Background: PGC1A and PGC1B encode transcriptional factors that regulate mitochondrial biogenesis... more Background: PGC1A and PGC1B encode transcriptional factors that regulate mitochondrial biogenesis and have been implicated as a link between telomeric and mitochondrial functions. Telomere dysfunction is associated with impaired mitochondrial biogenesis and increased generation of reactive oxygen species. Animal studies have also shown that loss of PGC1A protects against cancer as mitochondrial function is required for transformation and tumor growth. No human studies have examined the role of these genes in relation to cancer incidence. Our objective was to examine associations between PGC1A and PGC1B methylation in relation to cancer incidence. Methods: We studied 491 Normative Aging Study participants who had blood drawn 1-4 times from 1999 through 2012. After median 10.2-year follow-up, there were 125 incident cancers including 36 prostate. PGC1A and PGC1B methylation was measured at 30 and 27 sites, respectively, using the HumanMethylation450k assay. We used Cox proportional hazards models to examine associations between CpG methylation and cancer incidence measured at first blood draw as well as in time-dependent associations. All models adjusted for age, race, smoking status and pack-years, alcohol consumption, blood cell composition, and processing batch. Results: In single-site adjusted time-dependent models, we identified 13 instances of 11 individual loci (10 from PGC1A and 1 from PGC1B) significantly associated with all-cancer and prostate cancer incidence. In multi-site adjusted models, we showed cg0677257 from the 3’ UTR of PGC1A was inversely associated with first-visit (HR: 0.41; 95% CI: 0.21, 0.79) and time-dependent all-cancer incidence (HR: 0.42; 95% CI: 0.22, 0.81). Similarly, we showed cg0942771 and cg03281309 from the gene body of PGC1A were inversely associated with time-dependent all-cancer incidence (HR: 0.51; 95% CI: 0.30, 0.88; HR: 0.74; 95% CI: 0.56, 0.98, respectively). Finally, we found that cg1521939 from the 5’ UTR region of PGC1B was positively associated with time-dependent all-cancer incidence (HR: 3.27; 95% CI: 1.55, 6.90). Conclusion: This is the first study in humans to show site-specific PGC1A and PGC1B methylation is a significant predictor of cancer incidence. These findings point to a possible interplay between telomere and mitochondrial dysfunction, and epigenetics in carcinogenesis. Additional studies should also examine these associations in larger cohorts with greater racial/ethnic, gender, and socioeconomic diversity to validate these findings. Citation Format: Jacob K. Kresovich, Tao Gao, Brian T. Joyce, Pantel Vokonas, Joel Schwartz, Andrea A. Baccarelli, Lifang Hou. DNA methylation of mitochondrial biogenesis regulating genes: A possible link between telomeres, mitochondria, and cancer incidence [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4251. doi:10.1158/1538-7445.AM2017-4251
Background: Telomeres maintain genomic stability and regulate cellular senescence. Numerous studi... more Background: Telomeres maintain genomic stability and regulate cellular senescence. Numerous studies have examined blood telomere length (BTL) and cancer risk, including our own which previously identified a dynamic relationship between BTL and cancer risk over time: Relative to subjects who remained cancer-free, those who later developed cancer experienced accelerated telomere shortening until 3-4 years prior to cancer diagnosis/censoring, at which point BTL in the incident cancer group stabilized. Other studies have suggested that cancer cells are able to hijack telomere maintenance mechanisms (e.g., telomerase) in order to circumvent the Hayflick limit, which will otherwise induce senescence and limit cancer cells’ ability to multiply. As DNA methylation dysregulation is a known epigenetic hallmark of cancer, our objective was to understand whether DNA methylation of telomere biology genes is associated with cancer risk and BTL. Methods: We studied 915 blood measurements from 491 subjects in the Normative Aging Study taken between 1999 and 2013, who were cancer free in 1999 and randomly selected for whole-epigenome DNA methylation profiling using the Illumina 450K BeadChip array. Among these subjects, 125 cases developed cancer and 366 controls remained cancer free for our