The Journal of Clinical Endocrinology & Metabolism Vol. 86, No. 5 2320-2322
Copyright © 2001 by The Endocrine Society
Mutation Screening of the Neurogenin-3 Gene in Autosomal Dominant Diabetes1
Sung-Hoon Kim,
James H. Warram,
Andrzej S. Krolewski and
Alessandro Doria
Section on Genetics and Epidemiology, Research Division, Joslin
Diabetes Center (S.-H.K., J.H.W., A.S.K., A.D.), and Department of
Medicine, Harvard Medical School (S.-H.K., A.S.K., A.D.), Boston,
Massachusetts 02215
Address all correspondence and requests for reprints to: Alessandro Doria, M.D. Ph.D., Section on Genetics and Epidemiology, Joslin Diabetes Center, One Joslin Place, Boston, Massachusetts 02215. E-mail:
alessandro.doria{at}joslin.harvard.edu
 |
Abstract
|
|---|
We investigated whether genetic variability in neurogenin-3, a basic
helix-loop-helix transcription factor that is expressed in the
developing pancreas, contributes to the etiology of maturity-onset
diabetes of the young or other forms of autosomal dominant diabetes.
Ninety-one probands of families with autosomal dominant diabetes were
screened for neurogenin-3 mutations by dideoxy fingerprinting. Three
sequence differences were identified: a polymorphism not affecting the
amino acid sequence (L75L), a CA insertion/deletion in intron 1
(-44ins/del), and a C to T transition causing a serine to
phenylalanine substitution (S199F). None of these sequence differences
were more frequent in the family probands than in 179 nondiabetic
controls. In contrast, allele 199F was weakly, but significantly,
associated with common type 2 diabetes (199F frequencies = 0.436
in 132 cases with type 2 diabetes vs. 0.346 in the
family probands and 0.346 in controls; P = 0.05).
The relative risk of type 2 diabetes for 199F carriers was 1.7 (95%
confidence interval, 1.04–2.7). We conclude that sequence differences
in the neurogenin-3 gene do not play a major role in the development of
autosomal dominant diabetes. Rather, they might contribute to common
type 2 diabetes, although this finding must be replicated in other
populations.
|
Introduction
|
|---|
THE DEVELOPMENT of ß-cells is regulated
by a complex network of transcription factors organized in a
hierarchical fashion (1). Some of these factors also
control ß-cell physiology in the adult life by modulating the
expression of genes involved in glucose metabolism and the insulin gene
itself. The importance of these transcription factors is underscored by
the recent discovery that mutations in these genes cause maturity-onset
diabetes of the young (MODY), a rare form of diabetes characterized by
an early onset and an autosomal dominant mode of inheritance. To date,
MODY mutations have been identified in glucokinase and five
transcription factors, namely hepatocyte nuclear factor
(HNF)-1
, HNF-1ß, HNF-4, insulin promoter factor,
and NEUROD1 (2, 3, 4, 5, 6, 7). However, there is strong evidence that
additional MODY genes remain to be identified (8).
Neurogenin-3 (NGN3) is a member of the basic helix-loop-helix family of
transcription factors that is expressed in the nervous system and
developing pancreas (9). Mice homozygous for a null
NGN3 mutation fail to generate any pancreatic endocrine
cells and die soon after birth from diabetes (9). This
raises the hypothesis that less severe mutations, impairing, but not
abolishing, NGN3 function, may be responsible in humans for a reduction
of ß-cell mass, leading to the development of MODY or other forms of
autosomal dominant diabetes with a later onset. Here we report the
results of mutation screening of the NGN3 gene in 91 Joslin
families who were selected for an autosomal dominant pattern of
transmission of diabetes.
|
Subjects and Methods
|
|---|
Families and unrelated individuals
Ninety-one Joslin families with a pattern of occurrence of type
2 diabetes consistent with an autosomal dominant mode of inheritance
were included in the mutation screening. The ascertainment of these
families has been previously described (7). The screening
criteria used to identify these families were 1) an index case and at
least 1 sibling having type 2 diabetes diagnosed between ages 10 and 60
yr, 2) the treatment for diabetes in the index case for the initial 2
yr was diet or oral agents, and 3) diabetes occurred in at least 3
generations. Diabetes was diagnosed 1) if an individual was treated
with insulin or oral agents, 2) if results of an oral glucose tolerance
test met WHO criteria for diabetes, or 3) if the level of hemoglobin
A1c was more than 7.0% in individuals who
declined an oral glucose tolerance test or were not fasting when
examined. The study protocol and informed consent procedures were
approved by the human subjects committee of the Joslin Diabetes Center.
Of the 91 families, 43 families had early-onset diabetes (median age at
diagnosis in the family, <40 yr), and 48 families had diabetes
diagnosed in middle age (median age at diagnosis, 40–59 yr).
