A Molecular Approach to Estimating Time of Death

Stopwatch
I will admit that over the years, I have watched various crime scene investigation shows and read several books by Kathy Reichs and Patricia Cornwell because I was fascinated by forensic science. These same books and shows are a guilty pleasure because as a scientist, I know these portrayals do not accurately reflect how laboratory work is done. Answers are not so cut and dried as an exact estimation of time of death—for example, death was five hours before the body was found in an abandoned warehouse. However, scientists are always looking for ways to improve accuracy in time of death estimates, which are currently based on a few physical clues that are affected by environment and other factors. One approach taken by Sampaio-Silva et al. (1) was to assess the RNA degradation using reverse transcription quantitative PCR (RT-qPCR).

The authors of this PLOS ONE article wanted to determine if nucleic acid degradation could be used as a method to improve time of death estimates. Sampaio-Silva et al. started with post mortem factors they could eliminate: variability in temperature and contamination by microbes. Both of these parameters are known to confound the estimate for time of death. Earlier studies cited in the article showed variability in the rate of RNA degradation in different tissues, so when the authors euthanized mice, they took samples from eight tissues: heart, lung, spleen, femoral quadriceps, liver, stomach, pancreas and skin—organs that are relevant for forensic analysis in humans. The organs were stored in sterile conditions at 21°C, and RNA was extracted four and 24 hours post mortem and assessed for purity and integrity using spectrophotometry and electrophoresis, respectively. The tissues grouped into three categories according to the RNA Quality Indicator (RQI) determined by the electrophoresis assessment: Group 1 included heart, spleen and lung, which showed the greatest RNA stability even at 24 hours and had an RQI value of 6; group 2 included femoral quadriceps, liver and stomach, which had greater RNA degradation compared to the first group; and group 3 included pancreas and skin, which had an RQI value less than 4, reflecting the RNase-rich nature of these organs. All tissues showed time-dependent RNA degradation, but only organs in groups 1 and 2 were used for subsequent analysis due to concerns that RNA quality might affect RT-qPCR results.

Sampaio-Silva et al. looked at RNA degradation at regular intervals after death and selected heart, femoral quadriceps and liver tissues to analyze over 11 hours, with samples taken every hour. RNA from heart tissue was the most stable, with no degradation after the first four hours, while both femoral quadriceps and liver showed degradation from the first time point. However, RNA from all three tissues showed statistically significant correlation to the post mortem interval (PMI).

For the RT-qPCR analysis, 11 genes were used: evolutionarily conserved genes (Rps29 and Srp72), highly expressed liver mRNAs (Bhmt and Alb) and femoral quadriceps and cardiac muscle transcripts (Tpm1 and Mylk) as well as more ubiquitous transcripts (Actb, GAPDH, Hprt, Ppia and Cyp2E1). All reactions were normalized against Rps29 to control for variations in RNA extraction, reverse transcription and amplification. The Rps29 transcript had the least variation among the post mortem tissues [i.e., less than 1% coefficient of variation in cycle threshold (Ct) values, defined by the authors as the cycle number at which qPCR fluorescence rises above the background fluorescence]. Interestingly, the Ct values from the transcripts amplified from heart tissue were constant for samples taken once an hour for 11 hours, not much use for determining PMI. In contrast, mRNAs from both liver (Alb and Cyp2E1) and femoral quadriceps (Actb, Gapdh, Ppia and Srp72) showed statistically significant correlation to PMI.

Based on the results from the RT-qPCR analysis, the authors chose the four transcripts from the femoral quadriceps to develop a mathematical model for estimating PMI based on linear regression and a separate equation to estimate error by calculating the confidence interval for PMI. To confirm the utility of the regression equation for PMI, femoral quadriceps tissues from mice were sampled at 1, 4 and 10 hours post mortem. The results from the Ct values of the Actb, Gapdh, Ppia and Srp72 mRNAs were used in the PMI calculation, which estimated 1.90 ± 0.01, 4.10 ± 0.87 and 9.80 ± 1.87 hours for the three time intervals.

This PLOS ONE article is a first attempt at quantitating RNA transcripts from post mortem tissues to estimate time of death. This study used murine tissues stored aseptically under temperature-controlled conditions to eliminate for some of the variables that forensic analysts encounter when determining time of death for humans. I think this is an exciting first step in giving forensics another tool to determine PMI, but it is a long way from RNA analysis under controlled conditions in the lab using mice to useful tool for human forensic analysis.

Reference
1. Sampaio-Silva F., Magalhães T., Carvalho F., Dinis-Oliveira R.J., Silvestre R. and Kayser M. (2013). Profiling of RNA Degradation for Estimation of Post Mortem Interval, PLOS ONE, 8 (2) e56507. DOI:

The following two tabs change content below.
Avatar

Sara Klink

Technical Writer at Promega Corporation
Sara is a native Wisconsinite who grew up on a fifth-generation dairy farm and decided she wanted to be a scientist at age 12. She was educated at the University of Wisconsin—Parkside, where she earned a B.S. in Biology and a Master’s degree in Molecular Biology before earning her second Master’s degree in Oncology at the University of Wisconsin—Madison. She has worked for Promega Corporation for more than 15 years, first as a Technical Services Scientist, currently as a Technical Writer. Sara enjoys talking about her flock of entertaining chickens and tries not to be too ambitious when planning her spring garden.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.