Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Civil and Environmental Engineering | en_US |
| dc.contributor.advisor | Jin, Ling (CEE) | en_US |
| dc.creator | Liu, Xintong | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/14644 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Bioanalytical assessment of contaminant cocktails in marine cetaceans using species-specific cell lines | en_US |
| dcterms.abstract | The global marine environment increasingly acts as a sink for a complex array of chemical contaminants, originating from both intensified anthropogenic activities and naturally occurring biological processes. In subtropical coastal ecosystems, such as the waters surrounding Hong Kong adjacent, this pollution manifests as a chemical cocktail comprising legacy persistent organic pollutants (POPs), contaminants of emerging concerns (CECs) and pharmaceuticals and personal care products (PPCPs) originated from industrial discharges, and prevalent marine algal toxins produced by harmful algal blooms (HABs). Resident cetaceans, the Indo-Pacific humpback dolphin (Sousa chinensis) and the Indo-Pacific finless porpoise (Neophocaena phocaenoides), act as sentinel species for this ecosystem since their high trophic level, long lifespan, and extensive use of coastal habitats expose them to health risks from these complex mixtures. However, traditional ecotoxicological driven primary by in vitro single-chemical testing, cannot resolve the mixture effects of environmental contaminants, identify toxic drivers and link external environmental concentrations to biologically relevant internal doses. Furthermore, standard targeted screening analysis captures only a fraction of the chemical universe, leaving the majority of toxicity compounds in environmental mixtures unidentified. To address these critical knowledge gaps, this thesis establishes an integrated, mechanism-informed framework that combines the development of species-specific in vitro cell models, non-target screening (NTS) coupled with machine learning (ML), and physiologically based toxicokinetic (PBTK) modeling to provide a comprehensive risk assessment of seawater contaminant cocktails for Hong Kong's cetaceans. | en_US |
| dcterms.abstract | The research methodology involved the collection of 72 surface seawater samples from Hong Kong coastal waters across wet and dry seasons. To overcome the ethical and logistical constraints associated with studying endangered marine mammals, primary and immortalized fibroblast cell lines were successfully established from the skin of the humpback dolphin (CWDT), while cell lines from the finless porpoise (FPT) donated by Shantou University. These cell lines provided a biologically relevant platform for measuring cytotoxicity, intracellular reactive oxygen species (ROS) induction, and DNA damage caused by seawater extracts and 38 individual target contaminants. The study further conducted a five-compartment PBTK model to simulate the absorption, distribution, and elimination of contaminants, facilitating quantitative in vitro to in vivo extrapolation (QIVIVE) to predict internal tissue concentrations and refine risk assessments based on freely dissolved concentrations. Additionally, this study integrated a high-resolution Orbitrap mass spectrometry for non-target screening (NTS) and a random forest machine-learning model to train on experimental toxicity data to prioritize potential toxicants from tens of thousands of unknown chemical features. The in vitro bioassays revealed that seawater extracts elicited cytotoxicity and oxidative stress in both cetacean cell lines, with species-specific sensitivities. Specifically, the mean toxic units (TU) required to induce cytotoxicity and reactive oxygen species (ROS) generation were 0.0058 and 0.0269 in FPT, and 0.02 and 0.031 in CWDT. A pivotal finding of this research was the disproportionate role of natural toxins in driving observed toxicity. Algal toxins, specifically pectenotoxin-2 (PTX-2), okadaic acid (OA), and gymnodimine (GYM), were identified as the dominant drivers even at low detected concentration (Mean: 8.38 × 10⁻¹³ M, 6.88 × 10⁻¹³ M, and 5.33 × 10⁻¹³ M, respectively), far exceeding the toxicity contribution of anthropogenic pollutants. PTX-2 as extraordinarily toxic potent and explained majority of the cytotoxicity caused by known chemicals in mixture modeling. While PTX-2 drove cytotoxicity, other toxins like OA and GYM were the primary contributors to oxidative stress. Interestingly, anthropogenic chemicals such as per- and polyfluoroalkyl substances (PFAS) and ultraviolet (UV) filters, while less cytotoxic, contributed measurably to ROS induction, suggesting they pose a risk of oxidative injury and skin barrier disruption rather than acute lethality. Despite characterizing a broad suite of target compounds, the study confirmed a significant iceberg effect, where the majority of cytotoxicity and ROS induction observed in seawater extracts remained unexplained by targeted chemicals, highlighting the presence of unidentified toxic drivers. The application of the PBTK model provided a critical shift in risk perspective from external exposure to internal dose. The simulations revealed that skin and blubber serve as major reservoirs for semi-lipophilic pollutants, with contaminants like phenanthrene and octocrylene accumulating substantially in the skin due to slow clearance and lipophilicity. A divergence was observed between in vitro potency and in vivo accumulation. High toxic potency algal toxins showed low skin accumulation due to rapid hepatic clearance, whereas anthropogenic contaminants persisted in tissues. Consequently, when risk was assessed using external seawater concentrations, the mixture appeared to pose low risk. However, utilizing PBTK-derived internal skin concentrations dramatically increased risk estimates, with the internal risk for cytotoxicity and DNA damage approaching the threshold of concern. Notably, algal toxin still emerged as a critical internal risk driver, as pharmacokinetic factors amplified its contribution to internal cytotoxicity risk to in the skin compartment. | en_US |
| dcterms.abstract | To elucidate the unexplained toxicity observed in the bioassays, an integrated NTA and ML workflow successfully prioritized potential toxicity candidates from tens of thousands of unknown chemical features. The analysis revealed that the toxicological landscape is defined by endpoint-specific signatures, with the random forest (RF) models demonstrating a performance gradient where cytotoxicity prediction achieved the highest accuracy (80.0%). Specifically, the framework identified high-potency candidates including resiniferatoxin (a natural neurotoxin), clarithromycin (a macrolide antibiotic), and a cinnamoyl-iridoid glycoside pro-toxicant. These findings point to a complex source profile where domestic wastewater markers and natural toxins constitute a persistent chemical background. Mechanistically, the ultrapotent receptor-binding of resiniferatoxin and the potential DNA interaction of organoheterocyclics align well with the observed toxicological endpoints. This discovery highlights that the hidden toxicity stems from a small subset of high-potency drivers within a stable wastewater-plastic complex, necessitating a multi-endpoint approach to fully deconstruct coastal chemical risks. | en_US |
| dcterms.abstract | In conclusion, this thesis presents the first comprehensive, mechanism-based framework for assessing the dermal health risks of chemical mixtures to Hong Kong's resident cetaceans. The findings challenge the traditional regulatory focus on anthropogenic contaminants by demonstrating that natural algal toxins are primary drivers of acute toxicity, even in the absence of visible algal blooms. Simultaneously, widespread anthropogenic contaminants contribute to oxidative stress and accumulate significantly in cetacean skin. The integration of PBTK modeling underscores that external water monitoring drastically underestimates toxicological risk, as contaminants appearing safe at environmental levels can reach hazardous internal concentrations. Furthermore, the discovery of pharmaceutical residues and natural toxins via machine learning highlights the urgent need to expand monitoring programs to include these unknown pollutants. By elucidating the drivers of mixture toxicity, this research provides a scientific foundation for holistic monitoring strategies and internal dose-based frameworks, which are essential for the effective conservation of vulnerable Indo-Pacific cetacean populations. | en_US |
| dcterms.extent | 228 pages : color illustrations, map | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2026 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14644