Diabetes, Obesity and Metabolism 15 (Suppl. 3): 82–88, 2013.
© 2013 John Wiley & Sons Ltd
Infectious triggers in type 1 diabetes: is there a case for
G. Afonso1,2& R. Mallone1,2,3
1Cochin Institute, INSERM U1016, DeAR Lab Avenir, Paris, France
2Sorbonne Paris Cit´ e, Facult´ e de M´ edecine, Paris Descartes University, Paris, France
3Assistance Publique Hˆ opitaux de Paris, Hˆ otel Dieu, Service de Diab´ etologie, Paris, France
Environmental factors are the main contributors to type 1 diabetes (T1D) pathogenesis, yet they remain unidentified. Enteroviruses are proposed
candidate triggers due to temporal correlations between infection and T1D autoimmunity and to detection of viral proteins in diseased islets.
However, such correlations are not universal and may be relatively uncommon. Furthermore, evidence of a cause–effect relationship is lacking,
as infection of non-obese diabetic mice with Coxsackie enteroviruses can either trigger or blunt disease. The proposed mechanisms are either
non-antigen-specific (i.e. β-cell destruction and release of sequestered antigens, islet inflammation) or antigen-specific (i.e. epitope mimicry,
by which immune responses to enteroviruses may be diverted against homologous β-cell antigens). The case for the latter mechanisms is
even less stringent, as there is little evidence of promiscuous antigen recognition at the single T-cell level. Other infectious agents may thus be
implicated. Demonstration of their role will require fulfilling the Koch’s postulates, namely isolation of the agent preferentially in T1D patients,
including before disease onset; and T1D induction when the agent is inoculated into mice. The same is needed for cross-reactive T cells to
support epitope mimicry mechanisms. Generation of alternative (humanized) mouse models that could be challenged with candidate microbes
Keywords: antigen, autoimmunity, enterovirus, epitope, NOD mouse, T cells
Date submitted 22 March 2013; date of final acceptance 22 May 2013
The Role of Environmental Factors in Type 1
Type 1 diabetes (T1D) is a multifactorial disease in which
environmental triggers act on a genetic background of
predisposition to engender islet autoimmunity, leading to β-
cell destruction and insulin dependency. What is the relative
Several lines of evidence converge to support the predominant
identified at different genetic loci. While three of them, namely
alleles at the DQB1, INS variable number of tandem repeats
(VNTR) and PTPN22 loci carry disease odds ratios >2, all
the other predisposing alleles identified confer a more modest
risk. Even the DQ2 and/or DQ8 haplotypes which are more
strongly associated with T1D are found in a large (∼75%)
proportion of T1D patients, yet also in healthy individuals.
[2,3]. Third, migrant studies document that T1D incidence
adapts to that of the hosting region rather than to that of
origin . Fourth, comparison between regions with the same
genetic background and different environmental exposures
shows largely discordant T1D prevalence. One paradigmatic
Correspondence to: Roberto Mallone, MD, PhD, DeAR Lab Avenir, INSERM U1016, Hˆ opital
Saint Vincent de Paul, 82 avenue Denfert Rochereau, 75674 Paris Cedex 14, France.
example is given by urbanized Finland and the nearby rural
Russian Republic of Karelia, which host genetically similar
human populations, yet display a ∼sixfold higher incidence
of autoantibody positivity and T1D in Finland . Fifth,
the skyrocketing T1D incidence (∼4% increase/year in most
Western countries) and the shift towards younger ages of
diagnosis cannot be explained by any major drift in genetic
susceptibility . Rather, several reports documented that
the genetic background of T1D has changed in the opposite
in recent T1D cases [6–8]. Taken together, these observations
suggest that the environment plays a major role in T1D
development and that this role is gaining importance.
Which Are the Environmental Factors at
Despite this evidence for the prominent role of environmental
most disappointing gap in the knowledge that several decades
of T1D research have not filled in. Several candidate triggers
have been called to attention.
The contribution of factors promoting insulin-resistant
states (e.g. diet and physical activity) is well documented, as
insulin resistance indexes and weight gain have been proposed
DIABETES, OBESITY AND METABOLISM
II haplotypes [9,10]. However, such factors are regarded as
accelerators rather than initiators. Autoimmune β-cell failure
is more likely to reach the critical threshold under conditions
of increased functional demand.
Enteroviral infections (e.g. coxsackievirus B4 and rotavirus)
have received overarching attention as candidate triggers.
