Transcript Diabetes mellitus An Inborn Error of Metabolism ?

HKSPEM Clinical Meeting
8 May 2008
Diabetes Mellitus
An Inborn Error of Metabolism ?
Dr Grace Poon
Department of Paediatrics and Adolescent Medicine
Queen Mary Hospital
PKH
Male / 14 years
• Para 1 FT LSCS for fetal distress in Canossa Hospital
• Mother with GDM since 7 mths of gestation – on diet control
• IUGR – induction of labour
• Born SGA with birth weight of 2.4 kg
• PGM and MGF both have NIDDM
• Hospitalized for RSV bronchiolitis at 2.5 months
• Developed DKA whilst in hospital – pH 7.05, BE -23, blood
glucose 26.2 mmol/L, gross glucosuria and ketonuria
• Complicated by acute pulmonary oedema and mild
cerebral oedema
• Diagnosed to have neonatal diabetes mellitus
• Started on insulin therapy (Aug 94)
• USG pancreas normal (Sep 94)
• Could stop insulin after 5 months (Jan 95 at CA 8 months)
• HbA1c 7.6% (Oct 94)  5.7% (Jan 95)
OGTT (May 95 – CA 1 year)
0
30min
60min
90min
120min
150min
Insulin (mIU/L)
2.3
9.9
5.8
3.3
<1
<1
C-peptide
(fasting 0.5-3 ng/ml)
1.4
2.2
1.7
1.9
2.1
1.7
Glucose (mmol/L)
4.6
9.4
9
8.6
6.1
4.6
* Samples haemolysed: insulin and c-peptide could be spuriously low
• Islet cell Ab negative (Sep 00)
• Serial OGTT between 1996 to 2001 showed normal
glucose response
• HbA1c 5-6.1% (Mar 96 – Nov 02)
• Onset of puberty since July 2005
HbA1c
Mar
1996
Jun
1996
Feb
1999
Feb
2000
Jan
2001
Sep
2001
Nov
2002
Mar
2005
5%
5.6%
6%
6%
6.1%
5.9%
5.8%
6.5%
OGTT (Oct 05 – CA 11y 5m)
0
30min
120min
Insulin
(mIU/L)
2.8
7.5
21
Glucose
(mmol/L)
5.3
10.6
10.1
Impression: Impaired glucose tolerance
HbA1c 6.7%
Defaulted FU until June 06
OGTT (Jun 06 – CA 12y 1m)
0
30min
120min
Insulin (mIU/L)
5.8
21
21
Glucose
(mmol/L)
7.6
13.9
12.1
Impression: Diabetic response
Islet cell Ab negative
HbA1c 7.2%
Glucagon stimulation test (Aug 06)
Glucose
(mmol/L)
Insulin
(mIU/L)
C-peptide
(0.27-1.27
nmol/L)
T-5
8.5
4.2
0.41
T0
8.4
2.2
0.43
T5
9.4
30
1.15
T10
9.8
4.1
0.73
T15
10.9
4.6
0.5
Haemolysed samples
• Started on daily dose of Daonil (glibenclamide) 2.5mg
since Aug 2006 (CA 12y 3m)
• HbA1c
HbA1c
Started on Daonil

Jun
2006
Aug
2006
Jan
2007
Jun
2007
Aug
2007
Feb
2008
7.2%
7.8%
6.5%
6.2%
6.2%
7.4%
• Daonil increased to 5mg daily since April 2008
Neonatal diabetes and
abnormal development of the pancreas
Neonatal diabetes mellitus (NDM)
• First described by Kitselle in 1852
• Classically defined as diabetes mellitus occurring in the
first 6 weeks of life in term infants
• Some suggests a cut off for early onset DM at 6 months
• Single gene disorders rather than classical autoimmune
type 1 diabetes
• Incidence: 1 in 300,000-400,000 live births
• Transient (TNDM) vs Permanent (PNDM)
– based on the grounds of resolution or persistence
beyond the first year of life
Transient neonatal
diabetes mellitus (TNDM)
TNDM
• TNDM contributes 50-60% of cases of neonatal DM
• Develop diabetes in the first few weeks of life but go into
remission in a few months
• The pancreatic dysfunction may be maintained
throughout life, with possible relapse to a permanent
diabetes state usually initiated at times of metabolic
stress such as puberty or pregnancy or as adults
• Regular follow-up of TNDM is recommended and
parents should be informed of the risk of recurrence
• Remission of TNDM is temporary and more than 60% of
cases relapse at a median age of 11 years
• Fasting blood sugar and insulin, HbA1c and insulin
sensitivity and insulin response of IVGTT are normal
during the remission phase
Permanent neonatal
diabetes mellitus (PNDM)
PNDM
• Insulin secretory failure occurs in the late fetal or
early postnatal period
• Does not go into remission
Comparison of several features in TNDM and
PNDM cases in the French cohort (n=50)
TNDM (n=29)
PNDM (n=21)
p value
Gestational age (weeks)
38.2 ± 2.2
39.2 ± 1.6
p = 0.15
Birth weight (g)
1987 ± 510
2497 ± 690
