疏水性有机污染物在水生生物体内的生物富集作用是影响其生态环境风险的重要过程，其在水生生物体内的累积包括水相吸收（生物浓缩）和摄食吸收（生物富集）两个途径，但目前有关摄食吸收对不同营养级水生生物富集疏水性有机污染物的稳态浓度和动力学过程尚不清楚。多环芳烃（polycyclic aromatic hydrocarbons，PAHs）是一类典型的疏水性有机污染物，并且广泛存在于环境中。PAHs可以累积在水生生物体内并产生一系列潜在毒性效应。基于此，本论文采用室内暴露实验和模型模拟相结合的手段，研究了摄食吸收对水生生物富集多环芳烃的影响机制。

为了避免实验室背景污染，选取了4种氘代多环芳（PAHs-d₁₀）作为本论文的目标污染物，包括3环的氘代菲和氘代蒽以及4环的氘代荧蒽和氘代芘。构建了由小球藻、大型溞和斑马鱼组成的淡水浮游食物链。分别开展了小球藻、大型溞和斑马鱼的生物浓缩、生物富集和净化实验。运用被动给料技术控制暴露期间水相中氘代多环芳烃的自由溶解态浓度在环境浓度量级（ng L^-1），并维持恒定。在本论文中，将PAHs-d₁₀在水生生物体内的摄食吸收分为两个过程：（1）PAHs-d₁₀通过食物携带进入水生生物肠胃或肠道，并将这个过程定义为摄食作用；（2）PAHs-d₁₀在肠胃或肠道消化吸收进而进入体内。根据不同营养级水生生物不同的摄食行为特点，构建了氘代多环芳烃在不同营养级水生生物体内富集的动力学模型框架。通过动力学实验数据和模型拟合，估计了氘代多环芳烃在大型溞和斑马鱼体内生物富集过程中关键的动力学参数包括水相吸收速率常数、生物转化速率常数、同化效率等。此外，通过本论文新构建的多房室富集动力学模型探究了氘代多环芳烃在斑马鱼体内的吸收、分布、生物转化和排泄过程。通过模型计算，明确了氘代多环芳烃在大型溞和斑马鱼体内的排出机制。基于动力学模型框架和动力学参数的估计值，通过基于能量需求的蒙特卡洛模拟分析探究了氘代多环芳烃在小球藻-大型溞-斑马鱼食物链上的传递过程。主要研究结果如下：

（1）当水相中PAHs-d₁₀自由溶解态浓度恒定时，额外摄食吸收含有PAHs-d₁₀的大型溞或鱼饲料会增加PAHs-d₁₀在斑马鱼体内的富集稳态浓度，PAHs-d₁₀在鱼体内的生物富集过程不仅受控于脂水分配作用，还受其摄食量的影响。进一步揭示了摄食作用的离散性和随机性对斑马鱼累积PAHs-d₁₀过程的影响，由于摄食作用的离散性，摄食间隔内PAHs-d₁₀在斑马鱼体内存在一个峰值浓度。在考虑连续的水相吸收和离散的摄食吸收的基础上，构建了一个针对鱼类富集PAHs-d₁₀的动力学模型，该模型考虑了PAHs-d₁₀从食物到肠胃再到体内传输的过程。基于能量需求的蒙特卡洛模拟分析结果表明，相比于仅有水相吸收，额外摄食大型溞在很大程度上会增加PAHs-d₁₀在斑马鱼体内的稳态浓度，但不会产生生物放大现象。由于摄食作用的离散性与随机性，PAHs-d₁₀在斑马鱼体内的稳态浓度在一定范围内呈随机波动状态，这在一定程度上解释了自然界同一水生生态系统中鱼体内PAHs浓度存在较大差异的原因。

