Clinical Chemistry 47:9
1641–1648 (2001)
Molecular Diagnostics
and Genetics
Mitochondrial Gene Mutations in the
tRNALeu(UUR) Region and Diabetes: Prevalence
and Clinical Phenotypes in Japan
Kumiko Ohkubo,1* Akemi Yamano,2 Mariko Nagashima,1 Yumiko Mori,1 Keizo Anzai,1
Yuko Akehi,1 Riku Nomiyama,2 Takashi Asano,2 Akinori Urae,3 and Junko Ono1
Background: Mitochondrial gene mutations play a role
in the development of diabetes mellitus. We have assessed the frequency of the A3243G and other mitochondrial mutations in Japan and in the relationship to
clinical features of diabetes.
Methods: DNA was obtained from peripheral leukocytes of 240 patients with diabetes mellitus (39 with type
1; 188 with type 2; 13 with gestational diabetes) and 125
control subjects. We used PCR-restriction fragment
length polymorphism analysis (ApaI) for A3243G and
PCR-single-strand conformation polymorphism analysis to determine the mutations in the mitochondrial
gene including nucleotide position 3243.
Results: The A3243G mutation was found in seven
patients, and an inverse relationship was observed between the degree of heteroplasmy and the age at onset of
diabetes. A3156G, G3357A, C3375A, and T3394C were
detected in addition. Those who shared the same mutation showed similar clinical characteristics, thus representing a putative clinical subtype. The patients with
A3156G had a sudden onset of hyperglycemia and
showed a rapid progression to an insulin-dependent
state with positive anti-glutamic acid decarboxylase
antibody. Those with T3394C showed a mild defect in
glucose-stimulated insulin secretion, and hyperglycemia
appeared after adding such factors as aging or obesity.
Conclusions: The identification of mitochondrial gene
mutations allows preclinical diagnosis of diabetes and
1
Department of Laboratory Medicine, Fukuoka University School of
Medicine, 7-45-1, Nanakuma, Jonan-ku, Fukuoka, 814-0180, Japan.
2
The First Department of Internal Medicine, Fukuoka University School of
Medicine, 7-45-1, Nanakuma, Jonan-ku, Fukuoka, 814-0180, Japan.
3
Kyushu Clinical Pharmacology Research Clinic, 2-13-16, Jigyo, Chuo-ku,
Fukuoka, 810-0065, Japan.
*Author for correspondence. Fax 81-92-873-1050; e-mail kokubo@cis.
fukuoka-u.ac.jp.
Received April 9, 2001; accepted June 20, 2001.
prediction of the age at onset by evaluating the degree of
heteroplasmy in cases with A3243G. Mutation detection
may also be important for patient management and
identification of affected family members.
© 2001 American Association for Clinical Chemistry
Mitochondria are organelles that play an important role in
the energy production of a cell, and they also have
extranuclear genes, which are transmitted maternally.
Especially in pancreatic ␤ cells, mitochondria are responsible for glucose-induced insulin secretion because the
exocytosis of secretory granules is triggered by changes in
the intracellular ATP and ADP concentrations and subsequent increase in the ATP/ADP ratio as a result of
oxidative phosphorylation in the mitochondria caused by
glucose metabolism (1 ). Mutations in mitochondrial
genes may lead to disorders in ATP production, and
thereby impair the glucose-induced insulin secretion.
The mitochondrial DNA is a ringed, double-stranded
DNA, measuring 16 569 bp in overall length, and exists in
the mitochondrial matrix. A cell contains 103–104 molecules of mitochondrial DNA. The mitochondrial DNA
encodes 13 subunits of the mitochondrial respiratory
chain, 22 tRNAs, and 2 rRNAs. The mutations of mitochondrial DNA constitute one of the main causes of the
diseases that involve abnormalities in the brain and
muscular tissues. An impaired glucose metabolism is also
a symptom that is often seen in patients with mutations of
mitochondrial DNA. Among these diseases, mitochondrial encephalomyopathy with lactic acidosis and strokelike episodes (MELAS)4 is known to be a syndrome that
4
Nonstandard abbreviations: MELAS, mitochondrial encephalomyopathy
with lactic acidosis and stroke-like episodes; np, nucleotide position; PCRRFLP, PCR-restriction fragment length polymorphism; PCR-SSCP, PCR-single-strand conformation polymorphism; BMI, body mass index; and GAD,
glutamic acid decarboxylase.
1641
1642
Ohkubo et al.: Detecting Mitochondrial Gene Mutations in Japanese
frequently accompanies diabetes mellitus. The screening
of the A-to-G mutation at nucleotide position (np) 3243,
found in 1992 in patients with MELAS (2 ), has been
applied recently among diabetic patients. As a result, a
maternally transmitted clinical subtype of diabetes mellitus, called diabetes-deafness syndrome (OMIM 520000),
has been revealed. Different mutations based on the
substitution of one base pair have been found in this
region of mitochondrial DNA (3–5 ), thus showing it to be
a hot spot area for mutations. Patients with the A3243G
mutation have been reported to comprise 1–3% of all
diabetic patients in Japan (6 – 8 ). The clinical characteristics of the patients with the A3243G mutation include
impaired insulin secretion, sensorineural deafness, and
maternal inheritance of the disease. Once impaired insulin
secretion becomes overt, most cases progressively develop a disease state of insulin dependence. Therefore, the
early or preclinical diagnosis of this disease by detection
of such a mutation might help to prevent or delay the
onset of diabetes, while also allowing for the selection of
optimal treatment, thereby possibly avoiding diabetic
complications altogether.