Such attention stems from a number of epidemiological
studies, which documented a temporal correlation between
enteroviral infections and appearance of anti-islet auto-
antibodies (aAbs) [11,12]. The non-obese diabetic (NOD)
mouse model of T1D has further shown that enteroviral
infections may accelerate rather than initiate T1D progression,
because they are only effective only once autoimmune T cells
have already accumulated in the islets, otherwise exerting a
protective effect . Further supporting this conclusion for
humans, the Diabetes and Autoimmunity Study in the Young
(DAISY) showed that T1D progression in aAb+ children
was increased approximately sevenfold in the 4months after
detection of enteroviral RNA in serum, but not in rectal swabs
. These results need to be confirmed in larger cohorts,
because only three children progressed to T1D in the 4-
month interval after a positive serum enteroviral RNA test.
Nonetheless, it should also be noted that only 8% of the
children progressing to T1D had enteroviral RNA in their
serum, suggesting that ‘systemic’ infection leads to a more
rapid progression, but it may be relatively uncommon. On the
same lines, immunohistochemical studies have documented
coxsackievirus B4 capsid proteins in the pancreas of some T1D
observation comes from a large series of autoptic pancreatic
specimens, which showed that 61% of islets from recent-onset
T1D patients stain positive for the enteroviral capsid protein
vp1 versus 6% of healthy controls . However, similar
which may suggest that intra-islet enteroviral infection may
be a consequence rather than a cause of β-cell failure. Taken
together, these data underline that the association between
enteroviral infections and T1D is not universal, suggesting
that other infectious agents may be implicated; and that
definite demonstration of a cause–effect relationship between
enteroviruses and T1D is still lacking (see also below).
The proposed role of enteroviral infections may also be at
variance with the so-called ‘hygiene hypothesis’, according to
which infections could rather have a protective effect . The
specific pathogen-free conditions . This apparent paradox
could be reconciled by several factors affecting the protective
or promoting effect of viral infections on T1D, including that
of coxsackievirus in NOD mice . These factors include:
viral strains (e.g. coxsackievirus B4 vs. B3), which can display
β-cell antigens, different inflammatory potential and capacity
to develop into a chronic infection; viral load; timing of infec-
T cells and systemic and local inflammatory status at the time
of infection. These variables could explain the inductive or
protective effects of viral infections in mouse models and the
conflicting epidemiological observations in human T1D .
Another major environmental change that has occurred
of T1D has, however, been ruled out. In a population-based
studies of Danish children born between 1990 and 2000 ,
there was no evidence of increased T1D incidence associated
with vaccinations against Haemophilus influenzae, diphtheria,
tetanus, poliovirus, pertussis, measles, mumps and rubella,
siblings of T1D patients, as compared with unvaccinated
children. Furthermore, there was no evidence of any clustering
of cases 2–4years after vaccination with any of those vaccines.
Despite this lack of consensus about triggering environ-
mental factors, it has been suggested that the initial β-cell
antigen provision that primes autoreactive T cells and starts
the autoimmune cascade may come from a wave of neona-
tal β-cell apoptosis, which is part of the remodelling process
of many organs [20,21]. Indeed, this wave of neonatal β-cell
apoptosis takes place in the islets of young rodents, includ-
ing NOD mice, and reaches its zenith at 14–17days of age
. This wave of apoptosis correlates remarkably well with
the appearance of insulitis, which abruptly begins a few days
later in the NOD mouse. Indeed, the earliest signs of immune
infiltrates are detected at ∼15days in NOD mice and also in
accelerated transgenic models [20,23]. It is at this stage that
β-cell antigens can be detected in pancreatic lymph nodes. In
a recent report, Diana et al. further dissected the initial steps
by which β-cell death may initiate the autoimmune cascade
of T1D . These authors reported that this physiological
apoptosis induces the recruitment and activation of B cells (B-
1a subset), neutrophils and plasmacytoid dendritic cells (DCs)
to the pancreas of young NOD mice. Activated B-1a cells
secrete IgGs, which both directly bind double-stranded DNA
peptides, which also bind DNA. These immune complexes of
self-DNA, DNA-specific IgGs and DNA-binding antimicrobial
peptides subsequently activate plasmacytoid DCs, leading to
interferon-α production in pancreatic islets and propagation
of the inflammatory cascade.
