|Year : 2015 | Volume
| Issue : 2 | Page : 133-139
Identification of the Mislabeled Breast Cancer Samples by Mitochondrial DNA Haplotyping
Xiaogang Chen1, Di Lu2, Ji Zhang1, Feng Song1, Haibo Luo1, Zheng Ren1, Wei Wei1, Yiping Hou1
1 Institute of Forensic Medicine, West School of Basic Science and Forensic Medicine, Sichuan University, Chengdu 610041, Sichuan, China
2 Collaborative Innovation Center of Judicial Civilization, Institute of Evidence Law and Forensic Science, University of Political Science and Law, Beijing 100088, China
|Date of Web Publication||27-Nov-2015|
Ren Min Nan Road 3-17, Chengdu 610041
Source of Support: None, Conflict of Interest: None
The task to identify whether an archival malignant tumor specimen had been mislabeled or interchanged is a challenging one for forensic genetics. The nuclear DNA (nDNA) markers were affected by the aberration of tumor cells, so they were not suitable for personal identification when the tumor tissues were tested. In this study, we focused on a new solution - mitochondrial single nucleotide polymorphism (mtSNP) haplotyping by a multiplex SNaPshot assay. To validate our strategy of haplotyping with 25 mtSNPs, we analyzed 15 pairs of cancerous/healthy tissues taken from patients with ductal breast carcinoma. The haplotypes of all the fifteen breast cancer tissues were matched with their paired breast tissues. The heteroplasmy at 2 sites, 14783A/G and 16519C/T was observed in one breast tissue, which indicated a mixture of related mitochondrial haplotypes. However, only one haplotype was retained in the paired breast cancer tissue, which could be considered the result of proliferation of tumor subclone. The allele drop-out and allele drop-in were observed when 39 STRs and 20 tri-allelic SNPs of nDNA were applied. Compared to nDNA markers applied, 25 mtSNPs were more stable without interference from aberrance of breast cancer. Also, two cases were presented where the investigation of haplotype with 25 mtSNPs was used to prove the origin of biopsy specimen with breast cancer. The mislabeling of biopsy specimen with breast cancer could be certified in one case but could not be supported in the other case. We highlight the importance of stability of mtSNP haplotype in breast cancer. It was implied that our multiplex SNaPshot assay with 25 mtSNPs was a useful strategy to identify mislabeled breast cancer specimen.
Keywords: Breast cancer, forensic genetics, individual discrimination, mitochondrial single nucleotide polymorphism
|How to cite this article:|
Chen X, Lu D, Zhang J, Song F, Luo H, Ren Z, Wei W, Hou Y. Identification of the Mislabeled Breast Cancer Samples by Mitochondrial DNA Haplotyping. J Forensic Sci Med 2015;1:133-9
|How to cite this URL:|
Chen X, Lu D, Zhang J, Song F, Luo H, Ren Z, Wei W, Hou Y. Identification of the Mislabeled Breast Cancer Samples by Mitochondrial DNA Haplotyping. J Forensic Sci Med [serial online] 2015 [cited 2019 Mar 25];1:133-9. Available from: http://www.jfsmonline.com/text.asp?2015/1/2/133/170603
| Introduction|| |
Archival malignant tumor specimens may be submitted for forensic DNA profiling analysis to investigate whether pathology samples were either mislabeled or interchanged in medical testing facility. The biggest risk is the misinterpretation of the profile that is caused by genetic instability of cancer.,,
The nuclear DNA (nDNA) markers that are applied in the most commercial forensic DNA profiling kits have been manifested as loss of heterozygosity (LOH) or microsatellite instability in a number of malignancies such as breast, bladder, lung, cervix, ovary, colon, and prostate., In addition, the alteration of nDNA in tumor cells is complex, which makes the interpretation of genetic profile more difficult and less reliable, especially for the nonidentical profiles. As a result, the genetic profile analysis using core short tandem repeats (STR) markers would be questionable when the cancer tissues are the only available samples.
