Department of Otolaryngology-Head and Neck Surgery, Head and Neck Cancer Research Division, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Department of Otolaryngology-Head and Neck Surgery, Head and Neck Cancer Research Division, 720 Rutland Avenue, Johns Hopkins University School of Medicine, 818 Ross Research Building, Baltimore, MD 21205-2196
Human mitochondrial DNA (mtDNA) is a 16.6 kb circular double-stranded DNA molecule, which is present at a high copy number per cell, and the number varies widely with cell type. mtDNA encodes 13 polypeptides involved in respiration and oxidative phosphorylation, 2 rRNAs and a set of 22 tRNAs that are essential for protein synthesis in the mitochondria.1 Mitochondria have also been shown to play an important role in apoptosis, a fundamental biological process by which cells die in a controlled manner.2 Apoptosis plays a critical role in cancer development and in cellular response to anticancer agents. It is important to understand the biology of mitochondria and its contribution to tumorigenesis.
The replication of mtDNA is not synchronized with nuclear DNA replication, and the overall number of mitochondria per cell remains fairly constant for specific cell types during proliferation. Mitochondrial defects have long been suspected to play an important role in the development and progression of cancer.3 Over half a century ago, Warburg proposed that cancer cells undergo mitochondrial respiration alterations, and proposed a mechanism to explain how they evolve during the carcinogenic process.4 Since then, a number of cancer-related mitochondrial defects have been identified and described in the literature.5, 6, 7, 8 These defects include altered expression and activity of respiratory chain subunits and glycolytic enzymes, decreased oxidation of NADH-linked substrates, mtDNA mutations as well as altered mtDNA content. Many solid tumors have changes in the energy metabolism that are often associated with a reduction of the mitochondrial DNA content and the activity of enzymes of oxidative phosphorylation.9, 10, 11 This reduction in mtDNA levels is associated with the induction of nuclear DNA and mtDNA encoded oxidative phosphorylation genes,12 perhaps indicating a compensatory mechanism for responding to a reduced energy metabolism state. On the contrary oncocytes are known to accumulate a large number of mitochondria.11, 13, 14 This increase may be due to lack of coordination between the nuclear and mitochondrial genomes.14, 15 Other recent studies suggest, however, that increase in mtDNA content may also be due to a compensatory mechanism.1, 2, 3
There is increasing evidence that mitochondrial alterations are associated with various cancer types, although it is not clear whether these alterations are contributing factors in carcinogenesis or whether they simply arise as part of secondary effects in cancer progression.16, 17, 18 To determine whether there is an association between tumor progression and mtDNA content changes, we used real-time quantitative PCR to analyze mtDNA content in paired normal and tumor samples from breast and thyroid tissue samples. We show that mtDNA content is decreased in breast tumors, but increased in thyroid tumor samples. The relationship between mtDNA content and tumor aggressiveness is further discussed.
Material and Methods
Tissue samples and DNA isolation
Paired primary invasive breast carcinomas and adjacent normal tissues (frozen tissue), were obtained from the Surgical Pathology archives of The Johns Hopkins Hospital (Baltimore, MD) according to the Institutional Review Board protocol and DNA was isolated using standard phenol-chloroform protocol. Prof. Saraswati Sukumar at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University provided isolated DNA. Human thyroid tissues were obtained from Drs. G. Tallini, B. Haugen, and S.-C.J. Yeung and micro-dissected from fresh surgical samples or paraffin-embedded blocks as described previously.19 For paraffin-embedded tissues, samples were first treated with xylene for 8 hr at 48°C to remove the paraffin before proteinase K digestion. DNA was subsequently isolated from the digested tissues by phenol-chloroform extraction and ethanol precipitation.