A similar wave of perinatal apoptosis, which reaches
its maximum at birth, has been demonstrated in humans
. However, the initiation of productive (auto)immune
inflammatory signals that provide the immune system with a
‘danger’ cue driving DC activation (so called maturation)
and inducing a switch from steady state tolerance to active
further provide inflammatory signals for DC activation if they
had been activated or ‘stressed’ before undergoing apoptosis
how environmental factors enter this game and hence why,
despite physiological β-cell death, not all individuals develop
T1D. We may hypothesize that inflammatory signals may be
Volume 15Suppl. 3 September 2013doi:10.1111/dom.12166 83
DIABETES, OBESITY AND METABOLISM
of perinatal β-cell apoptosis, which would otherwise remain
a physiological phenomenon. Two types of environmental
islet-tropic infectious agents such as enteroviruses or other as
yet unidentified ones, which could induce local inflammation.
On the other, metabolic cues which could impose a functional
‘stress’ on β-cells, thus leading to immunogenic rather than
tolerogenic apoptosis. In other words, accessory inflammatory
signals may come from outside the β-cell (e.g. infections) or
play also in the ‘environment-protected’ NOD mouse models.
What type of metabolic cues could lead to β-cell stress and
immunogenic apoptosis? Weaning corresponds to a period of
high metabolic stress in mice, marking the transition from
a lipid-rich milk diet to a carbohydrate-rich chow diet. It is
thus possible that the metabolic stress imposed on β-cells by
these nutritional changes may add the critical inflammatory
of β-cell stress at weaning is so far available, increased β-cell
activity has been documented in neonatal NOD mice .
Could weaning have a similar effect in humans? Practices
have evolved towards a shorter duration of breast-feeding in
Western societies, which has been implicated as a risk factor
for β-cell autoimmunity and T1D [29,30]. However, the cow
milk used for infant formulas is actually poorer in fat and
lactose than human breast milk. It is thus unlikely that these
preparations may impose a functional stress on β-cells, and
their potential triggering role has been rather attributed to
early exposure to complex dietary proteins possibly eliciting a
molecular mimicry with β-cell antigens. This hypothesis has
been thoroughly tested in the past , but evidence remains
circumstantial at best. Additional candidate mechanisms may
involve reduced gut permeability, induction of the maturation
of regulatory T cells in the gut-associated lymphoid tissue
and/or modification of the gut microflora. Supporting these
possibilities, it was shown that a highly hydrolyzed formula
induced a decrease in autoimmune diabetes in the disease-
prone BioBreeding rat, in association with improved integrity
beneficial changes in gut microflora [32,33]. It is also possible
may be at play, as suggested by some animal models [34,35].
The protein at play does not seem to be (exclusively) gliadin,
which is a major target antigen in celiac disease, a condition
frequently associated with T1D. Indeed, diabetes development
is not completelyrestoredby gliadinsupplementationof cereal
protein-free diet in NOD mice .
Some prospective studies have not detected any association
between the duration of breast-feeding and the development
However, a single-cohort study involving children in the
general population showed that a short duration of breast-
group launched a T1D prevention trial that is directly testing
the hypothesis that supplementing breast milk with a highly
hydrolyzed milk formula rather than conventional, cow’s-
milk-based formulas may decrease the cumulative incidence
of islet aAbs and eventually T1D in children with genetic
susceptibility. While results about T1D incidence are eagerly
awaited, a first Finnish report covering a median observation
period of 10years documented a protective effect on islet aAb
seroconversion . The hazard ratio (adjusted for differences
in the duration of exposure) for ≥1 aAb in the hydrolysate
group was 0.51 (95% confidence interval 0.28–0.91).
Proposed Mechanisms and the Molecular
There are several mechanisms through which infectious trig-
explanation of how microbial infections may set off autoim-
mune diseases must take into account the observation that
all individuals harbour potentially autoreactive lymphocytes,
but that these cells remain innocuous unless somehow acti-
vated. Mechanisms by which infectious agents might activate
these cells fall into two major classes: non-antigen-specific and
Regarding non-antigen-specific mechanisms, infections can
cause host cell destruction, leading to release of normally
sequestered antigens which thus become visible for immune
recognition. In addition, microbes could improve the
antigen-presenting capacity of antigen-presenting cells such
as DCs and render them more effective by enhancing
the antigen-processing machinery and the expression of
major histocompatibility complex (MHC) and co-stimulatory
molecules. Microbial infections can also induce the synthesis
of inflammatory cytokines, which further activate DCs and
modify lymphocyte homing patterns. Finally, some infections
might provoke polyclonal lymphocyte activation via either a
is documented that they can promote local islet inflammation
sequestered antigens. Polymorphisms in the IFHI1 gene locus,
which participates in the innate immune responses to these
viruses, have also been associated with T1D .
