PhD Oral Exam - Keshava Praveena Neriya Hegade, Mechanical Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Cancer continues to be a predominant cause of global morbidity and mortality, with projections indicating an increase in incidence and prevalence due to factors such as population aging, growth, and lifestyle-related risks. Patient outcomes are significantly influenced by the stage at which diagnosis occurs, with late-stage detection proving particularly detrimental for aggressive cancers known for their rapid progression, resistance to therapy, and early dissemination. These trends highlight the urgent need for earlier, minimally invasive, and biologically informative detection strategies.

Aggressive cancers share traits, including uncontrolled growth, increased invasiveness, and the ability to spread to distant areas of the body before spreading. Conventional diagnostic approaches, such as tissue biopsy and imaging, are limited by invasiveness, sampling bias, and poor temporal resolution, which has driven growing interest in liquid biopsy approaches. However, established liquid biopsy analytes, such as circulating tumor DNA, circulating tumor cells, and soluble proteins, face challenges related to abundance, specificity, and biological interpretability in aggressive and heterogeneous disease contexts.

Extracellular vesicles (EVs) are promising as liquid biopsy analytes due to their abundance, stability, and role in intercellular communication. They carry molecular cargo that reflects their cells of origin and are linked to aggressive cancer processes, including immune modulation, matrix remodeling, angiogenesis, and metastasis. Nevertheless, EV-based liquid biopsy is analytically challenging due to the heterogeneity of circulating EV populations and the low fraction of tumor-derived vesicles, necessitating cautious interpretation of sensing results.

This dissertation focuses on the development of a colloidal gold nanoparticle–based nanoplasmonic sensing platform for EV detection using four cancer cell line models of two cancer types: U87MG glioblastoma and U87MG-EGFRvIII glioblastoma, MDA-MB-231 (aggressive breast cancer), and MCF-7 (non-aggressive breast cancer). Guided by cross-cancer biological insights, EGFR signaling is employed as a test-case marker to evaluate detectability rather than as a definitive diagnostic biomarker, while the Vn96 peptide is used as a pan-EV capture probe. A controlled streptavidin–biotin–mediated antibody functionalization strategy is implemented, enabling EV detection at concentrations in the millions per milliliter range within a sub-3-hour workflow.

Label-free optical biosensing platforms based on LSPR offer high sensitivity and simplified workflows for EV detection. However, this sensitivity also renders them susceptible to artefacts arising from surface chemistry, nanoparticle aggregation, and nonspecific interactions. Distinguishing biologically meaningful EV binding from chemistry-driven optical effects is therefore a critical challenge for reliable EV sensing.

The central original contribution of this thesis is a systematic investigation of carbodiimide activation chemistry in colloidal gold nanoparticle–based LSPR sensors. The study shows that uncontrolled activation leads to aggregation-driven plasmon coupling, which can overshadow LSPR responses and complicate biological interpretation. By defining an operating window that preserves colloidal stability and signal fidelity, this work establishes a mechanistic foundation for reproducible and interpretable EV sensing.

Using the optimized biosensing framework, proof-of-concept detection of EV-enriched populations derived from breast cancer and glioblastoma cell lines is demonstrated using complementary affinity probes. In breast cancer models, Vn96 detects EVs from both MDA-MB-231 and MCF-7 cell lines, whereas EGFR-dependent responses are observed only for EVs derived from the aggressive MDA-MB-231 line. In glioblastoma models, Vn96 similarly captures EVs from both U87MG and U87MG-EGFRvIII cells, whereas EGFR-dependent detection is observed exclusively for EVs derived from the EGFRvIII-expressing model. These detection outcomes highlight differences in EV surface accessibility and biological diversity.

This thesis lays the groundwork for a reliable, interpretable EV-based liquid biopsy. By merging EV biology with nanoplasmonic chemistry, it supports efforts to establish liquid biopsy as a credible method for studying aggressive cancer behavior.