Author ORCID Identifier
https://orcid.org/0000-0003-4873-5973
Defense Date
2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Integrative Life Sciences
First Advisor
Sarah Seashols-Williams
Second Advisor
Tal Simmons
Third Advisor
Joseph Reiner
Fourth Advisor
Laura Gaydosh Combs
Fifth Advisor
Kimberly Sturk-Andreaggi
Abstract
Skeletal remains frequently serve as the sole source of genetic material available for developing a profile in cases concerning unidentified human remains or mass casualties. Nevertheless, as unidentified skeletal remains age and are exposed to uncontrolled environmental conditions, the success rates of conventional methods of bone analysis diminish significantly. The physical and mechanical processing of skeletal remains into fine powder for DNA isolation and downstream analysis can substantially impact the quality and quantity of recoverable genetic information. This concern may be exacerbated when dealing with degraded forensic skeletal remains. Few studies have explored whether pulverization of bone samples contributes to DNA damage observed in profiles obtained from skeletal remains for forensic human identification.
This work developed and optimized a prolonged demineralization and slicing method for small cortical bone fragments of bovine bone. Given its dimensions, collagen composition, and comparatively reduced ethical restrictions, bovine was used as an experimental model for normal human bone. Bovine samples sourced for this work were collected from aged and weathered skeletonized remains found on rural properties, and the samples were exposed to uncontrolled environmental conditions for over five years before collection. After sample preparation, complete demineralization of small cortical bone fragments was performed, and samples were sliced manually with a razor blade before isolation with silica-based DNA extraction chemistry. For comparison, subsets of pulverized bovine samples were extracted using organic and silica-based chemistries. Bovine mitochondrial DNA quantification data and short tandem repeat (STR) profile quality were assessed for powdered bone samples and demineralized slices. Bovine profiles generated from demineralized slices yielded a higher average percentage of alleles detected (p < 0.05) and had greater retention of larger STR loci than pulverized samples. In conclusion, using demineralized slices resulted in more complete, balanced STR profiles, higher peak heights, and less degradation than powdered samples.
In addition, a portion of this work evaluated chemical excision and physical cell capture methods using demineralized, collagenase-digested bovine cortical bone slices as a powder-free alternative for cell isolation. The parameters for collagenase digestion were determined, and manual cell capture efforts were conducted to isolate endogenous cells and reduce microbial DNA within samples. Ultimately, it was evident that the combined approach of collagenase treatment of demineralized slices did not improve the total yields of recovered DNA or STR profile development from aged and weathered bovine cortical bone samples. Similarly, employing micromanipulation methods on the partially digested demineralized bone slices did not increase endogenous DNA or reduce exogenous DNA yields in the samples.
The efficacy of the developed demineralized slices method was assessed using human cortical skeletal samples in a comparative analysis with pulverized samples from the same donors. This study investigated the concordance of genetic markers through sequencing technologies to enhance the recovery of degraded DNA for human identification. Endogenous and exogenous DNA yields were evaluated through quantitative PCR (qPCR), and STR genotyping was conducted to assess traditional pulverization versus demineralized slices in 19th-century human skeletal remains. The average human DNA yields for the silica-extracted pulverized samples and demineralized slices were 0.032 ng and 0.692 ng, respectively. Demineralized slices ultimately recovered more amplifiable DNA than traditional homogenization methods for DNA analysis of skeletal remains (p < 0.05). The presence of amplification inhibition was assessed using the internal positive control (IPC) cycle threshold (Ct) values, and 59% of pulverized samples and 71% of demineralized slices were within the validated and acceptable range (27-31). Notably, due to failed large autosomal amplification for the pulverized samples, degradation indices (DI) could not be calculated. DI values for demineralized slices ranged from 3.855 to 247.081. Pulverized samples for organic and silica-based extraction chemistries yielded no STR profiles. In contrast, demineralized slices of the same donors yielded partial profiles for two of the seven unidentified individuals, ranging from 6 to 34 alleles across replicates. These data showcase the effectiveness of a modified processing method for forensic identification through successful STR profile generation from skeletal remains aged over a century.
In instances where STR analysis of degraded skeletal samples is unachievable, targeted sequencing is an effective alternative for forensic identification. Massively parallel sequencing (MPS) yields substantial quantities of pertinent genetic information from short DNA fragments, rendering it particularly advantageous for difficult samples. When MPS is coupled with targeted hybridization capture of single nucleotide polymorphisms (SNPs) and enrichment, the recovery of genetic information from highly degraded samples is improved. DNA extracts of demineralized slices and pulverized samples from 19th-century skeletal remains underwent DNA repair, library preparation, and hybridization capture with the FORensic Capture Enrichment (FORCE) panel at the Armed Forces Medical Examiner System’s Armed Forces DNA Identification Laboratory (AFMES-AFDIL) in Dover, DE. This study investigated the recovery of SNPs within DNA extracts of demineralized slices and pulverized samples using targeted sequencing to evaluate process-induced DNA damage within problematic skeletal samples. When assessing the sequencing metrics, the total number of reads covering target SNPs was higher in demineralized slices (p < 0.05), and the mean read length and mean insert size were greater than those in pulverized samples (p < 0.05). Demineralized slices outperformed those prepared with traditional pulverization methods (p < 0.05) at 1X coverage levels (82.22% and 37.36%, respectively).
The results of this dissertation demonstrate significant and compelling differences between the treatment groups for the model moderately aged bovine and 19th-century human skeletal samples. The findings of this work further exemplify the issue of method-induced DNA degradation within samples processed by pulverization. Homogenizing skeletal elements into a fine powder dramatically impacts the recoverable template DNA, as shown in both length-based and targeted sequencing results from historical skeletal remains. Extended demineralization followed by slicing presents an alternative option for skeletal samples at the start of the forensic DNA analysis workflow.
Rights
© Ciara N. Rhodes
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
12-12-2024