- Email: [email protected]
North American White Mitochondrial Haplogroups in Prostate and Renal Cancer
North American White Mitochondrial Haplogroups in Prostate and Renal Cancer Lyra M. Booker, Geoffrey M. Habermacher, Benjamin C. Jessie, Qi Carrie Sun, Amanda K. Baumann, Mahul Amin, So Dug Lim, Carina Fernandez-Golarz, Robert H. Lyles, Michael D. Brown, Fray F. Marshall* and John A. Petros† From the Departments of Urology (LMB, GMH, BCJ, QCS, AKB, CF-G, FFM, JAP) and Pathology and Laboratory Medicine (MA, SDL, JAP) and School of Public Health (RHL), Emory University and Atlanta Veterans Affairs Medical Center (JAP), Atlanta and Mercer University School of Medicine (MDB), Macon, Georgia
Purpose: While the mitochondrion is known to be a key mediator of apoptosis, there has been little inquiry into the inheritance pattern of mitochondria in patients with cancer. We compared the mtDNA haplotype in patients with prostate and renal cancer to that in controls to determine if there is an association between mitochondrial genotype and cancer. Materials and Methods: Haplotyping was performed using polymerase chain reaction/digest identification of key polymorphic sites in the mitochondrial genome. A total of 121 and 221 white men with renal and prostate cancer, respectively, were identified following pathological confirmation of cancer, while 246 white controls were selected randomly from a bank of cadaveric organ donor DNA. Statistical analysis was performed and ORs were calculated. Results: Mitochondrial haplogroup U was a highly significant risk factor for prostate and renal cancer vs controls (16.74% and 20.66% vs 9.35%, Fisher’s exact test p ⫽ 0.019 and 0.005, respectively). The association remained statistically significant in renal cancer even after Bonferroni adjustment for multiple comparisons. Haplogroup U carried an OR of 1.95 for prostate cancer and an OR of 2.52 for renal cancer. Conclusions: The inheritance of mitochondrial haplogroup U is associated with an approximately 2-fold increased risk of prostate cancer and 2.5-fold increased risk of renal cancer in white North American individuals. Therefore, individuals with this mitochondrial haplotype are in a high risk group. Because mitochondrial haplogroup U is found in 9.35% of the white United States population, there are more than 20 million individuals in this high risk group. Key Words: prostatic neoplasms; renal neoplasms; DNA, mitochondrial; European continental ancestry group; risk
role of the mitochondrion in the first committed step in apoptosis, that is the release of cytochrome c. In addition, reactive oxygen species are a natural byproduct of mitochondrial oxidative phosphorylation and reactive oxygen species have mutagenic as well as mitogenic effects relevant to malignant transformation.3 The mitochondrion contains a genome without introns that encodes the machinery of protein production, including a 12s ribosomal RNA, a 16s ribosomal RNA and a complete complement of transfer RNAs. In addition, coding sequences for 13 polypeptides are present, of which all are components of large protein complexes (respiratory complexes I to V) localized in the inner mitochondrial membrane. Unique sets of mitochondrial DNA polymorphisms define the ancestral origin or haplogroup of the mitochondrial DNA in each individual.4 Patients with certain mitochondrial haplogroups are more susceptible to nonmalignant mitochondrial diseases, eg Leber’s hereditary optic neuropathy, Alzheimer’s disease, multiple sclerosis and Parkinson’s disease.5 Because of the central role of mitochondria in regulating apoptosis, we hypothesized that the mitochondrial haplogroup in an individual may protect against or predispose to prostatic and renal carcinomas. A prediction of this hypothesis is that the mitochondrial lineage represented by the mtDNA haplogroup is different in cases and controls. We defined mitochondrial haplogroups in a group of white patients with
n 2005, 232,090 prostate cancer and 36,160 renal cancer cases were diagnosed and 30,350 men died of prostate cancer, while some 12,660 individuals died of renal cancer.1 While other cancers, eg leukemia, lymphoma and testicular carcinoma, are characterized primarily by excessive cellular proliferation, prostate and renal cancers are characterized primarily by resistance to apoptosis. The mitochondrion is a key player in the process of apoptosis. Therefore, we hypothesized that the inherited mitochondrial genotype in an individual may predispose to or protect patients from these cancer types, which would be reflected in different frequencies of mitochondrial genotypes in patients with cancer compared to controls. Because the relative activities of glycolysis and oxidative phosphorylation are frequently altered in cancer, the mitochondrion has been implicated in the biology of cancer for decades.2 Current understanding centers around the proven
I
Submitted for publication March 7, 2005 Supported by Grant DAMND 17-00-1-0080 from the Department of Defense and Grants CA96994 and CA98912 from the National Institutes of Health (JAP). * Financial interest and/or other relationship with Johnson & Johnson. † Correspondence and requests for reprints: Department of Urology, Emory University School of Medicine, 1365 Clifton Rd., Atlanta, Georgia 30322 (telephone: 404-778-4847; FAX: 404-778-5016; e-mail: [email protected]).
0022-5347/06/1752-0468/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
468
Vol. 175, 468-473, February 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00163-1
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
469
Haplogroup distribution in patients with prostate and renal cancer, and controls Haplogroup (polymorphic sites)
Primer Coordinates (forward, reverse)
H (⫺7025 AluI) I: ⫺1715DdeI ⫹8249AvaII ⫹10028AluI J (⫺16065HinfI) K: ⫺9052HaeII ⫹12308HinfI T: ⫹13366BamI-II ⫹15606AluI U (⫹12308HinfI) V (⫺4577NlaIII) W: ⫹8249AvaII ⫺8994HaeIII X (⫺1715DdeI) Other Totals
Prostate
Renal Ca
No. Control (%)
No. (%)
OR
No. (%)
OR
6890–6909, 7131–7115 7115 1615–1643, 1894–1874 8188–8207, 8366–8345 9911–9932, 10107–10088 15838–15857, 16261–16242
108 (43.90)
90 (40.72)
0.88
45 (37.19)
0.76
9 (3.66)
5 (2.26)
0.61
2 (1.65)
0.44
20 (8.13)
21 (9.50)
1.19
9 (7.44)
0.91
8829–8845, 9184–9163 12104–12124, 12338–12309
23 (9.35)
18 (8.15)
0.86
14 (11.57)
1.27
13172–13190, 13403–13384 15409–15428, 15701–15682 12104–12124, 12338–12309 4500–4519, 4680–4661
30 (12.20)
21 (9.50)
0.76
5* (4.13)
0.31
23 (9.35) 5 (2.03)
37† (16.74) 4 (1.81)
1.95 0.89
25‡ (20.66) 5 (4.13)
2.52 2.08
5 (2.03)
5 (2.26)
1.12
1 (0.83)
0.40
3 (1.22) 20 (8.13) 246
5 (2.26) 15 (6.79) 221
1.87 0.82
4 (3.31) 11 (9.09) 121
2.77 1.13
8188–8207, 8366–8345 8829–8845, 9184–9163 1615–1643, 1894–1874
* U over represented in renal cancer (p ⫽ 0.005). † U over represented in prostate cancer (p ⫽ 0.019). ‡ T under represented in renal cancer (p ⫽ 0.013).
