Semester

Spring

Date of Graduation

2019

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Biology

Committee Chair

Glen P. Jackson

Committee Co-Chair

Clifton P. Bishop

Committee Member

Clifton P. Bishop

Committee Member

Jennifer E. Gallagher

Committee Member

Shikha Sharma

Committee Member

Stephen J. Valentine

Abstract

In this work we demonstrate the power of compound-specific isotope analysis (CSIA) to analyze proteinaceous biological materials in three distinct forensic applications, including: 1) linking necrophagous blow flies in different life stages to their primary carrion diet; 2) identifying the harvesting area of oysters for food authentication purposes; and 3) the ability to predict biometric traits about humans from their hair.

In the first application, we measured the amino-acid-level fractionation that occurs at each major life stage of Calliphora vicina (Robineau-Desvoidy) (Diptera: Calliphoridae) blow flies. Adult blow flies oviposited on raw pork muscle, beef muscle, or chicken liver. Larvae, pupae and adult blow flies from each carrion were selected for amino acid CSIA. Canonical discriminant analysis showed that flies were correctly classified to specific carrion types in 100% (original rules) and 96.8% (leave-one-out cross-validation [LOOCV]) of cases. Regarding life stages, we obtained 100% and 71% of correct classification in original rules and LOOCV, respectively. Most of the essential amino acids did not significantly change between life stages (at 95% CI). However, some non-essential (Ala, Ser, and Glu) and conditionally essential amino acids (Gly and Pro) were isotopically depleted in the adult stage. Except for the essential amino acids, the amino acids in larvae and pupae were enriched in 13C and adult blow flies were depleted in 13C relative to the carrion on which they fed. These results make it possible to exclude potential sources of carrion as larval food. In addition, amino-acid-specific IRMS could help inform entomologists whether a fly has just arrived from another location to feed on a corpse or has emerged from a pupa whose feedstock was the corpse.

Regarding the source inference of oysters, we investigated the bulk, amino-acid compound-specific stable isotopes, cadmium and lead concentrations of the popular Eastern oyster, Crassostrea virginica. This species has been one of the most popular species for the oyster harvesting business in the United States, despite its claimed reduced availability due to excessive harvesting and some parasitic diseases. The results from specimens collected from different Gulf of Mexico bays were subjected to multivariate statistical analysis to assess whether we could predict the oysters’ harvest area. Our results indicate that the combination of trace elements and isotope ratios can predict geographic provenance of oysters with greater than 70% correct classification using LOOCV, which is superior to using only CSIA or only trace elements. The δ13C values of serine and glycine could also discriminate between two adjacent harvest areas within the same Apalachicola bay. One of these areas is fishable in the winter season and the other is fishable in the summer season, so the ability to differentiate oysters from these two areas is a valuable capability for the Florida Department of Agriculture, which is responsible for enforcement. The use of chemical signatures to identify harvest areas is a valuable tool to protect consumers from food fraud, food-borne diseases and to help regulatory agencies enforce harvesting regulations.

Finally, we describe the use of amino-acid CSIA and amino acid quantitation of scalp hair of American individuals to predict soft biometrics in humans. We measured the isotope ratios and respective quantities of 13 amino acid peaks. Correlation analysis of the multivariate data provided the degree of correlation between essential and non-essential amino acids with factors such sex and age of the hair donors. The isotope ratios of each amino acid were first corrected for the extent of C4-based carbon in the diet to reveal relationships between metabolic or phenotypic factors and the isotope ratios of 13C in the amino acids in the hair shafts. Multivariate analysis revealed that the sex of a donor could be correctly predicted with cross-validated accuracies of 80% and 89% using the isotope ratios or quantities of amino acids, respectively. The continuous dependent variables of donor age and body mass index (BMI) were also predicted using the amino acid isotope ratios or quantities, but the predictions were not as reliable as for sex determination. Unexpectedly, the δ13C values of hair reflected the frequency of alcohol consumption in two groups of subjects.

Embargo Reason

Publication Pending

Share

COinS