p < 0.006
Birth length (cm)
44.3 ± 3.4
47.5 ± 2.4
p < 0.006
Head circumference (cm)
31.5 ± 1.8
33 ± 1.9
p < 0.02
(n=20/27) 74%
(n=7/19) 36%
p < 0.03
Median age at diagnosis (days)
(range)
6 (1-81)
27 (1-127)
p < 0.01
Initial insulin dose (unit/kg/day)
0.6 ± 0.25
1.4 ± 1.2
p < 0.006
IUGR
Metz C et al. J Pediatr 2002;141:483-489
TNDM vs PNDM
• Comparing with infants with PNDM, patients with TNDM:
– More likely to have IUGR
– Less likely to develop ketoacidosis
– Younger at diagnosis of diabetes
– Have lower initial insulin requirements
• Considerable overlap occurs between the 2 groups and
TNDM cannot be distinguished from PNDM on the basis of
clinical features
Aetiology of transient neonatal
diabetes mellitus (TNDM)
Transient neonatal DM
Chromosome 6 anomalies
•
Paternal duplications
•
Paternal isodisomy
•
Methylation defect
Potassium channel activating mutations
•
ABCC8 (SUR1) and rarely KCNJ11 (Kir6.2) mutations
Unidentified genetic defects
Chromosome 6 anomalies
(OMIM 601410)
Chromosome 6 anomalies:
a disease linked to imprinting
• Uniparental isodisomy (UDP6)
– 2 haplo-identical copies of chromosome 6 inherited
from the father with no contribution from the mother
• Unbalanced duplication of paternal chromosome 6q24
• Loss of the normal methylation from maternal 6q24
Two imprinted genes in 6q24
• ZAC (Zinc finger protein associated with Apoptosis and
cell Cycle arrest)
• HYMAI (hydatidiform, mole-associated and imprinted
transcript)
ZAC
• A zinc finger protein which potently induces apoptosis
and cell cycle arrest and prevents tumour formation
• A key regulator of peroxisome proliferator-activated
receptor gamma (PPAR)
• Activation of PPAR is sufficient to inhibit -cell
proliferation, and PPAR overexpression significantly
compromises glucose-stimulated insulin secretion
• Overexpression of ZAC, either due to extra copies of
paternal origin or loss of the silencing effect of
methylation on the maternal allele is the likely cause of
6qTNDM
HYMAI
• Uncertain if HYMAI plays a role in TNDM
• It has no open reading frame and its function is not clear
Chromosome 6q anomalies
• Usually present in the first week of life with profound
diabetes requiring insulin therapy
• Significant IUGR (mean BW 1980 gm at term)
• Go into remission at a median age of 12 weeks
• Likely to relapse at times of metabolic stress such as
during puberty or intercurrent illness
• Although the relapse is characterized by loss of classical
insulin response to hyperglycaemia, there is evidence
that insulin is still available, as there is excellent insulin
response on glucagon stimulation, suggesting that the
G-coupled protein receptor response remains intact
through  cAMP (might be an area of therapeutic
intervention in future)
Chromosome 6q anomalies
• Some children have macroglossia and anterior
abdominal wall defects, as described in BeckwithWiedemann syndrome
• Not only do these children lose the maternal methylation
at 6q, they also have hypomethylation at other imprinted
loci, such as the Beckwith-Wiedemann locus, resulting in
significant phenotypic variation
(Mackay et al. Hum Genet 2006;120:262-269)
Chromosome 6 anomalies and the
potassium channel activating mutations
cause the vast majority of cases of TNDM
K+(ATP) channels in glucose metabolism
The islet KATP channel
• A critical regulator of -cell insulin secretion
• In the normal response to increased glucose
exposure and its metabolism within -cell, ATP
levels increase with a concomitant decrease in
magnesium-adenosine diphosphate (Mg-ADP)
levels, allowing closure of the KATP channel and
membrane depolarization, which allows calcium
influx into the cell and leads to insulin secretion
The islet KATP channel
• A hetero-octamer made up of 2 types of subunits: 4
regulatory sulphonylurea receptors (SURs) embracing 4
