moriyama H, Babaya N, Liu E, MiaoD, Yu L, Wegmann DR, Hutton JC, Elliott JF, et al. 2005.Prime role for an insulin epitope in the development oftype 1 diabetes in NOD mice. Nature 435: 220–223.

Noorchashm H, Noorchashm N, Kern J, Rostami SY, BarkerCF, Naji A. 1997. B-cells are required for the initiation ofinsulitis and sialitis in nonobese diabetic mice. Diabetes46: 941–946.

Palmer JP, Hirsch IB. 2003. What’s in a name: Latent auto-immune diabetes of adults, type 1.5, adult-onset, andtype 1 diabetes. Diabetes Care 26: 536–538.

Palmer JP, Asplin CM, Clemons P, Lyen K, Tatpati O, RaghuPK, Paquette TL. 1983. Insulin antibodies in insulin-de-pendent diabetics before insulin treatment. Science 222:1337–1339.

Peakman M, Stevens EJ, Lohmann T, Narendran P, DromeyJ, Alexander A, Tomlinson AJ, Trucco M, Gorga JC, ChiczRM. 1999. Naturally processed and presented epitopesof the islet cell autoantigen IA-2 eluted from HLA-DR4. J Clin Invest 104: 1449–1457.

Pescovitz MD, Greenbaum CJ, Krause-Steinrauf H, BeckerDJ, Gitelman SE, Goland R, Gottlieb PA, Marks JB,McGee PF, Moran AM, et al. 2009. Rituximab, B-lym-phocyte depletion, and preservation of b-cell function.N Engl J Med 361: 2143–2152.

Pietropaolo M, Eisenbarth GS. 2001. Autoantibodies in hu-man diabetes. Curr Dir Autoimmun 4: 252–282.

Pietropaolo M, Becker DJ, LaPorte RE, Dorman JS, RiboniS, Rudert WA, Mazumdar S, Trucco M. 2002. Progressionto insulin-requiring diabetes in seronegative prediabeticsubjects: The role of two HLA-DQ high risk haplotypes.Diabetologia 45: 66–76.

Pietropaolo M, Yu S, Libman IM, Pietropaolo SL, Riley K,LaPorte RE, Drash AL, Mazumdar S, Trucco M, BeckerDJ. 2005. Cytoplasmic islet cell antibodies remain valu-able in defining risk of progression to type 1 diabetes insubjects with other islet autoantibodies. Pediatr Diabetes6: 184–192.

Pietropaolo M, Barinas-Mitchell E, Kuller LH. 2007. Theheterogeneity of diabetes: Unraveling a dispute: Is sys-temic inflammation related to islet autoimmunity? Dia-betes 56: 1189–1197.

Pugliese A, Kawasaki E, Zeller M, Yu L, Babu S, Solimena M,Moraes CT, Pietropaolo M, Friday RP, Trucco M, et al.1999. Sequence analysis of the diabetes-protective humanleukocyte antigen-DQB1�0602 allele in unaffected, isletcell antibody-positive first degree relatives and in rarepatients with type 1 diabetes. J Clin Endocrinol Metab84: 1722–1728.

Rabinowe SL, Brown FM, Watts M, Kadrofske MM, VinikAI. 1989. Anti-sympathetic ganglia antibodies and pos-tural blood pressure in IDDM subjects of varying dura-tion and patients at high risk of developing IDDM. Dia-betes Care 12: 1–6.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Reijonen H, Mallone R, Heninger AK, Laughlin EM, KochikSA, Falk B, Kwok WW, Greenbaum C, Nepom GT. 2004.GAD65-specific CD4þ T-cells with high antigen avidityare prevalent in peripheral blood of patients with type 1diabetes. Diabetes 53: 1987–1994.

Roep BO, Duinkerken G, Schreuder GM, Kolb H, De VriesRR, Martin S. 1996. HLA-associated inverse correlationbetween T cell and antibody responsiveness to islet auto-antigen in recent-onset insulin dependent diabetes mel-litus. Eur J Immunol 26: 1285–1289.

Serreze DV, Chapman HD, Niens M, Dunn R, Kehry MR,Driver JP, Haller M, Wasserfall C, Atkinson MA. 2011.Loss of intra-islet CD20 expression may complicate effi-cacy of B-cell-directed type 1 diabetes therapies. Diabetes60: 2914–2921.

Shoda LK, Young DL, Ramanujan S, Whiting CC, AtkinsonMA, Bluestone JA, Eisenbarth GS, Mathis D, Rossini AA,Campbell SE, et al. 2005. A comprehensive review ofinterventions in the NOD mouse and implications fortranslation. Immunity 23: 115–126.