The cornerstone of the antigen-specific theory is epitope
mimicry: a microbial epitope is structurally similar to self-β-
cell epitope(s), although different enough to be recognized
as foreign by the host’s immune system. This initial
physiologic response against foreign invaders could thus divert
immune response against self-pancreatic tissues. This immune
cross-activation is further favoured by the fact that T-cell
receptor recognition of MHC-peptide complexes is extremely
degenerate, which probably represents the evolutionary down-
side of being able to respond to the widest possible array
of potential foreign antigens . Molecular mimicry has
frequently been called forth because some enteroviral epitopes
are highly homologous to β-cell epitopes derived from GAD
and tyrosine phosphatase-like insulinoma antigen 2 (IA-2)
[43–45] and, in some cases, give rise to cross-reactive T cells
is, by observing that β-cell and enteroviral T-cell reactivities
84 Afonso and MalloneVolume 15 Suppl. 3 September 2013
DIABETES, OBESITY AND METABOLISM
exception is the report by Honeyman et al. , who raised
CD4+ T-cell clones recognizing a rotavirus VP740–52epitope
and tested them for their cross-reactivity with the homologous
single Vβ13 T-cell receptor also proliferated upon stimulation
with the IA-2805–817 epitope. This report is thus the only
one providing evidence of cross-reactivity at the single T-cell
responsible for this promiscuous epitope recognition. Without
this evidence, it would remain possible that the autoimmune
and viral-specific reactivities are simply associated, but that
different T-cell clonotypes are responsible for these responses.
Several previous reports have, however, failed to reach similar
conclusions. In NOD mice, T-cell cross-reactivity between
Coxsackievirus B and GAD epitopes was found , but
Coxsackievirus infection of NOD mice had no effect on T-
cell reactivity to GAD or on diabetes incidence . In T1D
patients, CD4+ T-cell clones generated against a GAD epitope
showed no proliferation to a homologous Coxsackievirus
sequence,  and three CD4+ T-cell lines raised against a
peptides . Furthermore, HLA-A2-restricted CD8+ T-cell
lines activated to secrete interferon-γ by a Coxsackievirus
epitope did not respond to a homologous GAD sequence and
were not cytotoxic to target cells pulsed with the peptide .
In light of these discordant and anecdotal reports, enteroviral
has met with scepticism .
The hypothesis of a molecular mimicry trigger for
autoimmunity may gain novel momentum following a recent
report from the group of Davis and colleagues . This
group used state-of-the-art technologies such as HLA class
II tetramers and detection of rare epitope-specific CD4+ T
lymphocytes by magnetic enrichement of tetramer-binding
cells to quantify and functionally characterize CD4+ T cells
were derived from viruses [human immunodeficiency virus
(HIV-1), cytomegalovirus (CMV), and Herpes simplex) to
which these subjects had never been exposed. Surprisingly,
these CD4+ T cells were not only detected at high frequency,
but also displayed a memory rather than the expected na¨ ıve
phenotype. This was true in adult blood but not in cord blood,
where these cells were present but displayed the expected
na¨ ıve phenotype, witnessing that antigen exposure had not
yet taken place at birth. In other words, adult T cells had
clearly already encountered their cognate epitope during the
life of these individuals, despite lack of viral exposure. This
unexpected finding was explained by demonstrating that these
CD4+ T cells, analysed at the clonal level, were able to
cross-recognize homologous epitopes derived from common
environmental and commensal microbes. This was shown
with homologous epitopes and by staining with HLA class
II tetramers loaded with the same epitopes. To directly
demonstrate the potential for cross-reactivity upon exposure
to an unrelated antigen carrying homologous sequences,
healthy subjects were immunized with a flu vaccine. This
vaccination did not only boost flu-specific T-cell responses,
but also expanded T cells recognizing homologous epitopes
derived from Finegoldia magna (which is part of the normal
human flora that colonizes skin and the gastrointestinal tract)
and Trichomonas vaginalis (a widespread parasite infecting
the urogenital tract). Although T-cell responses against auto-
antigens were not thoroughly evaluated, these authors further
found that CD4+ T cells recognizing self-epitopes derived
from preproinsulin, fibrinogen, and the melanocyte protein
gp100 were also present at frequencies similar to T cells
recognizing foreign antigens and equally displayed a memory
phenotype, implying that, also in this case, they may have
been primed by homologous non-self-epitopes. Of further
note, these autoreactive memory-like T cells were present at
higher frequencies in females than in males, a sex difference
that could help explain the higher female penetrance of most
study highlights how the T-cell cross-reactivity that provides
its immunological basis is a very common phenomenon.