pore-forming inwardly rectifying potassium channels (Kir)
• A 1:1 SUR1:Kir6.2 stoichiometry is both necessary and
sufficient for assembly of active KATP channels
• SUR, a member of ABC transporter family, originates
from 2 separate genes and therefore occurs in several
spliced isoforms
• SUR1 is found in the pancreatic - cell and neurons
• SUR2A occurs in heart cells whilst SUR2B in smooth
muscle
The islet KATP channel
• Kir6.2 subunits form the channel pore in the majority of
tissues, such as pancreatic -cell, brain, heart and
skeletal muscle
• Whilst Kir6.1 can be found in smooth vascular muscle
and astrocytes
• These different channel forms have different pore
properties and adenine nucleotide sensitivities
• The KCNJ11 (potassium inwardly-rectifying channel,
subfamily J, member 11) and ABCC8 (ATP-binding
cassette, subfamily C, member 8) genes, encoding the
Kir6.2 and SUR1 respectively are located at 11p15
The channelopathies
Potassium channel activating mutations
• Mutations in KCNJ11 and ABCC8 genes, encoding the
Kir6.2 and SUR1 subunits of the pancreatic KATP
channel involved in regulation of insulin secretion,
account for 1/3 to 1/2 of the PNDM cases but also for
TNDM cases
• Molecular analysis of chromosome 6 anomalies and the
KCNJ11 and ABCC8 genes provides a tool for
distinguishing transient from permanent neonatal
diabetes mellitus in the neonatal period
KCNJ11 mutations
KCNJ11 mutations
• Mutations in KCNJ11 that encode the Kir6.2 subunit of
the KATP channel are the most common cause of PNDM
and account for 30% of all cases
• The KATP channel is variably unresponsive to  ATP,
making the membrane hyperpolarized and preventing
influx of calcium and efflux of insulin
KCNJ11 mutations
• Median age of presentation ~ 3-4 weeks of age as
opposed to babies with 6q anomalies who tend to
present in the first week of life
• Born SGA but not as small as those with 6q anomalies
(mean BW 2497g vs 1980g)
• All display low insulin levels despite dramatic
hyperglycaemia
• 30% with ketoacidosis
• 20% have associated neurological disease with
developmental delay
KCNJ11 mutations
• Epilepsy or muscle weakness is sometimes present,
indicating that the same potassium channels play a role
in the functioning of CNS
• Children at the most severe end of this spectrum can be
profoundly affected - ‘DEND’ (developmental delay,
epilepsy and neonatal diabetes)
• An intermediate form (i-DEND) associated with milder
developmental delay and no epilepsy
• The mutation causing isolated diabetes produce less
change in ATP sensitivity than those associated with
diabetes plus neurological disease (Q52R, V59G)
KCNJ11 mutations
• Sulphonylureas close KATP channels by an ATPindependent mechanism
• Some patients with diabetes due to Kir6.2 mutations have
been successfully transferred to oral sulphonylurea therapy
at doses ranging from 0.5-1.0 mg/kg/day of glyburide
• Pearson et al. demonstrated that 44 of 49 patients (90%)
aged 3 months to 36 years could be successfully
transferred to oral therapy with a highly significant and
sustained improvement in HbA1c (8.1% to 6.4%)
• Patients with neurological features are less likely to be
successful in managing their DM using sulphonylureas
NEJM 2006;355:467-477
ABCC8 mutations
ABCC8 mutations
• The basal Mg-nucleotide-dependent stimulatory action of
sulphonylurea receptor-1 (SUR1) encoded by ABCC8 is
increased, which effectively prevents closure of the KATP
channel
• Recently Babenko et al. demonstrated that activating
mutations in the ABCC8 gene encoding the SUR1
subunit of the channel can also cause permanent
neonatal DM that is responsive to sulphonylurea therapy
NEJM 2006;355:456-466
K+(ATP) channels and neonatal diabetes
Aetiology of permanent neonatal
diabetes mellitus (PNDM)
Permanent neonatal DM (1)
• Heterozygous activating mutation in KCNJ11 gene and
