0013-7227/01/$03.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 86(10):4920 – 4925
Copyright © 2001 by The Endocrine Society
Defective Mitochondrial ATP Synthesis in Oxyphilic
Thyroid Tumors
F. SAVAGNER, B. FRANC, S. GUYETANT, P. RODIEN, P. REYNIER,
AND
Y. MALTHIERY
Inserm EMI-U 00-18 (F.S., P.R., P.R., Y.M.), Laboratoire de Biochimie et Biologie Moléculaire, Angers F-49033; Laboratoire
d’Anatomie Pathologique (B.F.), Hôpital Ambroise Paré, Boulogne F-92104; Laboratoire d’Anatomie Pathologique (S.G.);
and Service d’Endocrinologie (P.R.), Nutrition et Médecine Interne, Angers F-49033, France
Oxyphilic tumors (oncocytomas or Hürthle cell tumors) form
a rare subgroup of thyroid tumors characterized by cells containing abundant mitochondria. The relationship between
the mitochondrial proliferation and the pathogenesis of these
tumors is unknown. We have assessed the expression of the
mitochondrial ND2 and ND5 (subunits of the nicotinamide
adenine dinucleotide dehydrogenase complex) genes and the
nuclear UCP2 (uncoupling protein 2) gene in 22 oxyphilic
thyroid tumors and matched controls. The consumption of
O
XYPHILIC THYROID TUMORS, also known as oncocytomas or Hürthle cell tumors, represent a rare subgroup of follicular thyroid neoplasms, characterized by cells
with a distinctive eosinophilic cytoplasm (1). The cytoplasmic eosinophilia is owing to the abundance of morphologically altered mitochondria in the majority of tumor cells (1,
2). However, the relationship between mitochondrial proliferation and the histogenesis of oxyphilic tumors is unknown.
The development of mitochondria involves the synthesis
of proteins encoded by mitochondrial DNA (mtDNA), which
carries the genes for the essential subunits of the respiratory
chain complexes, as well as by nuclear DNA (nDNA). Electrons from the nicotinamide adenine dinucleotide generated
by glycolysis in the cell are transported into the mitochondria
in which the flow of electrons between the respiratory chain
complexes supplies the energy used by ATP synthase to
produce ATP. An alternative source of energy, independent
of ATP synthase, is provided by the uncoupling protein
(UCP), which plays an important role in energy homeostasis
(3). The analysis of mtDNA in several oxyphilic tumors has
shown that the abnormally high histochemical activity of the
respiratory chain complexes is associated with a great increase in the amount of wild-type mtDNA (4).
The frequency of aneuploid or polyploid cells in oxyphilic
thyroid tumors suggests the presence of anomalies in nuclear
genes (5). Mitochondrial proliferation has also been reported
in mitochondrial diseases associated with respiratory chain
defects or coupling defects between the respiratory chain and
ATP production (6 –9). Among the UCPs that induce thermogenesis during mitochondrial respiration (3), UCP2 is the
unique form expressed in many tissues and cell types. In
particular, UCP2 is expressed in thyroid tissue, whereas the
expression of UCP1 and UCP3 is limited to brown fat adipose
Abbreviations: mtDNA, Mitochondrial DNA; nDNA, nuclear DNA;
rDNA, ribosomal DNA; UCP, uncoupling protein.
oxygen in mitochondria from tumors was determined by polarography. ATP assays were used to explore the mitochondrial respiratory chain activity and the oxidative phosphorylation coupling in seven fresh thyroid tumors and controls.
Adenosine triphosphate synthesis was significantly lower in
all the tumors, compared with controls, suggesting that a coupling defect in oxidative phosphorylation may be a cause of
mitochondrial hyperplasia in oxyphilic thyroid tumors. (J
Clin Endocrinol Metab 86: 4920 – 4925, 2001)
tissue and muscle/white fat tissues, respectively (10). However, the histochemical investigation of the key nuclear components involved in the oxidative phosphorylation process
has revealed no coupling defects in oxyphilic thyroid tumors
(11).