A Roadmap for Future Research
What lines of evidence need to be produced to definitely assess
the role of a given microbe on T1D pathogenesis? Such claims
should fulfil the Koch’s postulates in order to establish a causal
1 Infection with a given agent should be found predominantly
in T1D patients (and at-risk subjects before T1D
development) versus healthy and T2D controls.
2 The agent should be isolated from T1D patients and grown
3 The agent should be capable of inducing T1D when
introduced into a healthy mouse.
4 The agent should be re-isolated from inoculated mice.
Have these postulates been fulfilled in the case of enteroviral
infections? We believe that this is not the case. Regarding the
first two points (evidence of infection and isolation of the
responsible agentmore frequently inT1D patients andpreced-
more frequent or more persistent rotavirus infections in islet
that infection with a ubiquitous agent like rotavirus, if it con-
as it may most probably synergize with a fertile background
of genetic susceptibility and/or autoimmune priming prior to
infection, as suggested for the NOD mouse .
Regarding the third point (the agent should trigger disease
in animal models), the case for enteroviruses is a weak one,
because these infections can either induce diabetes or confer
protection [13,54], depending on factors such as timing of
infection and strains used . Going back to early reports,
the interest in the enteroviral (Coxsackievirus B4) hypothesis
was ignited by a case report published by Yoon et al. in 1979
. These authors described a T1D child who died soon after
Volume 15Suppl. 3 September 2013doi:10.1111/dom.12166 85
DIABETES, OBESITY AND METABOLISM
finding in T1D. The key observation is that Coxsackievirus B4
was isolated from pancreas homogenates, documenting local
infection which was associated with seroconversion at the time
of death. Inoculation of mice with the human isolate produced
hyperglycaemia and insulitis associated with detection of viral
antigens in β cells. While both the clinical picture and animal
studies may suggest that this T1D case was virus-induced,
several points are worth noting. First, only one patient was
studied, who apparently had a quite aggressive T1D onset.
Second, the Coxsackievirus strain used for subsequent mouse
studies underwent several in vitro passages before being used
for mouse in vivo studies. The same caveat applies to the
Coxsackievirus B4 Edwards strain used in the majority of
subsequent in vivo mouse studies. This is a mouse-adapted
variant of Coxsackievirus B4 which was not originally isolated
from human pancreas specimens, but in rhesus renal cell
cultures from myocardial tissue of an infant with encephalo-
hepato-myocarditis and focal necrosis and inflammation of
the pancreas . This virus was subsequently passed in HeLa
cells and in mice by successive intraperitoneal inoculation of
homogenized pancreatic tissue from previously infected mice
to obtain the final experimental strain .
be noted that not only the microbial agent, but also the cross-
This could be achieved with adoptive transfer experiments,
by which T-cell clones cross-recognizing microbial and β-
cell antigens are injected into healthy mice. Alternatively,
transgenic mice in which all T cells have been engineered
with a T-cell receptor recognizing the promiscuous epitope
could be obtained. Neither approach has been reported for
testing this hypothesis in the case of enteroviruses. On the
same lines, the fourth point (i.e. re-isolation of the agent from
inoculated mice) is more critical to address in the case of
molecular mimicry, that is, by showing that infection leads to
expansion of β-cell cross-reactive T cells and that such T cells
are capable of transferring disease. Also in this case, evidence is
of a cause–effect relationship between enteroviral infections
autoimmune diseases for which an association with microbial
infections has been called forth, for example, in the case of
Borrelia burgdorferi and Lyme’s arthritis . This gap in
knowledge is due to the lack of adequate animal models to test
the proposed pathogenic triggers and the mechanisms at play.
Development of suitable models is therefore a key priority to
which remains the mainstay T1D model and develops disease
only when kept protected from environmental cues under spe-
How could these alternative models be developed for T1D and
other autoimmune diseases? In the case of epitope mimicry,
humanized mice expressing the human variants of both the
cross-reactive self-antigen and the MHC molecules restricting
its recognition by T cells appear as the most suitable. These
humanized models would be highly relevant for thoroughly
testing the molecular mimicry hypothesis by: i) infecting ani-
mals with the candidate triggering microbe, without the risk
of missing aetiologic links due to differences in self-epitope
sequences between mouse and human and/or their presenta-
modifying them to express T-cell receptor(s) responsible for
the promiscuous epitope recognition under study.