in ABCC8 gene (Kir6.2 and SUR1 subunits of the
pancreatic KATP channel)
• Insulin (INS) gene mutation
• Severe pancreatic hypoplasia associated with insulin
promoter factor-1 (IPF-1) mutation and with cerebellar
hypoplasia due to pancreatic transcription factor 1A
(PTF1A) mutation
• Homozygous glucokinase gene mutation: insensitivity to
glucose
Permanent neonatal DM (2)
• IPEX syndrome and FOXP3 mutation: diffuse
autoimmunity
• Associated with epiphyseal dysplasia: Wolcott-Rallison
syndrome and EIF2AK3 gene mutation
• Mitochondrial disease
• Possibly associated with enterovirus infection
• Associated with hypothyroidism, glaucoma and GLIS3
mutation
Insulin gene mutations
Insulin (INS) gene mutations
• Stoy et al. reported 10 missense mutations in INS
gene in 21 patients (20 with neonatal DM and 1 with
type 2 diabetes) from 16 families in which NDM
seems to segregate as a dominant trait and ABCC8
and KCNJ11 mutations were not found
• Not associated with -cell autoantibodies
• INS mutations accounts for 15-20% of cases of
PNDM – similar to ABCC8 mutations (19%) but less
than KCNJ11 mutations (30%)
Stoy J et al. PNAS 2007;104:10540-15044
Insulin (INS) gene mutations
• These patients are older at diagnosis, presenting at a
median age of 9 weeks, with 15/16 being diagnosed in the
first 6 months, 3 diagnosed between 6 months and 1 year,
and the father of a proband at 30 years with type 2 DM
• Usually presenting with DKA or marked hyperglycaemia
and was treated with insulin from diagnosis
• These patients with permanent diabetes without
extrapancreatic features except for a low birth weight
Stoy J et al. PNAS 2007;104:10540-15044
Pancreatic agenesis/hypoplasia and the
insulin promoter factor-1 (IPF-1) gene
IPF-1 mutation and PNDM
• First described in 1997 by Stoffers et al. in a child with
PNDM and pancreatic exocrine insufficiency due to
pancreatic agenesis (Nat Genet 1997;15:106-110)
• The proband was homozygous for a mutation
(Pro63fsdelC) in IPF-1, the gene involved in the master
control of exocrine and endocrine pancreatic development,
being responsible for the coordinated development of the
pancreas in-utero and also for the continued functional
integrity of pancreatic  islet cells
IPF-1 mutation and PNDM
• Within the extended family were 8 individuals in 6
generations with early-onset diabetes akin to type 2
diabetes
• These were identified as heterozygotes for the same
mutation with the mutant truncated isoform of IPF-1
acting as a dominant negative inhibitor of wild type IPF-1
activity; the resultant illness was reassigned as MaturityOnset Diabetes of the Young (MODY) 4
• Additional studies have also identified that less severe
IPF-1 mutations can cause autosomal dominant lateonset forms of type 2 diabetes
IPF-1 mutation and PNDM
• Only one further case report of pancreatic agenesis has
been ascribed to an IPF-1 mutation and this was a
compound heterozygous mutation of the gene
• Since there is complete absence of pancreatic tissue
and exocrine function is also compromised, these
patients will require the use of pancreatic enzyme
supplementation
Pancreatic agenesis/hypoplasia
and the pancreatic transcription
factor 1A (PTF1A) gene
PNDM with cerebellar hypoplasia
• 3 members of a consanguineous Pakistani family with
neonatal diabetes and cerebellar hypoplasia were
described in 1999
• Suggestive of an autosomal recessive inheritance pattern
• The infants all died within a few months of birth from a
combination of metabolic dysfunction, respiratory
compromise and sepsis
• A further child of North European descent was later
identified with an identical phenotype and complete
pancreatic agenesis
PNDM with cerebellar hypoplasia
• This syndrome was found to be linked to mutations in
the transcription factor PTF1A, a major gene involved in
pancreatic development and also expressed in the
cerebellum