Twenty-one of the early-onset families had a median age at diagnosis
younger than 25 yr. All families were Caucasian, with the exception of
5 Hispanic, 2 African-American, and 1 Pacific Islander pedigrees. Known
MODY genes have been previously excluded as the loci responsible for
diabetes in these families. The characteristics of affected family
members are summarized in Table 1
. One
diabetic individual per family was included in the screening for
mutations in the NGN3 gene.
View this table:
[in this window]
[in a new window]
|
Table 1. Characteristics of affected member of Joslin
families with an autosomal dominant pattern of transmission of diabetes
|
|
The frequencies of the sequence differences identified in the families
were also determined in 132 unrelated individuals with common type 2
diabetes and in 179 nondiabetic controls. Type 2 diabetic subjects were
randomly selected from a sample of Joslin Clinic patients, aged 40–64
yr, who met the following criteria: 1) having diabetes that was
diagnosed after age 35 yr; and 2) being treated with diet or oral
agents for at least 2 yr after diagnosis of diabetes. Nondiabetic
controls consisted of nondiabetic spouses of family members. All of
these subjects were of Caucasian origin. The characteristics of these
two study groups are summarized in Table 2.
Mutation screening
The 91 family probands were screened for sequence differences by
dideoxy fingerprinting (ddF), a powerful modification of single strand
conformation polymorphism analysis (10). The whole coding
sequence of the NGN3 (645 bp in a single exon) plus 72 bp of intron 1
and 49 bp of the 3'-untranslated region were amplified from each
individual by PCR using primers 5'-CCCACCTAGCCTCGGAAT-3' and 5'-CCCTCT-
CCCTTACCCTTAGC-3'. PCR was performed from 40 ng DNA in a total volume
of 30 µL using the Advantage GC Genomic Polymerase Mix
(CLONTECH Laboratories, Inc., Palo Alto, CA) with a 0.4
µmol/L primer concentration and 1.1 Mg(OAC)2.
Reactions were carried out for 30 cycles at 94 C for 30 s, 55 C
for 45 s, and 68 C for 60 s. PCR products were purified from
agarose and subjected to Sanger’s dideoxy chain termination reaction
using ddG in a 10-µL reaction containing 20 ng primary PCR template,
0.15 µmol/L end-labeled ddF primer, 25 µmol/L deoxy-NTPs, 200
µmol/L ddG, 10 mmol/L Tris HCl (pH 8.3), 50 mmol/L KCl, 1.5 mmol/L
MgCl2, and 1 U Taq polymerase at 95 C
for 30 s, 55 C for 30 s, and 72 C for 1 min for 30 cycles. To
ensure that the whole fragment was covered by the screening, separate
reactions were run using both PCR primers as well as an internal primer
(5'-AAGAGGGAGGCTGCCGAG-3'). After denaturing the samples, 4 µL were
electrophoresed overnight at room temperature in a nondenaturing 0.75x
mutation detection enhancement gel (FMC Bioproducts, Rockland, ME) in
0.5x Tris borate EDTA on a sequencing apparatus at a constant power of
6 watts. Dried gels were autoradiographed overnight. Samples showing
aberrant bands were manually sequenced using the Sequiterm EXCEL II DNA
sequencing kit (Epicentre, Madison, WI). For each sequence difference
that was identified, genotyping was performed by PCR, dot-blotting, and
allele-specific hybridization. Genotype and allele frequencies in
family probands, type 2 diabetes subjects, and nondiabetic controls
were compared by 
2 test. Haplotype frequencies
were estimated by maximum likelihood methods.
|
Results
|
|---|
Three abnormal band patterns were identified in ddF gels: one with
the forward, one with the internal, and one with the reverse primer.
The polymorphism identified by the internal primer was a rare T to C
transition at position 544 of the NGN3 gene that did not
affect the amino acid sequence (L75L). The other two were frequent
polymorphisms: one a CA insertion/deletion at position -44 of intron
1, and the other a C to T transition at position 917 that caused the
substitution of phenylalanine for serine at amino acid position 199
(S199F). These two polymorphisms were in significant linkage
disequilibrium ([
[standardized coefficient of linkage
disequilibrium] = -0.48; P < 0.0001). Allele 199S
was observed in conjunction with either -44ins or -44del (haplotypes
-44ins/199S and -44del/199S), whereas allele 199F was found only in
association with -44ins (haplotype -44ins/199F).
