Research article

Performance of mitochondrial DNA mutations detecting early stage cancer

John P Jakupciak¹ , Samantha Maragh¹ , Maura E Markowitz² , Alissa K Greenberg⁵ , Mohammad O Hoque³ , Anirban Maitra³ , Peter E Barker¹ , Paul D Wagner⁴ , William N Rom⁵ , Sudhir Srivastava⁴ , David Sidransky³ and Catherine D O'Connell⁶

¹  Biochemical Science Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899, USA

²  Geo-Centers, Inc, Newton, Massachusetts, USA

³  Johns Hopkins University School of Medicine, 720 Rutland Ave, Baltimore, Maryland 21205, USA

⁴  Cancer Biomarkers Research Group, National Cancer Institute, Rockville, Maryland, USA

⁵  Division of Pulmonary and Critical Care Medicine, NYU, School of Medicine, New York, USA

⁶  Tetracore, Inc, Rockville, Maryland, USA

author email corresponding author email

BMC Cancer 2008, 8:285doi:10.1186/1471-2407-8-285

Published:	3 October 2008

Abstract

Background

Mutations in the mitochondrial genome (mtgenome) have been associated with cancer and many other disorders. These mutations can be point mutations or deletions, or admixtures (heteroplasmy). The detection of mtDNA mutations in body fluids using resequencing microarrays, which are more sensitive than other sequencing methods, could provide a strategy to measure mutation loads in remote anatomical sites.

Methods

We determined the mtDNA mutation load in the entire mitochondrial genome of 26 individuals with different early stage cancers (lung, bladder, kidney) and 12 heavy smokers without cancer. MtDNA was sequenced from three matched specimens (blood, tumor and body fluid) from each cancer patient and two matched specimens (blood and sputum) from smokers without cancer. The inherited wildtype sequence in the blood was compared to the sequences present in the tumor and body fluid, detected using the Affymetrix Genechip^® Human Mitochondrial Resequencing Array 1.0 and supplemented by capillary sequencing for noncoding region.

Results

Using this high-throughput method, 75% of the tumors were found to contain mtDNA mutations, higher than in our previous studies, and 36% of the body fluids from these cancer patients contained mtDNA mutations. Most of the mutations detected were heteroplasmic. A statistically significantly higher heteroplasmy rate occurred in tumor specimens when compared to both body fluid of cancer patients and sputum of controls, and in patient blood compared to blood of controls. Only 2 of the 12 sputum specimens from heavy smokers without cancer (17%) contained mtDNA mutations. Although patient mutations were spread throughout the mtDNA genome in the lung, bladder and kidney series, a statistically significant elevation of tRNA and ND complex mutations was detected in tumors.

Conclusion

Our findings indicate comprehensive mtDNA resequencing can be a high-throughput tool for detecting mutations in clinical samples with potential applications for cancer detection, but it is unclear the biological relevance of these detected mitochondrial mutations. Whether the detection of tumor-specific mtDNA mutations in body fluidsy this method will be useful for diagnosis and monitoring applications requires further investigation.