- Email: [email protected]
Plasma-derived cell-free mitochondrial DNA: A novel non-invasive methodology to identify mitochondrial DNA haplogroups in humans
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Plasma-derived cell-free mitochondrial DNA: A novel non-invasive methodology to identify mitochondrial DNA haplogroups in humans ⁎
Christopher Newella, , Stacey Humeb, Steven C. Greenwayc,e, Lynn Podemskib, Jane Shearerd, Aneal Khana,e a
Department of Medical Genetics, Cumming School of Medicine, University of Calgary, Alberta, Canada Department of Medical Genetics, University of Alberta, Alberta, Canada c Department of Cardiac Sciences, Libin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta, Canada d Faculty of Kinesiology, University of Calgary, Alberta, Canada e Department of Pediatrics, Alberta Children's Hospital Research Institute, Cumming School of Medicine, University of Calgary, Alberta, Canada b
A R T I C LE I N FO
A B S T R A C T
Keywords: Mitochondrial disease Cell-free mitochondrial DNA mtDNA Mutations mtDNA Haplotype
Background: Mitochondrial diseases are a clinically heterogeneous group of diseases caused by mutations in either nuclear or mitochondrial DNA (mtDNA). The diagnosis is challenging and has frequently required a tissue biopsy to obtain a sufficient quantity of mtDNA. Less-invasive sources mtDNA, such as peripheral blood leukocytes, urine sediment, or buccal swab, contain a lower quantity of mtDNA compared to tissue sources which may reduce sensitivity. Cellular apoptosis of tissues and hematopoetic cells releases fragments of DNA and mtDNA into the circulation and these molecules can be extracted from plasma as cell-free DNA (cfDNA). However, entire mtDNA has not been successfully identified from the cell free fraction previously. We hypothesized that the circular nature of mtDNA would prevent its degradation and a higher sensitivity method, such as next generation sequencing, could identify intact cf-mtDNA from human plasma. Methods: Plasma was obtained from patients with mitochondrial disease diagnosed from skeletal muscle biopsy (n = 7) and healthy controls (n = 7) using a specially cfDNA collection tube (Streck Inc.; La Vista, NE). To demonstrate the presence of mtDNA within these samples, we amplified the isolated DNA using custom PCR primers specific to overlapping fragments of mtDNA. cfDNA samples were then sequenced using the Illumina MiSeq sequencing platform. Results: We confirmed the presence of mtDNA, demonstrating that the full mitochondrial genome is in fact present within the cell-free plasma fraction of human blood. Sequencing identified the mitochondrial haplogroup matching with the tissue specimen for all patients. Conclusion: We report the existence of full length mtDNA in cell-free human plasma that was successfully used to perform haplogroup matching. Clinical applications for this work include patient monitoring for heteroplasmy status after mitochondrially-targeted therapies or haplogroup monitoring as a measure of stem cell transplantation.
1. Introduction Mitochondria are cellular organelles primarily responsible for the generation of ATP via oxidative phosphorylation (OXPHOS). The number of mitochondria within a given cell depends upon its metabolic requirements, ranging from tens to hundreds per cell [1]. Furthermore, each mitochondrion contains tens to hundreds of copies of a subject's mitochondrial DNA (mtDNA) [2]. The mtDNA encodes for 37 genes, 13 of which are proteins that are part of the OXPHOS machinery [3] and several mitochondrially derived peptides including humanin and
⁎
MOTS-c [4,5]. Differing from nuclear DNA, mtDNA is much smaller (~16.6 Kb vs. ~3.3 Mb), is arranged in a circular double-stranded form, lacks histones, relies on a single enzyme for replication (DNA polymerase γ; POLG), is maternally inherited, and lacks non-coding introns [3,6–8]. Some of these factors, alongside the close proximity to superoxide byproducts of OXPHOS, contribute to the high frequency of mutations which can accumulate in mtDNA [9]. To combat the high error rate and to manage the large pool of mtDNA that resides in each cell, the mitochondria undergo an organelle-specific form of autophagy termed mitophagy. This recycling
Correspondence: Department of Medical Genetics, Cumming School of Medicine, University of Calgary, 3330 Hospital Dr. NW, Calgary, AB T2N 4N1, Canada. E-mail address: [email protected] (C. Newell).
https://doi.org/10.1016/j.ymgme.2018.10.002 Received 16 July 2018; Received in revised form 10 October 2018; Accepted 10 October 2018 1096-7192/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: Newell, C., Molecular Genetics and Metabolism, https://doi.org/10.1016/j.ymgme.2018.10.002
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
mechanism helps to regulate the ratio of healthy and mutated mtDNA, known as DNA heteroplasmy [10]. The point in which the mtDNA mutation load exceeds the naturally defined threshold for that cell or tissue results in pathogenic manifestations. Mitochondrial disease encompasses a group of genetic metabolic disorders which are caused by nDNA or mtDNA mutations and affect at least 1 in 5000 individuals. Approximately 15% of mitochondrial disease cases are a result of mtDNA mutations [11]. Clinically, mitochondrial diseases manifest at any age, impact any tissue, and may involve either route of genetic inheritance (nDNA or mtDNA transmission). These challenges further complicate diagnosing patients with mitochondrial diseases. Clinical testing for mitochondrial disease is a laborious task which may involve multiple tissue biopsies due to the variability of mitochondrial populations between tissues. There is no gold standard for diagnosing mitochondrial disease. Primarily, this workup involves invasive muscle and skin biopsies in combination with buccal swabs and blood draws. Testing for mutations in nDNA can be accomplished using peripheral blood leukocytes using standard methods [12]. Mutations in mtDNA; however, are more problematic for detection. First, the mutation load can be variable due to mtDNA heteroplasmy, resulting in sampling bias and potential misrepresentation of mutation load. Second, muscle, skin, buccal and hematopoietic cells only contain cells from a single tissue source and may not represent mtDNA defects from another tissue source, such as the heart, which requires more invasive methods of tissue biopsy that most clinicians would consider too high a risk for diagnostic testing. Cell-free DNA (cfDNA) is the collection of nucleic acids found in the circulatory system. A by-product of cellular apoptosis and natural cell turnover, these ~150–200 base pair fragments are released from both hematopoietic and non-hematopoietic cells [13,14]. Current applications of cfDNA for use as a clinical biomarker have been applied to the fields of cancer biology, organ transplantation, and prenatal screening [15–17]. Application of this cellular process to study mtDNA may therefore enable a blood draw to assess mtDNA from cell sources other than leukocytes. Interestingly, no research has investigated whether full length mtDNA in present within the cfDNA fraction. As we suspected that the circular nature of mtDNA is protective, our objective was to confirm that entire cell-free mitochondrial DNA (cf-mtDNA) can be obtained from plasma.
Table 1 Patient characteristics at time of diagnosis confirmation. ID
Age (years)
Sex
Diagnosis
Heteroplasmy
P1
55
M
< 10%
P2
54
F
P3 P4
56 48
M M
P5
48
M
P6 P7
80 59
F M
Mitochondrial Deletion Syndrome (m.8215_m.16532del)⁎ Mitochondrial Deletion Syndrome (m.8753_m.16566del) Kearns-Sayre Syndrome Mitochondrial Deletion Syndrome (m.6342_m.14004del) Mitochondrial Deletion Syndrome (m.9090_m.16070del)⁎ MELAS (m.3243A > G) MELAS (m.3243A > G)
< 10% N/A 30% < 10% < 15% < 15%
Patient diagnoses and disease heteroplasmy levels confirmed using the Illumina MiSeq sequencing platform are denoted by *, the remaining patients were confirmed using Sanger sequencing. N/A: Heteroplasmy levels for P3 were clinically reported as unknown. MELAS; Mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes.
left undisturbed to prevent leukocyte DNA contamination of plasma. Aliquoted plasma was then centrifuged again at 13,000 rpm and 4 °C for 15 min, with the supernatant being aliquoted into a new 2 ml tube and stored at −80 °C until further use. cfDNA was isolated from 1.5 ml of thawed plasma using the QIAamp Circulating Nucleic Acid kit with modifications (Qiagen; Germantown, MD). Isolated cfDNA samples were quantified using the Qubit dsDNA assay (Life Technologies; Burlington, ON) and stored at −80 °C until further use. 2.3. Cell-free mitochondrial DNA (cf-mtDNA) amplification Custom PCR primers were designed to span the mtDNA in two overlapping fragments, with primers synthesized by the University of Calgary Core DNA Services (Table S1). Takara LA Taq was used in the reaction mixture as it contains a DNA proofreading polymerase commonly used to amplify large DNA fragments (Takara Bio; Mountain View, CA). PCR was performed in 25 μl reaction volumes using ~2.5 ng of isolated cfDNA on a Mastercycler ep Gradient S (Eppendorf; Mississauga, ON). PCR conditions were as follows: 94 °C-5 min, 30 cycles of (98 °C × 15 s, 68 °C × 10 s, 76 °C × 15 s, 68 °C × 11 min), then 72 °C × 10 min. Amplified PCR products were run undiluted on 0.8% agarose gels containing ethidium bromide (0.5 μg/ml).
2. Materials and methods 2.1. Subjects and tissues
2.4. Library preparation and sequencing All experimental procedures for this study were carried out following approval from the University of Calgary's Conjoint Health Research Ethics Board with informed consent obtained from all study subjects. Each subject provided written consent in accordance with the Declaration of Helsinki and the study protocol was approved by the University of Calgary's Conjoint Health Research Ethics Board (REB130753). Seven patients aged 48 to 80 years of age (5 males, 2 females) were seen in the Metabolic Clinics at the Alberta Children's Hospital (Calgary, Alberta, Canada) and investigated for mitochondrial disease (Table 1). Data were provided by the Metabolic Clinic at Alberta Children's Hospital (Calgary, Alberta, Canada) in conjunction with the Molecular Diagnostics Laboratory, Genetic Laboratory Services, Alberta Health Services (Edmonton, Alberta, Canada). Seven healthy volunteers (3 males, 4 females) aged 26 to 52 years of age were also recruited.