Analysis of mtDNA content by quantitative real-time PCR
The 7900HT sequence-detection system (Applied Biosystems, Foster City, CA) was used to amplify β-actin and mtDNA coding region, cytochrome c oxidase (Co I). Table I lists the primers and probes used to amplify the respective DNA regions. All primers were obtained from Invitrogen (Carlsbad, CA), and the mitochondrial Co I TaqMan probe (Applied Biosystems, Foster City, CA) was labeled with 5′-FAM (6-carboxyfluorescein, fluorescent reporter) and 3′-TAMRA (6-carboxy-tetramethylrhodamine, fluorescence quencher) whereas the β-actin probe was labeled with 5′-Vic (fluorescent reporter), and 3′-TAMRA (quencher). Duplex PCR amplifications Co I and β-actin were carried out in buffer containing 16.6 mM ammonium sulfate, 67 mM Tris base, 2.5 mM MgCl2, 10 mM 2-mercaptoethanol, 0.1% DMSO, 0.2 mM each of dATP, dCTP, dGTP, and dTTP, 600 nM each of forward and reverse primers, 200 nM TaqMan probe, 0.6 U Platinum Taq polymerase, and 2% Rox reference dye.20 DNA (2 ng) was used to amplify mitochondrial region and nuclear β-actin gene. The real-time PCR reactions were carried out in triplicate for each gene. Standard curves were obtained by using cell line DNA. The mtDNA content index was given as a ratio of mtDNA/nuclear DNA that was calculated by dividing the mitochondrial Co I signal for each gene by the corresponding β-actin signal.
Table I. Sequence of Primers and Probes used in the Quantitative Real-time PCR Analysis
Forward primer (5′-3′)
Reverse primer (5′-3′)
TaqMan probe (5′-3′)
Quantitative real-time PCR assay allows for the measurement of DNA in individual amplifiable DNA segments. The use of fluorescent probes in this assay allows for the detection of small differences in the available starting template using only nanogram amounts of DNA. Higher threshold cycle number (CT) or a shift of the amplification curve to the right indicates a low amount of starting template (decreased DNA content). Ratios of mtDNA/nuclear DNA were used to obtain the mtDNA content index, whereby a lower ratio represents less initial template, denoting a decrease in the amount of mtDNA. We selected the nuclear β-actin gene and mtDNA coding region, cytochrome c oxidase I (Co I). mtDNA content was analyzed by calculating the mtDNA/nDNA ratio after normalizing each mitochondrial signal to the corresponding nuclear β-actin signal. Standard curves for mtDNA Co I (Fig. 1a) were generated using H1299 cell line DNA (range = 10 pg–10 ng). The correlation coefficient (R2) of >0.99) provided evidence of linearity over the range of template concentration used. The ΔCT of 3.36 matched the 10-fold dilution from 10,000 pg to 1,000 pg where fold change is given by 2⁁ΔCT. We used quantitative real time PCR to analyze mtDNA content in paired normal and tumor breast and thyroid samples as well as DNA from none-paired thyroid samples. Table II shows the diagnosis of the patient samples used in our study and the average mtDNA content index for each group. Representative quantitative real-time PCR amplification curves are shown for mtDNA Co I gene (Fig. 1b) and β-actin (Fig. 1c) for a paired breast tumor and normal DNA from the same individual. Figure 1c shows a shift to the right by 0.5 cycles in the tumor breast sample compared to the normal β-actin. This shift corresponds to 1.4 (2⁁0.5) fold higher β-actin in normal. Figure 1b shows that amplification of the mitochondrial Co I gene was shifted to the right by about 3 cycles in the tumor, corresponding to a 8-fold decrease in mtDNA content in the tumor sample. After normalizing to β-actin, the effective decrease in mtDNA content in tumor was 5.6-fold (8/1.4). We also did not observe a significant difference between the β-actin amplification in tumor and normal samples.
