Download Identification and Functional Analysis of Mutations in the Hepatocyte

0021-972X/00/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2000 by The Endocrine Society
Vol. 85, No. 1
Printed in U.S.A.
Identification and Functional Analysis of Mutations in
the Hepatocyte Nuclear Factor-1a Gene in Anti-Islet
Autoantibody-Negative Japanese Patients with
Type 1 Diabetes
EIJI KAWASAKI, YASUNORI SERA, KENICHI YAMAKAWA, TAKAHIRO ABE,
MASAKO OZAKI, SHIGEO UOTANI, NARIYUKI OHTSU, HIROFUMI TAKINO,
HIRONORI YAMASAKI, YOSHIHIKO YAMAGUCHI, NOBUO MATSUURA, AND
KATSUMI EGUCHI
First Department of Internal Medicine, Nagasaki University School of Medicine, Nagasaki 852-8501;
and Department of Pediatrics, Kitasato University School of Medicine (N.O., N.M.), Sagamihara
228-8555, Japan
ABSTRACT
Mutations in the hepatocyte nuclear factor-1a (HNF-1a) gene are
the cause of maturity-onset diabetes of the young type 3 (MODY 3),
which is characterized by a severe impairment of insulin secretion
and early onset of the disease. Although the majority of patients with
type 1 diabetes have type 1A, immune-mediated diabetes, there is a
significant percentage of the patients who have no evidence of an
autoimmune disorder at the onset of disease. The aim of this study
was to estimate the prevalence of MODY 3 in antiislet autoantibody
negative patients with type 1 diabetes. From a large population-based
sample of unrelated Japanese patients with type 1 diabetes, 28 patients who lacked autoantibodies to glutamic acid decarboxylase, islet
cell antigen 512/insulinoma-associated antigen-2, phogrin (phosphate homolog of granules of insulinoma)/insulinoma-associated antigen-2b, and insulin at the onset of type 1 diabetes were examined
by PCR-based direct sequencing of the 10 exons, flanking introns, and
the promoter region of the HNF-1a gene. Two (7.1%) of 28 autoan-
tibody-negative patients with type 1 diabetes were identified as carrying mutations in the HNF-1a gene. One patient carried a frameshift
mutation (Pro379fsdelCT) in exon 6, and another patient carried a
novel 2-bp substitution at nucleotides 145 (G to A) and 146 (C to A)
from the transcriptional site of the promoter region. These mutations
were identified in heterozygous form and were not identified in 64
unrelated healthy control subjects or 54 unrelated islet autoantibodypositive patients with type 1 diabetes. Functional analysis of the
mutant HNF-1a gene indicated that the Pro379fsdelCT mutation had
no transcriptional trans-activation activity and acted in a dominant
negative manner. The 145/46 GC to AA mutation in the promoter
region showed reduced promoter activity by 10 –20% compared to the
wild-type sequence. In conclusion, about 7% of Japanese diabetic
patients lacking antiislet autoantibodies initially classified as having
type 1 diabetes could have diabetes caused by mutations in the
HNF-1a gene. (J Clin Endocrinol Metab 85: 331–335, 2000)
M
ULTIPLE TYPES of diabetes mellitus have been defined by the recent reports of an American Diabetes
Association Expert Committee and a WHO consultation
based on our current understanding of pathogenesis rather
than the requirement for insulin therapy (1, 2). Type 1 diabetes is often associated with chronic and progressive autoimmune destruction of islet b-cells with a long prodromal
phase (3). This type of type 1 diabetes is classified as immunemediated (type 1A) diabetes. The autoimmune phenomena
associated with type 1A diabetes include circulating serum
autoantibodies to various islet cell antigens, including glutamic acid decarboxylase (GAD), islet cell antibody 512
(ICA512)/insulinoma-associated antigen-2 (IA-2), and insulin. At least one of these autoantibodies is present at disease
onset in more than 90% of patients with type 1 diabetes.
However, a significant percentage (5–10%) of patients who
have no evidence of an autoimmune disorder at disease onset
exist and are classified as idiopathic (type 1B) diabetics (1).
Type 1B diabetes may be etiologically heterogeneous, including insulin secretory defects caused by extensive pancreatic islet b-cell destruction or b-cell dysfunction.
