Use of differential fluorescence induction and optical trapping to isolate environmentally induced genes

Life Sciences Rm 539, 1280 Main Street West,McMasters University, Hamilton, Ontario, Canada L8S4K1.SummaryThe techniques of differential fluorescence induction(DFI) and optical trapping (OT) have been combinedto allow the identification of environmentally inducedgenes in single bacterial cells. Designated DFI-OT,this technique allows the in situ isolation of genesdriving the expression of green fluorescent protein(Gfp) using temporal and spatial criteria. A series ofplasmid-based promoter probe vectors (pOT) wasdeveloped for the construction of random genomiclibraries that are linked to gfpUV or egfp. Bacteria thatdo not express Gfp on laboratory medium (i.e. non-fluorescent) were inoculated into the environment,and induced genes were detected with a combinedfluorescence/optical trapping microscope. Using thisselection strategy, rhizosphere-induced genes withhomology to thiamine pyrophosphorylase (thiE) andcyclic glucan synthase (ndvB) were isolated. Othergenes were expressed late in the stationary phase oras a consequence of surface-dependent growth,including fixND and metX, and a putative ABCtransporter of putrescine. This strategy provides aunique ability to combine spatial, temporal andphysical information to identify environmental reg-ulation of bacterial gene expression.IntroductionThe in situ physiology of soil microbes is difficult toanalyse because the complex natural environment cannotbe replicated in the laboratory. To overcome this difficulty,several new techniques have been developed to measuremicrobiological activities in situ (Osbourn et al., 1987;Mahan et al., 1993; Hensel ., 1995; Valdivia andFalkow, 1996; 1997; Pelz et al., 1999; Rainey, 1999;Ramos et al., 2000). These include signature-taggedmutagenesis (STM), in vivo expression technology (IVET)and differential fluorescence induction (DFI) (for reviews,see Chiang et al., 1999; Handfield and Levesque, 1999;Lee and Camilli, 2000). Each has its particular benefitsand weaknesses and has a role in investigating geneinduction in the environment.In DFI, expression of a modified green fluorescentprotein (Gfp) reporter fusion is linked to a fluorescence-activated cell sorter (FACS) (Valdivia and Falkow, 1996;1997). The benefits of DFI include the automatedscreening of large populations and the ability to alter thefluorescence window to detect changes in expressionlevels rather than simply identifying tightly regulatedpromoters. However, disadvantages of DFI includeproblems sorting bacteria, which are at the low end ofthe detection capabilities of flow cytometers (Valdivia andFalkow, 1996), and the difficulty of isolating bacteria fromnaturally fluorescent environments.Gfp has a number of advantages over other reporters ofgene expression. It is detectable in single cells using amicroscope and requires no cofactors (Gage et al., 1996;Cirillo et al., 1998). We considered that a powerfulstrategy would be to produce a versatile promoter probevector linked to gfp for use on agar or microtitre plates orunder the microscope and that this precision should becombined with IVET/DFI-like strategies to enable spatialinformation from microscopy to be linked with the ability toselect genes specifically induced in the environment. Theability to isolate single cells from a population has beenmade possible by the recent development of opticaltrapping microscopes (Ashkin, 1997; Allaway et al., 2000).Optical trapping uses the property of coherent light toproduce radiational pressure, sufficient to hold a bacter-ium within the light beam (Ashkin and Dziedzic, 1987;1989; Ashkin et al., 1987).Combining a DFI strategy with optical trapping (OT) hasallowed the development of a system that enables theidentification and isolation of single cells expressingenvironmentally induced gene fusions under a micro-scope for further characterization. Here, we describe theconstruction and testing of this strategy, designated