Introduction
A molecular autopsy is becoming an essential part of forensic practice in the past years [Castiglione]. A special challenge in forensic medical research are pathologies that are not encountered very often, thus limit the availability of research samples. In order to form an eligible sample group, researchers rely on aged and/or degraded samples, obtained and stored for different use and not always adequate for new emerging techniques. Nowadays, pathology of high interest for a forensic medical researcher is sudden cardiac death (SCD), especially in case of a negative autopsy (the absence of any structural change and negative histological and toxicological tests). Since SCD and negative autopsy pose a strong indicator of the presence of an inherited pathology of the heart, genetic testing is suggested [Torkamani]
Due to long storage time and degradation, the quality of biological samples is reduced [Madisen]. DNA analysis methods typically used in forensic medicine are based on short DNA fragments that are easily amplified even from biological samples with a high degree of degradation [Bukyya]. Unlike typical forensic analysis, massively parallel sequencing requires biological samples of higher quantity and quality.
Massively parallel sequencing (MPS) is a high-throughput method for DNA sequencing that enables the simultaneous analysis of millions of sequences [Nagy]. The quality of the starting sample is critical for the success of MPS, as poor-quality samples can lead to inaccurate and unreliable results. MPS platforms typically require high molecular weight DNA with minimal degradation and chemical modifications, and high purity and concentration. If the starting sample is of low quality, the sequencing library may contain biased or low-quality sequences, which can lead to errors, false positives, and reduced accuracy [Ug MAN0013432]. The quality of the starting sample can be assessed using various methods, such as gel electrophoresis, spectrophotometry, or fluorometry. Gel electrophoresis can be used to visualize the DNA fragmentation and assess the integrity, while spectrophotometry and fluorometry can be used to determine the concentration of the DNA, with the spectrophotometry additionally being able to detect chemical impurities [Nielsen]. Various pre-sequencing sample preparation protocols can be used depending on the type and quality of the starting sample, since this can affect the success of the consecutive steps such as DNA fragmentation, size selection, and adapter ligation. Overall, the quality of the starting sample is critical for the success of NGS, and careful sample preparation and quality control measures are essential to ensure accurate and reliable results.
The aim of the study is to determine the correlation between the storage time of post-mortem blood samples and extracted DNA quantity and quality, and to evaluate DNA sample eligibility for successive MPS analyses. Also, the aim is to establish proper quality control in sample preparation workflow for Massively Parallel Sequencing using Ion Torrent GeneStudio™ S5 platform.
Materials and methods
We analysed 48 post-mortem blood samples (peripheral whole blood without anticoagulant) with different storage time, short term (1-3 months) at 4 °C, followed by long term storage (3-122 months) at -20 °C. The analysed samples were taken from forensic-medical autopsies performed in a 10-year period (2013-2022) at the Institute for Forensic Medicine, Criminology and Medical Deontology at the Faculty of Medicine, St. Cyril and Methodius University in Skopje. All samples were obtained according to the standard protocols for forensic autopsy, biological sample processing and storage prescribed by the European Council of Forensic Medicine and R99 recommendations [Harmonization].
DNA extraction from peripheral whole blood
Blood used for DNA extraction was taken directly from a blood vessel of the deceased during autopsy and collected in sterile blood collection tubes without anticoagulant.
Two genomic DNA extraction protocols were applied. Eleven samples were extracted by using QIAamp DNA Micro Kit (QIAGEN, Hilden, Germany) according to manufacturer’s instructions, except for the starting volume of blood, which was increased from 100 µl to 200 µl, along with proportionally increasing the volume of lysis buffers. DNA was eluted in 100 µl of TE buffer. The rest of the samples (37) were extracted by using QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) according to manufacturer’s instructions, except for the elution volume, which was reduced from 200 µl to 100 µl.
DNA quantification
After DNA extraction, quantitation was initially done by using Qubit™ dsDNA HS Assay Kit on the Qubit 3.0 fluorimeter. Ten microliters of the DNA eluate were used for quantification.
To evaluate the amount of amplifiable DNA extracted from different post-mortem blood samples by Real-Time PCR, we used the Quantifiler™ Trio DNA Quantification Kit (Thermofisher Scientific) on the 7500 Real-Time PCR System, according to manufacturer’s instructions. The kit also gives indication of the degradation of the DNA, by comparing the amplification success of one small and one large target. The quantity of the DNA samples was estimated by using the quantity of the small autosomal target.
Results and Discussion
The results of the performed quantity measurements of DNA are shown in Table 1.
DNA was extracted using two extraction kits (QIAamp DNA Micro Kit & QIAamp DNA Mini Kit) and quantity was measured by fluorometric and Real-Time PCR approach. The results indicate high variability between the obtained DNA quantities from post-mortem blood samples with different storage time. This means that the storage time is probably not the only factor affecting the DNA quantity, and that other circumstances can influence the overall yield. Although some samples had low DNA yield, it was shown that silica-based DNA extraction method, such as QIAamp DNA Micro/Mini Kit, is adequate even for aged samples. Previous studies have also shown the applicability of silica-based DNA extraction methods for samples that have been stored for several years [Tagliaferro].
The maximum DNA concentration in our research was 160,902 ng/μl, obtained from sample kept for 6 months, extracted with the Mini Kit, as for the lowest concentration with the same kit was 0,018 ng/μl, from sample kept for 50 months. The highest concentration obtained with Micro Kit is 41,744 ng/μl from sample kept for 19 months, and the lowest with the Micro kit was 0,483 ng/μl from sample kept for 106 months.