Maturity-onset diabetes of the young (MODY) is a monogenic form of diabetes characterized by autosomal dominant
inheritance, an early age of onset (usually ,25 yr of age), and
b-cell dysfunction (4, 5). MODY is genetically heterogeneous,
resulting from mutations in at least five genes, the hepatocyte
nuclear factor-4a (HNF-4a) for MODY1 (6), glucokinase for
MODY2 (7), HNF-1a for MODY3 (8), the insulin promoter
factor-1 for MODY4 (9), and the HNF-1b for MODY5 (10),
respectively. The most commonly identified cause of MODY
in most racial groups is a mutation in the HNF-1a gene
(MODY3). Although the mechanism of hyperglycemia in
MODY3 is not fully understood, the phenotypic characterization of MODY3 families has shown a deficient insulin
secretory response to glucose (11, 12). Because of the early
age of onset, the severe hyperglycemia, and the deficient
insulin secretion of patients with MODY3, we hypothesized
that some patients classified as having type 1B diabetes could
Received July 14, 1999. Revision received September 15, 1999. Accepted September 23, 1999.
Address all correspondence and requests for reprints to: Eiji Kawasaki, M.D., First Department of Internal Medicine, Nagasaki University School of Medicine, 1–7-1 Sakamoto, Nagasaki 852-8501, Japan.
E-mail: [email protected].
331
332
KAWASAKI ET AL.
have MODY3. In this study we screened autoantibody-negative Japanese patients with type 1 diabetes at disease onset
for mutations in the HNF-1a gene and analyzed the functional properties of mutant genes.
Subjects and Methods
Subjects
From a large population-based sample from patients with type 1
diabetes all of the antiislet autoantibody-negative, unrelated, new-onset
patients (n 5 28; 15 males and 13 females) were selected from 231 (92
males and 139 females; median age of onset, 17.0 yr) type 1 diabetic
patients screened for antiislet autoantibodies. Their median age at onset
and body mass index were 14.0 yr (range, 0.1–59.0 yr) and 18.8 kg/m2
(range, 13.1–22.8 kg/m2), respectively. They were all clinically diagnosed as having type 1 diabetes according to WHO criteria and required
insulin therapy from the time of diagnosis (13). Two and 5 patients had
at least 1 family member with type 1 and type 2 diabetes, respectively.
Seventy-one percent (20 of 28) of these patients had DQA1*0301/
DQB1*0303, DQA1*0301/DQB1*0401, or both, which are the high risk
human leukocyte antigen (HLA)-DQ haplotypes for type 1 diabetes in
Japanese populations (14). Sera were obtained from patients within 2
weeks after beginning insulin treatment.
An additional 54 patients with type 1A diabetes (23 males and 31
females) and 64 healthy control subjects (35 males and 29 females) were
screened for those mutations that were found in autoantibody-negative
patients with type 1 diabetes. The median age at onset in patients with
type 1A diabetes was 19.5 yr (range, 3.0 –59.0 yr), and 83% (45 of 54) were
positive for GAD autoantibody, 52% (28 of 54) were positive for ICA512/
IA-2 autoantibody, and 41% (22 of 54) were positive for phogrin (phosphatase homolog of granules of insulinoma)/IA-2b autoantibody, respectively. All subjects gave informed consent, and the protocol was
approved by the institutional review boards of the Nagasaki University
School of Medicine. Sera were stored at 220 C until use.