Compared to the nuclear genome, the genetic instability of mitochondrial DNA (mtDNA) in tumor cells is relatively simpler. The mutation and the insertion/deletion are the main forms of mtDNA alterations observed in tumor cells. This leads to the hypothesis that mtDNA analysis can be applied in personal identification of tumor samples. Mutations in D-loop region of mtDNA have been identified in various types of cancer., As regards the breast cancer, the mutations concentrated in the non-coding hypervariable regions of D-loop have been reported by three researchers.,, Therefore, forensic mtDNA sequence analysis of the hypervariable region within D-loop is not applicable for personal identification when breast cancer samples are tested. Köhnemann et al. reported that analysis of mitochondrial single nucleotide polymorphism (mtSNP) showed a higher potential for personal identification when STR-analysis and sequencing of the mtDNA hypervariable region led to an ambiguous conclusion., Therefore, an expanded investigation of mtDNA sequence information, mtSNPs out of hypervariable regions in D-loop, was proposed as an applicable approach for the personal identification of breast cancer tissues. Moreover, the interference from the somatic mutation of breast cancer cells could be minimized and the satisfactory results are obtained by a number of specific mtSNPs.
In the present study, a multiplex SNaPshot system with 25 mtSNPs was applied to verify our hypothesis. The system had been constructed successfully and verified validly in healthy individuals from Chinese Han population in our Lab. We analyzed 15 pairs of cancerous/healthy tissues taken from patients with ductal breast carcinoma. The haplotypes of 15 breast cancer tissues were matched with that of paired breast tissues. Then, two cases were presented where the disputed biopsy specimens with breast cancer were tested using the same multiplex system to verify whether the specimens were mislabeled with identity code of the patient who had accepted mastectomy.
| Materials and Methods|| |
A female patient, whose biopsy diagnosis of invasive ductal carcinoma was positive, was surgically treated by a modified mastectomy technique. However, the postoperative pathological diagnosis failed to reveal the presence of any tumor tissues. Review of the disputed biopsy tissue confirmed the presence of ductal carcinoma. The discrepancy of pathological diagnosis aroused suspicion that the biopsy material was unintentionally interchanged, which led to the incorrect treatment.
A female patient accepted mastectomy with the diagnosis of invasive ductal carcinoma several months after a biopsy diagnosis of intra-cystic papillary tumor of the breast. The biopsy specimen was re-examined, and finally diagnosis was revised to the ductal carcinoma. The origination of biopsy specimen was needed to ascertain the misdiagnosis of biopsy.
Samples and DNA extraction
The disputed preoperative biopsy specimens in case 1 and case 2 were formalin-fixed and paraffin-embedded breast ductal carcinoma tissues. The tissue block and the blood sample from each patient were sent to our lab to identify whether both samples were belonged to the same person.
To test the reliability of the multiplex SNaPshot system with 25 mtSNPs in the analysis of breast cancer samples, 15 pairs of matched breast cancer and breast tissues were obtained from the patient with diagnosis of breast ductal carcinoma. The tissues were collected after the radical mastectomy for breast cancer with informed consent. The breast tissues were taken from cutting edges apart from the margin of breast cancer at least 3 cm. Breast cancer tissues and breast tissues without cancer were determined by a senior pathologist under the microscopy. All the tissues provided by the pathologist immediately after removal from the subject were preserved in 70% ethanol until analysis.
A small amount of tissue was picked up from the paraffin-embedded block. After grinding, the samples were transferred into 200 µl of 5% Chelex-100 (pH 9.0) containing 40 µg proteinase K. These samples were incubated at 56°C overnight, and then boiled at 100°C for 8 min before centrifugation at 13,000 rpm for 3 min. Samples were stored at 4°C until analysis. For the ethanol preserved tissues, the DNA extraction procedure was the same as that of paraffin-embedded tissues using 100 µl of 5% Chelex-100 (pH 9.0) containing 100 µg proteinase K. DNAs from the blood samples were extracted according to the classical chelex-100 method. The DNAs were purified with the QIAquick polymerase chain reaction (PCR) Purification Kits (Qiagen) according to the manufacturer's instructions.