We developed a rapid diagnostic system using PCRrestriction fragment length polymorphism (PCR-RFLP)
and PCR-single-strand conformation polymorphism
(PCR-SSCP) analysis for screening mutations in the
tRNALeu(UUR) region of mitochondrial DNA. We determined the detection limit of our method, which was
essential for applying this system as a screening test
because mutant DNA and wild-type DNA could coexist
in a single cell known as heteroplasmy. A total of 240
diabetic patients and 125 controls were examined by use
of this screening method.
Subjects and Methods
patients and controls
Peripheral blood was obtained from 240 unrelated Japanese with diabetes mellitus (188 with type 2; 39 with type
1; 13 with gestational diabetes). The mean age was 52.8 ⫾
16.2 years. Individual patients were considered to be
affected if they were being treated for type 2 diabetes or if
the results of a standard 75-g oral glucose tolerance test
indicated diabetes mellitus. Type 1 diabetes had been
clinically diagnosed at the onset based on WHO criteria
(9 ). The nondiabetic control group consisted of 125 unrelated Japanese (mean age, 22.9 ⫾ 2.7 years) who showed
fasting blood glucose of ⬍7 mmol/L and body mass index
(BMI) of ⬍24 kg/m2. Informed consent, according to the
medical ethics criteria of Fukuoka University, was obtained from all participants.
materials
A restriction enzyme, ApaI, was purchased from Takara.
AmpliTaq DNA polymerase was from Roche Molecular
Systems. SYBR Green I was obtained from FMC Bioproducts, and PhastGels, PhastGel Buffer Strips Native, and
PhastGel DNA silver staining reagent sets were from
Amersham Pharmacia Biotech. Plasmid vector pT7Blue T
was from Novagen. SepaGene™ was from Sanko Junyaku. Other reagents were of analytical grade.
preparation of dna and pcr amplification
Total DNA was extracted from peripheral leukocytes with
SepaGene according to the manufacturer’s instructions.
The fragments of mitochondrial DNA encompassing np
3243 were amplified by PCR with AmpliTaq DNA polymerase. PCR was carried out in a total volume of 100 ␮L
containing 100 ng of extracted DNA, 200 ␮M each dNTP,
10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and
1.0 U of Taq polymerase. The forward primers were
5⬘-CGTTTGTTCAACGATTAAAG-3⬘ (MT1) and 5⬘-AGGACAAGAGAAATAAGGCC-3⬘ (MT3), and the reverse
primers were 5⬘-AGCGAAGGGTTGTAGTAGCC-3⬘ (MT2)
and 5⬘-CACGTTGGGGCCTTTGCGTA-3⬘ (MT4). The DNA
was initially denatured at 94 °C for 5 min and subjected to
30 PCR cycles of 94 °C for 1 min, 55 °C for 1 min, and
72 °C for 1 min. The PCR products were electrophoresed
on 4% agarose gel and stained with ethidium bromide.
rflp analysis
The 422-bp fragment amplified by the primers MT1 and
MT2 was digested by ApaI to identify any A-to-G mutation at np 3243 and electrophoresed on 4% agarose gels.
After electrophoresis, the gels were stained with ethidium
bromide or SYBR Green I. Fluoroscopic scans and analyses were performed with ChemiImager (Alpha Innotech).
sscp analysis
After PCR with the primers MT3 and MT4, the mixture
containing the 294-bp amplified fragment was diluted
twofold with a solution containing 950 mL/L formamide,
20 mmol/L EDTA, 0.5 g/L xylene cyanol, and 0.5 mL/L
bromphenol blue. The sample was heated to 95 °C for 5
min, and after cooling on ice, 1 ␮L was separated on 20%
polyacrylamide gels with PhastGel native buffer strips
(0.25 mol/L Tris, 0.88 mol/L l-alanine, pH 8.8) on the
PhastSystemTM (Amersham Pharmacia Biotech) (10 ). After electrophoresis, the gel was fixed and stained with a
PhastGel DNA silver staining reagent set.
dna sequencing
Mutations detected by PCR-RFLP (ApaI) and/or PCRSSCP were confirmed by direct DNA sequencing of PCR
products or DNA sequencing after cloning into plasmid
vectors. The sequencing reactions were carried out by use
of the BigDye terminator cycle sequencing method (Perkin-Elmer), and reaction products were analyzed on an
ABI-Prism 377 automated sequencer (Perkin-Elmer).
preparation for calibrators of the pcr products
encompassing the mutation of a3243g
To determine the detection limit of the A3243G mutation,
we constructed the mixture of PCR products with known
1643
Clinical Chemistry 47, No. 9, 2001
Fig. 1. Detection limit of the A3243G mutation by
PCR-RFLP (ApaI).