entire follow up (median 10.1 years). Our analysis focused on 2,731 CpG sites on 142 genes related to telomere length maintenance. We used linear models to identify CpGs differentially methylated by cancer status at the first blood draw only, followed by mixed models using our full data to determine the final CpG sites of interest for Cox models of cancer risk, and another set of mixed models to compare trajectories over time of methylation in cancer patients vs. controls. Finally an additional mixed model explored the temporal relationship between methylation measured at the first visit and BTL measured at the second visit. Results: Our screening identified 167 CpG sites, 10 of which were associated with time-dependent cancer risk at FDR <0.05. Methylation of one CpG on each of CDC73, NSMCE2, and RPA1 was associated with decreased cancer risk. Methylation of one CpG on each of DCLRE1C, DDB1, H2AFY, HNRNPA2B1, and TEP1, as well as two CpGs on TERF2, was associated with increased cancer risk. DNA methylation trajectory at three CpGs was significantly different across cancer status (cg20772347 on DDB1, cg23157637 on TEP1, and cg04818274 on TERF2). One CpG on NSMCE2 and one on TERF2 were associated with BTL in cancer-free subjects, while another CpG on TERF2 was associated with BTL in subjects who later developed cancer. Conclusion: These findings suggest that methylation changes in telomere maintenance genes may be a mechanism by which cancer cells alter telomere length, thus undermining genomic stability and protecting cancer cells from senescence. Future studies should confirm these findings, and explore these CpG sites and genes as potential early detection biomarkers and therapeutic targets. Citation Format: Brian T. Joyce, Yang Li, Yinan Zheng, Lei Liu, Hushan Yang, Chad Achenbach, Pantel Vokonas, Joel Schwartz, Andrea Baccarelli, Lifang Hou. Telomere biology gene methylation and cancer risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2264. doi:10.1158/1538-7445.AM2017-2264
INTRODUCTION: Environmental exposures to heavy metals are common in the general population and ar... more INTRODUCTION: Environmental exposures to heavy metals are common in the general population and are related to a wide range of adverse health effects. Possible mechanisms include metal-induced chang...
Table S1: Detailed information on all CpGs of interest Table S2: Cancer-associated CpGs in TRGs b... more Table S1: Detailed information on all CpGs of interest Table S2: Cancer-associated CpGs in TRGs by pathway at FDR < 0.05, with unspecified skin malignancies excluded Table S3: Methylation trajectory over time by cancer status Table S4: Logistic regression results for DNA methylation measured over 8 years before diagnosis/censoring Table S5: Regulatory elements enrichment analysis results Figure S1: Plots of Beta-coefficients by genomic location for all CpGs on six genes of interest Figure S2: DNA methylation by years to cancer diagnosis/censoring and cancers status with 95% confidence intervals Figure S3: DNA methylation by years to cancer diagnosis/censoring and cancer status for CpG additional sites Figure S4: Regulatory enrichment analysis results
Low levels of physical activity are associated with increased risk and mortality of common cancer... more Low levels of physical activity are associated with increased risk and mortality of common cancers including breast and colon, with strong evidence for increasing risk with age. Epigenetic age is a recently developed concept using DNA methylation measurements to describe biological age at the level of human tissues, cells, and organs. Epigenetic age represents deviation from expected DNA methylation patterns associated with chronological age (defined as Δ_age), and is suggested as a possible biological mechanism driving observed associations between physical activity and cancer risk. However, no studies have evaluated associations between physical activity level and Δ_age. The Normative Aging Study (NAS) is a cohort study of healthy aging conducted by the Veterans Affair Administration. Among NAS participants (n=656), we conducted multivariate linear regression to prospectively evaluate associations between physical activity (MET-hours per week) and Δ_age across two follow-up visits. The visits were on average 3.6 years apart, with a range of 2-8 years. Δ_age was estimated using Horvath’s