Mutation analysis
The 10 exons, flanking introns, and the minimal promoter (the 354 bp
upstream from the start codon) of the HNF-1a gene were amplified by
PCR using genomic DNA obtained from peripheral blood and sequencespecific primers (15). PCR was performed in a 100-mL volume containing
10 mmol/L Tris-HCl (pH 8.3), 50 mmol/L KCl, 1 mmol/L MgCl2, 200
mmol/L deoxy-NTPs, 1 mmol/L of each primer, 0.25 U AmpliTaq Taq
polymerase (Perkin-Elmer Corp., Foster City, CA), and 50 ng DNA. The
cycling conditions were 1 min at 94 C, followed by 35 cycles consisting
of 1 min at 94 C, 1 min at 60 C, and 1 min at 72 C. The PCR products
were purified using a Microcon-100 (Amicon, Inc., Beverly, MA) before
both strands were sequenced using a BigDye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer Corp. PE Applied Biosystems). The reactions were analyzed on an ABI Prism 377 DNA Sequencer. The sequence of each mutation was confirmed by cloning the
PCR product into pGEM-T vector (Promega Corp., Madison, WI) and
sequencing clones representing both alleles. In the screening for specific
mutations in MODY family members, patients with immune-mediated
type 1 diabetes, and healthy control subjects, the nucleotide substitution
of GC to AA in the promoter region was determined by PCR restriction
fragment length polymorphism with the restriction enzyme Fnu4HI. The
Pro379fsdelCT mutation was detected by a rapid screening technique
using fluorescently labeled forward primer (59-TCCCCTCGTAGGTCTCACGCAG-39) and modified reverse primer (59-GTTTCCAGGAAGTGAGGCCATCATG), generating a 137-bp fragment in nonmutated samples and a 135-bp fragment from mutation alleles (16). The PCR products
were analyzed on a 4% polyacrylamide denaturing gel on an ABI Prism
377 DNA Sequencer, and the difference in length between normal and
mutant alleles was detected by Genescan Analysis software (ABI PerkinElmer Corp.).
Autoantibody assays
Autoantibodies to GAD were analyzed using RIA kits with 125Ilabeled native GAD purified from pig brain (RIP anti-GAD Hoechst,
Hoechst-Behring, Tokyo, Japan) as previously described (17). The in-
JCE & M • 2000
Vol 85 • No 1
sulin autoantibody assay was performed using a fluid phase radioassay
with competition with cold insulin and precipitation with polyethylene
glycol as previously described with some modifications (17). Autoantibodies to protein tyrosine phosphatase-like proteins, ICA512/IA-2 and
phogrin/IA-2b, were analyzed by radioassay using in vitro translated
35
S-labeled ICA512bdc (amino acids 256 –979 of IA-2) and the cytoplasmic domain of phogrin (amino acids 640-1015), respectively (18). The
cut-off value was 5 U/mL for GAD autoantibodies, 80 mU/mL for
insulin autoantibody assay, index 0.023 for ICA512/IA-2 autoantibodies, and index 0.031 for phogrin/IA-2b autoantibodies, respectively. We
participated in the international autoantibody proficiency programs for
autoantibodies to GAD, insulin, and ICA512/IA-2 using these assays,
and laboratory sensitivity and specificity for each assay were 100% and
100%, respectively.
Preparation of DNA constructs and functional study of the
mutant HNF-1a gene
A human HNF-1a complementary DNA (cDNA) clone including the
entire coding region in expression vector that has the cytomegalovirus
promoter was provided by Dr. Y. Yamada (Kyoto University School of
Medicine, Kyoto, Japan). The Pro379fsdelCT mutation was introduced by
the Sculptor in vitro mutagenesis kit (Amersham International, Aylesbury, UK). Three kilobases of human albumin (ALB) gene promoter
containing the HNF-1a-binding site (21773 to 21789, 2358 to 2342, and
265 to 249 relative to the cap site) originally inserted in pBR-CAT vector
(19) were subcloned into the HnidIII site of the pGL3-Basic vector (Promega Corp.). Ninety-two base pairs of the HNF-1a gene promoter (111
to 1102 relative to the transcriptional start site), which includes the
wild-type (WT) and mutant activating protein-1 (AP-1)-binding site,
were amplified by PCR using genomic DNA from the proband in family
B and subcloned into a pGL3-Basic reporter vector (Promega Corp.). The
sequences of the constructs were confirmed on an ABI Prism 377 DNA
Sequencer.
HeLa cells were transfected with the indicated amounts of expression
and reporter vectors together with 50 ng pRL-simian virus 40 (SV40)
vector (Promega Corp.) as an internal control using SuperFect transfection reagent (QIAGEN, Tokyo, Japan). The trans-activation activities
of WT-HNF-1a and Pro379fsdelCT-HNF-1a were measured after 48 h
using the Dual Luciferase Reporter Assay System (Promega Corp.) and
TD-20/20 Luminometer (Turner Designs, Sunnyvale, CA). To analyze
the effect of the AP-1-binding site mutation 145/46 GC to AA, MIN6
cells and HuH7 cells were transfected with 2 mg WT or mutant HNF-1a
promoter-pGL3, and the trans-activation activities were measured after
48 h. Each study was repeated three or four times.