Extracted cfDNA and cf-mtDNA were quantified using the Qubit dsDNA assay (Life Technologies), and amplicon size was verified using a TapeStation D1000 (Agilent Technologies; Santa Clara, CA). A maximum of 1 μg of each sample was sonically sheared with the addition of 1× TE buffer, with equal amounts of each fragment being used for each sample. Sequencing libraries were prepared using the NEBNext Ultra II DNA library preparation kit for Illumina, per the manufacturer's protocol (New England BioLabs; Ipswich, MA). The products were then quantified using the Qubit dsDNA assay (Life Technologies). Amplicon size was assessed using the TapeStation D1000 preparation (Agilent Technologies) and samples were then normalized to 4 nM. Barcoded libraries were then pooled before being denatured and diluted to a final concentration of 10 pM per target. cfDNA samples were prepared using the MiSeq v3 150 cycle sequencing kit with a paired-end read length of 2 × 75 bp, whereas cf-mtDNA samples were prepared using the MiSeq v2 Nano 300 cycle sequencing kit with a paired-end read length of 2 × 150 bp. All samples were run using the Illumina MiSeq instrument (Illumina Inc.; San Diego, CA) with cfDNA samples being run on a standard flow cell and cf-mtDNA samples on a nano flow cell. The resulting sequences were mapped with Burrows-Wheeler Aligner 0.6.1 to the GRCh37, with the Revised Cambridge Reference Sequence (rCRS)
2.2. Plasma cell-free DNA (cfDNA) isolation Whole blood (3–5 ml) was drawn from a peripheral vein using standard phlebotomy techniques into a cfDNA blood collection tube (Streck Inc.; La Vista, NE). Within 1-h of collection, the blood was centrifuged at 4000 rpm and 4 °C for 15 min, the plasma was then removed and transferred into a sterile 2 ml tube(s). The buffy coat was 2
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
for mtDNA. 2.5. mtDNA haplogroup Each haplogroup was assigned using a custom pipeline integrating the HaploGrep2 workflow [18]. The subject's haplotype, as generated by HaploGrep2 according to PhyloTree Build 16 [19], was then linked to the MITOMASTER database to compare frequency information for local and global private mutations in Genbank [20]. Sequence alignment data were exported as individual Binary Alignment/Map (BAM) formatted files and uploaded into the mtDNA-Server to determine data quality and to examine heteroplasmic variant calling [21]. 2.6. Statistical analyses Statistical analysis was performed using GraphPad Prism for Windows, Version 7.02 (GraphPad Software Inc.; La Jolla, CA). Differences between treatment groups were determined by MannWhitney U tests assuming and non-Gaussian distribution where p < 0.05 was significant. Data are expressed as mean ± SEM. 3. Results 3.1. Total cell-free DNA (cfDNA) The baseline cfDNA concentration comparing control and patient plasma samples is reported in Fig. 1. The results indicate that patients with mitochondrial disease have an elevated level of circulating nucleic acids compared to healthy controls.
Fig. 2. A schematic representation of the two overlapping amplicons mapped to the mitochondrial genome following PCR.
mtDNA in the cfDNA fraction (n = 3 per group). Since both nDNA and mtDNA are known to be present in whole cfDNA the MiSeq v3 150 cycle sequencing kit was used to provide appropriate sequencing read depth. Each sequenced cfDNA sample had coverage spanning the entire mitochondrial genome, suggesting that full length mtDNA was present in cfDNA isolates. However, the percentage of total mapped reads to nonmitochondrial DNA, presumably nDNA, (> 96%) was vastly greater than the mtDNA due to utilization of whole cfDNA (Fig. 4A). Next, PCR amplified cf-mtDNA samples from the same cohort of patients and controls were sequenced to provide improved sequencing efficiency and read depth to the mitochondrial reference genome (n = 7 per group). Long range amplification of two mitochondrial fragments in cfDNA increased the proportion of total mapped reads for mtDNA to a similar percentage (ie. ~96%) as identified for non-mitochondrial DNA using raw cfDNA samples (Fig. 4B). mtDNA coverage for both cfDNA and cfmtDNA samples followed a Poisson distribution (data not shown).
3.2. Amplification of cf-mtDNA Isolated cfDNA samples were amplified using PCR in two overlapping fragments spanning the mitochondrial genome, as represented in Fig. 2. Following amplification, MTL1 (9239 bp in size) and MTL2 (11,216 bp in size) amplicons were resolved on representative 0.8% agarose gels to confirm successful mtDNA amplification. Amplified samples from a representative set of control and patient samples each migrated to their appropriate band on the protein ladder (slightly below 10 Kb for MTL1 and slightly above 10 Kb for MTL2) (Fig. 3). The term cf-mtDNA refers to cfDNA isolates following MTL1/2 PCR amplification specific to the mitochondrial genome. 3.3. cfDNA and cf-mtDNA sequencing First, unamplified cfDNA samples from healthy controls and mitochondrial disease patients were sequenced to ensure presence of full
3.4. mtDNA haplogroup analysis Patient haplogroup data was collected from previous clinical diagnostic reports performed at the Genetic Laboratory Services, Alberta Health Services, Molecular Diagnostics Laboratory. As part of the diagnostic pathway for a suspected mitochondrial disease, tissues from skeletal muscle biopsy, buccal mucosa swab, or blood leukocytes were collected and DNA was extracted. After mtDNA genome sequencing the variants and were annotated by assigning a mitochondrial haplogroup (Table 2). This clinical data was used to compare the accuracy and quality of variant calling to the respective cf-mtDNA sample. Each of the cf-mtDNA (n = 7) samples corroborated clinical haplogroup findings. The haplogroup for cf-mtDNA (n = 7) samples from healthy controls was also obtained, although the diagnostic laboratory had not previously sequenced these samples. Mapping and quantification of mtDNA variants was used to identify how closely clinical tissues and cfmtDNA yielded the same mtDNA haplogroup (Table 2). Interestingly, there was variability in the number of detected variants from each
Fig. 1. A comparison of total cell-free DNA (cfDNA) between control and patient samples. cfDNA was isolated and quantified from blood plasma of 7 healthy control and 7 patients with mitochondrial disease. Data are presented as mean ± SEM with * indicating a significant difference between control and patient samples at p < 0.05. 3
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
Fig. 3. Demonstration of cell-free mitochondrial DNA (cf-mtDNA) amplification in two overlapping fragments. Each panel depicts a series of alternating healthy control (C1-C3) and mitochondrial disease patient (P1-P3) samples following PCR amplification using the custom PCR primer pairs. MTL1 product of ~9239 bp (A) and MTL2 product of ~11,216 bp.
Fig. 4. Percent of sequencing reads mapping back to the non-mitochondrial (presumably nDNA) and mitochondrial (mtDNA) genomes. Panel A describes the percent of non-mitochondrial DNA and mtDNA reads after sequencing cfDNA samples of both healthy control and mitochondrial disease patients (n = 3 per group). Panel B describes the percent of non-mitochondrial DNA and mtDNA reads after sequencing cf-mtDNA samples of both healthy control and mitochondrial disease patients (n = 7 per group). All sequencing was performed using the Illumina MiSeq platform at the University of Calgary Core DNA Services facilities. Data are presented as mean ± SEM with * indicating a significant difference between non-mitochondrial DNA and mtDNA respective to control or patient samples at p < 0.001.
tissue. These can likely be the result of the different environmental conditions for cf-mtDNA (in circulation) and the lack of pressure to conserve as non-functional DNA. 4. Discussion Diagnosing mitochondrial disease is particularly challenging as mutations may arise in two distinct genomes, and additionally mitochondrial genome mutations may be acquired in a tissue-specific manner and may exhibit varying degrees of heteroplasmy [22]. While peripheral blood leukocytes, urine sediment, skin fibroblasts and buccal swab samples can be incorporated fairly easily into the diagnostic workflow, skeletal muscle biopsy remains the gold standard for obtaining mtDNA [23]. Our internal data also shows that muscle mtDNA has a higher sensitivity than peripheral blood with several cases where the diagnosis was missed on blood DNA analysis alone. However, obtaining a muscle biopsy is an invasive procedure and in children, invariably requires sedation with a general anesthetic. Furthermore, muscle as a single tissue source for mtDNA may not represent disease manifestations in other organs such as the liver and heart. With the development of NGS technologies, nuclear DNA diagnoses can be typically obtained with peripheral blood DNA reducing the need for muscle biopsy but the issue of obtaining sufficient mtDNA without muscle tissue remains unresolved. We hypothesized that if mtDNA could be obtained from plasma with a more sensitive technique, such as NGS, then even fewer cases would require invasive tissue biopsy. The methods we employed suggest that this may be possible. Firstly, there are limitations to the presented work that should be addressed. The primary limitation in this study is the small number of
Table 2 Mitochondrial haplogroup assessed using mtDNA sequence variants.
