Am J Physiol Endocrinol Metab
280: E1007–E1014, 2001.
Depletion of mitochondrial DNA alters glucose
metabolism in SK-Hep1 cells
KYU-SANG PARK,1 KYUNG-JAY NAM,1 JUN-WOO KIM,1 YOUN-BOK LEE,1
CHANG-YEOP HAN,1 JUNE-KEY JEONG,2 HONG-KYU LEE,1,3 AND YOUNGMI KIM PAK1
1
Division of Metabolic Disease, Department of Biomedical Sciences, National Institute of Health,
Seoul 122-701; and Departments of 2Nuclear Medicine and 3Internal Medicine,
School of Medicine, Seoul National University, Seoul 110-744, Korea
Received 28 June 2000; accepted in final form 5 February 2001
(DM) is a genetically heterogeneous
disorder but exhibits common features of glucose intolerance and hyperglycemia (25). It is generally considered that various genetic factors, although not all are
known, cause DM. Maternally inherited mitochondrial
DNA (mtDNA), which does not follow the classical
Mendelian genetics, has been considered to be one of
the genetic factors for developing diabetes (14).
Approximately 0.5–1.5% of all diabetic patients exhibit pathogenic mtDNA mutations such as duplications (2), point mutations (18), and large-scale deletions (3). Also, diabetics show not only mutations but
also quantitative changes of mtDNA. Antonetti et al.
(1) reported that mtDNA copy number was ⬃50% decreased in skeletal muscles of both type 1 and type 2
diabetic patients as estimated by Southern blot analy-
ses. Lee et al. (22) also reported that the quantitative
decrease of mtDNA in lymphocytes preceded the type 2
diabetic development, suggesting that the decreased
content of mtDNA might be a causal factor for type 2
diabetes. However, there is as yet no convincing evidence whether the reduction of mtDNA copy number
causes enough disturbance in the glucose metabolism
in peripheral tissues such as liver cells to participate in
the development of diabetes.
In pancreatic ␤-cells, glucose-induced ATP production stimulates insulin secretion via mitochondrial oxidative phosphorylation. The depletion of mtDNA impairs this glucose-stimulated insulin secretion and
causes glucose intolerance (16, 31). The impairment of
glucose uptake via glucose transporter in peripheral
tissues may also contribute to the diabetic pathogenesis, especially of type 2 (5). For example, population
studies have suggested that genetic variation in the
GLUT-1 gene, an isoform of glucose transporter, is
associated with an increased risk for developing type 2
diabetes (23). To our knowledge, however, there has
been no study showing that the alteration of mtDNA
copy number affects glucose utilization and metabolism in peripheral tissues.
In this study, we established mtDNA-depleted (␳0)
human hepatoma SK-Hep1 cells and investigated the
effects of mtDNA depletion on glucose metabolism to
test whether the decrease in mtDNA could participate
in the diabetic pathogenesis. The ␳0 cells, established
by long-term treatment with ethidium bromide (EtBr),
have been important tools in investigating the function
of mtDNA (21). EtBr intercalates into circular DNA
and inhibits mtDNA replication and transcription at a
low concentration (0.1–2 ␮g/ml) without any detectable effect on nuclear DNA division (16, 35). We succeeded in isolating the ␳0 human hepatoma cell line,
which lost the ability of oxidative phosphorylation. We
observed that cellular ATP production and glucose
uptake were attenuated in ␳0 cells. Glucose transporter
expression and glucose metabolizing enzyme activities
in ␳0 cells were also lower than in control cells. These
Address for reprint requests and other correspondence: Y. K. Pak,
Div. of Metabolic Disease, Dept. of Biomedical Sciences, National
Institute of Health, 5 Nokbun-Dong, Eunpyung-Ku, Seoul, Korea
122-701 (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
oxidative phosphorylation; glucose uptake
DIABETES MELLITUS
http://www.ajpendo.org
0193-1849/01 $5.00 Copyright © 2001 the American Physiological Society
E1007
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
Park, Kyu-Sang, Kyung-Jay Nam, Jun-Woo Kim,
Youn-Bok Lee, Chang-Yeop Han, June-Key Jeong,
Hong-Kyu Lee, and Youngmi Kim Pak. Depletion of mitochondrial DNA alters glucose metabolism in SK-Hep1 cells.
Am J Physiol Endocrinol Metab 280: E1007–E1014, 2001.—
Maternally inherited mitochondrial DNA (mtDNA) has been
suggested to be a genetic factor for diabetes. Reports have
shown a decrease of mtDNA content in tissues of diabetic
patients. We investigated the effects of mtDNA depletion on
glucose metabolism by use of ␳0 SK-Hep1 human hepatoma
cells, whose mtDNA was depleted by long-term exposure to
ethidium bromide. The ␳0 cells failed to hyperpolarize mitochondrial membrane potential in response to glucose stimulation. Intracellular ATP content, glucose-stimulated ATP
production, glucose uptake, steady-state mRNA and protein
levels of glucose transporters, and cellular activities of glucose-metabolizing enzymes were decreased in ␳0 cells compared with parental ␳⫹ cells. Our results suggest that the
quantitative reduction of mtDNA may suppress the expression of nuclear DNA-encoded glucose transporters and enzymes of glucose metabolism. Thus this may lead to diabetic
status, such as decreased ATP production and glucose utilization.