The proliferation of mitochondria in oxyphilic tumors
might result from the induction of genes involved in mitochondrial biogenesis. The patterns of mitochondrial transcripts (especially ND2 and ND5) as well as nuclear transcripts from certain genes coding for proteins involved in
oxidative phosphorylation differ according to whether the
tumor is a renal oxyphilic tumor or a salivary gland oxyphilic
tumor (12). Changes in mtDNA transcription and mitochondrial mRNA stability were observed in the former but not in
the latter, suggesting that the process of mitochondrial proliferation varies according to the origin of the tumor.
We examined thyroidectomy specimens from 22 anonymous patients. In each case, tissue removed from the normal
part of the thyroid gland served as a control for the oxyphilic
thyroid tumor. For each set of paired specimens, we determined the gene expression profiles of the mitochondrial ND2
and ND5 genes as well as the nuclear UCP2 gene involved
in energy production. Fresh tissue samples, obtained from 7
of the 22 patients, were analyzed by polarography to investigate the mitochondrial respiratory chain activity, the rate of
oxidative phosphorylation (ADP/oxidation ratio) and ATP
assays were used to determine the mitochondrial ATP
synthesis.
Materials and Methods
Thyroid tissue samples
Twenty-two benign or malignant oxyphilic thyroid tumors, diagnosed between 1992 and 2000 at the Ambroise Paré Hospital, Paris (15
cases) and the University Hospital, Angers (7 cases), were included in
the study. All the samples used were rendered anonymous (i.e. all
patient identifiers were deleted before the study). The cases were con-
4920
Savagner et al. • ATP Synthesis in Oxyphilic Thyroid Tumors
secutive and unselected apart from exclusions on account of insufficient
material or association of the tumors with chronic thyroiditis. Nineteen
of the tumors were follicular oxyphilic adenomas (six of which were
trabecular) and three were follicular oxyphilic carcinomas. Five of the
adenomas were associated with a multinodular goiter. The diagnoses
were made according to the World Health Organization classification
(1). Oxyphilic adenoma was distinguished from carcinoma on the basis
of vascular or capsular invasion or metastasis. The patients were 2 men
and 20 women, with a mean age of 53 yr (range 27– 82 yr).
The average size of the tumor was 37.8 ⫾ 20.5 mm (mean ⫾ sd; range
15–90 mm). In addition to neoplastic thyroid samples, normal thyroid
samples were taken sufficiently distant from the tumors to serve as
controls. All the samples were immediately stored in liquid nitrogen
until extraction of high-molecular-weight DNA and RNA.
The 22 tumor samples and controls were fixed in formalin, embedded
in paraffin, and stained with hematoxylin and eosin. Immunohistochemistry was performed on paraffin-embedded sections. Tissue samples from the seven fresh tumors (six adenomas and one carcinoma) and
matched controls were kept in a preservative medium [100 mm sucrose,
1 mm EGTA, 20 mm 3-[N-morpholino]propanesulfonic acid (pH 7.4), 1
g/liter bovine albumin] to prepare the mitochondria for polarographic
studies and ATP measurement.
Tumoral and control tissues were compared using a nonparametric
test for matched-pair samples (Wilcoxon test), the differences being
considered statistically significant at P less than or equal to 0.05. All the
numerical values below are expressed as means ⫾ sd.
Immunohistochemistry
After morphological examination of hematoxylin- and eosin-stained
sections, corresponding 3-␮m sections of the paraffin blocks were prepared for the detection of mitochondrial antigen expression as a semiquantitative index of mitochondrial biogenesis. A monoclonal antibody
113–1 was used, recognizing an unknown 60-kDa nonglycosylated protein component of human cell mitochondria (BioGenex Laboratories,
Inc., San Ramon, CA). Immunostaining was performed with the standard avidin-biotin peroxidase technique with antigen retrieval. For negative control slides, the primary antibody was either omitted or replaced
by a suitable concentration of normal IgG of the same species.
Polarographic studies
Oxygen consumption was measured with a Clark electrode at 30 C
in a 2-mL chamber (Oxygraph OROBOROS, Anton Paar, Innsbrück,
Austria). The chamber was isolated from contact with the atmosphere
by a close-fitting cap so that the electrode current was proportional to
the partial pressure of oxygen in the sample. Substrates for the different
complexes of the respiratory chain were introduced into the chamber
and consumed by mitochondria in the oxyphilic tumors or matched
controls. Mitochondria were isolated from seven fresh tissue samples
and matched controls using the standard procedure (13). Oxygen uptake
was measured after the addition of mitochondria (200 ␮g protein) to 2
ml incubation medium [300 mm mannitol, 10 mm KH2PO4 (pH 7.2), 10
mm KCl, and 5 mm MgCl2] to determine the basal respiratory activity.