prostate and renal cancer, and compared them to haplogroups in a Georgia region North American white population control group, thereby, testing this prediction and hypothesis. MATERIALS AND METHODS Cases and controls. We analyzed 121 consecutive white patients 17 to 88 years old (average age ⫾ SD 58.40 ⫾ 13.58) who had undergone radical nephrectomy in the 1990s for renal cancer and 221 consecutive patients with prostate cancer 40 to 81 years old (average age 56.71 ⫾ 8.49) accrued between 1997 and 2001. Data gathering and analysis were done in compliance with federal and institutional regulations (Health Insurance Portability and Accountability Act and the Emory University Institutional Review Board). Because there are slight differences in mitochondrial haplogroups in white individuals from different regions of the United States due to immigration patterns, we chose a control group of 246 patients 10 months to 69 years old (average age 34.21 ⫾ 16.55) from the same region as the patients with cancer, making the differences in haplogroup distribution between cases and controls due only to disease status and to not regional variation or ethnic origin. Source of DNA. Total genomic DNA was prepared from fresh frozen tissue procured prospectively as part of ongoing tissue banking in the 121 patients with renal cancer and 157 of those with prostate cancer. An additional 64 prostate cancer tissue DNAs were prepared from paraffin embedded tissues. Control DNAs were prepared from frozen tissues obtained from 246 organ donors at Emory University Hospital. Because mitochondrial haplogroup assignment depends on the identification of a limited set of precisely defined polymorphisms that are equally present in every cell in the body, no attempt was made to differentiate cancerous from normal tissue. Determination of mtDNA haplogroup. North American individuals of European origin belong to 1 of 9 mitochondrial
haplogroups, namely H, I, J, K, T, U, V, W or X.6 Primers were designed to amplify the region of mitochondrial DNA that defines each haplogroup (see table). Standard polymerase chain reaction amplification conditions were used, followed by digests that define European haplogroups. A haplogroup assignment was confirmed when a polymorphic mutation was present in mitochondrial DNA (see table). The frequency of individual haplogroups in patients with cancer was then statistically compared to that in controls using Fisher’s 2-tailed exact test. Given the 9 haplogroups considered, there were 9 tests performed in the comparison between each group of patients with cancer and controls. To consider the effects of multiple testing the ␣ level was adjusted downward by applying a conservative Bonferroni correction. ORs were also calculated in patients with cancer in each haplogroup. Due to the relatively low prevalence of prostate and renal cancers (prostate cancer is one of the most commonly diagnosed malignancies in men but its incidence is relatively low compared to that of nonmalignant diseases) OR estimates are valid estimates of the cancer relative risk for the different haplogroups (see table).
RESULTS AND DISCUSSION A total of 206 of the 221 patients with prostate cancer and 110 of the 121 with renal cancer were in the 9 mitochondrial haplogroups analyzed in this study, namely H, I, J, K, T, U, V, W and X. All remaining patients with cancer were grouped as other, a group that is expected to contain white North American individuals who may not be of European heritage, eg haplotype N1b. Following the assignment of haplogroup the prevalence of each haplogroup was determined in each cancer group. There were no significant differences between cancer groups and the regional control distributions for haplogroups H, I, J, K, V, W and X. Haplogroup U was significantly over represented in patients with prostate and renal cancer, while haplogroup T was under represented in patients with renal cancer (see table). This association just missed statistical significance. To assess the
470
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
FIG. 1. Percent of patients with prostate and renal cancer, and controls in each haplogroup. A, U haplogroup was significantly over represented in patients with prostate cancer (p ⫽ 0.019). B, U haplogroup was significantly over represented in patients with renal cancer (p ⫽ 0.005), while T haplogroup was significantly under represented (p ⫽ 0.013).
effect of the other group on these results statistical analyses were repeated, excluding this group. The results obtained in terms of significance and the ␣ level were essentially the same and remained significant (data not shown). When the haplogroup distribution in patients with prostate cancer was compared to that in the control group, a significantly greater percent of patients with prostate cancer was in mitochondrial haplogroup U. Of the prostate cancer population 16.74% of patients (37 of 221) compared to 9.35% of the regional control population (23 of 246) were members of the U haplogroup ( Fisher’s exact test p ⫽ 0.019). The OR for prostate cancer in individuals in mitochondrial haplogroup U was 1.95, suggesting that the U haplogroup is a prostate cancer risk factor (see table and fig. 1, A). When comparing the haplogroup distribution in patients with renal cancer to that in the control group, there was again a significantly greater percent of patients with renal cancer in mitochondrial haplogroup U (20.66% or 25 of 121 vs 9.35% or 23 of 246, Fisher’s exact test p ⫽ 0.005). The p value for the increased prevalence of the U haplogroup in patients with renal cancer also remained statistically significant at the 0.05 level after a conservative Bonferroni ad-
justment. The OR for renal cancer in individuals in mitochondrial haplogroup U was 2.52, suggesting that the U haplogroup is a renal cancer risk factor (see table and fig. 1, B). A significantly smaller percent of patients with renal cancer was also noted in the T haplogroup designation. Of the renal cancer population 4.13% (5 of 121 patients) compared to 12.20% of the control population (30 of 246) were members of the T haplogroup (Fisher’s exact test p ⫽ 0.013). When a conservative Bonferroni adjustment was applied, the p value for the decreased prevalence of the T haplogroup in patients with renal cancer did not remain statistically significant. However, the significance level obtained with Fisher’s exact test combined with the low OR value in individuals in mitochondrial haplogroup T made this haplogroup a potential protective factor against renal cancer (OR ⫽ 0.31, see table and fig. 1, B). More studies would be needed to clarify this association. To ensure that DNA prepared from tissues showed the same haplotyping result as DNA prepared from buffy coat DNA we performed haplotyping polymerase chain reaction and digests in 15 samples for which cancer tissue and buffy coat DNA were available. In all cases the haplogroup result was the same whether tissue or blood was used as the source of DNA. This confirms that haplotyping is accurate when tissues are used as the source of genomic DNA. The restriction pattern was the same in blood and tissues (fig. 2). Disease status in the control group is not known. Because some controls were likely to have or subsequently have prostate or renal cancer, the differences that we found are the minimal possible differences and unidentified patients with cancer in the control group would only have enhanced the statistical significance of the distributional differences if patients could have been culled from the control group. The mean age of controls was less than that of patients with cancer (34 vs 56 to 58 years). The effect of this age difference was minimal since disease status was not determined and controls were chosen because they represented the inherited mtDNA haplogroup distribution of the regional population, which does not change with age or disease state. In this study we document that the distributions of mitochondrial haplogroups in white North American patients with renal and prostate cancer are different from those in a regional control group of North American white individuals. Over representation of the U haplogroup was seen in the renal cancer group at highly significant levels, an association that remained significant even after conservative adjustment for the possible effect of multiple comparisons. Under representation of haplogroup T in patients with renal cancer and over representation of haplogroup U in patients with prostate cancer were also observed, although they were not significant following conservative statistical (Bonferroni) analysis. These findings confirm that mitochondrial composition might serve as a predisposing or protective factor for the subsequent development of these 2 common adult solid tumors. While to our knowledge no previous groups have evaluated the association of inherited mitochondrial haplogroup with cancer, extensive precedent exists for the association of other (benign) diseases with the mitochondrial haplogroup. The H haplogroup has been linked to late onset Alzheimer’s disease and the J haplogroup is connected to Liber’s hereditary optic neuropathy and increased European longevity.7
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
471
A. Haplogroup H Lane 1 2 3 4 5 6 7 8 9 10 11 12 100 1 b.p. ladder
2
3
4
5
6
7
8
9
Lane
HinfI
2
1 2 3 4 5 6 3
4
5
6
7
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue Cancer tissue
10 11 12
B. Haplogroup K
1
Sample number 1 2 3 4 5 + control H20 blank 1 2 3 4 5
8
9
10
7 8 9 10
Sample number 1 5 8 10 + control H20 blank 1 5 8 10
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue
HaeII Lane
1
2
3
4
5
6
7
8
9
10 11
12
1 2 3 4 5 6 7 8 9 10 11 12
Sample number 1 5 8 10 14 + control H20 blank 1 5 8 10 14
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue Cancer tissue
FIG. 2. Representative digests from set of 15 blood/tissue pairs showing haplogroup defining digest in each case for blood and cancer tissue in same patient. A, haplogroup H was defined by loss of AluI site at np 7025. Lanes 1 to 5, AluI digests of blood DNA from 5 patients. Lanes 2 to 4, lanes assigned to haplogroup H whether blood or tissue was genomic DNA source. Lane 6, haplotype H positive control. Lane 7, water blank. Lanes 8 to 12, AluI digests in same patients using DNA from their cancer tissue. B to D, similar results. B, haplogroup K was defined by HinfI site gain at np 12308 and HaeII site loss at np 9052. For HinfI digest blood (lanes 1 to 4) and tissue (lanes 7 to 10) represent same 4 patients and showed identical results. HaeII digests for blood and tissues from same 4 patients demonstrated concordant results, placing patient 2 in haplogroup K. C, haplogroup J was defined by loss of HinfI site at np 16065. Note identical results obtained from blood (lanes 1 to 4) and tissue (lanes 7 to 10), placing patients 2 and 3 in haplogroup J. D, haplogroup T was defined by BamHI site gain at np 13366 and AluI site gain at np 15606. In blood and tissue BamHI and AluI digestions placed each patient in haplogroup T. Identical results were obtained in all 15 patients whose blood and tissue were tested.