PCR products containing various percentages of A3243G
mutation indicated above each lane were digested with ApaI
for 1 h at 37 °C. After electrophoresis on a 4% agarose gel,
they were stained with ethidium bromide (lanes 1–9) or
SYBR Green I (lanes 10 –14). The arrow indicates the two
fragments overlapping after digestion with ApaI. MW, molecular weight.
proportions of amplified DNA that contained both the
A3243G mutation and no mutation. The PCR products
amplified with DNA from a patient with the heteroplasmic A3243G mutation, using primers MT1 and MT2, were
subcloned into a plasmid vector, pT7Blue T. After cloning
of the plasmid vector inserted with the mitochondrial
DNA containing A3243G or no mutation, the inserted
DNA was amplified with the primers MT1 and MT2 by
PCR. The products containing only the A3243G mutation
or no mutation were extracted with chloroform, precipitated with ethanol, resolubilized in distilled water, and
quantified by measurement of the absorbance at 260 nm.
The products from DNA containing the A3243G mutation
or no mutation were diluted to a concentration of 150
ng/␮L with distilled water and mixed at ratios of 0:100,
1:99, 2:98, 5:95, 10:90, 20:80, 40:60, 50:50, and 100:0 by
volume. Each 150 ng of the mixed DNA was digested with
10 U of ApaI in 10 ␮L of the reaction mixture for 1 h at
37 °C. The digested DNA was electrophoresed on a 4%
agarose gel and stained with ethidium bromide or SYBR
Green I. Fluoroscopic scans and analyses were performed
with ChemiImager. For PCR-SSCP, PCR products with or
without mutation, amplified with the primers MT3 and
MT4, were mixed as above and electrophoresed by use of
the PhastSystem as described above.
statistical analysis
The statistical analysis was performed with the Stat
View-J 4.02 program (Abacus Concepts). Fisher exact test
or the ␹2 test was used for statistical analyses to compare
the frequencies of mutations between the diabetics and
controls. A regression analysis was performed to determine the correlation between the percentage of heteroplasmy and the age at onset of diabetes in the patients
with A3243G.
Results
detection limit for a3243g mutation
The PCR product amplified by the primers MT1 and MT2
was 422 bp in length. After digestion with ApaI, it was
separated into 210 bp or 212 bp if it had a substitution of
A to G at np 3243 to constitute the recognition site for
ApaI. When electrophoresed on a 4% agarose gel, both the
210 bp and 212 bp overlapped and appeared as a single
band, thus leading to an improvement in the sensitivity of
the PCR-RFLP analysis.
After analysis of the calibrators, which consisted of
DNA fragments containing A3243G and no mutation in
the various known ratios, it turned out that PCR-RFLP
could detect the A3243G fragment at a percentage as low
as 1.0% when stained with ethidium bromide and 0.2%
Fig. 2. Detection limit for the A3243G mutation by
PCR-SSCP.
PCR products containing various percentages of A3243G
mutation indicated above each lane were electrophoresed
on a 20% polyacrylamide gel and stained with silver. The
arrow accompanying an asterisk shows the additional band
resulting from the PCR products having A3243G.
1644
Ohkubo et al.: Detecting Mitochondrial Gene Mutations in Japanese
with SYBR Green I (Fig. 1). PCR-SSCP could detect 5.0%
fragments of A3243G (Fig. 2). The detection limit of RFLP
was lower than that of SSCP; however, the detection rates
of the A3243G mutation among patients were the same
between these two methods because all of the A3243G
mutations showed heteroplasmy of ⬎5%.
screening of the mutations
Of 240 diabetic patients examined, five kinds of one-point
mutations including A3243G were detected by PCR-SSCP
(Figs. 3 and 4). They consisted of A3156G, A3243G,
G3357A, C3375A, and T3394C mutations. Only A3243G is
heteroplasmic, whereas the others are homoplasmic mutations.
A3243G, A3156G, and G3357A were found only in
diabetic patients, whereas C3375A and T3394C were also
detected in the controls. No statistically significant differences in the frequencies of mutations were observed
between the diabetic patients and controls (Table 1),
although the frequencies of the A3243G mutation tended
to be higher in the diabetic patients than in the controls
(P ⫽ 0.057).
The A3243G mutation was found in seven patients
(one male and six females). The ages of the patients
diagnosed to have A3243G mutations were 29 – 62 years.
All patients showed hearing loss and a low height. The
mean height of the six female patients was 146.3 ⫾ 2.9 cm,
and one male patient was 155 cm in height, all belonging
to the lower 2.5 percentile of Japanese. Their clinical
profiles are summarized in Table 2. The percentage of
A3243G mutations were widely distributed, from 7.4% to
38.9%. The degree of heteroplasmy for A3243G was
significantly correlated with the age at onset of hyperglycemia (P ⬍0.05; Fig. 5).