Substrates and inhibitors were introduced into the oxygraph chamber
through the stopper port. First, 2 ␮l EDTA (20 mm) and 2 ␮l rotenone
(5 mm) were added to inhibit exogenous ATPase and complex I (nicotinamide adenine dinucleotide ubiquinone oxidoreductase of the respiratory chain), respectively. Then, 20 ␮l succinate (1 m) and 10 ␮l ADP
(30 mm) were successively added to the sample to determine the rate of
oxidative phosphorylation (ADP/O ratio). Finally, 10 ␮l potassium cyanide (200 mm) were added to stop oxygen uptake by inhibiting complex
IV (cytochrome c oxidase of the respiratory chain).
ATP measurement
Mitochondrial ATP was measured by bioluminescence using the
luciferin-luciferase reaction (Enliten, Promega Corp., Madison, WI) (14).
Mitochondria were isolated, using the same methods as for the polarographic studies, from the seven fresh oxyphilic tumor samples and their
matched controls. After incubation of mitochondria with 10 mm glutamate and malate for 10 min, the rate of ATP synthesis (expressed per
J Clin Endocrinol Metab, October 2001, 86(10):4920 – 4925 4921
milligram of mitochondrial protein) was determined for intact and permeabilized mitochondria.
DNA isolation and Southern blot analysis
DNA was isolated using the phenol-chloroform procedure. The samples were digested overnight with RNase A (20 ␮g/ml) and proteinase
K (20 mg/ml) at 37 C in Tris-HCl 10 mm, EDTA (pH 8) 0.1 m, and SDS
0.5%. The proteins were removed by organic extraction followed by
ethanol precipitation with NaCl 0.2 m and centrifugation for 15 min at
10,000 g. Five micrograms of DNA were digested with the restriction
enzyme XbaI (Biolabs, Beverly, MA). Southern blotting was performed
according to standard methods. Probes labeled by digoxigenin were
obtained by multirandom priming and were revealed with antidigoxigenin antibodies labeled by alkaline phosphatase (DigDNA labeling and
detection kit, Roche, Basel, Switzerland). The mitochondrial and nDNA
were detected by using the probes 12S ribosomal DNA (rDNA) (nt
592-1344) and 18S rDNA (nt 1201–1811), respectively. For each sample,
the intensities of the mtDNA signals, and the corresponding nDNA
signals, were quantified by densitometric analysis (Molecular Analyst,
Bio-Rad Laboratories, Inc., Cambridge, MA).
RNA isolation and cDNA synthesis
RNA was isolated using the guanidinium isothiocyanate procedure
(Trizol Reagent, Life Technologies, Inc., Gaithersburg, MD). Residual
DNA was removed by DNase treatment: 5 ␮g total RNA were incubated
with 2 U RNase-free DNase I for 1 h at 37 C.
To generate cDNA, 1 ␮g of RNA was first denatured at 70 C with 1
␮m of oligodT (Promega Corp.) for 5 min before quenching on ice; then
0.5 mm of each of the 4 dNTPs, 10 mm dithiothreitol, 10 U RNase
inhibitor, and 200 U superscript II (Life Technologies, Inc.) were added
to the 5⫻ buffer to make up a final volume of 20 ␮l reaction mix. The
reaction mix was incubated for 1 h at 42 C. The reverse transcriptase was
inactivated at 70 C for 15 min.
Quantitative PCR analysis
Real-time quantitative PCR with an external standard was used to
determine the gene copy number (Lightcycler, Roche). Standard PCR
products for each gene were generated by amplifying nuclear cDNA or
mtDNA templates. PCR products were purified by the phenol-chloroform method and the copy number in the final sample was determined
by two independent methods (i.e. spectrophotometry and gel analysis).
For each gene tested, a sequence-specific standard curve was plotted
using serial dilutions of the target gene standard PCR product, and the
same primers were used to amplify the cDNA.