（2）揭示了在组织水平下，摄食作用的离散性对PAHs-d₁₀在斑马鱼体内累积过程的影响，并构建了一个考虑连续的水相吸收和离散的摄食吸收的多房室富集动力学模型。在该模型中，斑马鱼被分为8个房室，包括肠胃、肝脏、血液、鳃、皮肤、肌肉、卵巢（包括鱼卵）和残体（包括心脏、胆囊、脾、肾、脑和骨等），考虑了PAHs-d₁₀从食物到肠胃再到体内的传输途径以及肝肠循环过程。暴露实验结果表明，由于摄食作用的离散性，在摄食间隔期间PAHs-d₁₀在斑马鱼体内不同组织和器官中的浓度存在一个峰值，达到峰值的时间以及峰值的大小在不同组织和器官中存在差异。实验和模型模拟结果表明，水相中PAHs-d₁₀通过皮肤吸收被斑马鱼累积的过程不可忽略，基于皮肤的水相吸收速率常数占总水相吸收速率常数的20%。PAHs-d₁₀主要分布在斑马鱼的血液、皮肤和肌肉中，稳态条件下分别占斑马鱼体内总累积量的20％-27％、20％-26％和16％-22％。与肠胃相比，肝脏是PAHs-d₁₀在斑马鱼体内进行生物转化的主要场所。PAHs-d₁₀在斑马鱼体内的分布不仅受到每个组织或器官中脂肪含量的影响，而且还受到其在不同组织器官与血液之间的传输过程和生物转化速率的影响。

（3）当PAHs-d₁₀的自由溶解态浓度恒定时，额外摄食含有PAHs-d₁₀的小球藻会增加PAHs-d₁₀在大型溞体内的富集稳态浓度。摄食吸收对PAHs-d₁₀在大型溞体内富集稳态浓度的相对贡献随暴露体系内小球藻密度的增加而增加，这是由于大型溞对小球藻的摄食速率随小球藻密度的升高而升高。进一步发现大型溞对小球藻的摄食间隔非常短暂，几乎每隔3 s就有一次摄食行为发生，因此可以认为大型溞摄食小球藻是一个近似连续的过程。由此，本研究构建了一个新的针对浮游动物富集PAHs-d₁₀的动力学模型，考虑连续的水相吸收和近似连续的摄食吸收，以及PAHs-d₁₀从食物到肠道再到体内的传输过程，该模型能很好拟合暴露实验动力学数据。模型拟合结果表明，排泄过程可能是摄食条件下PAHs-d₁₀在大型溞体内排出的主要途径之一。与传统的单房室和双房室动力学模型相比，该模型能更好估计大型溞对小球藻体内PAHs-d₁₀的同化效率以及区分PAHs-d₁₀的排出途径，并可提供更多更准确的动力学信息。

（4）提出摄食作用具有从无（对浮游植物）到有（对浮游动物和鱼类），从近似连续（对浮游动物）到离散（对鱼类）的特征，并且该摄食动力学特征会显著影响较易代谢的疏水性有机污染物（如PAHs）在不同营养级水生生物体内的生物富集过程，并构建了考虑水相吸收和摄食吸收的不同营养级水生生物富集动力学模型框架。近似连续的食物摄取导致PAHs-d₁₀在大型溞肠道和全身（除肠道）中的浓度逐渐升高至稳态，而间歇性食物摄入导致PAHs-d₁₀在斑马鱼肠胃和全身（除肠胃）中存在一个峰值浓度。本模型框架估算出的斑马鱼对大型溞体内PAHs-d₁₀的同化效率比大型溞对小球藻体内PAHs-d₁₀的同化效率高出近一倍。生物转化是PAHs-d₁₀在较高营养级生物斑马鱼体内排出的主要途径，而排泄和净化过程是PAHs-d₁₀在较低营养级生物大型溞体内排出的主要途径，生物转化也起到一定的作用。此外，暴露实验结果表明，在仅有水相吸收时，PAHs-d₁₀在小球藻-大型溞-斑马鱼食物链上发生营养级稀释现象。基于能量需求的蒙特卡洛模拟分析结果表明，尽管额外的摄作用不会导致营养级放大现象的发生，但减少了营养级稀释的程度，且减少程度随小球藻密度增加而增加，另外，摄食吸收对PAHs-d₁₀在斑马鱼体内富集稳态浓度的贡献高于大型溞，由此说明摄食吸收在高营养级鱼类累积PAHs的过程中扮演更重要的角色。