An A-to-G mutation at np 3156 is located in the 3⬘
portion of 16S ribosomal RNA. The patient carrying this
mutation (case 1 in Table 3) showed the clinical features of
type 1 diabetes. The same mutation was shared with her
mother and younger sister, and her mother abruptly
developed hyperglycemia with a high titer of anti-glutamic acid decarboxylase (GAD) antibody, fell into an
insulin-dependent state, and was diagnosed as having
type 1 diabetes mellitus 4 years after the detection of the
A3156G mutation. The patient’s younger sister, who is 43
Fig. 3. PCR-RFLP and PCR-SSCP of the A3243G mutation.
PCR products were analyzed by PCR-RFLP (ApaI) and
PCR-SSCP. Controls are represented in lanes 1 (PCR-RFLP),
5, and 6 (PCR-SSCP). The A3243G mutation is indicated in
lanes 2 and 7 (case 1 in Table 2), lanes 3 and 8 (case 2), and
lanes 4 and 9 (case 3).
Fig. 4. PCR-SSCP of mitochondrial gene mutations other than A3243G.
PCR products were electrophoresed as described in Subjects and Methods.
Lanes 1 and 6 are controls. Four kinds of mutations are shown: A3156G in lane
2, G3357A in lane 3, C3375A in lane 4, and T3394C in lane 5.
years of age, currently shows an impaired glucose tolerance.
G3357A, C3375A, and T3394C mutations are located in
the coding region of ND1, the gene that encodes NADH
dehydrogenase subunit 1. G3357A and T3394C are missense mutations (Met17Ileu and Tyr30His, respectively),
and C3375A is a silent mutation. The patient with G3357A
(case 2) had type 1 diabetes and was also positive for the
anti-GAD antibody. The C3375A mutation was detected
in two patients with type 2 diabetes. The T3394C mutation
was found in six patients, all of whom had type 2
diabetes. Four patients (cases 5, 6, 7, and 8) were diagnosed with hyperglycemia between the ages of 51 and 63
years for the first time. They have all maintained relatively good insulin secretion and thus have only undergone diet therapy or drug therapy with hypoglycemic
agents. The remaining two patients (cases 9 and 10)
became diabetic in their youth and have been treated with
insulin. These two patients also showed a relatively high
BMI, and their clinical features are summarized in Table 3.
Discussion
Mitochondrial DNA is predominantly inherited maternally. Less than 0.1% of the mitochondrial DNA is contributed by a sperm. Each cell contains hundreds of
mitochondria and thousands of copies of the mitochondrial DNA genome. Therefore, cells can harbor a mixture
of mutant and wild-type mitochondrial DNAs, a phenom-
1645
Clinical Chemistry 47, No. 9, 2001
Table 1. Prevalence of mitochondrial gene mutations in
diabetics and controls.
Mutation
Diabetic patients, %
Controls, %
P
A3243G
A3156G
G3357A
C3375A
T3394C
2.9 (7/240)
0.4 (1/240)
0.4 (1/240)
0.8 (2/240)
2.5 (6/240)
0 (0/125)
0 (0/125)
0 (0/125)
0.8 (1/125)
3.2 (4/125)
0.057
0.471
0.471
0.974
0.706
enon called heteroplasmy. Over multiple cell divisions,
the proportion drifts toward either predominantly mutant
or wild-type mitochondrial DNAs, leading to homoplasmy. The percentage of mitochondrial DNA with
mutations varies from tissue to tissue, and is observed to
be the highest in the affected tissues (7, 11, 12 ). In the case
of diabetic patients with the A3243G mutation, which was
responsible for a deterioration in insulin secretion, pancreatic ␤ cells showed the highest percentage of heteroplasmy (13 ), and therefore ␤ cells were considered the
most suitable tissue for the examination. However, obtaining ␤ cells by a biopsy of the pancreas is a hazardous
and difficult procedure. The peripheral blood leukocytes
are relatively easy to obtain and, therefore, are used most
frequently to screen for gene mutations, although the
observed percentages of heteroplasmy might be lower
than those in the affected tissue (14, 15 ). In our diagnostic
system, DNA is extracted from peripheral blood leukocytes, and then PCR is performed, followed by SSCP or
RFLP. It takes 4.5–5.0 h to complete all processes, which
makes it possible to diagnose mitochondrial DNA mutations the same day the sample is taken. The lower limit of
detection for A3243G heteroplasmy by PCR-RFLP is 1.0%
with ethidium bromide staining and 0.2% with SYBR
Green I staining, whereas in PCR-SSCP, the lower limit of
detection is 5.0% in our system. By the use of ligationmediated PCR method, which can detect 0.01% heteroplasmy, only diabetic patients, and not controls, have
been reported to show ⬎0.01% heteroplasmy (16 ). As a
result, the possibility of overlooking mutated mitochondrial DNAs if their number is ⬍0.2% cannot be ruled out.
However, ligation-mediated PCR takes a much longer
Fig. 5. Relationship between the degree of heteroplasmy and the age
at onset of diabetes in the patients with A3243G.
The degree of heteroplasmy (percentage) was significantly associated with the
age at onset of diabetes (y ⫽ 35.4 ⫺ 0.5x; r ⫽ ⫺0.63; P ⬍0.05).
time than PCR-RFLP or PCR-SSCP, and thus is not
suitable for routine screening methods.