Work reviewed in this article was supported by grants from the
Societ´ eFrancophoneduDiab` ete,INSERMAvenir,Programme
Hospitalier de Recherche Clinique of the Assistance Publique
des Hopitaux de Paris and the Aviesan program ‘Diabetes and
the vessel wall injury’.
Conflict of Interest
None to declare.
1. Brezar V, Carel JC, Boitard C, Mallone R. Beyond the hormone: insulin as an
autoimmune target in type 1 diabetes. Endocr Rev 2011; 32: 623–669.
2. Hyttinen V, Kaprio J, Kinnunen L, Koskenvuo M, Tuomilehto J. Genetic
liability of type 1 diabetes and the onset age among 22,650 young Finnish
twin pairs: a nationwide follow-up study. Diabetes 2003; 52: 1052–1055.
3. Redondo MJ, Jeffrey J, Fain PR, Eisenbarth GS, Orban T. Concordance for
islet autoimmunity among monozygotic twins. N Engl J Med 2008; 359:
4. Bodansky HJ, Staines A, Stephenson C, Haigh D, Cartwright R. Evidence for
an environmental effect in the aetiology of insulin dependent diabetes in
a transmigratory population. BMJ 1992; 304: 1020–1022.
5. Kondrashova A, Viskari H, Kulmala P et al. Signs of beta-cell autoimmunity
in nondiabetic schoolchildren: a comparison between Russian Karelia with
a low incidence of type 1 diabetes and Finland with a high incidence rate.
Diabetes Care 2007; 30: 95–100.
6. Hermann R, Knip M, Veijola R et al. Temporal changes in the frequencies of
HLA genotypes in patients with Type 1 diabetes--indication of an increased
environmental pressure? Diabetologia 2003; 46: 420–425.
7. Gillespie KM, Bain SC, Barnett AH et al. The rising incidence of childhood
type 1 diabetes and reduced contribution of high-risk HLA haplotypes.
Lancet 2004; 364: 1699–1700.
8. Fourlanos S, Varney MD, Tait BD et al. The rising incidence of type 1
diabetes is accounted for by cases with lower-risk human leukocyte
antigen genotypes. Diabetes Care 2008; 31: 1546–1549.
9. Fourlanos S, Narendran P, Byrnes GB, Colman PG, Harrison LC. Insulin
resistance is a risk factor for progression to type 1 diabetes. Diabetologia
2004; 47: 1661–1667.
10. Couper JJ, Beresford S, Hirte C et al. Weight gain in early life predicts risk
of islet autoimmunity in children with a first-degree relative with type 1
diabetes. Diabetes Care 2009; 32: 94–99.
11. Lonnrot M, Korpela K, Knip M et al. Enterovirus infection as a risk factor
for beta-cell autoimmunity in a prospectively observed birth cohort: the
Finnish Diabetes Prediction and Prevention Study. Diabetes 2000; 49:
12. Honeyman MC, Coulson BS, Stone NL et al. Association between rotavirus
infection and pancreatic islet autoimmunity in children at risk of developing
type 1 diabetes. Diabetes 2000; 49: 1319–1324.
86 Afonso and Mallone Volume 15 Suppl. 3 September 2013
DIABETES, OBESITY AND METABOLISM
13. Serreze DV, Ottendorfer EW, Ellis TM, Gauntt CJ, Atkinson MA. Acceleration
of type 1 diabetes by a coxsackievirus infection requires a preexisting
critical mass of autoreactive T-cells in pancreatic islets. Diabetes 2000; 49:
14. Stene LC, Oikarinen S, Hyoty H et al. Enterovirus infection and progression
Study in the Young (DAISY). Diabetes 2010; 59: 3174–3180.
15. Dotta F, Censini S, van Halteren AG et al. Coxsackie B4 virus infection of
beta cells and natural killer cell insulitis in recent-onset type 1 diabetic
patients. Proc Natl Acad Sci U S A 2007; 104: 5115–5120.
16. Richardson SJ, Willcox A, Bone AJ, Foulis AK, Morgan NG. The prevalence
of enteroviral capsid protein vp1 immunostaining in pancreatic islets in
human type 1 diabetes. Diabetologia 2009; 52: 1143–1151.
17. Bach JF. The effect of infections on susceptibility to autoimmune and
allergic diseases. N Engl J Med 2002; 347: 911–920.
