| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Gene in Typical Familial Type 2 Diabetes: Evidence for Novel Mutations in Exons 8 and 101
Division of Endocrinology and Metabolism, John L. McClellan Memorial Veterans Hospital, and University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205
Address all correspondence and requests for reprints to: Steven C. Elbein, M.D., Endocrinology 111J/LR, John L. McClellan Memorial Veterans Hospital, 4300 West 7th Street, Little Rock, Arkansas 72205. E-mail: sce{at}nidgene1.uams.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
(HNF1) gene are an
important cause of autosomal dominant diabetes with onset before age 25
yr [maturity-onset diabetes of the young (MODY)], and some regions of
the HNF1 gene appear to be hot spots for mutations. To evaluate the
role of HNF1 in the more common familial type 2 diabetes, we studied
62 families of Northern European origin by linkage analysis and
molecular screening. Linkage was rejected under dominant models
consistent with either late-onset type 2 diabetes or early-onset
dominant diabetes. We used single strand conformation polymorphism
analysis to screen 53 diabetic members of 36 families who reported
diabetes diagnosed before age 40 yr, 9 members of 2 Utah families with
typical MODY, and 24 additional members of families with possible
linkage. One MODY family showed the previously reported frameshift
mutation (P291fsinsC) in exon 4. Among the individuals with more
typical type 2 diabetes, we identified the previously reported common
polymorphisms, a new intronic polymorphism, and 3 common amino acid
variants. We also identified 2 novel missense mutations that segregated
with type 2 diabetes in 1 family each: lysine for glutamic acid
substitution at codon 619 in exon 10 (E619K), and an arginine for
threonine substitution at codon 537 in exon 8 (R537T) in a second
family. The exon 8 mutation showed relatively low penetrance, and the
role in this family remains uncertain. No coding mutations were
identified in the family members screened on the basis of linkage but
without early-onset diabetes. Although HNF1 mutations are not a
common cause of familial type 2 diabetes, they may account for 5% of
families in which at least 1 member has onset of type 2 diabetes before
age 40 yr. Incomplete penetrance and a high sporadic frequency make
linkage an inefficient screening tool. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
(HNF1) (6), or HNF4 (7). These mutations cause three subsets of
diabetes, MODY2, MODY3, and MODY1, respectively, which differ in the
severity of diabetes, but share a common defect in insulin secretion
(8, 9, 10).
Although MODY2 (glucokinase) appears to be the most common form of
early-onset, autosomal dominant diabetes in France (11), both linkage
studies and molecular screening have identified few mutations in
families with more typical diabetes (12, 13, 14). MODY1 mutations have been
demonstrated in relatively few families (6, 15, 16), although several
linkage studies suggest possible loci for typical type 2 diabetes on
chromosome 20 (2, 17, 18). Initial linkage studies of the MODY3 region
did not suggest any role in French (19) or Hispanic sibling pairs (20).
However, Mahtani et al. (3) reported linkage to the region
of MODY3 in a subset of Botnian Finnish families with the lowest
quartile of insulin secretion. This locus appears to be the most common
cause of MODY outside of France (21, 22), and at least 41 mutations
have been identified (23). Of interest, several mutations have been
identified in multiple independent and unrelated families, suggesting
the presence of potential mutational hot spots (21, 24). Furthermore, a
recent German study (21) found HNF1 mutations in 9 of 25 type 2
diabetic individuals who had diabetes onset before age 35 yr but
otherwise did not fit a typical MODY pattern of inheritance, whereas an
English study found a common mutation in 2 of 32 diabetic individuals
who had onset before age 40 yr (22).
Based on the high frequency of HNF1 mutations among early-onset
diabetic families, the evidence for a potential subset of insulinopenic
type 2 diabetics with a susceptibility gene near MODY3, and
the potential for mutational hot spots in this gene, we hypothesized
that a subset of typical type 2 diabetic families with relatively early
onset have HNF1 mutations. We initially tested this hypothesis by
linkage analysis of markers near the MODY3 locus. We
supplemented linkage studies with molecular screening experiments of
all 10 HNF1 exons in 53 individuals from 36 of the same families who
had type 2 diabetes diagnosed before age 40 yr, an additional 9 members
of 2 families with more typical MODY, and 22 individuals from 6
families in which the individual LOD scores suggested linkage to MODY3.