Among the 240 patients investigated, we identified 7
patients who had the A3243G mutation. The prevalence of
this mutation was 2.9%, and this finding correlates with
reports that the prevalence was within a range of 1–3% in
Japanese diabetic populations (6 – 8, 17 ). In patients with
the A3243G mutation, impaired insulin secretion tends to
become overt between 20 and 30 years of age. Once a
patient becomes diabetic, the amount of secretable insulin
progressively declines. Thus, all cases in whom we detected the A3243G mutation had to receive insulin therapy. Among the six female patients carrying the mutation,
four had experienced an abnormal pregnancy, leading to
either still birth or premature birth. In these cases, it
appears that hyperglycemia might have occurred during
pregnancy for the first time. Among the population of
Japanese women attending a diabetic pregnancy clinic, it
was reported (18 ) that the prevalence of the A3243G
mutation was higher than that observed in randomly
selected Japanese diabetic patients. Moreover, the prevalence of spontaneous abortions in patients with A3243G
was also higher than in those without this mutation.
Table 2. Clinical characterization of the diabetic patients having the A3243G mutation.
Case
Type of
diabetes
mellitus
Therapy
Sex
Age,
years
Age at
onset, years
Height,
cm
BMI,
kg/m2
F-IRI,a
pmol/L
F-CPR,
pmol/L
Heteroplasmy
of A3243G, %
Hearing
loss
Cardiomyopathy
1
2
3
4
5
6
7
Type 2
Type 2
Type 2
Type 1
Type 2
Type 1
Type 2
Insulin
Insulin
Insulin
Insulin
Insulin
Insulin
Insulin
F
F
F
F
F
F
M
62
51
29
42
42
49
45
58
27
23
24
30
30
37
150
147
144
142
148
147
155
14.2
13.9
16.4
16.9
15.5
15.3
14.6
60.0
34.8
12.0
ND
23.4
ND
ND
0.50
0.30
ND
⬍0.10
0.36
0.10
0.33
7.4
15.0
38.9
18.9
18.2
13.6
18.6
⫹b
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫺
ND
⫺
⫹
⫺
⫺
a
b
F-IRI, fasting immunoreactive insulin; F-CPR, fasting reactive C peptide immunoreactivity; ND, not determined.
⫹, positive; ⫺, negative.
Abnormal
pregnancy
⫺
⫹
⫹
⫺
⫹
⫹
1646
Ohkubo et al.: Detecting Mitochondrial Gene Mutations in Japanese
Table 3. Clinical characterization of the diabetic patients with mitochondrial gene mutations other than A3243G.
Case
Mutation
1
2
3
4
5
6
7
8
9
10
A3156G
G3357A
C3375A
a
b
T3394C
Type of
diabetes mellitus
Type
Type
Type
Type
Type
Type
Type
Type
Type
Type
1
1
2
2
2
2
2
2
2
2
Therapy
Sex
Age,
years
Age at
onset, years
BMI,
kg/m2
F-IRI,a
pmol/L
F-CPR,
pmol/L
Family
history
Insulin
Insulin
Diet
Diet
Diet
Diet
Acarbose
Glibenclamide
Insulin
Insulin
F
F
M
F
F
M
F
M
F
F
45
35
72
8
66
65
69
61
51
27
42
16
66
8
63
51
60
59
34
24
21.3
23.9
20.2
14.2
21.44
26.4
20.4
24.3
27.2
28.0
21.6
ND
33.0
ND
48.0
58.8
62.4
34.8
28.2
ND
0.07
0.70
1.03
ND
1.26
0.89
0.93
ND
0.89
0.30
⫹b
⫹
⫹
⫹
⫹
⫺
⫺
⫹
⫹
⫹
F-IRI, fasting immunoreactive insulin; F-CPR, fasting reactive C peptide immunoreactivity; ND, not determined.
⫹, positive; ⫺, negative.
Therefore, the diagnosis of the A3243G mutation before a
hyperglycemic state develops seems to be very important
in female patients not only to prevent hyperglycemia, but
also to prevent abnormal pregnancy. The percentages of
heteroplasmy for the A3243G mutation that were detected
with amplified DNA from peripheral leukocytes were
significantly associated with age at the onset of hyperglycemia (P ⬍0.05; Fig. 5). Although various percentages of
heteroplasmy were observed in the tissues and organs,
heteroplasmy in leukocytes was thought to be associated
with heteroplasmy in the pancreatic ␤ cells. These findings may make it possible to predict the onset of hyperglycemia for the siblings and descendants of the patients
who were unaffected at the time of diagnosis, while also
allowing for a better control of blood glucose and the
prevention of diabetic complications.
Regarding the etiology of impaired glucose tolerance,
the A3243G mitochondrial gene mutation is thought to be
a cause of impaired, glucose-induced insulin secretion.