European males with decreased sperm motility have been found to belong preferentially to haplogroup T.8 Patients with multiple sclerosis9 and those with the rare disease DIDMOAD syndrome10 have an increased frequency of the T haplogroup. The U haplogroup has been associated with occipital stroke11 as well as with progressive tubulointerstitial nephritis.12 The U haplogroup is unusually common in Finland, where its population frequency reaches 21%.13 Finland has the second highest prostate cancer incidence in Europe and cancer of the prostate is the leading primary site of cancer in Finnish men (http://www-dep.iarc.fr and http:// www.cancerregistry.fi). The unique characteristics of the Finnish population (genetic homogeneity, a clear European genetic profile and unusually high representation of the U haplogroup) provide an excellent opportunity for further investigation into the possible role of U as a cancer predisposition factor. A consequence of having a single mitochondrial haplogroup that is predisposed to prostate and renal cancer is that there should be an increased incidence of renal cancer in patients with prostate cancer and an increased rate of pros-
tate cancer in patients with renal cancer. A review of the literature confirms that each case is true. A population based study of all 3,675 men diagnosed with prostate cancer in the metropolitan Atlanta area between 1975 and 1982 clearly documented an increased rate of renal cancer in the study group compared to that in the local general population.14 A similar study was performed using the New South Wales, Australia central cancer registry for 1972 to 1991 to determine the risk of second malignancies following initial renal cancer or initial prostate cancer. There was a reciprocal relationship of an increased risk of prostate cancer following initial renal cancer and an increased risk of renal cancer following an initial prostate cancer (RR 1.7 and 1.2, respectively).15 A small cohort study of 164 men with prostate cancer in Connecticut revealed an increased rate of subsequent renal cancers compared to the general incidence in that state.16 A cohort of 763 patients with renal cancer treated surgically in New York showed an increased risk of prostate cancer after an initial diagnosis of renal cancer (papillary histology).17
472
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
Two studies of familial cancers in Scandinavian countries are particularly interesting. In the first study 62 Swedish families with hereditary prostate carcinoma were studied and the overall cancer risk was observed in 1,364 firstdegree relatives.18 The standardized incidence ratio was defined as the ratio between the observed and the expected number of cases. There was a significantly increased risk of renal cancer but only in female first-degree relatives (standardized incidence ratio 4.60, 95% CI 1.84 to 9.48). This was the highest ratio of all cancers studied. The second group analyzed the relative risk of all cancers among relatives of 371 Icelandic men with prostate cancer.19 They found that the risk of kidney cancer was higher in first-degree and second-degree female relatives (RR 2.5 and 2.67, respectively). The risk of kidney cancer was not statistically significantly greater in male relatives. The increased risk of cancer in female relatives only is consistent with a mitochondrial predisposition factor.
CONCLUSIONS We report that inheriting mitochondrial haplogroup U confers an increased risk of prostate and renal cancer in North American white individuals. To our knowledge for the first time mitochondrial haplogroups have been shown to have a role as predisposing factors in human cancer. We conclude that the mitochondrial haplogroup U predisposes to renal cancer, while haplogroup T may be protective for renal cancer and haplogroup U may predispose to prostate cancer.
ACKNOWLEDGMENTS C. R. Robinson provided specimens from patients with cancer, Keith Kerstann provided technical advice and support, and Kim Kerstann performed the statistical analysis. Controls were provided by the Histocompatibility and Molecular Immunogenetics Laboratory, Emory University Hospital.
Abbreviations and Acronyms ⫹ ⫽ restriction site gain ⫺ ⫽ restriction site loss mtDNA ⫽ mitochondrial DNA
REFERENCES 1. Cancer Facts and Figures 2005. Atlanta: American Cancer Society, 2005 2. Warburg, O.: On the origin of cancer cells. Science, 123: 309, 1956 3. Gao, N., Ding, M., Zheng, J. Z., Zhang, Z., Leonard, S. S., Liu, K. J. et al: Vanadate-induced expression of hypoxia-inducible factor 1 alpha and vascular endothelial growth factor through phosphatidylinositol 3-kinase/Akt pathway and reactive oxygen species. J Biol Chem, 277: 31963, 2002 4. Wallace, D. C., Lott, M. T., Brown, M. D. and Kerstann, K.: Mitochondria and neuro-opthalmologic diseases. In: The Metabolic and Molecular Bases of Inherited Disease, 8th ed. Edited by C. R. Scriver, A. L. Beaudet, D. Valle, W. S. Sly, B. Childs, K. W. Kinzler et al. New York: McGraw-Hill, pp. 2425-2491, 2001
5. Kalman, B., Li, S., Chatterjee, D. O’Connor, J., Voehl, M. R., Brown, M. D. et al: Large scale screening of the mitochondrial DNA reveals no pathogenic mutations but a haplotype associated with multiple sclerosis in Caucasians. Acta Neurol Scand, 99: 16, 1999 6. Torroni, A., Huoponen, K., Francalacci, P., Petrozzi, M., Morelli, L., Scozzari, R. et al: Classification of European mtDNAs from an analysis of three European populations. Genetics, 144: 1835, 1996 7. De Benedicits, G., Rose, G., Carrieri, G., De Luca, M., Falcone, E., Passarino, G. et al: Mitochondrial DNA inherited variants are associated with successful aging and longevity in humans. FASEB J, 13: 1532, 1999 8. Ruiz-Pesini, E., Lapena, A. C., Diez-Sanchez, C., Perez-Martos, A., Montoya, J., Alvarez, E. et al: Human mtDNA haplogroups associated with high or reduced spermatozoa motility. Am J Hum Genet, 67: 682, 2000 9. Kalman, B., Lublin, F. D. and Alder, H.: Mitochondrial DNA mutations in multiple sclerosis. Mult Scler, 1: 32, 1995 10. Hofmann, S., Bezold, R., Jaksch, M., Obermaier-Kusser, B., Mertens, S., Kaufhold, P. et al: Wolfram (DIDMOAD) syndrome and Leber hereditary optic neuropathy (LHON) are associated with distinct mitochondrial DNA haplotypes. Genomics, 39: 8, 1997 11. Majamaa, K., Finnila, S., Turkka, J. and Hassinen, I. E.: Mitochondrial DNA haplogroup U as a risk factor for occipital stroke in migraine. Lancet, 352: 455, 1998 12. Zsurka, G., Ormos, J., Ivanyi, B., Turi, S., Endreffy, E., Magyari, M. et al: Mitochondrial mutation as a probable causative factor in familial progressive tubulointerstitial nephritis. Hum Genet, 99: 484, 1997 13. Richards, M., Corte-Real, H., Forster, P., Macaulay, V., Wilkinson-Herbots, H., Demaine, A. et al: Paleolithic and neolithic lineages in the European mitochondrial gene pool. Am J Hum Genet, 59: 185, 1996 14. Greenberg, R. S., Rustin, E. D. and Clark, W. S.: Risk of genitourinary malignancies after cancer of the prostate. Cancer 61: 396, 1988 15. McCredie, M., Macfarlane, G. L., Stewart, J. and Coates, M.: Second primary cancers following cancers of the kidney and prostate in New South Wales (Australia), 1972-91. Cancer Causes Control, 7: 337, 1996 16. Johnstone, P. A. S., Powell, C. R., Riffenburgh, R., Rohde, D. C. and Kane, C. J.: Second primary malignancies in T1-3N0 prostate cancer patients treated with radiation therapy with 10-year followup. J Urol, 159: 946, 1998 17. Rabbani, F, Reuter, V. E., Katz, J. and Russo, P.: Second primary malignancies associated with renal cell carcinoma influence of histologic type. Urology, 56: 399, 2000 18. Gronberg, H., Bergh, A., Damber, J. E. and Emanuelsson, M.: Cancer risk in families with hereditary prostate carcinoma. Cancer, 89: 1315, 2000 19. Eldon, B. J., Jonsson, E., Tomasson, J., Tryggvadottir, L. and Tulinius, H.: Familial risk of prostate cancer in Iceland. BJU Int, 92: 915, 2003