Mitochondrial single nucleotide polymorphism haplotyping
The multiplex mtSNP assay was performed according to our published literature. This assay included 9 hot-spot mtSNPs that were used to increase discrimination power of the multiplex assay and 16 coding mtSNPs that defined the main haplogroups in Chinese population. All the 25 mtSNPs, mtSNP 16519 in D-loop and the other mtSNPs in coding region, were out of hypervariable regions.
nDNA markers genotyping
Twenty tri-allelic SNPs located at 16 autosomal chromosomes were selected and analyzed according to our published literature. The tri-allelic SNPs tested were: rs10045, rs10811897, rs11141033, rs140676, rs1630312, rs17287498, rs2032582, rs2069945, rs2278786, rs2298556, rs2307223, rs3091244, rs356167, rs3743842, rs3812847, rs3816662, rs385780, rs4540055, rs6001030, and rs941454.
A total of 39 polymorphic autosomal STR loci, including D3S1358, D13S317, D7S820, D16S539, TPOX, TH01, vWA, D21S11, D18S51, CSF1PO, FGA, D8S1179, D5S818, Penta D, D6S1043, D12S391, D6S474, D12ATA63, D22S1045, D10S1248, D1S1677, D11S4463, D1S1627, D3S4529, D2S441, D6S1017, D4S2408, D19S433, D17S1301, D1GATA113, D18S853, D20S482, D14S1434, D9S1122, D2S1776, D10S1435, D5S2500, PentaE, and D2S1338, were selected from two commercial PCR amplification kits such as AGCU Express marker 22 PCR Amplification Kit and AGCU 21 + 1 STR PCR Amplification Kit. All the STR loci were detected according to the manufacturers' recommendations. PCR reactions were conducted using the GeneAmp® PCR System 9700 (Applied Biosystems, Foster City, CA, USA). All the products were analyzed using the automated ABI3100 Genetic Analyzer platform and GeneMapper Software (Applied Biosystem, USA).
| Results|| |
All validation samples, including 15 pairs of matched breast cancer and breast tissues, were successfully genotyped by our multiplex SNaPshot assay. The haplotypes of 15 breast cancer tissues were matched with that of paired breast tissues [Table 1]. There was one breast tissue that was characterized by a mixture of mitochondrial haplotypes rather than a single haplotype, reflected by the two heteroplasmic SNPs observed such as 14783A/G and 16519C/T. However, only one haplotype was discovered in the paired breast cancer tissue, and the others were lost [Figure 1].
|Table 1: mtSNP haplotypes of fifteen pairs bct/bt, together with the biopsy specimens and the blood samples of two cases|
Click here to view
|Figure 1: Electropherograms of 25 mitochondrial single nucleotide polymorphisms for a pair of breast cancer/breast tissues (upper) and the case 1 (lower). The x-axis represents the length of the polymerase chain reaction products in base pairs and the y-axis represents the peak height in fluorescence units. The solid arrow (↓) marks 14783G of the heteroplasmic breast tissue that was absent in the paired breast cancer tissue. The difference between the biopsy specimen with breast cancer and blood of the labeled patient was indicated with thin arrows (↓)|
Click here to view
The nDNA markers, including 39 STRs and 20 tri-allelic SNPs, were applied to obtain genotypes of 15 pairs of breast cancer and breast tissues. As affected by the unequal peak height of heterozygote alleles, the genotyping of 15 cancerous samples with 39 STRs was difficult. The allele drop-in was observed in 1 cancerous sample at D10S1435 and D1GATA113. The complete LOH was observed in another cancerous sample at 10 STR loci [Table 2]. The partial LOH (pLOH), that was attributed to a peak intensity ratio (peak ratio in the tumor tissue/peak ratio in the healthy tissue) of less than 0.5 or higher than 2.0, was observed in 8 cancerous samples. Among 39 STR loci analyzed, only D2S441, Penta D, D1S1677, D21S11, TPOX, D6S1017, and D19S433 never exhibited LOH-type alterations, while the most altered locus was D17S1301 and D11S4463 (showing five LOH) [Supplement [Table 1]. Four cancerous samples were classified as high-frequency LOH, showing LOH-type alterations at ≥33% of the heterozygote STR loci. There were 4 pairs that failed to be genotyped by 4 nuclear SNPs. The LOH was observed in 2 cancerous samples and the allele drop-in in 1 cancerous sample when the 20 tri-allelic SNPs were applied [Table 2].