The clonal cybrid cells between the cells lacking mitochondrial DNA and fibroblasts from a patient with the
A3243G mutation showed poor respiration, an increased
ratio of lactate to pyruvate, and marked defects in the
mitochondrial morphology and respiratory chain complex I and IV activities (19 ). Janssen et al. (20 ) indicated
that cells harboring patient-derived mitochondria with
the A3242G mutation displayed a severe loss of respiration, whereas the rate of mitochondrial translation was
not seriously affected despite the low degree of leucylation of tRNALeu (UUR). Chomyn et al. (21 ) observed that
cell lines that were nearly homoplasmic for this mutation
exhibited a strong (70 –75%) reduction in the degree of
aminoacylated tRNALeu(UUR), as well as a decrease in
mitochondrial protein synthesis. They also revealed a
reduced association of mRNA with ribosomes, possibly
attributable to the presence of a tRNALeu(UUR) aminoacylation defect. The cultured ␤-cell line MIN6 with knockedout mtDNA (␳0 MIN6) showed a loss of mitochondrial
transcription, translation, and respiration activity and a
glucose-stimulated increase in insulin secretion (22 ). Sim-
ilar results were also obtained for the mitochondrial
DNA-deficient ␳0 INS-1 cells established by Kennedy et al.
(23 ).
In addition to the A3243G mutation, we detected four
types of one-point mutations in this region of the mitochondrial gene. These mutations were A3156G, G3357A,
C3375A, and T3394C. A3156G, G3357A, and C3375A were
novel mutations. The np 3156 in the mitochondrial gene is
located in the region close to the 3⬘ end of 16S rRNA,
whereas 3357, 3375, and 3394 are located in the coding
region of NADH dehydrogenase subunit 1 (ND1). The
two mutations, G3357A and T3394C, induce an exchange
of amino acids. In the case of G3357A, methionine is
substituted for isoleucine and, in T3394C, tyrosine for
histidine. It has not yet been determined whether these
mutations cause mitochondrial dysfunctions, which reduce insulin secretion in the affected patients. However,
some groups of the patients with the same mutation
tended to show very similar clinical manifestations, including a sudden onset of hyperglycemia and a rapid
progression to an insulin-dependent state with positive
anti-GAD antibody in the case of the A3156G mutation,
and a mild defect in glucose-stimulated insulin secretion
and the appearance of hyperglycemia with aging or
obesity in the cases of the T3394C mutation. These diabetic
states are suggested to result from the same etiologies
related to the same mutations in the mitochondrial gene.
The A3156G mutation was detected in three subjects in
one family. The proband (case 1 in Table 3) and her
mother became abruptly hyperglycemic and fell into an
insulin-dependent state at 42 and 70 years of age, respectively, thus showing the typical clinical features of type 1
diabetes mellitus. The mother had anti-GAD antibody
with a high titer at onset, whereas the proband was
negative at the time of the investigation, 3 years after her
own onset of the disease. Her younger sister had the same
mutation and showed impaired glucose tolerance at the
time of the investigation when the sister was 42 years of
age. Although we could not complete the pedigree of the
A3156G mutation because the extended family declined
1647
Clinical Chemistry 47, No. 9, 2001
to participate in further studies, the examined subjects
who demonstrated the A3156G mutation might represent
a group of phenotypes with slowly progressive insulindependent diabetes mellitus (24, 25 ). Recent studies have
shown that mitochondria are frequently involved in a
variety of key events of apoptosis, including the release of
such caspase activators as cytochrome c, changes in electron transport, a loss of mitochondrial transmembrane
potential, altered cellular oxidation-reduction, and participation of pro- and antiapoptotic Bcl-2 family proteins
(26, 27 ). The mitochondrial dysfunction, possibly resulting from the translational disorders produced by the
mutated 16S rRNA with A3156G, could be associated
with the apoptosis of ␤ cells in islets, which is considered
one of the causes of type 1 diabetes mellitus (28, 29 ).
Apoptotic features were observed in the muscle fibers of
patients carrying a high percentage of single mtDNA
deletions and tRNA point mutations including A3243G
(30 ), but not in the islet ␤ cells of patients with A3243G
(13, 31 ).
A case of type 1 diabetes with the G3357A mutation
(case 2) showed a high titer of anti-GAD antibody and
was complicated with Down syndrome. Her mother
showed impaired glucose tolerance by a 75-g oral glucose
tolerance test.
The C3375A mutation was detected in two patients
with type 2 diabetes. They showed a slight increase of
blood glucose, and diet therapy has been introduced, thus
leading to good control of the glucose concentration. The
proband, an 8-year-old girl with the C3375A mutation
(case 4) had an early onset of type 2 diabetes, whereas her
mother showed normal glucose tolerance with a low
insulin response after the 75-g oral glucose tolerance test.
Her elder brother and sister also showed impaired glucose tolerance in their early teens.
The T3394C mutation has previously been reported to
be implicated in Leber hereditary optic neuropathy
(32, 33 ) and also in patients with type 2 (non-insulindependent) diabetes mellitus in the United Kingdom (34 )
and in Japan (35 ). All patients with the T3394C mutation
who were detected by our screening process also showed
the clinical features of type 2 diabetes. Among the four
cases whose onset of disease started at more than 50 years
of age, three showed a normal BMI and a mild disturbance in glucose tolerance and were treated with either
dietary treatment or with acarbose or glibenclamide (cases 5, 7, and 8). On the other hand, patients who demonstrated diabetes before 40 years of age (cases 9 and 10) and
had BMIs of 27.2 and 28.0, respectively, were placed on
insulin therapy. In the case of this T3394C mutation, the
disorder of insulin secretion might become overt when
combined with such factors as aging or obesity. The
prevalence of this mutation was 2.5% among diabetic
patients and 3.2% among the controls according to our
screening, whereas the prevalence was 4.9% among diabetic patients and 1.3% among the controls in another
study on Japanese patients with non-insulin-dependent
diabetes mellitus (35 ). Because the controls in our study
were younger than the patients and were not agematched to the patient group, there is a possibility that the
control group might have contained some individuals
who might become diabetic in middle age by aging or
obesity. In addition, it is also possible that this T3394C
mutation is a modifier locus for one or many other genes.