The expression of two mitochondrial genes, ND2 and ND5, and two
nuclear genes, UCP2 and ␤-ACTIN, was analyzed using the PCR primer
sets indicated in Table 1. The amount of RNA determined for each
sample was normalized by the quantification of the ␤-ACTIN transcripts.
Two microliters of master mix containing Taq DNA polymerase,
dNTPs, and SYBR green I (DNA Master SYBR Green I kit, Roche) were
incubated for 5 min at room temperature with 0.16 ␮l of Taqstart antibody. The PCR reaction was then started by adding MgCl2 4 mm and
forward and reverse primers (0.5 ␮m) to the capillary tubes of the
Lightcycler apparatus containing the master mix and 2 ␮L of template
(cDNA or a standard with a known copy number) in a final volume of
20 ␮l.
TABLE 1. Oligonucleotide pairs used for quantitative PCR
Primer pairs
5⬘-GCACCCCTCTGACATCC-3⬘
5⬘-CGGTCGGCGAACATCAGTGG-3⬘
5⬘-GGGGATTGTGCGGTGTGTG-3⬘
5⬘-CTTCTCCTATTTATGGGGGT-3⬘
5⬘-CCAGTGCGCGCGCTACAGTCA-3⬘
5⬘-GTGGTGCTGCCTGCTAGGAG-3⬘
5⬘-CGACATGGAGAAAATCTGGC-3⬘
5⬘-AGGTCCAAGACGCAGGATGG-3⬘
Genes
ND2
ND5
UCP2
␤-ACTIN
4922
J Clin Endocrinol Metab, October 2001, 86(10):4920 – 4925
Savagner et al. • ATP Synthesis in Oxyphilic Thyroid Tumors
Results
mitochondria, attesting to the functionality of ATP/ADP
translocase in exporting ATP from intact mitochondria.
Immunohistochemistry
The 60-kDa mitochondrial protein was present in the cytoplasm of all the 22 oxyphilic thyroid tumor samples, both
homogeneous (19 tumors) and heterogeneous (3 tumors)
distributed (histochemistry score data not shown), confirming the increased mitochondrial biogenesis in the tumor samples. In contrast, the immunostaining of the matched control
samples was extremely weak or undetectable. The homogeneous distribution of mitochondrial immunostaining with
no difference between adenomas vs. carcinomas has been
previously described for oxyphilic tumors (15).
Polarographic analysis
Seven tumors and matched controls (n ⫽ 14) were used for
the functional analysis of the respiratory chain. The mitochondria were prepared within 1 h after thyroidectomy so as
to obtain interpretable results. Polarographic analysis of
complexes I to IV produced no evidence of respiratory chain
defects in the seven oxyphilic tumors vs. control tissue. The
respiratory control indices were calculated by dividing the
rate of oxygen uptake in state 3 (stimulated by ADP addition)
by the rate of oxygen uptake in state 4 (when all the ADP is
converted to ATP). Using succinate as substrate, the indices
were 3.8 ⫾ 0.8 for tumors and controls (n ⫽ 14). The ADP/O
ratio in the tumor samples was 1.2 ⫾ 0.3 (n ⫽ 7), whereas it
was 1.6 ⫾ 0.4 in the matched controls (n ⫽ 7), but this
difference is not significant (Table 2).
ATP synthesis
Because the analysis of ATP synthesis can be usefully
performed exclusively on mitochondria from fresh tissues,
only the mitochondria from the seven tumors and their
matched controls (n ⫽ 14), as used for the polarographic
studies, were used for ATP measurement. The mitochondrial
ATP synthesis, adjusted to the mitochondrial protein level
after addition of the substrate, was 5.8 ⫾ 1.4 ␮mol/mg per
10 min in oxyphilic tumors samples, compared with 12.1 ⫾
3.1 ␮mol/mg per 10 min in matched controls. This represents
a low rate of mitochondrial ATP synthesis in comparison
with the basal rate for normal thyroid tissue we had previously measured (13.5 ⫾ 2.8) (unpublished data). The Wilcoxon test showed that the decrease observed in ATP synthesis was highly significant (P ⫽ 0.018, Table 2). There was
no difference in ATP levels between intact and permeabilized
MtDNA quantification
Twenty-two tumors and matched controls were explored
for mtDNA quantification. DNA was preserved from rRNA
contamination by buffered RNase during extraction. The
ratio of mtDNA to nDNA in oxyphilic tumor samples and
matched controls was determined by Southern blot analysis.