Bioaccumulation of hydrophobic organic compounds (HOCs) includes waterborne uptake and dietary uptake, which plays an important role in affecting their environmental risks. Still, the effects of dietary uptake on HOC bioaccumulation steady-state concentration and kinetics in aquatic organisms at different trophic levels are not clear. Polycyclic aromatic hydrocarbons (PAHs) are a typical class of HOCs, and they broadly exist in the environment. Moreover, PAHs could be accumulated by aquatic organisms and pose potential toxicity on them. Therefore, the present study investigated the influencing mechanisms of dietary uptake on the bioaccumulation of PAHs by aquatic organisms through laboratory-based exposure experiments and model-based simulations.

To avoid the potential interference of background PAHs, deuterated PAHs (PAHs-d₁₀) including 3-ring PAHs-d₁₀ such as phenanthrene-d₁₀ and anthracene-d₁₀, as well as 4-ring PAHs-d₁₀ such as fluoranthene-d₁₀ and pyrene-d₁₀ were chosen in this study. A simple artificial freshwater pelagic food chain was built composed of algae Chlorella vulgaris (C. vulgaris), zooplankton Daphnia magna (D. magna), and zebrafish (Danio rerio). Bioconcentration, bioaccumulation, and depuration experiments of PAHs-d₁₀ by C. vulgaris, D. magna and Danio rerio were quantified in a passive dosing exposure system separately, where the PAH-d₁₀ freely dissolved concentrations were maintained constant at environmentally relevant concentrations (ng L^?1). In the present study, dietary uptake was divided into two processes including ingestion of PAHs via food ingestion and the following digestion and uptake by gastrointestinal (GI) tract or gut. A toxicokinetic model framework was developed for the bioaccumulation of PAHs-d₁₀ by aquatic organisms at different trophic levels considering different feeding intervals. Through fitting the model to concentration-time profile data, the key toxicokinetic parameters were estimated such as waterborne uptake rate constant, biotransformation rate constant, and assimilation efficiency etc. Moreover, a new multi-compartmental toxicokinetic model was built to investigate the absorption, distribution, metabolism (biotransformation), and excretion (ADME) processes in zebrafish. Based on the model simulations, the elimination pathway of PAHs-d₁₀ in D. magna and zebrafish was determined. Based on the toxicokinetic model framework and estimated parameters, energy demand-based Monte Carlo simulations were run to investigate the trophic transfer of PAHs-d₁₀ in the simplified pelagic food chain composed of C. vulgaris, D. magna, and zebrafish. The main conclusions are as follows:

1.        When freely dissolved concentrations of the PAHs-d₁₀ were constant in water, the additional dietary uptake from contaminated D. magna or spiked commercial fish food could increase their steady-state concentrations in zebrafish. The bioaccumulation of PAHs-d₁₀ in zebrafish was controlled not only by lipid-water partitioning, but also by the intake amount of PAHs-d₁₀. Our study revealed that the discontinuity and randomness of dietary uptake would have a substantial impact on the bioaccumulation of PAHs-d₁₀ in zebrafish. Due to the intermittent food ingestion, there was a peak concentration in zebrafish during the feeding interval. A new bioaccumulation kinetic model for fish was established to take into account discrete dietary uptake, and it considered the transportation of PAHs-d₁₀ from ingested food to the GI tract then to the body. Energy demand-based Monte Carlo simulation results showed that the additional dietary uptake from contaminated D. magna could increase the steady-state concentrations of PAHs-d₁₀ in zebrafish, but it could not induce biomagnification. Due to the discontinuity and randomness of dietary uptake, the steady-state concentration would fluctuate within a range, which may partly explain the differences of PAH steady-state concentrations in fish in the same aquatic ecosystem.

