Mitochondria and Cancer

Systemic oxidative stress is associated with a wide range of pathological conditions. Oxidative DNA damage is frequently measured incirculatinglymphocytes.MitochondrialDNA(mtDNA)isknowntobemoresensitivetooxidativedamagethannuclearDNAbutisrarelyusedfordirectmeasurementofDNAdamageinclinicalstudies.Basedonthesupercoiling-sensitivereal-timePCRmethod,weproposeanewapproachforthenoninvasivemonitoringofsystemicoxidativestressbyquantifyingthemtDNAstructuraldamageandcopynumberchangeinisolatedlymphocytesinasingletest.Weshowthatlymphocyteshavesigni�cantlylessmtDNAcontentandrelativelylowerbaselinelevelsofdamagethancancercelllines.Inan ex vivo challenge experiment, we demonstrate, for the �rst time, that exogenous H 2 O 2 induces a signi�cant increase in mtDNA damage in lymphocytes from healthy individuals, but no repair activity is observed aer 1h recovery. We further demonstrate that whole blood may serve as a convenient alternative to the isolated lymphocytes in mtDNA analysis. us, the blood analysis with the multiple mtDNA end-points proposed in the current study may provide a simple and sensitive test to interrogate the nature and extent of systemic oxidative stress for a broad spectrum of clinical investigations. Mitochondrial DNA (mtDNA) mutations have been found in many cancers but the physiological derangements caused by such mutations have remained elusive. Prostate cancer is associated with both inherited and somatic mutations in the cytochrome c oxidase (COI) gene. We present a prostate cancer patient-derived rare heteroplasmic mutation of this gene, part of mitochondrial respiratory complex IV. Functional studies indicate that this mutation leads to the simultaneous decrease in cytochrome oxidation, increase in reactive oxygen, and increased reactive nitrogen. ese data suggest that mitochondrial DNA mutations resulting in increased reactive oxygen and reactive nitrogen generation may be involved in prostate cancer biology. Alterations in the mitochondrial genome have been chronicled in most solid tumors, including breast cancer. e intent of this paper is to compare and document somatic mitochondrial D-loop mutations in paired samples of ductal carcinoma in situ (DCIS) and invasive breast cancer (IBC) indicating a potential breast ductal epithelial cancerization �eld eﬀect. Paired samples of these histopathologies were laser-captured microdissected (LCM) from biopsy, lumpectomy, and mastectomy tissues. Blood samples were collected as germplasm control references. For each patient, hypervariable region 1 (HV1) in the D-loop portion of the mitochondrial genome (mt�enome) was sequenced for all 3 clinical samples. Speci�c parallel somatic heteroplasmic alterations between these histopathologies, particularly at sites 16189, 16223, 16224, 16270, and 16291, suggest the presence of an epithelial, mitochondrial cancerization �eld eﬀect. ese results indicate that further characterization of the mutational pathway of DCIS and IBC may help establish the invasive potential of DCIS. Moreover, this paper indicates that bio�uids with low cellularity, such as nipple aspirate �uid and/or ductal lavage, warrant further investigation as early and minimally invasive detection mediums of a cancerization �eld eﬀect within breast tissue. Intrinsic oxidative stress through increased production of reactive oxygen species (ROS) is associated with carcinogenic transformation, cell toxicity, and DNA damage. Mitochondrial DNA (mtDNA) is a natural surrogate to oxidative DNA damage. MtDNAdamageresultsinthelossofitssupercoiledstructureandisreadilydetectableusinganovel,supercoiling-sensitivereal-timePCRmethod.OurstudieshavedemonstratedthatmtDNAdamage,asmeasuredbyDNAstrandbreaksandcopynumberdepletion,isverysensitivetoexogenousH 2 O 2 but independent of endogenous ROS production in both prostate cancer and normal cells. In contrast, aggressive prostate cancer cells exhibit a more than 10-fold sensitivity to H 2 O 2 -induced cell toxicity than normal cells, and a cascade of secondary ROS production is a critical determinant to the diﬀerential response. We propose a new paradigm to account for diﬀerent mechanisms governing cellular oxidative stress, cell toxicity, and DNA damage with important rami�cations in devising new techniques and strategies in prostate cancer prevention and treatment. p55PIK, regulatory subunit of class IA phosphatidylinositol 3-kinase (PI3K), plays a crucial role in cell cycle progression by interaction with tumor repressor retinoblastoma (Rb) protein. A recent study showed that Rb protein can localize to the mitochondria in proliferative cells. Aberrant p55PIK expression may contribute to mitochondrial dysfunction in cancer progression. To reveal the mechanisms of p55PIK transcriptional regulation, the p55PIK promoter characteristics were analyzed. e data show that myeloid zinc �nger 1, MZF1, is necessary for p55PIK gene transcription activation. ChIP (Chromatin immunoprecipitation) assay shows that MZF1 binds to the cis-element “TGGGGA” in p55PIK promoter. In MZF1 overexpressed cells, the promoter activity, expression of p55PIK, and cell proliferation rate were observed to be signi�cantly enhanced. Whereas in MZF1-silenced cells, the promoter activity and expression of p55PIK and cell proliferation level was statistically decreased. In CRC tissues, MZF1 and p55PIK mRNA expression were increased ( 𝑃𝑃 𝑃 𝑃𝑃𝑃𝑃𝑃 , 𝑃𝑃 𝑃 𝑃𝑃𝑃𝑃𝑃 , resp.). A strong positive correlation ( 𝑅𝑅𝑅𝑅 𝑃 𝑃𝑃𝑅𝑃 ) between MZF1 and p55PIK mRNA expression was observed. Taken together, we concluded that p55PIK is transcriptionally activated by MZF1, resulting in increased proliferation of colorectal cancer cells.

e future of mitochondrial DNA research has the potential to uncover new insights on genetic diseases and open new opportunities to discover ways to control mitochondria and their in�uence on the human health and cancer. e outcomes of this work will expand the understanding of cellular respiration and disease risk. In this special issue, we give examples of strategies for the measurement of oxidative stress; a critical factor in tumor progression. While the mitochondrial genome has been well characterized, the associations of the broad spectrum of mitochondrial genotypes remains a relatively rich �eld of study.
Genetic tools are beginning to be realized which characterize mitochondrial populations and link their associations with normal and malignant cells. ere are many largescale deletions which require further investigation. ere are populations of genotypes that rise and fall with tissue �eld effects. Mutations in the mitochondrial genome can alter the cellular biochemical behavior, changing the conditions for potential tumor growth. In this special issue we also address roles of mitochondrial interactions which are central to the physiological processes involved in malignant transformation. Measurements of DNA damage associated with prostate and other cancers can be normalized by comparative measurements of mitochondrial subpopulations. is can be used to assess DNA damage and somatic mutations under physiological and pathological conditions and can serve as a strategy to measure cell toxicity as a guide for devising innovative cancer preventions and treatments.
On account that mitochondrial DNA is accessible across various tissues, noninvasive collection and analyses are possible. From such investigations, it has been demonstrated that there is a progression of change in mitochondria through the tissues, as a �eld effect, that is, associated with the tumor tissue progression. Tumorigenic effects related to increasing ROS are well known in prostate and other solid tumor cancers.
Mapping mutations across the entire mitochondrial genome are fundamental to the future work on mitochondrial "omic" investigations. Mitochondrial whole genome sequencing pioneered the concepts of conducting whole genome analyses to understand forensics. is has resulted in efforts to go beyond simple STR typing and to type the entire chromosome. It is because of the increased resolution achieved by sequencing whole metagenomes that, by extension, other �elds of diagnostics, personalized medicine, and bacterial and viral forensics have emerged (J. P. Jakupciak unpublished data).
rough the study of mitochondria, mechanisms of cancer are emerging. e in�uence of mitochondria on the metastatic potential of cancer cell lines points to a promising future and in vivo characterization of populations of mitochondria will function as a "looking-glass" into monitoring the modulation events and even predicting changes in metastatic capabilities. Mitochondrial genomics is poised to enhance the over all �eld of omics and contribute signi�cantly to the advent of personalized medicine �1].
In this special issue, the authors present some of the latest �ndings in this exciting and rapidly expanding area of genomic research: (i) speci�c heteroplasmic somatic alterations in the mitochondrial genome contribute to the cell proliferation; (ii) a new paradigm for oxidative stress and cell and DNA damage has important implications for both cancer prevention and treatment; (iii) an assay for gauging systemic oxidative stress using peripheral blood; (iv) upregulation of a nuclear gene whose molecular interactions contribute to mitochondrial dysfunction, promoting cell proliferation; (v) a cancerization �eld e�ect described by progressive mitochondrial mutations in noninvasiveand invasive breast cancer.

Past
As far back as 1850 scientists identi�ed the existence of structures within cells that today we call mitochondria [1]. However, it was not until 1898 that the these structures were given the term mitochondria by Carl Benda [2]. Cytologists worked hard to identify the function of mitochondria and in 1912 the �rst reference to a possible link between mitochondria and respiration was made by Kingsbury [2]. is link was made exclusively from morphological observations. What followed was 30-40 years of intense biochemical analyses before the characterization of mitochondria as the "Powerhouse of the Cell" by Siekevitz in 1957 [3]. Leading up to this there were a number of key events. In 1909 Correns and Baur independently identi�ed the �rst cases of extracellular inheritance. e source and site of this was unknown at the time, though mitochondria were the prime suspect. Ephrussi's laboratory had been working with yeast and their key publication in 1949 [4] used genetic analysis to show that respiration-de�cient baker's yeast harboured mutations found in the cytoplasm, and not the nucleus. Soon thereaer Slonimski and Ephrussi investigated this area further and showed the de�ciency was due to mitochondrial dysfunction.
On the back of these exciting �ndings, one of the most important discoveries in mitochondrial research was made in 1963, when the identi�cation of the existence of mitochondrial DNA (mtDNA) was made by M. M. Nass and S. Nass [5]. Using electron microscopy they showed conclusively that chick embryo mitochondria contained DNA. e importance of this discovery cannot be overstated; as it renewed interest in the evolutionary origin of mitochondria. ese �ndings were con�rmed biochemically in 1964 when Schatz and Klima [6] showed that baker's yeast mitochondria also contained DNA. is of course led to more questions, speci�cally does the mtgenome interact with the nuclear genome, and if so how does this occur? e mtDNA sequence was �rst published as being 16,569 base pairs long in 1981 by Anderson et al. [7], the sequence was later revised by Andrews et al. in 1999 [8]. Following the publication of the mtDNA sequence there was a focus on mitochondrial genomics that has been sustained till today. is work initially focused on myopathies and neuropathies by Wallace et al. [9,10]. ese so called "mitochondrial diseases" were due to both mitochondrial and nuclear mutations, where a symbiotic relationship exists between the mitochondrial and nuclear genomes. Understanding the complexity of these interactions is the key to detecting dysfunction within the cell.
Mitochondria control various metabolic functions and synthesizes 95% of cellular metabolic energy, while 1,200 nuclear genes drive and participate in mitochondrial function. ere are 37 genes coded for by the mtgenome, 24 of which are dedicated to processing 13 genes within the mtgenome (mtgenome) itself which produce the subunits essential to electron transport. ese 13 key genes work in conjunction with 93 nuclear proteins. In cancer cells certain mutations in the mtgenome can alter the biochemical behaviour of mitochondrial/nuclear protein complexes, thereby increasing pools of reactive oxygen species (ROS) which in turn enable tumour growth and may provide proliferative advantage to the cell [11]. us despite its small size relative to the nuclear genome, somatic mutations that occur in the mtgenome are able to contribute directly to the process of tumourigenesis.
ere is now a signi�cant body of literature describing the interactions between mitochondria and the nucleus. It appears that the somatic mutations which can alter these interactions occur early in the disease process, appearing in histologically normal tissues [12]. is leaves us with a key question with regard to mitochondrial mutations, are they causative factors of disease or a simple record of the development of disease ? In 1924 Otto Heinrich Warburg postulated that cancer and tumour growth are in part caused by a change in the way the cells generate their energy. In normal healthy cells, ATP is generated mainly by oxidative breakdown of pyruvate following non-oxidative breakdown of glucose during the process of glycolysis. In contrast, malignant cell metabolism stalls aer glycolysis a phenomenon that Warburg reported as the fundamental difference between normal and cancerous cells. ese observed differences in the ratio of glycolysis to respiration have become known as the Warburg effect [13]. It has since become clear that these metabolic differences adapt cancer cells to the hypoxic conditions inside solid tumours. us it may be theorized that rather than causing cancer, perhaps these changes are characteristics produced by key cancer-causing mutations in certain genes involved in the aforementioned symbiotic relationship between nucleus and mitochondria. ese �ndings have led to a new theory known as the Warburg theory of cancer, which suggests that the main driver of tumorigenesis is an insufficient cellular respiration caused by insult to mitochondria [14].
Clearly mitochondria are key to the function of both normal and malignant cells, and there has been much speculation about the origins of mitochondria, with perhaps the most prevalent theory being that of an endosymbiotic origin. Mitochondria have many features in common with prokaryotes, and this theory is generally accepted today. At its very basic level, the endosymbiotic theory hypothesises that mitochondria, chloroplasts and perhaps other organelles of eukaryotic cells, originated as independent organisms which were taken inside what became eukaryotic cells as endosymbionts. It has been suggested that mitochondria as we know them today developed from proteobacteria, speci�cally the SAR11 lineage [15]. Although the endosymbiotic theory has been around for over one hundred years, it remains a controversial and developing hypothesis with new evidence for and against the theory still appearing today. Although beyond the scope of this publication, this controversy certainly highlights mitochondria as a hot topic for research in many diverse �elds of study.

Present
Mitochondria play a central role in the regulation of cellular function, metabolism, and cell death in cancer cells. Several important functional changes to cancer cell mitochondrial have been observed that implicate the organelle in tumour formation including increased production of reactive oxygen species (ROS), decreased oxidative phosphorylation, and a corresponding increase in glycolysis [16,17]. However, the speci�c role of mitochondria in tumourigenesis remains unclear as these changes could represent either key mechanisms in tumour initiation, promotion, or simply secondary effects of tumourigenesis. In 2011, deregulated cellular energetics was considered as an additional emerging hallmark of cancer [18]. Cancer cell signaling that is regulated by kinases and phosphatases are guided by cellular redox status and may be a key in malignant transformation. is section brie�y examines the role that mitochondria play in the cancer cell phenotype by relating the physiological process of the organelle to genomic and proteomic studies.