18. Filippi CM, von Herrath MG. Viral trigger for type 1 diabetes: pros and cons.
Diabetes 2008; 57: 2863–2871.
19. Hviid A, Stellfeld M, Wohlfahrt J, Melbye M. Childhood vaccination and type
1 diabetes. N Engl J Med 2004; 350: 1398–1404.
20. Trudeau JD, Dutz JP, Arany E, Hill DJ, Fieldus WE, Finegood DT. Neonatal
beta-cell apoptosis: a trigger for autoimmune diabetes? Diabetes 2000;
Nature 2001; 414: 792–798.
22. Turley S, Poirot L, Hattori M, Benoist C, Mathis D. Physiological beta cell
death triggers priming of self-reactive T cells by dendritic cells in a type-1
diabetes model. J Exp Med 2003; 198: 1527–1537.
23. Hoglund P, Mintern J, Waltzinger C, Heath W, Benoist C, Mathis D. Initiation
of autoimmune diabetes by developmentally regulated presentation of
islet cell antigens in the pancreatic lymph nodes. J Exp Med 1999; 189:
24. Diana J, Simoni Y, Furio L et al. Crosstalk between neutrophils, B-1a cells
and plasmacytoid dendritic cells initiates autoimmune diabetes. Nat Med
2013; 19: 65–73.
25. Kassem SA, Ariel I, Thornton PS, Scheimberg I, Glaser B. Beta-cell
proliferation and apoptosis in the developing normal human pancreas
and in hyperinsulinism of infancy. Diabetes 2000; 49: 1325–1333.
26. Feng H, Zeng Y, Graner MW, Katsanis E. Stressed apoptotic tumor cells
stimulate dendritic cells and induce specific cytotoxic T cells. Blood 2002;
27. Johansson U, Walther-Jallow L, Smed-Sorensen A, Spetz AL. Triggering
of dendritic cell responses after exposure to activated, but not resting,
apoptotic PBMCs. J Immunol 2007; 179: 1711–1720.
28. Throsby M, Coulaud J, Durant S, Homo-Delarche F. Increased transcriptional
preproinsulin II beta-cell activity in neonatal nonobese diabetic mice: in
situ hybridization analysis. Rev Diabet Stud 2005; 2: 75–83.
29. Virtanen SM, Knip M. Nutritional risk predictors of beta cell autoimmunity
and type 1 diabetes at a young age. Am J Clin Nutr 2003; 78: 1053–1067.
30. Sadauskaite-Kuehne V, Ludvigsson J, Padaiga Z, Jasinskiene E, Samuelsson
U. Longer breastfeeding is an independent protective factor against
development of type 1 diabetes mellitus in childhood. Diabetes Metab Res
Rev 2004; 20: 150–157.
31. Cheung R, Karjalainen J, Vandermeulen J, Singal DP, Dosch HM. T cells
from children with IDDM are sensitized to bovine serum albumin. Scand J
Immunol 1994; 40: 623–628.
32. Brugman S, Visser JT, Hillebrands JL, Bos NA, Rozing J. Prolonged exclusive
breastfeeding reduces autoimmune diabetes incidence and increases
regulatory T-cell frequency in bio-breeding diabetes-prone rats. Diabetes
Metab Res Rev 2009; 25: 380–387.
33. Visser JT, Lammers K, Hoogendijk A et al. Restoration of impaired intestinal
barrier function by the hydrolysed casein diet contributes to the prevention
of type 1 diabetes in the diabetes-prone BioBreeding rat. Diabetologia
2010; 53: 2621–2628.
34. Schmid S, Koczwara K, Schwinghammer S, Lampasona V, Ziegler AG,
Bonifacio E. Delayed exposure to wheat and barley proteins reduces
diabetes incidence in non-obese diabetic mice. Clin Immunol 2004; 111:
35. Mueller DB, Koczwara K, Mueller AS, Pallauf J, Ziegler AG, Bonifacio E.
Influence of early nutritional components on the development of murine
autoimmune diabetes. Ann Nutr Metab 2009; 54: 208–217.
36. Norris JM, Barriga K, Klingensmith G et al. Timing of initial cereal exposure
in infancy and risk of islet autoimmunity. JAMA 2003; 290: 1713–1720.
37. Ziegler AG, Schmid S, Huber D, Hummel M, Bonifacio E. Early infant feeding
and risk of developing type 1 diabetes-associated autoantibodies. JAMA
2003; 290: 1721–1728.