Statistical analysis
Results are expressed as the mean 6 sd unless otherwise indicated.
Data in a luciferase reporter assay were analyzed by unpaired t test. P ,
0.05 was considered statistically significant.
Results
Identification of HNF-1a gene mutations in patients with
type 1 idiopathic diabetes
Mutations were identified in 2 (7.1%) of 28 unrelated patients with type 1 diabetes who lacked antiislet autoantibodies at disease onset by sequencing of the 10 exons, flanking
introns, and the minimal promoter of the HNF-1a gene. The
deletion of CT at codon 379 for Pro (CCT) in exon 6
(Pro379fsdelCT) that would be expected to generate a truncated protein of 416 amino acids was identified in 1 family.
A novel 2-bp substitution at nucleotide 145 (G to A) and
nucleotide 146 (C to A) from the transcriptional start site,
which is located in an AP-1-binding site of the promoter
region and is conserved in the sequences of human, rat,
mouse, and chicken HNF-1a (15, 20), was identified in another family. Both mutations were identified in heterozygous
HNF-1a MUTATIONS IN TYPE 1B DIABETES
form and were not identified in 64 unrelated healthy control
subjects (128 chromosomes) and 54 unrelated patients with
type 1A diabetes (108 chromosomes).
Clinical profiles of subjects with mutations and their
family members
The diabetic proband with frameshift mutation
Pro379fsdelCT was a 27-yr-old woman who had a body mass
index of 20.5 kg/m2. Her diabetes was initially noticed by
urine glucose screening at school at 12 yr of age and was
diagnosed by a subsequent oral glucose tolerance test
(OGTT). The low insulin secretory responses to glucose [fasting immunoreactive insulin (IRI), 30.1 pmol/L; 30 min IRI,
74.2 pmol/L) and arginine (fasting IRI, 22.8 pmol/L; 30 min
IRI, 67.8 pmol/L) were observed at the onset of the disease.
Insulin treatment was started 5 months after diagnosis. Thus,
she was initially diagnosed as a slow-onset patient with type
1 diabetes. The amount of exogenous insulin required was
rapidly increased to 0.98 –1.25 U/kgzday at 15 yr of age. She
is a homozygote of HLA-DQA1*0301/DQB1*0401, which is
one of the highest risk haplotypes of type 1 diabetes in
Japanese population. She has severe diabetic retinopathy,
with vitreous hemorrhage and neovascular glaucoma and
diabetic neuropathy. The Pro379fsdelCT mutation was also
identified in her brother, father, and two paternal cousins
from whom the genomic DNA was able to be obtained (Fig.
1, family A). The proband, two aunts, and two cousins examined had overt diabetes, and the father had impaired
glucose tolerance as measured during an OGTT. The brother
333
showed normal glucose tolerance with impaired insulin secretory response (fasting IRI, 13.4 pmol/L; 30 min IRI, 106.9
pmol/L) at the time of examination. None of the examined
subjects carrying the mutation had autoantibodies to GAD,
ICA512/IA-2, or phogrin/IA-2b.
The patient with a novel 2-bp substitution at nucleotides
145 (G to A) and 146 (C to A) of the promoter region was
a 16-yr-old girl with a body mass index of 19.4 kg/m2. She
developed acute hyperglycemia (34.6 mmol/L) at 16 yr of
age with severe symptoms, including thirst, polydipsia, and
polyuria. A low insulin secretory response to glucose (fasting
IRI, 12.0 pmol/L; 30 min IRI, 18.0 pmol/L) was observed at
the onset of the disease. She is a homozygote for HLADQA1*0301/DQB1*0401. No diabetic complications were
observed. The same mutation was found in her mother and
her sister, but not in her father or a brother. The mother
showed impaired glucose tolerance as measured by an
OGTT, but the sister, a 14-yr-old girl, had normal glucose
tolerance (Fig. 1, family B).