|Table 2: The difference of typing results of nDNA markers in fifteen pairs bct/bt|
Click here to view
In the two cases, the disputed biopsy specimens with breast cancer and the control blood samples were successfully genotyped by all the 39 STRs and 45 SNPs, including 20 nuclear SNPs and 25 mtSNPs. In the case 1, the mtSNP haplotype of biopsy specimen with breast cancer was non-matching with the labeled patient. There were 11 mtSNPs that were different between the biopsy specimen under suspicion and the blood sample from the patient [Figure 1]. Except the variation of 5147 (G→G/A), the differences were observed at 10 SNPs, including 3010, 3970, 4883, 6392, 8701, 10398, 12705, 13708, 13928, and 14783, which presented in the homoplasmic state [Table 1]. The apparent difference of the mtSNP haplotypes was difficult to be explained as the mutations according to the result of the prior test and the review of literature.,,,, Also, the differences were obviously observed when the nuclear genetic markers were applied. A total of 13 nSNPs and 37 STRs were non-identical [Supplementary [Table 2]. Hence, our conclusion was that the patient could be excluded as origin of the biopsy specimen with breast cancer. In the case 2, the mtSNP haplotype of biopsy specimen with breast cancer was matching with the blood of the patient [Table 1]. Consulting the criteria of mtDNA SNPs comparison proposed by Köhnemann and Pfeiffer, the biopsy specimen cannot be excluded as potentially being from the patient. The genotyping result of nDNA markers further supported the conclusion drawn on that of mtSNPs. Although pLOH was observed at 14 loci among 30 heterozygote STR loci [Supplementary [Table 1], the genotypes with the nuclear genetic markers were matched between the biopsy specimen and the blood sample [Supplementary [Table 2].
| Discussion|| |
Malignant tumor specimens are a great challenge in the forensic genetic analysis. The trouble comes from the genomic aberrations in tumor cells, which may affect the genetic markers used in forensic casework (e.g., STRs and mtDNA sequence of the hypervariable region).,,,,,, When the disputed biopsy samples with breast cancer were sent to our laboratory for personal identification, we developed and applied a novel and reliable strategy in addition to performing traditional genotyping with nuclear markers. As the aberration of mtDNA in tumor cells, including somatic mutation and insertion/deletion, was simpler compared to the nuclear genome,, the haplotype composed of mtSNPs without interference of somatic mutations seems to be feasible in the identification of the origination of breast cancer samples.
To test our hypothesis, we first analyzed the nDNA markers including 39 STRs and 20 tri-allelic SNPs using 15 pairs of breast cancer/breast tissues. As we expected, allele drop-ins and allele drop-outs were observed. Also, a few samples failed to be genotyped by some nuclear SNPs despite repeated testing. Hence, the nuclear genetic markers were not applicable to breast cancer specimen when the personal discrimination was concerned.