Role of Mitochondria in Oncogenesis.
Recent studies of mitochondrial involvement in cancer have uncovered a plethora of differences in the structure and function of these organelles upon comparing metastatic mitochondria to those belonging to nontransformed cells. Notably, modern research has largely upheld the metabolic observations of Warburg and his successors, while re�ning and greatly expanding the breadth of mechanistic knowledge of mitochondrial state and function in tumour development. e comprehensive mechanisms of the Warburg effect have not yet been isolated; however, multiple intertwining causative and responsive mechanisms have recently been characterized. is understanding of the indicative features of cancer cell metabolism has also directly been applied to current clinical care through the increasingly widespread adoption of positron emission tomography (PET) imaging using glucose analogues to identify cancerous lesions that are characterized by high glucose uptake [19].
Study of the mechanisms of the Warburg effect has revealed that the characteristic metabolic shi towards aerobic glycolysis and increased glucose uptake imparts several BioMed Research International 3 functional advantages to the cancer cell. ese advantages permit rapid growth and survival in conditions that would be potentially lethal to noncancerous cells. Perhaps the most signi�cant shi� in the understanding of the �arburg effect in recent years has been the abandonment of the view that aerobic glycolysis is a metabolic defect of cancer cells, in favor of the theory that cancer cell metabolism is maintained through regulatory control, and better �ts the metabolic pro�le of rapidly dividing cells [20]. e most recognized of these adaptations is the utilization of abundant glycolytic pathway intermediates in multiple anabolic reactions critical to the survival and growth of rapidly dividing cells. e demand for glucosederived carbon skeletons for macromolecule synthesis of molecules such as glycogen, phospholipids, triglycerides, and malate, exceeds the demand for efficient ATP production [21]. Glucose is also alternatively metabolized in cancer cells through an enhanced pentose phosphate pathway that results in the synthesis of nucleotides and antioxidant nicotinamide adenine dinucleotide phosphate (NADPH) [20].
e cancer cell's lack of reliance on oxidative phosphorylation for ATP generation also permits cellular survival in conditions of inconsistent oxygen supply, an environment that is typical for rapidly expanding tumours which can, at times, experience inadequate angiogenesis [22]. Local acid-i�cation of the tumour microenvironment is also induced through the glycolytic generation of excess bicarbonic and lactic acids. e resultant pH change is recognized to favour tumour growth and invasion through the activation of cancer cell-derived cathepsins and metalloproteinases [23], and the inhibition of the subsequent host immune responses [24]. Additionally, the ensuing excessive production of lactate can be converted to pyruvate in cancer-associated stromal cells to fuel oxidative phosphorylation within these cells [25].
It is also well established that the enhanced production of mitochondrial ROS, most notably superoxide, hydroxyl radicals, and hydrogen peroxide, is a prominent byproduct of cancer cell metabolism. Increased oxidative phosphorylation in pre-metastatic cells therefore increases the production of mitochondrial ROS, which may be an initiative factor in carcinogenesis [26]. e mechanisms of ROS production and their signi�cant downstream effects have become important topics in current mitochondrial research in cancer. Excess ROS act not only as mutagens and initiators of oxidative stress, but are also signi�cant inter-and intracellular signaling molecules, responsible for a host of nuclear and mitochondrial changes in gene expression, the details of which are reviewed by Verschoor et al. [27].
e well-established addiction to glutamine as an energy source of proliferating cancer cells is yet another key hallmark of cancer cell metabolism. In the cytoplasm, glutamine is converted to glutamate by glutaminase, and transported into the mitochondria where it is converted to the TCA cycle intermediate -ketoglutarate, and also acts as a source of substrates for macromolecule synthesis [28]. Glutamate is also a key substrate of the glutathione-dependent antioxidant system that is the primary intracellular antioxidant mechanism, and that is critical to cellular protection from ROS. Recent studies have also shown that glutamine is an important component in several signaling pathways involved in cell growth including mTOR and ERK pathways, and glutamine uptake and degradation are controlled via c-myc regulation [29].
e increased mitochondrial ROS production in metastatic cells has also been associated with the corresponding upregulation of cellular antioxidant defense mechanisms [30]. In particular, the synthesis of GSH is signi�cantly increased by the enhanced uptake of ratelimiting cystine through the action of the cystine/glutamate antiporter system − , which is frequently upregulated in cancer cells [31]. is increased antioxidant capacity has been implicated in cellular resistance to chemotherapy and radiation therapy that induce cancer cell death by initiating oxidative stress [32]. In addition, it was recently found that the abundant glutamate excreted as a byproduct of cystine uptake by system − in cancer cells may also initiate several signi�cant pathologies of metastatic tumours including excitotoxic cell death in tumours of the CNS, and disruption of bone cell signaling in metastatic tumours in the bone [33,34].

Genomics of Mitochondrial DNA: Mutations and
Polymorphisms. e circular mtgenome encodes 37 genes including several components of the electron transport chain (ETC), tRNAs, and rRNAs. Additionally, mtDNA contains a noncoding region comprised of two hypervariable regions within a displacement loop (D-loop), which is the location of the origin of replication and transcriptional promoters. Mutations in mtDNA are frequently observed in cancer, likely due to the lack of introns, lack of histone protection, and close proximity to damaging ROS. Each cell contains multiple copies of mitochondrial genes, giving rise to mitochondrial homoplasmy, where all the mitochondria of a cell have the same genomic composition, or heteroplasmy, where wildtype and mutant mtDNA coexist [35]. us is it possible for a mutation that confers a distinct advantage for cancer cells, such as accelerated growth or enhanced survival, to be clonally expanded to become a homoplasmic mutant and to predominate within a population of cancer cells. Alternatively, Coller and colleagues [36] used a mathematical model to show that random segregation of mtgenomes during rapid tumour development could result in a mutant homoplasmic population without the need for a selective advantage. Regardless of the existence of background homoplasmic mutations that confer no functional consequence, there are numerous mtDNA mutations that result in signi�cant alterations in mitochondrial function that affect tumour development and progression.
In certain cancer tissues mtDNA mutations were more readily detectable and abundant than mutated nuclear p53 DNA, suggesting that mtDNA mutations could serve as excellent cancer biomarkers, particularly for early detection [35]. e most commonly mutated or deleted region of mtDNA in cancer is within the D-loop at the D310 tract, which is a mononucleotide cytidine repeat at position 310 [37]. As the D-loop is involved in mitochondrial replication, mutations in this region could also affect mtDNA copy number, though this theory has yet to be proven empirically.
In one study, colorectal cancer patients with D-loop mutations were found to have signi�cantly lower overall survival rates and increased chemotherapeutic resistance compared to patients who's mtDNA did not harbour such mutations [38]. e high frequency of D-loop deletion or insertion somatic mutations in cancer render these mutations unlikely to confer any functional impairment to mitochondria, and so the uncertain functional consequences of these mutations should remain an important area for mitochondrial research in cancer.
e importance of mitochondrial polymorphisms in cancer development and risk is intimately related to evolutionary haplogroups, and has recently been a contentious area of research. Haplogroups are characterized by a speci�c mutation that occurs widely within individuals of a particular population, and are further divided into haplotypes generally based on restriction fragment length polymorphisms [39]. Among the main European haplotypes, the A12308G mutation in tRNA Leu2 common to haplotype U was associated with increased risk of both renal and prostate cancers [40]. e NADH-ubiquinone oxioreductase chain 3 (ND3) substitution mutation at G10398A has been associated with increased breast cancer risk in both African American and Indian women [41][42][43][44][45]. In European-American women the A10398G ND3 substitution conferred increased risk of breast cancer, as did the T16519C D-loop polymorphism [46]. A comprehensive study of pancreatic cancer risk revealed associations with the A331T substitution in mitochondrial ND2 [47]. Despite these promising �ndings, and because the majority of mtDNA polymorphisms are functionally inconsequential, associations with speci�c polymorphisms and cancer risk have been subject to heated debate. Several older studies involving association of speci�c polymorphisms with cancer risk have been heavily scrutinized due to erroneous experimental design, interpretation, and poor data quality [35]. However, due to the potential usefulness of somatic mtDNA mutational pro�ling as a diagnostic tool, the study of mitochondrial somatic mutations and associations with cancer should remain an important focus of cancer biomarker research pending proper study design, population strati�cation, and independent replication of results. Interestingly, one study reports that one well characterized pathological mtgenome alteration, A3243G, drives mtDNA depletion [48].
Mitochondrial depletion, a hallmark of cancer initiation and malignant development, is characterized by a wide range of mtDNA deletions [49,50]. In prostate cancer, a cascade of both large and small-scale deletions reduce cellular mtDNA. is reduction is associated with androgen independence which facilitates disease progression [51]. Consistent with these �ndings, a 3.4kb mtgenome deletion is currently being used by many urologists to identify the presence and/or absence of prostate cancer in patients with an initial benign biopsy [52,53]. Contrary to a negative outcome these patients remain highly suspicious for disease by other clinical parameters.

Mithondrial Genome Sequencing.
Homoplasmic and heteroplasmic mutations have been reported in the mtgenomes of patient tumors [54], and improved patient outcomes have been demonstrated using mtDNA mutation identi�cation for early detection of solid tumours [55]. Clinically, the detection of mtDNA mutations could be reliably used to compare differences in healthy and cancerous tissues, used to monitor mutations in high-risk, asymptomatic patients, or to monitor cancer patients for recurrence of the disease. Although mtDNA mutations have been reported in a wide variety of human cancers extending early detection to cancer prevention has proved problematic with regards to linking homoplasmic or heterplasmic variations with the etiology of cancer. us the characterization of populations of mtDNA variation would facilitate broad acceptance of mtDNA analysis.
Initial studies on whole mtgenome analysis established protocols for directly sequencing entire mtgenomes to detect sequence changes. For example, age-matched individuals with lung cancer had strikingly different mtgenome signatures, suggesting that these variants could be cancerassociated changes [56]. To evaluate progression of mtDNA mutation load associated with tumor stage progression, mtDNA mutation type and location across the entire mtgenome were evaluated between individuals with different stages and different types of cancer. Sequence variants were identi�ed in stage I to stage IV tumor samples, and these mutations were distributed across the entire mtgenome with no indication of a hotspot or speci�c site of mutation associated with speci�c cancer types or stages. Analysis across larger genome regions indicated a signi�cant clustering of mtDNA mutations in the ND gene complex, while 10% of mitochondrial mutations were found in the D-loop region.
e importance of whole genome analysis can be recognized in analogous measurements of entire mtgenomes. In human forensics, sequencing entire mtgenome is more effective because polymorphisms in mtgenomes can be useful for resolving individuals who have the identical hypervariable (HV) HV1 and HV2 control region sequences [57]. Using the whole genome as a potential source of mutations improves the discrimination power of forensic assays [58], and by extension cancer diagnostics, prognostics and tumor pro�ling. Figure 1 illustrates the advantage of whole genome analysis. All ten samples have identical D-loop mutation patterns and types, thus these samples are not distinguishable with only partial mtgenome analysis. As whole mtgenome sequencing is ubiquitous, easy to perform, and high-throughput for even small genomes, whole mtgenome sequencing should be the standard.

Considerations for Mitochondrial DNA Analysis.
Proper analysis of mtgenome mutations is required for accurate correlation of homoplasmic mutations with tumor tissue and stage ( Figure 2). Mutations are detected by comparing DNA sequence of tumor tissue to that of normal tissue or blood from the same individual [59]. It is important to use blood as control tissue measurements because it is necessary to subtract out mtDNA variation that arises from accumulation of damage over the lifespan, for example, due to aging. Direct analysis of haplogroups associated with listed across the top. Transformation of sequence information to a number enables a bar-code description of the samples. e HV regions do contain mutations, but they are identical and hence of no informative value and are not shown. On the contrary, mutations across the entire mtgenome demonstrates that whole genome analysis has clear utility.

A T G G A T G T C T G T T G G A A T C G G
T F 2: Samples were analyzed by whole genome sequence comparison of (1) tumour tissue and (2) matched patient blood. e sequence of a small region of the mtgenome is indicated across the top of the �gure. e red box indicates position and type of mutation observed in the tumour specimen. cancer does not establish a correlation with variation of mtgenomes metastasis [60]. Comparison of control tissue to tumor tissue must be conducted with such samples collected from the same individual. Accurate detection of mtDNA mutations must account for other sources of errors, for example nuclear mitochondrial pseudogenes (numts) are sources of contamination during �CR ampli�cation. is warrants careful experimental design and cautious interpretation of heteroplasmic results. Hence, mtgenome disease-associated biomarkers must be authenticated to preclude false-positive detection of paralogous nuclear pseudogenes [61].

Cancer Field Effect.
Another important reason to characterize entire mtgenomes is because of occurrence of mtDNA mutations that could be part of a cancer �eld effect within tissue. ese mutations could be biomarkers for progressive mutation patterns in lesions. However, correlation between single mutation sites and speci�c gradings remains loosely associated. Instead of attempting to de�ne hotspots for mutations, the gradual accumulation of mutation(s) distributed across entire genomes could be considered as an "individualized" marker of progression. Studies have reported no correlation of tumor-associated mtDNA mutations with respect to patient age. e mutation load or population are uniquely attributed to each starting point, and thereaer constrained by risk and rate based on initial populations. While mutation load is individualized, DNA damage may not be compartmentalized to one site. Although low sensitivity assays and limited sampling have plagued the majority of mtgenome comparative studies, not every tumor possessed sequence variants, while some samples contained a number of variants [58,62].
Analysis for identical damage in different tissues (cancer versus control) could be more apparent with analysis of populations of mtgenomes that indicate tissue predisposed to cancer. Absence of a 1 : 1 correlation between the mutation patterns of tumor progression is likely a result of tissue sampling. Hence, population analyses would facilitate characterization of speci�c locations and surrounding tissues. In summary, there are salient considerations for mtDNA mutation identi�cation comparative studies between potential cancer tissue, different types and stages of tumors, as well as non-invasive collected samples and experimental controls. It is imperative to incorporate analysis of the entire mtgenome and accurate identi�cation of heteroplasmic mitochondrial populations. Additionally, the sampling of bodily �uids and tissues surrounding cancerous tissue will facilitate de�ning the extent of the cancer �eld effect.
e derivation of sequence changes in the mtgenome in cancer remains unclear. Should these changes prove to be clonal expansions of a heteroplasmy already present in the tissue rather than tumour-associated in situ mutations, early detection of cancer may thus rely on the ability to differentiate levels of heteroplasmy. In general, studies of mtDNA mutations in cancer indicate the presence of sequence variants spanning the entire mtgenome, and therefore full genome sequencing will provides the cancer diagnostic community with a useful biomarker discovery approach. Such characterization of populations would be useful to de�ne PCR panels for inexpensive triage and screening of human populations. e sensitivity of mutation detection, rates as little as 2% contribution to the admixture of normal and tumor DNA, indicate heterogeneous biological samples such as bodily �uids, lavage specimens, �ne-needle aspirates, or biopsies can potentially be analyzed for cancerassociated mitochondrial DNA mutations. e identi�ed heteroplasmic and homoplasmic sequence variants from tumors and blood (control) and urine (for matched bladder cancer) and bronchoalveolar lavage (for matched lung cancer) were measured from the same individual. It is a reasonable assumption that heteroplasmies should comprise multiple subpopulations of mutated mtDNA molecules. Research that incorporated whole mtgenome analysis using sensitive methods describes sequence variant identi�cation of both heteroplasmic and homoplasmic sequence variants in clinical samples distributed across the mtgenome. Hence, small cohort studies that use incomplete mtgenome sequencing or methods designed to scan discrete portions of the genome, miss important sequence variants associated with cancer or other diseases. It is now possible to screen populations to understand speci�c frequencies and distributions, and to compare sequences of entire mtgenomes in order to have a comprehensive characterization of sample material. Finally, on account mtDNA mutations are well validated, mitochondrial research is beginning to concentrate on understanding the link between mitochondrial function and pathological states. ere are a few studies that have begun to address the association of mitochondrial function with change in homeostasis [63,64], as well as mitochondrial redox state [65]. As technology improves to allow the accurate assessment of cellular interaction and hemodynamics, monitoring the effects of mitochondrial dysfunction in combination with using mtgenome mutation diagnostics on the pathophysiology of cancer cells will begin to support medical decisionmaking.
2.6. Mitochondrial Proteomics. e majority of mitochondrial proteins are encoded by the nuclear genome and imported to the mitochondria to perform their speci�c functions. us the mitochondrial proteome is the result of complex crosstalk between both nuclear and mitochondrial programs, and is greatly in�uenced by pathological conditions including cancer. In the past decade, the mitochondrial proteome has been characterized from highly puri�ed mitochondria resulting in a comprehensive list of over 1,000 mitochondrial proteins (as reviewed in [66]). Using the wealth of knowledge from such studies, numerous databases have been created such as MitoInteractome [67], MitoP2 [68], HMPDb, and MitoMiner [69]. e MitoInteractome database contains 6,549 protein sequences derived from mutltiple databases (SwissProt, MitoP, MitoProteome, HPRD, GO) from several different species creating a comprehensive protein-protein interaction network. Certainly one of the most extensive databases, MitoP2 contains data from a wide breadth of mitochondrial proteomic studies spanning from single protein studies to extensive proteome-wide mapping and expression studies. e HMPDb (Human Mitochondrial Protein Database) provides consolidated information on mitochondrial DNA sequences, polymorphisms, disease-related proteins, and 3-D mitochondrial protein structures. Collectively these databases serve as wonderful utilities for the discovery and characterization of novel mitochondrial biomarkers for diagnosis and molecular targets for drug treatments.
Extensive protein expression differences have been found in mitochondrial glycolytic enzymes, heat-shock proteins, cytoskeleton proteins, and antioxidant enzymes through comparative proteomic analysis. In regards to metabolism, proteins of the glycolytic and pentose phosphate pathways tend to be induced, along with reductions in oxidative phosphorylation pathways [66,70] [71]. e most differentially expressed mitochondrial proteins between normal and cancerous cells included cytochrome oxidase subunit 5B, malate dehydrogenase, and elongation factor Tu. Several proteomic studies have shown a signi�cant correlation between high levels of heat-shock protein 70 (HSP-70) in a variety of cancers including gastric adenocarcinoma, hepatocarcinoma, and oesophageal cancer [66]. HSP-70 functions as a mediator of cell proliferation, cellular senescence, and cellular immortalization, and when concentrated in to cytoplasm sequesters p53 and activates Ras-Raf signaling which controls cell proliferation [72,73].
Recent observations and interest in mitochondrial research has generated a lot of enthusiasm and hope that novel therapeutic agents will be identi�ed that are effective for cancer therapy. In a general sense a dynamic and bidirectional exchange between the metabolic status of the cell generated by mitochondria and genetic pro�le of a cell will provide a better understanding of metabolites and unexplored signaling mechanisms. Hence a complete understanding of the mitochondrial proteome and its regulation by metabolites, including ROS, will provide a better understanding of the symbiotic relationship that has evolved in eukaryotic cells. Additionally, recent advances in high-throughput technologies, such as next-gen sequencing and Mitochip [74], have allowed for the rapid and accurate detection of mtDNA mutations, polymorphisms, or copy number variations in a variety of tissues and bodily �uids [55,[75][76][77][78][79].