DNA methylation online calculator and based upon DNA methylation patterns at the 71 cytosine-phosphate-guanine (CpGs) loci used in Hannum’s model. The calculator accounts for cell proportion of white blood cells. We adjusted models for chronological age, body mass index, education, smoking, alcohol consumption, and prevalent cancer status. Overall, we observed statistically significant associations between high physical activity levels and lower Δ_age, or younger epigenetic age. Among individuals reporting the highest level of physical activity (≥ 30 MET-hours per week), Δ_age was 1.44 (95% CI: 0.43 to 2.45) and 2.20 (95% CI: 1.04 to 3.36) years lower compared to those reporting low activity (<12 MET-hours per week) at the first and second visit, respectively. Furthermore, compared to low physical activity, high activity at the first visit was associated with 2.46 (95% CI: 1.25 to 3.67) years lower Δ_age at the second visit, and 3.72 (95% CI: 1.08 to 6.36) months/calendar year lower Δ_age rate of change, or decelerated epigenetic aging over time. In summary, the results of the current study demonstrate that physical activity influences DNA methylation patterns at the CpG loci associated with epigenetic age. Future studies are necessary to characterize further the relationship between physical activity and epigenetic age, especially as a potential biological mechanism of action in carcinogenesis. However, epigenetic age represents a unique set of markers with the potential to identify high-risk populations and, in turn, develop interventions to mitigate risk of cancer. Citation Format: Elizabeth A. Hibler, Yinan Zheng, Lei Liu, Wei Zhang, Frank J. Penedo, Siobhan M. Phillips, David E. Conroy, Joel Schwartz, Pantel Vokonas, Andrea Baccarelli, Qi Dai, Lifang Hou. Physical activity and epigenetic age among normative aging study participants. [abstract]. In: Proceedings of the AACR Special Conference: Developmental Biology and Cancer; Nov 30-Dec 3, 2015; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Res 2016;14(4_Suppl):Abstract nr A17.
Background: PGC1A and PGC1B encode transcriptional factors that regulate mitochondrial biogenesis... more Background: PGC1A and PGC1B encode transcriptional factors that regulate mitochondrial biogenesis and have been implicated as a link between telomeric and mitochondrial functions. Telomere dysfunction is associated with impaired mitochondrial biogenesis and increased generation of reactive oxygen species. Animal studies have also shown that loss of PGC1A protects against cancer as mitochondrial function is required for transformation and tumor growth. No human studies have examined the role of these genes in relation to cancer incidence. Our objective was to examine associations between PGC1A and PGC1B methylation in relation to cancer incidence. Methods: We studied 491 Normative Aging Study participants who had blood drawn 1-4 times from 1999 through 2012. After median 10.2-year follow-up, there were 125 incident cancers including 36 prostate. PGC1A and PGC1B methylation was measured at 30 and 27 sites, respectively, using the HumanMethylation450k assay. We used Cox proportional hazards models to examine associations between CpG methylation and cancer incidence measured at first blood draw as well as in time-dependent associations. All models adjusted for age, race, smoking status and pack-years, alcohol consumption, blood cell composition, and processing batch. Results: In single-site adjusted time-dependent models, we identified 13 instances of 11 individual loci (10 from PGC1A and 1 from PGC1B) significantly associated with all-cancer and prostate cancer incidence. In multi-site adjusted models, we showed cg0677257 from the 3’ UTR of PGC1A was inversely associated with first-visit (HR: 0.41; 95% CI: 0.21, 0.79) and time-dependent all-cancer incidence (HR: 0.42; 95% CI: 0.22, 0.81). Similarly, we showed cg0942771 and cg03281309 from the gene body of PGC1A were inversely associated with time-dependent all-cancer incidence (HR: 0.51; 95% CI: 0.30, 0.88; HR: 0.74; 95% CI: 0.56, 0.98, respectively). Finally, we found that cg1521939 from the 5’ UTR region of PGC1B was positively associated with time-dependent all-cancer incidence (HR: 3.27; 95% CI: 1.55, 6.90). Conclusion: This is the first study in humans to show site-specific PGC1A and PGC1B methylation is a significant predictor