WO2001032858A1 - A high throughput screening (hts) method - Google Patents

Abstract

The inventors have found that a combination of High Troughput Screening (HTS) assays in a sequential manner have allowed a much improved screening process. Certain HTS-assays are able to screen extremely large numbers of clones for materials or activities of interest, such as Substrate Reloading assays, whereas Array based HTS assays are only efficient with a smaller number of clones; all relative to screening capacity. The method of this invention combines the advantages of at least two different HTS-assays, for instance a pre-screen Substrate Reloading assay will first semi-quantitatively select clones with an activity of interest, whereupon an Array based assay will rank the isolated active clones in a more quantitative manner according to activity.

Description

A high throughput screening (HTS) method

Field of invention Microbial organisms are widely used for the manufacture of various industrial products and optimizing the production in microbial hosts with respect to yield and production cost has become a highly competitive area of research. The microbial production of such products as industrial enzymes or pharmaceu- ticals is usually achieved by a fermentation process followed by one or more purification steps.

Microbial products that are secreted from the producing microorganism into the fermentation broth are particularly preferred in the industry, since secretion minimizes the required post -fermentative purification. Methods to efficiently screen microorganisms or gene libraries for production of secreted products have consequently become of great importance.

Background Technologies such as DNA shuffling, random mutagenesis and in vivo recombination have allowed the generation of enormous populations of variant cells which produce variants of a certain protein, RNA, or small molecule. In addition, it has become possible to establish large DNA libraries of natural proteins from other organisms in production strains. Together, this has created a need to develop assays by which large numbers of variant cells can efficiently and accurately be screened. Also, automatization has become necessary to handle these large populations. Use of a Fluorescence Activated Cell Sorter (FACS) to screen for enzymes is described in the art. WO 98/58085 describes screening gel microdrops (GMD's) for an intracellular enzyme (β-galactosidase) . Nir et al . (Nir, R., Y. Yisraeli, R. Lamed, and E. Sahar. 1990. Flow cytometry sorting of viable bacteria and yeasts according to beta-galactosidase activity. Appl . Environ. Microbiol . 56: 3861-3866) developed a method for sorting cells on the basis of expression of β-galactosidase using a fluorescent substrate. The β-galactosidase assay required partial permeabilization of cells and ' cross-talk' (i.e. exchange of metabolites) between GMD's was minimized by carrying out the β-galactosidase assay at 4°C. Intracellular esterase activity has been measured in Pseudomonas aeruginosa using 6- carboxy-fluorescein-diacetat , and an assay measuring the disappearance of fluorescence from GMD's containing fluorescently labeled casein was described for elastase secreted by Pseudomo- nas aeruginosa (Sahar, E., R. Nir, and R. Lamed. 1994. Flow cy- tometric analysis of entire microbial colonies. Cytometry 15:213-221). Cid et al (V. J. Cid, A. M. Alvarez, A. I. Santos, C. Nombela, and M. Sanchez. 1994. Yeast exo-beta-glucanases can be used as efficient and readily detectable reporter genes in Saccharomyces cereviεiae . Yeast 10 (6) : 747-756) describe a FACS based reporter system using a β-glucanase that partially accumulates in the yeast periplasm. WO 99/10539 describes an assay for bioactive substances by co-encapsulation of library and target cells and sorting on the basis of e.g. live/dead staining. A similar approach using co-encapsulation of cells in a screening for natural compounds was described in WO 98/41869. WO 98/49286 discloses a surface display system where a protease (OmpT) is retained on the cell surface and a substrate is modified to bind to the cell by electrostatic interactions. Prote- ase action is detected by a FACS via de-quenching of a Bodipy- FL tri-methyl-rhodamine amino acid substrate where the BIDOPY- FL part is retained at the cell surface. In US 4,401,755 coating GMD's is indicated as a way to make GMD's essentially impermeable to the substrates or products of the activity screened for.

Recently, methods have been described, involving the attachment of the substrate of the target reaction to a protein with potential catalytic activity towards the attached substrate (Pedersen et al . , Proc. Natl. Acad. Sci., US, 1998, vol. 95, pp. 10523-10528; Jestin et al . , 1999, Angew. Chem. Int. Ed., vol. 38, pp. 1124-1127; Demartis et al . , 1999, JMB, vol. 286, pp. 617-633; Neri et al . , 1997, WO 97/40141). Upon intramolecular conversion of the substrate, the active catalyst can be isolated by means of the attached product. This scheme is very general. However, since successful catalysts need only perform one turn-over during the selective process/round, which is typically in the order of minutes, there is no selective advantage for efficient catalysts. For the same reason, it presumably is not possible to distinguish enzymes with slightly different specific activity with this selection scheme.

