2018-10-10 Epigenetic Age Signatures In Saliva: Age Prediction Using Methylation SNaPshot and Massively Parallel Sequencing Sae Rom Hong 1,2 , Sang-Eun Jung 1 , Eun Hee Lee 1 , Kyoung-Jin Shin 1,2 , Woo Ick Yang 1 , Hwan Young Lee 1 1 Department of Forensic Medicine, Yonsei University College of Medicine, Seoul, South Korea 2 Department of Forensic Medicine and Brain Korea 21 PLUS Project for Medical Science, Yonsei University, Seoul, South Korea DNA Methylation • Addition of a methyl group to cytosine followed by guanine • 5’-CG-3’ DNMT Cytosine 5-methyl Cytosine 1
2018-10-10 DNA Methylation Cell differentiation Aging [ ] [ ] Body Fluid Age Identification Prediction Genetic factor Environmental factor [ ] [ ] Genetic Behavior Traits Habits G Age Prediction • Age-related molecular changes ‒ Telomere shortening ‒ Mitochonrial DNA deletion ‒ sjTREC ‒ DNA methylation • DNA methylation-based age predictors ‒ Various tissues - Koch & Wagner. Aging (Albany NY) (2011) - Horvath. Genome biol. (2013) ‒ Blood - Hannum et al . Mol. Cell. (2013) - Zbiec-Piekarska et al . FSI Genet. (2015) ‒ Semen - Lee et al . FSI Genet. (2015) ‒ Saliva & Buccal swab - Bocklandt et al . PLoS One. (2011) - Eipel et al . Aging (Albany NY) (2016) 2
2018-10-10 Method • Saliva samples ‒ 280 samples (18-73 years) • HumanMethylation450 BeadChip Array ‒ 54 males (18-73 years) • Targeted Bisulfite Sequencing info Training Set Testing Set Total Male 47 70 117 Female 48 61 109 Total 95 131 226 ‒ Multiplex methylation SNaPshot (226 samples; Both sets) ‒ Massively parallel sequencing (95 samples; Training set) ‒ Multivariate linear regression analysis using SPSS HumanMethylation450 BeadChip Array • Details ‒ 54 males (18-73 years) ‒ GSE92767 ‒ 445,791 CpGs • Selection of marker candidates Criteria No. CpGs FDR_p < 0.05 74,807 R 2 value > 0.65 80 |β-score MAX – β-score min | ≥ 0.1 62 Advancing age 3
2018-10-10 Age-associated CpG candidates Stepwise linear regression analysis Additional candidates + Cell Type-specific Marker 0.7 • Buccal-Cell-Signature (ϐ) 0.6 ‒ Eipel et al. Aging (Albany NY). (2016) ‒ cg07380416 ( CD6 ) 0.5 Methylation at cg18384097 ‒ cg20837735 ( SERPINB5 ) ‒ Percentage of buccal epithelial cells 0.4 R² = 0.9286 99.8 × � ���������� + 1.92 ϐ = 0.3 2 −98.12 × � ���������� + 88.54 + 0.2 2 N = 54 0.1 • cg18384097 ( PTPN7 ) Spearman’s rho = 0.955 ‒ Souren et al. Genome Biol. (2013) 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 ‒ High in buccal epithelial cell Buccal-Cell-Signature (ϐ) ‒ Low in blood cell (predicted epithelial cell compositions) ‒ PTPN7 gene - Protein tyrosine phosphatase (PTP) - Preferentially expressed in hematopoietic cells 4
2018-10-10 Detail Workflow Hong et al . FSI Genet. (2017) A G DNA Methylation Analysis Multiplex PCR Multiplex SBE 10ng Multiplex Methylation SNaPshot (N=226=95+131) Bisulfite conversed Massively Parallel Sequencing (N=95) DNA Multiplex PCR Indexing PCR DNA Methylation Analysis Read Sequence Index Sequence Lee et al . FSI Genet. (2016) Methylation SNaPshot (N=226) Methylation SNaPshot (N=95) cg00481951 (SST) cg19671120 (CNGA3) cg14361627 (KLF14) cg00481951 (SST) cg19671120 (CNGA3) cg14361627 (KLF14) 0.4 0.4 0.4 0.4 0.4 0.4 R² = 0.6347 R² = 0.4791 R² = 0.2882 R² = 0.6313 R² = 0.5412 R² = 0.2541 0.3 0.3 0.3 0.3 0.3 0.3 Methylation Methylation Methylation Methylation Methylation Methylation 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0 0 