EDITORIAL COMMENT In this case-control study the authors report an association between the U haplogroup of mtDNA, and prostate and renal cancers. Maternally inherited germline alterations in mtDNA have been linked to several inherited neurological disorders but not to the risk of developing malignancies. Since mitochondria are intimately involved in apoptosis and responsible for generating oxygen free radicals known to be associated with carcinogenesis, the association of mitochondrial haplogroups with prostate and renal cancers seems
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER reasonable at face value. However, caution should be exercised when interpreting this study. While the authors should be complimented for using a control group of similar race and from the same geographic area as the cases, this does not ensure that there are not differences in the ethnic composition of the 2 groups that could account for a false association. The United States is truly a melting pot, and differences among ethnic groups of the same race due to founder effects will be reflected in the mtDNA haplotype. Therefore, the differences between the case and control groups might merely reflect selection biases in the ethnic composition of the 2 groups that are not captured by race and are due to chance alone. The possibility that biases in the case and control groups confound the association is reflected in the relatively weak statistical association between haplogroup and cancer, although this could also reflect the low sample size of the study. With the Bonferroni adjustment only the association between the U haplogroup and renal cancer remains significant (p ⫽ 0.05). Before accepting the conclusions of this study it will be critical to test the association between mtDNA haplogroup and cancer in another population, preferably one that is more homogeneous. If true, does the association between mtDNA haplogroup, and renal and prostate cancer imply that the haplogroup is somehow involved in carcinogenesis? The mitochondrial proteins and pathways involved in apoptosis and free radical generation are well understood and have little to do with genes encoded by the mitochondrial genome. As of now, there are no plausible biological explanations for the contribution of the mtDNA U haplotype to carcinogenesis. A more likely explanation for the association is that the mtDNA haplogroup tracks with (is a surrogate marker for) other germline alterations in nuclear DNA that predispose to prostate and renal cancers. Again, mtDNA haplogroup reflects founder effects and germline mutations in the nuclear genes of a small ethnic group, such as BRCA1/2 mutations in Ashkenazi Jews, are the real culprits in predisposing to malignancies.1 While the weak association between renal and prostate cancer might reflect the shared risk of harboring the U haplogroup, as the authors assert, it more likely reflects the risk that patients with prostate or renal cancer will undergo imaging and laboratory tests that reveal the other malignancy, thanks to their urologists.2 James D. Brooks Department of Urology Stanford University School of Medicine Stanford, California 1. King, M. C., Marks, J. H. and Mandell, J. B.: Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science, 302: 643, 2003 2. Barocas, D. A., Rabbani, F. and Scherr, D. S.: Population-based study of renal cell carcinoma and prostate cancer in the same patient. J Urol, suppl., 173: 73, abstract 264, 2005
473
REPLY BY AUTHORS While we agree that our findings are novel, we disagree that inherited germ line alterations of mtDNA have not been linked to the risk of developing cancer. In early 2005 we published the first report of germ line (inherited) mtDNA mutations in patients with prostate cancer.1 In that report we demonstrated decreased apoptosis and increased in vivo tumorigenesis in prostate cancers containing a mtDNA mutation compared to nonmutant controls. In similar work from Japan Shidara et al also found that tumors with mutant mtDNA were resistant to apoptosis and grew faster.2 They also demonstrated that the mutant protein was directly responsible for the increase in malignant phenotype by reversion of that phenotype when the wild type gene was reintroduced. The Japanese study argues strongly against the statement that mitochondrial genotype is simply a surrogate marker for a (causative) nuclear DNA alteration. More recently in a population based case control study of breast cancer conducted at the University of North Carolina (The Carolina Breast Cancer Study) researchers from the Vanderbilt University Center for Human Genetics Research evaluated more than 1,200 individuals and found that an inherited mtDNA polymorphism influenced breast cancer susceptibility in black women.3 Finally, the data we present in our article on 588 individuals reveal a statistically significant association between mitochondrial haplogroup U and the risk of developing cancer. This is a clinically significant level of increased risk similar in magnitude to having a first-degree relative with prostate cancer (reference 18 in manuscript).4 This risk is most pronounced in renal cell carcinoma, a disease that currently lacks effective screening tools. Because the incidence of renal cell carcinoma continues to increase, the development of effective means of early detection and diagnosis will remain important for the foreseeable future.5 Because the performance of any screening test depends on the prevalence of disease in the screened population, the accurate identification of a high risk group (mitochondrial haplogroup U) is particularly important. 1. Petros, J. A., Baumann, A. K., Ruiz-Pesini, E., Amin, M. B., Sun, C. Q., Hall, J. et al: MtDNA mutations increase tumorigenicity in prostate cancer. Proc Nate Acad Sciences USA, 102: 719, 2005 2. Shidara, Y., Yamagata, K., Kanamori, T., Nakano, K., Kwong, J. Q., Manfredi, G. et al: Positive contribution of pathogenic mutations in the mitochondrial genome to the promotion of cancer by prevention of apoptosis. Cancer Res, 65: 1655, 2005 3. Canter, J. A., Kallianpur, A. R., Parl, F. F. and Millikan, R. C.: Mitochondiral DNA G10398A polymorphism and invasive breast cancer in African-American women. Cancer Res, 65: 8028, 2005 4. Gronberg, H., Bergh, A., Damber, J. E. and Emanuelsson, M.: Cancer risk in families with hereditary prostate carcinoma. Cancer 89: 1315, 2000 5. Hock, L., Lynch, J. and Balaji, K. C.: Increasing incidence of all stages of kidney cancer in the last 2 decades in the United States: an analysis of surveillance, epidemiology and end results program data. J Urol, 167: 57, 2002
