Portal de Periódicos da CAPES

The Laboratory in a Droplet

2005; Elsevier BV; Volume: 12; Issue: 12 Linguagem: Inglês

10.1016/j.chembiol.2005.11.006

1879-1301

Helen M. O’Hare, Kai Johnsson,

Innovative Microfluidic and Catalytic Techniques Innovation

In this issue of Chemistry & Biology, the groups of Tawfik [1Ahranoi A. Amitai G. Bernath K. Magdassi S. Tawfik D. Chem. Biol. 2005; 12 (this issue): 1281-1289Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar] and Griffiths [2Mastrobattista E. Taly V. Chanudet E. Treacy P. Kelly B.T. Griffiths A.D. Chem. Biol. 2005; 12 (this issue): 1291-1300Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar] present fluorescence-activated cell sorting of double emulsions as a generally applicable screen for enzyme activity. This novel methodology increases the throughput of a typical enzyme screen by two orders of magnitude. In this issue of Chemistry & Biology, the groups of Tawfik [1Ahranoi A. Amitai G. Bernath K. Magdassi S. Tawfik D. Chem. Biol. 2005; 12 (this issue): 1281-1289Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar] and Griffiths [2Mastrobattista E. Taly V. Chanudet E. Treacy P. Kelly B.T. Griffiths A.D. Chem. Biol. 2005; 12 (this issue): 1291-1300Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar] present fluorescence-activated cell sorting of double emulsions as a generally applicable screen for enzyme activity. This novel methodology increases the throughput of a typical enzyme screen by two orders of magnitude. Numbers are all important in the world of high-throughput protein technology, as the size of the throughput can make the difference between success (striking gold with a rare positive event) and failure (not digging deep enough!). Today's biological questions often concern scales of millions and billions: the human interactome is currently estimated at 280,000 functional interactions from a total of 2.4 × 108 possible pairwise combinations, whereas screening for tomorrow's antibody pharmaceuticals is performed on libraries of 109–1013 molecules (www.cambridgeantibody.com). Current techniques for assaying enzyme activity, however, fall far short of the mark. Typical high-throughput screens are performed in microplates at scales of 103–105 reactions, and the investment of handling time and reagent costs fixes a ceiling of 106–107. Microplates permit parallel reactions of soluble molecules without diffusion, cross contamination, or loss of identity of each well, but trends toward miniaturization and increased throughput will eventually reach their limits. Nature suggests an alternative compartmentalization, namely the cell, which retains the advantages of the microplate, but at drastically reduced size. Cells keep together the genes, the RNAs, and the proteins that they encode, and the products of their activities, thus linking genotype to phenotype. Inspired by this property, an in vitro system of cell-like compartments was developed by Tawfik and Griffiths in 1998 [3Tawfik D.S. Griffiths A.D. Nat. Biotechnol. 1998; 16: 652-656Crossref PubMed Scopus (764) Google Scholar]. An aqueous in vitro transcription/translation reaction is emulsified to give a water-in-oil (w/o) emulsion of 1010 droplets. Since DNA is added at limiting dilution, each aqueous droplet contains a single gene, and acts as a unique, independent reaction vessel. Initially the method was demonstrated by in vitro expression of DNA methyltransferases, which then methylated their own genes, providing a selection criterion for active enzymes. In the last few years, activity-based selections of other DNA-associated enzymes have been developed, including polymerases [4Ghadessy F.J. Ong J.L. Holliger P. Proc. Natl. Acad. Sci. USA. 2001; 98: 4552-4557Crossref PubMed Scopus (303) Google Scholar] and restriction enzymes [5Doi N. Kumadaki S. Oishi Y. Matsumura N. Yanagawa H. Nucleic Acids Res. 2004; 32: e95Crossref PubMed Scopus (41) Google Scholar]. Heat-stable w/o emulsion formulations allow PCR in emulsion as a method for superior unbiased amplification of complex samples of genomic DNA or RNA [6Nakano M. Nakai N. Kurita H. Komatsu J. Takashima K. Katsura S. Mizuno A. J. Biosci. Bioeng. 2005; 99: 293-295Crossref PubMed Scopus (46) Google Scholar], and the generation of libraries of DNA on beads as a viable alternative to cloning [7Nakano M. Komatsu J. Matuura S. Takashima K. Katsura A. Mizuno A. J. Biotechnol. 2003; 102: 117-124Crossref PubMed Scopus (189) Google Scholar, 8Kojima T. Takei Y. Ohtsuka M. Kawarasaki Y. Yamane T. Nakano H. Nucleic Acids Res. 2005; 33: e150Crossref PubMed Scopus (109) Google Scholar, 9Samatov T.R. Chetverina H.V. Chetverin A.B. Nucleic Acids Res. 2005; 33: e145Crossref PubMed Scopus (19) Google Scholar]. In vitro expression in emulsion is an efficient method to generate large libraries of proteins physically linked to their encoding genes either by direct covalent attachment or via beads [10Sepp A. Tawfik D.S. Griffiths A.D. FEBS Lett. 2002; 18: 455-458Abstract Full Text Full Text PDF Scopus (101) Google Scholar]. Microfluidic channels offer improved methods to generate and manipulate emulsions of precisely controlled droplet size, and the contents of droplets can be interrogated by confocal microscopy [11Dittrich P.S. Jahnz M. Schwille P. ChemBioChem. 