We found the previously described exon 4 frameshift mutation in 1 of 2
Utah MODY families, but we report novel amino acid substitutions in
exons 8 and 10 in families with more typical type 2 diabetes that were
not found in 100 unrelated diabetic individuals or 93 control
individuals.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Linkage studies were performed on 62 families ascertained for grandparents of Northern European ancestry, 2 siblings with onset of typical type 2 diabetes before age 65 yr, and no more than 1 parent known to be affected. As described previously (25, 26), each available sibling or offspring of a diabetic family member over age 18 yr was tested with a standard 75-g oral glucose tolerance test. For purposes of linkage studies, individuals were considered affected if they met WHO criteria (27) for either diabetes (fasting glucose >7.8 mmol/L or 2-h postchallenge glucose >11.1 mmol/L) or impaired glucose tolerance before age 45 yr (2-h postchallenge glucose between 7.8–11.1 mmol/L), or diabetes between ages 45–65 yr. After age 65 yr, we considered a 2-h postchallenge glucose in excess of 220 mg/dL (11.1 mmol/L) to be diagnostic of diabetes in the absence of fasting hyperglycemia. All other individuals who were not normal by glucose tolerance testing were considered of unknown affection status. The current studies include 20 families similarly ascertained but not included in earlier reports.
Molecular screening for MODY3/HNF1 mutations was conducted initially
on 53 family members who had type 2 diabetes by WHO criteria (27) with
reported age of diagnosis at or before 40 yr. These individuals were
not different from other diabetic family members. They represented 36
families and comprised 21 males and 32 females with mean age of onset
35.3 ± 4.7 yr, mean age at time of study of 50.9 ± 13.4 yr,
and body mass index of 31.5 ± 6.4 kg/m2.
Subsequently, we included additional diabetic family members regardless
of age of onset based on suggestive linkage to at least 1 marker under
at least 1 model. These criteria added 4 families in which no member
met the age criteria. Finally, we included 9 diabetic members of 2
typical MODY family (onset before age 25 yr in 2 or more diabetic
members, apparent autosomal dominant transmission) ascertained from the
same population. Additional diabetic patients screened to determine the
prevalence of specific mutations were ascertained similarly to the 62
families studied, but did not have other relatives available for study.
Nondiabetic control samples comprised spouses of family members who had
normal glucose tolerance tests.
Linkage studies
Four markers were typed in the vicinity of HNF1: D12S79,
D12S86, D12S76, and D12S324 (19, 28, 29), using methods described
previously (25, 26). Linkage analysis was conducted under four dominant
models that include a range of allele frequencies, penetrance
functions, and sporadic frequencies (details available from S. C.
Elbein), including models described previously (25, 26). Only typical
type 2 diabetic families were tested for linkage.
Molecular scanning of HNF1 gene
Each of the 10 exons of the HNF1 gene was scanned for
mutations using single strand conformation polymorphism analysis of
enzymatically amplified DNA (SSCP), as described by Orita et
al. (30) and in previous studies from our laboratory (31, 32). The
primers were those described by Yamagata et al. (6), except
for exon 9, where we used the alternative primer sequences described by
Kaisaki et al. (21). PCR products for exons 1, 2, 4, 5, and
7 were digested with enzymes BglI, PstI,
RsaI, PstI, and BglI, respectively,
before analysis. All resulting fragments were less than 400 bp, which
in our experience leads to high sensitivity. Amplification products
separated on 5% nondenaturing polyacrylamide gel (Accugel, National
Diagnostics, Atlanta, GA) at 8 watts under four conditions: room
temperature and 4 C, with and without 10% glycerol. PCR product was
also examined by heteroduplex analysis on 1 x MDE (mutation
detection enhancement) gel (FMC BioProducts, Rockland, ME) at a
constant voltage of 20 V/cm.
Sequence analysis
Genomic DNA was reamplified using the same primer set, and the
amplification product was isolated on an agarose gel and column
purified (PCR Purification Kit, Qiagen, Santa Clarita, CA). Dideoxy
sequence analysis was conducted on double-stranded PCR product using
ThermoSequenase (Amersham, Arlington Heights, IL) with
-32P (ICN Pharmaceuticals) end labeled forward or
reverse primers. Sequence variants were determined by comparison with
published sequence (6, 21), and those of individuals were determined
without SSCP variants.