of cancer incidence. These findings point to a possible interplay between telomere and mitochondrial dysfunction, and epigenetics in carcinogenesis. Additional studies should also examine these associations in larger cohorts with greater racial/ethnic, gender, and socioeconomic diversity to validate these findings. Citation Format: Jacob K. Kresovich, Tao Gao, Brian T. Joyce, Pantel Vokonas, Joel Schwartz, Andrea A. Baccarelli, Lifang Hou. DNA methylation of mitochondrial biogenesis regulating genes: A possible link between telomeres, mitochondria, and cancer incidence [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4251. doi:10.1158/1538-7445.AM2017-4251
Background: Telomeres maintain genomic stability and regulate cellular senescence. Numerous studi... more Background: Telomeres maintain genomic stability and regulate cellular senescence. Numerous studies have examined blood telomere length (BTL) and cancer risk, including our own which previously identified a dynamic relationship between BTL and cancer risk over time: Relative to subjects who remained cancer-free, those who later developed cancer experienced accelerated telomere shortening until 3-4 years prior to cancer diagnosis/censoring, at which point BTL in the incident cancer group stabilized. Other studies have suggested that cancer cells are able to hijack telomere maintenance mechanisms (e.g., telomerase) in order to circumvent the Hayflick limit, which will otherwise induce senescence and limit cancer cells’ ability to multiply. As DNA methylation dysregulation is a known epigenetic hallmark of cancer, our objective was to understand whether DNA methylation of telomere biology genes is associated with cancer risk and BTL. Methods: We studied 915 blood measurements from 491 subjects in the Normative Aging Study taken between 1999 and 2013, who were cancer free in 1999 and randomly selected for whole-epigenome DNA methylation profiling using the Illumina 450K BeadChip array. Among these subjects, 125 cases developed cancer and 366 controls remained cancer free for our entire follow up (median 10.1 years). Our analysis focused on 2,731 CpG sites on 142 genes related to telomere length maintenance. We used linear models to identify CpGs differentially methylated by cancer status at the first blood draw only, followed by mixed models using our full data to determine the final CpG sites of interest for Cox models of cancer risk, and another set of mixed models to compare trajectories over time of methylation in cancer patients vs. controls. Finally an additional mixed model explored the temporal relationship between methylation measured at the first visit and BTL measured at the second visit. Results: Our screening identified 167 CpG sites, 10 of which were associated with time-dependent cancer risk at FDR <0.05. Methylation of one CpG on each of CDC73, NSMCE2, and RPA1 was associated with decreased cancer risk. Methylation of one CpG on each of DCLRE1C, DDB1, H2AFY, HNRNPA2B1, and TEP1, as well as two CpGs on TERF2, was associated with increased cancer risk. DNA methylation trajectory at three CpGs was significantly different across cancer status (cg20772347 on DDB1, cg23157637 on TEP1, and cg04818274 on TERF2). One CpG on NSMCE2 and one on TERF2 were associated with BTL in cancer-free subjects, while another CpG on TERF2 was associated with BTL in subjects who later developed cancer. Conclusion: These findings suggest that methylation changes in telomere maintenance genes may be a mechanism by which cancer cells alter telomere length, thus undermining genomic stability and protecting cancer cells from senescence. Future studies should confirm these findings, and explore these CpG sites and genes as potential early detection biomarkers and therapeutic targets. Citation Format: Brian T. Joyce, Yang Li, Yinan Zheng, Lei Liu, Hushan Yang, Chad Achenbach, Pantel Vokonas, Joel Schwartz, Andrea Baccarelli, Lifang Hou. Telomere biology gene methylation and cancer risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2264. doi:10.1158/1538-7445.AM2017-2264
INTRODUCTION: Environmental exposures to heavy metals are common in the general population and ar... more INTRODUCTION: Environmental exposures to heavy metals are common in the general population and are related to a wide range of adverse health effects. Possible mechanisms include metal-induced chang...