P1 P2 P3 P4 P5 P6 P7
Patient haplogroup (muscle)
Patient haplogroup (cf-mtDNA)
Variant total (muscle)
Variant total (cf-mtDNA)
H16b K1a4a1a K1a4a1a H1a3c T2b5 U5a1b3b U5a1b3
H16b K1a4a1a K1a4a1 H1a3c T2b5 U5a1b3 U5a1b3
33 36 38 19 37 25 23
34 37 38 17 34 27 27
mtDNA haplogroup for healthy control and mitochondrial disease patients performed using a custom pipeline incorporating the existing HaploGrep2 workflow. Each clinically derived patient haplogroup was compared to cfmtDNA samples. Mapping of variants to the mitochondrial genome was used to identify how closely each group of variants mapped to the expected haplogroup variant list. No differences were detected between controls and patients. Clinical haplogroup were collected from skeletal muscle biopsies (n = 5), buccal mucosa swab (n = 1), or blood leukocytes (n = 1). a P3 didn't consent to a muscle biopsy and therefore the haplotype and variant total were derived from leukocyte DNA. b P6 also didn't undergo a muscle biopsy and the haplotype and variant total were derived using buccal mucosal DNA. 4
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
in > 96% of sequencing reads subsequently mapping to the rCRS for the mitochondrial genome. This finding confirms that cfDNA cannot be used to diagnose a mitochondrial-genome disorder without prior amplification of the mitochondrial genome. The next question was whether cf-mtDNA variants could properly identify the patient haplogroup when compared to the gold standard using skeletal muscle biopsy. Haplogroup analysis was performed on sequenced cf-mtDNA samples, comparing data to previously collected clinical haplogroup results from skeletal muscle biopsies (n = 5), buccal mucosa swab (n = 1), or blood leukocytes (n = 1). We expected that the total number of variants detected in the cf-mtDNA may not match the variants detected in clinical tissues, caused by missing variants in cf-mtDNA or extra variants due to differences in the compared tissues. This hypothesis proved correct since the variant number between ct-mtDNA and clinical tissues was different (Table 2). A difference in variants found by cf-mtDNA can likely of low allelic fraction and therefore are attributed to somatically acquired. However, each mtDNA haplogroup from cf-mtDNA correctly matched the mtDNA haplogroup from the clinical gold standard using skeletal muscle tissue, or buccal mucosa swab, or blood leukocytes. These findings demonstrate that cf-mtDNA from plasma can accurately determine mtDNA haplogroup in patients with mitochondrial disease. Beyond utility to the field of evolutionary genetics, mtDNA haplogrouping provides valuable genetic information for potential therapeutic and diagnostic opportunities in patients with mitochondrial disease. Interestingly, recent research in liver cell therapy utilized cfDNA to measure response to therapy, comparing host and donor nDNA levels up to 6 months following cell transplantation [16]. In a similar manner, sequencing cf-mtDNA could be clinically applied to the identification of donor versus host mtDNA in response to cell therapy. Prior to cell therapy, haplogroup analysis of the donor should confirm a sufficiently different haplogroup to permit differentiation of donor from patient. In summary, this technique is advantageous as it represents a noninvasive method to provide mtDNA from multiple tissue sources for genetic analyses which should theoretically increase the sensitivity for the detection of mitochondrial disease. Further research efforts will develop cf-mtDNA as a prospective method for the evaluation of mtDNA haplogroup variants and response to future transplantation or mitochondrial therapies. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ymgme.2018.10.002.
control and patient samples that were analyzed (n = 7 per group). Although the prevalence of rare diseases inhibits researchers from achieving large sample sizes as documented in more common diseases, our small sample size makes it difficult to conclude whether the effect would be reproducible in a larger patient cohort. Collecting and analyzing samples from a multitude of haplotypes would also help to ensure that each unique haplotype can be identified using our proposed method. Secondly, there remains risk of contamination during both the blood collection and plasma processing. To ensure that the current method is avoiding leukocyte contamination, samples could be compared from a separate cohort of individuals using a sample from each of the blood draw regions and compare these findings to exclude leukocyte contamination. Initially identified in 1948, circulating nucleic acids have been suggested as a clinical diagnostic tool since the application to cancer in 1996 [24]. Since this time the application of cfDNA in diagnostics has broadened to include the fields of organ transplantation and prenatal screening [15,16]. These applications have focused on the quantification of cfDNA, the identification of traditional genetic mutations and the detection of epigenetic patterning [25–27]. Although cfDNA was previously known to be increased in cancer patients [28], we decided to investigate cfDNA levels in patients with mitochondrial disease. Our mitochondrial disease patients had significantly elevated cfDNA concentrations compared to healthy controls suggesting higher mtDNA turnover. Building upon the differences in cfDNA between patients and controls, we investigated the presence of mtDNA within the cfDNA fraction. Previous research indicated that both nuclear- and mitochondrialderived cfDNA fragments are identifiable using commercially available cfDNA isolation protocols [26,29]. Although there are unique structural and functional differences between nDNA and mtDNA, existing research has focused on short-length mtDNA amplicons, similar to the investigation of nDNA from cfDNA. Since mtDNA is a circular doublestranded molecule and it lacks histones, we hypothesized that intact mtDNA persists in the cfDNA fraction as the circular nature of intact mtDNA delays the rapid degradation of cfDNA by circulating nucleases [30]. The lack of histones also contributes by forgoing the targeted excision of DNA in the classical ~150–200 bp fragment pattern seen in nDNA, thus enabling mtDNA to remain whole [13]. As we wanted to ensure that we isolated mitochondria from the cell-free portion of the plasma, precautions were taken to avoid contamination with mitochondrial DNA from intact leukocytes. First, a specific cfDNA blood collection tube (Streck Inc.) was utilized to collect blood. These tubes contain a proprietary cell preservative which stabilizes nucleated cells and thereby prevents the release of nDNA and mtDNA for up to 14 days after collection [31]. Second, a ~10 mm gap between the supernatant plasma and buffy coat was left undisturbed to prevent mechanical contamination of plasma with disturbed leukocytes. Using these precautions, our results show that full-length cf-mtDNA can be amplified from plasma of both mitochondrial disease patients and healthy controls. These data, provide support to previous work confirming that mtDNA exists in both particle-associated and non-particle associated forms in cfDNA [32]. Population geneticists use mitochondrial haplogroups to map the migration of humans throughout the world. This collection of maternally inherited mtDNA variants are usually homoplasmic changes associated with a branch of the mitochondrial phylogenetic tree [33]. Originally ranging from letters A-Z, mtDNA haplogroups are now further defined by their sub-haplogroup due to the recent abundance of data provided by whole mtDNA genome sequencing studies [19]. We decided to use this evolutionarily conserved feature and sequence both cfDNA and cf-mtDNA samples, using the mtDNA haplogroup to then differentiate between individuals. Initially comparing whole cfDNA to cf-mtDNA, we demonstrated that a significant percentage of mapped reads corresponded to non-mitochondrial DNA (> 96%) in the cfDNA samples. Amplification of mtDNA circumvented this problem, resulting
Author contributions CN, SCG and AK designed and developed the research. CN, SH, LP, SCG and AK conducted experiments, collected and analyzed data. CN, JS and AK wrote the manuscript and all authors read and approved the final manuscript.
Competing interests The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. CN, SH, SCG and AK have applied for a patent related to the described methodology.
Funding This study was supported by the Alberta Children's Hospital Research Foundation, Department of Pediatrics and MitoCanada (AK). This research was supported by PhD funding to CN from MitoCanada and an Alberta Innovates – Health Solutions MD/PhD Studentship. 5
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
Ethics approval statement [17]
All experimental procedures for this study were carried out in accordance with the recommendations of the University of Calgary's Conjoint Health Research Ethics Board (REB13-0753). Patient consent statement All subjects gave written informed consent in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).
[18]
Acknowledgements [19]
The authors would like to thank Dr. Richard Pon, Shelly Wegener and Dr. Paul Gordon at the University of Calgary Core DNA Services for sample sequencing, and Elizabeth Newell for editing an earlier draft of this manuscript.