PCR-RFLP analysis of exon 10 Glu619Lys (E619K) mutation
The exon 10 Glu619Lys (E619K) mutation was confirmed by failure of TaqI to digest the 248-bp exon 10 PCR product into bands of 160 and 88 bp.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We next tested by molecular screening 53 members who reported the diagnosis of type 2 diabetes before age 40 yr. The members represented 36 of 62 families tested for linkage. We tested all diabetic members of 6 families based on possible linkage (an additional 22 members) and 9 members of 2 families ascertained for onset before age 25 yr. Variant patterns were found in exons 1, 2, 3, 4, 6, 7, 8, and 10, including the common amino acid polymorphisms Ile27Leu (exon 1), Ala98Val (exon 1), and Ser487Asn (exon 7). Each expected common variant was detected by our SSCP assay, with the exception of a silent variant at codon 550 (GAG to CAG) (22) in exon 9 that was not seen in our sample. Additionally, a T for C substitution at position -27 of intron 5 segregated with type 2 diabetes in 1 family selected for a single individual with onset at age 40 yr and possible linkage. This sequence variation does not appear to introduce a cryptic splice site and is expected to be silent.
The previously described (6, 21, 24, 33) frameshift mutation P291fsinsC
in exon 4 was detected by SSCP and confirmed by sequence analysis in
all three diabetic members of one of two MODY families. No abnormality
was detected in the second family. Ages of onset for the P291fsinsC
ranged from 13–37 yr (Table 1
). No other
previously described mutations were detected. Novel variant patterns
were seen in exon 8 (room temperature without glycerol) in members of 1
family and in exon 10 (4 C without glycerol) in a second family.
Sequence analysis of exon 10 showed the common silent nucleotide
substitution (C/T) in intron 9 at position -24, and a heterozygous
adenine (A) for guanine (G) substitution at the first position of codon
619, resulting in a lysine for glutamic acid substitution (E619K; Fig. 1). Analysis of HNF1 sequences of rat,
mouse, hamster, and chicken show that this region of the serine-rich
activation domain is highly conserved, and the glutamic acid is
invariant. We used the loss of a TaqI site to confirm the
segregation of the E619K allele in members of family 25 (Figs. 2 and 3).
The proband, who was heterozygous, did not share the variant with his
diabetic sister, but transmitted it to three affected offspring. The
clinical characteristics of the six diabetic family members (Table 1)
were remarkable only for a relatively early onset among E619K carriers
and a variety of therapies.
|
|
|
|
;
T537R). The threonine is invariant in this position in rat, hamster,
mouse, and chicken. We used the SSCP pattern to confirm segregation of
the variant with haplotypes from markers D12S79 and D12S86 (D12S76 was
uninformative) in family 2 (Fig. 5). The
clinical characteristics of the family members, including diabetic and
nondiabetic carriers, are shown in Table 1
. Although all four diabetic
individuals carried the T537R allele, the allele was also carried by
five nondiabetic individuals, including an obligate carrier (individual
7; Fig. 6) who had impaired glucose
tolerance (2-h postchallenge glucose between 7.8–11.1 mmol/L). Four
carriers had normal glucose tolerance despite an age at testing of up
to 54 yr (Table 1 and Fig. 6).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
gene are an important cause of MODY
worldwide, with mutations now demonstrated in France, England, Germany,
Finland, Japan, and the United States (21, 22, 24, 34, 35, 36). Unlike
glucokinase mutations, which account for half of all MODY in France
(34) but little diabetes elsewhere, HNF1 mutations are a frequent
cause of early-onset, autosomal dominant diabetes. Several mutations
have appeared in multiple independent families, and a site in exon 4
appears to be a mutational hot spot (21, 24). We now report that the
mutation P291fsinsC, a single base pair insertion that introduces a
frame shift in exon 4 with premature termination, is also present in a
typical MODY family from Utah. As in other families with this mutation,
two of the three family members had diabetes onset before age 20 yr,
and thus are quite distinct from the other two missense mutations that
we report.
The role of HNF1 mutations in more typical diabetes, even with early
onset, is unclear. Mutations of HNF1 accounted for 9 of 25 German
individuals whose diabetes onset was before age 35 yr (21), but they
were not common among late-onset individuals from Sweden and Finland
(24), Denmark (37), or Japan (33). Mahtani et al. (3)
reported linkage of the MODY3 region to a subgroup of type 2
diabetes with low insulin secretory response to glucose, although no
mutations have been reported. Linkage analysis of the MODY3
region in France (19), in Hispanic sibling pairs (20), and in our
sample have all been negative, suggesting that MODY3 is not
a major diabetes gene in unselected type 2 diabetic families. However,
detection of the small number of families in which HNF1 plays a role
may be difficult by linkage, particularly in the setting of sporadic
cases and incomplete penetrance.