Three hybridization bands were detected by the mitochondrial 12S rRNA probe (7.5 and 1.7 kb) and the nuclear 18S
rRNA probe (1.5 kb). The densitometric analysis of the 1.7and 1.5-kb bands showed that the 12S rDNA/18S rDNA ratio
was 1.67 ⫾ 0.09 in oxyphilic thyroid tumors, compared with
0.54 ⫾ 0.19 in controls. This increase in mtDNA content for
tumors was highly significant, using the Wilcoxon test (P ⬍
0.001). Fig. 1 shows the 1.7- and 1.5-kb bands of several
oxyphilic tumors samples and matched controls.
Deletions in mtDNA were explored by long PCR analysis
using a standard procedure (16). The common mtDNA4977
deletion was found in two of the tumors as well as in the
matched controls, with an identical level of heteroplasmy.
All the other samples were free from this deletion.
Mitochondrial and nuclear gene expression
Twenty-two tumors and matched controls were explored
for mtDNA quantification. The expression of ND2 and ND5
mitochondrial genes was 12 times higher in oxyphilic thyroid
tumor samples than in controls. When adjusted to the
mtDNA/nDNA ratio, the relative mitochondrial transcript
ratio was 3.8 times higher in the tumor samples than in
controls.
For the nuclear gene, we observed a 2-fold increase of
UCP2 in oxyphilic tumors samples, compared with controls.
Fig. 2 shows the different patterns of mitochondrial and
nuclear gene expression in the tumor samples and controls.
Table 3 summarizes the histology and the gene expression
pattern of the different samples. Table 2 sums up the statistical analysis of the results.
Discussion
Several authors have suggested that defective energyproducing mechanisms of oxyphilic cells may be responsible
for mitochondrial proliferation (11, 17). This hypothesis
stems from the observation that the active metabolism of
TABLE 2. Statistical analysis of mitochondrial (ND2, ND5) and nuclear (UCP2) gene expression and ATP synthesis in oxyphilic thyroid
tumors, compared with matched controls
ND2b
ND5b
UCP2b
ADP/O ratio
ATP (␮mol/mg protein per 10 min)
Control tissuea
(n ⫽ 22)
Oxyphilic tumorsa
(n ⫽ 22)
658 ⫾ 290
612 ⫾ 302
272 ⫾ 119
n⫽7
1.6 ⫾ 0.4
12.1 ⫾ 3.1
6799 ⫾ 3215
7354 ⫾ 3208
503 ⫾ 191
n⫽7
1.2 ⫾ 0.3
5.8 ⫾ 1.4
Mean ⫾ SD.
Copy number/␤-ACTIN copy number.
ADP/O ratio, Rate of oxydative phosphorylation; NS, not significant.
a
b
Wilcoxon test
P ⬍ 0.001
P ⬍ 0.001
P ⬍ 0.001
NS
P ⫽ 0.018
Savagner et al. • ATP Synthesis in Oxyphilic Thyroid Tumors
J Clin Endocrinol Metab, October 2001, 86(10):4920 – 4925 4923
FIG. 1. Quantitation of mtDNA and
nDNA for several oxyphilic thyroid tumors and matched controls by Southern
blot analysis. Five micrograms of total
DNA digested with XbaI was hybridized first with an 18S rRNA nuclear
probe and then rinsed and hybridized
with a 12S rRNA mtDNA probe. The
intensities of the 1.5-kb (18S) and
1.7-kb (12S) bands were quantified by a
RadioImager (Cyclone, Packard, Downers Grove, IL). C, Control tissue; O, oxyphilic thyroid tumor.
FIG. 2. Quantification of mitochondrial and nuclear gene transcripts
from 22 oxyphilic thyroid tumors and matched controls. 2a, Mean (⫾
SD) of the ratio of the mitochondrial ND2 and ND5 cDNA copy numbers vs. the ␤-ACTIN cDNA copy number. 2b, Mean (⫾ SD) of the ratio
of the nuclear UCP2 cDNA copy number vs. the ␤-ACTIN cDNA copy
number. The cDNA copies were determined by real-time quantitative
PCR analysis (Lightcycler, Roche, Basel, Switzerland) after reverse
transcription of the total RNA of each tumor and control sample.