Purpose: While the mitochondrion is known to be a key mediator of apoptosis, there has been little inquiry into the inheritance pattern of mitochondria in patients with cancer. We compared the mtDNA haplotype in patients with prostate and renal cancer to that in controls to determine if there is an association between mitochondrial genotype and cancer. Materials and Methods: Haplotyping was performed using polymerase chain reaction/digest identification of key polymorphic sites in the mitochondrial genome. A total of 121 and 221 white men with renal and prostate cancer, respectively, were identified following pathological confirmation of cancer, while 246 white controls were selected randomly from a bank of cadaveric organ donor DNA. Statistical analysis was performed and ORs were calculated. Results: Mitochondrial haplogroup U was a highly significant risk factor for prostate and renal cancer vs controls (16.74% and 20.66% vs 9.35%, Fisher’s exact test p ⫽ 0.019 and 0.005, respectively). The association remained statistically significant in renal cancer even after Bonferroni adjustment for multiple comparisons. Haplogroup U carried an OR of 1.95 for prostate cancer and an OR of 2.52 for renal cancer. Conclusions: The inheritance of mitochondrial haplogroup U is associated with an approximately 2-fold increased risk of prostate cancer and 2.5-fold increased risk of renal cancer in white North American individuals. Therefore, individuals with this mitochondrial haplotype are in a high risk group. Because mitochondrial haplogroup U is found in 9.35% of the white United States population, there are more than 20 million individuals in this high risk group. Key Words: prostatic neoplasms; renal neoplasms; DNA, mitochondrial; European continental ancestry group; risk
role of the mitochondrion in the first committed step in apoptosis, that is the release of cytochrome c. In addition, reactive oxygen species are a natural byproduct of mitochondrial oxidative phosphorylation and reactive oxygen species have mutagenic as well as mitogenic effects relevant to malignant transformation.3 The mitochondrion contains a genome without introns that encodes the machinery of protein production, including a 12s ribosomal RNA, a 16s ribosomal RNA and a complete complement of transfer RNAs. In addition, coding sequences for 13 polypeptides are present, of which all are components of large protein complexes (respiratory complexes I to V) localized in the inner mitochondrial membrane. Unique sets of mitochondrial DNA polymorphisms define the ancestral origin or haplogroup of the mitochondrial DNA in each individual.4 Patients with certain mitochondrial haplogroups are more susceptible to nonmalignant mitochondrial diseases, eg Leber’s hereditary optic neuropathy, Alzheimer’s disease, multiple sclerosis and Parkinson’s disease.5 Because of the central role of mitochondria in regulating apoptosis, we hypothesized that the mitochondrial haplogroup in an individual may protect against or predispose to prostatic and renal carcinomas. A prediction of this hypothesis is that the mitochondrial lineage represented by the mtDNA haplogroup is different in cases and controls. We defined mitochondrial haplogroups in a group of white patients with
n 2005, 232,090 prostate cancer and 36,160 renal cancer cases were diagnosed and 30,350 men died of prostate cancer, while some 12,660 individuals died of renal cancer.1 While other cancers, eg leukemia, lymphoma and testicular carcinoma, are characterized primarily by excessive cellular proliferation, prostate and renal cancers are characterized primarily by resistance to apoptosis. The mitochondrion is a key player in the process of apoptosis. Therefore, we hypothesized that the inherited mitochondrial genotype in an individual may predispose to or protect patients from these cancer types, which would be reflected in different frequencies of mitochondrial genotypes in patients with cancer compared to controls. Because the relative activities of glycolysis and oxidative phosphorylation are frequently altered in cancer, the mitochondrion has been implicated in the biology of cancer for decades.2 Current understanding centers around the proven
I
Submitted for publication March 7, 2005 Supported by Grant DAMND 17-00-1-0080 from the Department of Defense and Grants CA96994 and CA98912 from the National Institutes of Health (JAP). * Financial interest and/or other relationship with Johnson & Johnson. † Correspondence and requests for reprints: Department of Urology, Emory University School of Medicine, 1365 Clifton Rd., Atlanta, Georgia 30322 (telephone: 404-778-4847; FAX: 404-778-5016; e-mail: [email protected]).
0022-5347/06/1752-0468/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
468
Vol. 175, 468-473, February 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00163-1
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
469
Haplogroup distribution in patients with prostate and renal cancer, and controls Haplogroup (polymorphic sites)
Primer Coordinates (forward, reverse)
H (⫺7025 AluI) I: ⫺1715DdeI ⫹8249AvaII ⫹10028AluI J (⫺16065HinfI) K: ⫺9052HaeII ⫹12308HinfI T: ⫹13366BamI-II ⫹15606AluI U (⫹12308HinfI) V (⫺4577NlaIII) W: ⫹8249AvaII ⫺8994HaeIII X (⫺1715DdeI) Other Totals
Prostate
Renal Ca
No. Control (%)
No. (%)
OR
No. (%)
OR
6890–6909, 7131–7115 7115 1615–1643, 1894–1874 8188–8207, 8366–8345 9911–9932, 10107–10088 15838–15857, 16261–16242
108 (43.90)
90 (40.72)
0.88
45 (37.19)
0.76
9 (3.66)
5 (2.26)
0.61
2 (1.65)
0.44
20 (8.13)
21 (9.50)
1.19
9 (7.44)
0.91
8829–8845, 9184–9163 12104–12124, 12338–12309
23 (9.35)
18 (8.15)
0.86
14 (11.57)
1.27
13172–13190, 13403–13384 15409–15428, 15701–15682 12104–12124, 12338–12309 4500–4519, 4680–4661
30 (12.20)
21 (9.50)
0.76
5* (4.13)
0.31
23 (9.35) 5 (2.03)
37† (16.74) 4 (1.81)
1.95 0.89
25‡ (20.66) 5 (4.13)
2.52 2.08
5 (2.03)
5 (2.26)
1.12
1 (0.83)
0.40
3 (1.22) 20 (8.13) 246
5 (2.26) 15 (6.79) 221
1.87 0.82
4 (3.31) 11 (9.09) 121
2.77 1.13
8188–8207, 8366–8345 8829–8845, 9184–9163 1615–1643, 1894–1874
* U over represented in renal cancer (p ⫽ 0.005). † U over represented in prostate cancer (p ⫽ 0.019). ‡ T under represented in renal cancer (p ⫽ 0.013).