DNA polymorphisms in the HNF-1a gene
In addition to the diabetes-associated mutations described
above, we found 11 nucleotide substitutions, of which 4 were
located in exons and 7 in introns (Table 1). All of these have
been described previously and were not associated with
MODY (8, 21). Seven polymorphisms, including Ala983 Val,
Gly288, GGG3 GGC, Thr515, ACG3 ACA, and Gly574,
GGC3 AGC; intron 1 nucleotide 291, A3 G, intron 5 nucleotide 247, C3 T, and intron 9 nucleotide 144, C3 T,
which have been reported in whites and African-Americans
(8, 21, 22), were not found in our Japanese subjects.
Functional properties of the mutant HNF-1a gene
FIG. 1. Families with mutations in the HNF-1a gene. Patients with
diabetes are indicated by closed symbols, nondiabetic subjects are
indicated by open symbols, and undiagnosed subjects are indicated by
gray symbols. Subjects with impaired glucose tolerance are indicated
by half-filled symbols. Arrows indicate the proband from each pedigree who was examined for mutations. The HNF-1a genotype, if
known, is indicated below the symbol (N, normal; M, mutant). The age
at diagnosis (age of examination) or age at performance of an OGTT
is indicated as well.
To analyze the functional properties of the Pro379fsdelCT
mutant HNF-1a, a luciferase reporter assay was performed
using human cervical carcinoma HeLa cells, which do not
have endogenous HNF-1a (23). HeLa cells were transfected
with constructs encoding WT-HNF-1a and Pro379fsdelCTHNF-1a together with the human ALB promoter-luciferase
reporter gene. The mutant HNF-1a did not stimulate transcription of the ALB-luciferase reporter, whereas WTHNF-1a stimulated transcription and generated a significant
increase in reporter gene activity (Fig. 2A). Increasing
amounts of the Pro379fsdelCT-HNF-1a inhibited luciferase
activity up to 53% of the control value in a dose-dependent
manner, suggesting that Pro379fsdelCT has a dominant negative effect on HNF-1a activity (Fig. 2B).
The effect of the AP-1-binding site mutation 145/46 GC
to AA was also analyzed using mouse insulinoma MIN6 cells
and human hepatoma HuH7 cells. The promoter activity of
the 145/46 GC to AA construct was decreased by 20% and
12% compared to that of a WT-promoter construct in a MIN6
and a HuH7 cell, respectively (Fig. 2C).
Discussion
Although the underlying molecular mechanisms for the development of type 1 diabetes have not been fully clarified, the
most common form of type 1 diabetes is believed to be an
334
JCE & M • 2000
Vol 85 • No 1
KAWASAKI ET AL.
TABLE 1. DNA polymorphisms in HNF-1a gene in 28 patients with type 1B diabetes
Exon/Intron
Exon 1
Exon 4
Exon 7
Exon 8
Exon 9
Intron 1
Intron 2
Intron 5
Intron 7
Intron 9
Codon/nt
Nucleotide change
Codon 17
Codon 27
Codon 98
Codon 288
Codon 459
Codon 487
Codon 515
Codon 574
nt 291
nt 242
nt 251
nt 223
nt 19
nt 247
nt 242
nt 17
nt 144
nt 224
CTC (Leu) 3 CTG (Leu)
ATC (Ile) 3 CTC (Leu)
GCC (Ala) 3 GTC (Val)
GGG (Gly) 3 GGC (Gly)
CTG (Leu) 3 TTG (Leu)
AAC (Asn) 3 AGC (Ser)
ACG (Thr) 3 ACA (Thr)
GGC (Gly) 3 AGC (Ser)
A3G
G3A
A3T
C3T
C3G
C3T
G3T
G3A
C3T
T3C
Frequency
C 5 0.52
A 5 0.70
C 5 1.00
G 5 1.00
C 5 0.63
A 5 0.54
G 5 1.00
G 5 1.00
A 5 0.00
G 5 0.77
A 5 0.16
C 5 0.57
C 5 0.95
C 5 1.00
G 5 0.61
G 5 0.57
C 5 1.00
T 5 0.57
G 5 0.48
C 5 0.30
T 5 0.00
C 5 0.00
T 5 0.37
G 5 0.46
A 5 0.00
A 5 0.00
G 5 1.00
A 5 0.23
T 5 0.84
T 5 0.43
G 5 0.05
T 5 0.00
T 5 0.39
A 5 0.43
T 5 0.00
C 5 0.43
DNA polymorphisms found in introns are noted with respect to the splice donor or acceptor site. Sites that are shown as being monomorphic
in our Japanese subjects are polymorphic in white or African American subjects (8, 21, 22).