In addition to the P291fsinsC MODY mutation, we have identified two
novel mutations that segregate with typical type 2 diabetes in 2 of 36
families in which an individual had onset of diabetes before age 40 yr.
Although our study is most comparable to that of Kaisaki et
al. (21), in that our individuals have diabetic first degree
relatives and relatively early onset, HNF1 mutations are not as
significant a cause of type 2 diabetes in these Utah families as in the
German population. Nonetheless, our data suggest that among similarly
selected families of Northern European decent, 5% may have mutations
of the HNF1 gene. Thus, we concur with the conclusions of Kaisaki
et al. (21) that it may be beneficial to screen for
mutations in HNF1 in individuals with early-onset diabetes who do
not meet the criteria for MODY, even when such individuals are obese.
Although linkage was suggestive in individual families, the
cosegregation of markers in the MODY3 region with type 2
diabetes was less helpful than the age of onset in selecting families
with mutations. The only variant detected in one of these families was
an intronic substitution that does not alter a splice site or introduce
a cryptic splice site, and thus seems unlikely to be causing
diabetes.
In addition to the two rare alleles, we identified a large number of previously described variants and the exon 4 frameshift mutation. Thus, our SSCP-based methods were sufficiently sensitive to detect most potential mutations. This sensitivity should be improved further by our inclusion of heteroduplex screening for all exons. The two common amino acid variants, Ile27Leu in exon 1 and Asp487Ser in exon 7 have not been associated with type 2 diabetes, although the Ala98Val allele was reported to diminish insulin secretion (37). No common variant segregated with diabetes in our study (data not shown).
The codon 515 and 619 mutations detected in our study arise in exons
for which only one mutation each was previously described (23) and are,
to our knowledge, novel. The exon 10 E619K allele shows the high
penetrance typical of MODY3/HNF1 mutations; each carrier has
diabetes, although the age of diagnosis is variable. This segregation,
the highly conserved nature of this region, the nonconservative amino
acid substitution (acidic to basic residue), and the absence of this
mutation in normal individuals all argue that the E619K allele is
causative of the diabetes in this family. The family is atypical for
MODY, in that the affected individuals are considerably obese.
Furthermore, the sibling pair upon which the family was ascertained is
discordant for this mutation. Finally, the unrelated mother of the
three young-onset diabetic individuals was herself diabetic upon
testing. Thus, inspection of family 25 suggested that bilineal
transmission in combination with obesity could easily account for the
relatively early onset of type 2 diabetes. The apparent simultaneous
segregation of MODY genes with typical late-onset type 2 diabetes or
phenocopies is well described (6, 21, 38) and would diminish evidence
of linkage.
The role of the exon 8, T537R mutation is less certain. This codon is
also highly conserved among mammals and results in substitution of a
basic amino acid (arginine) for a neutral amino acid (threonine). In
contrast, the three observed amino acid polymorphisms are conservative.
Like the exon 10 mutation, T537R occurs in the
trans-activation domain of HNF1. The exact effect of
these mutations in causing diabetes is uncertain and will require
detailed studies of the protein. The R537 allele segregates in each
diabetic member of family 2, but also occurs in several individuals who
show no evidence of diabetes on oral glucose tolerance testing, even in
their mid-fifties. Incomplete penetrance has been observed in both
MODY3 (6, 22, 34) and other mutations causing early-onset diabetes (38, 39). The penetrance of T537R in the heterozygous state in family 2 may
be under 60% at age 50 yr and only 30% at age 20 yr. Lower penetrance
may result from the difference in the way our families were ascertained
(onset before age 40 yr) in contrast to studies of MODY (onset before
age 25 yr). Even among members of our family with true MODY, age of
onset varied from 13–37 yr, with a mutation that can be expected to
completely inactivate the protein.
We have previously reported a relatively common variant at codon 985 in exon 17 (methionine) of the insulin receptor in family 2 that increased postchallenge glucose levels (14). This insulin receptor variant also appeared to predispose to hyperglycemia in a Dutch population study (40). We found no evidence of epistatic interaction with T537R, and several individuals carried both variant alleles without evidence of diabetes.