E1008
GLUCOSE METABOLISM IN ␳0 CELLS
results suggest that the decrease in mtDNA copy number may suppress glucose uptake and metabolism,
which may lead to the development of glucose intolerance and diabetes.
MATERIALS AND METHODS
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
Cells and cell culture. Cells from the human hepatoma cell
line SK-Hep1 (American Type Culture Collection, HTB-52,
Rockville, MD) were grown in DMEM containing 1 mM
pyruvate and supplemented with 10% fetal bovine serum
(FBS), penicillin (100 IU/ml), and streptomycin (100 ␮g/ml).
They were incubated at ambient oxygen concentrations in
the presence of 5% CO2 at 37°C. The mtDNA-depleted (␳0)
cell line was isolated by treating SK-Hep1 cells with 0.1
␮g/ml EtBr and 50 ␮g/ml uridine for ⬎10 wk in DMEM
supplemented with 10% FBS. The control parental SK-Hep1
cells were maintained for the same time period in normal
culture medium.
PCR and RT-PCR analyses. To compare the mtDNA content between control and ␳0 SK-Hep1 cells, total cellular
DNA was extracted using QIAmp Tissue Kit (Qiagen, Hilden,
Germany). The PCR amplification of human mtDNA was
performed in a volume of 50 ␮l. This contained 0.25 units of
Taq polymerase (Perkin-Elmer, Norwalk, CT) and 10 pmol of
oligonucleotide primer pair, which was sequence specific for
human mtDNA: upstream 5⬘-CCT AGG GAT AAC AGC GCA
AT-3⬘ and downstream 5⬘-TAG AAG AGC GAT GGT GAG
AG-3⬘ (630-bp product). The PCR conditions for mtDNA included an initial denaturing for 5 min and then 30 cycles as
follows: denaturing for 30 s at 94°C, annealing for 30 s at
60°C, and extending for 40 s at 72°C using GeneAmp (PerkinElmer). Aliquots of the PCR reactions were analyzed by 1.1%
agarose gel electrophoresis and were examined for the sizepredicted products by EtBr staining.
For quantification of mRNA expression, RT-PCR was performed. Total RNA from the cells was prepared using a
modified guanidium thiocyanate-phenol-chloroform extraction method. Random hexamer-primed cDNA synthesis was
performed in a final volume of 25 ␮l containing 20 units of
RNase inhibitor and 200 units of murine leukemia virus
reverse transcriptase (Promega, Madison, WI) at 37°C for 60
min. Then, PCR amplification of human glucose transporter
genes was performed with Taq polymerase and oligonucleotide primers. Sequence-specific primers for cytochrome c oxidase-1 (COX-1) and glucose transporters were as follows:
COX-1, forward 5⬘-ACA CGA GCA TAT TTC ACC TCC G-3⬘
and reverse 5⬘-GGA TTT TGG CGT AGG TTT GGT C-3⬘
(336-bp product); GLUT-1, forward 5⬘-CCG CAT CAT CTG
CCC ACT-3⬘ and reverse 5⬘-CTG GAT GAC CGA AAA GCT
A-3⬘ (352-bp product); GLUT-2, forward 5⬘-CAA TGA ACC
CAA AAC CAA CC-3⬘ and reverse 5⬘-CAG CTG ATG AAA
AGT GCC AA-3⬘ (383-bp product); GLUT-3, forward 5⬘-ATC
ACA GTT GCT ACA ATC GG-3⬘ and reverse 5⬘-AAT TCC
AAC AAC GAT GCC CA-3⬘ (439-bp product); GLUT-4, forward 5⬘-ACC AAC TGG CCA TTG TTA TC-3⬘ and reverse
5⬘-CGA AGA TGC TGG TCG AAT AA-3⬘ (415-bp product).
Amplification of the human ␤-actin gene was performed using commercial primers (Clontech, Palo Alto, CA) to enable
semiquantitative normalization. In preliminary experiments, we demonstrated that the product amplification by
PCR was linear at the number of cycles. As described, cycle
numbers corresponding to the exponential phase of the reaction were determined to be 30 cycles for ␤-actin, COX-1, and
GLUT-1, 40 cycles for GLUT-2 at 50°C annealing temperature, and 35 cycles at 58°C annealing temperature for
GLUT-3 and GLUT-4. Aliquots of PCR reactions were loaded
onto a 1.1% agarose gel containing EtBr, electrophoresed,
visualized, and quantified by densitometry using Bioprint
(Vilber Lourmat, France).
Staining of mitochondria. To study mitochondrial structure and distribution, the control and ␳0 SK-Hep1 cells were
incubated with MitoTracker Red (Molecular Probes, Eugene,
OR) for 15–45 min at 37°C and then washed three times with
prewarmed phosphate-buffered saline (PBS). Cells were
fixed for 10 min in 3.7% formaldehyde in PBS at pH 7.4,
washed again with PBS, and then incubated with Sytox
Green (Molecular Probes) for staining the nucleus. The cell
preparations were visualized and photographed in a fluorescence microscope (Carl Zeiss, Jena, Germany).