[20]
References
[21]
[1] A.M. D'Erchia, A. Atlante, G. Gadaleta, G. Pavesi, M. Chiara, C. De Virgilio, C. Manzari, F. Mastropasqua, G.M. Prazzoli, E. Picardi, C. Gissi, D. Horner, A. Reyes, E. Sbisà, A. Tullo, G. Pesole, Tissue-specific mtDNA abundance from exome data and its correlation with mitochondrial transcription, mass and respiratory activity, Mitochondrion 20 (2015) 13–21, https://doi.org/10.1016/j.mito.2014.10.005. [2] N.R. Phillips, M.L. Sprouse, R.K. Roby, Simultaneous quantification of mitochondrial DNA copy number and deletion ratio: a multiplex real-time PCR assay, Sci. Rep. 4 (2014) 3887, , https://doi.org/10.1038/srep03887. [3] J.-W. Taanman, The mitochondrial genome: structure, transcription, translation and replication, Biochim. Biophys. Acta Bioenerg. 1410 (1999) 103–123, https:// doi.org/10.1016/S0005-2728(98)00161-3. [4] C. Lee, J. Zeng, B.G. Drew, T. Sallam, A. Martin-Montalvo, J. Wan, S.-J. Kim, H. Mehta, A.L. Hevener, R. de Cabo, P. Cohen, The mitochondrial-derived peptide MOTS-c promotes metabolic homeostasis and reduces obesity and insulin resistance, Cell Metab. 21 (2015) 443–454, https://doi.org/10.1016/j.cmet.2015.02. 009. [5] C. Lee, K. Yen, P. Cohen, Humanin: a harbinger of mitochondrial-derived peptides? Trends Endocrinol. Metab. 24 (2013) 222–228, https://doi.org/10.1016/j.tem. 2013.01.005. [6] J.D. Stumpf, R.P. Saneto, W.C. Copeland, Clinical and molecular features of POLGrelated mitochondrial disease, Cold Spring Harb. Perspect. Biol. 5 (2013) a011395, https://doi.org/10.1101/cshperspect.a011395. [7] R.W. Taylor, D.M. Turnbull, Mitochondrial DNA mutations in human disease, Nat. Rev. Genet. 6 (2005) 389–402, https://doi.org/10.1038/nrg1606. [8] F.M. Yakes, B. Van Houten, Mitochondrial DNA damage is more extensive and persists longer than nuclear DNA damage in human cells following oxidative stress, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 514–519 http://www.ncbi.nlm.nih.gov/ pubmed/9012815 (accessed April 2, 2017). [9] S.D. Cline, Mitochondrial DNA damage and its consequences for mitochondrial gene expression, Biochim. Biophys. Acta 1819 (2012) 979–991, https://doi.org/10. 1016/j.bbagrm.2012.06.002. [10] D.C. Wallace, D. Chalkia, Mitochondrial DNA genetics and the heteroplasmy conundrum in evolution and disease, Cold Spring Harb. Perspect. Biol. 5 (2013) a021220, https://doi.org/10.1101/cshperspect.a021220. [11] S. Dimauro, G. Davidzon, Mitochondrial DNA and disease, Ann. Med. 37 (2005) 222–232, https://doi.org/10.1080/07853890510007368. [12] C.-C. Liew, J. Ma, H.-C. Tang, R. Zheng, A.A. Dempsey, The peripheral blood transcriptome dynamically reflects system wide biology: a potential diagnostic tool, J. Lab. Clin. Med. 147 (2006) 126–132, https://doi.org/10.1016/j.lab.2005.10. 005. [13] M.W. Snyder, M. Kircher, A.J. Hill, R.M. Daza, J. Shendure, Cell-free DNA comprises an in vivo nucleosome footprint that informs its tissues-of-origin, Cell 164 (2016) 57–68, https://doi.org/10.1016/j.cell.2015.11.050. [14] J.C.H. Tsang, Y.M. Dennis Lo, Circulating nucleic acids in plasma/serum, Pathology 39 (2007) 197–207, https://doi.org/10.1080/00313020701230831. [15] A.B. Sparks, C.A. Struble, E.T. Wang, K. Song, A. Oliphant, Noninvasive prenatal detection and selective analysis of cell-free DNA obtained from maternal blood: evaluation for trisomy 21 and trisomy 18, Am. J. Obstet. Gynecol. 206 (2012) 319.e1–319.e9, https://doi.org/10.1016/j.ajog.2012.01.030. [16] P.M.K. Gordon, A. Khan, U. Sajid, N. Chang, V. Suresh, L. Dimnik, R.E. Lamont, J.S. Parboosingh, S.R. Martin, R.T. Pon, J. Weatherhead, S. Wegener, D. Isaac, S.C. Greenway, An algorithm measuring donor cell-free DNA in plasma of cellular and solid organ transplant recipients that does not require donor or recipient
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
6
genotyping, Front. Cardiovasc. Med. 3 (2016) 33, , https://doi.org/10.3389/fcvm. 2016.00033. C. Bettegowda, M. Sausen, R.J. Leary, I. Kinde, Y. Wang, N. Agrawal, B.R. Bartlett, H. Wang, B. Luber, R.M. Alani, E.S. Antonarakis, N.S. Azad, A. Bardelli, H. Brem, J.L. Cameron, C.C. Lee, L.A. Fecher, G.L. Gallia, P. Gibbs, D. Le, R.L. Giuntoli, M. Goggins, M.D. Hogarty, M. Holdhoff, S.-M. Hong, Y. Jiao, H.H. Juhl, J.J. Kim, G. Siravegna, D.A. Laheru, C. Lauricella, M. Lim, E.J. Lipson, S.K.N. Marie, G.J. Netto, K.S. Oliner, A. Olivi, L. Olsson, G.J. Riggins, A. Sartore-Bianchi, K. Schmidt, l.-M. Shih, S.M. Oba-Shinjo, S. Siena, D. Theodorescu, J. Tie, T.T. Harkins, S. Veronese, T.-L. Wang, J.D. Weingart, C.L. Wolfgang, L.D. Wood, D. Xing, R.H. Hruban, J. Wu, P.J. Allen, C.M. Schmidt, M.A. Choti, V.E. Velculescu, K.W. Kinzler, B. Vogelstein, N. Papadopoulos, L.A. Diaz, Detection of circulating tumor DNA in early- and late-stage human malignancies, Sci. Transl. Med. 6 (2014), https://doi.org/10.1126/scitranslmed.3007094 224ra24-224ra24. H. Weissensteiner, D. Pacher, A. Kloss-Brandstätter, L. Forer, G. Specht, H.J. Bandelt, F. Kronenberg, A. Salas, S. Schönherr, HaploGrep 2: mitochondrial haplogroup classification in the era of high-throughput sequencing, Nucleic Acids Res. 44 (2016) W58–W63, https://doi.org/10.1093/nar/gkw233. M. van Oven, M. Kayser, Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation, Hum. Mutat. 30 (2009) E386–E394, https://doi.org/ 10.1002/humu.20921. M.C. Brandon, E. Ruiz-Pesini, D. Mishmar, V. Procaccio, M.T. Lott, K.C. Nguyen, S. Spolim, U. Patil, P. Baldi, D.C. Wallace, MITOMASTER: a bioinformatics tool for the analysis of mitochondrial DNA sequences, Hum. Mutat. 30 (2009) 1–6, https:// doi.org/10.1002/humu.20801. H. Weissensteiner, L. Forer, C. Fuchsberger, B. Schöpf, A. Kloss-Brandstätter, G. Specht, F. Kronenberg, S. Schönherr, mtDNA-Server: next-generation sequencing data analysis of human mitochondrial DNA in the cloud, Nucleic Acids Res. 44 (2016) W64–W69, https://doi.org/10.1093/nar/gkw247. C.L. Alston, M.C. Rocha, N.Z. Lax, D.M. Turnbull, R.W. Taylor, The genetics and pathology of mitochondrial disease, J. Pathol. 241 (2017) 236–250, https://doi. org/10.1002/path.4809. R.H. Haas, S. Parikh, M.J. Falk, R.P. Saneto, N.I. Wolf, N. Darin, L.-J. Wong, B.H. Cohen, R.K. Naviaux, The in-depth evaluation of suspected mitochondrial disease, Mol. Genet. Metab. 94 (2008) 16–37, https://doi.org/10.1016/j.ymgme. 2007.11.018. A.K.C. Chan, R.W.K. Chiu, Y.M.D. Lo, Clinical Sciences Reviews Committee of the Association of Clinical Biochemists, Cell-free nucleic acids in plasma, serum and urine: a new tool in molecular diagnosis, Ann. Clin. Biochem. 40 (2003) 122–130, https://doi.org/10.1258/000456303763046030. E. Bohers, P.J. Viailly, S. Dubois, P. Bertrand, C. Maingonnat, S. Mareschal, P. Ruminy, J.-M. Picquenot, C. Bastard, F. Desmots, T. Fest, K. Leroy, H. Tilly, F. Jardin, Somatic mutations of cell-free circulating DNA detected by next-generation sequencing reflect the genetic changes in both germinal center B-cell-like and activated B-cell-like diffuse large B-cell lymphomas at the time of diagnosis, Haematologica 100 (2015) e280–e284, https://doi.org/10.3324/haematol.2015. 123612. C. Kohler, R. Radpour, Z. Barekati, R. Asadollahi, J. Bitzer, E. Wight, N. Bürki, C. Diesch, W. Holzgreve, X. Zhong, Levels of plasma circulating cell free nuclear and mitochondrial DNA as potential biomarkers for breast tumors, Mol. Cancer 8 (2009) 105, https://doi.org/10.1186/1476-4598-8-105. R. Lehmann-Werman, D. Neiman, H. Zemmour, J. Moss, J. Magenheim, A. VakninDembinsky, S. Rubertsson, B. Nellgård, K. Blennow, H. Zetterberg, K. Spalding, M.J. Haller, C.H. Wasserfall, D.A. Schatz, C.J. Greenbaum, C. Dorrell, M. Grompe, A. Zick, A. Hubert, M. Maoz, V. Fendrich, D.K. Bartsch, T. Golan, S.A. Ben Sasson, G. Zamir, A. Razin, H. Cedar, A.M.J. Shapiro, B. Glaser, R. Shemer, Y. Dor, Identification of tissue-specific cell death using methylation patterns of circulating DNA, Proc. Natl. Acad. Sci. 113 (2016) E1826–E1834, https://doi.org/10.1073/ pnas.1519286113. X.Y. Zhong, A. Ladewig, S. Schmid, E. Wight, S. Hahn, W. Holzgreve, Elevated level of cell-free plasma DNA is associated with breast cancer, Arch. Gynecol. Obstet. 276 (2007) 327–331, https://doi.org/10.1007/s00404-007-0345-1. L. Li, H.-W. Hann, S. Wan, R.S. Hann, C. Wang, Y. Lai, X. Ye, A. Evans, R.E. Myers, Z. Ye, B. Li, J. Xing, H. Yang, Cell-free circulating mitochondrial DNA content and risk of hepatocellular carcinoma in patients with chronic HBV infection, Sci. Rep. 6 (2016) 23992, , https://doi.org/10.1038/srep23992. Y.M.D. Lo, J. Zhang, T.N. Leung, T.K. Lau, A.M.Z. Chang, N.M. Hjelm, Rapid clearance of fetal DNA from maternal plasma, Am. J. Hum. Genet. 64 (1999) 218–224, https://doi.org/10.1086/302205. S.E. Norton, J.M. Lechner, T. Williams, M.R. Fernando, A stabilizing reagent prevents cell-free DNA contamination by cellular DNA in plasma during blood sample storage and shipping as determined by digital PCR, Clin. Biochem. 46 (2013) 1561–1565, https://doi.org/10.1016/j.clinbiochem.2013.06.002. R.W.K. Chiu, L.Y.S. Chan, N.Y.L. Lam, N.B.Y. Tsui, E.K.O. Ng, T.H. Rainer, Y.M.D. Lo, Quantitative analysis of circulating mitochondrial DNA in plasma, Clin. Chem. 49 (2003), http://clinchem.aaccjnls.org/content/49/5/719.long (accessed April 11, 2017). J.B. Stewart, P.F. Chinnery, The dynamics of mitochondrial DNA heteroplasmy: implications for human health and disease, Nat. Publ. Gr. 16 (2015), https://doi. org/10.1038/nrg3966.