Future
e mitochondrion, as the biochemical nexus of the cell, is a critical consideration in the genomics era of the new millennium. Although much of the current funding is aligned to continuing to further understand the functional details of the nuclear genome, the mitochondrion and its modest complement of DNA and protein is emerging as a crucial component of the biological networking of nuclear pathways. In addition to generating 95% of the chemical fuel �ring cellular metabolism through carbohydrate breakdown, mitochondrial perform and mediate a number of events including ROS generation, retrograde calcium signaling and intrinsic apoptosis. Most of these pathways are fundamentally altered during malignant transformation. For example, the apoptotic pathway is severed and carbohydrate metabolism is preempted for anaerobic respiration. Mitochondria shroud a multitude of unidenti�ed proteins, suggesting yet to be understood, deeper biological functions [80]. is important information is not yet fully detailed; however, it is rich with promise. It is likely that these discoveries will provide new approaches to cancer treatment, diagnosis and prognosis. For example, the accelerated mutation rate of the mtgenome offers early identi�cation of malignant transformation by identi�cation of a �eld effect in normal appearing tissue [81]. is molecular conditioning is well attested particularly in prostate cancer [52,82,83]. In addition, the mtgenome has a high copy number and an increased somatic mutation rate in comparison to its nuclear counterpart, providing multiple target copies with target markers. is threesome of characteristics (�eld effect, copy number, mutation rate) will enable monitoring of vulnerable epithelium in organs such as the lungs, colon, breasts, prostate, and ovaries. Resection of both tumors and the surrounding �eld may have important implications for recurrence [79]. Since malignant transformation is a 20 to 30 year process, in most cases a shi to a �eld effect could allow prediction of the development of intraepithelial neoplasia (IEN). Speci�c markers may indicate which IEN lesions and molecular renovations may progress towards a malignant phenotype. e American Association for Cancer Research (AACR) Task Force on the Treatment and Prevention of IEN published the following statement in 2002: "e AACR IEN Task Force recommends focusing on established precancers as the target for new agent development because of the close association between dysplasia and invasive cancer and because a convincing reduction in IEN burden provides patient bene�t by reducing cancer and because a convincing reduction in IEN burden provides patient bene�t by reducing cancer risk and/or by decreasing the need for invasive interventions [84]. " In addition, the traits of mtDNA have successfully led to identi�cation of mitochondrial mutations in low cellular bio�uids such as nipple aspirate �uid [85]. Signi�cant resolution between bladder cancer stages Ta, T1 and T2 was obtained using the SNP counts in whole mtgenome sequencing of urine cell pellets in 20 of 31 patients [86]. Notably, circulating cell-free mitochondrial DNA in peripheral blood has diagnostic utility for breast cancer, urological malignancies, and predicting prostate cancer recurrence. [87][88][89].
Due to its central role in cell physiology, speci�c alterations in the mtgenome may indicate the status of speci�c pathways or impact biological outcomes. For example, mutations in mitochondrial respiratory complexes may in�uence the induction of apoptosis [90] and promote metastatic behavior in both prostate and breast cancers [49]. ese studies suggest that the sequence of bases in the mtgenome are �nely ordered to the point that even some sequence speci�c haplogroups may be more susceptible to malignancies [50]. is concept should not be surprising since "natural selection mediated by climate has contributed to shape the current distribution of mtDNA" [91]. Hence mitochondria are dynamic, adaptable molecules able to mitigate biological compromise given metabolic parameters. Disease susceptibility may be tolerated due to imposed climatic constraints.
e cellular ganglion of mitochondria, plethora of pathways and high volume molecular trafficking have been recognized as ideal chemotherapeutic targets [52]; however, this approach draws the proverbial "double-edged sword. " For instance, the adjuvant treatment of estrogen receptor positive breast cancer with tamoxifen requires intact and fully operational mitochondria [92]. Importantly, mitochondrial toxicity is a major implication in the failure of chemotherapeutic agents in the late stages of drug development [93]. Careful consideration of mitochondrial and compound interactions is imperative to both target mitochondria for therapeutic indications, while avoiding off-target effects of other therapeutic molecules. Disruption of key mitochondrial molecular transport molecules, such as SCaMC-1, or SLC25A1, in proliferating cells has been suggested as a mitochondrial speci�c approach to tumor treatment [89,90].

Conclusion
Mitochondria have a critical role to play in the successful conquest of cancer. Further and deeper investigations of this organelle assure profound insights into the missing molecular mechanisms of malignancy. e oen referred to "powerhouse of the cell" is beginning to look more like a well ordered neighborhood of sprawling metabolic mansions. Some areas contain décor dating from the earliest of antiquities, while others have yet to be opened and thoroughly explored for the elusive, but ultimate answers to cancer biology; however, many have hurried through the biological lobby of this complex like tourists on a bus schedule. We must now committee to taking the grand tour; more magni�cent biological vistas await. Mitochondria may yet be found to be the master of the cellular orchestra.

Introduction
Oxidative stress is a state of physiological imbalance between oxidant production and antioxidant defence at different biological levels. It is implicated in the development of many pathological conditions such as aging, neurodegenerative diseases, and cancer initiation and progression [1][2][3][4][5][6]. Many diseases are suspected to be linked to oxidative stress, but procurement of disease tissues may be difficult due to its invasive nature and the scarcity of available tissues. However, researchers have mitigated this problem by using the systemic oxidative stress in peripheral tissues, such as circulating blood, as a noninvasive surrogate. Extrinsic factors such as in�ammation, nutrient imbalance, and hypoxic environment could affect inter and intracellular redox homeostasis, therefore altering systemic oxidative stress levels; new efforts are made to investigate the interactions between systemic oxidative stress and pathogenesis of many disease conditions [7][8][9][10][11][12][13]. For example, several recent studies suggest a correlation between increased systemic oxidative stress and prostate cancer risk and progression [14][15][16]. Similar results are reported in lung cancer [17], head and neck cancer [18], and other human cancers [19,20]. us, enhanced oxidative stress is not only a common property of the diseased cells, but may also be re�ected in the peripheral tissues.
Systemic oxidative stress has been analyzed in serum and blood cells using different biomarkers and assay systems. Genomic DNA in circulating lymphocytes is a widely used target in measuring different end-points of oxidative DNA damage, such as 8-oxoguanine (8-oxo-G) base lesions or DNA strand breaks detected with the comet assay [14][15][16][17][18]. e mitochondrial DNA (mtDNA) in lymphocytes is an attractive alternative target to determine systemic oxidative stress. MtDNA is a circular, multicopy cytoplasmic DNA, semiautonomously maintained in mitochondria. It is known to be more sensitive to oxidative damage than nuclear DNA [21][22][23] and has been increasingly used for evaluating systemic oxidative stress with various assays. Similarly to nuclear DNA, 8-oxo-G base lesions can be assessed in puri�ed mtDNA from lymphocytes [24]. Extracellular circulating mtDNA in serum is another marker recently used for evaluating genetic integrity and cancer risk. Elevated levels of free �oating mtDNA detected in the plasma or serum are found to be associated with poor prognoses for prostate and testicular germ cell cancers [25][26][27]. However, the source and nature of this circulating mtDNA are not fully elucidated. Oxidative stress can also affect the total mtDNA content in lymphocytes under various diseased conditions [28]. For example, signi�cant alterations in mtDNA content were detected in lymphocytes from patients with renal cell carcinoma, hyperlipidemia, and Huntington's disease when compared to control populations [19,29,30]. However, the relationship between different mtDNA end-points reported in lymphocytes is not clear and the direct measurement of mtDNA strand breaks in lymphocytes has not been reported. We previously developed a sensitive in vitro assay to quantify mtDNA structural damage induced by strand breaks, repair and copy number change in prostate cancer cell lines using a supercoiling-sensitive real-time PCR (ss-qPCR) [6,31]. We showed that oxidative damage can induce single-or doublestrand breaks (SSB or DSB), which lead to the disruption of the supercoiled conformation, and that the resulting relaxed conformation is a better qPCR substrate for signi�cantly increased ampli�cation than the supercoiled conformation, even if the starting mtDNA molecules remain the same [31]. Additionally, we observed that prolonged exposure to 95 ∘ C heat also introduced strand breaks in the mtDNA. is particular property was advantageously used to disrupt all structural features of mtDNA for precise quanti�cation of the total mtDNA content [31].
e objectives of this current study were to test if the ss-qPCR method could be applied to the lymphocytes and to explore a quantitative strategy to measure multiple mtDNA end-points in circulating blood cells for the study of systemic oxidative stress. We developed an absolute quanti�cation method for precise measurement of mtDNA structural damage, copy number change, and repair activity in blood cells. We demonstrated that mtDNA has low levels of both copy number and baseline damage in lymphocytes as compared to cancer cell lines, and that exogenous H 2 O 2 led to a signi�cant increase in mtDNA damage but with little repair activity in inactivated lymphocytes in ex vivo experiments.

Blood Collection and Lymphocyte Preparation.
Healthy male volunteers ranging from 28 to 45 years old were recruited for this pilot study through an institutional review board (REB) approved protocol at the McGill University Health Center. Blood (10 to 15 mL) was collected into 9 mL collection tubes coated with EDTA (Vacu K3EDTA PULL LAV) (Fisher, Monroe, NC). For experiments with whole blood, the samples were immediately stored at −80 ∘ C in 10% dimethyl sulfoxide (DMSO) prior to analysis. For experiments with isolated lymphocytes, blood was submitted to Ficoll-Paque Plus (GE Healthcare, Buckinghamshire, England) to recover the lymphocytes [33], then stored at −80 ∘ C in 40% RPMI media 1640 supplemented with 50% FBS and 10% DMSO prior to analysis. As per manufacturer's speci�cations, the extracted sample is composed in majority of lymphocytes (75-93%), with a remaining fraction of monocytes (7-25%) and minimal contaminants from granulocytes, erythrocytes, and platelets (3 ± 2%, 5 ± 2%, and < 0.5%, resp.).

H 2 O 2 Challenge Experiments with Lymphocytes and
Whole Blood. Frozen lymphocytes were thawed in a 37 ∘ C water bath for 1-2 min and washed with 5 volumes of icecold wash medium (50% FBS and 50% RPMI 1640). e lymphocytes were counted and cell viability was assessed under microscope using the trypan blue dye (average of over 90% viability). A total of ∼3 × 10 6 lymphocytes were incubated in 50 mL conical tubes with RPMI-1640 complete medium for 30 min prior to the experiment. e cell suspension was split into three groups of ∼1 × 10 6 cells each, treated with 0 (control) or 120 M H 2 O 2 for 15 min for exposure or allowed to recover in fresh medium for 60 min. e concentration of 120 M H 2 O 2 was chosen to be in the lower-middle range of concentrations used in similar treatments in the literature (50 to 500 M) [14,34]. Aerwards, the lymphocyte samples were washed with PBS, spun down to a pellet, and then stored at −80 ∘ C before DNA preparation.
Frozen whole blood was thawed in a 37 ∘ C water bath for 1-2 min and washed with 5 volumes of ice-cold PBS wash medium. e whole blood cells were counted with trypan blue dye prior to incubation (average of over 90% viability). A total of ∼15 × 10 6 whole blood cells were incubated in RPMI-1640 complete medium in 50 mL conical tubes for 30 min prior to the experiment. Whole blood samples were separated into three groups with ∼5 × 10 6 cells each and treated with 0 or 120 M H 2 O 2 as in the lymphocyte experiment. Whole blood samples were collected aer treatment and stored at −80 ∘ C before DNA preparation. Kit according to the manufacturer's instructions with minor modi�cations to ensure that both mtDNA and nuclear DNA were collected together [31,35]. Total DNA was quanti�ed with a NanoDrop spectrophotometer. DNA template solutions of 1 ng/ L were prepared for each sample with 1X Tris/EDTA Buffer Solution (pH 8.0). Each template solution was split into two equal parts with half serving as an original template for the measurement of the damaged/relaxed mtDNA fraction and the other half heat-treated (95 ∘ C for 6 min on a PCR machine) to quantify total amount of mtDNA [31].

Nuclear DNA and mtDNA Standards Preparation for
Absolute �uanti�cation. MtDNA standards were prepared for absolute quanti�cation. A 3.3 kb mtDNA fragment containing the CO2 gene and a 2.5 kb fragment containing the D-loop region were ampli�ed from the immortalized normal human prostate cell line, RWPE-1, using primers listed in Table 1. PCR reactions were performed using the GeneAmp PCR 9700 system (ABI) with recombinant ermus thermophilus (rTth) DNA polymerase (ABI). e ampli�cation program was performed as follows: preheat samples to reach 75 ∘ C; add rTth DNA polymerase and incubate for 2 min; denature at 94 ∘ C for 1 min, followed by 30 cycles of 94 ∘ C for 15 sec, 60 ∘ C for 30 sec., and 72 ∘ C for 3.5 min; then 72 ∘ C for 5 min and cool down to 10 ∘ C. Ampli�ed DNA fragments were puri�ed with the �IAGEN PCR Puri�cation Kit. e puri�ed products were carefully quanti�ed with the Nanodrop spectrophotometer, and the average of three readings was used for calculating precise copy number according to the following equation ( Figure 1): Six or seven serial dilutions were made ranging from 3 × 10 6 to 30 or 3 × 10 7 to 30 copies with a dilution factor of 10 depending on the experiment. e 6-point standard was used for the mtDNA quanti�cation in blood samples, while the 7-point standard was used to demonstrate the dynamic range and linearity of the assay. e original stock solutions were made into small aliquots and stored at −80 ∘ C to prevent repeated freeze and thaw. e nuclear DNA standards were similarly prepared. e nuclear primer sequences are listed in Table 1. A 2.7 kb nuclear fragment containing the calicin gene was ampli�ed from RWPE-1. Calicin is a single-copy nuclear gene that encodes for a basic protein of the sperm head cytoskeleton.

2.�. �uanti�cation of mtDNA Damage and Copy
Number Using the Absolute ss-qPCR Method. e amount of relaxed/damaged mtDNA and total copy number were measured by quantifying the original and preheated DNA templates, respectively. e nuclear DNA marker calicin was quanti�ed using the original templates. e qPCR was performed using the Applied Biosystems7500 Fast Real-Time PCR System (ABI) with Power SYBR Fast Green PCR MASTER MIX (ABI) [35]. e original DNA templates and preheated DNA templates and standards were analyzed in triplicates on the same plate. e two-step PCR ampli�cation program for both nuclear DNA and mtDNA was 95.0 ∘ C for 30 sec, followed by 40 cycles of 95.0 ∘ C for 3 sec and 60.0 ∘ C for 30 sec. A melt curve analysis was enabled at the end of ampli�cation. e primer sequences are listed in Table 1. e absolute copy numbers of CO2, D-loop, and calicin were calculated based on the standard curves. Since calicin is a single copy nuclear gene, the cell number could be calculated with the following equation with the assumption that the nuclear equivalent is representative of the cell number ( e exact copies of damaged and total mtDNA per cell were calculated from: mtDNAcopies cell = CO2 or D-Loop copy number cell number * , (3) * cell number and nuclear equivalent will be used interchangeably from this point.