FIG. 2. Functional
properties
of
Pro379fsdelCT mutant and 146/46 GC
to AA mutant promoter. A, Trans-activation activity of ALB promoter in
HeLa cells. Ten to 100 ng WT or
Pro379fsdelCT-HNF-1a expression vector were transfected with 2 mg ALBreporter gene and 10 ng pRL-SV40
DNA.
E,
WT-HNF-1a;
F,
Pro379fsdelCT-HNF-1a. B, Fifty nanograms of WT-HNF-1a plasmid were
transfected with increasing amounts
(50, 200, 500, and 1000 ng) of the
Pro379fsdelCT-HNF-1a expression vector. The total amount of DNA added was
adjusted by empty vector. C, Promoter
activity assay of WT and 145/46 GC to
AA mutant promoter in MIN6 cells and
HuH7 cells. Two micrograms of WT or
mutant promoter reporter gene were
transfected with 50 ng pRL-SV40. Luciferase activity was normalized by the
activity of pRL-SV40. Each experiment
was repeated three or four times, and a
representative result is shown. The
mean 6 SD are shown. *, P , 0.05; **,
P , 0.01.
autoimmune disorder caused by the T cell-mediated selective
destruction of pancreatic b-cells that leads to an absolute insulin
deficiency. At present, the best tool for identifying the autoimmune process of pancreatic b-cells is the detection of autoantibodies to biochemically defined islet autoantigens, including
insulin, GAD, and the protein tyrosine phosphatase-like molecules, ICA512/IA-2 and phogrin/IA-2b (24, 25). With combinatorial antiislet autoantibody determination, greater than 90%
of Japanese new-onset patients with type 1 diabetes express one
or more of these three autoantibodies (insulin, GAD, and IA2/phogrin), whereas about 10% of patients are negative for all
of these autoantibodies and are classified as type 1B diabetics
(17). MODY3 caused by the mutations in the HNF-1a gene is
characterized by rapid progress to overt diabetes in childhood
or adolescence and severe insulin secretory defects in response
to glucose (12, 26, 27). We report here identification of the
mutations in the HNF-1a gene in 2 of 28 antiislet autoantibodynegative patients with type 1 diabetes, indicating that some of
the MODY3 patients treated with insulin would be misclassified as having type 1 diabetes. The results highlight the difficulties in distinguishing between insulin-dependent diabetic
patients with MODY3 and patients with type 1B diabetes
because of the lack of antiislet autoantibodies in these patients
(12, 28, 29).
HNF-1a MUTATIONS IN TYPE 1B DIABETES
The frameshift mutation Pro379fsdelCT identified in exon
6 of the HNF-1a gene would generate a mutant truncated
protein of 416 amino acids, which affects the COOH-terminal
trans-activation domain of HNF-1a (30). The functional studies indicate that this frameshift mutation has no trans-activation activity and acts in a dominant negative manner.
Sourdive and co-workers reported that loss of residues 416 –
474 of HNF-1a not only impaired trans-activation but also
abolished nuclear transportation of the HNF-1a protein (31),
suggesting that this mutant is also lacking in proper nuclear
targeting signals. The mutation was present in five subjects;
three were diabetic, one had impaired glucose tolerance, and
one had normal glucose tolerance. The father in this family
carried the mutation, although he was nondiabetic at the age
of 66 yr, indicating incomplete penetrance.
The novel mutation in the promoter region (GC3 AA) at
nucleotides 145 and 146 from the transcriptional start was
found in the other nuclear family. This mutation disrupts the
binding site for the AP-1, which has been shown to be important in activating the transcription factor HNF-4a gene
(32). The functional studies using MIN6 cells and HuH7 cells
suggest that the 145/46 GC to AA mutation could lead to
reduced promoter activity of HNF-1a gene and thus might
lead to lower than normal levels of HNF-1a protein expression and consequently decreased transcription of target
genes encoding proteins that play a key role in the insulin
secretory response to glucose.