Measurement of mitochondrial membrane potential. The
mitochondrial membrane potential (⌬⌿m) was measured using the fluorescent probe rhodamine 123 (Sigma Chemical,
St. Louis, MO) (20). Because rhodamine 123 is a cationic dye,
it accumulates in the mitochondria driven by ⌬⌿m. Under
appropriate loading conditions, the concentration of rhodamine 123 within the mitochondria reaches sufficiently high
levels that it quenches its own fluorescence. If the mitochondria depolarize, rhodamine 123 leaks out into the cytoplasm
and is associated with a reduction in the amount of quenching. Thus the changes in ⌬⌿m are revealed as changes in
total fluorescence intensity (8).
The control and ␳0 SK-Hep1 cells were detached from
culture dishes with trypsin and washed twice with KrebsRinger Henseleit (KRH) buffer (containing in mM: 125 NaCl,
1.2 KH2PO4, 1.2 MgSO4, 6 glucose, 25 HEPES, 2 CaCl2, pH
7.4) and supplemented with 0.2% BSA. Cell suspensions (2 ⫻
105 cells/ml) were incubated in KRH buffer with 10 ␮g/ml
rhodamine 123 at 37°C for 10 min and washed with glucosefree KRH buffer. The measurements of ⌬⌿m were carried out
in a thermostatically regulated and magnetically stirred fluorimeter cuvette (SFM-25, Bio-Tek Kontron Instruments,
Milan, Italy) at a concentration of 2 ⫻ 105 cells/ml. The
changes in fluorescence intensities were measured at an
excitation wavelength of 490 nm and an emission wavelength
of 530 nm.
ATP assay. The cellular contents of ATP were measured
using the luciferin-luciferase reaction with an ATP bioluminescence assay kit (Sigma). The harvested control and ␳0
SK-Hep1 cells were suspended with KRH buffer containing
0.1 mM glucose and 0.2% BSA. These cell suspensions were
incubated in a 37°C shaking water bath. After addition of an
assay mixture containing luciferin and luciferase, luminescence was measured immediately in a bioluminometer
equipped with an injector (Lumat LB 9501, Berthold, Germany). The amounts of ATP were determined by running an
internal standard, expressed as moles per milligram of protein by normalizing to the amount of cellular protein. A
Bradford assay was used to determine the protein content.
Glucose uptake using 2-fluoro-2-deoxy- D -[ 3 H]glucose.
2-Fluoro-2-deoxy-D-[5,6-3H]glucose ([3H]FDG; specific activity 30 Ci/mM or 1.1 TBq/mM) was used for the uptake studies
(12). Human hepatoma SK-Hep1 cells or the established ␳0
cells (5 ⫻ 105 cells/well) were cultured on a 6-well plate in
DMEM until 95% confluent. The medium was changed to 1
ml of glucose-free Hanks’ balanced salt solution (HBSS), and
1 ␮Ci (37 kBq) of [3H]FDG was added and then incubated at
37°C for 1 h. Addition of ice-cold HBSS stopped the incorporation of [3H]FDG. The cells were washed three times with
HBSS and dissolved in 0.5 ml of 0.3 N NaOH with 10% SDS.
Aliquots of cell lysates were mixed in 10 ml of scintillation
fluid (Hionic Fluor, Packard Instruments, Meriden, CT), and
bound 3H activities were measured using a liquid scintillation counter (Packard Instruments). Calibration standards
GLUCOSE METABOLISM IN ␳0 CELLS
brane. The membrane was incubated with goat polyclonal
antibodies against human GLUT-1, -3, or -4 (1:1,000; Santa
Cruz, Santa Cruz, CA) and with anti-goat IgG conjugated
with horseradish peroxidase (1:1,000; Santa Cruz) after
three washings in PBS. Bound antibodies were visualized
with an enhanced chemiluminescence system (Amersham,
Buckinghamshire, England). The antibody against ␤-actin
was utilized to confirm the equal loading of the sample.
Statistical analysis. Data shown are expressed as means ⫾
SE. Statistical significance was evaluated by paired or unpaired Student’s t-test, and P ⬍ 0.05 was considered significant.
RESULTS
Establishment of ␳0 SK-Hep1 cells. To establish the
mtDNA-depleted cells, the SK-Hep1 human hepatoma
cells were treated with the medium containing 0.1
␮g/ml EtBr and 50 ␮g/ml uridine for 10 wk, and the
mtDNA contents of the EtBr-treated cells were examined by PCR. Figure 1A shows that mtDNA was not
amplified from genomic DNAs of EtBr-treated cells,
Fig. 1. Mitochondrial DNA (mtDNA) content and fluorescence microscopic characterization of mitochondria
in the human hepatoma SK-Hep1 and mtDNA-depleted
(␳0) SK-Hep1 cells. A: genomic DNAs were isolated from
both cells, and then mtDNA and ␤-actin nuclear DNA
were amplified by PCR. The PCR products were visualized on agarose gel after ethidium bromide (EtBr)
staining. B: phase micrographs of the SK-Hep1 (a) and
␳0 SK-Hep1 cells (c) are shown. SK-Hep1 (b) and ␳0
SK-Hep1 (d) cells were stained with MitoTracker Red, a
mitochondrial membrane potential-dependent dye, and
Sytox Green, a nucleic acid-staining dye. Original magnification is ⫻1,000.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
were also used. The tracer uptake was expressed as counts
per minute per milligram of cellular protein.