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
Plasma-derived cell-free mitochondrial DNA: A novel non-invasive methodology to identify mitochondrial DNA haplogroups in humans ⁎
Christopher Newella, , Stacey Humeb, Steven C. Greenwayc,e, Lynn Podemskib, Jane Shearerd, Aneal Khana,e a
Department of Medical Genetics, Cumming School of Medicine, University of Calgary, Alberta, Canada Department of Medical Genetics, University of Alberta, Alberta, Canada c Department of Cardiac Sciences, Libin Cardiovascular Institute of Alberta, University of Calgary, Calgary, Alberta, Canada d Faculty of Kinesiology, University of Calgary, Alberta, Canada e Department of Pediatrics, Alberta Children's Hospital Research Institute, Cumming School of Medicine, University of Calgary, Alberta, Canada b
A R T I C LE I N FO
A B S T R A C T
Keywords: Mitochondrial disease Cell-free mitochondrial DNA mtDNA Mutations mtDNA Haplotype
Background: Mitochondrial diseases are a clinically heterogeneous group of diseases caused by mutations in either nuclear or mitochondrial DNA (mtDNA). The diagnosis is challenging and has frequently required a tissue biopsy to obtain a sufficient quantity of mtDNA. Less-invasive sources mtDNA, such as peripheral blood leukocytes, urine sediment, or buccal swab, contain a lower quantity of mtDNA compared to tissue sources which may reduce sensitivity. Cellular apoptosis of tissues and hematopoetic cells releases fragments of DNA and mtDNA into the circulation and these molecules can be extracted from plasma as cell-free DNA (cfDNA). However, entire mtDNA has not been successfully identified from the cell free fraction previously. We hypothesized that the circular nature of mtDNA would prevent its degradation and a higher sensitivity method, such as next generation sequencing, could identify intact cf-mtDNA from human plasma. Methods: Plasma was obtained from patients with mitochondrial disease diagnosed from skeletal muscle biopsy (n = 7) and healthy controls (n = 7) using a specially cfDNA collection tube (Streck Inc.; La Vista, NE). To demonstrate the presence of mtDNA within these samples, we amplified the isolated DNA using custom PCR primers specific to overlapping fragments of mtDNA. cfDNA samples were then sequenced using the Illumina MiSeq sequencing platform. Results: We confirmed the presence of mtDNA, demonstrating that the full mitochondrial genome is in fact present within the cell-free plasma fraction of human blood. Sequencing identified the mitochondrial haplogroup matching with the tissue specimen for all patients. Conclusion: We report the existence of full length mtDNA in cell-free human plasma that was successfully used to perform haplogroup matching. Clinical applications for this work include patient monitoring for heteroplasmy status after mitochondrially-targeted therapies or haplogroup monitoring as a measure of stem cell transplantation.
1. Introduction Mitochondria are cellular organelles primarily responsible for the generation of ATP via oxidative phosphorylation (OXPHOS). The number of mitochondria within a given cell depends upon its metabolic requirements, ranging from tens to hundreds per cell [1]. Furthermore, each mitochondrion contains tens to hundreds of copies of a subject's mitochondrial DNA (mtDNA) [2]. The mtDNA encodes for 37 genes, 13 of which are proteins that are part of the OXPHOS machinery [3] and several mitochondrially derived peptides including humanin and
⁎
MOTS-c [4,5]. Differing from nuclear DNA, mtDNA is much smaller (~16.6 Kb vs. ~3.3 Mb), is arranged in a circular double-stranded form, lacks histones, relies on a single enzyme for replication (DNA polymerase γ; POLG), is maternally inherited, and lacks non-coding introns [3,6–8]. Some of these factors, alongside the close proximity to superoxide byproducts of OXPHOS, contribute to the high frequency of mutations which can accumulate in mtDNA [9]. To combat the high error rate and to manage the large pool of mtDNA that resides in each cell, the mitochondria undergo an organelle-specific form of autophagy termed mitophagy. This recycling
Correspondence: Department of Medical Genetics, Cumming School of Medicine, University of Calgary, 3330 Hospital Dr. NW, Calgary, AB T2N 4N1, Canada. E-mail address: [email protected] (C. Newell).
https://doi.org/10.1016/j.ymgme.2018.10.002 Received 16 July 2018; Received in revised form 10 October 2018; Accepted 10 October 2018 1096-7192/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: Newell, C., Molecular Genetics and Metabolism, https://doi.org/10.1016/j.ymgme.2018.10.002
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
mechanism helps to regulate the ratio of healthy and mutated mtDNA, known as DNA heteroplasmy [10]. The point in which the mtDNA mutation load exceeds the naturally defined threshold for that cell or tissue results in pathogenic manifestations. Mitochondrial disease encompasses a group of genetic metabolic disorders which are caused by nDNA or mtDNA mutations and affect at least 1 in 5000 individuals. Approximately 15% of mitochondrial disease cases are a result of mtDNA mutations [11]. Clinically, mitochondrial diseases manifest at any age, impact any tissue, and may involve either route of genetic inheritance (nDNA or mtDNA transmission). These challenges further complicate diagnosing patients with mitochondrial diseases. Clinical testing for mitochondrial disease is a laborious task which may involve multiple tissue biopsies due to the variability of mitochondrial populations between tissues. There is no gold standard for diagnosing mitochondrial disease. Primarily, this workup involves invasive muscle and skin biopsies in combination with buccal swabs and blood draws. Testing for mutations in nDNA can be accomplished using peripheral blood leukocytes using standard methods [12]. Mutations in mtDNA; however, are more problematic for detection. First, the mutation load can be variable due to mtDNA heteroplasmy, resulting in sampling bias and potential misrepresentation of mutation load. Second, muscle, skin, buccal and hematopoietic cells only contain cells from a single tissue source and may not represent mtDNA defects from another tissue source, such as the heart, which requires more invasive methods of tissue biopsy that most clinicians would consider too high a risk for diagnostic testing. Cell-free DNA (cfDNA) is the collection of nucleic acids found in the circulatory system. A by-product of cellular apoptosis and natural cell turnover, these ~150–200 base pair fragments are released from both hematopoietic and non-hematopoietic cells [13,14]. Current applications of cfDNA for use as a clinical biomarker have been applied to the fields of cancer biology, organ transplantation, and prenatal screening [15–17]. Application of this cellular process to study mtDNA may therefore enable a blood draw to assess mtDNA from cell sources other than leukocytes. Interestingly, no research has investigated whether full length mtDNA in present within the cfDNA fraction. As we suspected that the circular nature of mtDNA is protective, our objective was to confirm that entire cell-free mitochondrial DNA (cf-mtDNA) can be obtained from plasma.
Table 1 Patient characteristics at time of diagnosis confirmation. ID
Age (years)
Sex
Diagnosis
Heteroplasmy
P1
55
M
< 10%
P2
54
F
P3 P4
56 48
M M
P5
48
M
P6 P7
80 59
F M
Mitochondrial Deletion Syndrome (m.8215_m.16532del)⁎ Mitochondrial Deletion Syndrome (m.8753_m.16566del) Kearns-Sayre Syndrome Mitochondrial Deletion Syndrome (m.6342_m.14004del) Mitochondrial Deletion Syndrome (m.9090_m.16070del)⁎ MELAS (m.3243A > G) MELAS (m.3243A > G)
< 10% N/A 30% < 10% < 15% < 15%
Patient diagnoses and disease heteroplasmy levels confirmed using the Illumina MiSeq sequencing platform are denoted by *, the remaining patients were confirmed using Sanger sequencing. N/A: Heteroplasmy levels for P3 were clinically reported as unknown. MELAS; Mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes.
left undisturbed to prevent leukocyte DNA contamination of plasma. Aliquoted plasma was then centrifuged again at 13,000 rpm and 4 °C for 15 min, with the supernatant being aliquoted into a new 2 ml tube and stored at −80 °C until further use. cfDNA was isolated from 1.5 ml of thawed plasma using the QIAamp Circulating Nucleic Acid kit with modifications (Qiagen; Germantown, MD). Isolated cfDNA samples were quantified using the Qubit dsDNA assay (Life Technologies; Burlington, ON) and stored at −80 °C until further use. 2.3. Cell-free mitochondrial DNA (cf-mtDNA) amplification Custom PCR primers were designed to span the mtDNA in two overlapping fragments, with primers synthesized by the University of Calgary Core DNA Services (Table S1). Takara LA Taq was used in the reaction mixture as it contains a DNA proofreading polymerase commonly used to amplify large DNA fragments (Takara Bio; Mountain View, CA). PCR was performed in 25 μl reaction volumes using ~2.5 ng of isolated cfDNA on a Mastercycler ep Gradient S (Eppendorf; Mississauga, ON). PCR conditions were as follows: 94 °C-5 min, 30 cycles of (98 °C × 15 s, 68 °C × 10 s, 76 °C × 15 s, 68 °C × 11 min), then 72 °C × 10 min. Amplified PCR products were run undiluted on 0.8% agarose gels containing ethidium bromide (0.5 μg/ml).
2. Materials and methods 2.1. Subjects and tissues
2.4. Library preparation and sequencing All experimental procedures for this study were carried out following approval from the University of Calgary's Conjoint Health Research Ethics Board with informed consent obtained from all study subjects. Each subject provided written consent in accordance with the Declaration of Helsinki and the study protocol was approved by the University of Calgary's Conjoint Health Research Ethics Board (REB130753). Seven patients aged 48 to 80 years of age (5 males, 2 females) were seen in the Metabolic Clinics at the Alberta Children's Hospital (Calgary, Alberta, Canada) and investigated for mitochondrial disease (Table 1). Data were provided by the Metabolic Clinic at Alberta Children's Hospital (Calgary, Alberta, Canada) in conjunction with the Molecular Diagnostics Laboratory, Genetic Laboratory Services, Alberta Health Services (Edmonton, Alberta, Canada). Seven healthy volunteers (3 males, 4 females) aged 26 to 52 years of age were also recruited.