Data Analysis.
All statistical analyses were performed with the aid of Graphpad Prism version 4 soware. Unless speci�ed otherwise, the data was analyzed with one-way ANOVA with Dunnett post test, and a . is considered signi�cant.

�.�. A Ne� Strategy for the Absolute �uanti�cation of �otal and Damaged mtDNA.
We have devised a new approach for the absolute quanti�cation of mtDNA structural damage and total copy number in a single analysis. e protocol, illustrated in Figure 1, was comprised of four main steps.
where m is the mass of a single copy and n is the target size in base pairs. • Standard solutions range: 3 × 10 6 to 30 copies/cell with a 10× dilution factor. e �rst step consisted in the construction of mtDNA and nuclear DNA standards (Figure 1(a)). Two to three kb DNA fragments containing mtDNA (CO2 or D-loop) and nuclear DNA (calicin) were ampli�ed by PCR from a normal prostate cell line, RWPE-1. e concentration (copies/ L) of these long DNA fragments were quanti�ed and calculated according to (1). e second step was to prepare the DNA templates for qPCR analysis (Figure 1(b)). Each DNA template was split into two equal halves. One half was used for the quanti�cation of relaxed mtDNA and calicin nuclear DNA copies. e other half was pretreated at 95 ∘ C for 6 min to unfold any structure and was used for quantifying total mtDNA. e third step consisted in the absolute quanti�cation using qPCR ( Figure 1(c)). To obtain mtDNA content per cell, the exact amount of mtDNA and nuclear DNA copies were quanti�ed and calculated from the standard curves according to the equation: copies = "10 (Ct− , " where the cell number was derived from (2). e �nal step was the interpretation of the data (Figure 1(d)). With this approach, the amount of damaged mtDNA copies/cell, total mtDNA copies/cell, and baseline mtDNA damage (ratio of damaged mtDNA/total mtDNA) were quanti�ed simultaneously. 0.9996 for calicin. A single uniform melting peak at 76 ∘ C, 75 ∘ C, and 79 ∘ C was observed for CO2, D-loop, and calicin, respectively, demonstrating the high speci�city of the primers (Figures 2(b), 2(d), and 2(f)). e intra-assay reproducibility of the standard was analyzed by calculating the coefficient of variation (CV) of the triplicates. e intra-assay median CV were 0.27%, 0.17%, and 0.12% for CO2, D-loop, and calicin, respectively ( Table 2). e interassay CV was calculated with data from two or more independent experiments: the CV were 0.33%, 0.10%, and 0.62% for CO2, D-loop, and calicin, respectively ( cell lines, LNCaP and C4-2, were analyzed for mtDNA content and baseline damage. e prostate cancer cell lines served a reference in method development because the ss-qPCR method was previously developed with these cell lines [6,31,35]. In lymphocytes, the total mtDNA content was quanti�ed at an average of 153.25 ± 21.02 copies/cell from 4 individual samples, among which the amount of damaged mtDNA molecules was averaged at 41.44 ± 7.87 copies/cell (Figure 3(a)). In comparison, signi�cantly higher mtDNA contents were detected in prostate cancer cells C4-2 (1495.35 ± 12.45, 0.01) and LNCaP (3086.61 ± 48.27, 0.01). e damaged mtDNA copies were 466.44 ± 8.64 and 990.41 ± 6.77 copies/cell, respectively. e baseline damage was calculated with the ratio of damaged mtDNA over total mtDNA: 27.04% of the total mtDNA content was damaged for lymphocytes versus 31.19% and 32.09% for C4-2 and LNCaP, respectively (Figure 3(b)). is assay was highly reproducible; the median intra-and interassay CV were 0.74% and 1.20% for cell lines and 1.87% and 2.33% for lymphocytes, respectively (Table 3). Furthermore, the use of different mtDNA markers, CO2 and D-loop, generated near identical results in terms of total mtDNA content, damaged mtDNA, and baseline damage detected ( Figure  3(c)). Indeed, the average CV value obtained between CO2 and D-loop markers was calculated at 0.51%. us, these two mtDNA markers were highly consistent and interchangeable in quantitative mtDNA analyses. It was interesting to note that the absolute number of damaged mtDNA molecules was proportionally higher in samples with increased total copy numbers (Figure 3(a)). As such, the ratio between damaged and total copy numbers was a better indicator of the baseline level of DNA damage in a cell (Figure 3(b)). Taken together, the new quanti�cation platform developed in this study provided a highly reproducible method for simultaneous analysis of absolute mtDNA copy number, damaged molecules, and baseline damage in both isolated lymphocytes and cancer cell lines.

Ex Vivo mtDNA Damage Responses to Exogenous H 2 O 2 in
Isolated Lymphocytes and in Whole Blood. Isolated lymphocytes from 9 healthy men were treated with 0 or 120 M H 2 O 2 for 15 min to evaluate induced mtDNA damage and repair activity aer 60 min of recovery. e average mtDNA copy number of the untreated control samples was 161.78 ± 31.67 copies/cell (Figure 4(a)). e total mtDNA copy number was not affected by H 2 O 2 treatment and remained stable across all treatment groups (Figure 4(a)). However, rapid mtDNA damage response was observed. e average baseline damage of untreated control samples was 27.63% (Figure 4(b)). Upon H 2 O 2 exposure, the fraction of damaged mtDNA increased to 58.19% in lymphocytes, representing a 110.6% increase in induced damage from the control ( 0.001). Interestingly, the induced damage was not repaired aer 60 min of recovery, suggesting a lack of repair activity during the recovery period.
As an alternative to the isolated lymphocytes, a small amount of whole blood samples ( 1 mL each) from four healthy subjects was tested using the same procedure. e average total mtDNA content of untreated control was 109.4± 22.40 copies/cell ( Figure 5(a)). Similar to lymphocytes, the average baseline mtDNA damage of the untreated control samples was 26.6% ( Figure 5(b)). When treated with 120 M H 2 O 2 , the damaged fraction of mtDNA increased to 36.7%, representing a 38.0% increase in induced damage as compared to the baseline levels ( 0.05), while the total mtDNA content remained the same ( Figure 5(b)). An absence of repair activity was also observed within 60 min recovery aer the H 2 O 2 treatment. However, the whole blood samples had slightly lower mtDNA content and less pronounced mtDNA damage responses as compared to the lymphocytes. is could be caused by the complexity of different types of white blood cells present in whole blood samples. Despite this difference, the overall stress response pattern was similar between the isolated lymphocytes and the whole blood. us, the latter may serve as a convenient alternative to isolated lymphocytes in the analysis of mtDNA stress responses in circulating blood.

Discussion
Based on our previously developed ss-qPCR method [6,31,35], we propose a quantitative approach for precise and rapid detection of mtDNA structural damage and copy number change in isolated lymphocytes in a single analysis. We have demonstrated that the new approach had a wide dynamic range and was highly speci�c and reproducible. A relatively low mtDNA content and baseline level of damage were observed in lymphocytes of healthy men, and the lymphocytes were shown for the �rst time to exhibit a signi�cant increase in mtDNA damage, followed by little repair activity aer 1 h of recovery in an ex vivo challenge experiment with H 2 O 2 . is lack of repair activity to H 2 O 2 -induced damage aer 1 h of recovery is consistent with a study from Collins et al. in which nuclear DNA repair activity was only observed aer several hours (>2 h) [34]. Moreover, we showed that 1 mL of whole blood may serve as a convenient alternative  ), and lymphocytes ( ) were analyzed by ss-qPCR for total mtDNA content, damaged mtDNA number, and level of baseline damage. e cell number was calculated from the copy numbers of calicin, a single copy nuclear marker. (a) With mtDNA CO2 marker, the original (CO2 original) and preheated (CO2-heated) DNA templates were quanti�ed for damaged copies and total mtDNA copies, respectively. (b) e baseline damage level was obtained by dividing damaged copies over total copies. (c) Comparison of CO2 versus D-loop markers in lymphocyte samples ( ).
to the isolated lymphocytes in the mtDNA analysis. us, mtDNA in blood may be explored as a sensitive surrogate to systemic oxidative stress by simultaneous analysis of multiple end-points in a single test. e absolute quanti�cation system developed in this study provides a standard method for the reliable quanti�cation of the precise mtDNA copy number in lymphocytes and whole blood cells. is is achieved through well-de�ned mtDNA and single-copy nuclear DNA markers and by taking into account the DNA structural effects on qPCR ampli�cation [31]. e relatively low mtDNA copy number revealed in isolated lymphocytes is consistent with very limited data reported in the literature. For example, one study detected ∼87 to 579 copies/cell with a different real-time PCR method [36] and the other ∼70 to 320 copies/cell with competitive PCR in lymphocytes [37]. Many studies report mtDNA content on a relative scale [19,[38][39][40][41]. However, the relative analysis is limited by the difficulty of comparing results from one  study to another and by the signi�cant variations observed in mtDNA content between individuals [42]. To account for the inhibitory effect of the supercoiled DNA structure on qPCR ampli�cation, we have taken steps to ensure an accurate measurement by disrupting the supercoiled mtDNA conformation with a preheating step prior to qPCR analysis. is step is necessary for precise quanti�cation of mtDNA content but has largely been ignored in previous reports.
Depending on the manufacturers of the qPCR machinery and chemistry kits, there are wide variations in the duration of the initial hot-activation step for hot-start DNA polymerases, which varies from 10 min to as short as 20 sec (e.g., ABI 7500 Fast System). We have shown that shorter denaturation time at 95 ∘ C was insufficient to disrupt all the supercoiled mtDNA conformation in a time-and dose-dependent experiment in prostate cancer cell lines [31]. erefore, the inclusion of a preheating step in the template preparation is crucial for the accurate mtDNA measurement. In addition to measuring the total mtDNA content, our new system also provides a novel approach for direct quanti�cation of the absolute copies of damaged mtDNA with qPCR. is is in contrast to mtDNA conformational study based on gel electrophoresis coupled with Southern Blot, which requires tedious post-PCR manipulations and is semiquantitative in nature. On the other hand, popular assays such as the comet test for detecting nuclear DNA strand breaks are not applicable to mtDNA due to its small size [34]. e quanti�cation of structural mtDNA damage reported in this study mainly re�ects the damage caused by single-and double-stranded breaks, as it was shown that other type of DNA damage such as base lesions or abasic sites had little, if any, effect on the structure [31]. It is interesting to note that the amount of damaged mtDNA changes with the total mtDNA content in a cell. e direct comparison of the damaged mtDNA molecules from different individuals can be compounded by variations in the total content. To normalize this variation, we propose to calculate mtDNA damage based on the percentage of damaged versus total mtDNA molecules in a cell; this ratio of damage is relatively stable and more informative for comparative studies [43]. Moreover, the ratio of damage can be used to infer the baseline or endogenous damage in the isolated lymphocytes or whole blood from the untreated samples; it also quanti�es induced mtDNA damage in isolated lymphocytes or a small amount of whole blood cells under oxidative stress. e ability of our approach to measure both endogenous and induced damage/repair responses in ex vivo treatments may be used to explore the state of oxidative defence and/or repair capability of individuals with different disease conditions. Indeed, previous studies have suggested that there is an association between systemic oxidative stress and diseases, such as an association of the high prostate cancer risk with severe damage response and poor repair capacity of nuclear DNA in lymphocytes [14].
In conclusion, we have developed an absolute quanti�cation system for rapid measurement of mtDNA structural damage, copy number change, and damage response in isolated lymphocytes and whole blood cells. Systemic oxidative stress is associated with diverse pathological conditions, ranging from neurodegenerative diseases to many types of cancers. It is conceivable that the blood analysis with the multiple mtDNA end-points proposed in the this study may provide a simple and sensitive test to interrogate the nature and extent of systemic oxidative stress for a broad spectrum of clinical investigations, especially when coupled with other established tests such as cell-free circulating mtDNA, the comet assays targeting the nuclear DNA, and the detection of 8-oxo-G base lesions.

Introduction
e mitochondrion contains the only functional DNA outside the eukaryotic cell nucleus and mutations in this genome have been linked to a vast array of human disease including pediatric neurologic disease, degenerative muscular disease, and blindness in the pediatric population and more recently to chronic diseases of adults including diabetes, Alzheimer's dementia, Parkinson's disease, cardiovascular disease, and cancer [1]. Because of the critical role the mitochondria play in normal human cell physiology, severe mutations that cause dramatic derangements of mitochondrial function can be fatal [2]. Indeed, recent evidence indicates that severe mutations are culled from the germ line through depletion of mutant germ cells in the ovary [3]. Severe mutations can exist and even be passed between generations if they exist in a state of heteroplasmy where some copies of the genome contain the mutation and some are wild type. For this reason, one indication that a mitochondrial DNA (mtDNA) mutation is potentially pathogenic is that it exists in a heteroplasmic state, though this is not strictly necessary in all cases. Prostate cancer is one example of an adult disease tightly linked to ageing for which there is evidence linking both inherited and somatic mtDNA mutations to disease [4]. Mutations in the mitochondrial cytochrome c oxidase subunit I (COI) gene are found at disproportionately high rates in prostate cancer patients compared to either the general population or rigorously de�ned controls without prostate cancer [5]. A population-based case-control study of African Americans found 2 single nucleotide polymorphisms in this gene signi�cantly associated with prostate cancer ( ) and in strong linkage disequilibrium with each other ( 6) [6]. MtDNA mutations are easily detected in early stage prostate cancer tissues and can also be found in the serum and urine from the same individuals [7]. MtDNA content has been linked to androgen responsiveness of prostate cancer cells in vitro [8] and analysis of clinical samples suggests that mtDNA content is higher in prostate cancer compared to adjacent normal prostatic tissue in the same individual [9]. A speci�c mtDNA deletion mutation may have utility in identifying adult cancer not apparent in clinical prostate biopsies [10].
While multiple independent investigators have con-�rmed mtDNA mutations in prostate cancer, there is little understanding of the cell biologic and biochemical consequences of speci�c prostate cancer-associated mtDNA mutations. We investigated the mtDNA from a patient with prostate cancer and found a heteroplasmic missense mutation in the mitochondrial COI gene that impairs the oxidation of cytochrome c (respiratory complex IV inhibition) and increases the generation of both reactive oxygen species (ROS) and reactive nitrogen species (RNS).

Cytochrome c Oxidase Subunit I (COI) Gene Sequencing.
e mtDNA region encompassing COI was ampli�ed using a forward primer starting at nucleotide position (np) 5772 (5 � AGG TTT GAA GCT GCT TCT TC 3 � ) and a reverse primer ending at np 7600 (5 � CGC TGC ATG TGC CAT TAA GA 3 � ). e template was denatured at 95 ∘ C for 7 min and primers extended for 35 cycles of 94 ∘ C for 1 min, then 55 ∘ C for 1 min, and 72 ∘ C for 1 min. Both strands of the COI polymerase chain reaction (PCR) product were cyclesequenced using the slip primers in the forward direction starting at np 6080 (5 � TCT ACA ACG TTA TCG TCA CA 3 � ) and at np 6930 (5 � TGC AGT GCT CTG AGC CCT AG 3 � ) and in the reverse direction starting at np 6340 (5 � CTA GGT GTA AGG AGA AGA TG 3 � ) and at np 7150 (5 � GAT TTA CGC CGA TGA ATA TG 3 � ). e templates were denatured at 96 ∘ C and primers extended in the presence of "Big Dye Terminators" for 25 cycles of 96 ∘ C for 10 sec, then 55 ∘ C for 5 sec, and 60 ∘ C for 4 min. e reactions were chilled to 4 ∘ C, and the excess dye terminators removed by Centri-Sep Columns. e trace �les were determined using an Applied Biosystems (ABI) Prism 3100 genetic analyzer, analyzed using Sequencher gene analysis soware v 4.1 (Gene Codes, Ann Arbor, MI), and interpreted within the context of MITOMAP (http://www.mitomap.org/). Full mtDNA sequencing was performed as described above using primer sets that span the full length of the mitochondrial DNA [11].

Epstein-Barr Transformation.
Lymphocytes were isolated from whole blood by centrifugation and diluted with phosphate buffered saline (PBS). Red cells were lysed by the addition of H 2 O. Aer 20 s, osmolarity was restored with 10x concentrated PIPES (piperazine-N, � -bis-[2-ethanosulfonic-acid]) buffer, centrifuged, and the pellet was resuspended in RPMI with fetal bovine serum (FBS) and incubated for 45 min in 5% CO 2 at 37 ∘ C. e nonadherent cells were washed and collected and lymphocytes pelleted. Lymphocytes were infected with the B95-8 strain of Epstein-Barr Virus (EBV) and cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM l-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin, at 37 ∘ C in a humid atmosphere saturated with 5% CO 2 until cryopreservation or use [12].