Determination of enzyme activity. COX (EC 1.9.3.1) activity was determined according to the method described by
Madden and Storrie (24). Briefly, the enzyme reaction was
initiated by the addition of cell lysate to the reaction mixture,
and the absorbance change at 550 nm was recorded in a spectrophotometer for a period of 1 min (Spectramax 250, Molecular
Devices, Sunnyvale, CA). Succinate dehydrogenase (SDH; EC
1.3.99.1) activity was assayed by determining the absorbance
change at 600 nm (28). Lactate dehydrogenase (LDH; EC
1.1.27), glyceraldehyde-3-phosphate dehydrogenase (GAPDH;
EC 1.2.1.12), glucose-6-phosphate dehydrogenase (G6PDH; EC
1.1.1.49), and hexokinase (EC 2.7.1.1) were assayed as described (10, 17, 26, 33). Specific activities of all enzymes were
expressed as milliunits per milligram of protein, where 1 mU
was defined as the amount of enzyme that catalyzes the formation of 1 nmol product/min under standard assay conditions.
Western blot analysis. Western blot analysis was performed to estimate expression level of glucose transporter in
␳0 cells. Total cell lysates (50 ␮g protein) were subjected to
10% SDS-PAGE and transferred onto a nitrocellulose mem-
E1009
E1010
GLUCOSE METABOLISM IN ␳0 CELLS
Fig. 2. Changes of mitochondrial membrane potential (⌬⌿m). Cell
suspensions were loaded with rhodamine 123 (10 ␮g/ml for 10 min)
in Krebs-Ringer Henseleit (KRH) buffer. A: representative traces
illustrating that ⌬⌿m changes were evoked by the addition (arrows)
of glucose (10 mM) and carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone (FCCP; 1 ␮M) in SK-Hep1 (n ⫽ 10) and its ␳0 cells (n ⫽
6). B: percent changes of fluorescence by glucose and FCCP are
expressed as means ⫾ SE; **P ⬍ 0.01.
produced 0.44 ⫾ 0.09 nmol ATP/mg protein in the
control SK-Hep1 cells, whereas only 0.04 ⫾ 0.07 nmol
ATP/mg protein was produced in ␳0 cells (Fig. 4).
Reduced glucose uptake and glucose transporter expression in ␳0 cells. To investigate whether mtDNA
depletion impairs glucose utilization, glucose uptake
was assessed using [3H]FDG, a radiolabeled, nondegradable glucose (12). The resting and insulin-stimulated [3H]FDG uptakes in ␳0 cells were decreased by
⬃30% compared with the control cells (Fig. 5).
It is possible that the decrease in glucose uptake may
result from an alteration of glucose transporter expressions. The steady-state levels of mRNA and proteins of
the glucose transporter isoforms, GLUT-1, -2, -3, and
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
which are different from the control cells. On the other
hand, the ␤-actin gene, a nuclear DNA control, was
amplified in both control and EtBr-treated cells. This
observation demonstrated that the EtBr-treated SKHep1 cells were entirely devoid of mtDNA (␳0 cells). We
stained the ␳0 or control SK-Hep1 cells with MitoTracker, a mitochondrial membrane potential-dependent fluorescent dye, to visualize the mitochondrial
structure. As shown in Fig. 1B, ␳0 cells lost the reticulum structure of mitochondria in cytoplasm compared
with the control SK-Hep1 cells, suggesting that they
might lose a functional structure of mitochondria.
Loss of ⌬⌿m in ␳0 cells. To confirm whether ␳0 cells
lost oxidative phosphorylation activity, we measured
⌬⌿m using rhodamine 123. In control SK-Hep1 cells,
an application of 10 mM glucose hyperpolarized ⌬⌿m,
and the subsequent treatment with carbonyl cyanide
p-(trifluoromethoxy)phenylhydrazone (FCCP; 1 ␮M), a
mitochondrial uncoupler, rapidly depolarized ⌬⌿m
(n ⫽ 10). However, the ␳0 cells did not hyperpolarize
⌬⌿m upon glucose application, indicating that the ␳0
cells may be unable to utilize glucose metabolites as
substrates for ATP production through the tricarboxylic acid cycle. Furthermore, the ␳0 cells failed to change
⌬⌿m when treated with FCCP alone, which reflected
the amplitude of resting ⌬⌿m (n ⫽ 6; Fig. 2A). Figure
2B shows the quantitative changes of ⌬⌿m in the
control or ␳0 cells by either glucose or FCCP. This
result suggests that the ␳0 cells lack the basal proton
gradient across the mitochondrial membrane as well as
the glucose-stimulated gradient generation.
Decreased COX activity in ␳0 cells. The three subunits of COX complex (COX-1, -2, and -3) in the mitochondrial electron transfer system are encoded by
mtDNA, whereas the other 10 subunits are encoded by
nuclear DNA. We determined the steady-state mRNA
levels of COX-1 in the control and ␳0 cells by RT-PCR.
