Epigenome-wide cross-tissue correlation of human bone and blood DNA methylation – can blood be used as a surrogate for bone?

ABSTRACT

Difficulty in obtaining bone tissue is an obstacle to studying epigenetics to understand gene–environment interactions, and their role in disease pathogenesis. Blood is an obvious alternative and in this proof of principle study, our aim was to systematically investigate whether blood is a viable surrogate for bone. We measured epigenome-wide DNA methylation at 850 K CpG sites in matched trabecular bone and peripheral blood collected from the same patients at the same time-point (n = 12 women; 66–85y), to investigate the between-tissue correspondence. What constituted a CpG site with corresponding methylation in both tissues was stringently defined. Only sites highly correlated (r² > 0.74; FDR q-value <0.05) and at least 80% similarity in methylation level (Δβ <0.2) between paired samples were retained. In total, 28,549 CpG sites were similarly methylated in bone and blood. Between 33% and 49% of loci associated with bone phenotypes through GWAS were represented among these sites, and major pathways relevant to bone regulation were enriched. The results from this study indicate that blood can mirror the bone methylome and capture sites related to bone regulation. This study shows that in principal, peripheral blood is a feasible surrogate for bone tissue in DNA methylation investigations. As the first step, this will provide a platform for future studies in bone epigenetics, and possibly for larger-scale epidemiological studies.

KEYWORDS:

• DNA methylation
• bone
• blood
• epigenetics
• surrogate tissue

Acknowledgments

Support by NBIS (National Bioinformatics Infrastructure Sweden) is gratefully acknowledged. We would also like to thank Prof. Andrew E. Teschendorff and Dr Petr Volkov for valuable comments and constructive discussions, Prof. Shafaat A. Rabbani for sharing additional data from their publication, and finally Lisa Jansson and Karin Önnby for sample collection and technical support.

All the authors have contributed to the study design, study conduct, and data interpretation. PE has performed data collection and data analysis. PE, KÅ, FM have drafted the manuscript, and all the authors have revised the manuscript content and approve the final version. PE takes responsibility for the integrity of the data analysis.

Disclosure statement

The authors report no conflict of interest relevant to this manuscript.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Additional information

Funding

This work was supported by the Swedish Research Council under grant K2015-52X-14691-13-4; Greta and Johan Kock Foundation; Royal Physiographic Society of Lund; Herman Järnhardt foundation; Thelma Zoegas Foundation; Skåne University Hospital Research Fund; and the Research and Development Council of Region Skåne, Sweden.