Research paperHuman age estimation from blood using mRNA, DNA methylation, DNA rearrangement, and telomere length
Introduction
The age of individual is highly relevant when it comes to personal identification and also matters in legal affairs such as for the applicable law being that for adults or not. Establishing the age of unknown persons can provide important leads in police investigations, disaster victim identification, and identity fraud cases, amongst others, to trace and finally identify the person. Age estimation of known persons with unclear age can be relevant in law enforcement and other legal affairs. However, age estimation for forensic applications is commonly performed with techniques that rely on morphological analyses of teeth or skeletal features in living or dead individuals [1], [2], and thus require the availability of such materials for forensic testing. Molecular methods of forensic age estimation allow for the use of human materials that do not contain morphological age features. Previously proposed methods based on chemical and physical principles, such as aspartic acid racemization and radiocarbon analysis [3], respectively, provide highly accurate age estimation. However, they are limited by the type of human biological samples suitable for analysis (most often tooth dentin or enamel) and require specialized equipment. To expand the forensic use of individual age estimation it is desirable to use molecular methods that allow retrieving a person’s age information from various human biological samples including typical crime scene traces such as bloodstains.
Until recently, only few DNA-based molecular methods were proposed, including the quantification of particular mitochondrial (mt) DNA mutations that accumulate with increasing age [4], and the estimation of telomere repeat length that decreases with increasing age [5]. While the use of mtDNA mutations is considered of limited value for practical forensic applications because of technical issues and limited prediction accuracy [6], the suitability of telomere length as forensic age marker is discussed controversially [7], [8]. In 2010, we introduced a simple and reliable DNA test for age estimation from blood that is based on the quantification of particular T-cell specific DNA rearrangements (sjTREC) yielding a high and statistically significant correlation coefficient R2 of 0.835 (SE ± 8.9) [9], and somewhat lower correlations were obtained in parallel studies [10], [11]. The sjTREC-based DNA test we developed controlled for the amount of genomic DNA in blood, delivered high prediction accuracies expressed as area under curve (AUC) of around and above 0.9 when predicting categorical age i.e., age groups separated by 20 years [9].
Over the last few years, several studies have indicated that changes in DNA methylation patterns are correlated with age [12], [13], [14]. Studies of whole genome DNA methylation using microarray profiling confirmed that changes in DNA methylation status are deeply involved in the human ageing processes [14], [15], [16]. Several age-correlated DNA methylation markers were suggested for blood [17], [18] as well as for some other tissues [19], [20]. In addition to general epigenetic studies on ageing, applied forensically oriented studies on particular DNA methylation markers for forensic age estimation from blood have started to appear [21], [22], [23], [24], [25], [26]. However, many of these studies used rather small sample size [25], [27], [28], which is especially critical when proposing new markers without additional confirmation in independent samples. Consequently, the overlap in proposed age-predictive DNA methylation markers between studies is low, with the exception of CpGs from the ELOVL2 gene identified and confirmed by several genome-wide epigenetic ageing studies [18], [29], [30], [31] and verified in multiple forensic age estimation studies [21], [24], [26], currently reflecting the most promising biomarkers for age estimation from blood.
Although gene expression is long known to correlate with age [32], [33], mRNA markers are practically not explored for forensic age estimation as of yet, except for fetal hemoglobin proposed for identification of newborns [34].
In the present study, we investigated different types of human molecular biomarkers, including mRNA, DNA methylation, sjTREC, and telomere length, for their suitability to estimate age from blood. To identify new blood biomarkers of age, we first performed genome-wide gene expression and DNA methylation microarray analyses in peripheral blood of the same young and old healthy volunteers. The most age-informative candidate mRNA and methylated DNA markers obtained from the microarray screenings were then technically validated via different analysis methods and biologically validated in blood samples from large independent sets of individuals of wide age ranges. The predictive power of the validated biomarkers for age estimation was evaluated together with previously reported DNA methylation, telomere length, and sjTREC biomarkers of age. Statistical models for age estimation were built and tested and a final model was proposed for estimating human age from blood.
Section snippets
Sample collection, ethical statement and nucleic acid isolation
Peripheral venous blood was collected in PaxGene Blood DNA and RNA tubes (PreAnalytix GmbH, HiIden, Germany) in one setting for the subsequent isolation of the respective nucleic acids. The samples from a total of 350 healthy male members of the population based Erasmus Rucphen Family (ERF) study [35], ranging from 22 to 84 years of age were obtained. Additionally, blood samples from 40 younger donors (4–22 years) were obtained from healthy volunteers. Due to sample availability constrains not
Genome-wide search for mRNA age markers via gene expression microarray analysis
Aiming to investigate age-dependent changes in gene expression patterns and to reveal new age-informative candidate mRNA markers, we carried-out genome-wide expression analysis using Affymetrix microarray technology in 47 healthy male human samples of young (23–28 years) and old (66–72 years) age (Suppl. Tab 1). Because previous studies showed that conditions and duration of the storage of collected peripheral blood samples are crucial for obtaining unbiased gene expression profiles [43], in
Discussion
Although the previous application of genome-wide gene expression and DNA methylation profiling revealed various new insights into the molecular biology of ageing [47], [48], [49], [50], an area not having received much attention in the field of molecular ageing research is the systematic identification of the most age-correlated mRNA and methylated CpG sites as biomarkers for molecular age estimation. Identifying minimal numbers of biomarkers that provide maximal age information is especially
Conflict of interest
None.
Acknowledgements
The authors are grateful to the study participants, the staff from the ERF Study and the Rotterdam Study, and the participating general practitioners and pharmacists. We are particularly grateful to Petra Veraart for the help in collecting ERF Study blood samples for this study purpose. This study was financially supported in parts by the Netherlands Forensic Institute (NFI), the Erasmus MC University Medical Center Rotterdam, and the Netherlands Genomics Initiative (NGI)/Netherlands
References (57)
- et al.
Age estimation
Forensic Sci. Int.
(2007) - et al.
Age estimation in forensic sciences: application of combined aspartic acid racemization and radiocarbon analysis
Mol. Cell. Prot.
(2010) - et al.
Molecular pathology and age estimation
Forensic Sci. Int.
(2010) - et al.
Exploring the utility of genetic markers for predicting biological age
Leg. Med. (Tokyo)
(2012) - et al.
Age estimation via quantification of signal-joint T cell receptor excision circles in Koreans
Leg. Med. (Tokyo)
(2014) - et al.
Epigenetics and aging: the targets and the marks
Trends Genet.
(2007) Impact of aging on DNA methylation
Ageing Res. Rev.
(2003)- et al.
Age-related DNA methylation changes for forensic age-prediction
Int. J. Legal Med.
(2014) - et al.
Developing a DNA methylation assay for human age prediction in blood and bloodstain
Forensic Sci Int Genet.
(2015) - et al.
Development of a forensically useful age prediction method based on DNA methylation analysis
Forensic Sci. Int. Genet.
(2015)