As expected, the COX-1 mRNA was nearly absent,
whereas ␤-actin was normally expressed in ␳0 cells
(Fig. 3). The total cellular enzyme activity of COX was
found to be significantly decreased in ␳0 cells compared
with that in the control cells (2.4 ⫾ 0.6 vs. 1.0 ⫾ 0.7
mU/mg protein; n ⫽ 4, P ⬍ 0.01). In contrast, the
specific activity of SDH, of which all subunits are
nuclear DNA encoded, were slightly increased by
mtDNA depletion (353 ⫾ 15 vs. 400 ⫾ 28 mU/mg
protein; n ⫽ 5, P ⬍ 0.05).
Attenuated ATP production in ␳0 cells. There are
several reports showing that the adipocytes and muscle cells initially decrease the intracellular ATP content by inhibiting oxidative phosphorylation but subsequently recovering it (4). Results from mtDNA
depletion were different from this. In the case of ␳0
SK-Hep1 cells, the resting intracellular ATP contents
were decreased 80% compared with the control cells
[0.35 ⫾ 0.03 (n ⫽ 13) vs. 1.84 ⫾ 0.09 nmol/mg protein
(n ⫽ 11), P ⬍ 0.01]. Furthermore, the glucose-stimulated ATP production was also largely attenuated in ␳0
cells. The application of glucose (10 mM) for 10 min
GLUCOSE METABOLISM IN ␳0 CELLS
-4, were determined by RT-PCR and Western blot. The
levels of GLUT-1, GLUT-3, and GLUT-4 mRNA in ␳0
cells were reduced by 10.2 ⫾ 4.2, 35.5 ⫾ 8.9, and 57.2 ⫾
8.4%, respectively, compared with those in control cells
(Fig. 6). However, the mRNA of GLUT-2 was not
present in either control or ␳0 SK-Hep1 cells (data not
shown). To a similar degree, the protein levels of
GLUT-1, GLUT-3, and GLUT-4 were attenuated in ␳0
cells (Fig. 7). These data demonstrate that the mtDNA
depletion downregulated the expression of glucose
transporters, which were nuclear encoded, resulting in
a reduction of glucose utilization.
Activities of glucose-metabolizing enzymes. The effects of mtDNA depletion on glycolysis and pentose
phosphate shunt were determined by cellular activities
of GAPDH and G6PDH. The specific activities of
GAPDH and G6PDH in ␳0 cells were reduced to 75 and
Fig. 4. Intracellular ATP contents at basal state and after glucose
stimulation (10 mM; arrows) were compared between SK-Hep1 (F,
n ⫽ 13) and its ␳0 cells (E, n ⫽ 11), which were measured by
bioluminometer using luciferin-luciferase reaction. ATP contents
were normalized to cellular protein and expressed as nanomoles per
milligram of cellular protein. Data are expressed as means ⫾ SE.
Fig. 5. Basal and insulin-stimulated glucose uptake. Both SK-Hep1
and its ␳0 cells (n ⫽ 11) were incubated with 2-fluoro-2-deoxy-D[3H]glucose ([3H]FDG), and then cellular 3H activities were determined by a liquid scintillation counter. Data are expressed as
means ⫾ SE; **P ⬍ 0.01.
30% of the control cells, respectively (Table 1). Interestingly, the activity of hexokinase in ␳0 cells was only
10% of the control cells. The activity of LDH, an indicator for the intracellular NADH accumulation due to
dysfunction of the respiratory chain, was 53% enhanced in ␳0 cells. The increased LDH activity in ␳0
cells implies that the minimal energy production for
cell survival might be achieved via anaerobic metabolism.
DISCUSSION
Mitochondria are intracellular organelles that are
the site of oxidative phosphorylation and the primary
source of cellular energy (34). It is known that mutations, deletions, and insertions of mtDNA result in
mitochondrial dysfunction associated with the pathogenesis of various mitochondrial diseases, including
DM (3). There are also several lines of evidence to
support the idea that the alterations of mtDNA quantity may cause mitochondrial diseases. First, depletions of mtDNA in muscle, liver, and kidney were
reported in fatal mitochondrial encephalopathies (27).
Second, streptozotocin, one of the diabetogenic agents,
reduced the levels of mtDNA content and its transcripts in pancreatic islets (11). Third, ␤-cells of adult
Goto-Kakizaki rats, a genetic model of defective insulin
secretion and hyperglycemia, had a significantly
smaller mitochondrial volume compared with the
␤-cells of control rats (29). Fourth, mtDNA depletion
inhibited glucose-stimulated increases of the intracellular free Ca2⫹ concentration and insulin secretion in
mouse insulinoma cells, which led to glucose intolerance and hyperglycemia (31). Finally, in our laboratory, the decrease of mtDNA content preceded the
development of diabetes (22), suggesting that the
quantitative reduction of mtDNA be considered as one
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
Fig. 3. Steady-state mRNA levels of cytochrome c oxidase (COX)-1.
Total RNAs were isolated from SK-Hep1 and ␳0 SK-Hep1 cells, and
the mRNA levels of COX-1 and ␤-actin were determined by RT-PCR,
as described in MATERIALS AND METHODS. The products were visualized on agarose gel after EtBr staining.
E1011
E1012
GLUCOSE METABOLISM IN ␳0 CELLS
Fig. 7. Protein expression levels of glucose transporters. Total cell
lysates (50 ␮g protein) were subjected to 10% SDS-PAGE. Western blot
analysis was performed using goat polyclonal antibody against human
glucose transporters (GLUT-1, -3, and -4). Bound antibodies were visualized with enhanced chemiluminescence. The antibody against human
␤-actin was utilized to confirm the equal loading of the samples.
of the possible causative factors, not a consequence, of
diabetes.