Summary of the invention

The inventors have found that a combination of High Troughput Screening (HTS) assays in a sequential manner have allowed a much improved screening process. Certain HTS-assays are able to screen extremely large numbers of clones for materials or activities of interest, such as Substrate Reloading assays, whereas Array based HTS assays are only efficient with a smaller number of clones; all relative to screening capacity. The method of this invention combines the advantages of at least two different HTS-assays, for instance a pre-screen Substrate Reloading assay will first semi-quantitatively select clones with an activity of interest, whereupon an Array based assay will rank the isolated active clones in a more quantitative manner according to activity. Accordingly, in a first aspect, the present invention relates to a High Troughput Screening (HTS) Method for a microbi- ally produced material of interest, comprising sequentially performing at least two different HTS-assays chosen from the group of HTS-assays consisting of FACS-assays, Array based as- says, Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays.

In a second aspect the present invention relates to a High Troughput Screening (HTS) Method for a microbially produced material of interest, the method comprising the steps of: a) choosing at least two different HTS assays from the group of HTS-assays consisting of FACS-assays, Array based assays, Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays; and b) performing the assays chosen in step a) in a sequential manner in the following prioritized order: 1) Substrate Re- placement assays or Substrate Reloading assays; 2) FACS- assays or Colony Picking assays; and 3) Array based assays. And finally in a third aspect the present invention relates to a High Troughput Screening (HTS) Method for a microbi- ally produced material of interest, the method comprising the steps of: a) choosing at least two different HTS assays from the group of HTS-assays consisting of FACS-assays, Array based assays, Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays; and b) performing the assays chosen in step a) in a sequential manner in the following prioritized order: 1) FACS-assays or Colony Picking assays; 2) Substrate Replacement assays or Substrate Reloading assays; and 3) Array based assays.

Detailed description of the invention

Definitions of the terms associated with the aspects and embodiments of the invention are found below or by illustration of the invention in the non-limiting examples.

A preferred embodiment relates to a method of the second or third aspects, wherein at least two assays chosen in step a) belong to the same order of priority in step b) and where those assays belonging to the same order of step b) are performed sequentially in no particular order.

A further preferred embodiment relates to all aspects, wherein an assay chosen is performed multiple times before another different chosen assay is performed.

Yet another preferred embodiment relates to all aspects, wherein the material of interest is a polynucleotide, preferably DNA, more preferably cDNA, or wherein the material of in- terest is a peptide or a polypeptide, preferably an antimicrobial peptide, a growth promoting peptide, a neuropeptide, or a pharmaceutical peptide, and more preferably an enzyme, most preferably selected from the group consisting of proteases, cellulases (endoglucanases) , β-glucanases, hemicellulases, lipases, peroxidases, laccases, α-amylases, glucoamylases, cuti- nases, pectinases, reductases, oxidases, phenoloxidases, ligni- nases, pullulanases, pectate lyases, xyloglucanases, xylanases, pectin acetyl esterases, polygalacturonases, rhamnogalacturo- nases, pectin lyases, mannanases, pectin methylesterases, cel- lobiohydrolases, transglutaminases and phytases . Another preferred embodiment relates to all aspects of the invention, wherein the material of interest is a small molecule, a pharmaceutical compound, a primary or secondary metabolite, or a cellular component.

Further a preferred embodiment relates to all aspects of the invention, wherein the material of interest originates from or is produced in bacterial cells, preferably the bacterial cells belong to a strain selected from the group consisting of the species Bacillus alkalophilus , Bacillus agaradhaerenε , Bacillus amyloliquefaciens, Bacillus brevis, Bacillus clausii , Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacil lus lentus, Bacillus licheniformis , Bacillus megaterium, Bacil lus stearothermophilus, Bacillus subtilis, Bacillus thuringien- sis, Streptomyces lividans and Streptomyces murinus .

A preferred embodiment relates to all aspects of the in- vention, wherein the material of interest originates from or is produced in fungal cells, preferably the fungal cells belong to a strain selected from the group consisting of the genera Acre- monium, Aspergillus , Fusarium, Humicola, Myceliophthora, Neuro- spora, Penicillium, Thielavia, Tolypocladium, Trichoderma, Eu- penicillium, Emericella, Eurotium, Allomyces, Blastocladiella, Coelomomyces, Achlya, Candida, Al ternaria, Rhizopus and Mucor; preferably the species Aspergillus awamori , Aspergillus foe- tidus, Aspergillus japonicus , Aspergillus niger, Aspergillus nidulanε or Aspergillus oryzae. Another preferred embodiment relates to all aspects of the invention, where at least a Substrate Reloading assay and an Array based assay is chosen, preferably the material of interest is an amylase, and the Array based assay is an Array assay for detergent α-amylase.

A final preferred embodiment relates to all aspects of 5 the invention, where at least a Colony Picking assay and an Array based assay is chosen, preferably where the material of interest is an amylase, and the Array based assay is an Array assay for detergent α-amylase.