As a result of this screening method, we found both
novel and known mutations of the mitochondrial gene
and determined the potential clinical subtypes based on
the site of the mitochondrial gene mutations. Our findings
regarding mitochondrial gene mutations should provide
valuable information related to the preclinical diagnosis
of diabetes mellitus, the treatment and follow-up of such
diabetic patients, and the identification of their relatives
with maternally transmitted mitochondrial gene mutations.
This work was supported in part by funds (No. 986006)
from the Central Research Institute of Fukuoka University.
References
1. Gerbitz K-D, Gempel K, Brdiczka D. Mitochondria and diabetes.
Genetic, biochemical, and clinical implications of the cellular
energy circuit. Diabetes 1996;45:113–26.
2. van den Ouweland JM, Lemkes HHJ, Ruitenbeek W, Snadkuijl LA,
de Vijlder MF, Struyvenberg PA, et al. Mutation in mitochondrial
tRNALeu(UUR) gene in a large pedigree with maternally transmitted
type II diabetes mellitus and deafness. Nat Genet 1992;1:368 –
71.
3. Suzuki Y, Hosokawa K, Suzuki S, Shimada A, Hinokio Y, Asahina
T, et al. Diabetes associated with a novel 3264 mitochondrial
tRNALeu(UUR) mutation. Diabetes Care 1997;20:1138 – 40.
4. Tsukuda K, Suzuki Y, Kameoka K, Osawa N, Goto Y, Katagiri H, et
al. Screening of patients with maternally transmitted diabetes for
mitochondrial gene mutations in the tRNALeu(UUR) region. Diabet
Med 1997;14:1032–7.
5. Nakano S, Fukuda M, Hotta F, Ito T, Ishii T, Kitazawa M, et al.
Mitochondrial DNA point mutation at nucleotide pair 3316 in a
Japanese family with heterogeneous phenotypes of diabetes.
Endocr J 1998;45:625–30.
6. Katagiri H, Asano T, Ishihara H, Inukai K, Anai M, Yamanouchi T,
et al. Mitochondrial diabetes mellitus: prevalence and clinical
characterization of diabetes due to mitochondrial tRNALeu(UUR)
gene mutation in Japanese patients. Diabetologia 1994;37:504 –
10.
7. Tokunaga M, Mita S, Sakuta R, Nonaka I, Araki S. Increased
mitochondrial DNA in blood vessels and ragged-red fibers in
mitochondrial myopathy, enchephalopathy, lactic acidosis, and
stroke-like episodes (MELAS). Ann Neurol 1993;33:275– 80.
8. Kadowaki T, Kadowaki H, Mori Y, Tobe K, Sakuta R, Suzuki Y, et
al. A subtype of diabetes mellitus associated with a mutation of
mitochondrial DNA. N Engl J Med 1994;330:962– 8.
9. Alberti KGMM, Zimmet PZ, for the WHO Consultation. Definition,
diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus
provisional report of a WHO consultation. Diabet Med 1998;15:
539 –53.
10. Dockhorn-Dworniczak B, Dworniczak B, Brommelkamp L, Bulles J,
1648
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Ohkubo et al.: Detecting Mitochondrial Gene Mutations in Japanese
Horst J, Bocker WW. Non-isotopic detection of single strand
conformation polymorphism (PCR-SSCP): a rapid and sensitive
technique in diagnosis of phenylketonuria. Nucleic Acids Res
1991;19:2500.
Moraes CT, Ciacci F, Bonilla E, Jansen C, Hirano M, Rao N, et al.
Two novel pathogenic mitochondrial DNA mutations affecting
organelle number and protein synthesis. J Clin Invest 1993;92:
2906 –15.
Smith ML, Hua XY, Marsden DL, Liu D, Kennaway NG, Ngo KY, et
al. Diabetes and mitochondrial encephalomyophathy with lactic
acidosis and stroke-like episodes (MELAS): radiolabeled polymerase chain reaction is necessary for accurate detection of low
percentages of mutation. J Clin Endocrinol Metab 1997;82:
2826 –31.
Kobayashi T, Nakanishi K, Nakase H, Kajio H, Okubo M, Murase
T, et al. In situ characterization of islets in diabetes with a
mitochondrial DNA mutation at nucleotide position 3243. Diabetes 1997;46:1567–71.
Chinnery PF, Zwijnenburg PJ, Walker M, Howell N, Taylaor RW,
Lightowlers RN, et al. Nonrandom tissue distribution of mutant
mtDNA. Am J Med Genet 1999;85:498 –501.