Mitochondrial Preparation and Isolation.
Mitochondria were isolated by cell fractionation and centrifugation. All procedures were carried out on ice. Cells were pelleted and washed in cold PBS and resuspended in 10 volumes of isolation buffer (250 mM sucrose, 10 mM HEPES, 1 mM EDTA, pH 7.35). Cell lysis was performed with 5 passes of a hand held Dounce homogenizer and centrifuged at 500 g (4 ∘ C) for 5 min. e supernatant was reserved and the pellet was resuspended in isolation buffer, rehomogenized, and centrifuged. is process was repeated a third time and supernatants combined and centrifuged 12,000 g at 4 ∘ C for 10 min to pellet mitochondria which were then washed in isolation buffer and repelleted three times. Following the �nal wash, mitochondria were resuspended in reaction buffer (1 mL/mL of original cell pellet), and frozen until assayed.

Cytochrome c Oxidase
Activity. Cytochrome c oxidase activity of isolated mitochondrial preparations was measured by dual wavelength spectrophotometry (550/540 nm) as previously described [13]. Brie�y, mitochondria were suspended in reaction buffer (250 mM Sucrose, 10 mM HEPES, 1 mM MgCl 2 , pH 7.35). An aliquot of mitochondria suspension (∼300 ug mitochondrial protein) was added to the cuvette containing reaction buffer plus ∼5 uM reduced cytochrome c in a �nal volume of 700 uL. e oxidation of cytochrome c was recorded over time and the cytochrome c oxidase speci�c activity was calculated. Mitochondrial protein was determined by the Bradford method [14].

Cybrid Formation.
Mitochondrial donors (patient lymphoblasts) were enucleated by short term incubation with Actinomycin D, which was subsequently removed from the culture medium by centrifugation and washing. ese cytoplasts were then rescued by polyethylene glycol (PEG)induced fusion with 143B rho zero cells. Fusions are monitored by phase contrast microscopy and isolated by ring cloning. Clones were expanded and genotyped to assure that the donor mtDNA had been incorporated (sequencing) and that a single nucleus was present (Karyotyping).

Flow Cytometric Assay for ROS and RNS. Peroxides
Assay: Lymphoblasts were grown in suspension in �asks. When cells were in growth phase, 1 × 10 6 cells per sample were removed, pelleted, and resuspended in 2 M CM-DCFDA (5-(and -6)-chloromethyl-2 � ,7 � -dichlorodihydro-�uorescein diacetate) (Life Technologies, Grand Island, N�). Cells were placed in a dark room at room temperature while gently rocking for 30 minutes (CM-DCFDA). Following this incubation, samples were kept on ice until counting on a Becton Dickinson FACScan �ow cytometer. 10,000 cells were counted and analyzed by FlowJo 7.6.4 to compare mean values of DCF �uorescence intensity. All samples were repeated in triplicate. Some sample tubes also contained 10 M FCCP (Mesoxalonitrile 4-tri�uoromethoxyphenylhydrazone) and were preincubated with the lymphoblasts prior to the addition of CM-DCFDA. Cybrid cell lines were also treated with CM-DCFDA. Brie�y, adherent cells were grown to 70% con�uency, washed with PBS, and cells were removed from the plate by the addition of trypsin-EDTA (Mediatech, Inc, Manassas, VA). Cells were pelleted at 400 g for 5 minutes at room temperature and the cells were resuspended in CM-DCFDA and assay continued as described above. Mitochondrial Superoxide Assay: Media (RPMI 1640 containing 10% fetal bovine serum) was replaced on cells in growth phase with media containing 5 M MitoSOX (Life Technologies). Cells were incubated in the dark at 37 ∘ C, 5% CO 2 for 10 minutes. Following this incubation, cells were harvested by trypsin digestion, pelleted, and resuspended in HANKS Buffer containing 10% FBS. Samples were kept on ice until counting on a Becton Dickinson LSRII �ow cytometer as described above. NO Assay: Media (RPMI 1640 containing 10% fetal bovine serum) was replaced on cells in growth phase with media containing 4 M 4-amino-5-methylamino-2 � ,7 � -di�uoro�uorescein diacetate (DAF-FM DA) (Life Technologies). Cells were incubated in the dark at 37 ∘ C, 5% CO 2 for 30 minutes, DAF-FM DA was removed and replaced with media and incubated in the dark at 37 ∘ C, 5% CO 2 for a additional 20 minutes. Following this incubation, cells were harvested by trypsin digestion, pelleted, and resuspended in HANKS Buffer containing 10% FBS. Samples were kept on ice until counting on a Becton Dickinson LSRII �ow cytometer as described above. Time between dye loading and reading was 30 minutes and all samples were treated identically. e DAF experiment was done in the presence of serum.  dilution. Immunodetection was completed by using the corresponding secondary horseradish peroxide-conjugated antibodies (Amersham, Piscataway, NJ). Horseradish peroxide activity was detected using enhanced chemiluminescence from ECL Western Blotting Analysis System (Amersham).

Mice.
Male nu/nu mice, 6-8-weeks old, were purchased from Charles River Laboratories (Wilmington, MA) and housed in ventilated cages under sterile conditions. For surgical manipulation, mice were anesthetized with an intramuscular injection of a mixture of ketamine hydrochloride, Xylazine, and Acepromazine.

In Vivo Tumor Study.
Mice were injected subcutaneously in the neck with 3 separate 6124WT clones for a total of 29 mice and 3 separate 6124Mut clones for a total of 30 mice. 2.5 × 10 5 cells, resuspended in PBS, were injected per mouse. Mice were checked daily for tumor growth. When tumors were visually observed they were measured twice a week and the tumor volumes calculated using the formula [length × (width) 2 ]/2 [15]. Experiment was repeated 2 additional times. Mice were sacri�ced when the tumors reached 10% of body weight. Tumor tissues were dissected for further study.

Mutation Analysis.
A 60-year old patient underwent radical prostatectomy aer presenting to his primary care physician with an elevated-serum-prostate-speci�c antigen (PSA = 5.5 ng/mL) that prompted a prostate biopsy revealing a Gleason 6 prostatic adenocarcinoma. Histopathologic examination of the radical prostatectomy specimen revealed a Gleason's score 6 (moderately differentiated-grade G2) conventional (acinar) prostatic adenocarcinoma involving ∼5% of the le lobe of the prostate with negative surgical margins (pathologic stage pT2a). Follow-up of over 7 years with history, physical examinations and serum PSA determinations found the patient to be without evidence of prostate cancer recurrence. ere was no family history of prostate cancer and the patient was otherwise healthy. DNA from the prostate tissue was the �rst specimen to be sequenced from this patient and this demonstrated a heteroplasmic point mutation at nucleotide position (np) 6124 of the mitochondrial genome. e sequencing chromatogram demonstrated approximately equal proportions of the wild type base (T) and mutant base (C) (Figure 1(a)). e mutation changes amino acid 74 from the nonpolar, methionine (sulfur side chain) to the polar, threonine (hydroxyl side chain). is is a highly conserved amino acid, being methionine in 95% of all species ( ) for which the sequence has been determined. is conservation index (CI) of 95 is at the very upper end of what is observed for "adaptive mutations" (CI = 85 ± 9%) and signi�cantly above the CI of "neutral polymorphisms" (CI = 23 ± 5%), making it highly likely to be functional in terms of mitochondrial physiology [1]. Another indicator of the uniqueness of this base and amino acid change is that this patient remains the only example of this base change in MITOMAP and mutation at this nucleotide position has not been found in the 2704 complete mtDNA sequences reported in the online database mtDB [16], where 2704 out of 2704 individuals have the wild type base (T). Subsequent sequencing of patient peripheral blood DNA, lymphoblast cell lines, and laser capture microdissection of various cellular compartments of the prostate all revealed similar levels of BioMed Research International 5 heteroplasmic mutation at this base, con�rming inheritance in the germ line and maintenance of the mutation in the prostate. e entire mitochondrial DNA sequence for this patient was determined and all changes from the revised Cambridge reference sequence are shown in Table 1. e mutation at n.p. 6124 was the only heteroplasmic mutation and the only mutation that had not been previously found in the 2407 complete mtDNA sequences published in the online database mtDB.

Cytochrome Oxidase Enzyme Activity.
In order to study the biochemical consequences of this mutation we studied reactive oxygen production and enzymatic activity of respiratory complex (RC) IV (cytochrome oxidoreductase). e COI polypeptide forms the catalytic core of this enzyme. Cytochrome oxidation was measured in mitochondria isolated from the patient's lymphoblast cell line and compared to other lymphoblast cell lines from two unrelated, individuals' lymphoblast lines, sequence proven to be wild type at the COI locus (Figure 1(b)). e mutation is associated with a 29% decrease in COI activity.

Reactive Oxygen.
Mutations in the COI polypeptide that lead to decreased RCIV activity could potentially lead to increased reactive oxygen [17]. In order to determine if this was occurring, we assayed the lymphoblast cell line of the patient for increased peroxide levels using dichloro�uorescin (DCF) and compared the results to the two-wild type patients described above. ere was a marked increase (1.75 fold) in DCF �uorescence in the lymphoblast cell line containing the mutation compared to wildtype (Figure 1(c), gray bars). e ma�ority of the increase in DCF �uorescence can be attributed to the mitochondria as demonstrated when oxidative phosphorylation is inhibited with FCCP (white bars), DCF �uorescence levels are decreased to similar levels. More detailed analysis of the effect of the 6124 mutation on ROS generation was obtained in the cybrid experiments, results highlighted below ("Cybrid ROS and RNS").

Cybrid Generation.
In order to eliminate the potential confounding effects of the nuclear genome we made cytoplasmic hybrids (cybrids) that combined either pure mutant or pure wild type genomes from this patient with a stable nuclear background (143 B cells). e resultant pair of cybrids thus have the exact same nucleus (different from the patient), and the exact same mitochondrial DNA sequence except for the single base mutation at n.p. 6124. MtDNA genotypes were sequence veri�ed in cybrids.

Cybrid
In Vitro Proliferation. ree cybrids 6124 wild type clones and three 6124 mutant clones were analyzed for proliferation in culture. All three mutant 6124 cybrid clones grew faster than the three wild type clones with an average doubling time of 1.57 ± 0.12 days compared to 2.81 ± 0.56 days, mutant to wild type, respectively, ( 0.0 8) ( Figure  2). T 1: Patient mitochondrial DNA mutations. Patient peripheral blood, lymphoblast, and cybrid mtDNA was sequenced in its entirety. All changes from rCRS are shown as is the region in which the change occurs and the amino acid alteration when applicable. AI (allelic index) is the measure of the frequency of the mutation when compared to mtDB-Human Mitochondrial Genome Database [16]. Cybrid mtDNA was identical to peripheral blood with the exception of n.p. 6124 at which point the cybrids were determined to be either homoplasmic mutant or wild type.

Cybrid ROS and RNS.
We then compared multiple wild type and mutant clones for reactive oxygen species (ROS) and reactive nitrogen species generation using �ow cytometry and ROS and RNS sensitive �uorescent dyes. In order to study the cellular peroxide levels in cells that were either entirely mutant or entirely wild type at this base, cybrid cell lines were made and multiple clones assayed by DCF �uorescence. Overall, 6 individual mutant clones showed a consistent increase in peroxide levels when compared to 5 individual wildtype clones (data not shown). When averaging the relative level of �uorescence in all wildtype clones compared to mutant clones, the mutation is associated with a statistically signi�cant ( ) increase in DCF �uorescence (Figure 3(a)). Similarly, in order to determine mitochondrial superoxide levels, there was a slight overall decrease in mitochondrial superoxide in the 6124Mut cell lines when compared to 6124WT ( ) (Figure 3(b)). A substantial and signi�cant increase was observed in nitric oxide (NO) levels in the same 6124Mut and 6124WT cell lines (Figure 3(c)). When treated with DAF-FM diacetate, 6124Mut cell lines showed an average of 5.2-fold increase in NO levels compared to 6124WT ( ). For comparison, NO was also measured cybrids containing 8993WT or Mut. 8993Mut cells had 1.8-fold higher NO when compared to 8993WT ( ). Finally, there were no differences in the overall levels of hydroxyl radicals and peroxynitrite anions as measured by hydroxyphenyl�uorescein (HPF) (Figure 3(d)).

Nitric Oxide Synthases (NOS) in Cybrids Cell
Lines. In order to determine the possible source of the increased NO, we harvested RNA from the 6124WT and 6124Mut cybrids cell lines, followed by reverse transcription and PCR. ere were no detectable levels of NOS1 or NOS3 observed (data not shown). However, NOS2 was demonstrated to be in both WT and Mut cell lines (Figure 4). e breast cancer cell line BT474 is used as a robustly positive control for NOS2 RNA.
3.8. Apoptosis. PAPR (poly ADP ribose polymerase) cleavage was analyzed to determine if changes in cellular proliferation were in part due to changes in apoptosis. Western blot analysis of PARP cleavage demonstrates that PARP cleavage was reduced in 6124Mut cybrids compared to 6124WT ( Figure 5).

3.9.
In Vivo Proliferation. 6124Mut cybrid cell lines grew faster in vivo when compared to 6124WT cybrids cells ( Figure  6). ere was some variability within each group as to the number and size of tumor, but at all time points there was a statistically signi�cant increase in the size of tumors from 6124Mut cybrids cell lines ( at day 32). e animals injected with cybrids showed no systemic symptoms in response to tumor injection or growth.

