A microfluidic chemotaxis assay to study microbial behavior in diffusing nutrient patches

The nutrient environment experienced by planktonic microorganisms is patchy at spatiotemporal scales commensurate with their motility (μm – cm), and the efficiency with which chemotactic microbes can exploit this heterogeneous seascape influences trophodynamics and nutrient cycling rates in aquatic environments. Yet, methodological limitations have largely prevented direct examinations of microbial behavior within heterogeneous microenvironments. We used soft lithography to fabricate a microfluidic-based chemotaxis assay to study the foraging response of aquatic microbes to diffusing microscale nutrient patches. The transparency, biocompatibility, and simplicity of microfluidic devices make them ideally suited for microbial ecology studies. A microinjector was used to create a 300 μm-wide nutrient band, simulating a pulse release of solutes. The chemotactic response of microbes to the diffusing patch was measured at the population and single-cell level. In contrast to traditional chemotaxis assays, this technique permits the assessment and quantification of chemotaxis toward a potential attractant in real time, enabling rapid screening of multiple chemicals. Furthermore, detailed information on chemotactic behavior can be obtained by tracking individual organisms. Here, we applied this microassay to study the chemotactic behavior of a range of aquatic microorganisms, including three marine bacterial isolates, a species of phagotrophic flagellate, and a species of phytoplankton. Each organism exhibited a rapid chemotactic response to a variety of chemical compounds, suggesting that many marine microbes are adapted to life within patchy microenvironments. The chemotaxis assay described here was found to be a flexible platform for studying both the specific case of microbes foraging within patchy habitats and as a broadly applicable tool for rapidly assessing and quantifying microbial chemotaxis. *Corresponding author: E-mail: [email protected] Acknowledgments We are grateful to D. Hunt, M. Polz, and R. Belas for providing bacterial isolates, S. Stransky and T. Clay for the implementation of BacTrack, and three anonymous referees for insightful comments on the manuscript. This research was funded by NSF grant OCE-0526241. TA acknowledges support from a Martin Fellowship for Sustainability. Limnol. Oceanogr.: Methods 6, 2008, 477–488 © 2008, by the American Society of Limnology and Oceanography, Inc. LIMNOLOGY and OCEANOGRAPHY: METHODS Bacterial chemotaxis first was observed over a century ago (Pfeffer 1888), and has since been studied widely and demonstrated to be a ubiquitous adaptation among a broad range of species. Several techniques have been developed to assess chemotaxis. The most frequently used is the capillary assay technique (Pfeffer 1888; modified by Adler 1969), based on immersing a capillary tube filled with a potential attractant (or repellent) into a suspension of bacteria and enumerating the cells that migrate into the capillary after a prescribed time. Other chemotaxis assays include swarm plates (Adler 1966), stop-flow diffusion chambers (Ford et al. 1991; Ford and Lauffenburger 1991), and cell tethering assays (Silverman and Simon 1974). However, most techniques are either organismspecific, time-consuming, or do not allow direct observation of behavior. Due to the myriad of environments inhabited by microorganisms (Fenchel 2003), the ecological relevance of a chemotaxis assay depends on the phenotypic characteristics of the specific organisms and the physicochemical nature of their natural habitat. For instance, aquatic microbes often experience a heterogeneous and nutrient-deprived environment, characterized by sporadic and ephemeral microscale (μm – cm) patches of nutrients generated from several sources including the lysis of cells (Blackburn et al. 1997, 1998), excretions and exudations from other organisms (Lehman and Scavia 1982; Azam and Ammerman 1984; Mitchell et al. 1985), and leakage of substrates from sinking organic particles (Kiørboe and Jackson 2001). Hence, at sub-centimeter scales, the ocean is characterized by marked nutrient heterogeneity and the chemotactic foraging behavior of planktonic bacteria potentially can provide significant advantage in nutrient uptake (Blackburn et al. 1998; Kiørboe and Jackson 2001; Fenchel 2002; Stocker et al. 2008). Importantly, the typical life span of these microscale nutrient patches will be limited by the erosive effects of diffusion, so obtaining metabolic gains from a patch then becomes a race against time (Blackburn et al. 1997, 1998; Blackburn and Fenchel 1999a; Stocker et al. 2008). An understanding of marine microbial behavior within this dynamic and patchy nutrient landscape is important because efficient exploitation of nutrient patches by motile bacteria could have profound consequences for biogeochemical transformation rates (Azam 1998; Fenchel 2002). However, observations of microbial chemotactic foraging in patchy habitats have been hampered by methodological limitations and our current perception of these processes is based mostly on theoretical assumptions (Mitchell et al. 1985), numerical modeling (Bowen et al. 1993; Blackburn et al. 1997; Kiørboe and Jackson 2001), and evidence of bacteria clustering around randomly occurring and uncharacterized nutrient sources (Blackburn et al. 1998). Recent advances in microfabrication techniques (Whitesides et al. 2001) and the application of microfluidic systems to study microbial ecology have opened new doors for studying behavior of microorganisms at environmentally relevant spatiotemporal scales (Park et al. 2003; Keymer et al. 2006; Marcos and Stocker 2006; Weibel et al. 2007; Stocker et al. 2008). Microfluidic devices consist of small chips onto which complex patterns, including microchannels, can be constructed accurately with micrometer precision (Whitesides et al. 2001; Weibel et al. 2007). Microfluidic chemotaxis assays have been proposed recently as sensitive and convenient alternatives to traditional assays (Mao et al. 2003; Diao et al. 2006; Koyama et al. 2006; Cheng et al. 2007). Here, we describe the design, fabrication, and application of a novel microfluidic chemotaxis assay, designed specifically for creating microscale nutrient pulses with dimensions and diffusive characteristics matching those predicted to occur in the ocean. We demonstrate how this device can serve both as a rapid and flexible qualitative chemotaxis assay with broad applicability to aquatic microorganisms, and as an effective quantitative platform for studying their behavioral responses to ephemeral resource patches.