0 0 0 0 0 20 40 60 80 0 20 40 60 80 0 0 20 20 40 40 60 60 80 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) cg07547549 (SLC12A5) cg08928145 (TSSK6) cg12757011 (TBR1) cg07547549 (SLC12A5) cg08928145 (TSSK6) cg12757011 (TBR1) 1 1 0.6 0.6 0.6 0.6 R² = 0.167 R² = 0.5486 R² = 0.4341 R² = 0.5928 R² = 0.4193 R² = 0.2139 0.5 0.5 0.5 0.5 0.8 0.8 Methylation Methylation Methylation Methylation Methylation Methylation 0.4 0.4 0.4 0.4 0.6 0.6 0.3 0.3 0.3 0.3 0.4 0.4 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0 0 0 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) Chronological Age (years) 5
2018-10-10 Model – Multiplex Methylation SNaPshot Training Set (N=95) Testing Set (N=131) Target ID Coefficient 80 80 -24.521 (intercept) cg18384097 -31.111 Predicted Age (years) 60 Predicted Age (years) 60 cg00481951 6.718 40 40 cg19671120 23.760 cg14361627 81.053 20 20 24.325 cg08928145 MAD = 3.03 MAD = 3.43 RMSE = 4.03 RMSE = 4.36 cg12757011 53.634 0 0 0 20 40 60 80 0 20 40 60 80 cg07547549 89.415 Chronological Age (years) Chronological Age (years) MAD: Mean Absolute Deviation RMSE: Root Mean Square Error MPS (N=95) – Read Count Read count per sample AVG depth per marker 1,000,000 cg07547549 500,000 0 cg12757011 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1,000,000 cg08928145 500,000 0 25 26 27 28 29 30 31 32 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 cg14361627 1,000,000 500,000 cg19671120 0 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 cg00481951 1,000,000 500,000 cg18384097 0 74 75 76 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 100 AVG 0 30000 60000 6
2018-10-10 MPS (N=95) – Methylation value cg00481951 (SST) cg19671120 (CNGA3) cg14361627 (KLF14) 0.2 0.2 0.2 R² = 0.661 R² = 0.3209 Methylation Methylation Methylation R² = 0.5738 0.1 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) cg08928145 (TSSK6) cg12757011 (TBR1) cg07547549 (SLC12A5) 1 0.5 0.5 R² = 0.4047 0.8 0.4 0.4 R² = 0.5808 Methylation Methylation Methylation R² = 0.2218 0.6 0.3 0.3 0.4 0.2 0.2 0.2 0.1 0.1 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Chronological Age (years) Chronological Age (years) Chronological Age (years) SNaPshot vs MPS (N=95) G (Methylated) A (Unmethylated) 7
2018-10-10 SNaPshot vs MPS (N=95) SNaPshot vs MPS (N=95) Model - SNaPshot 80 Target ID Coefficient (intercept) -24.521 60 cg18384097 -31.111 cg00481951 6.718 Predicted Age (years) 40 cg19671120 23.760 MPS cg14361627 81.053 SNaPshot 24.325 cg08928145 20 cg12757011 53.634 cg07547549 89.415 0 0 20 40 60 80 -20 Chronological Age (years) Model – MPS MPS model (N=95) Target ID Coefficient 80 (intercept) -7.282 cg18384097 -18.170 Predicted Age (years) 60 131.995 cg00481951 40 cg19671120 71.822 cg14361627 138.619 20 20.377 cg08928145 MAD = 3.67 RMSE = 4.80 cg12757011 -1.307 0 0 20 40 60 80 78.467 cg07547549 Chronological Age (years) 8
2018-10-10 Conclusion • A cell type-specific marker (cg18384097) and 6 age-associate markers (cg00481951, cg19671120, cg14361627, cg08928145, cg12757011, and cg07547549) enabled age prediction in saliva with high accuracy. • Markers can be applied to both Multiplex methylation SNaPshot and MPS. • The model should be altered as the platform differs. Acknowledgement Yonsei DNA Profiling Group This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation of Korea (NRF) funded by the Korean government ( NRF- 2014M3A9E1069992 ). 9
Recommend
More recommend