The effectiveness and qualification of haplotyping with 25 mtSNPs in healthy Chinese Han individuals have been evaluated in our laboratory. The multiplex SNaPshot assay was adjusted a little when applied to breast cancer samples. The mtSNP 709 was removed as it had been reported as a tumor-specific homo→homo somatic mutation in breast cancer. Fifteen pairs of breast cancer/breast tissues were further genotyped to test the reliability of the use of mtSNP haplotypes in the identification of breast cancer samples. The results supported our hypothesis. The haplotypes of 15 breast cancer tissues were matched with their paired breast tissues. For the individual who have 2 heteroplasmic SNPs, 14783A/G and 16519C/T, in breast tissue, a single haplotype with 14783A and 16519T retained in the paired breast cancer tissue. We have not found any previous studies on tumorigenic properties of mutations at site 14783 and 16519 in breast cancer, so the retained haplotype in breast cancer cells may link to the tumorigenic mutations at other sites of mtDNA that promoted tumor cell proliferation and permitted tumors to adapt to new environments. Single haplotype retaining in breast cancer sample could be interpreted as a result of a replicative advantage for tumorigenic mtDNA copies and/or a growth advantage for a tumor subclone containing certain mtDNA haplotypes.,,,
As a result of careful selection of mtSNPs in our study, the homo→homo and homo→hetero somatic mutations were not observed in 15 paired samples. For the tumor-specific somatic mtDNA mutations reported, the homo→homo somatic mutations should be avoided when constructing the system, since it will interfere with the interpretation of the result of haplotyping. However, Parrella et al. reported that the original homoplasmic state of breast tissue could be observed in tumor specimens that were dissected under microscope and contained at least 70% neoplastic cells when the homo→hetero somatic mutation was concerned. Hence, mtSNPs with homo→hetero somatic mutations reported in breast cancer still could be used to construct our multiplex system.
One limitation of our research was the sample size. Clearly 15 pairs of samples were not enough to make generalizations about the 25 mtSNPs analyzed. However, from the results of those limited number of breast cancer samples, a clear pattern emerged where haplotypes of mtSNPs out of mutation regions and sites were less influenced by tumor genome instability than nDNA markers. Therefore, haplotyping with mtSNPs was an applicable approach for the personal identification of tumor tissues.
Köhnemann and Pfeiffer. suggested that exclusion could be reported if at least one of the confirmed mtDNA SNPs was different to the victim or the suspect when the SNaPshot assay was applied to get the mtSNPs haplotype of the non-neoplastic tissues. As the somatic mutations and heteroplasmic state instability of mtDNA were popular in tumors, we considered that the following criteria could be supplemented when the breast cancer samples were analyzed: (1) The mtSNP analyzed should exclude tumor-specific homo→homo somatic mutations reported in breast cancer. (2) Single mtSNP haplotype retained in breast cancer should be judged as consistent when the heteroplasmy was observed in healthy tissue. Nowadays, researches on mtDNA mutations in breast cancer are mostly small sample sized. As the tumor genetic instability is relative and not absolute, the possibility of erroneous non-identity should be taken into account without further supporting data.
In the case 1, the haplotype of biopsy specimen with breast cancer was not identical with the labeled patient at 11 mtSNPs [Figure 1]. Hence, our conclusion was that the biopsy specimen that was under suspicion was excluded as originated from the patient. A vast difference was also observed between the biopsy specimen under suspicion and the blood sample of the patient when the nuclear genetic markers were applied. In the other case, the haplotype of biopsy specimen with breast cancer was identical with the patient at all 25 mtSNPs, so the ownership could not be excluded. The profile of nDNA markers seemed to support the conclusion as well, but the LOH and the possibility of allele drop-in limited the application of nDNA markers. Since we cannot discriminate the genetic instability of cancer cells with real polymorphisms undoubtedly, the value of nDNA markers was less than mtSNPs at least in terms of our experiments.
As the coding mutations of mtDNA may occur during embryonic development, and the heteroplasmic mutations presented in a stem cell should become either homoplasmic or be lost within 70 generations, the multiple mtSNPs system that we recommended should only be used to compare the breast cancer sample with samples taken from the patient herself but not the relatives without further validation testing.
| Conclusion|| |
The task of identifying mislabeling and interchange of archival malignant tumor specimens is a challenging one for forensic genetics because of aberration of tumor cells. We attempted a new solution employing haplotyping with 25 mtSNPs from coding and noncoding regions of the mtDNA. Our result showed the haplotypes of 25 mtSNPs were less influenced by tumor genome instability than nDNA markers. It was implied that our multiplex SNaPshot assay with 25 mtSNPs was a useful strategy to identify mislabeled breast cancer samples. This will be useful to throw light on the identification of mislabeled tumor specimens.