2005; 6: 811-814Crossref PubMed Scopus (165) Google Scholar]. Recently the massively parallel nature of emulsified reactions was exploited by two independent groups to sequence an entire bacterial genome in a single reaction [12Shendure J. Porreca G.J. Reppas N.B. Lin X. McCutcheon J.P. Rosenbaum A.M. Wang M.D. Zhang K. Mitra R.D. Church G.M. Science. 2005; 309: 1728-1732Crossref PubMed Scopus (1016) Google Scholar, 13Margulies M. Egholm M. Altman W.E. Attiya S. Bader J.S. Bemben L.A. Berka J. Braverman M.S. Chen Y.J. Chen Z. et al.Nature. 2005; 437: 376-380Crossref PubMed Scopus (5963) Google Scholar]. Each group developed an innovative method of sequencing by synthesis, and in both cases, the role of the emulsion is in single-molecule “cloning” and template preparation on beads. Emulsion technology offers unprecedented high throughput of compartmentalized reactions and will undoubtedly be taken up, adapted, and applied for diverse new methods, some as yet unimagined. Nevertheless, it has suffered from one serious drawback so far: the lack of a general method to assay the contents of each droplet and retrieve those of interest. Directed evolution of enzyme function in emulsions was either obligately linked to gene survival (selection rather than screening), with the inherent compromises in scope of reaction and dynamic range, or was reliant on bead capture and subsequent screening of the beads. To overcome this limitation, the groups of Tawfik and Griffiths have turned back to the analogy of cells. Fluorescence-activated cell sorting (FACS) is a cornerstone technique in biology, as single cells can be selected at a rate of 107 per hour on the basis of one (or several) fluorescence properties. Fluorescence assays are extremely sensitive, and the display of proteins on the surface of yeast, followed by capture of fluorescent ligands, is becoming an important screen for binding affinity [14Boder E.T. Midelfort K.S. Wittrup K.D. Proc. Natl. Acad. Sci. USA. 2000; 97: 10679-10681Crossref PubMed Scopus (551) Google Scholar]. Although fluorogenic substrates exist for many enzymes, FACS has not been widely applied to screen enzyme activity because of the difficulty in capturing the fluorescent products after the reaction. For individual enzymes, particular strategies have been devised to capture products on the cell surface [15Varadarajan N. Gam J. Olsen M.J. Georgiou G. Iverson B.L. Proc. Natl. Acad. Sci. USA. 2005; 102: 6855-6860Crossref PubMed Scopus (123) Google Scholar], or on beads [16Griffiths A.D. Tawfik D.S. EMBO J. 2003; 22: 24-35Crossref PubMed Scopus (253) Google Scholar], but the need to design and synthesize a new substrate for each enzyme severely limits the scope of FACS for application to new enzymes, and limits the accessibility to specialist laboratories. The breakthrough described in this issue of Chemistry & Biology [1Ahranoi A. Amitai G. Bernath K. Magdassi S. Tawfik D. Chem. Biol. 2005; 12 (this issue): 1281-1289Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar, 2Mastrobattista E. Taly V. Chanudet E. Treacy P. Kelly B.T. Griffiths A.D. Chem. Biol. 2005; 12 (this issue): 1291-1300Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar] lies in the formation of double water-in-oil-in-water (w/o/w) emulsions to provide stable linkage between genes, enzymes, and enzyme products. The oil layer which surrounds each droplet has low, controllable permeability, allowing fully soluble reactions to be performed in complete isolation from each other. The external aqueous phase renders the emulsions nonviscous and amenable to manipulation in a standard flow sorter (Figure 1). These papers pave the way for FACS screening of almost any fluorogenic assay. The two groups demonstrate the efficiency of the method with unrelated enzymes: Tawfik and coworkers assayed thiolactonase activity by coupling of the reaction product to a thiol-reactive fluorogenic dye, whereas Griffiths and coworkers detected β-galactosidase by the release of fluorescein from fluorescein di-β-D-galactopyranoside. Test mixtures of active and inactive genes were screened to demonstrate detection of enzyme activity, enrichment of genes encoding active enzymes (by 300-fold), and the ability to distinguish different rates of enzyme activity by the difference in fluorescence intensity. Both groups chose an enzyme that had previously been studied by directed evolution, thus safeguarding the success of the experiment while proving the worth of the system by evolving hundred-fold improvements on the low activities of the parent proteins (PON1 [1Ahranoi A. Amitai G. Bernath K. Magdassi S. Tawfik D. Chem. Biol. 2005; 12 (this