Discussion
e mitochondrially encoded COI gene was �rst implicated in cancer biology in 1998 when a somatically acquired chain termination mutation was reported in colon cancer by the Vogelstein group at Johns Hopkins [18]. In 2005 an analysis of 260 patients with prostate cancer revealed that 31 (12%) had an inherited mutation in COI compared to <2% of no-cancer controls. Similarly, a case-control analysis of African-American men revealed that two COI gene single nucleotide polymorphisms (SNPs) (T6221C and T7389C) were signi�cantly associated with prostate cancer ( < ) and in strong linkage disequilibrium with each other ( 2 > 6). It is therefore likely that COI gene mutations predispose individuals to the development of prostate cancer. For this reason we have sequenced the COI gene from 482 prostate cancer patients and identi�ed missense mutations in 116 (24.1%) [19]. e patient reported in this paper is part of that cohort.
e biochemical analysis of mtDNA mutations requires that viable cells are obtained from patients with such mutations and that potentially confounding nuclear events are controlled. is is made possible by the combination of capturing clinically relevant mutations in patient lymphoblast cell lines and the subsequent formation of cybrids with a common nuclear background. e cybrid formation process has the further advantage of allowing the mutant base to be studied separately from the wild type base. is paper documents that a prostate cancer-associated COI mutation affects the normal functioning of respiratory complex (RC) IV (cytochrome oxidase) in at least two distinct ways: decreasing the rate of mitochondrial cytochrome c oxidation and increasing the rate of ROS and NO generation in intact cells. It therefore seems likely that both of these effects are a direct consequence of the mutation. As Figure 3 depicts, not all ROS are affected equally by the 6124 mutation. In particular, peroxides are elevated, superoxide is depressed and there is no difference in hydroxyl radicals. One possible explanation for the decrease in superoxide levels in the presence of abundant superoxide dismutase that results in the rapid conversion of superoxide to hydrogen peroxide. Hydrogen peroxide is stable and able to accumulate while superoxide anion is transient and highly reactive. ). For comparison, 143B cybrids cell lines containing a separate patient mtDNA with either a mutation at position 8993 (8993Mut) or wild type 8993 (8993WT) is shown. Error bars represent the standard error of the mean of 9 data points each 6124WT, 6124Mut, and 3 data points each 8993WT and 8993Mut. (d) Hydroxyl radicals and peroxynitrite anions as measured by hydroxyphenyl�uorescein (HPF) remain unchanged. Error bars represent the standard error of the mean of 9 data points each 6124WT and 6124Mut.
e possible tumorigenic effects of increases in ROS are well known and include (at least) an increased rate of DNA mutations and ROS-induced promitogenic signaling [20][21][22]. ese are common �ndings in prostate cancer and other solid tumors [23,24]. It is likely that the cumulative effect over a lifetime of a tonic increase in cellular ROS increases the chances of malignant transformation. It is uncertain whether the prostate is more susceptible to this in�uence than other  organs, but the association of COI mutations with disease suggests some speci�city. e possible mechanism by which a decreased efficiency of oxidative phosphorylation is related to malignant transformation is less obvious. It is probably relevant to the so called "Warburg Effect" wherein tumor cells exhibit defective oxidative phosphorylation and increased glycolysis as a primary means of ATP generation [25]. It is possible that inherited or somatically acquired mtDNA mutations are partially or wholly responsible for this effect. It is possible that the decrease in oxidative phosphorylation is not in and of itself causally related to the increased risk of cancer, but that the relation to cancer risk is conferred predominantly by the increased ROS generation and that compromises to oxidative phosphorylation are merely a bystander effect of no direct consequence to tumorigenesis. is remains to be determined.
e possible tumorigenic effects of increases in RNS are also well known. Nitric oxide (NO) is generated enzymatically by synthases (NOS), which oxidize L-arginine to Lcitrulline. e inducible form, iNOS, is present in a variety of cell types. Over the past 20 years iNOS expression has been associated with various human tumors including: breast, brain, lung, prostate, colorectal, and melanoma [26][27][28][29][30][31][32]. NO action is concentration-dependent. Increased cGMPmediated ERK phosphorylation is associated with low levels of NO in cancer cells whereas HIF-1a stabilization is associated with intermediate levels [33]. ough NO has been shown to induce apoptosis it can transcriptionally enhance MMP-1 via the ERK and p38 MAPK pathways resulting in tumor progression and also can result in the overproduction of VEGF [33]. Although NO levels are transient, iNOS generated NO �uctuations varying from seconds to days can in�uence the antiapoptotic�proapoptotic effects of NO [33].
We have demonstrated that the T6124C (M74T) mutation was inherited in a heteroplasmic state by a patient that developed prostate cancer and that the mutation not only causes increased ROS and nitric oxide but also induces increased cellular proliferation, decreased apoptosis, and In 2008 Ishikawa demonstrated that mtDNA mutations determined the metastatic potential of lung cancer cell lines independent of the characteristics of the cancer cell nucleus. Speci�cally, high-metastatic cell lines lost their metastatic potential when the mtDNA was replaced with mtDNA from low-metastatic cell lines and low-metastatic cell lines acquired high-metastatic capabilities when mtDNA from high metastatic lines were inserted [34]. e metastatic capacity could be eliminated by treatment with antioxidants indicating the central importance of ROS-induced signaling in metastasis. Similarly, an ATP6 (ATP synthase subunit 6) missense mutation in the mitochondrial genome (T8993G) causes increased ROS in prostate cancer cell lines and increased tumor growth [5].

Conclusions
Mitochondrial DNA from a prostate cancer patient with an inherited-heteroplasmic-mtDNA mutation in COI, the catalytic core of mitochondrial respiratory complex IV was studied. In the laboratory, this mutation was found to simultaneously decrease the activity of the respiratory complex as measured by the rate of cytochrome c oxidation and to increase the rate of mitochondrial reactive oxygen generation. Other mutation-induced biochemical changes included increased generation of nitric oxide. Cells harboring the mutation proliferated faster in vitro and caused increased tumor growth in vivo. ese �ndings suggest a possible molecular substrate (mtDNA mutation) for the "Warburg effect" of anaerobic metabolism exhibited by some tumors and increased cellular reactive oxygen, a common �nding in solid tumors.

Introduction
In 2010, close to 207,090 new cases of IBC were diagnosed in the �nited States, and DCIS was identi�ed in an additional 54,010 women who did not yet have IBC [1]. ese statistics indicate that breast cancer and the oen associated precursor lesion, DCIS, are global health problems.
Although DCIS masses are oen small by comparison to IBC masses, DCIS is typically detected through mammography or self-examination. However, there are signi�cant shortcomings to these methods. Mammography is generally used to identify breast masses within a resolution limit of 1 cm, and the Van Nuys prognostic index (VNPI) for DCIS does not score tumors less than 1.5 cm. is means that a subset of smaller lesions which may have signi�cant future clinical impact remain undetected and/or evaluated. Breast self-examination has also been extensively shown to be an ineffective detection tool for Asian women [2,3].
Currently, there is no clinical means to distinguish between the heterogeneous types of DCIS and recognize the carcinomas that will progress into invasive, metastatic breast cancer. e mechanism that drives this transformation from DCIS to IBC is also not well understood. Hence, when mammography detects DCIS, a full diagnostic workup and treatment is required [4]. As such, a huge need exists for the development of early detection procedures or tools for preinvasive lesions. If a link is established between smaller DCIS lesions and larger IBC lesions, or if a distinction can be made between invasive and noninvasive masses, it may then be possible to apply this knowledge to the development of early detection tools and chemopreventive treatment for women at risk. e progression of DCIS is poorly understood because the technology used to detect it relies on tissue mass. Indeed, the identi�cation of DCIS via mammography is low compared to larger tumors. If a signi�cant proportion of IBC cases originate as DCIS, then successful detection and strat-i�cation of these lesions will assist the clinician and the patient with determining potential monitoring and treatment strategies. A recent review articulated the need for a combined research effort directed towards this clinical need [5].
is study proposes using somatic mitochondrial D-loop mutations in paired samples of DCIS and IBC to identify a potential breast ductal epithelial "cancerization" �eld effect. Alterations in the mitochondrial genome have been chronicled in most solid tumors, including breast cancer [6]. Since the mtGenome has an accelerated mutation rate in association with the beginning or presence of malignant transformation, patient-matched characterization of this genome in both DCIS and IBC may reveal a common related or progressive mutation pattern between these two lesions.
Mitochondrial D-loop mutations can be evaluated using tissue samples from solid tumors. Using bio�uids with low cellularity such as nipple aspirate �uid (NAF) or ductal lavage (DL) represents a much less invasive route for developing early detection tests. e mtGenome is ideal for these investigations because it has a high copy number per cell, when compared to the nuclear archive of DNA.
ere are other characteristics suggesting that the mtGenome may be an ideal "biosensor" as follows: (1) each copy of the mtGenome is clonal; (2) the mtGenome has a maternal inheritance pattern which precludes generational recombination; (3) somatic mutations appearing in a subset of mtGenomes, known as heteroplasmy, afford early disease detection; (4) the modest size of the mtGenome (16,568 bp) allows inexpensive, targeted, and concentrated genetic analyses; (5) the mtGenome has a 10-100-fold copy advantage over the nuclear genome; (6) the mitochondrial organelle is the center of ATP synthesis and is the mediator of cell apoptosis, and for successful tumorigenesis to occur, energy production must be replaced by an alternative process and apoptosis must be by-passed; (7) mitochondrial DNA (mtDNA) has an accelerated somatic mutation rate in which mutations occur within years, and perhaps months, from when molecular pathways are altered by early molecular changes associated with malignant transformation; (8) mutations in the mtGenome have been attested in a wide variety of solid tumors.

Patients and Samples.
Women who were referred to a surgical oncologist for a clinical breast examination and had a biopsy with positive results were recruited to this study. Patients having a biopsy, lumpectomy, or mastectomy were selected based on a pathology report which identi-�ed both DCIS and IBC. Two patients had both a biopsy and a secondary procedure (lumpectomy and mastectomy). All patients were procured in accordance with the ethical guidelines of the under Bay Regional Health Sciences Research Ethics Board in adherence to the Tri-Council Policy Statement on Ethical Conduct for Research Involving Humans. Written consent was obtained from the patients for publication of the study. Patients were selected based on review of biopsy and/or surgical pathology reports. A total of 34 patients were identi�ed, however, upon sectioning of requested samples, only 15 had sufficient quantities of both IBC and DCIS to warrant LCM. Aer complete sample processing (extraction through sequencing), 5 patients were further eliminated due to sample drop-out. A total of 34 samples, including blood, were contributed by a suite of 10 patients (Table 1). Blood from a �nger prick was collected on IsoCode cards (Whatman, Piscataway, NJ). DNA was extracted using a QIAcube (Qiagen, Germantown, MD) and QIAamp DNA mini kit (Qiagen, Germantown, MD) using the protocol for DNA puri�cation for dried blood.

Laser Capture Microdissection.
Requested tissues (biopsy and mastectomy samples) were sectioned from formalin-�xed paraffin-embedded (FFPE) blocks and processed for LCM. LCM was performed by two quali�ed, gowned, gloved, and masked technicians who captured both DCIS and IBC from each patient. By direct observation of the process, about 3-4 cells were harvested per laser pulse, or capture event, and approximately 2,000 captures were recovered from each tissue type. DNA was liberated from LCM samples by an overnight digestion at 65 ∘ C in 50 L of 100 mM Tris-HCl (pH 8.0), 10 mM EDTA, 1% Tween 20, and 20 mg/mL Proteinase K. e following morning, the reactions were inactivated at 95 ∘ C for 10 minutes. A total of 24 FFPE samples were processed. DNA was extracted using a QIAcube and QIAamp DNA mini kit tissue protocol, with the addition of heating each sample at 90 ∘ C for 1 hour aer incubation of the sample at 56 ∘ C with 180 L of Buffer ATL plus 20 L Proteinase K. e samples were eluted in 200 L of Buffer AE. Samples were dried down and resuspended in 30 L of ddH 2 0.

2.�. Mitochondrial D�Loop �mpli�cation.
A portion of the D-loop was ampli�ed with primer sets MT1, 2, 3 forward and reverse (MitoScreen Assay Kit, Transgenomic, Omaha, NE) using the following reagent concentrations per reaction: 1X FastStart High Fidelity Reaction Buffer, 1.8 mM MgCl 2 , and 0.25 U FastStart High Fidelity Enzyme Blend (Roche, Burgess Hill, UK); 0.2 mM of each dNTP; 0.3 M of each primer; 2.5 L of tissue extract or 1 L of DNA recovered from blood, with the �nal reaction volume ad�usted to 25 L with ddH 2 0. Reactions were activated at 95 ∘ C for 6 minutes, then ampli�ed with the following pro�le for 42 cycles: 95 ∘ C, 30 seconds; 56 ∘ C, 30 seconds; 72 ∘ C, 1 minute; followed by a �nal extension for 7 minutes at 72 ∘ C.

Denaturing High Performance Liquid Chromatography
(DHPLC). Following ampli�cation, DHPLC was used to 2.6. Sequencing. Primer set MT2/MT19 (15424-102) was used to generate template for nested ampli�cation with D1 primers (15898-16525). Both sets of primers were tested for null ampli�cation against Rho 0 derived template [9] using the PCR conditions described previously to preclude the possibility of coampli�cation of numts. is mandatory precaution has been chronicled elsewhere [10,11]. In addition, results were compared to the HV1 sequence signature of everyone directly involved with handling the samples to detect any incidental contamination by laboratory personnel. Finally, the corresponding germ plasma-derived DNA was ampli�ed and sequenced from each patient as a direct comparative to control for actual somatic mutations as opposed to maternal variation.
Ampli�ed template was sequenced at Genevision (Newcastle Upon Tyne, UK). Both Geneious bioinformatics soware (Biomatters) and Sequencher 4.5 (Gene Codes) were used for sequence analyses.

Statistical Analyses.
Analyses were performed on HV1 mutation patterns and all applicable parameters listed in the pathology report: age, receptor status, tumor grade, nuclear grade, tubule formation, mitotic score, modi�ed Bloom-Richardson grade, and presence or absence of extensive intraductal component. Attempts to correlate the diagnostic rankings and per-site mutation results were made using point-biserial and rank-biserial statistics. Pearson rank correlation was used to identify the strength of the relationship between HVR1 relative substitution rates and the prevalence of each mutation site in the patient data. IBC and DCIS sample populations were considered separately in order to determine if any patterns existed in the mutation load of the individual sample types as well as to discover the presence T 2: HV1 somatic mutations are bolded, while mutations persisting in all patient samples are also italicized. Patient histologies are compared to the corresponding sequence of their germplasm or blood (B) to detect mutations. Only those sites appearing in all histologies for a given patient are identi�ed .   93 126 188 189 192 203 223 224 249 270 291 298 304 311 319 357 362 390  RCRS  T  T  C  T  C  A  C  T  T  C  C  T  T  T  G  T  T  G  33 B  T  T  C  T  T  A  C  T  T  T  C  T  C  T  G  T  T  G  of any interactions between the two tissue types. Again, Pearson correlations were used as statistics for this analysis.

Results
Mutations were identi�ed in HV1 which was reampli�ed with Rho 0 null primers and sequenced. All patients in this study demonstrate heteroplasmy in all of the associated histologies in comparison to germ plasma, or blood. It is important to note that 18 sites had homoplasmic and/or heteroplasmic mutation sites in common between DCIS and IBC lesions recovered from the same patient. All patients had at least 1 corresponding homoplasmic and/or heteroplasmic site in both DCIS and IBC. ese results parallel similar observations noting that other biomarkers are held in common between DCIS and IBC [12]. Two patients (43 and 74) had equivalent mutations in both biopsy samples and tissue from follow-up procedures (lumpectomy and mastectomy). See Tables 1 and 2 for an overview of clinical pathology and HV1 somatic mutations, respectively. No exogenous contamination from laboratory personnel, via comparison to HV1 sequence from germplasm, was observed.

Clinical Correlation and Mutation
Load. ere appears to be no statistically signi�cant correlation between single individual mutation sites and speci�c gradings� namely, the modi�ed Bloom-Richardson grade, nuclear grade, tubule formation, and mitotic score. e mutation loads of the IBC and DCIS samples were similar, even though up to a third of the mutations for a given patient differed. e average mutation load per patient was the same.
Considering IBC and DCIS mutation load from a persite perspective, the two populations strongly correlate ( 0.929, 0.00 ), meaning that the mutation load at a given site is consistent, regardless of tissue type. is may imply that the same damage is occurring in both tissues and that the disease processes may be similar.

Discussion
e observed frequency of mutations in the study population indicates a medium correlation with the relative mutation rates in HV1. All of the identi�ed sites have estimated relative rates greater than zero, and 65% of the sites are classi�ed as "fast" by multiple studies since they have a greater tendency to mutate than other neighboring sites. Using the same metric (substitution rate >2), 88% of the identi�ed sites could be classi�ed as "fast" [7]. e mutation sites identi�ed by this study appear predisposed towards mutation. Since sites such as 16189 and 16224 are present in almost every patient, they demonstrate near con�uence in this small cohort. is is perhaps due to a biological propensity to rapid mutation. As such, this attribute could be used as a breast cancer marker if this behavior is consistent in transforming breast tissue.
ese results are consistent with a �eld effect demonstrated in epithelial tissues in general, including those cells lining the mammary ducts [13]. is �eld effect was also observed by Xu et al. in a small segment of the D-loop referred to as D310 [14]. is idea is demonstrated in multiple matching heteroplasmic and homoplasmic changes in HV1 in corresponding patient-matched DCIS and IBC samples from 10 patients in the study. A gland-wide in�uence is further suggested by the results of patients 43 and 74. Here, common mutations are observed in tissues from separate clinical procedures. Patient 43 has 3 mutations which occur in both biopsy and lumpectomy samples, in DCIS and IBC captured from biopsy and DCIS taken from a later lumpectomy. Patient 74 has 5 parallel alterations between IBC and DCIS from biopsy and DCIS recovered aer a mastectomy. e IBC from mastectomy share 2 of these sites. is sample also has 2 unique changes.
Unfortunately, only patients 43 and 74 had follow-up procedures allowing this level of comparative analyses. e IBC and DCIS from the remaining 8 study participants were associated with 1 procedure, a biopsy, lumpectomy, or mastectomy. Absence of a 1 : 1 correlation between the mutation patterns of IBC and DCIS for a given patient and between separate procedures is likely a result of capturing ducts from tissue cross-sections and the convoluted anatomy of ductal tissue (i.e., patient 43). e extent and effect of the �eld may vary among associated, parallel ducts. Also, heteroplasmic signal detection up to 20% may not have been reached in all comparative patient samples.
Both telomere content (TC) and allelic imbalance (AI) have been documented in histologically normal breast tissue at 1 cm from a tumor focus. At 5 cm from a focus, TC and AI re�ect normal parameters. is �eld could be much wider than 1 cm, since data was collected only at 1 and 5 cm intervals [15]. Similar epithelial �eld attributes have also been noted in lung cancer [16]. Also, extensive cancerization �elds have been described in both head and neck cancers (7 cm in diameter) and colon cancer (3-10 cm in diameter) [17,18]. e size of these �elds may depend on the biological characteristics of the speci�c biomarkers.
It has been reported that D-loop mutations are associated with tumors which are both estrogen and progesterone receptor negative in women 50 years of age or older [18]. at pattern was not seen here which means that the Dloop alterations identi�ed in this study would be suitable for use in a broad age range of women. Moreover, women with alterations in the D-loop experience poorer outcomes than those free of mutations [19]. is suggests that HV1 mutations found in both DCIS and IBC, when found in patients with DICS only, may be indicators of DCIS with potential aggressive behavior.
In other work, NAF was successfully retrieved from 82% of the participants with 96% yielding �uid from both breasts [20]. Given that the alterations displayed by the mtGenome demonstrate a �eld effect in breast tissue, there is merit in assessing NAF or DL recovered from women with both DCIS and IBC histopathologies. is applies to other abnormal breast histopathologies as well, such as atypical ductal hyperplasia. Both NAF and DL have been investigated as a source of biomarkers and for biological indications of breast cancer [16,[20][21][22][23][24][25][26][27][28][29][30][31]. Given the high copy number of the mtGenome and its rapid mutation rate, sequence analysis of the D-loop may identify mutations associated with these lesions in glandular organ-associated bio�uids which are low in both volume and cellularity. Full mtGenome sequencing was successful for NAF and blood from 19 women referred to a surgical oncologist for a clinical breast examination and who had a nonmalignant outcome. A subset of these patients had a single mutation each (4/19, 21%) in the entire mtGenome. Unfortunately, no follow-up information was available for these women, and thus, comments regarding the association of the mutations observed with a disease state could not be reported.