This work was supported by the grants from the Five-twelfth National Science and Technology Support Program of China (2012BAK16B01) and from the National Natural Science Foundation of China (81330073).
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kalfoglou EA, Faikoglu R, Özcan S, Petridis G, Yükseloglu H, Atasoy S. DNA analysis as a tool for breast cancer malpractice determination: an interdisciplinary approach. Oncol Rep 2006;16:203-6.
Page K, Graham EAM. Cancer and forensic microsatellites. Forensic Sci Med Pathol 2008;4:60-6.
Ananian V, Tozzo P, Ponzano E, Nitti D, Rodriguez D, Caenazzo L. Tumoural specimens for forensic purposes: comparison of genetic alterations in frozen and formalin-fixed paraffin-embedded tissues. Int J Legal Med 2011;125:327-32.
Carew JS, Huang P. Mitochondrial defects in cancer. Mol Cancer 2002;1:9.
Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer. Oncogene 2006;25:4647-62.
Fendt L, Niederstätter H, Huber G, Zelger B, Dünser M, Seifarth C, et al.
Accumulation of mutations over the entire mitochondrial genome of breast cancer cells obtained by tissue microdissection. Breast Cancer Res Treat 2011;128:327-36.
Tan DJ, Bai RK, Wong LJ. Comprehensive scanning of somatic mitochondrial DNA mutations in breast cancer. Cancer Res 2002;62:972-6.
Zhu W, Qin W, Bradley P, Wessel A, Puckett CL, Sauter ER. Mitochondrial DNA mutations in breast cancer tissue and in matched nipple aspirate fluid. Carcinogenesis 2005;26:145-52.
Köhnemann S, Pfeiffer H. Application of mtDNA SNP analysis in forensic casework. Forensic Sci Int Genet 2011;5:216-21.
Köhnemann S, Hohoff C, Pfeiffer H. An economical mtDNA SNP assay detecting different mitochondrial haplogroups in identical HVR 1 samples of Caucasian ancestry. Mitochondrion 2009;9:370-5.
Ren Z, Luo H, Song F, Wei W, Yang Y, Zhai X, et al.
Developing a multiplex mtSNP assay for forensic application in Han Chinese based on mtDNA phylogeny and hot spot. Electrophoresis 2015;36:633-9.
Zha L, Yun L, Chen P, Luo H, Yan J, Hou Y. Exploring of tri-allelic SNPs using pyrosequencing and the SNaPshot methods for forensic application. Electrophoresis 2012;33:841-8.
Poetsch M, Petersmann A, Woenckhaus C, Protzel C, Dittberner T, Lignitz E, et al.
Evaluation of allelic alterations in short tandem repeats in different kinds of solid tumors – Possible pitfalls in forensic casework. Forensic Sci Int 2004;145:1-6.
Parrella P, Xiao Y, Fliss M, Sanchez-Cespedes M, Mazzarelli P, Rinaldi M, et al.
Detection of mitochondrial DNA mutations in primary breast cancer and fine-needle aspirates. Cancer Res 2001;61:7623-6.
Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW. Cancer genome landscapes. Science 2013;339:1546-58.
Fliss MS, Usadel H, Caballero OL, Wu L, Buta MR, Eleff SM, et al.
Facile detection of mitochondrial DNA mutations in tumors and bodily fluids. Science 2000;287:2017-9.
Habano W, Sugai T, Nakamura SI, Uesugi N, Yoshida T, Sasou S. Microsatellite instability and mutation of mitochondrial and nuclear DNA in gastric carcinoma. Gastroenterology 2000;118:835-41.
He Y, Wu J, Dressman DC, Iacobuzio-Donahue C, Markowitz SD, Velculescu VE, et al.
Heteroplasmic mitochondrial DNA mutations in normal and tumour cells. Nature 2010;464:610-4.
[Table 1], [Table 2]