Many modern applications in molecular biology require pure and intact genomic DNA as a first choice. Therefore, the efficiency of a DNA extraction method will be affected by its robustness and ability to yield optimal amount of clean and intact genomic DNA. The total count of WBC in a normal blood in adult human is ranged from 4.5-10 × 103/μL, thus the total amount of DNA per μL of blood has been calculated to be within the range of 29.48-65.5 ng approximately [Guha]. In our research, only 14.5 % of samples fit in this range, all of them being stored for less than 24 months. The specimens with longer storage fell far behind the quantity and quality of fresh blood samples.
Figure 1. Pearson correlation coefficient Qubit concentration vs. storage time.
When comparing the DNA concentration of the blood samples kept for 1 year, the highest noted concentration was 160,902 ng/μl (as measured by qPCR) and 124 ng/μl (as measured by Qubit). The highest noted concentration from the samples kept for 10 years is 2,747 ng/μl (as measured by qPCR) and 3,12 ng/μl (as measured by Qubit).
Statistical analyses were performed to check for correlation between storage time and quantity of extracted DNA (Figures 1-3). By applying the Pearson correlation coefficient, it was determined that there is a significant negative relationship between the duration of storage of the blood samples and the obtained concentration of DNA measured with the Qubit fluorometer, i.e., the longer the storage time of the material, the lower the concentration of isolated DNA in a blood sample. (r2= -0.346, p=0.017).
Figure 2. Pearson correlation coefficient between the small autosomal target DNA concentration vs. storage time.
By applying the Pearson correlation coefficient, it was determined that between the duration of storage of the blood samples and the concentration of the small autosomal target obtained by Real-Time PCR, there is a significantly negative dependence, that is, the longer the time of storage of the sample, the lower the concentration of the small DNA fragment in a blood sample (r2= -0.536, p < 0.001).
By applying the Pearson correlation coefficient, it was determined that between the duration of storage of the material and the concentration of the long fragment obtained by PCR, there is a significantly negative dependence, i.e. the longer the time of storage of the material, the concentration of the long DNA fragment in a blood sample is lower (r2= -0.550, p < 0.001)
By applying the Pearson correlation coefficient, it was determined that the measured concentration between the Qubit and Quantifiler Trio DNA Quantification Kit, there is a positive correlation, which means there is a tendency for high Qubit measured concentration to go with high Quantifiler Trio DNA Quantification Kit (and vice versa). (r2= 0.7063, p < 0.0001)
Figure 3. Pearson correlation coefficient between the large autosomal target DNA concentration vs. storage time.
Comparing the results obtained from different quantification methods, some deviation has been noted. Namely,
in some samples, fluorometric measurement overestimated or underestimated the quantity of the extracted DNA.
Figure 4. Pearson correlation coefficient Qubit concentration vs. Quantifiler Trio DNA Quantification Kit.
The high variability between the obtained DNA quantities, were incentive to perform additional estimation of the DNA quantity and quality by Real-time PCR, even though most of the library preparation workflows used, such as Precision ID DL8 Kit (Applied Biosystems™) in Ion Torrent S5 sequencing, do not require precise DNA quantification by Real-Time PCR [Ug MAN0013432].
The Quantifiler Trio DNA Quantification Kit which was used in this study, measures total DNA present and inhibitors in a DNA sample by measuring the quantity of one short DNA fragment, one long DNA fragment and one internal positive control. The kit can be used to establish a degradation index of the DNA (useful in routine forensic DNA typing) [Ug 4485354].
The deviation between the values determined by Qubit can be attributed to different factors, such as the variable degradation levels of the samples and the possible inhibition in the samples.
The Qubit HS dsDNA assay is regarded as a highly specific and sensitive DNA quantification assay and is preferred when quantifying DNA for downstream NGS analysis. However, when dealing with sample sets with different DNA integrity and possible presence of PCR inhibitors, overestimation or underestimation of the DNA concentration can arise. Library preparation workflow requires equal starting DNA input in order to achieve equal amplification of all gene targets, and thus to provide complete gene coverage for all of the samples. Also, when the measured DNA concentration is overestimated / underestimated, the PCR amplification of the overloaded samples will be more efficient than the rest of the samples, thus the final library and the sequencing chip will be oversaturated with amplicons from the overloaded samples, which in turn can result with poor coverage of the rest of the samples. This is particularly difficult in MPS platforms that utilize automated library preparation (such as Thermofischer’s Ion S5 Sequencer coupled with Chef Instrument for library preparation, template preparation and chip loading), which do not allow for individual library quantification, but provide an 8-samples library pool which can be then diluted to the optimal concentration [Ug 4485354]. According to our experience, this issue can be avoided only with the use of real-time PCR quantification step before library preparation and diluting samples to equal DNA quantity to ensure equal input in library preparation reactions.
Conclusions
The Qiagen Mini Kit & Micro Kit are applicable for DNA extraction from aged blood samples with various storage periods and degradation levels.
Qubit measurement is recommended for assessment of DNA concentration by library preparation kit’s manuals, however our study showed that total DNA measured by Qubit should be taken with caution, and when possible, real-time PCR quantification should be performed in order to establish the total amplifiable DNA in the sample.
Older samples sometimes tend to have higher degradation levels, thus, the DNA input and cycle numbers in further steps in library preparation workflow should be adjusted carefully to achieve optimal library concentration for subsequent sequencing.