Moller and co-workers recently reported that in Danish
patients with type 1 diabetes without one of the high risk
HLA haplotypes, DR3 or DR4, about 10% of the patients
carried a mutation in the HNF-1a gene (29). However, both
patients with type 1B diabetes carrying the HNF-1a mutation
reported here had HLA-DQA1*0301/DQB1*0401, a haplotype that is a high risk HLA haplotype in Japanese subjects
for type 1 diabetes (14). Therefore, to avoid the misclassification of MODY3 patients as type 1 diabetics, in whom an
absolute deficiency of insulin secretion is due to pancreatic
b-cell destruction, clinically type 1 diabetic patients without
antiislet autoantibodies should be examined for mutations in
the HNF-1a gene even if they have high risk HLA haplotypes
for type 1 diabetes.
Acknowledgments
We are grateful to Dr. G. S. Eisenbarth (Barbara Davis Center for
Childhood Diabetes, Denver, CO) for providing the cDNA for human
ICA512/IA-2 and phogrin, and to Dr. F. Nishibe (Yamasa Corp., Tokyo,
Japan) and K. Ohgushi for their excellent technical assistance. We also
thank Dr. Y. Goto and Y. Maeda for their contribution to this work, and
Dr. Y. Yamada (Kyoto University, Kyoto, Japan) for providing the cDNA
for human HNF-1a.
References
1. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.
1997 Report of the Expert Committee on the Diagnosis and Classification of
Diabetes Mellitus. Diabetes Care. 20:1183–1197.
2. Alberti KG, Zimmet PZ. 1998 Definition, diagnosis and classification of diabetes mellitus and its complications. I. Diagnosis and classification of diabetes
mellitus provisional report of a WHO consultation. Diabetes Med. 15:539 –553.
3. Eisenbarth GS. 1986 Type I diabetes mellitus: a chronic autoimmune disease.
N Engl J Med. 314:1360 –1368.
4. Fajans S. 1990 Scope and heterogeneous nature of MODY. Diabetes Care.
13:49 – 64.
335
5. Velho G, Froguel P. 1997 Maturity-onset diabetes of the young (MODY),
MODY genes and non-insulin-dependent diabetes mellitus. Diabetes Metab.
23(Suppl 2):34 –37.
6. Yamagata K, Furuta H, Oda N, et al. 1996 Mutations in the hepatocyte nuclear
factor-4a gene in maturity-onset diabetes of the young. Nature. 384:458 – 460.
7. Froguel P, Zouali H, Vionnet N, et al. 1993 Familial hyperglycemia due to
mutations in glucokinase: definition of a subtype of diabetes mellitus. N Engl
J Med. 328:697–702.
8. Yamagata K, Oda N, Kaisaki PJ, et al. 1996 Mutations in the hepatocyte
nuclear factor-1a gene in maturity-onset diabetes of the young. Nature.
384:455– 458.
9. Stoffers DA, Ferrer J, Clarke WL, Habener JF. 1997 Early-onset type-II diabetes mellitus (MODY4) linked to IPF1. Nat Genet. 17:138 –139.
10. Horikawa Y, Iwasaki N, Hara M, et al. 1997 Mutation in hepatocyte nuclear
factor-1b gene (TCF2) associated with MODY. Nat Genet. 17:384 –385.
11. Byrne MM, Sturis J, Fajans SS, et al. 1995 Altered insulin secretory responses
to glucose in subjects with a mutation in the MODY1 gene on chromosome 20.
Diabetes. 44:699 –704.
12. Lehto M, Tuomi T, Mahtani MM, et al. 1997 Characterization of the MODY3
phenotype. Early-onset diabetes caused by an insulin secretion defect. J Clin
Invest. 99:582–591.
13. WHO. 1985 Diabetes mellitus: report of a WHO study group. Geneva: WHO;
Tech. Rep. Ser. no. 727.
14. Awata T, Kuzuya T, Matsuda A, Iwamoto Y, Kanazawa Y. 1992 Genetic
analysis of HLA class II alleles and susceptibility to type I (insulin-dependent)
diabetes mellitus in Japanese subjects. Diabetologia. 35:419 – 424.