In this study, we established the mtDNA-depleted ␳0
cells using SK-Hep1, a rapidly proliferating human
hepatoma cell, to study whether decreased mtDNA
might disturb the glucose metabolism in liver cells,
resulting in diabetes. First, the ␳0 SK-Hep1 cells
showed a slower growth rate than the parental SKHep1 cells, which was similar to the case of ␳0 HeLa
cells (7). Because of the low levels of intracellular ATP
content and G6PDH, which are determinants of
NADPH production and are essential for cell growth, it
may be natural for ␳0 cells to grow slowly (32). However, growth retardation was not observed in all
mtDNA-depleted cells; the growth rate of human ␳0
osteosarcoma cells was normal (21). This finding indicates that different cells exhibit different sensitivities
to intracellular changes elicited by mtDNA depletion.
Table 1. Specific activities of enzymes involved in
glucose metabolism
Enzymes
SK-Hep1 Cells
␳0 SK-Hep1 Cells
GAPDH
G6PDH
Hexokinase
LDH
2,526 ⫾ 405
66.5 ⫾ 14.3
1,550 ⫾ 291
1,462 ⫾ 157
1,857 ⫾ 370*
20.9 ⫾ 8.1*
120 ⫾ 35†
2,233 ⫾ 201*
Data are means ⫾ SE, expressed as mU/mg cellular protein, of 4–9
different experiments. ␳0, cells depleted of mitochondrial DNA;
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; G6PDH, glucose-6-phosphate dehydrogenase; LDH, lactate dehydrogenase.
* P ⬍ 0.005, † P ⬍ 0.001.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
Fig. 6. Steady-state mRNA levels of glucose transporters. Total RNA
was isolated from SK-Hep1 or its ␳0 cells. The steady-state mRNA levels
of GLUT-1 (n ⫽ 9), GLUT-3 (n ⫽ 5), GLUT-4 (n ⫽ 4), and ␤-actin were
determined by RT-PCR. A: representative pictures are shown. GLUT-2
was not expressed in either group of cells. B: band intensities of the
transporters were normalized to the intensity of ␤-actin and expressed
as the ratio (GLUT/␤-actin). Data are means ⫾ SE; *P ⬍ 0.05,
**P ⬍ 0.01.
Next, we determined which mitochondrial functions
were altered by mtDNA depletion. It was interesting
that the ␳0 cell showed no response to FCCP in ⌬⌿m,
indicating that the ␳0 cells lose the proton gradient
across the mitochondrial membrane. There are reports
of discrepancies on ⌬⌿m of ␳0 cells, although the impaired oxidative phosphorylation may account for the
loss of ⌬⌿m. Kennedy et al. (20) reported that ⌬⌿m was
abolished in ␳0 insulinoma cells, whereas Buchet and
Godinot (7) reported that ⌬⌿m was maintained in ␳0
HeLa cells through the functional F1-ATPase and adenine nucleotide translocator. It is unknown whether
␳0 insulinoma or SK-Hep1 cells contain normal activities of F1-ATPase and adenine nucleotide translocator.
Defects in glucose uptake into muscles and adipose
tissues are generally accepted as one of the major
characteristics of type 2 DM. A decrease in hepatic
glucose uptake was also reported in type 2 diabetic
patients and animal models (5, 30). The basal [3H]FDG
uptake into ␳0 cells was decreased by 30% compared
with the control cells, whereas insulin-stimulated uptake did not occur in either ␳0 or control cells. Because
GLUT-4 is known to play a major role in insulinresponsive glucose transport in peripheral tissues, including muscles and adipose tissues, GLUT-4 might
contribute to total glucose uptake relatively less than
others (19) in SK-Hep1 cells. This is consistent with
the fact that GLUT-3 and GLUT-4 are expressed at a
relatively lower level in SK-Hep1 cells than GLUT-1.
GLUT-1, which is dominant in fetal hepatocytes (15), is
the most abundant type of glucose transporter in SKHep1 cells. We found that the mRNA of GLUT-2, a
GLUCOSE METABOLISM IN ␳0 CELLS
The authors thank Dr. Soo-Young Park for precious support on
this work.
This study was supported by Grant no. 00–4 of the National
Institute of Health, Seoul, Korea.
REFERENCES
1. Antonetti DA, Reynet C, and Kahn CR. Increased expression
of mitochondrial-encoded genes in skeletal muscle of humans
with diabetes mellitus. J Clin Invest 95: 1383–1388, 1995.
2. Ballinger SW, Shoffner JM, Gebhart S, Koontz DA, and
Wallace DC. Mitochondrial diabetes revisited. Nat Genet 7:
458–459, 1994.
3. Ballinger SW, Shoffner JM, Hedaya EV, Trounce I, Polak
MA, Koontz DA, and Wallace DC. Maternally transmitted
diabetes and deafness associated with a 10.4 kb mitochondrial
DNA deletion. Nat Genet 1: 11–15, 1992.
4. Bashan N, Burdett E, Guma A, Sargeant R, Tumiati L, Liu
Z, and Klip A. Mechanism of adaptation of glucose transporters
to changes in the oxidative chain of muscle and fat cells. Am J
Physiol Cell Physiol 264: C430–C440, 1993.