2.        The present study revealed the effects of dietary uptake discontinuity on PAH-d₁₀ bioaccumulation by zebrafish at tissue level, and built a multi-compartmental toxicokinetic model considering continuous waterborne uptake and discrete dietary uptake. In this model, zebrafish was divided into eight compartments including GI tract, liver, blood, gill, skin, fillet, ovary, and residual (heart, gall bladder, spleen, kidney, brain, and bone), and the GI tract-liver-blood circulation transfer process for dietborne PAHs-d₁₀ in zebrafish was considered. The experimental results showed that there was a peak concentration in each compartment of zebrafish after every dietary uptake as a result of the intermittent food ingestion, and the peak concentration as well as the time point reaching the peak concentration varied among different compartments. The experimental and fitting results indicated that skin contributed to waterborne uptake and should not be ignored for zebrafish, the waterborne uptake rate constant of PAHs-d₁₀ through skin was about 20% of the total waterborne uptake rate constant through both skin and gill. The vast majority of PAH-d₁₀ amount was distributed in the blood, skin, and fillet of zebrafish at steady-state. Separately, the percentage of the amount in blood, skin, and fillet was 20%?27%, 20%?26%, and 16%?22%, respectively. Compared with GI tract, liver is considered as the main site for biotransformation of PAH-d₁₀ in zebrafish. The distribution of PAH-d₁₀ in zebrafish was not only affected by the lipid content in each compartment, but was also affected by the transportation between blood and each compartment and biotransformation.

3.        When freely dissolved concentrations of PAH-d₁₀ were constant in water, additional dietary uptake from contaminated C. vulgaris could increase the steady-state concentrations of PAH-d₁₀ in D. magna. The relative contribution of dietary uptake to the steady-state concentration increased with increasing algal density in the exposure medium because of the increasing ingestion rate. Furthermore, the observed feeding interval of D. magna on C. vulgaris was very short within seconds (pulse ingestion), and the ingestion occurred every 3 seconds. Therefore, the feeding of D. magna on C. vulgaris could be viewed as an approximately continuous process. A new bioaccumulation kinetic model for zooplankton was established to take into account continuous waterborne uptake and approximately continuous dietary uptake, and it considered the transportation of PAHs-d₁₀ from ingested food to the gut then to the body. The new established model fitted well with experimental concentration-time profile data. The fitting results indicated that the excretion would be mainly responsible for elimination of PAH-d₁₀ in D. magna when there was dietary uptake. Compared with traditional one-compartment or two-compartment toxicokinetic model, the new established model could accurately estimate assimilation efficiency, distinguish elimination pathway, and provide more kinetic information.

4.        We postulated that dietary uptake was expected to range from none for phytoplankton to approximately continuous for zooplankton to discrete for fish, and such feeding interval could significantly affect the bioaccumulation of HOCs that are prone to be biotransformed such as PAHs by aquatic organisms at different trophic levels. Moreover, a toxicokinetic model framework was established considering such difference in feeding interval. The approximately continuous food ingestion would cause a gradual increase of PAH-d₁₀ concentrations in the gut and the body except gut of D. magna. In contrast, the intermittent food ingestion would induce an instantaneous increase of PAH-d₁₀ concentrations in the GI tract of zebrafish, and a convex variation (first increase then decrease) of concentrations in the whole body except GI tract of zebrafish. The estimated assimilation efficiency of PAH-d₁₀ in zebrafish was almost one time higher than their counterpart in D. magna. Biotransformation might dominate the elimination of PAHs-d₁₀ in zebrafish. Excretion and depuration would be mainly responsible for elimination of PAHs-d₁₀ in D. magna, and biotransformation might also play a role. The experimental results showed that trophic dilution of PAHs-d₁₀ occurred along the artificial freshwater pelagic food chain composed of C. vulgaris, D. magna, and Danio rerio when there was waterborne-only uptake. Energy demand-based Monte Carlo simulation results indicated that additional dietary uptake could significantly increase the body burden of PAHs-d₁₀ in D. magna and zebrafish, but such increase could not lead to a trophic magnification of PAHs-d₁₀ in the pelagic food chain. Nevertheless, it largely mitigated the extent of their trophic dilution, and the extent increased with increasing algal density. Moreover, the relative contribution of dietary uptake to the bioaccumulation steady-state concentrations of PAHs-d₁₀ in zebrafish was higher than that in Daphnia magna, suggesting that dietary uptake played a more important role in bioaccumulation of PAHs by higher trophic level organisms.