Analysis of genome-wide 5-hydroxymethylation of blood samples stored in different anticoagulants: opportunities for the expansion of clinical resources for epigenetic research.
Author(s): Gao L, Zhang Z, Cui XL, West-Szymanski D, Ye C, He C, Zhang W, Bissonnette M
Publication: Epigenetics, 2023, Vol. 18, Page 2271692
PubMed ID: 37898998 PubMed Review Paper? No
Purpose of Paper
This paper compared cfDNA yields and 5-hydroxymethylcytosine Selective Chemical Labelling (5hmC-Seal) sequencing metrics results and profiles in case-matched EDTA and heparin plasma from patients diagnosed with colorectal cancer within the past 1 y (cases) and individuals that were free of cancer in the 15 years following blood collection (controls). The effect of patient age, sex and diagnosis on the 5hmC-Seal profile was also investigated.
Conclusion of Paper
cfDNA yields were higher from heparin plasma than EDTA plasma (P<0.01), but only when results were stratified by diagnosis; higher cfDNA yields were observed in plasma from patients diagnosed with CRC within 1 year (P<0.01). EDTA and heparin plasma libraries had comparable percentages of fragments between 200-1000 bp and electropherograms, but heparin plasma displayed a trend toward higher duplication rates compared to EDTA plasma. In principal component analysis (PCA), specimens did not cluster based on anticoagulant, diagnosis, patient sex or patient age (<80 versus ≥80 years). The number of covered genes with >10, >20 and >30 unique reads that were mapped to gene bodies, promoters and the histone modifications marking enhancers (H3K4me1 and H3K27ac) were all comparable between EDTA and heparin plasma; profiles were more strongly correlated between case-matched specimens collected in tubes with different anticoagulants than non-matched specimens that used the same anticoagulant (0.988 versus 0.981). Importantly, most differences observed between colorectal cases and healthy controls were common among the two anticoagulants. The authors conclude that EDTA plasma is suitable for 5hmC-Seal analysis.
Studies
-
Study Purpose
This study compared cfDNA yields and 5hmC-Seal sequencing metrics results and profiles in case-matched EDTA and heparin plasma from patients diagnosed with colorectal cancer within 1 y (cases) and individuals that were free of cancer in the 15 years following blood collection (controls). The effect of patient age, sex and diagnosis on 5hmC-Seal profile was also investigated. This study used case-matched EDTA and heparin plasma specimens from 30 individuals that were diagnosed with colorectal cancer within 1 year of blood sampling and 30 age-, gender- and race-matched controls who did not develop cancer in the 15 years following blood collection. All specimens were collected between 1998 and 2004 as part of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial and were stored for 18-24 y. Further details of plasma separation and storage were not provided in this publication. cfDNA was isolated from plasma using the QIAamp Circulating Nucleic Acid Kit and quantified using the Qubit dsDNA High Sensitivity Assay. To construct 5-hydroxymethylcytosine sequencing libraries, DNA underwent end-repair and A-tailing with the KAPA Hyper Prep Kit, KAPA Unique Dual-Indexed Adapter ligation, 5hmC modification labelling with UDP-azide-glucose and T4 β-glucosyltransferase, reaction with DBCO-PEG4-biotin, and enrichment using magnetic beads (Dynabeads M-270) prior to library construction with the KAPA Hyper Prep Kit. Libraries were pair-end sequenced using a NovaSeq6000. 5hmC read counts were compared between cases and controls using DESeq2 separately for each anticoagulant and then lists of differentially expressed genes were compared between the two anticoagulants.
Summary of Findings:
cfDNA yields were higher from heparin plasma than EDTA plasma (P<0.01), but only when results were stratified by diagnosis; higher cfDNA yields were observed in plasma from patients diagnosed with CRC within 1 year of blood collection (P<0.01). The percentage of fragments between 200-1000 bp were comparable in sequencing libraries prepared from cfDNA from EDTA and heparin plasma. Additionally, the electropherograms of 5hmC-Seal libraries from EDTA and heparin plasma were comparable. The duplication rates were higher in libraries from heparin plasma than EDTA plasma, but the difference was not significant in the complete data set or when stratified by diagnosis (cases/controls). In principal component analysis (PCA), specimens did not cluster based on anticoagulant, diagnosis, patient sex or patient age (<80 versus ≥80 years). The number of covered genes with >10, >20 and >30 unique reads that were mapped to gene bodies, promoters and the histone modifications marking enhancers (H3K4me1 and H3K27ac) were all comparable between EDTA and heparin plasma. Importantly, 94.4% of gene bodies with >50 reads were observed in cfDNA from both EDTA and heparin specimens (95.2% for CRC cases and 94.9% for controls). The gene bodies were more strongly correlated between matched specimens collected and stored in different anticoagulants than non-matched specimens in the same anticoagulant (0.988 versus 0.981). Importantly, most differences observed between colorectal cases and healthy controls were common among the two anticoagulants. Modeling found more than 10-fold enrichment in the number of differentially expressed genes shared than the null distribution.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Neoplastic - Carcinoma
- Not specified
- Normal
Platform:
Analyte Technology Platform DNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Preaquisition Patient age <80 years
≥ 80 years
Biospecimen Acquisition Anticoagulant EDTA
Heparin
Preaquisition Diagnosis/ patient condition Diagnosed with colorectal carcinoma within 1 y
Not diagnosed with cancer within 15 y
Preaquisition Patient gender Female
Male