Skyler JS. 2011. Immune intervention for type 1 diabetesmellitus. Int J Clin Pract Suppl 2011: 61–70.

Skyler JS, Cefalu WT, Koudries IA, Landschulz WH, Balag-tas CC, Cheng SL, Gelfand RA. 2001. Efficacy of inhaledhuman insulin in type 1 diabetes mellitus: A randomisedproof-of-concept study. Lancet 357: 331–335.

Soderlin MK, Kastbom A, Kautiainen H, Leirisalo-Repo M,Strandberg G, Skogh T. 2004. Antibodies against cycliccitrullinated peptide (CCP) and levels of cartilage oligo-meric matrix protein (COMP) in very early arthritis: Re-lation to diagnosis and disease activity. Scand J Rheuma-tol 33: 185–188.

Solimena M, Folli F, Denis-Donini S, Comi GC, Pozza G, DeCamilli P, Vicari AM. 1988. Autoantibodies to glutamicacid decarboxylase in a patient with stiffman syndrome,epilepsy, and type I diabetes mellitus. N Engl J Med 318:1012–1020.

Stadinski BD, Delong T, Reisdorph N, Reisdorph R, PowellRL, Armstrong M, Piganelli JD, Barbour G, Bradley B,Crawford F, et al. 2010. Chromogranin A is an autoan-tigen in type 1 diabetes. Nat Immunol 11: 225–231.

Steck AK, Johnson K, Barriga KJ, Miao D, Yu L, Hutton JC,Eisenbarth GS, Rewers MJ. 2011. Age of islet autoanti-body appearance and mean levels of insulin, but not GADor IA-2 autoantibodies, predict age of diagnosis of type 1diabetes: Diabetes autoimmunity study in the young.Diabetes Care 34: 1397–1399.

Todd JA, Walker NM, Cooper JD, Smyth DJ, Downes K,Plagnol V, Bailey R, Nejentsev S, Field SF, Payne F, et al.2007. Robust associations of four new chromosome re-gions from genome-wide analyses of type 1 diabetes. NatGenet 39: 857–864.

Verge CF, Gianani R, Kawasaki E, Yu L, Pietropaolo M,Jackson RA, Chase PH, Eisenbarth GS. 1996. Predictionof type I diabetes mellitus in first degree relatives using acombination of insulin, glutamic acid decarboxylase andICA512bdc/IA-2 autoantibodies. Diabetes 45: 926–933.

Verge CF, Stenger D, Bonifacio E, Colman PG, Pilcher C,Bingley PJ, Eisenbarth GS. 1998. Combined use of auto-antibodies (IA-2 autoantibody, GAD autoantibody, in-sulin autoantibody, cytoplasmic islet cell antibodies) in

type 1 diabetes: Combinatorial Islet Autoantibody Work-shop. Diabetes 47: 1857–1866.

Wasmeier C, Hutton JC. 1996. Molecular cloning of phog-rin, a protein tyrosyne phosphatase homologue localizedto insulin secretory granule membranes. J Biol Chem 271:18161–18170.

Wenzlau JM, Juhl K, Yu L, Moua O, Sarkar SA, Gottlieb P,Rewers M, Eisenbarth GS, Jensen J, Davidson HW, et al.2007. The cation efflux transporter ZnT8 (Slc30A8) is amajor autoantigen in human type 1 diabetes. Proc NatlAcad Sci 104: 17040–17045.

Wenzlau JM, Moua O, Liu Y, Eisenbarth GS, Hutton JC,Davidson HW. 2008. Identification of a major humoralepitope in Slc30A8 (ZnT8). Ann NYAcad Sci 1150: 252–255.

Wijesekara N, Chimienti F, Wheeler MB. 2009. Zinc, a reg-ulator of islet function and glucose homeostasis. DiabetesObes Metab 11: 202–214.

Wijesekara N, Dai FF, Hardy AB, Giglou PR, BhattacharjeeA, Koshkin V, Chimienti F, Gaisano HY, Rutter GA,Wheeler MB. 2010. b cell-specific Znt8 deletion inmice causes marked defects in insulin processing, crys-tallisation and secretion. Diabetologia 53: 1656–1668.

Winer S, Tsui H, Lau A, Song A, Li X, Cheung RK, SampsonA, Afifiyan F, Elford A, Jackowski G, et al. 2003. Autoim-mune islet destruction in spontaneous type 1 diabetes isnot b-cell exclusive. Nat Med 9: 198–205.

Xiu Y, Wong CP, Bouaziz JD, Hamaguchi Y, Wang Y, Pop SM,Tisch RM, Tedder TF. 2008. B lymphocyte depletion byCD20 monoclonal antibody prevents diabetes in non-obese diabetic mice despite isotype-specific differencesin FcgR effector functions. J Immunol 180: 2863–2875.