prostate and renal cancer, and compared them to haplogroups in a Georgia region North American white population control group, thereby, testing this prediction and hypothesis. MATERIALS AND METHODS Cases and controls. We analyzed 121 consecutive white patients 17 to 88 years old (average age ⫾ SD 58.40 ⫾ 13.58) who had undergone radical nephrectomy in the 1990s for renal cancer and 221 consecutive patients with prostate cancer 40 to 81 years old (average age 56.71 ⫾ 8.49) accrued between 1997 and 2001. Data gathering and analysis were done in compliance with federal and institutional regulations (Health Insurance Portability and Accountability Act and the Emory University Institutional Review Board). Because there are slight differences in mitochondrial haplogroups in white individuals from different regions of the United States due to immigration patterns, we chose a control group of 246 patients 10 months to 69 years old (average age 34.21 ⫾ 16.55) from the same region as the patients with cancer, making the differences in haplogroup distribution between cases and controls due only to disease status and to not regional variation or ethnic origin. Source of DNA. Total genomic DNA was prepared from fresh frozen tissue procured prospectively as part of ongoing tissue banking in the 121 patients with renal cancer and 157 of those with prostate cancer. An additional 64 prostate cancer tissue DNAs were prepared from paraffin embedded tissues. Control DNAs were prepared from frozen tissues obtained from 246 organ donors at Emory University Hospital. Because mitochondrial haplogroup assignment depends on the identification of a limited set of precisely defined polymorphisms that are equally present in every cell in the body, no attempt was made to differentiate cancerous from normal tissue. Determination of mtDNA haplogroup. North American individuals of European origin belong to 1 of 9 mitochondrial
haplogroups, namely H, I, J, K, T, U, V, W or X.6 Primers were designed to amplify the region of mitochondrial DNA that defines each haplogroup (see table). Standard polymerase chain reaction amplification conditions were used, followed by digests that define European haplogroups. A haplogroup assignment was confirmed when a polymorphic mutation was present in mitochondrial DNA (see table). The frequency of individual haplogroups in patients with cancer was then statistically compared to that in controls using Fisher’s 2-tailed exact test. Given the 9 haplogroups considered, there were 9 tests performed in the comparison between each group of patients with cancer and controls. To consider the effects of multiple testing the ␣ level was adjusted downward by applying a conservative Bonferroni correction. ORs were also calculated in patients with cancer in each haplogroup. Due to the relatively low prevalence of prostate and renal cancers (prostate cancer is one of the most commonly diagnosed malignancies in men but its incidence is relatively low compared to that of nonmalignant diseases) OR estimates are valid estimates of the cancer relative risk for the different haplogroups (see table).
RESULTS AND DISCUSSION A total of 206 of the 221 patients with prostate cancer and 110 of the 121 with renal cancer were in the 9 mitochondrial haplogroups analyzed in this study, namely H, I, J, K, T, U, V, W and X. All remaining patients with cancer were grouped as other, a group that is expected to contain white North American individuals who may not be of European heritage, eg haplotype N1b. Following the assignment of haplogroup the prevalence of each haplogroup was determined in each cancer group. There were no significant differences between cancer groups and the regional control distributions for haplogroups H, I, J, K, V, W and X. Haplogroup U was significantly over represented in patients with prostate and renal cancer, while haplogroup T was under represented in patients with renal cancer (see table). This association just missed statistical significance. To assess the
470
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
FIG. 1. Percent of patients with prostate and renal cancer, and controls in each haplogroup. A, U haplogroup was significantly over represented in patients with prostate cancer (p ⫽ 0.019). B, U haplogroup was significantly over represented in patients with renal cancer (p ⫽ 0.005), while T haplogroup was significantly under represented (p ⫽ 0.013).
effect of the other group on these results statistical analyses were repeated, excluding this group. The results obtained in terms of significance and the ␣ level were essentially the same and remained significant (data not shown). When the haplogroup distribution in patients with prostate cancer was compared to that in the control group, a significantly greater percent of patients with prostate cancer was in mitochondrial haplogroup U. Of the prostate cancer population 16.74% of patients (37 of 221) compared to 9.35% of the regional control population (23 of 246) were members of the U haplogroup ( Fisher’s exact test p ⫽ 0.019). The OR for prostate cancer in individuals in mitochondrial haplogroup U was 1.95, suggesting that the U haplogroup is a prostate cancer risk factor (see table and fig. 1, A). When comparing the haplogroup distribution in patients with renal cancer to that in the control group, there was again a significantly greater percent of patients with renal cancer in mitochondrial haplogroup U (20.66% or 25 of 121 vs 9.35% or 23 of 246, Fisher’s exact test p ⫽ 0.005). The p value for the increased prevalence of the U haplogroup in patients with renal cancer also remained statistically significant at the 0.05 level after a conservative Bonferroni ad-
justment. The OR for renal cancer in individuals in mitochondrial haplogroup U was 2.52, suggesting that the U haplogroup is a renal cancer risk factor (see table and fig. 1, B). A significantly smaller percent of patients with renal cancer was also noted in the T haplogroup designation. Of the renal cancer population 4.13% (5 of 121 patients) compared to 12.20% of the control population (30 of 246) were members of the T haplogroup (Fisher’s exact test p ⫽ 0.013). When a conservative Bonferroni adjustment was applied, the p value for the decreased prevalence of the T haplogroup in patients with renal cancer did not remain statistically significant. However, the significance level obtained with Fisher’s exact test combined with the low OR value in individuals in mitochondrial haplogroup T made this haplogroup a potential protective factor against renal cancer (OR ⫽ 0.31, see table and fig. 1, B). More studies would be needed to clarify this association. To ensure that DNA prepared from tissues showed the same haplotyping result as DNA prepared from buffy coat DNA we performed haplotyping polymerase chain reaction and digests in 15 samples for which cancer tissue and buffy coat DNA were available. In all cases the haplogroup result was the same whether tissue or blood was used as the source of DNA. This confirms that haplotyping is accurate when tissues are used as the source of genomic DNA. The restriction pattern was the same in blood and tissues (fig. 2). Disease status in the control group is not known. Because some controls were likely to have or subsequently have prostate or renal cancer, the differences that we found are the minimal possible differences and unidentified patients with cancer in the control group would only have enhanced the statistical significance of the distributional differences if patients could have been culled from the control group. The mean age of controls was less than that of patients with cancer (34 vs 56 to 58 years). The effect of this age difference was minimal since disease status was not determined and controls were chosen because they represented the inherited mtDNA haplogroup distribution of the regional population, which does not change with age or disease state. In this study we document that the distributions of mitochondrial haplogroups in white North American patients with renal and prostate cancer are different from those in a regional control group of North American white individuals. Over representation of the U haplogroup was seen in the renal cancer group at highly significant levels, an association that remained significant even after conservative adjustment for the possible effect of multiple comparisons. Under representation of haplogroup T in patients with renal cancer and over representation of haplogroup U in patients with prostate cancer were also observed, although they were not significant following conservative statistical (Bonferroni) analysis. These findings confirm that mitochondrial composition might serve as a predisposing or protective factor for the subsequent development of these 2 common adult solid tumors. While to our knowledge no previous groups have evaluated the association of inherited mitochondrial haplogroup with cancer, extensive precedent exists for the association of other (benign) diseases with the mitochondrial haplogroup. The H haplogroup has been linked to late onset Alzheimer’s disease and the J haplogroup is connected to Liber’s hereditary optic neuropathy and increased European longevity.7
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
471
A. Haplogroup H Lane 1 2 3 4 5 6 7 8 9 10 11 12 100 1 b.p. ladder
2
3
4
5
6
7
8
9
Lane
HinfI
2
1 2 3 4 5 6 3
4
5
6
7
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue Cancer tissue
10 11 12
B. Haplogroup K
1
Sample number 1 2 3 4 5 + control H20 blank 1 2 3 4 5
8
9
10
7 8 9 10
Sample number 1 5 8 10 + control H20 blank 1 5 8 10
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue
HaeII Lane
1
2
3
4
5
6
7
8
9
10 11
12
1 2 3 4 5 6 7 8 9 10 11 12
Sample number 1 5 8 10 14 + control H20 blank 1 5 8 10 14
DNA Origin Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Blood (buffy coat) Cancer tissue Cancer tissue Cancer tissue Cancer tissue Cancer tissue
FIG. 2. Representative digests from set of 15 blood/tissue pairs showing haplogroup defining digest in each case for blood and cancer tissue in same patient. A, haplogroup H was defined by loss of AluI site at np 7025. Lanes 1 to 5, AluI digests of blood DNA from 5 patients. Lanes 2 to 4, lanes assigned to haplogroup H whether blood or tissue was genomic DNA source. Lane 6, haplotype H positive control. Lane 7, water blank. Lanes 8 to 12, AluI digests in same patients using DNA from their cancer tissue. B to D, similar results. B, haplogroup K was defined by HinfI site gain at np 12308 and HaeII site loss at np 9052. For HinfI digest blood (lanes 1 to 4) and tissue (lanes 7 to 10) represent same 4 patients and showed identical results. HaeII digests for blood and tissues from same 4 patients demonstrated concordant results, placing patient 2 in haplogroup K. C, haplogroup J was defined by loss of HinfI site at np 16065. Note identical results obtained from blood (lanes 1 to 4) and tissue (lanes 7 to 10), placing patients 2 and 3 in haplogroup J. D, haplogroup T was defined by BamHI site gain at np 13366 and AluI site gain at np 15606. In blood and tissue BamHI and AluI digestions placed each patient in haplogroup T. Identical results were obtained in all 15 patients whose blood and tissue were tested.