Suzuki Y, Goto Y, Taniyama M, Nonaka I, Murakami N, Hosokawa
K, et al. Muscle histopathology in diabetes mellitus associated
with mitochondrial tRNALeu(UUR) mutation at position 3243. J Neurol Sci 1997;145:49 –53.
Urata M, Wakiyama M, Iwase M, Yoneda M, Kinosita S, Hamasaki
N, et al. New sensitive method for the detection of the A3243G
mutation of human mitochondrial deoxyribonucleic acid in diabetes mellitus patients by ligation-mediated polymerase chain reaction. Clin Chem 1998;44:2088 –93.
Fukui M, Nakano K, Obayashi H, Kitagawa Y, Nakamura N, Mori H,
et al. High prevalence of mitochondrial diabetes mellitus in
Japanese patients with major risk factors. Metabolism 1997;46:
793–5.
Yanagisawa K, Uchigata Y, Sanaka M, Sakura H, Minei S, Shimizu
M, et al. Mutation in the mitochondrial tRNALeu at position 3243
and spontaneous abortions in Japanese women attending a clinic
for diabetic pregnancies. Diabetologia 1995;38:809 –15.
van den Ouweland JM, Maechler P, Wollheim CB, Attardi G,
Maassen JA. Functional and morphological abnormalities of mitochondria harbouring the tRNALeu(UUR) mutation in mitochondria
derived from patients with maternally inherited diabetes and
deafness (MIDD) and progressive kidney disease. Diabetologia
1999;42:485–92.
Janssen GMC, Maassen JA, van den Ouweland JM. The diabetesassociated 3243 mutation in the mitochondrial tRNALeu(UUR) gene
causes severe mitochondrial dysfunction without a strong decrease in protein synthesis rate. J Biol Chem 1999;274:
29744 – 8.
Chomyn A, Enriquez JA, Micol V, Fernandez-Silva P, Attardi G. The
MELAS syndrome-associated human mitochondrial tRNALeu(UUR)
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
mutation causes aminoacylation deficiency and concomitant reduced association of mRNA with ribosomes. J Biol Chem 2000;
275:19198 –209.
Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, Oka Y, et al.
Mitochondrial DNA is required for regulation of glucose-stimulated
insulin secretion in a mouse pancreatic cell line, MIN6. J Biol
Chem 1996;271:26194 –9.
Kennedy ED, Maechler P, Wollheim CB. Effects of depletion of
mitochondrial DNA on metabolism-secretion coupling in INS-1
cells. Diabetes 1998;47:374 – 80.
Kobayashi T, Itoh T, Kosaka K, Tsuji K. Time course of islet cell
antibodies and ␤-cell function in non-insulin-dependent stage of
type I diabetes. Diabetes 1987;36:510 –7.
Kobayashi T. Subtype of insulin-dependent diabetes mellitus
(IDDM) in Japan: slowly progressive IDDM–the clinical characteristics and pathogenesis of the syndrome. Diabetes Res Clin Pract
1994;24(Suppl):S95–9.
Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;
281:1309 –12.
Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers
GM, et al. Molecular characterization of mitochondrial apoptosisinducing factor. Nature 1999;397:441– 6.
Loweth AC, Williams GT, James RF, Scarpello JH, Morgan NG.
Human islets of Langerhans express Fas ligand and undergo
apoptosis in response to interleukin-1␤ and Fas ligation. Diabetes
1998;47:727–32.
Mauricio D, Mandrup-Poulsen T. Apoptosis and the pathogenesis
of IDDM. Diabetes 1998;47:1537– 43.
Mirabella M, Giovanni SD, Silvestri G, Tonali P, Servidei S.
Apoptosis in mitochondrial encephalomyopathies with mitochondrial DNA mutations: a potential pathogenic mechanism. Brain
2000;123:93–104.
Otabe S, Yasuda K, Mori Y, Shimokawa K, Kadowaki H, Jimi A, et
al. Molecular and histological evaluation of pancreata from patients with mitochondrial gene mutation associated with impaired
insulin secretion. Biochem Biophys Res Commun 1999;259:
149 –56.
Johns DR, Neufeld MJ, Park RD. An ND-6 mitochondrial DNA
mutation associated with Leber hereditary optic neuropathy. Biochem Biophys Res Commun 1992;187:1551–7.
Brown MD, Voljavec AS, Lott MT, MacDonald I, Wallace DC.
Leber’s hereditary optic neuropathy: a model for mitochondrial
neurogenerative disease. FASEB J 1992;6:2791–9.
Thomas AW, Edwards A, Sherratt EJ, Majid A, Gagg J, Alcolado JC.
Molecular scanning of candidate mitochondrial tRNA genes in
type2 (non-insulin dependent) diabetes mellitus. J Med Genet
1996;33:253–5.
Hirai M, Suzuki S, Onoda M, Hinokio Y, Ai L, Hirai A, et al.
Mitochondrial DNA 3394 mutation in the NADH dehydrogenase
subunit 1 associated with non-insulin-dependent diabetes mellitus. Biochem Biophys Res Commun 1996;219:951–5.