(Table 1 shows the PCR primers used.)
oncocytic cells, with their high levels of oxidative enzymes,
does not correspond to high thyroid cell function (18). However, in these histochemical studies, the respiratory enzymes
were functional, and a protein uncoupling the oxidative
phosphorylation process (UCP1) was not present in oxyphilic tumors (11).
MtDNA alterations are associated with several mitochondrial degenerative diseases (19). Large mtDNA deletions,
such as the most common mtDNA4977 deletion, result in a
decline of the oxidative phosphorylation capacity and accumulate progressively in aging normal tissues (20). An increased frequency of the mtDNA4977 deletion in oncocytic
tumors has been described (18). We were unable to identify
any mtDNA deletions that might provide a replicative advantage over wild-type mtDNA. The “common” mtDNA4977 deletion was found in 2 of the 22 oxyphilic tumors investigated,
but in each case the same deletion was also detected in the
corresponding controls and corresponded to two elderly patients (63 and 75 yr). Thus, the mtDNA4977 deletion might be
associated with cellular aging rather than to the development
of an oncocytic phenotype, as previously suggested (21, 22). The
increase in mtDNA content observed in 22 thyroid tumors
(3.10 ⫾ 0.29) was 25% lower than that indicated by other authors
(4.31 ⫾ 1.09) (21).
The analysis of mitochondrial gene expression showed
that tumors had a 12-fold increase in ND2 and ND5 transcripts, compared with control tissue. The expression of these
two mitochondrial genes has already been associated with
abnormal mitochondrial biogenesis in oncocytic tumors (12).
The gene expression ratio was adjusted to the mtDNA/
nDNA ratio calculated by Southern blot analysis to obtain a
more accurate estimate of the real increase of mitochondrial
gene expression. The mtRNA/mtDNA ratio thus determined for the oxyphilic thyroid tumors may be compared
with that given for oxyphilic tumors in other tissues. We
found an mtRNA/mtDNA ratio of about 4:1 in oxyphilic
thyroid tumors, compared with controls, whereas this ratio
was 1:1 in the case of oxyphilic salivary gland tumors and 1:5
in the case of oxyphilic renal tumors (12). In the study of a
cell line derived from a thyroid oncocytoma, we found an
mtRNA/mtDNA ratio as high as 2:1, compared with a control thyroid cell line (23). These large differences in the
mtRNA/mtDNA ratio suggest tissue-specific regulation of
mitochondrial transcription and replication. It might therefore be relevant to investigate the nuclear factors involved in
this regulation in various tissues.
Polarographic analysis produced no evidence of respiratory chain defects in oxyphilic thyroid tumors, compared with control tissue. The respiratory chain ratios in
mitochondria isolated from seven oxyphilic tumors were
consistent with the indices published for mitochondria in
the normal thyroid (24). However, the ADP/O ratio was
only 75% of the normal value. The oxidative phosphorylation coupling defect revealed by polarography might be
related to the 2-fold increase in UCP2 expression observed
in oxyphilic tumors, compared with controls. After verifying that UCP1 was not expressed in oxyphilic thyroid
tumors (data not shown), we investigated the expression
of UCP2, the role of which has been established in the
uncoupling process (25).
4924
J Clin Endocrinol Metab, October 2001, 86(10):4920 – 4925
Savagner et al. • ATP Synthesis in Oxyphilic Thyroid Tumors
TABLE 3. Case reports
Case
Histology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
OTA
OFA
OFC
OFA
OFA
OFC
OFA
OTA
OFA
OFA
OTA
OTA
OTA
OTA
OFA
OFA
OFC
OFA
OFA
OFA
OFA
OFA
ND2#
ND5#
UCP2#
C
O
C
O
C
O
752
267
402
1,140
760
804
589
549
705
941
875
1,236
865
1,130
456
442
583
345
321
238
643
436
7,890
5,749
2,592
14,587
8,403
8,702
3,015
4,057
9,587
5,783
2,378
12,700
5,640
6,389
8,653
7,833
6,421
10,054
3,890
3,420
7,804
4,042
812
320
130
1,250
548
647
950
450
521
203
875
694
549
253
159
377
590
1,050
858
665
947
635
6,850
7,760
1,390
11,760
11,540
12,501
5,470
5,690
12,589
2,547
7,646
11,345
5,578
3,543
3,211
6,489
7,345
8,897
6,789
5,780
8,085
8,988
355
123
420
201
341
201
258
507
204
438
321
365
532
130
147
230
184
163
208
201
220
251
469
198
972
495
507
357
398
980
480
508
542
603
685
603
365
521
475
295
265
385
430
541
#, Copy number/␤-ACTIN copy number; OFA, oxyphilic follicular adenoma; OFC, oxyphilic follicular carcinoma; OTA, oxyphilic follicular
adenoma (trabecular pattern); C, control surrounding tissue; O, oxyphilic thyroid tumor.