5. Basu A, Basu R, Shah P, Vella A, Johnson CM, Nair KS,
Jensen MD, Schwenk WF, and Rizza RA. Effects of type 2
diabetes on the ability of insulin and glucose to regulate splanchnic and muscle glucose metabolism: evidence for a defect in
hepatic glucokinase activity. Diabetes 49: 272–283, 2000.
6. Brdiczka D. Function of the outer mitochondrial compartment
in regulation of energy metabolism. Biochim Biophys Acta 1187:
264–269, 1994.
7. Buchet K and Godinot C. Functional F1-ATPase essential in
maintaining growth and membrane potential of human mitochondrial DNA-depleted rho degrees cells. J Biol Chem 273:
22983–22989, 1998.
8. Buckler KJ and Vaughan-Jones RD. Effects of mitochondrial
uncouplers on intracellular calcium, pH and membrane potential
in rat carotid body type I cells. J Physiol (Lond) 513: 819–833,
1998.
9. Cesar MD and Wilson JE. Further studies on the coupling of
mitochondrially bound hexokinase to intramitochondrially compartmented ATP, generated by oxidative phosphorylation. Arch
Biochem Biophys 350: 109–117, 1998.
10. DeMoss RD. Glucose-6-phosphate and 6-phosphogluconic dehydrogenase from Leuconostoc mesenteroides. In: Methods in Enzymology, edited by Colowick SP and Kaplan NO. New York:
Academic, 1955, vol. 1, p. 328.
11. Eizirik DL, Sandler S, Ahnstrom G, and Welsh M. Exposure
of pancreatic islets to different alkylating agents decreases mitochondrial DNA content but only streptozotocin induces longlasting functional impairment of ␤-cells. Biochem Pharmacol 42:
2275–2282, 1991.
12. Gallagher BM, Fowler JS, Gutterson NI, MacGregor RR,
Wan CN, and Wolf AP. Metabolic trapping as a principle of
oradiopharmaceutical design: some factors responsible for the
biodistribution of [18F]2-deoxy-2-fluoro-D-glucose. J Nucl Med
19: 1154–1161, 1978.
13. Gerbitz KD, Gempel K, and Brdiczka D. Mitochondria and
diabetes. Genetic, biochemical, and clinical implications of the
cellular energy circuit. Diabetes 45: 113–126, 1996.
14. Gerbitz KD, van den Ouweland JM, Maassen JA, and
Jaksch M. Mitochondrial diabetes mellitus: a review. Biochim
Biophys Acta 1271: 253–260, 1995.
15. Grobholz R, Hacker HJ, Thorens B, and Bannasch P.
Reduction in the expression of glucose transporter protein GLUT
2 in preneoplastic and neoplastic hepatic lesions and reexpression of GLUT 1 in late stages of hepatocarcinogenesis. Cancer
Res 53: 4204–4211, 1993.
16. Hayakawa T, Noda M, Yasuda K, Yorifuji H, Taniguchi S,
Miwa I, Sakura H, Terauchi Y, Hayashi J, Sharp GW,
Kanazawa Y, Akanuma Y, Yazaki Y, and Kadowaki T.
Ethidium bromide-induced inhibition of mitochondrial gene
transcription suppresses glucose-stimulated insulin release in
the mouse pancreatic beta-cell line ␤HC9. J Biol Chem 273:
20300–20307, 1998.
17. Hirota Y, Cohen EM, and Bing OH. Lactate dehydrogenase
and isoenzyme changes in rats with experimental thiamine
deficiency. Metabolism 25: 1–8, 1976.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
major glucose transporter in normal adult hepatocytes,
was not detected in control SK-Hep1 cells, as it has
been reported that the expression of GLUT-2 was severely decreased throughout hepatocarcinogenesis
(15). Also, depletion of mtDNA significantly attenuated
the expression of all types of glucose transporters that
are encoded by nuclear DNA. Further studies are
needed to show how alterations of mtDNA could affect
the expression of nuclear DNA-encoded proteins involved in glucose transport but not a structural protein
like ␤-actin.
It is possible that the decreased glucose uptake reduces the glucose supply for enzymes involved in glucose metabolic pathways. These enzymes may be inactivated or downregulated. As expected, we found that,
in ␳0 cells, the activities of hexokinase, GAPDH, and
G6PDH were decreased by 90, 25, and 70%, respectively. These data suggest that the depletion of mtDNA
decreased glucose utilization by suppressing glucose
metabolism in addition to reducing glucose uptake.
Furthermore, the attenuated activities of glycolytic
enzymes could consequently reduce ATP production,
which may aggravate ATP depletion in ␳0 cells in a
vicious cycle.