European males with decreased sperm motility have been found to belong preferentially to haplogroup T.8 Patients with multiple sclerosis9 and those with the rare disease DIDMOAD syndrome10 have an increased frequency of the T haplogroup. The U haplogroup has been associated with occipital stroke11 as well as with progressive tubulointerstitial nephritis.12 The U haplogroup is unusually common in Finland, where its population frequency reaches 21%.13 Finland has the second highest prostate cancer incidence in Europe and cancer of the prostate is the leading primary site of cancer in Finnish men (http://www-dep.iarc.fr and http:// www.cancerregistry.fi). The unique characteristics of the Finnish population (genetic homogeneity, a clear European genetic profile and unusually high representation of the U haplogroup) provide an excellent opportunity for further investigation into the possible role of U as a cancer predisposition factor. A consequence of having a single mitochondrial haplogroup that is predisposed to prostate and renal cancer is that there should be an increased incidence of renal cancer in patients with prostate cancer and an increased rate of pros-
tate cancer in patients with renal cancer. A review of the literature confirms that each case is true. A population based study of all 3,675 men diagnosed with prostate cancer in the metropolitan Atlanta area between 1975 and 1982 clearly documented an increased rate of renal cancer in the study group compared to that in the local general population.14 A similar study was performed using the New South Wales, Australia central cancer registry for 1972 to 1991 to determine the risk of second malignancies following initial renal cancer or initial prostate cancer. There was a reciprocal relationship of an increased risk of prostate cancer following initial renal cancer and an increased risk of renal cancer following an initial prostate cancer (RR 1.7 and 1.2, respectively).15 A small cohort study of 164 men with prostate cancer in Connecticut revealed an increased rate of subsequent renal cancers compared to the general incidence in that state.16 A cohort of 763 patients with renal cancer treated surgically in New York showed an increased risk of prostate cancer after an initial diagnosis of renal cancer (papillary histology).17
472
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER
Two studies of familial cancers in Scandinavian countries are particularly interesting. In the first study 62 Swedish families with hereditary prostate carcinoma were studied and the overall cancer risk was observed in 1,364 firstdegree relatives.18 The standardized incidence ratio was defined as the ratio between the observed and the expected number of cases. There was a significantly increased risk of renal cancer but only in female first-degree relatives (standardized incidence ratio 4.60, 95% CI 1.84 to 9.48). This was the highest ratio of all cancers studied. The second group analyzed the relative risk of all cancers among relatives of 371 Icelandic men with prostate cancer.19 They found that the risk of kidney cancer was higher in first-degree and second-degree female relatives (RR 2.5 and 2.67, respectively). The risk of kidney cancer was not statistically significantly greater in male relatives. The increased risk of cancer in female relatives only is consistent with a mitochondrial predisposition factor.
CONCLUSIONS We report that inheriting mitochondrial haplogroup U confers an increased risk of prostate and renal cancer in North American white individuals. To our knowledge for the first time mitochondrial haplogroups have been shown to have a role as predisposing factors in human cancer. We conclude that the mitochondrial haplogroup U predisposes to renal cancer, while haplogroup T may be protective for renal cancer and haplogroup U may predispose to prostate cancer.
ACKNOWLEDGMENTS C. R. Robinson provided specimens from patients with cancer, Keith Kerstann provided technical advice and support, and Kim Kerstann performed the statistical analysis. Controls were provided by the Histocompatibility and Molecular Immunogenetics Laboratory, Emory University Hospital.
Abbreviations and Acronyms ⫹ ⫽ restriction site gain ⫺ ⫽ restriction site loss mtDNA ⫽ mitochondrial DNA
REFERENCES 1. Cancer Facts and Figures 2005. Atlanta: American Cancer Society, 2005 2. Warburg, O.: On the origin of cancer cells. Science, 123: 309, 1956 3. Gao, N., Ding, M., Zheng, J. Z., Zhang, Z., Leonard, S. S., Liu, K. J. et al: Vanadate-induced expression of hypoxia-inducible factor 1 alpha and vascular endothelial growth factor through phosphatidylinositol 3-kinase/Akt pathway and reactive oxygen species. J Biol Chem, 277: 31963, 2002 4. Wallace, D. C., Lott, M. T., Brown, M. D. and Kerstann, K.: Mitochondria and neuro-opthalmologic diseases. In: The Metabolic and Molecular Bases of Inherited Disease, 8th ed. Edited by C. R. Scriver, A. L. Beaudet, D. Valle, W. S. Sly, B. Childs, K. W. Kinzler et al. New York: McGraw-Hill, pp. 2425-2491, 2001
5. Kalman, B., Li, S., Chatterjee, D. O’Connor, J., Voehl, M. R., Brown, M. D. et al: Large scale screening of the mitochondrial DNA reveals no pathogenic mutations but a haplotype associated with multiple sclerosis in Caucasians. Acta Neurol Scand, 99: 16, 1999 6. Torroni, A., Huoponen, K., Francalacci, P., Petrozzi, M., Morelli, L., Scozzari, R. et al: Classification of European mtDNAs from an analysis of three European populations. Genetics, 144: 1835, 1996 7. De Benedicits, G., Rose, G., Carrieri, G., De Luca, M., Falcone, E., Passarino, G. et al: Mitochondrial DNA inherited variants are associated with successful aging and longevity in humans. FASEB J, 13: 1532, 1999 8. Ruiz-Pesini, E., Lapena, A. C., Diez-Sanchez, C., Perez-Martos, A., Montoya, J., Alvarez, E. et al: Human mtDNA haplogroups associated with high or reduced spermatozoa motility. Am J Hum Genet, 67: 682, 2000 9. Kalman, B., Lublin, F. D. and Alder, H.: Mitochondrial DNA mutations in multiple sclerosis. Mult Scler, 1: 32, 1995 10. Hofmann, S., Bezold, R., Jaksch, M., Obermaier-Kusser, B., Mertens, S., Kaufhold, P. et al: Wolfram (DIDMOAD) syndrome and Leber hereditary optic neuropathy (LHON) are associated with distinct mitochondrial DNA haplotypes. Genomics, 39: 8, 1997 11. Majamaa, K., Finnila, S., Turkka, J. and Hassinen, I. E.: Mitochondrial DNA haplogroup U as a risk factor for occipital stroke in migraine. Lancet, 352: 455, 1998 12. Zsurka, G., Ormos, J., Ivanyi, B., Turi, S., Endreffy, E., Magyari, M. et al: Mitochondrial mutation as a probable causative factor in familial progressive tubulointerstitial nephritis. Hum Genet, 99: 484, 1997 13. Richards, M., Corte-Real, H., Forster, P., Macaulay, V., Wilkinson-Herbots, H., Demaine, A. et al: Paleolithic and neolithic lineages in the European mitochondrial gene pool. Am J Hum Genet, 59: 185, 1996 14. Greenberg, R. S., Rustin, E. D. and Clark, W. S.: Risk of genitourinary malignancies after cancer of the prostate. Cancer 61: 396, 1988 15. McCredie, M., Macfarlane, G. L., Stewart, J. and Coates, M.: Second primary cancers following cancers of the kidney and prostate in New South Wales (Australia), 1972-91. Cancer Causes Control, 7: 337, 1996 16. Johnstone, P. A. S., Powell, C. R., Riffenburgh, R., Rohde, D. C. and Kane, C. J.: Second primary malignancies in T1-3N0 prostate cancer patients treated with radiation therapy with 10-year followup. J Urol, 159: 946, 1998 17. Rabbani, F, Reuter, V. E., Katz, J. and Russo, P.: Second primary malignancies associated with renal cell carcinoma influence of histologic type. Urology, 56: 399, 2000 18. Gronberg, H., Bergh, A., Damber, J. E. and Emanuelsson, M.: Cancer risk in families with hereditary prostate carcinoma. Cancer, 89: 1315, 2000 19. Eldon, B. J., Jonsson, E., Tomasson, J., Tryggvadottir, L. and Tulinius, H.: Familial risk of prostate cancer in Iceland. BJU Int, 92: 915, 2003