In our study of seven fresh oxyphilic tumors, the increased
expression of UCP2 was probably responsible for the coupling defect reflected by a significant decrease in ATP synthesis. Mitochondrial proliferation could therefore be an
adaptive response to a primary nuclear abnormality, namely
the overexpression of UCP2. However, the overexpression of
UCP2 might itself be a response to the proliferation of mitochondria compensating for the decreased mitochondrial
ATP synthesis. In the latest case, the proliferation of mitochondria leads to overproduction of reactive oxygen species,
which could be counteracted by an increase in UCP2 expression (26).
Another line of reasoning suggests that the metabolism in
oxyphilic tumors has switched to a glycolytic status (12). The
decrease in mitochondrial ATP synthesis we noticed could
lead to a shift toward anaerobic metabolism. Because the
defect is measured in oxyphilic adenoma as well as in carcinoma, we suggest that the metabolism switch is an early
event in the oxyphilic thyroid tumor progression. Thus, the
oxyphilic cell might be early resistant to hypoxia, which
could explain the aggressive clinical behavior of these tumors
(22).
In conclusion, the defective ATP synthesis we observed in
seven oxyphilic thyroid tumors might explain the mitochondrial proliferation found in the tumor cells. Because the expression of UCP2 was higher in all the 22 oxyphilic thyroid
tumors, compared with controls, we suggest that the oxidative phosphorylation coupling defect we detected may be
associated with mitochondrial proliferation in oxyphilic thyroid tumors. It would therefore be of interest to further investigate the factors involved in the transcription and replication of mitochondrial DNA in oxyphilic tumors of the
thyroid gland.
Acknowledgments
We are grateful to C. Savagner for the statistical analysis and to K.
Malkani for critical reading of the manuscript. We thank Anne, Dominique, and Florence for continuous support during the study.
Received January 26, 2001. Accepted June 5, 2001.
Address all correspondence and requests for reprints to: F. Savagner, Inserm EMI-U 00-18, Laboratoire de Biochimie et Biologie
Moléculaire, Chu, 4 rue Larrey, F-49033 Angers cedex 01, France.
E-mail: [email protected].
This work was supported by grants from l’Association pour la Recherche sur le Cancer.