Extracted cfDNA and cf-mtDNA were quantified using the Qubit dsDNA assay (Life Technologies), and amplicon size was verified using a TapeStation D1000 (Agilent Technologies; Santa Clara, CA). A maximum of 1 μg of each sample was sonically sheared with the addition of 1× TE buffer, with equal amounts of each fragment being used for each sample. Sequencing libraries were prepared using the NEBNext Ultra II DNA library preparation kit for Illumina, per the manufacturer's protocol (New England BioLabs; Ipswich, MA). The products were then quantified using the Qubit dsDNA assay (Life Technologies). Amplicon size was assessed using the TapeStation D1000 preparation (Agilent Technologies) and samples were then normalized to 4 nM. Barcoded libraries were then pooled before being denatured and diluted to a final concentration of 10 pM per target. cfDNA samples were prepared using the MiSeq v3 150 cycle sequencing kit with a paired-end read length of 2 × 75 bp, whereas cf-mtDNA samples were prepared using the MiSeq v2 Nano 300 cycle sequencing kit with a paired-end read length of 2 × 150 bp. All samples were run using the Illumina MiSeq instrument (Illumina Inc.; San Diego, CA) with cfDNA samples being run on a standard flow cell and cf-mtDNA samples on a nano flow cell. The resulting sequences were mapped with Burrows-Wheeler Aligner 0.6.1 to the GRCh37, with the Revised Cambridge Reference Sequence (rCRS)
2.2. Plasma cell-free DNA (cfDNA) isolation Whole blood (3–5 ml) was drawn from a peripheral vein using standard phlebotomy techniques into a cfDNA blood collection tube (Streck Inc.; La Vista, NE). Within 1-h of collection, the blood was centrifuged at 4000 rpm and 4 °C for 15 min, the plasma was then removed and transferred into a sterile 2 ml tube(s). The buffy coat was 2
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
for mtDNA. 2.5. mtDNA haplogroup Each haplogroup was assigned using a custom pipeline integrating the HaploGrep2 workflow [18]. The subject's haplotype, as generated by HaploGrep2 according to PhyloTree Build 16 [19], was then linked to the MITOMASTER database to compare frequency information for local and global private mutations in Genbank [20]. Sequence alignment data were exported as individual Binary Alignment/Map (BAM) formatted files and uploaded into the mtDNA-Server to determine data quality and to examine heteroplasmic variant calling [21]. 2.6. Statistical analyses Statistical analysis was performed using GraphPad Prism for Windows, Version 7.02 (GraphPad Software Inc.; La Jolla, CA). Differences between treatment groups were determined by MannWhitney U tests assuming and non-Gaussian distribution where p < 0.05 was significant. Data are expressed as mean ± SEM. 3. Results 3.1. Total cell-free DNA (cfDNA) The baseline cfDNA concentration comparing control and patient plasma samples is reported in Fig. 1. The results indicate that patients with mitochondrial disease have an elevated level of circulating nucleic acids compared to healthy controls.
Fig. 2. A schematic representation of the two overlapping amplicons mapped to the mitochondrial genome following PCR.
mtDNA in the cfDNA fraction (n = 3 per group). Since both nDNA and mtDNA are known to be present in whole cfDNA the MiSeq v3 150 cycle sequencing kit was used to provide appropriate sequencing read depth. Each sequenced cfDNA sample had coverage spanning the entire mitochondrial genome, suggesting that full length mtDNA was present in cfDNA isolates. However, the percentage of total mapped reads to nonmitochondrial DNA, presumably nDNA, (> 96%) was vastly greater than the mtDNA due to utilization of whole cfDNA (Fig. 4A). Next, PCR amplified cf-mtDNA samples from the same cohort of patients and controls were sequenced to provide improved sequencing efficiency and read depth to the mitochondrial reference genome (n = 7 per group). Long range amplification of two mitochondrial fragments in cfDNA increased the proportion of total mapped reads for mtDNA to a similar percentage (ie. ~96%) as identified for non-mitochondrial DNA using raw cfDNA samples (Fig. 4B). mtDNA coverage for both cfDNA and cfmtDNA samples followed a Poisson distribution (data not shown).
3.2. Amplification of cf-mtDNA Isolated cfDNA samples were amplified using PCR in two overlapping fragments spanning the mitochondrial genome, as represented in Fig. 2. Following amplification, MTL1 (9239 bp in size) and MTL2 (11,216 bp in size) amplicons were resolved on representative 0.8% agarose gels to confirm successful mtDNA amplification. Amplified samples from a representative set of control and patient samples each migrated to their appropriate band on the protein ladder (slightly below 10 Kb for MTL1 and slightly above 10 Kb for MTL2) (Fig. 3). The term cf-mtDNA refers to cfDNA isolates following MTL1/2 PCR amplification specific to the mitochondrial genome. 3.3. cfDNA and cf-mtDNA sequencing First, unamplified cfDNA samples from healthy controls and mitochondrial disease patients were sequenced to ensure presence of full
3.4. mtDNA haplogroup analysis Patient haplogroup data was collected from previous clinical diagnostic reports performed at the Genetic Laboratory Services, Alberta Health Services, Molecular Diagnostics Laboratory. As part of the diagnostic pathway for a suspected mitochondrial disease, tissues from skeletal muscle biopsy, buccal mucosa swab, or blood leukocytes were collected and DNA was extracted. After mtDNA genome sequencing the variants and were annotated by assigning a mitochondrial haplogroup (Table 2). This clinical data was used to compare the accuracy and quality of variant calling to the respective cf-mtDNA sample. Each of the cf-mtDNA (n = 7) samples corroborated clinical haplogroup findings. The haplogroup for cf-mtDNA (n = 7) samples from healthy controls was also obtained, although the diagnostic laboratory had not previously sequenced these samples. Mapping and quantification of mtDNA variants was used to identify how closely clinical tissues and cfmtDNA yielded the same mtDNA haplogroup (Table 2). Interestingly, there was variability in the number of detected variants from each
Fig. 1. A comparison of total cell-free DNA (cfDNA) between control and patient samples. cfDNA was isolated and quantified from blood plasma of 7 healthy control and 7 patients with mitochondrial disease. Data are presented as mean ± SEM with * indicating a significant difference between control and patient samples at p < 0.05. 3
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
Fig. 3. Demonstration of cell-free mitochondrial DNA (cf-mtDNA) amplification in two overlapping fragments. Each panel depicts a series of alternating healthy control (C1-C3) and mitochondrial disease patient (P1-P3) samples following PCR amplification using the custom PCR primer pairs. MTL1 product of ~9239 bp (A) and MTL2 product of ~11,216 bp.
Fig. 4. Percent of sequencing reads mapping back to the non-mitochondrial (presumably nDNA) and mitochondrial (mtDNA) genomes. Panel A describes the percent of non-mitochondrial DNA and mtDNA reads after sequencing cfDNA samples of both healthy control and mitochondrial disease patients (n = 3 per group). Panel B describes the percent of non-mitochondrial DNA and mtDNA reads after sequencing cf-mtDNA samples of both healthy control and mitochondrial disease patients (n = 7 per group). All sequencing was performed using the Illumina MiSeq platform at the University of Calgary Core DNA Services facilities. Data are presented as mean ± SEM with * indicating a significant difference between non-mitochondrial DNA and mtDNA respective to control or patient samples at p < 0.001.
tissue. These can likely be the result of the different environmental conditions for cf-mtDNA (in circulation) and the lack of pressure to conserve as non-functional DNA. 4. Discussion Diagnosing mitochondrial disease is particularly challenging as mutations may arise in two distinct genomes, and additionally mitochondrial genome mutations may be acquired in a tissue-specific manner and may exhibit varying degrees of heteroplasmy [22]. While peripheral blood leukocytes, urine sediment, skin fibroblasts and buccal swab samples can be incorporated fairly easily into the diagnostic workflow, skeletal muscle biopsy remains the gold standard for obtaining mtDNA [23]. Our internal data also shows that muscle mtDNA has a higher sensitivity than peripheral blood with several cases where the diagnosis was missed on blood DNA analysis alone. However, obtaining a muscle biopsy is an invasive procedure and in children, invariably requires sedation with a general anesthetic. Furthermore, muscle as a single tissue source for mtDNA may not represent disease manifestations in other organs such as the liver and heart. With the development of NGS technologies, nuclear DNA diagnoses can be typically obtained with peripheral blood DNA reducing the need for muscle biopsy but the issue of obtaining sufficient mtDNA without muscle tissue remains unresolved. We hypothesized that if mtDNA could be obtained from plasma with a more sensitive technique, such as NGS, then even fewer cases would require invasive tissue biopsy. The methods we employed suggest that this may be possible. Firstly, there are limitations to the presented work that should be addressed. The primary limitation in this study is the small number of
Table 2 Mitochondrial haplogroup assessed using mtDNA sequence variants.