issue): 1281-1289Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar] and Ebg [2Mastrobattista E. Taly V. Chanudet E. Treacy P. Kelly B.T. Griffiths A.D. Chem. Biol. 2005; 12 (this issue): 1291-1300Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar]). In each case, the improved mutants were equal to the best mutants isolated by traditional methods of bacterial colony screening on agar plates. Furthermore, Griffiths and coworkers discovered new beneficial mutations in the Ebg system, which was thought to have been mutated to exhaustion. An important difference between the two experiments lies in the means of protein expression. Griffiths and coworkers used in vitro translation, which combines the advantages of cloning-free library preparation, lack of bias due to clonal selection, control of the reaction conditions and constituents, and the potential to work with toxic substrates or toxic proteins [2Mastrobattista E. Taly V. Chanudet E. Treacy P. Kelly B.T. Griffiths A.D. Chem. Biol. 2005; 12 (this issue): 1291-1300Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar]. Tawfik and coworkers emulsified intact E. coli, expressing enzyme variants either in the cytoplasm or on the cell surface, to obtain higher enzyme concentrations [1Ahranoi A. Amitai G. Bernath K. Magdassi S. Tawfik D. Chem. Biol. 2005; 12 (this issue): 1281-1289Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar]. It seems likely that, for future applications, the choice between a cell-based or in vitro approach will be dictated by the activity in question, and that both approaches might be used in turn during the evolution of a single enzyme, perhaps taking advantage of the high enzyme concentrations in bacteria for the initial evolution of activity against a new substrate and then using a cloning-free and more chemically defined in vitro system for iterative improvements to fine tune substrate selectivity. In vitro display technologies such as bead display or ribosome or mRNA display should also be compatible with water/oil/water emulsions, and the technique might also be adapted to screen ribozymes for multiple turnover. The enormous battery of available fluorogenic substrates and coupled assays should allow flow sorting of emulsions to be applied to achieve new goals with many different biocatalysts. Each double emulsion approach offers massively parallel, yet individual and identifiable, femtoliter reaction compartments, combined with minimal handling time. High-throughput enzyme screening is a limiting factor, not just in directed evolution, but in other areas such as enzyme discovery by function-based cloning, functional genomics, and drug discovery. For selection of enzymes to catalyze commercially valuable transformations, or for assays of small amounts of natural products or drug leads, the small reaction volume (50 microlitres of aqueous phase forms 1010 droplets), and high local concentrations, may be particularly important. This breakthrough increases the throughput of a typical enzyme screen by over two orders of magnitude, and application to other biocatalysts promises to be simple. Apart from a flow sorter, the process does not require specialized equipment and could become as widely adopted as microplate screening. High-Throughput Screening of Enzyme Libraries: Thiolactonases Evolved by Fluorescence-Activated Sorting of Single Cells in Emulsion CompartmentsAharoni et al.Chemistry & BiologyDecember, 2005In BriefSingle bacterial cells, each expressing a different library variant, were compartmentalized in aqueous droplets of water-in-oil (w/o) emulsions, thus maintaining a linkage between a plasmid-borne gene, the encoded enzyme variant, and the fluorescent product this enzyme may generate. Conversion into a double, water-in-oil-in-water (w/o/w) emulsion enabled the sorting of these compartments by FACS, as well as the isolation of living bacteria cells and their enzyme-coding genes. We demonstrate the directed evolution of new enzyme variants by screening >107 serum paraoxonase (PON1) mutants, to yield 100-fold improvements in thiolactonase activity. Full-Text PDF Open ArchiveHigh-Throughput Screening of Enzyme Libraries: In Vitro Evolution of a β-Galactosidase by Fluorescence-Activated Sorting of Double EmulsionsMastrobattista et al.Chemistry & BiologyDecember, 2005In BriefWe describe a completely in vitro high-throughput screening system for directed evolution of enzymes based on in vitro compartmentalization (IVC). Single genes are transcribed and translated inside the aqueous droplets of a water-in-oil emulsion. Enzyme activity generates a fluorescent product and, after conversion into a water-in-oil-in-water double emulsion, fluorescent droplets are sorted using a fluorescence-activated cell sorter (FACS). Earlier in vivo studies have demonstrated that Ebg, a protein of unknown function, can evolve to allow Escherichia coli lacking the lacZ β-galactosidase gene to grow on lactose. Full-Text PDF Open Archive