Both polymorphisms were similarly frequent in the 91 family probands
and 179 nondiabetic controls, indicating that these sequence
differences did not contribute to MODY or other forms of autosomal
dominant diabetes (Table 3). Results were
unchanged when the 8 non-Caucasian probands were excluded from the
analysis or when families with early-onset and later-onset disease were
analyzed separately. A weak, but significant, association was detected
between allele 199F and common type 2 diabetes (P =
0.05; Table 3). Because of the linkage disequilibrium between positions
199 and -44, allele -44ins also had a tendency to be associated with
type 2 diabetes, but this was not statistically significant
(P = 0.07). Carriers of allele 199F had a 1.7-fold risk
of type 2 diabetes compared with noncarriers (95% confidence interval,
1.04–2.7). No significant relation with age at diagnosis or body mass
index was found.
|
Discussion
|
|---|
Our data indicate that genetic variability in the NGN3
gene is unlikely to play a major role in the etiology of MODY or other
forms of autosomal dominant diabetes. The diabetes-causing mutations in
the MODY genes that have been identified to date are mostly rare DNA
sequence differences affecting the amino acid sequence
(2, 3, 4, 5, 6, 7). In our 91 families, we did identify 1 private
mutation, but it did not change the amino acid sequence. The unlikely
possibility of a contribution of common polymorphisms was also ruled
out by the absence of any significant difference in allele frequencies
between family probands and nondiabetic controls.
In considering these negative results, one must take into account the
limitations of our study. Only one proband per family was screened for
mutations. If any of them was a phenocopy that happened by chance to be
in the family, diabetes-causing mutations would have been missed. Also,
mutations might have been missed by our screening method. Although the
sensitivity of ddF is higher than that of single strand conformation
polymorphism analysis, it is less than 100% for sequence differences
located far away from the ddF primer (10). It is also
important to note that the mutation screening was limited to the coding
sequence, and therefore, the role of mutations in regulatory regions
was not evaluated. Finally, these negative results only apply to the
Caucasian population, because the number of non-Caucasian families that
were included in the study was too small to draw meaningful
conclusions.
The finding of a significant association between allele 199F and type 2
(nonautosomal-dominant) diabetes is intriguing, because the S199F
substitution affects a residue that is conserved across multiple
species. The fact that the frequency of the 199F allele was not
increased in the family probands might seem inconsistent with this
finding, but it is not if one considers the very stringent selection
criteria for these families. The family probands represent an uncommon
form of diabetes that is transmitted as a Mendelian autosomal dominant
trait with high penetrance, such as successive generations are
affected. Susceptibility genes that have moderate penetrance will not
produce such pedigrees, but may well be represented in a collection of
common type 2 diabetes patients. Of course, one cannot exclude that
this finding was due to chance or to unrecognized population
stratification. Thus, this association will have to be validated
through replication studies in other populations and by investigating
the effect of the S199F substitution on NGN3 transcriptional
activity.
|
Footnotes
|
|---|
1 This work was supported by NIH Grants DK-55523 (to A.D.) and
DK-47475 (to A.S.K.) and Joslin Diabetes and Endocrinology Research
Center Grant DK-36836 (Genetics Core). 
Received November 9, 2000.
Revised January 9, 2001.
Accepted January 29, 2001.
|
References
|
|---|
-
Edlund H. 1998 Transcribing pancreas. Diabetes. 47:1817–1823.[Abstract]
-
Vionnet N, Stoffel M, Takeda J, et al. 1992 Nonsense mutation in the glucokinase gene causes early-onset
non-insulin-dependent diabetes mellitus. Nature. 356:721–722.[CrossRef][Medline]
-
Yamagata K, Oda N, Kaisaki PJ, et al. 1996 Mutations in the hepatocyte nuclear factor-1 gene in maturity-onset
diabetes of the young. Nature. 384:455–458.[CrossRef][Medline]
-
Yamagata K, Furuta H, Oda N, et al. 1996 Mutations
in the hepatocyte nuclear factor-4 gene in maturity-onset diabetes
of the young. Nature. 384:458–460.[CrossRef][Medline]
-
Stoffers DA, Ferrer J, Clarke WL, Habener JF. 1997 Early-onset type-II diabetes mellitus (MODY4) linked to IPF1. Nat
Genet. 17:138–139.[CrossRef][Medline]
-
HorikawaY, Iwasaki N, Hara M, et al. 1997 Mutation
in hepatocyte nuclear factor-1ß gene (TCF2) associated with MODY. Nat
Genet. 17:384–385.[CrossRef][Medline]
-
Malecki M, Jhala US, Antonellis A, et al. 1999 Mutations in NEUROD1 are associated with the development of type 2
diabetes mellitus. Nat Genet. 23:323–328.[CrossRef][Medline]
-
Doria A, Yang Y, Malecki M, et al. 1999 Clinical
characteristics of early-onset, autosomal dominant type-2 diabetes
unlinked to known MODY genes. Diabetes Care. 22:253–261.[Abstract/Free Full Text]
-
Gradwohl G, Dierich A, LeMeur M, Guillemot F. 2000 Neurogenin3 is required for the development of the four endocrine cell
lineages of the pancreas. Proc Natl Acad Sci USA. 97:1607–1611.[Abstract/Free Full Text]
-
Sarkar G, Yoon HS, Sommer SS. 1992 Dideoxy
fingerprinting (ddF): a rapid and efficient screen for the presence of
mutations. Genomics. 13:441–443.[CrossRef][Medline]
Top
Abstract
Introduction