P1 P2 P3 P4 P5 P6 P7
Patient haplogroup (muscle)
Patient haplogroup (cf-mtDNA)
Variant total (muscle)
Variant total (cf-mtDNA)
H16b K1a4a1a K1a4a1a H1a3c T2b5 U5a1b3b U5a1b3
H16b K1a4a1a K1a4a1 H1a3c T2b5 U5a1b3 U5a1b3
33 36 38 19 37 25 23
34 37 38 17 34 27 27
mtDNA haplogroup for healthy control and mitochondrial disease patients performed using a custom pipeline incorporating the existing HaploGrep2 workflow. Each clinically derived patient haplogroup was compared to cfmtDNA samples. Mapping of variants to the mitochondrial genome was used to identify how closely each group of variants mapped to the expected haplogroup variant list. No differences were detected between controls and patients. Clinical haplogroup were collected from skeletal muscle biopsies (n = 5), buccal mucosa swab (n = 1), or blood leukocytes (n = 1). a P3 didn't consent to a muscle biopsy and therefore the haplotype and variant total were derived from leukocyte DNA. b P6 also didn't undergo a muscle biopsy and the haplotype and variant total were derived using buccal mucosal DNA. 4
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
in > 96% of sequencing reads subsequently mapping to the rCRS for the mitochondrial genome. This finding confirms that cfDNA cannot be used to diagnose a mitochondrial-genome disorder without prior amplification of the mitochondrial genome. The next question was whether cf-mtDNA variants could properly identify the patient haplogroup when compared to the gold standard using skeletal muscle biopsy. Haplogroup analysis was performed on sequenced cf-mtDNA samples, comparing data to previously collected clinical haplogroup results from skeletal muscle biopsies (n = 5), buccal mucosa swab (n = 1), or blood leukocytes (n = 1). We expected that the total number of variants detected in the cf-mtDNA may not match the variants detected in clinical tissues, caused by missing variants in cf-mtDNA or extra variants due to differences in the compared tissues. This hypothesis proved correct since the variant number between ct-mtDNA and clinical tissues was different (Table 2). A difference in variants found by cf-mtDNA can likely of low allelic fraction and therefore are attributed to somatically acquired. However, each mtDNA haplogroup from cf-mtDNA correctly matched the mtDNA haplogroup from the clinical gold standard using skeletal muscle tissue, or buccal mucosa swab, or blood leukocytes. These findings demonstrate that cf-mtDNA from plasma can accurately determine mtDNA haplogroup in patients with mitochondrial disease. Beyond utility to the field of evolutionary genetics, mtDNA haplogrouping provides valuable genetic information for potential therapeutic and diagnostic opportunities in patients with mitochondrial disease. Interestingly, recent research in liver cell therapy utilized cfDNA to measure response to therapy, comparing host and donor nDNA levels up to 6 months following cell transplantation [16]. In a similar manner, sequencing cf-mtDNA could be clinically applied to the identification of donor versus host mtDNA in response to cell therapy. Prior to cell therapy, haplogroup analysis of the donor should confirm a sufficiently different haplogroup to permit differentiation of donor from patient. In summary, this technique is advantageous as it represents a noninvasive method to provide mtDNA from multiple tissue sources for genetic analyses which should theoretically increase the sensitivity for the detection of mitochondrial disease. Further research efforts will develop cf-mtDNA as a prospective method for the evaluation of mtDNA haplogroup variants and response to future transplantation or mitochondrial therapies. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ymgme.2018.10.002.
control and patient samples that were analyzed (n = 7 per group). Although the prevalence of rare diseases inhibits researchers from achieving large sample sizes as documented in more common diseases, our small sample size makes it difficult to conclude whether the effect would be reproducible in a larger patient cohort. Collecting and analyzing samples from a multitude of haplotypes would also help to ensure that each unique haplotype can be identified using our proposed method. Secondly, there remains risk of contamination during both the blood collection and plasma processing. To ensure that the current method is avoiding leukocyte contamination, samples could be compared from a separate cohort of individuals using a sample from each of the blood draw regions and compare these findings to exclude leukocyte contamination. Initially identified in 1948, circulating nucleic acids have been suggested as a clinical diagnostic tool since the application to cancer in 1996 [24]. Since this time the application of cfDNA in diagnostics has broadened to include the fields of organ transplantation and prenatal screening [15,16]. These applications have focused on the quantification of cfDNA, the identification of traditional genetic mutations and the detection of epigenetic patterning [25–27]. Although cfDNA was previously known to be increased in cancer patients [28], we decided to investigate cfDNA levels in patients with mitochondrial disease. Our mitochondrial disease patients had significantly elevated cfDNA concentrations compared to healthy controls suggesting higher mtDNA turnover. Building upon the differences in cfDNA between patients and controls, we investigated the presence of mtDNA within the cfDNA fraction. Previous research indicated that both nuclear- and mitochondrialderived cfDNA fragments are identifiable using commercially available cfDNA isolation protocols [26,29]. Although there are unique structural and functional differences between nDNA and mtDNA, existing research has focused on short-length mtDNA amplicons, similar to the investigation of nDNA from cfDNA. Since mtDNA is a circular doublestranded molecule and it lacks histones, we hypothesized that intact mtDNA persists in the cfDNA fraction as the circular nature of intact mtDNA delays the rapid degradation of cfDNA by circulating nucleases [30]. The lack of histones also contributes by forgoing the targeted excision of DNA in the classical ~150–200 bp fragment pattern seen in nDNA, thus enabling mtDNA to remain whole [13]. As we wanted to ensure that we isolated mitochondria from the cell-free portion of the plasma, precautions were taken to avoid contamination with mitochondrial DNA from intact leukocytes. First, a specific cfDNA blood collection tube (Streck Inc.) was utilized to collect blood. These tubes contain a proprietary cell preservative which stabilizes nucleated cells and thereby prevents the release of nDNA and mtDNA for up to 14 days after collection [31]. Second, a ~10 mm gap between the supernatant plasma and buffy coat was left undisturbed to prevent mechanical contamination of plasma with disturbed leukocytes. Using these precautions, our results show that full-length cf-mtDNA can be amplified from plasma of both mitochondrial disease patients and healthy controls. These data, provide support to previous work confirming that mtDNA exists in both particle-associated and non-particle associated forms in cfDNA [32]. Population geneticists use mitochondrial haplogroups to map the migration of humans throughout the world. This collection of maternally inherited mtDNA variants are usually homoplasmic changes associated with a branch of the mitochondrial phylogenetic tree [33]. Originally ranging from letters A-Z, mtDNA haplogroups are now further defined by their sub-haplogroup due to the recent abundance of data provided by whole mtDNA genome sequencing studies [19]. We decided to use this evolutionarily conserved feature and sequence both cfDNA and cf-mtDNA samples, using the mtDNA haplogroup to then differentiate between individuals. Initially comparing whole cfDNA to cf-mtDNA, we demonstrated that a significant percentage of mapped reads corresponded to non-mitochondrial DNA (> 96%) in the cfDNA samples. Amplification of mtDNA circumvented this problem, resulting
Author contributions CN, SCG and AK designed and developed the research. CN, SH, LP, SCG and AK conducted experiments, collected and analyzed data. CN, JS and AK wrote the manuscript and all authors read and approved the final manuscript.
Competing interests The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. CN, SH, SCG and AK have applied for a patent related to the described methodology.
Funding This study was supported by the Alberta Children's Hospital Research Foundation, Department of Pediatrics and MitoCanada (AK). This research was supported by PhD funding to CN from MitoCanada and an Alberta Innovates – Health Solutions MD/PhD Studentship. 5
Molecular Genetics and Metabolism xxx (xxxx) xxx–xxx
C. Newell et al.
Ethics approval statement [17]
All experimental procedures for this study were carried out in accordance with the recommendations of the University of Calgary's Conjoint Health Research Ethics Board (REB13-0753). Patient consent statement All subjects gave written informed consent in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).
[18]
Acknowledgements [19]
The authors would like to thank Dr. Richard Pon, Shelly Wegener and Dr. Paul Gordon at the University of Calgary Core DNA Services for sample sequencing, and Elizabeth Newell for editing an earlier draft of this manuscript.