Table S1: Detailed information on all CpGs of interest Table S2: Cancer-associated CpGs in TRGs b... more Table S1: Detailed information on all CpGs of interest Table S2: Cancer-associated CpGs in TRGs by pathway at FDR < 0.05, with unspecified skin malignancies excluded Table S3: Methylation trajectory over time by cancer status Table S4: Logistic regression results for DNA methylation measured over 8 years before diagnosis/censoring Table S5: Regulatory elements enrichment analysis results Figure S1: Plots of Beta-coefficients by genomic location for all CpGs on six genes of interest Figure S2: DNA methylation by years to cancer diagnosis/censoring and cancers status with 95% confidence intervals Figure S3: DNA methylation by years to cancer diagnosis/censoring and cancer status for CpG additional sites Figure S4: Regulatory enrichment analysis results
Low levels of physical activity are associated with increased risk and mortality of common cancer... more Low levels of physical activity are associated with increased risk and mortality of common cancers including breast and colon, with strong evidence for increasing risk with age. Epigenetic age is a recently developed concept using DNA methylation measurements to describe biological age at the level of human tissues, cells, and organs. Epigenetic age represents deviation from expected DNA methylation patterns associated with chronological age (defined as Δ_age), and is suggested as a possible biological mechanism driving observed associations between physical activity and cancer risk. However, no studies have evaluated associations between physical activity level and Δ_age. The Normative Aging Study (NAS) is a cohort study of healthy aging conducted by the Veterans Affair Administration. Among NAS participants (n=656), we conducted multivariate linear regression to prospectively evaluate associations between physical activity (MET-hours per week) and Δ_age across two follow-up visits. The visits were on average 3.6 years apart, with a range of 2-8 years. Δ_age was estimated using Horvath’s DNA methylation online calculator and based upon DNA methylation patterns at the 71 cytosine-phosphate-guanine (CpGs) loci used in Hannum’s model. The calculator accounts for cell proportion of white blood cells. We adjusted models for chronological age, body mass index, education, smoking, alcohol consumption, and prevalent cancer status. Overall, we observed statistically significant associations between high physical activity levels and lower Δ_age, or younger epigenetic age. Among individuals reporting the highest level of physical activity (≥ 30 MET-hours per week), Δ_age was 1.44 (95% CI: 0.43 to 2.45) and 2.20 (95% CI: 1.04 to 3.36) years lower compared to those reporting low activity (<12 MET-hours per week) at the first and second visit, respectively. Furthermore, compared to low physical activity, high activity at the first visit was associated with 2.46 (95% CI: 1.25 to 3.67) years lower Δ_age at the second visit, and 3.72 (95% CI: 1.08 to 6.36) months/calendar year lower Δ_age rate of change, or decelerated epigenetic aging over time. In summary, the results of the current study demonstrate that physical activity influences DNA methylation patterns at the CpG loci associated with epigenetic age. Future studies are necessary to characterize further the relationship between physical activity and epigenetic age, especially as a potential biological mechanism of action in carcinogenesis. However, epigenetic age represents a unique set of markers with the potential to identify high-risk populations and, in turn, develop interventions to mitigate risk of cancer. Citation Format: Elizabeth A. Hibler, Yinan Zheng, Lei Liu, Wei Zhang, Frank J. Penedo, Siobhan M. Phillips, David E. Conroy, Joel Schwartz, Pantel Vokonas, Andrea Baccarelli, Qi Dai, Lifang Hou. Physical activity and epigenetic age among normative aging study participants. [abstract]. In: Proceedings of the AACR Special Conference: Developmental Biology and Cancer; Nov 30-Dec 3, 2015; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Res 2016;14(4_Suppl):Abstract nr A17.
Uploads
Papers by Pantel Vokonas