Circulation of hydrothermal fluids in upper oceanic crust may support one of the most extensive, but least understood, of the Earth's biogeochemical systems. However, little is known about non-living organic matter carried in crustal hydrothermal fluids or its possible impact on the carbon cycle of the overlying ocean. Elevated concentrations of particulate organic carbon with distinct composition are present near venting sites 3 , but particulate organic carbon composes only a small fraction of organic matter in most natural waters, with dissolved organic carbon (DOC) making up most of it 4 . Particulate organic carbon is typically both more labile and surface-active than DOC, so that particle compositions tend to reflect local conditions and relatively short-timescale processes 5 . DOC probably represents most organic matter carried by crustal fluids and subsequently vented to the overlying ocean, with its chemical and isotopic composition providing an integrated view of sources and processes remote from local venting sites.

We studied the mechanism(s) by which a deposit feeder can solubilize PAH from contaminated sediments as well as the implications of these mechanisms for factors controlling PAH bioavailability. Arenicola marina digestive fluids solubilize 4.6 μg mL-1 phenanthrene and 2.0 μg mL-1 benzo[a]pyreneconcentrations greater than the PAH's seawater solubilitieswhen incubated with pure PAH solids. This enhanced solubilization is largely due to surfactant micelles in the digestive fluid. In experiments with contaminated sediments that repeat the incubation or vary the solid−fluid ratio, these and other PAHs saturate at much lower concentrations (often between 0.01 and 0.1 μg mL-1). Less solubilization is likely due to sorption of digestive surfactants by sedimentary organic matter and competition from other sedimentary hydrophobic solutes, such as aliphatic hydrocarbons, for remaining micellar space. Nevertheless, gut fluid concentrations of high molecular weight PAHs are greater than those predicted from equilibrium partitioning theory, indicating the importance of the digestive pathway for hydrophobic organic contaminant exposure and bioaccumulation.

To develop a simple and cost-effective bioavailability test for sediment-bound contaminants, the solubilization strengths of mixtures of four commercially available surfactants and four proteins were compared to that of digestive fluids from a deposit-feeding benthic polychaete Arenicola marina. Initial tests indicated that sodium taurocholate, a vertebrate bile salt, was the most accurate mimic of A. marina gut fluids' solubilization of individual polycyclic aromatic hydrocarbons (PAH). Further testing with nutritional lipids and other hydrophobic contaminants confirmed the similarities of these fluids. Bovine serum albumin (BSA) solubilization of PAH was the most efficient of all the proteins tested. A cocktail of sodium taurocholate and BSA was compared to A. marina's solubilization of 12 PAH from four different contaminated sediments (from Boston, Charleston, Jacksonville, and San Diego harbors). The two solutions released most PAH to similar extents; 40 of 48 PAH-sediment combinations were released at amounts within a factor of 2 in cocktail and gut fluid solutions. Therefore, the cocktail may serve as a surrogate for real gut fluids and allow easier adoption of the in vitro incubation approach to bioavailability testing.