Table II. Sample Profile Included in this Study
Average mtDNA content index
Paired normal tissue
Benign thyroid neoplasms
Non-paired papillary thyroid carcinoma
Paired papillary thyroid carcinoma
Figure 2 shows the mtDNA content in 57 papillary thyroid carcinoma samples (PTC), 21 benign thyroid neoplasms (BTN) and 20 controls (N). The mtDNA content in these subjects varied widely. The mean mtDNA content index was 0.87 in the papillary thyroid carcinoma (PTC), 0.64 in the benign thyroid neoplasms (BTN), and 0.45 in normal thyroid tissue (N), indicating increased mtDNA content in the PTC samples, with a p-value of 0.00021 (PTC vs. normal). We further analyzed the mtDNA content in 20 paired tumor and normal DNA from patients with PTC (Fig. 3). Our results showed that of the 20 pairs, 13 of 20 (65%) had increased mtDNA content in the tumor samples, 5 of 20 (25%) harbored less mtDNA content in the tumor and 2 of 20 (10%) did not show any difference between the normal and the tumor DNA. The mean mtDNA content index ratio was 0.45 for the normal and 0.89 for the PTC samples. We also analyzed the mtDNA content in 25-paired normal and primary tumor breast tissue DNA (Fig. 4). The majority (20 of 25; 80%) of the paired breast samples showed a reduction in mtDNA in the tumor DNA when compared to the corresponding normal. Contrary to the thyroid tumors, only 16% (4 of 25) of the breast pairs had increased mtDNA in tumor, whereas 1 of 25 (4%) showed no difference between the tumor and the corresponding normal. The mean mtDNA content index ratio was 0.80 for the normal and 0.58 for tumor samples (p = 0.012).
We found that mtDNA content varied widely depending on the type of tumor and sample. mtDNA alterations have been suggested as a potentially sensitive and specific biomarker for several cancer types. To that effect, mtDNA mutations have been reported for specific cancers,5, 21, 22 and were detected in bodily fluids,5 suggesting that monitoring mtDNA changes could provide a potentially sensitive method for the early detection and monitoring of cancer. We show in our studies that mtDNA content was reduced (though modestly) in 20 of 25 (80%) of breast tumors compared to their normal breast tissue taken form the same individual. No significant differences were observed between β-actin amplification in the tumor and normal samples, suggesting that the observed differences in mtDNA content were not due to β-actin input or amplification. To our knowledge, there are currently no studies that have examined mtDNA content in breast tumor tissue or bodily fluids from breast cancer patients. This is the first study indicating that breast carcinogenesis may involve a decrease in mtDNA content. Reduced mtDNA content has been reported in other cancer types. Specifically, renal carcinomas were reported to have reduced mtDNA content whereas renal oncocytomas harbored increased mtDNA content.11 The decrease of mtDNA in renal carcinomas was associated with a 5–10-fold decrease in mtDNA transcripts relative to control kidney samples,11 suggesting that mitochondrial transcript levels depend on mtDNA content. Similarly, Selvanayagam et al.,23 also reported a decrease of mtDNA and the mRNA coding for NADH dehydrogenase subunit 3 in 8 of 13 (61%) kidney tumor tissues obtained from patients with renal cell carcinoma. This phenomenon was also observed in 5 of 6 renal carcinoma cell lines. Recently, Meierhofer et al.24 reported a decrease in mtDNA content in 34 of 37 (91%) of renal cell carcinomas compared to control kidneys. The frequency with which this phenomenon occurs in renal cell carcinomas suggests that this may be an important phenotype associated with renal cell neoplastic transformation.
Other none-urological tumors like hepatomas have been reported to have large reductions in mitochondrial content25, 26 despite showing a paradoxical increase in the cellular abundance of transcripts encoding mitochondrial proteins.26 Cuezva et al.27 also showed that carcinogenesis in the liver was associated with a depletion of the cellular mitochondrial content. Yin et al.28 also reported reduced mtDNA copy number and the content of mitochondrial respiratory proteins in hepatocellular carcinoma (HCC) as compared to the corresponding normal liver. In those studies, mtDNA copy number was significantly reduced in female HCC but not in male HCC, suggesting that reduced mtDNA copy number and impaired mitochondrial biogenesis are important events during carcinogenesis of HCC, and further that mtDNA content may be regulated in a gender specific manner. Yin et al.28 further suggest that the differential alterations in mtDNA content of male and female HCC may contribute to the differences in the clinical manifestation between female and male HCC patients. Similar to liver and renal cancers, we propose that a decrease in mtDNA may confer a growth advantage in breast cancers, which are commonly characterized by hypoxic tissue.29 Changes in mtDNA levels did not correlate with tumor grade and metastasis, suggesting that these alterations may occur in the early stages of breast tumorigenesis.