References
1. Hedinger C, Williams ED, Sobin LH 1989 The WHO histological classification
of thyroid tumors: a commentary on the ed. 2. Cancer 63:908 –911
2. Katoh R, Harach HR, Williams ED 1998 Solitary, multiple, and familial oxyphil tumours of the thyroid gland. J Pathol 186:292–299
3. Ricquier D, Bouillaud F 2000 The uncoupling protein homologues: UCP1,
UCP2, UCP3, StUCP and AtUCP. Biochem J 345:161–179
4. Tallini G 1998 Oncocytic tumours. Virchows Arch 433:5–12
5. Bronner MP, Clevenger CV, Edmonds PR, Lowell DM, McFarland MM,
LiVolsi VA 1988 Flow cytometric analysis of DNA content in Hürthle cell
adenomas and carcinomas of the thyroid. Am J Clin Pathol 89:764 –769
6. Bouillaud F, Ricquier D, Thibault J, Weissenbach J 1985 Molecular approach
to thermogenesis in brown adipose tissue: cDNA cloning of the mitochondrial
uncoupling protein. Proc Natl Acad Sci USA 82:445– 448
7. DiMauro S, Bonilla E, Zeviani M, Nakagawa M, DeVivo DC 1985 Mitochondrial myopathies. Ann Neurol 17:521–538
8. Shoffner JM, Wallace DC 1990 Oxidative phosphorylation diseases. Disorders
of two genomes. Adv Hum Genet 19:267–330
9. Heddi A, Lestienne P, Wallace DC, Stepien G 1993 Mitochondrial DNA
expression in mitochondrial myopathies and coordinated expression of nuclear genes involved in ATP production. J Biol Chem 268:12156 –12163
10. Flier J, Lowell B 1997 Obesity research springs a proton leak. Nat Genet
15:223–224
11. Ebner D, Rodel G, Pavenstaedt I, Haferkamp O 1991 Functional and molecular analysis of mitochondria in thyroid oxyphil tumour. Virchows Arch B
Cell Pathol Incl Mol Pathol 60:139 –144
12. Heddi A, Faure-Vigny H, Wallace DC, Stepien G 1996 Coordinate expression
of nuclear and mitochondrial genes involved in energy production in carcinoma and oxyphil tumour. Biochim Biophys Acta 1316:203–209
13. Chretien D, Bourgeron T, Rotig A, Munnich A, Rustin P 1990 The measure-
Savagner et al. • ATP Synthesis in Oxyphilic Thyroid Tumors
14.
15.
16.
17.
18.
19.
ment of the rotenone-sensitive NADH cytochrome c reductase activity in
mitochondria isolated from minute amount of human skeletal muscle. Biochem Biophys Res Commun 173:26 –33
Lemasters JJ, Hackenbrock CR 1976 Continuous measurement and rapid
kinetics of ATP synthesis in rat liver mitochondria, mitoplasts and inner
membrane vesicles determined by firefly-luciferase luminescence. Eur J Biochem 67:1–10
Tallini G, Carcangiu ML, Rosai J 1992 Oncocytic neoplasms of the thyroid
gland. Acta Pathol Jpn 42:305–315
Reynier P, Chretien MF, Savagner F, et al. 1998 Long PCR analysis of human
gamete mtDNA suggests defective mitochondrial maintenance in spermatozoa and supports the bottleneck theory for oocytes. Biochem Biophys Res
Commun 254:373–377
Heddi A, Stepien G, Benke PJ, Wallace DC 1999 Coordinate induction of
energy gene expression in tissues of mitochondrial disease patients. J Biol
Chem 274:22968 –22976
Welter C, Kovacs G, Seitz G, Blin N 1989 Alteration of mitochondrial DNA
in human oxyphil tumours. Genes Chromosomes Cancer 1:79 – 82
Wallace DC 1992 Diseases of the mitochondrial DNA. Annu Rev Biochem
61:1175–1212
J Clin Endocrinol Metab, October 2001, 86(10):4920 – 4925 4925
20. Cortopassi G, Shibata D, Soong N, Arnheim N 1992 A pattern of accumulation of a somatic deletion of mitochondrial DNA in aging human tissues. Proc
Natl Acad Sci USA 89:7370 –7374
21. Tallini G, Ladanyi M, Rosai J, Jhanwar SC 1994 Analysis of nuclear and
mitochondrial DNA alterations in thyroid and renal oncocytic tumors. Cytogenet Cell Genet 66:253–259
22. Maximo V, Sobrinho-Simoes M 2000 Hürthle cell tumours of the thyroid. A
review with emphasis on mitochondrial abnormalities with clinical relevance.
Virchows Arch 437:107–115
23. Savagner F, Chevrollier A, Loiseau D, et al. 2001 Mitochondrial activity in
XTC.UC1 cells derived from thyroid oncocytoma. Thyroid 11:327–333
24. Lamy FM, Rodesch FR, Dumont JE 1967 Action of thyrotropin on thyroid
energetic metabolism. VI. Regulation of mitochondrial respiration. Exp Cell
Res 46:518 –532
25. Ricquier D, Bouillaud F 2000 Mitochondrial uncoupling proteins: from mitochondria to the regulation of energy balance. J Physiol 529:3–10
26. Arsenijevic D, Onuma H, Pecqueur C, et al. 2000 Disruption of the uncoupling
protein-2 gene in mice reveals a role in immunity and oxygen species production. Nat Genet 26:435– 439