• These patients have pancreatic hypoplasia associated
with microcephaly linked to cerebellar hypoplasia
• About 19 cases of pancreatic agenesis/hypoplasia have
been published, but most cases remain unexplained at
the molecular level
Glucokinase gene mutations
Glucokinase (GCK) mutations
• MODY 2 is caused by mutations in the glucokinase gene
and usually leads to mild hyperglycaemia in affected
individuals
• Glucokinase is a key regulator of glucose metabolism in
islet cells controlling the levels of insulin secretion
• Homozygous GCK mutations have been described but
are uncommon causes of PNDM
• This diagnosis should be considered in families with a
history of glucose intolerance, MODY in first degree
relatives and especially if consanguinity is suspected
IPEX
(OMIM 304790)
IPEX (Immune dysregulation, Polyendocrinopathy,
Enteropathy, X-linked syndrome)
• The only one of the currently identified neonatal diabetes
syndromes in which autoimmunity plays a central role
• Mutations in 2 genes have been identified: FOXP3 and
CD25 (interleukin-2 receptor alpha)
• Both genes are important for the normal function of
regulatory T cells that play a central role in regulating
adaptive immune responses to maintain tolerance to
host tissues
• In IPEX, this self-tolerance is lost and the result is a
devastating autoimmune response
IPEX (Immune dysregulation, Polyendocrinopathy,
Enteropathy, X-linked syndrome)
• Type 1 diabetes auto-antibodies (GAD, IAA, ICA) are
frequently described, as are those directed against the
thyroid gland and various other organs
• Enteropathy (100%), failure to thrive (>90%) and earlyonset (often neonatal) diabetets (>90%) occur in almost
all cases
• Other features frequently described include exfoliative
dermatitis, haemolytic anaemia, thrombocytopenia,
Addison’s disease and autoimmune hypothyroidism
IPEX (Immune dysregulation, Polyendocrinopathy,
Enteropathy, X-linked syndrome)
• There are several treatment options for IPEX:
– Supportive therapy
– Immunosuppressive agents such as tacrolimus, steroid
and cyclosporine have shown varying degrees of
efficacy, but toxicity to other organs such as the kidney
has been problematic
– BMT
• Prognosis remains guarded for children with this condition
Wolcott-Rallison syndrome
(OMIM 226980)
Wolcott-Rallison syndrome
• An autosomal recessive disorder characterized by
infancy-onset diabetes (often within the neonatal period)
associated with a spondylo-epiphyseal dysplasia
• A constellation of other features includes hepatomegaly,
mental retardation, renal failure and early death
• Mutation of the gene encoding eukaryotic translation
initiation factor 2-alpha kinase 3 (EIF2AK3) on
chromosome 2p12 has been shown to cause this
syndrome
Wolcott-Rallison syndrome
• EIF2AK3 is highly expressed in islet cells, liver, kidneys
and developing bone
• It has a role in protein translation and regulates the
synthesis of unfolded proteins in the endoplasmic
reticulum which could account for the multiple system
involvement reported in this condition
Mitochondrial diabetes
Mitochondrial diabetes
• Neonatal diabetes may also exist in the context of a
mitochondrial disorder
• Usually associated with other organ dysfunction, which
may be recognized after the diagnosis of neonatal DM
• Commonly associated with sensorineural deafness
• Characterized by progressive non-autoimmune -cell
failure
Mitochondrial diabetes
• Maternal transmission of mutated mitochondrial DNA can
result in maternally inherited diabetes
• Several mutations have been implicated but the
strongest evidence relates to a point substitution at
nucleotide position 3243 (A to G) in the mitochondrial
tRNA [leucine (UUR)] gene
Other very rare forms of NDM
Heterozygous Hepatocyte Nuclear Factor
(HNF) -1 mutation
• Heterozygous mutations of the transcription factor HNF1 (HNF1 homeobox B) gene have been associated with