Conclusions
is study was able to identify mtGenome alterations that occur in both DCIS and IBC within individual patients that are suggestive of a cancerization �eld effect, and DCIS that may be aggressive in nature. Other work demonstrates that large amounts of genetic information can be recovered from the high-copy-number mtGenome in low volume bio�uids [20]. Identi�cation of biomarkers with early detection and/or diagnostic capacity that utilize the mtGenome and its characteristics, in combination with the epithelial �eld effect and the use of NAF and/or DL as the detection medium, may have important clinical applications. Further studies are warranted to help unravel the mechanisms linking DCIS and IBC, as well as the mechanism that drives the transition from the smaller DCIS lesions to larger IBC lesions.

Introduction
Persistent oxidative stress due to reactive oxygen species (ROS) has been associated with carcinogenesis and cancer progression [1][2][3], along with various aggressive cancer cell phenotypes [4]. e superoxide anion (O 2 •− ), the primary type of ROS generated through various cellular metabolic pathways and through exposure to ionizing radiation [5], is converted into hydrogen peroxide (H 2 O 2 ) and hydroxyl radical (OH • ) via biological and antioxidant processes within the cell [6]. e hydroxyl radical, generated through the Fenton, or Haber-Weiss, reaction, is more reactive than either superoxide or hydrogen peroxide and causes direct damage to DNA and other macromolecules [7][8][9], resulting in DNA strand breaks and mutations.
e electron transport chain (ETC), one of the intracellular sources of ROS production, is located in the inner mitochondrial membrane and involved in the production of cellular energy through oxidative phosphorylation [10,11]. Due to its proximity to the ETC, mitochondrial DNA (mtDNA) is sensitive to oxidative stress-related damage, which may be responsible for altered mitochondrial gene expression and somatic mutations in many human cancers [12][13][14][15][16]. A mitochondrial mutator phenotype has been proposed to account for the accumulation of extensive somatic mutations in clinical tumors [17]. Mitochondrial DNA is a supercoiled, closed-circular molecule with multiple copies and an average of 100 negatively superhelical turns [18]. e supercoiled structure has been identi�ed as a functional substrate for mtDNA replication and transcription initiation in cells [19][20][21]. It is thus logical that disruptions to the supercoiled structure (i.e., strand breaks) would have direct effects on mitochondrial bioenergetics and mutagenesis.
In addition to DNA damage, cellular ROS can also induce cell proliferation and toxicity. e high levels of ROS generation and accumulation can lead to cell toxicity and death, making some tumor cells possible targets for ROS-induced apoptosis [22,23], while the low levels of ROS activate signal pathways that lead to cell growth and proliferation [24]. Many human cancer cells, such as prostate, breast, colon, and malignant mesothelioma, as well as mouse colon and liver hepatoma, are shown to have increased levels of indigenous ROS and are thus under persistent oxidative stress [4,[25][26][27]. is may be due to upregulation of membrane-bound NAD(P)H oxidases (NOX) [28], altered energy metabolism associated with mitochondrial dysfunction [12,[29][30][31][32], and reduced antioxidant activities in superoxide dismutase and GSH pathways [6,25]. erefore, a bell-shaped response curve has been proposed to account for the relationship between the level of ROS and the rate of cell proliferation [25]. e increased baseline levels of ROS in tumor cells can lead to differential responses to further oxidative injury as compared to normal cells [25,33].
In this paper, we will introduce the use of real-time PCR as a new method of assessing mitochondrial DNA damage through quanti�cation of damaged forms (relaxed circular and linear) of supercoiled mtDNA, a concept previously introduced by our group [34]; examine some surprising interactions among cytotoxicity, ROS production, and oxidative DNA damage in prostate cancer and normal cells [33]; propose a new paradigm to explain these intriguing phenomena.

A Novel Method for Sensitive �uanti��ation of mt�NA �amage� �epair� and Copy Number Change
e real-time PCR is a valuable tool oen used to quantify starting amounts of nucleic acids in a PCR reaction without post-PCR manipulation [35,36]. For mtDNA quanti�cation, the relative mtDNA content is calculated as the ratio of mtDNA versus a reference nuclear gene. However, we have previously shown that different structural conformations of mtDNA (supercoiled, nicked circular, linear) have different effects on real-time PCR quanti�cation [34]. Based on this observation, we have developed a supercoiling-sensitive quantitative PCR assay (ss-qPCR) to quantify oxidative damage in the supercoiled DNA [34,37]. e principle of this new approach can be illustrated by model molecules. e supercoiled pBR322 plasmid DNA is previously shown to be a reliable model for mtDNA conformational studies [38]. In a comparative analysis undertaken previously [34], both supercoiled DNA molecules were digested with enzymes that altered their supercoiled DNA structure. pBR322 was treated with EcoR 1 (a single restriction site in pBR322 DNA) to generate its linear form and with N.BstNB1 (two nicking sites in pBR322 DNA) to generate a nicked (or relaxed) circular form. Total genomic DNA (containing mtDNA) from the LNCaP cells, a type of androgen responsive prostate cancer cell, was treated with EcoR 1 to linearize mtDNA. Together, two nuclear DNA, multiple mtDNA, and two plasmid DNA markers were analyzed for real-time PCR ampli�cation using the MyiQ real-time PCR system as well as the SYBR Green I intercalation dye [34]. In this analysis, we observed that there was a 6-fold increase in the ampli�cation of nicked circular and linear forms of plasmid DNA as compared to its untreated and supercoiled form and a 2-fold increase in the ampli�cation of the closed-circular form of DNA when the same amount of starting template material was used (see Figure 1C in [34]). A 2-fold increase in ampli�cation was observed in EcoR-1-treated mtDNA as compared to ampli�cation of its untreated form in the same study. �e concluded that the negatively supercoiled structure of DNA was a poor substrate for real-time PCR ampli�cation and that a disruption of the supercoiled structure by either cutting (producing a linear molecule) or nicking (nicked circular form produced by single-strand breaks) the doublestranded molecule signi�cantly increased the e�ciency of qPCR ampli�cation. As such, ampli�cation e�ciency can be used to determine the degree to which a supercoiled DNA sample is damaged. A heat-denaturing step at the onset of qPCR ampli�cation can be used to introduce strand breaks into all initial mtDNA, thus enabling accurate measurement of total initial mtDNA copy number in a sample without interference from the supercoiled structure. Coupled with the quanti�cation of DNA structural damage, the percentage of damaged mtDNA in a total sample can be calculated and the degree of oxidative stress the cell is subjected to can be inferred. A quantitative evaluation of mtDNA degradation through copy number loss can also be achieved with qPCR.
�hile useful, the ss-qPCR protocol used for the quanti�cation of mtDNA damage has the potential of introducing arti�cial strand breaks into the sample. e initial heatdenaturing step at 95 ∘ C for three minutes, while necessary to initiate the ampli�cation process, was observed to significantly increase qPCR ampli�cation in mtDNA [34]. In an effort to reduce artifact, we subsequently developed a new two-step qPCR protocol with a shortened initial denaturing time at 95 ∘ C followed by a signi�cantly lowered denaturing temperature of 80 ∘ C for the remaining cycles [33]. is 2step procedure, by reducing both the duration and intensity of heat imposed upon the DNA substrates, led to a reduced baseline reading of mtDNA damage and increased sensitivity with regard to detecting induced mtDNA damage in different prostate cell lines. Indeed, an almost 2-fold decrease in baseline mtDNA damage levels from 44.2% to 24.6% was detected between the regular protocol and the new two-step protocol (Figure 1(a)) ( Figure 1A in [33]). e total mtDNA content per cell for different prostate cell lines and the number of damaged (or relaxed) mtDNA copies per cell can be determined using an absolute quanti�cation approach (Figure 1(b)) ( Figure 1B in [33]). Although the absolute number of mtDNA copies varies signi�cantly, the percentage of damaged mtDNA is relatively stable across cell lines (Figure 1(c)). As such, we propose to present the level of mtDNA damage as the percentage of damaged mtDNA in the total DNA content as opposed to the absolute number of damaged molecules. Since different cell lines have different amount of mitochondria and mtDNA damage is induced within each mitochondrion, cell lines with increased levels of mitochondria may exhibit greater absolute mtDNA damage when in fact each mitochondrion has the same amount of DNA damage. us, the percentage of damaged mtDNA is independent from the mitochondrial content in each cell. is provides a method of evaluating the constitutively different cell lines and tissues on equal footing and gives F 1: Two-phased, supercoiling-sensitive qPCR for improved mtDNA damage detection [33]. e percentage of relaxed/damaged mtDNA in C4-2 cancer cell line detected by a two-phased protocol and protocols previously reported as Fast and Regular ones (a). e absolute copy numbers of damaged and total mtDNA molecules were detected in normal RWPE-1 and three prostate cancer cell lines (LNCaP, C4-2, and PC-3) (b). e basal levels of mtDNA damage were calculated as the ratio of damaged versus total mtDNA copy numbers (c). Student's t-test was used for signi�cant analysis. ( * , * * ).
a more accurate view into oxidative mtDNA damage between tumor and normal cells.

Differential Responses to Oxidative Injury between Prostate Cancer and Normal Cells
It is increasingly recognized that cellular ROS have a signaling role in stimulating cancer growth [2]. erefore, intrinsic oxidative stress through enhanced levels of endogenous ROS in prostate and other cancers may be a phenotype actively selected for in cancer progression. However, which cellular processes contribute to ROS propagation in cancer and how such changes modify cancer cell responses to further oxidant injury remain to be fully elucidated. Using H 2 O 2 as an exogenous stimulus, we previously investigated differential responses to cell toxicity, cellular ROS production, and oxidative DNA damage between prostate cancer and normal cell lines [33]. We demonstrated that aggressive prostate cancer cells exhibited a low threshold effect and increased susceptibility to extrinsic oxidative injury. propagation speci�c to aggressive cancer cells is likely associated with the activation of NAD(P)H oxidases. Indeed, apocynin, a speci�c inhibitor of NAD(P)H oxidases, markedly reduced •− propagation and cytotoxicity in aggressive C4-2 cells [33]. is �nding is consistent with the stimulating effect of H 2 O 2 on NAD(P)H oxidases-mediated O 2 •− production reported in human SMC, endothelial and cancer cells [39][40][41], and is further supported by upregulation of several isoforms of NAD(P)H oxidases in prostate cancer cell lines and tumor tissues [4,28]. It is interesting to point out that other sources of O 2 •− production downstream of NAD(P)H oxidases may be required to maintain persistent O 2 •− accumulation in aggressive cancer cells, such as impaired mitochondrial respiration and oxidative DNA damage. us, upregulation of NAD(P)H oxidases likely confers an increased metastatic potential through enhanced levels of endogenous ROS in aggressive cancer cells, but subjects the same cells to increased susceptibility to oxidant toxicity through NAD(P)H oxidases-mediated O 2 •− burst.

mtDNA Damage Is Sensitive to Exogenous H 2 O 2 but Independent of Cellular ROS Production
e mitochondrion is both a major source of ROS production and a primary target of oxidative damage in the cell. e supercoiled mtDNA serves as a natural surrogate to oxidative DNA damage due to its close proximity to the site of ROS production [34]. e supercoiling-sensitive qPCR method provides a new opportunity to investigate whether oxidative DNA damage contributes to H 2 O 2 -induced differential cell toxicity or associates with cellular ROS production [33]. •− production in prostate cell lines. is observation is further supported by the lack of induced mtDNA damage in both C4-2 and RWPE-1 cell lines when treated by O 2 •− -producing agents, diphenyleneiodonium (DPI), and rotenone [33]. DPI is known to reduce O 2 •− production by inhibiting NAD(P)H oxidases but to increase cellular O 2 •− by impairing mitochondrial respiration depending on the dose and cell type used [42][43][44]. We demonstrated that DPI induced dose-dependent ROS production and growth inhibition in C4-2 and RWPE-1 cell lines. However, no effect was detected on either mtDNA structural damage or copy number change in both cell lines treated by DPI [33]. is result was further corroborated by the rotenone treatment that targeted speci�cally complex 1 of the mitochondrial ETC [33]. It is conceivable that DPI induces imbalanced O 2 •− accumulation without being converted to HO • through H 2 O 2 . Indeed, DPI has been shown to induce O 2 •− production in many cell types [43,44] but to suppress H 2 O 2 production in several prostate cancer cell lines [4] and in the mitochondria of rat skeletal muscle [45]. us, contrary to the common assumption that cellular O 2 •− is tightly balanced with H 2 O 2 in a cell, we suggest that imbalanced accumulation of different ROS species may occur under stressed conditions, leading to very different functional consequences.