Xu P, Wu Y, Zhu Y, Dagne G, Johnson G, Cuthbertson D,Krischer JP, Sosenko JM, Skyler JS. 2010. Prognostic per-formance of metabolic indexes in predicting onset of type1 diabetes. Diabetes Care 33: 2508–2513.

Yu L, Liu Y, Miao D, Wenzlau J, Davidson H, Hutton J,Eisenbarth GS. 2010. Triple chimeric islet autoantigenIA2-ZnT8WR to facilitate islet autoantibody determina-tion. J Immunol Methods 353: 20–23.

Yu L, Herold K, Krause-Steinrauf H, McGee PL, Bundy B,Pugliese A, Krischer J, Eisenbarth GS. 2011. Rituximabselectively suppresses specific islet antibodies. Diabetes60: 2560–2565.

Yu L, Miao D, Scrimgeour L, Johnson K, Rewers M, Eisen-barth GS. 2012. Distinguishing persistent insulin auto-antibodies with differential risk: Nonradioactive bivalentproinsulin/insulin autoantibody assay. Diabetes 61:179–186.

Zekavat G, Rostami SY, Badkerhanian A, Parsons RF, Koe-berlein B, Yu M, Ward CD, Migone TS, Yu L, EisenbarthGS, et al. 2008. In vivo BLyS/BAFF neutralization ame-liorates islet-directed autoimmunity in nonobese diabet-ic mice. J Immunol 181: 8133–8144.

Zimmet PZ, Tuomi T, Mackay IR, Rowley MJ, Knowles W,Cohen M, Lang A. 1994. Latent autoimmune diabetesmellitus in adults (LADA): The role of antibodies to glu-tamic acid decarboxylase in diagnosis and prediction ofinsulin dependency. Diabet Med 11: 299–303.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



August 17, 20122012; doi: 10.1101/cshperspect.a012831 originally published onlineCold Spring Harb Perspect Med

Massimo Pietropaolo, Roberto Towns and George S. Eisenbarth and Detection of Distinct Disease SubtypesHumoral Autoimmunity in Type 1 Diabetes: Prediction, Significance,

Subject Collection Type I Diabetes

DiabetesThe Pathogenesis and Natural History of Type 1

Mark A. Atkinson Disease SubtypesPrediction, Significance, and Detection of Distinct Humoral Autoimmunity in Type 1 Diabetes:

S. EisenbarthMassimo Pietropaolo, Roberto Towns and George

Diabetes and Other Autoimmune Diseases?Do MHCII-Presented Neoantigens Drive Type 1

Philippa Marrack and John W. KapplerDegeneration, and Diabetes

-CellβEndoplasmic Reticulum Stress, Pancreatic

Feroz R. Papa

Treatment of Type 1 DiabetesClinical Immunologic Interventions for the

Anette-G. ZieglerLucienne Chatenoud, Katharina Warncke and

Localization, and RegulationIslet Autoantigens: Structure, Function,

al.Peter Arvan, Massimo Pietropaolo, David Ostrov, et

Update on Islet TransplantationMichael McCall and A.M. James Shapiro

Environmental Triggers of Type 1 DiabetesMikael Knip and Olli Simell

Destruction in the Diagnosis of Type 1 Diabetes-CellβImmunologic and Metabolic Biomarkers of

Jasmin Lebastchi and Kevan C. HeroldFar

The Story So−− Cells from Stem CellsβGenerating

Matthias Hebrok

Tissues in Immunodeficient MiceDiabetes by Engraftment of Functional Human Advancing Animal Models of Human Type 1

Shultz, et al.Michael A. Brehm, Alvin C. Powers, Leonard D.

Antigen Targets of Type 1 Diabetes AutoimmunityBart O. Roep and Mark Peakman

Diabetes in Mice and HumansBreakdown in Peripheral Tolerance in Type 1

BluestoneLukas T. Jeker, Hélène Bour-Jordan and Jeffrey A.

Innate ImmunityConnecting Type 1 and Type 2 Diabetes through

Justin I. Odegaard and Ajay Chawla

1 DiabetesAntigen-Specific Therapeutic Approaches in Type

et al.Xavier Clemente-Casares, Sue Tsai, Carol Huang, Diabetes

Increased Frequency of Insulin-Dependent The Hygiene Hypothesis: An Explanation for the

Jean-François Bach and Lucienne Chatenoud

http://perspectivesinmedicine.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2012 Cold Spring Harbor Laboratory Press; all rights reserved


http://perspectivesinmedicine.cshlp.org/cgi/collection/

http://perspectivesinmedicine.cshlp.org/cgi/collection/


humoral autoimmunity in type 1 diabetes: prediction...

Documents