a form of MODY5, which is characterized by dominantly
inherited diabetes mellitus with renal cysts
• This has been described in one child with PNDM and
some small renal cysts whilst the other sibling only had
transient hyperglycaemia but with more profound renal
dysplasia
GLI-similar protein 3 (GLIS3) mutations
• In 2006, Senee et al. described a frameshift mutation or
deletions in the transcription factor GLIS3 in 3
consanguineous families with a history of neonatal
diabetes, congenital hypothyroidism and facial
dysmorphology (large, flat, square-shaped face with a
thin and bird-shaped curved nose)
• Additional features in some but not all patients included
congenital glaucoma, liver fibrosis and cystic kidneys
Possible association with enterovirus infection
• A single case report has suggested that maternal
enterovirus (echovirus 6) infection in pregnancy (end of
first trimester) can lead to autoimmune, neonatal-onset
diabetes with the presence of anti-insulin and glutamic
acid decarboxylase antibodies at or very soon after birth
• Ruling out IPEX, the pancreas in this female child was
very hypoplastic and the authors suggested a role for
maternally transmitted enterovirus either by direct
influence on pancreatic organogenesis or through
aggressive -cell-targeted autoimmune attack
Diabetologia 2000;43:1235-1238
Main types of neonatal diabetes, their genetic basis, co-morbidities, treatment,
outcome and relative frequencies as potential diagnoses
Treatment options in
neonatal diabetes mellitus
• Insulin therapy and high caloric intake are the basis of
treatment
• All neonates presenting acutely with diabetes should be
started on insulin therapy
• In the case of 6q anomalies, diabetes will be transient but
insulin is required until remission to prevent dehydration and
allow normal growth
• In permanent neonatal diabetes due to conditions other than
KCNJ11 and possibly ABCC8 mutations, insulin is also
required long-term
• Some patients with mutations in KCNJ11 and ABCC8 may
be transferred from insulin therapy to sulphonylureas
Genetic counselling
The risk of recurrence is different depending on whether
it is the transient or permanent form of neonatal diabetes
and the different molecular mechanisms identified
Chromosome 6 anomalies
• Cases of UPD6 are sporadic with low recurrence risk to
siblings and offspring
• In familial cases of unbalanced duplication of 6q24,
affected male will have 50% risk of passing it to an
offspring; the risk of passing a duplication of 6q24 from
the mother to her offspring is low
Imprinting anomaly: logic says that the mother should pass it on but so far
no familial case is known and the risk of transmission is unknown. The
cause of imprinting relaxation is not identified and the identified children with
this anomaly are too young to procreate
• The sibling and offspring risks for sporadic cases of
methylation defect are not known
PNDM and Mendelian inheritance
should be counselled accordingly
• Recurrence risk is 25% in the autosomal recessive
disorders ( EIF2AK, GLIS3, IPF-1 and PTF1A genes)
• IPEX is an x-linked disorder
• Mutations in genes encoding the potassium channel
subunits are transmitted in the heterozygous state in a
dominant way
Conclusions
Conclusions
• Effective treatment of neonatal diabetes requires thorough
understanding of the underlying disease processes
• Successful studies illuminating these processes have not
only improved our knowledge of pancreatic development
and physiology, but also have revolutionized the treatment
options for some patients
• The progress made in our scientific and clinical
understanding of these extremely rare diseases is a
perfect example of how studying seemingly rare illnesses
can improve our overall knowledge of much more common
conditions
Thank you