38. VirtanenSM,KenwardMG,Erkkola Metal.Age atintroduction of newfoods
susceptibility to type 1 diabetes. Diabetologia 2006; 49: 1512–1521.
39. Holmberg H, Wahlberg J, Vaarala O, Ludvigsson J. Short duration of breast-
feeding as a risk-factor for beta-cell autoantibodies in 5-year-old children
from the general population. Br J Nutr 2007; 97: 111–116.
40. Knip M, Virtanen SM, Seppa K et al. Dietary intervention in infancy and later
signs of beta-cell autoimmunity. N Engl J Med 2010; 363: 1900–1908.
41. Smyth DJ, Cooper JD, Bailey R et al. A genome-wide association study of
nonsynonymous SNPs identifies a type 1 diabetes locus in the interferon-
induced helicase (IFIH1) region. Nat Genet 2006; 38: 617–619.
42. Mason D. A very high level of crossreactivity is an essential feature of the
T-cell receptor. Immunol Today 1998; 19: 395–404.
43. Oldstone MB, Nerenberg M, Southern P, Price J, Lewicki H. Virus infection
triggers insulin-dependent diabetes mellitus in a transgenic model: role of
anti-self (virus) immune response. Cell 1991; 65: 319–331.
44. Atkinson MA, Bowman MA, Campbell L, Darrow BL, Kaufman DL,
Maclaren NK. Cellular immunity to a determinant common to glutamate
decarboxylase and coxsackie virus in insulin-dependent diabetes. J Clin
Invest 1994; 94: 2125–2129.
45. Honeyman MC, Stone NL, Harrison LC. T-cell epitopes in type 1 diabetes
autoantigen tyrosine phosphatase IA-2: potential for mimicry with rotavirus
and other environmental agents. Mol Med 1998; 4: 231–239.
46. Honeyman MC, Stone NL, Falk BA, Nepom G, Harrison LC. Evidence
for molecular mimicry between human T cell epitopes in rotavirus and
pancreatic islet autoantigens. J Immunol 2010; 184: 2204–2210.
47. Tian J, Lehmann PV, Kaufman DL. T cell cross-reactivity between
coxsackievirus and glutamate decarboxylase is associated with a murine
diabetes susceptibility allele. J Exp Med 1994; 180: 1979–1984.
48. Horwitz MS, Bradley LM, Harbertson J, Krahl T, Lee J, Sarvetnick N. Diabetes
induced by Coxsackie virus: initiation by bystander damage and not
molecular mimicry. Nat Med 1998; 4: 781–785.
49. Schloot NC, Willemen SJ, Duinkerken G, Drijfhout JW, de Vries RR, Roep
BO. Molecular mimicry in type 1 diabetes mellitus revisited: T-cell clones
to GAD65 peptides with sequence homology to Coxsackie or proinsulin
peptides do not crossreact with homologous counterpart. Hum Immunol
2001; 62: 299–309.
50. Marttila J, Juhela S, Vaarala O et al. Responses of coxsackievirus B4-specific
T-cell lines to 2C protein-characterization of epitopes with special reference
to the GAD65 homology region. Virology 2001; 284: 131–141.
51. Varela-Calvino R, Skowera A, Arif S, Peakman M. Identification of a naturally
processed cytotoxic CD8 T-cell epitope of coxsackievirus B4, presented by
HLA-A2.1 and located in the PEVKEK region of the P2C nonstructural protein.
J Virol 2004; 78: 13399–13408.
Volume 15 Suppl. 3 September 2013doi:10.1111/dom.12166 87
review article Download full-text
DIABETES, OBESITY AND METABOLISM
52. Benoist C, Mathis D. Autoimmunity provoked by infection: how
good is the case for T cell epitope mimicry? Nat Immunol 2001;
53. Su LF, Kidd BA, Han A, Kotzin JJ, Davis MM. Virus-specific CD4(+) memory-
phenotype T cells are abundant in unexposed adults. Immunity 2013; 38:
54. Serreze DV, Wasserfall C, Ottendorfer EW et al. Diabetes acceleration or
prevention by a coxsackievirus B4 infection: critical requirements for both
interleukin-4 and gamma interferon. J Virol 2005; 79: 1045–1052.
55. Yoon JW, Austin M, Onodera T, Notkins AL. Isolation of a virus from the
pancreas of a child with diabetic ketoacidosis. N Engl J Med 1979; 300:
88 Afonso and MalloneVolume 15 Suppl. 3 September 2013