One common feature of many cancer cells is their abnormal bioenergetics, characterized by increased glycolytic capacity accompanied by an impaired respiration.30 This abnormality has been ascribed to a marked deficit in the cellular content of mitochondria coupled with a decrease in the activities of oxidative phosphorylation enzyme complexes.24, 30 Simonnet et al.9 reported a correlation of renal cell carcinoma aggressiveness with a decrease in content of oxidative phosphorylation complexes. Furthermore, decreased mtDNA content may result in decreased oxidative phosphorylation capacity that in turn may favor faster growth or increased invasiveness as suggested in studies by Simonnet et al.9 In general, decreased mitochondrial activity seems to be an adaptation to environmental conditions of solid tumors, which have to endure hypoxia during their development. Low mitochondrial activity leads to lower oxidative stress under hypoxic conditions and might therefore represent an advantage for carcinoma progression.
Unlike the breast cancer results, we observed that thyroid tumors had increased mtDNA content in paired as well as none-paired samples. It has been shown that patterns of mitochondrial transcripts coding for proteins involved in oxidative phosphorylation differ depending on the tissue type or the origin of the tumor.11 Mitochondrial proliferation has also been reported in mitochondrial diseases associated with respiratory chain defects or coupling defects between the respiratory chain and ATP production.31, 32, 33, 34 The proliferation of mitochondria in thyroid tumors might result from the induction of genes involved in mitochondrial biogenesis. It has also been shown that thyroid oncocytomas are characterized by cells containing abundant but morphologically altered mitochondria.35, 36 Similarly, salivary gland oncocytoma have increased mtDNA content that is paralleled with an increase in mtDNA transcripts.11 Other findings from our laboratory also suggest an increase in mtDNA content in primary tumors and in saliva from advanced tumors in head and neck squamous cell carcinoma patients (unpublished results). Other previous studies from our laboratory have shown an increased abundance of mutated mtDNA in lung cancer, suggesting an enrichment of mutated mtDNA and increased mtDNA copy number in those specific lung cancer samples.5 Our current findings and previous results indicate that thyroid, head and neck, and perhaps lung cancers may be characterized by increased mtDNA content. It is interesting to note that these cancers are usually influenced by external/environmental factors, whereas breast cancers are not. Mitochondrial proliferation found in the thyroid tumors cells may be explained as compensation for defective ATP synthesis often associated with these tumors. A decreased rate of ATP synthesis has recently been documented in thyroid oncocytoma37 despite a 3- to 4-fold increase in mtDNA and mitochondrial transcripts. We did not find any correlation between tumor stage and increased mtDNA in paired thyroid samples. Thyroid samples of advanced stage harbored increased mtDNA content, suggesting that alterations in mtDNA content might be an early event in carcinoma formation. Increased mtDNA may cause increased mitochondrial biogenesis attempting to compensate for the loss of oxidative phosphorylation function. The elevated mitochondrial biogenesis might be similar to the often observed up-regulation in the muscle tissue of patients suffering from mitochondrial myopathies.38
To elucidate the relationship between tumorigenesis and the mitochondrial DNA alterations, we analyzed mtDNA content in paired normal and tumor samples from breast and thyroid tissue samples. A significant reduction of mtDNA content was detected in breast tumors, whereas thyroid tumors exhibited increased mtDNA content. mtDNA changes were not correlated with aggressive disease in both breast and thyroid cancers, suggesting that mtDNA alterations may occur early during tumorigenesis. Our data indicate that changes in mtDNA content may be regulated in a tumor specific manner during carcinogenesis. Prospective studies will examine mtDNA content changes in bodily fluids of cancer patients to elucidate the role of mtDNA content in none-invasive disease monitoring.
The authors thank Dr. Z. Guo for insightful discussions and help during the preparation of this manuscript and Dr. D. Goldenberg for help with the clinical information database.