[20]
References
[21]
[1] A.M. D'Erchia, A. Atlante, G. Gadaleta, G. Pavesi, M. Chiara, C. De Virgilio, C. Manzari, F. Mastropasqua, G.M. Prazzoli, E. Picardi, C. Gissi, D. Horner, A. Reyes, E. Sbisà, A. Tullo, G. Pesole, Tissue-specific mtDNA abundance from exome data and its correlation with mitochondrial transcription, mass and respiratory activity, Mitochondrion 20 (2015) 13–21, https://doi.org/10.1016/j.mito.2014.10.005. [2] N.R. Phillips, M.L. Sprouse, R.K. Roby, Simultaneous quantification of mitochondrial DNA copy number and deletion ratio: a multiplex real-time PCR assay, Sci. Rep. 4 (2014) 3887, , https://doi.org/10.1038/srep03887. [3] J.-W. Taanman, The mitochondrial genome: structure, transcription, translation and replication, Biochim. Biophys. Acta Bioenerg. 1410 (1999) 103–123, https:// doi.org/10.1016/S0005-2728(98)00161-3. [4] C. Lee, J. Zeng, B.G. Drew, T. Sallam, A. Martin-Montalvo, J. Wan, S.-J. Kim, H. Mehta, A.L. Hevener, R. de Cabo, P. Cohen, The mitochondrial-derived peptide MOTS-c promotes metabolic homeostasis and reduces obesity and insulin resistance, Cell Metab. 21 (2015) 443–454, https://doi.org/10.1016/j.cmet.2015.02. 009. [5] C. Lee, K. Yen, P. Cohen, Humanin: a harbinger of mitochondrial-derived peptides? Trends Endocrinol. Metab. 24 (2013) 222–228, https://doi.org/10.1016/j.tem. 2013.01.005. [6] J.D. Stumpf, R.P. Saneto, W.C. Copeland, Clinical and molecular features of POLGrelated mitochondrial disease, Cold Spring Harb. Perspect. Biol. 5 (2013) a011395, https://doi.org/10.1101/cshperspect.a011395. [7] R.W. Taylor, D.M. Turnbull, Mitochondrial DNA mutations in human disease, Nat. Rev. Genet. 6 (2005) 389–402, https://doi.org/10.1038/nrg1606. [8] F.M. Yakes, B. Van Houten, Mitochondrial DNA damage is more extensive and persists longer than nuclear DNA damage in human cells following oxidative stress, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 514–519 http://www.ncbi.nlm.nih.gov/ pubmed/9012815 (accessed April 2, 2017). [9] S.D. Cline, Mitochondrial DNA damage and its consequences for mitochondrial gene expression, Biochim. Biophys. Acta 1819 (2012) 979–991, https://doi.org/10. 1016/j.bbagrm.2012.06.002. [10] D.C. Wallace, D. Chalkia, Mitochondrial DNA genetics and the heteroplasmy conundrum in evolution and disease, Cold Spring Harb. Perspect. Biol. 5 (2013) a021220, https://doi.org/10.1101/cshperspect.a021220. [11] S. Dimauro, G. Davidzon, Mitochondrial DNA and disease, Ann. Med. 37 (2005) 222–232, https://doi.org/10.1080/07853890510007368. [12] C.-C. Liew, J. Ma, H.-C. Tang, R. Zheng, A.A. Dempsey, The peripheral blood transcriptome dynamically reflects system wide biology: a potential diagnostic tool, J. Lab. Clin. Med. 147 (2006) 126–132, https://doi.org/10.1016/j.lab.2005.10. 005. [13] M.W. Snyder, M. Kircher, A.J. Hill, R.M. Daza, J. Shendure, Cell-free DNA comprises an in vivo nucleosome footprint that informs its tissues-of-origin, Cell 164 (2016) 57–68, https://doi.org/10.1016/j.cell.2015.11.050. [14] J.C.H. Tsang, Y.M. Dennis Lo, Circulating nucleic acids in plasma/serum, Pathology 39 (2007) 197–207, https://doi.org/10.1080/00313020701230831. [15] A.B. Sparks, C.A. Struble, E.T. Wang, K. Song, A. Oliphant, Noninvasive prenatal detection and selective analysis of cell-free DNA obtained from maternal blood: evaluation for trisomy 21 and trisomy 18, Am. J. Obstet. Gynecol. 206 (2012) 319.e1–319.e9, https://doi.org/10.1016/j.ajog.2012.01.030. [16] P.M.K. Gordon, A. Khan, U. Sajid, N. Chang, V. Suresh, L. Dimnik, R.E. Lamont, J.S. Parboosingh, S.R. Martin, R.T. Pon, J. Weatherhead, S. Wegener, D. Isaac, S.C. Greenway, An algorithm measuring donor cell-free DNA in plasma of cellular and solid organ transplant recipients that does not require donor or recipient
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
6
genotyping, Front. Cardiovasc. Med. 3 (2016) 33, , https://doi.org/10.3389/fcvm. 2016.00033. C. Bettegowda, M. Sausen, R.J. Leary, I. Kinde, Y. Wang, N. Agrawal, B.R. Bartlett, H. Wang, B. Luber, R.M. Alani, E.S. Antonarakis, N.S. Azad, A. Bardelli, H. Brem, J.L. Cameron, C.C. Lee, L.A. Fecher, G.L. Gallia, P. Gibbs, D. Le, R.L. Giuntoli, M. Goggins, M.D. Hogarty, M. Holdhoff, S.-M. Hong, Y. Jiao, H.H. Juhl, J.J. Kim, G. Siravegna, D.A. Laheru, C. Lauricella, M. Lim, E.J. Lipson, S.K.N. Marie, G.J. Netto, K.S. Oliner, A. Olivi, L. Olsson, G.J. Riggins, A. Sartore-Bianchi, K. Schmidt, l.-M. Shih, S.M. Oba-Shinjo, S. Siena, D. Theodorescu, J. Tie, T.T. Harkins, S. Veronese, T.-L. Wang, J.D. Weingart, C.L. Wolfgang, L.D. Wood, D. Xing, R.H. Hruban, J. Wu, P.J. Allen, C.M. Schmidt, M.A. Choti, V.E. Velculescu, K.W. Kinzler, B. Vogelstein, N. Papadopoulos, L.A. Diaz, Detection of circulating tumor DNA in early- and late-stage human malignancies, Sci. Transl. Med. 6 (2014), https://doi.org/10.1126/scitranslmed.3007094 224ra24-224ra24. H. Weissensteiner, D. Pacher, A. Kloss-Brandstätter, L. Forer, G. Specht, H.J. Bandelt, F. Kronenberg, A. Salas, S. Schönherr, HaploGrep 2: mitochondrial haplogroup classification in the era of high-throughput sequencing, Nucleic Acids Res. 44 (2016) W58–W63, https://doi.org/10.1093/nar/gkw233. M. van Oven, M. Kayser, Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation, Hum. Mutat. 30 (2009) E386–E394, https://doi.org/ 10.1002/humu.20921. M.C. Brandon, E. Ruiz-Pesini, D. Mishmar, V. Procaccio, M.T. Lott, K.C. Nguyen, S. Spolim, U. Patil, P. Baldi, D.C. Wallace, MITOMASTER: a bioinformatics tool for the analysis of mitochondrial DNA sequences, Hum. Mutat. 30 (2009) 1–6, https:// doi.org/10.1002/humu.20801. H. Weissensteiner, L. Forer, C. Fuchsberger, B. Schöpf, A. Kloss-Brandstätter, G. Specht, F. Kronenberg, S. Schönherr, mtDNA-Server: next-generation sequencing data analysis of human mitochondrial DNA in the cloud, Nucleic Acids Res. 44 (2016) W64–W69, https://doi.org/10.1093/nar/gkw247. C.L. Alston, M.C. Rocha, N.Z. Lax, D.M. Turnbull, R.W. Taylor, The genetics and pathology of mitochondrial disease, J. Pathol. 241 (2017) 236–250, https://doi. org/10.1002/path.4809. R.H. Haas, S. Parikh, M.J. Falk, R.P. Saneto, N.I. Wolf, N. Darin, L.-J. Wong, B.H. Cohen, R.K. Naviaux, The in-depth evaluation of suspected mitochondrial disease, Mol. Genet. Metab. 94 (2008) 16–37, https://doi.org/10.1016/j.ymgme. 2007.11.018. A.K.C. Chan, R.W.K. Chiu, Y.M.D. Lo, Clinical Sciences Reviews Committee of the Association of Clinical Biochemists, Cell-free nucleic acids in plasma, serum and urine: a new tool in molecular diagnosis, Ann. Clin. Biochem. 40 (2003) 122–130, https://doi.org/10.1258/000456303763046030. E. Bohers, P.J. Viailly, S. Dubois, P. Bertrand, C. Maingonnat, S. Mareschal, P. Ruminy, J.-M. Picquenot, C. Bastard, F. Desmots, T. Fest, K. Leroy, H. Tilly, F. Jardin, Somatic mutations of cell-free circulating DNA detected by next-generation sequencing reflect the genetic changes in both germinal center B-cell-like and activated B-cell-like diffuse large B-cell lymphomas at the time of diagnosis, Haematologica 100 (2015) e280–e284, https://doi.org/10.3324/haematol.2015. 123612. C. Kohler, R. Radpour, Z. Barekati, R. Asadollahi, J. Bitzer, E. Wight, N. Bürki, C. Diesch, W. Holzgreve, X. Zhong, Levels of plasma circulating cell free nuclear and mitochondrial DNA as potential biomarkers for breast tumors, Mol. Cancer 8 (2009) 105, https://doi.org/10.1186/1476-4598-8-105. R. Lehmann-Werman, D. Neiman, H. Zemmour, J. Moss, J. Magenheim, A. VakninDembinsky, S. Rubertsson, B. Nellgård, K. Blennow, H. Zetterberg, K. Spalding, M.J. Haller, C.H. Wasserfall, D.A. Schatz, C.J. Greenbaum, C. Dorrell, M. Grompe, A. Zick, A. Hubert, M. Maoz, V. Fendrich, D.K. Bartsch, T. Golan, S.A. Ben Sasson, G. Zamir, A. Razin, H. Cedar, A.M.J. Shapiro, B. Glaser, R. Shemer, Y. Dor, Identification of tissue-specific cell death using methylation patterns of circulating DNA, Proc. Natl. Acad. Sci. 113 (2016) E1826–E1834, https://doi.org/10.1073/ pnas.1519286113. X.Y. Zhong, A. Ladewig, S. Schmid, E. Wight, S. Hahn, W. Holzgreve, Elevated level of cell-free plasma DNA is associated with breast cancer, Arch. Gynecol. Obstet. 276 (2007) 327–331, https://doi.org/10.1007/s00404-007-0345-1. L. Li, H.-W. Hann, S. Wan, R.S. Hann, C. Wang, Y. Lai, X. Ye, A. Evans, R.E. Myers, Z. Ye, B. Li, J. Xing, H. Yang, Cell-free circulating mitochondrial DNA content and risk of hepatocellular carcinoma in patients with chronic HBV infection, Sci. Rep. 6 (2016) 23992, , https://doi.org/10.1038/srep23992. Y.M.D. Lo, J. Zhang, T.N. Leung, T.K. Lau, A.M.Z. Chang, N.M. Hjelm, Rapid clearance of fetal DNA from maternal plasma, Am. J. Hum. Genet. 64 (1999) 218–224, https://doi.org/10.1086/302205. S.E. Norton, J.M. Lechner, T. Williams, M.R. Fernando, A stabilizing reagent prevents cell-free DNA contamination by cellular DNA in plasma during blood sample storage and shipping as determined by digital PCR, Clin. Biochem. 46 (2013) 1561–1565, https://doi.org/10.1016/j.clinbiochem.2013.06.002. R.W.K. Chiu, L.Y.S. Chan, N.Y.L. Lam, N.B.Y. Tsui, E.K.O. Ng, T.H. Rainer, Y.M.D. Lo, Quantitative analysis of circulating mitochondrial DNA in plasma, Clin. Chem. 49 (2003), http://clinchem.aaccjnls.org/content/49/5/719.long (accessed April 11, 2017). J.B. Stewart, P.F. Chinnery, The dynamics of mitochondrial DNA heteroplasmy: implications for human health and disease, Nat. Publ. Gr. 16 (2015), https://doi. org/10.1038/nrg3966.