The majority of previously described HNF1 mutations have clustered
in exons 1, 2, 4, 6, and 9. This finding may represent a selection bias
for the most severe mutations imposed by screening members of MODY
families (diabetes before age 25 yr in lean individuals). Although both
occur highly conserved regions and result in nonconservative amino acid
substitutions in the trans-activation domain, the mutations
in exons 8 and 10 may be less disruptive of HNF1 function and thus
require other genetic or environmental factors (advanced age or
obesity) for expression. More sophisticated clinical studies of
nondiabetic carriers may provide additional information about the roles
of these mutations in diabetes pathogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received November 18, 1997.
Revised January 28, 1998.
Accepted February 26, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
gene in maturity-onset
diabetes of the young (MODY3). Nature. 384:455–458.[CrossRef][Medline]
gene in maturity-onset diabetes of
the young (MODY1). Nature. 384:458–460.[CrossRef][Medline]
/MODY1 gene. J Clin Invest. 100:1400–1405.[Medline]
gene in MODY and
early-onset NIDDM: evidence for a mutational hotspot in exon 4. Diabetes. 46:528–535.[Abstract]
gene are a common cause
of maturity onset diabetes of the young in the U.K. Diabetes. 46:720–725.[Abstract]
binding site in the promoter of the
hepatocyte nuclear factor-1 gene. Diabetes. 46:1648–1651.[Abstract]
gene (MODY3) are not a
major cause of late-onset NIDDM in Japanese subjects. Diabetes. 46:1512–1513.[Medline]
gene associated with maturity-onset diabetes of the young (MODY3). Hum Mol Genet. 6:583–586./MODY3 gene in Japanese subjects with
early- and late-onset NIDDM. Diabetes. 46:1504–1508.[Abstract]
gene is associated with reduced serum C-peptide and insulin
responses to an oral glucose challenge. Diabetes. 46:912–916.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  H. S. Freitas, G. F. Anhe, K. F. S. Melo, M. M. Okamoto, M. Oliveira-Souza, S. Bordin, and U. F. Machado Na+-Glucose Transporter-2 Messenger Ribonucleic Acid Expression in Kidney of Diabetic Rats Correlates with Glycemic Levels: Involvement of Hepatocyte Nuclear Factor-1{alpha} Expression and Activity Endocrinology, February 1, 2008; 149(2): 717 - 724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Elbein and M. A. Karim Does the Aspartic Acid to Asparagine Substitution at Position 76 in the Pancreas Duodenum Homeobox Gene (PDX1) Cause Late-Onset Type 2 Diabetes? Diabetes Care, August 1, 2004; 27(8): 1968 - 1973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Das, S. J. Hasstedt, Z. Zhang, and S. C. Elbein Linkage and Association Mapping of a Chromosome 1q21-q24 Type 2 Diabetes Susceptibility Locus in Northern European Caucasians Diabetes, February 1, 2004; 53(2): 492 - 499. [Abstract] [Full Text]
|
|
|
|
|
|
S. C. Elbein Perspective: The Search for Genes for Type 2 Diabetes in the Post-Genome Era Endocrinology, June 1, 2002; 143(6): 2012 - 2018. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Aguilar-Salinas, E. Reyes-RodrÍguez, Ma. L. Ordóñez-Sánchez, M. A. Torres, S. Ramírez-Jiménez, A. Domínguez-López, J. R. MartÍnez-Francois, Ma. L. Velasco-Pérez, M. Alpizar, E. GarcÍa-GarcÍa, et al. Early-Onset Type 2 Diabetes: Metabolic and Genetic Characterization in the Mexican Population J. Clin. Endocrinol. Metab., January 1, 2001; 86(1): 220 - 226. [Abstract] [Full Text]
|
|
|
|
|
|
K. C. Chiu, L.-M. Chuang, J. M. Ryu, G. P. Tsai, and M. F. Saad The I27L Amino Acid Polymorphism of Hepatic Nuclear Factor-1{alpha} Is Associated with Insulin Resistance J. Clin. Endocrinol. Metab., June 1, 2000; 85(6): 2178 - 2183. [Abstract] [Full Text]
|
|
|
|
 |
|
 Q.-X. Hua, M. Zhao, N. Narayana, S. H. Nakagawa, W. Jia, and M. A. Weiss Diabetes-associated mutations in a beta -cell transcription factor destabilize an antiparallel "mini-zipper" in a dimerization interface PNAS, February 29, 2000; 97(5): 1999 - 2004. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
