Multi-stable (C/N/H/O) isotope signatures in humans to discern dietary habits, geographical origin and biological sex for forensic identification

Agrawal, Gunjan (2026) Multi-stable (C/N/H/O) isotope signatures in humans to discern dietary habits, geographical origin and biological sex for forensic identification. PhD thesis, Indian Institute of Science Education and Research Kolkata.

Text (PhD thesis of Gunjan Agrawal (20RS074)) 20RS074.pdf - Submitted Version Restricted to Repository staff only Download (8MB)

Abstract

Stable isotope analysis of human tissues examines the isotopic composition of bioelements such as carbon, nitrogen, hydrogen, and oxygen preserved within the tissue matrix. During tissue formation, proteins, carbohydrates and lipids incorporate these elements from diet, drinking water, and the surrounding environment. The resulting isotopic composition of the tissues preserves an integrated record of an individual’s diet, and geography over the period of tissue growth. As a result, stable isotope analysis has become an important tool in forensic science, archaeology, and nutrition research for reconstructing dietary habits, geographic origin, and migration history. Beyond this, stable isotope analysis has also emerged as an indicator of physiological and pathological conditions. Since metabolic processes regulate the assimilation and incorporation of dietary elements into body tissues, alterations in normal metabolism due to disease or physiology can modify tissue isotope ratios. Such isotopic changes have been found to be linked with a range of conditions such as liver cirrhosis, eating disorders, obesity, diabetes, starvation, and nutritional stress and phenotypic characteristics such as biological sex, age, and body mass index (BMI). For example, liver cirrhosis has been associated with decreased δ¹⁵N values, eating disorders with elevated δ¹⁵N values, and nutritional stress arising from starvation or undernourishment with increased δ¹⁵N and decreased δ¹³C values. Human scalp hair and fingernails are particularly well suited for stable isotope investigations because they are continuously growing, keratinized tissues that become metabolically inert after formation. During growth, they incorporate dietary and environmental elements into the keratin matrix, preserving the isotopic signatures present at the time of tissue synthesis. Their predictable growth rates provide a chronological record spanning weeks to months, while their metabolic stability, resistance to post-mortem degradation, and ease of non-invasive collection make them valuable substrates for forensic human identification, dietary reconstruction, and biomedical investigations. This thesis investigates the application of bulk and compound-specific stable isotope analysis of human scalp hair and fingernails to infer dietary habits, geographical origin and biological sex in the Indian population, addressing a significant gap in isotope reference data for South Asia. Bulk tissue carbon (δ¹³C) and nitrogen (δ¹⁵N) isotope analysis of hair and nail samples were performed using Elemental Analyser–Isotope Ratio Mass Spectrometry (EA-IRMS), whereas oxygen (δ18O) isotope ratios were measured using High Temperature Conversion EA-IRMS. Hydrogen (δ²H) and oxygen isotope analysis of water was carried out using GasBench-IRMS. Finally, compound-specific carbon isotope analysis of amino acids (CSIA-AA) was analysed using Gas Chromatography–Combustion–Isotope Ratio Mass Spectrometry (GC-C-IRMS). The δ¹³C values of human tissues revealed a diet predominantly based on C₃ plant-derived foods, whereas increasing consumption of animal protein resulted in corresponding enrichment of both δ¹⁵N and δ¹³C values. Both the amount and source of dietary protein influenced tissue isotope composition, with lacto-vegetarians exhibiting δ¹⁵N values approximately 0.9% lower than omnivores. Moreover, a dietary shift of only 5% produced measurable changes of approximately 0.4 – 0.5% in both δ¹⁵N and δ¹³C values. The study further established that the oxygen isotope ratios of scalp hair and fingernails is a reliable indicator of geographic residence. A strong relationship between the δ¹⁸O values of drinking water and hair was observed highlighting that approx. 45% of the oxygen in hair keratin was derived directly from locally consumed drinking water. In addition, water and hair isoscapes for capital city of India, Delhi was developed which enabled the identification of predominant drinking water sources across the city. The analysis suggested that groundwater is the principal source of drinking water for most residents despite efforts to reduce groundwater abstraction by the government. Compound-specific δ¹³C analysis of amino acids provided a higher level of biochemical resolution than bulk isotope analysis by identifying the amino acids most responsive to dietary habits, and biological sex. Branched-chain amino acids (leucine, valine, isoleucine) and the aromatic amino acids (phenylalanine, tyrosine) accounted for the greatest isotopic variation among individuals. The δ¹³C values of essential amino acids, particularly leucine and valine emerged as the most reliable dietary markers. The δ¹³C values of all essential amino acids enabled classification of lacto-vegetarians and omnivores with 74 - 79% accuracy in scalp hair and fingernails using a random forest model. Biological sex exerted a comparatively smaller influence on amino acid δ¹³C values. However, significant differences in tyrosine and phenylalanine in nail tissue, enabled biological sex classification with an accuracy of approximately 66%. These findings establish baseline stable isotope signatures for healthy Indian adults and demonstrate the value of bulk and compound-specific (amino acid) stable isotope analysis of human tissue in distinguishing dietary habits (lacto-vegetarians vs omnivores), tracing geographic origin and urban water use, and improving forensic human identification (diet and biological sex). The resulting baseline dataset also provides an important reference against which isotopic alterations associated with metabolic disorders, including diabetes, liver cirrhosis, eating disorders, and malnutrition, can be evaluated in future clinical and forensic applications.

Actions (login required)

View Item