APPLICATION OF TOXICITY-BASED FRACTIONATION TECHNIQUES AND STRUCTURE-ACTIVITY RELATIONSHIP MODELS FOR THE IDENTIFICATION OF PHOTOTOXIC POLYCYCLIC AROMATIC-HYDROCARBONS IN SEDIMENT PORE-WATER

Recent studies conducted at our laboratory have shown that sediments c ontaminated with complex mixtures of polycyclic aromatic hydrocarbons (PAHs) can exhibit enhanced toxicity (lethality) to a variety of aquat ic species when the samples are tested under ultraviolet (UV) light de signed to mimic the wavelengths present in sunlight. However, because these contaminated sediments can contain literally thousands of chemic als, it is difficult to use conventional analytical techniques to iden tify those compounds responsible for photo-induced toxicity. The purpo se of this study was to adapt existing toxicity identification evaluat ion methods to attempt to identify those compounds contributing to the phototoxicity observed in our sediment samples. Pore water obtained f rom sediments collected near an oil refinery discharge was toxic to Lu mbriculus variegatus following exposure to UV light, while organisms e xposed to the same pore water, but without subsequent UV treatment, sh owed no toxic effect. Solid-phase extraction disks and high-performanc e liquid chromatography were used, in conjunction with toxicity tests with L. variegatus, to extract and fractionate phototoxic chemicals fr om the pore water. Phototoxic fractions analyzed by gas chromatography -mass spectrometry revealed the presence of a number of aliphatic hydr ocarbons, substituted PAHs, and PAHs containing heteroatoms. Chemicals were screened for their phototoxic potential based on empirical data and predictive models. A refined list of PAHs was then evaluated on th e basis of their phototoxic potency as defined by a recently developed quantitative structure-activity relationship model and estimates of t heir bioaccumulation potential. Based on the model predictions of pote ncy and bioaccumulation, nine likely phototoxic chemicals were identif ied.