15. Kaisaki PJ, Menzel S, Lindner T, et al. 1997 Mutations in the hepatocyte
nuclear factor-1a gene in MODY and early-onset NIDDM: evidence for a
mutational hotspot in exon 4. Diabetes. 46:528 –535.
16. Frayling TM, Bulman MP, Appleton M, Hattersley AT, Ellard S. 1997 A rapid
screening method for hepatocyte nuclear factor 1a frameshift mutations; prevalence in maturity-onset diabetes of the young and late-onset non-insulin
dependent diabetes. Hum Genet. 351–354.
17. Sera Y, Kawasaki E, Abiru N, et al. 1999 Autoantibodies to multiple islet
autoantigens in patients with abrupt onset type 1 diabetes and diabetes diagnosed with urinary glucose screening. J Autoimmun. 13:257–265.
18. Kawasaki E, Eisenbarth GS, Wasmeier C, Hutton JC. 1996 Autoantibodies to
protein tyrosine phosphatase-like proteins in type I diabetes: overlapping
specificities to phogrin and ICA512/IA-2. Diabetes. 45:1344 –1349.
19. Hayashi Y, Chan J, Nakabayashi H, Hashimoto T, Tamaoki T. 1992 Identification and characterization of two enhancers of the human albumin gene.
J Biol Chem. 267:14580 –14585.
20. Gragnoli C, Lindner T, Cockburn BN, et al. 1997 Maturity-onset diabetes of
the young due to a mutation in the hepatocyte nuclear factor-4a binding site
in the promoter of the hepatocyte nuclear factor-1 a gene. Diabetes.
46:1648 –1651.
21. Urhammer SA, Rasmussen SK, Kaisaki PJ, et al. 1997 Genetic variation in the
hepatocyte nuclear factor-1a gene in Danish Caucasians with late-onset
NIDDM. Diabetologia. 40:473– 475.
22. Boutin P, Gresh L, Cisse A, et al. 1999 Missense mutaion Gly574Ser in the
transcription factor HNF-1a is a marker of atypical diabetes mellitus in African-American children. Diabetologia. 42:380 –381.
23. Nicosia A, Tafi R, Monaci P. 1992 Trans-dominant inhibition of transcription
activator LFB1. Nucleic Acids Res. 20:5321–5328.
24. Verge CF, Gianani R, Kawasaki E, et al. 1996 Prediction of type I diabetes in
first-degree relatives using a combination of insulin, GAD, and ICA512bdc/
IA-2 autoantibodies. Diabetes. 45:926 –933.
25. Bingley PJ, Christie MR, Bonifacio E, et al. 1994 Combined analysis of autoantibodies improves prediction of IDDM in islet cell antibody-positive relatives. Diabetes. 43:1304 –1310.
26. Vaxillaire M, Boccio V, Philippi A, et al. 1995 A gene for maturity onset
diabetes of the young (MODY) maps to chromosome 12q. Nat Genet.
9:418 – 423.
27. Velho G, Vaxillaire M, Boccio V, Charpentier G, Froguel P. 1996 Diabetes
complications in NIDDM kindreds linked to the MODY3 locus on chromosome
12q. Diabetes Care. 19:915–919.
28. Yamada S, Nishigori H, Onda H, et al. 1997 Identification of mutations in the
hepatocyte nuclear factor (HNF)-1a gene in Japanese subjects with IDDM.
Diabetes. 46:1643–1647.
29. Moller AM, Dalgaard LT, Pociot F, Nerup J, Hansen T, Pedersen O. 1998
Mutations in the hepatocyte nuclear factor-1a gene in Caucasian families
originally classified as having type I diabetes. Diabetologia. 41:1528 –1531.
30. Tronche F, Yaniv M. 1992 HNF1, a homeoprotein member of the hepatic
transcription regulatory network. BioEssays. 14:579 –587.
31. Sourdive DJD, Chouard T, Yaniv M. 1993 The HNF-1 C-terminal domain
contributes to transcriptional activity and modulates nuclear localisation. C R
Acad Sci [D] (Paris). 316:385–394.
32. Kuo CJ, Conley PB, Chen L, Sladek FM, Darnell JEJ, Crabtree G. 1992 A
transcriptional hierarchy involved in mammalian cell-type specification. Nature. 355:457– 461.