Chemical composition of biodegradable dissolved organic matter in streamwater

Plug-flow biofilm reactors colonized by microorganisms in streamwater were used to measure the concentration and composition of biodegradable dissolved organic C (BDOC) in White Clay Creek. During the 4-month study period, DOC ranged from 0.8 to 10.4 mg C liter-’ and was, on average, composed of 75% humic substances, 13% carbohydrates, 2% amino acids, and 18% > 100 kDa. The carbohydrates were predominantly polysaccharides, nearly all amino acids were present in the combined form, and most carbohydrates and amino acids were humic bound. BDOC ranged from 0.2 to 2.9 mg C liter-‘, averaged 25% of the DOC, and was composed of 75% humic substances, 30% carbohydrates, 4% amino acids, and 39% DOC >lOO kDa. The carbohydrate portion of the BDOC was primarily polysaccharide or humic bound. Similarly, the amino acid portion of the BDOC was overwhelmingly present in the combined form and primarily humic bound. Glycine and aspartic acid were the dominant amino acids in White Clay Creek DOC and in the BDOC pool. Our data broaden the perspective on substrates important to microbial metabolism and energy flow in streams and provide the first direct evidence that humic substances, although largely refractory, are an important component of streamwater BDOC. Dissolved organic matter (DOM) comprises most of the reduced carbon in aquatic ecosystems and provides energy and carbon resources for the metabolism of heterotrophic bacteria. Not all DOM is biologically labile or even biodegradable. Although numerous investigations in both freshwater and marine environments have reported on the quantity and composition of DOM, fewer studies have addressed the biodegradable fraction. Identifying biodegradable DOM (BDOM) constituents and quantifying their contribution to heterotrophic metabolism can increase our understanding of ecosystem function and bacterial ecophysiology. From a limited number of studies we know that BDOM in streams and rivers includes both low-molecular-weight (Kaplan and Bott 1983) and high-molecular-weight (Meyer 1 Present address: American Water Works Service Company, Belleville, Illinois. 2 Corresponding author: Stroud Water Research Center, Academy of Natural Sciences of Philadelphia, 970 Spencer Rd., Avondale, Pennsylvania 193 11. Acknowledgments Technical assistance was provided by H.E.P. Brooks and S. L. Roberts. This work was supported by the Stroud Foundation and the Boyer Fund for Environmental Research. The National Science Foundation (DEB-9308000) provided additional support for L.A.K. and the Philadelphia Suburban Water Company provided additional support for C.J.V. Funds for the construction of biofilm reactors were provided by the American Water Works Association Research Foundation, contract 729-9 1. et al. 1987; Amon and Benner 1996) components. It is also known that humic substances dominate the DOM pool in streamwater (Wallis and Ladd 1983), and it has been suggested that humic substances play an important role in ecosystem metabolism (Wetzel 1992). Our data are the first to directly measure the biodegradation of humic substances in streamwater. Measuring the contribution of different DOM constituents to the metabolism of heterotrophic bacteria has been hindered by methodological problems. The presence of poorly characterized organic complexes and numerous individual organic molecules at low concentrations makes comprehensive measures of uptake rates difficult on a molecular basis. Composite measurements, such as dissolved organic C (DOC) or humic substance analyses, are typically not sensitive enough to detect significant uptake without extended periods of incubation, concentration of substrate (Meyer et al. 1987), or both (Moran and Hodson 1990, 1994). As a result, the utilization of DOM components is often inferred from the growth of bacteria suspended in batch cultures of concentrated DOM fractions (Tranvik 1988; Leff and Meyer 1991). The application of biofilm reactor methods to the study of DOC utilization, wherein high densities of attached microorganisms are exposed to relatively low fluxes of organic substrates in a once-through mode, has made it possible to measure net DOC uptake in unamended streamwater over a period of hours (Kaplan and Newbold 1995). The objectives of the present study were to apply the bioreactor measure-