A New Paradigm of Oxidative Injury and Its Implications
We propose a new paradigm of oxidative injury in prostate cancer cells (Figure 2). Aggressive cancer cells exhibit intrinsic oxidative stress based on the type and source of different ROS [4,6,12,[25][26][27][28][29][30][31][32]. We have demonstrated that H 2 O 2 exposure induces differential cell toxicity and sensitive oxidative DNA damage in prostate cancer cells through different mechanisms. A cascade of cellular O 2 •− production is shown to be a critical determinant of selective toxicity in aggressive cancer cells [33], which is mediated by the activation of elevated NAD(P)H oxidases [28] and by crosstalk with impaired mitochondrial respiration in cancer cells [13] (Figure 2(a)). Conversely, the resistance to H 2 O 2 -induced cytotoxicity in normal cells may be attributed to a low cellular level of NAD(P)H oxidases and the normal mitochondrial function (Figure 2(b)). On the other hand, a signi�cant level of oxidative DNA damage is induced by exogenous H 2 O 2 in both cancer and normal cells, but is independent of cellular O 2 •− steady-state levels. We propose that selective accumulation of cellular O 2 •− (e.g., DPI) is cytotoxic regardless of cell types [43] but is independent to HO • -mediated DNA damage [33]. e O 2 •− accumulation may be exacerbated by a difference in the rate of O 2 •− accumulation and conversion to H 2 O 2 and HO • in stressed cancer cells. Alternatively, O 2 •− may be metabolized promptly with other reactive species such as nitric oxide (NO). NO is shown to interact with O 2 •− to generate peroxynitrite anions (ONOO-) and nitrogen oxides (NO ) [22], which could attenuate the formation of the highly reactive HO • and oxidative DNA damage.
e new paradigm has important implications in designing new strategies in cancer prevention and therapy. Selective production of cellular O 2 •− rather than HO • is a promising strategy for preferential killing of aggressive cancer cells. is can be achieved by targeted activation of NAD(P)H oxidases in prostate cancer cells, which may be further sensitized by modulating other sources of cellular O 2 •− production. In contrast, HO • is a potent mutagen that may cause mtDNA damage and copy number depletion under subtoxic conditions; these active responses to DNA damage provide a new explanation to the accumulation of extensive mtDNA damage and somatic mutations in clinical tumors and aging tissues under physiological and/or pathological conditions [11,17]. us, minimizing cellular HO • production is better suited for cancer prevention by reducing the long-term accumulation of oxidative DNA damage and mutagenesis. Besides, the recognition of cellular partitioning and functional separation of major ROS in prostate cancer cells will likely shed new lights on the evolution of aggressive phenotype in prostate cancer. Additional investigations are desirable to test the applicability of the implications in other human cancers.
e development of a very sensitive approach to analyze structure-mediated DNA damage using real-time PCR provides a powerful, quantitative new approach to the study of mtDNA damage, repair, and copy number change in a single test [34]. is quantitative approach is in contrast to the semiquantitative analysis on mtDNA structural damage based on gel electrophoresis and southern blot [46]. erefore, this new technical platform may �nd broad applications to study oxidative stress in cultured cells, clinical samples, and model animals. Indeed, mtDNA may serve as a sensitive surrogate for precise quanti�cation of oxidative DNA damage locally in diseased tissues and systemically in circulating blood of cancer patients. We have developed a comprehensive strategy to measure multiple mtDNA endpoints in circulating lymphocytes to study systemic stress in clinical investigations [47] and to study the in�uence of microsurgical varicocelectomy on human sperm mtDNA copy number [48]. Finally, this method also has the potential to quantify speci�c oxidative base lesions accumulated in supercoiled mtDNA when coupled with lesion-speci�c repair enzymes.

Conclusion
Cellular ROS are natural byproducts of metabolic processes, but persistent accumulation of ROS can lead to cellular oxidative injury, including DNA damage and cell toxicity. Damage to mtDNA results in the loss of its supercoiled structure, which is readily detectable with a twostep, supercoiling-sensitive qPCR assay. Our previous studies have demonstrated that mtDNA damage is very sensitive to exogenous H 2 O 2 but independent of endogenous ROS accumulation in both prostate cancer and normal cells. In contrast, aggressive prostate cancer cells exhibit a more than 10-fold sensitivity to H 2 O 2 -induced cell toxicity than normal cells, suggesting a very different mechanism of action. We propose a new paradigm to account for different mechanisms governing oxidative stress, cell toxicity and DNA damage with important rami�cations in devising new techniques and strategies in cancer prevention and treatment.

Introduction
Activation of the phosphatidylinositol 3-kinase (PI3K)/AKT pathway is thought to play a crucial role in the development of a variety of human cancers. Several academic efforts are underway to de�ne therapeutic inhibitors of the pathway components [1,2]. PI3K interacts with phosphatidylinositol-3-phosphate at the cell membrane and catalyzes the phosphorylation of downstream effector(s) such as Akt [1]. Class IA PI3Ks, consisting of a catalytic subunits p110 and regulatory subunit (p85, p55, and p50), play a critical role in cell proliferation and cell survival [3][4][5][6].
e p55PIK, also known as p55 , is encoded by the pik3r3 gene [7,8]. We previously reported that p55PIK, the N-terminal 24-amino-acid of which is associated with tumor suppressor retinoblastoma protein (Rb), may play an important role in cell cycle control [9]. Ectopic expression of N-terminal 24-amino-acid of p55PIK inhibited cell cycle progression in several cell lines, such as colorectal (HT29) and thyroid (FTC236) cancer cells [10]. One study reported that elevated p55PIK mRNA expression was observed in ovarian, liver, prostate, and breast cancers. A recent study showed that Rb protein can localize to the mitochondria in proliferative cells [11]. Aberrant p55PIK expression may contributes to mitochondrial dysfunction in cancer progression. In addition, apoptosis was observed in p55PIK downregulated ovarian cancer cell lines [12]. Further more, the insulin-like growth factor 2 (IGF2)-p55PIK interaction involved in promoting the growth of a subset of proliferative glioblastomas that lack EGF receptor ampli�cation [13]. ese �ndings suggest that p55PIK was aberrantly expressed in several human cancers and p55PIK may act as an important target for cancer treatment. e detailed mechanism, especially its transcriptional regulation mechanism, remains unknown. Disclosure of the factor(s) that contribute to the regulation of p55PIK expression may be useful in cancer treatment targeting p55PIK. Primer name Sequence pGL3-p55PIK-(−651/+45) up Several factor(s) have been reported to regulate p55PIK expression in different human disease models, such as mammary cancer [14] and cerebral ischemia-reperfusion [15]. One study has shown that p55PIK expression increased in the presence of doxorubicin, an anthracycline antibiotic that is used abroad in cancer chemotherapy, in breast cancer MDA-MB-231 cells but not in MCF-7 cells [14]. e genetic factors altered in MDA-MB-231 cells, which are p53/estrogen receptor/progesterone receptor negative, may be involved in regulation of p55PIK expression. Another study reported that the Insulin-like growth factor 2 (IGF2) and p55PIK are overexpressed in more proliferative glioblastomas [13]. In fatty acid and cholesterol biosynthesis, Sterol-regulatory element binding protein-1 (REBP-1) and Platelet-derived growth factor (PDGF) induce the expression of p55PIK in AG01518 human foreskin �broblasts [16]. A recent study revealed that berberine, an effective candidate neuroprotective agent in clinical ischemic stroke, enhances p55PIK promoter activity during cerebral ischemia-reperfusion [15]. In Mycobacterium tuberculosis model of WI-38 cells, downregulated p55PIK expression was observed by recombinant Mycobacterium tuberculosis CFP-10/ESAT-6 protein treatment [17]. Despite the clari�cation of these factors, little is known about the mechanism of p55PIK transcriptional regulation.
e aim of the present study is to identify the ciselements and transcription factor(s) involved in p55PIK transcriptional activation in colorectal cancer cells (CRCs). Firstly, we made in silico analysis and deletion analysis of the p55PIK gene promoter and determined the transcriptional factor(s) that may regulate p55PIK transcription. We also evaluated the in�uence of the transcriptional factors(s) on PI3K expression and the cell growth of CRC cells. Based on the results of this study, the transcription factor(s)-p55PIK axis may be suggested as the potentially crucial target(s) of CRC treatment.

Ethics Statement.
All research involving human participants has been approved by the Huazhong University of Science and Technology Ethics committee. We obtained informed, written consent from all participants involved in this study.

Cell Culture and Transfection.
Cell lines HepG2, HeLa, SW480, and LoVo were purchased from the American Type Culture Collection (Manassas, VA, USA) and cultured in DMEM supplemented with 10% fetal bovine serum (HyClone, Logan, UT, USA). ese cell lines were cultured at 37 ∘ C in 5% CO 2 /air atmosphere. Transfection was done using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions.

Small
Interfering RNA. Synthetic siRNA targeting human MZF1 (RuiBo, Guangzhou, China) was transfected into cultured cells. Transfection was done using Lipofectamine 2000 following manufacturer's instructions. Cells were cultured in 24-well plates in antibiotic-free 10% fetal bovine serum plus medium and transfected with 50 nmol/L siRNA at 70-80% con�uency. Expression of MZF1 or p55PIK was detected at 24 h or 48 h aer-transfection.

Site-Directed Mutagenesis.
Constructs bearing mutant promoter variants of p55PIK were generated by PCR using the wildtype p55PIK reporter construct (−1243/+45)-p55PIK as template. Underlined nucleotides in Table 2 indicate mutated sequences. Primers were designed according to manufacturer's instructions and produced by Invitrogen. Site-directed mutagenesis was done according to manufacturer's protocol for the Quick Change site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA). All mutants of (−1243/+45)-p55PIK were veri�ed correctly by sequencing.
2.6. Dual-Luciferase Reporter Gene Assay. Cells were seeded in 24-well plates. Aer culturing for 24 h, cells were cotransfected with luciferase reporter plasmids and Renilla vector (pRL-TK) (Promega, Madison, WI, U.S.A). Luciferase activities were measured at 24 h post-transfection, using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, U.S.A). Luciferase activity was normalized for transfection efficiency using the corresponding Renilla luciferase activity. All experiments were performed independently at least four times.
C T G C C T T T G T C C C G C T T A G T A A C T G G G C T C T G A C C A A T C G C C G C C T T A C C GGCCGGCGCCGGCTGTCTCCTGATTGGCCCGCTTGGGGCGCGCAGCCGTCTCCCGCT C C T T C G C A G G A G C A A G G C A G A C A A A A G A C G C A G G C T C C A G G C T C C G G C A A C C G G C C T C A G C C A A T C A C C A C C C A C C T T G C C T C T C T T C C C T C G C C C C A C T C C C T C C T C G C C T G G C T C G C G T G C T C C C G G C A C T G A C T C C T G G A  and the −1243/−840 region were responsible for p55PIK gene transcription. We mapped the p55PIK promoter to �nd its most activated fragments. A series of 5 ′ �anking of p55PIK promoter DNA fragments (−1633/+45, −1243/+45, −1064/+45, −839/+45, and −651/+45) were cloned into luciferase-reporter construct and the relative luciferase activity was measured in cancer cells. e results show that −1243/+45 fragment confers highest promoter activity and −840/+45 fragment presents a signi�cant decrease compared with −1243/+45 fragment. e current study indicate that the cis-elements were located at the −1243/−840 region of p55PIK promoter.

3.�. Identi�cation of Cis�Acting Elements Controlling p55PIK
Expression. To identify the critical cis-enhancing elements in the −1243/−840 region, we generated various mutant reporter based on the (−1243/+45)-p55PIK plasmid with substitution mutations by site-directed mutagenesis. As shown in Figure 2(a), the potential transcription factor binding sites of YY1, MZF1, Runx1, ADR1, IRF1, Delta1, and p300 were found in the −1243/−840 region based on motif analysis. e introduction of a TGGGGA site mutation (from TGGGGA to CTAGTG) which located at −901/−896 markedly reduced the luciferase activity of (−1243/+45)-p55PIK (Figure 2(d)), whereas mutation of YY1, RUNX1, or other MZF1 binding sites did not affect the promoter activity of (−1243/+45)-p55PIK (Figures 2(b), 2(c)). ese results demonstrated that the putative MZF1 binding site TGGGGA located at −901/−896 region was crucial for functioning of the p55PIK promoter. e focus of the second set of experiments was to identify the cis-element(s) of p55PIK promoter and corresponding transcription factor(s). e preliminary sequence analysis of the domain −1243/−840 reveals the presence of several nonoverlapping cis-elements and corresponding transcription factors. It was reported that MZF1 binds to the 5 ′ -AGTGGGGA-3 ′ or 5 ′ -CGGGnGAGGGGGAA-3 ′ sequence of gene promoters to regulate the expression of a target gene [19]. YY1, which may bind to the 5 ′ -ACCATTC-3 ′ site of the p55PIK promoter, is a ubiquitously distributed zinc-�nger-type transcription factor, involved in regulating a variety of promoters [16][17][18]. Runx1, which may bind to the 5 ′ -CACCACCC-3 ′ sequence of the p55PIK promoter, is essential for hematopoietic development [19,20]. Interferon regulatory factor 1 (IRF1), which may interact with the 5 ′ -AGACGC-3 ′ DNA binding site, was initially described as a transcription factor able to activate expression of the cytokine interferon beta. IRF-1 plays important roles in immune response [21,22], apoptosis [23,24], and tumor suppression [25]. e p300/CBP coactivator family interacts with transcription factors p53 [26] and STAT3 [27] to transcriptionally activate the expression of their target genes. Only the mutant MZF1-mut1 shows decreased luciferase activity compared with the wildtype promoter plasmids.

MZF1 Binding on the TGGGGA Site Located at −901/−896
Region of the p55PIK Promoter in Colon Cancer Cell and Tissues. Next, to further verify whether MZF1 is involved in p55PIK transcription, we employed primers spanning the putative MZF1-binding site of the p55PIK promoter to perform chromatin immunoprecipitation (ChIP) assays, con�rming the presence of endogenous MZF1 bound to this region in CRC cell line SW480 or CRC tissues. In Figure  2(e), the speci�c sequence within p55PIK promoter was precipitated from cell lysates by anti-MZF1 antibody but not by control IgG in both CRC cell line SW480 and CRC tissues. us, the data strongly indicate that MZF1 binds to TGGGGA domain in the proximal promoter of p55PIK in CRC cells or tissues.

p55PIK Is Transcriptionally Activated by MZF1.
To assess the role of MZF1 in transcriptional activity of p55PIK, we measured the luciferase activity of the p55PIK promoter construct aer transfection of MZF1 expression plasmids or MZF1-SiRNA in CRC cell SW480 and LoVo. Increased luciferase activity was observed in MZF1 overexpressed cells (Figure 3(a)). Figure 3(b) shows decreased luciferase activity aer MZF1 SiRNA transfection. p55PIK mRNA expression was evaluated in MZF1 over-expressing or MZF1-silencing CRC cells. As shown in Figures 3(c) and 3(d), p55PIK mRNA increased aer MZF1 was overexpressed and decreased aer MZF1 was silenced, respectively. By Western Blot analysis, MZF1 over-expressing CRC cell SW480 and LoVo showed increased p55PIK expression (Figures 3(e), 3(f)). Altogether, these �ndings indicate that MZF1 functions at least in part as a transcriptional regulator of p55PIK.

Transcriptional Activation of p55PIK by MZF1 Resulting in Accelerated Cell Proliferation in CRC Cell
Lines. Myeloid �inc �nger 1 (MZF1), a transcription factor belonging to the Kr�ppel �inc �nger protein family, was previously reported as an important factor whose aberrant expression disturbed hematopoietic cell proliferation and cell tumorigenesis [19,28,29]. Transcription factor MZF1, binding on the DNA-binding consensus sequence of 5 ′ -AGTGGGGA-3 ′ or 5 ′ -CGGGnGAGGGGGAA-3 ′ [19], regulated several genes' expression which play important role in cancer migration, invasion, or cancer differentiation. Human liver cancer cell line treated with MZF1 antisense oligonucleotide showed repressed protein kinase C expression and inhibited subcutaneous tumor growth in nude mice [30]. MZF1 transcriptionally regulates Axl receptor tyrosine kinase gene in human colon cancer or cervical cancer, which induced migration and metastasis of colon cancer in vitro and in vivo [31]. mRNA expression of MZF1 and A�L, with signi�cant correlation, were both upregulated in colorectal cancer [31]. Raised cell cycling and loss of contact inhibition were detected in MZF-1 overexpressed NIH 3T3 cells [19]. erefore, MZF1 may function as an oncogene in solid cancer. Next, to determine the effects of the MZF1 induced p55PIK transcriptional activation on the growth, the CRC cell lines were examined with the MTT assay. We found that MZF1-GFP could induce acceleration of the proliferation of CRC (Figure 4(a)) and that MZF1-SiRNA could induce inhibition of growth of CRC (Figure 4(b)) as well.

Relationship between Expression of MZF1 and p55PIK
in CRC Tissues. Finally, to demonstrate the relationship between MZF1 and p55PIK expression in human CRC, we examined endogenous expression of MZF1 and p55PIK in 10 resected CRC tissue samples and corresponding normal mucosal tissues from the same patient received therapeutic surgery. Tumors from 7 of 10 patients showed signi�cant increased expression of MZF1 and p55PIK compared with normal tissues ( and , resp.); MZF1 gene expression was positively and signi�cant correlated with p55PIK expression in the resected tumor tissues (Rs = 0.94; ), indicating that MZF1 and p55PIK are involved in tumorigenesis ( Figure 5).

Conclusion
In summary, we have shown that the transcription factor MZF1, which directly binds to its cis-element within the p55PIK promoter, activates p55PIK expression and acts as a growth accelerator in CRC cells. We also demonstrate that the expression of MZF1 and p55PIK is signi�cant correlated, and they are both overexpressed in resected CRC tissues. is investigation may suggest a strategy for development of therapies on p55PIK-associated cancer especially mitochondrion associated cancer.