EDITORIAL COMMENT In this case-control study the authors report an association between the U haplogroup of mtDNA, and prostate and renal cancers. Maternally inherited germline alterations in mtDNA have been linked to several inherited neurological disorders but not to the risk of developing malignancies. Since mitochondria are intimately involved in apoptosis and responsible for generating oxygen free radicals known to be associated with carcinogenesis, the association of mitochondrial haplogroups with prostate and renal cancers seems
MITOCHONDRIAL HAPLOGROUP IN PROSTATE AND RENAL CANCER reasonable at face value. However, caution should be exercised when interpreting this study. While the authors should be complimented for using a control group of similar race and from the same geographic area as the cases, this does not ensure that there are not differences in the ethnic composition of the 2 groups that could account for a false association. The United States is truly a melting pot, and differences among ethnic groups of the same race due to founder effects will be reflected in the mtDNA haplotype. Therefore, the differences between the case and control groups might merely reflect selection biases in the ethnic composition of the 2 groups that are not captured by race and are due to chance alone. The possibility that biases in the case and control groups confound the association is reflected in the relatively weak statistical association between haplogroup and cancer, although this could also reflect the low sample size of the study. With the Bonferroni adjustment only the association between the U haplogroup and renal cancer remains significant (p ⫽ 0.05). Before accepting the conclusions of this study it will be critical to test the association between mtDNA haplogroup and cancer in another population, preferably one that is more homogeneous. If true, does the association between mtDNA haplogroup, and renal and prostate cancer imply that the haplogroup is somehow involved in carcinogenesis? The mitochondrial proteins and pathways involved in apoptosis and free radical generation are well understood and have little to do with genes encoded by the mitochondrial genome. As of now, there are no plausible biological explanations for the contribution of the mtDNA U haplotype to carcinogenesis. A more likely explanation for the association is that the mtDNA haplogroup tracks with (is a surrogate marker for) other germline alterations in nuclear DNA that predispose to prostate and renal cancers. Again, mtDNA haplogroup reflects founder effects and germline mutations in the nuclear genes of a small ethnic group, such as BRCA1/2 mutations in Ashkenazi Jews, are the real culprits in predisposing to malignancies.1 While the weak association between renal and prostate cancer might reflect the shared risk of harboring the U haplogroup, as the authors assert, it more likely reflects the risk that patients with prostate or renal cancer will undergo imaging and laboratory tests that reveal the other malignancy, thanks to their urologists.2 James D. Brooks Department of Urology Stanford University School of Medicine Stanford, California 1. King, M. C., Marks, J. H. and Mandell, J. B.: Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science, 302: 643, 2003 2. Barocas, D. A., Rabbani, F. and Scherr, D. S.: Population-based study of renal cell carcinoma and prostate cancer in the same patient. J Urol, suppl., 173: 73, abstract 264, 2005
473
REPLY BY AUTHORS While we agree that our findings are novel, we disagree that inherited germ line alterations of mtDNA have not been linked to the risk of developing cancer. In early 2005 we published the first report of germ line (inherited) mtDNA mutations in patients with prostate cancer.1 In that report we demonstrated decreased apoptosis and increased in vivo tumorigenesis in prostate cancers containing a mtDNA mutation compared to nonmutant controls. In similar work from Japan Shidara et al also found that tumors with mutant mtDNA were resistant to apoptosis and grew faster.2 They also demonstrated that the mutant protein was directly responsible for the increase in malignant phenotype by reversion of that phenotype when the wild type gene was reintroduced. The Japanese study argues strongly against the statement that mitochondrial genotype is simply a surrogate marker for a (causative) nuclear DNA alteration. More recently in a population based case control study of breast cancer conducted at the University of North Carolina (The Carolina Breast Cancer Study) researchers from the Vanderbilt University Center for Human Genetics Research evaluated more than 1,200 individuals and found that an inherited mtDNA polymorphism influenced breast cancer susceptibility in black women.3 Finally, the data we present in our article on 588 individuals reveal a statistically significant association between mitochondrial haplogroup U and the risk of developing cancer. This is a clinically significant level of increased risk similar in magnitude to having a first-degree relative with prostate cancer (reference 18 in manuscript).4 This risk is most pronounced in renal cell carcinoma, a disease that currently lacks effective screening tools. Because the incidence of renal cell carcinoma continues to increase, the development of effective means of early detection and diagnosis will remain important for the foreseeable future.5 Because the performance of any screening test depends on the prevalence of disease in the screened population, the accurate identification of a high risk group (mitochondrial haplogroup U) is particularly important. 1. Petros, J. A., Baumann, A. K., Ruiz-Pesini, E., Amin, M. B., Sun, C. Q., Hall, J. et al: MtDNA mutations increase tumorigenicity in prostate cancer. Proc Nate Acad Sciences USA, 102: 719, 2005 2. Shidara, Y., Yamagata, K., Kanamori, T., Nakano, K., Kwong, J. Q., Manfredi, G. et al: Positive contribution of pathogenic mutations in the mitochondrial genome to the promotion of cancer by prevention of apoptosis. Cancer Res, 65: 1655, 2005 3. Canter, J. A., Kallianpur, A. R., Parl, F. F. and Millikan, R. C.: Mitochondiral DNA G10398A polymorphism and invasive breast cancer in African-American women. Cancer Res, 65: 8028, 2005 4. Gronberg, H., Bergh, A., Damber, J. E. and Emanuelsson, M.: Cancer risk in families with hereditary prostate carcinoma. Cancer 89: 1315, 2000 5. Hock, L., Lynch, J. and Balaji, K. C.: Increasing incidence of all stages of kidney cancer in the last 2 decades in the United States: an analysis of surveillance, epidemiology and end results program data. J Urol, 167: 57, 2002