It is interesting to note the large decrease of hexokinase activity in ␳0 cells. In contrast to other glycolytic
enzymes, hexokinase is known to be associated with
the mitochondrial outer membrane through its interaction with porin and uses intramitochondrial ATP
supplied by oxidative phosphorylation as a substrate
for glucose phosphorylation (9). Moreover, the activation of hexokinase depends on a contact site-specific
structure of the pore, which is voltage dependent and
influenced by the electrical potential of the mitochondrial inner membrane (13). It has been reported that
mitochondria lacking a membrane potential induced
by a mitochondrial uncoupler such as dinitrophenol
decreases the contacts and hexokinase activity in hepatocytes (6). We also observed that the hexokinase protein was significantly decreased in mitochondrial fraction from ␳0 cells, and this was determined by Western
blotting assay (unpublished data). Therefore, we can
infer that the defects in intramitochondrial ATP production and depolarized ⌬⌿m in ␳0 cells could inactivate hexokinase. Furthermore, the decreased hexokinase activity in ␳0 cells may be one of the fundamental
causes of the disturbed glucose metabolism.
This study demonstrates for the first time that
mtDNA depletion may disturb the cellular capacity for
glucose utilization, at least in liver cells, although the
study was performed in a single cell line and not over
a wide variety of cell types. Further studies on other
changes in various types of mtDNA-depleted cell lines
would be helpful in extending our knowledge about the
pathogenic role of quantitative changes in mtDNA.
E1013
E1014
GLUCOSE METABOLISM IN ␳0 CELLS
28.
29.
30.
31.
32.
33.
34.
35.
variable tissue expression: a novel genetic abnormality in mitochondrial diseases. Am J Hum Genet 48: 492–501, 1991.
Owen P and Freer JH. Factors influencing the activity of
succinate dehydrogenase in membrane preparations from Micrococcus lysodeikticus. Biochem J 120: 237–243, 1970.
Serradas P, Giroix MH, Saulnier C, Gangnerau MN, Borg
LA, Welsh M, Portha B, and Welsh N. Mitochondrial deoxyribonucleic acid content is specifically decreased in adult, but not
fetal, pancreatic islets of the Goto-Kakizaki rat, a genetic model
of non-insulin-dependent diabetes. Endocrinology 136: 5623–
5631, 1995.
Shiba Y, Yamasaki Y, Kubota M, Matsuhisa M, Tomita T,
Nakahara I, Morishima T, Kawamori R, and Hori M.
Increased hepatic glucose production and decreased hepatic
glucose uptake at the prediabetic phase in the Otsuka LongEvans Tokushima fatty rat model. Metabolism 47: 908–914,
1998.
Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, Oka
Y, and Hayashi JI. Mitochondrial DNA is required for regulation of glucose-stimulated insulin secretion in a mouse
pancreatic beta cell line, MIN6. J Biol Chem 271: 26194–
26199, 1996.
Tian WN, Braunstein LD, Pang J, Stuhlmeier KM, Xi QC,
Tian X, and Stanton RC. Importance of glucose-6-phosphate
dehydrogenase activity for cell growth. J Biol Chem 273: 10609–
10617, 1998.
Velick SF. Glyceraldehyde-3-phosphate dehydrogenase. In: The
Enzymes, edited by Boyer PD, Lardy H, and Myrback K. New
York: Academic, 1961, vol. 7, p. 243.
Wallace DC. Mitochondrial genetics: a paradigm for aging and
degenerative diseases? Science 256: 628–632, 1992.
Zylber E, Vesco C, and Penman S. Selective inhibition of the
synthesis of mitochondria-associated RNA by ethidium bromide.
J Mol Biol 44: 195–204, 1969.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
18. Kadowaki T, Kadowaki H, Mori Y, Tobe K, Sakuta R,
Suzuki Y, Tanabe Y, Sakura H, Awata T, and Goto Y. A
subtype of diabetes mellitus associated with a mutation of mitochondrial DNA. N Engl J Med 330: 962–968, 1994.
19. Kahn BB. Lilly lecture 1995. Glucose transport: pivotal step in
insulin action. Diabetes 45: 1644–1654, 1996.
20. Kennedy ED, Maechler P, and Wollheim CB. Effects of
depletion of mitochondrial DNA in metabolism secretion coupling in INS-1 cells. Diabetes 47: 374–380, 1998.
21. King MP and Attardi G. Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 246: 500–503, 1989.
22. Lee HK, Song JH, Shin CS, Park DJ, Park KS, Lee KU, and
Koh CS. Decreased mitochondrial DNA content in peripheral
blood precedes the development of non-insulin-dependent diabetes mellitus. Diabetes Res Clin Pract 42: 161–167, 1998.
23. Li SR, Baroni MG, Oelbaum RS, Stock J, and Galton DJ.
Association of genetic variant of the glucose transporter with
non-insulin-dependent diabetes mellitus. Lancet 2: 368–370,
1988.
24. Madden EA and Storrie B. The preparative isolation of mitochondria from Chinese hamster ovary cells. Anal Biochem 163:
350–357, 1987.
25. Martin BC, Warram JH, Krolewski AS, Bergman RN,
Soeldner JS, and Kahn CR. Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a
25-year follow-up study. Lancet 340: 925–929, 1992.
26. Miwa I, Mita Y, Murata T, Okuda J, Sugiura M, Hamada Y,
and Chiba T. Utility of 3-O-methyl-N-acetyl-D-glucosamine, an
N-acetylglucosamine kinase inhibitor, for accurate assay of glucokinase in pancreatic islets and liver. Enzyme Protein 48: 135–
142, 1994.
27. Moraes CT, Shanske S, Tritschler HJ, Aprille JR, Andreetta
F, Bonilla E, Schon EA, and DiMauro S. mtDNA depletion with