10 Definitions

In accordance with the present invention there may be employed conventional molecular biology, microbiology, and recombinant DNA techniques within the skill of the art. Such techniques are explained fully in the literature. See, e.g., Sam- is brook, Fritsch & Maniatis, Molecular Cloning: A Laboratory Manual , Second Edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York (herein "Sambrook et al . , 1989") DNA Cloning: A Practical Approach, Volumes I and II /D.N. Glover ed. 1985) ; Oligonucleotide Synthesis (M. J. Gait ed. 20 1984); Nucleic Acid Hybridization (B.D. Hames & S.J. Higgins eds (1985)); Transcription And Translation (B.D. Hames & S.J. Higgins, eds. (1984)); Animal Cell Cul ture (R.I. Freshney, ed. (1986)); Immobilized Cells And Enzymes (IRL Press, (1986)); B. Perbal, A Practical Guide To Molecular Cloning (1984) . 25 When applied to a protein, the term "isolated" indicates that the protein is found in a condition other than its native environment, such as apart from blood and animal tissue. In a preferred form, the isolated protein is substantially free of other proteins, particularly other proteins of animal origin. 30 It is preferred to provide the proteins in a highly purified form, i.e., greater than 95% pure, more preferably greater than 99% pure. When applied to a polynucleotide molecule, the term "isolated" indicates that the molecule is removed from its natural genetic milieu, and is thus free of other extraneous or 35 unwanted coding sequences, and is in a form suitable for use within genetically engineered protein production systems. Such isolated molecules are those that are separated from their natural environment and include cDNA and genomic clones. Isolated DNA molecules are free of other genes with which they are ordinarily associated, and may include naturally occurring 5' and 3' un-translated regions such as promoters and terminators. The identification of associated regions will be evident to one of ordinary skill in the art (see for example, Dynan and Tijan, Nature 316: 774-78, 1985) .

The term "polynucleotide" denotes a single- or double- stranded polymer of deoxyribonucleotide or ribonucleotide bases read from the 5' to the 3' end. Polynucleotides include RNA and DNA, and may be isolated from natural sources, synthesized in vi tro, or prepared from a combination of natural and synthetic molecules . The term "nucleic acid molecule" refers to the phosphate ester polymeric form of ribonucleosides (adenosine, guanosine, uridine or cytidine; "RNA molecules") or deoxyribonucleosides

(deoxyadenosine, deoxyguanosine, deoxythymidine, or deoxy- cytidine,- "DNA molecules") in either single stranded form, or a double-stranded helix. Double stranded DNA-DNA, DNA-RNA and RNA-RNA helices are possible. The term nucleic acid molecule, and in particular DNA or RNA molecule, refers only to the primary and secondary structure of the molecule, and does not limit it to any particular tertiary or quaternary forms. Thus, this term includes double-stranded DNA found, inter alia , in linear or circular DNA molecules (e.g., restriction fragments), plasmids, and chromosomes. In discussing the structure of particular double-stranded DNA molecules, sequences may be described herein according to the normal convention of giving only the sequence in the 5' to 3' direction along the non- transcribed strand of DNA (i.e., the strand having a sequence homologous to the mRNA) . A "recombinant DNA molecule" is a DNA molecule that has undergone a molecular biological manipulation. The term DNA "coding sequence" denotes a double-stranded DNA sequence which is transcribed and translated into a poly- peptide in a cell in vi tro or in vivo when placed under the control of appropriate regulatory sequences. The boundaries of the coding sequence are determined by a start codon at the 5' (amino) terminus and a translation stop codon at the 3' (car- boxyl) terminus. A coding sequence can include, but is not limited to, prokaryotic sequences, cDNA from eukaryotic mRNA, ge- nomic DNA sequences from eukaryotic (e.g., mammalian) DNA, and even synthetic DNA sequences. If the coding sequence is intended for expression in a eukaryotic cell, a polyadenylation signal and transcription termination sequence will usually be located 3' to the coding sequence.

The variant or host cells or variant or host colonies

The present invention relates to screening of variant or host cells and recombinant variant or host cells comprising a DNA sequence of interest or nucleic acid sequence which are advantageously used in the recombinant production of the material of interest. The term "variant cell" or "host cell" encompasses any progeny of a parent cell which is not identical to the parent cell due to mutations that occur during replication. The cell is preferably transformed with a vector comprising a nucleic acid sequence followed by integration of the vector into the host chromosome .

Transformation means introducing a vector comprising a nucleic acid sequence into a host cell so that the vector is maintained as a chromosomal integrant or as a self-replicating extra-chromosomal vector. Integration is generally considered to be an advantage as the nucleic acid sequence is more likely to be stably maintained in the cell. Integration of the vector into the host chromosome may occur by homologous or non- homologous recombination as described above.

The choice of a variant or host cell will to a large extent depend upon the gene encoding the polypeptide and its source. The host cell may be a unicellular microorganism, e . g. , a pro- karyote, or a non-unicellular microorganism, e.g., a eukaryote. Useful unicellular cells are bacterial cells such as gram posi- tive bacteria including, but not limited to, a Bacillus cell, e. g. Bacillus alkalophilus , Bacillus agaradhaerens , Bacillus amyloliquefaciens, Bacillus brevis, Bacillus clausii , Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacillus len- tus, Bacillus licheniformis, Bacillus megaterium, Bacillus stearothermophilus, Bacillus subtilis, and Bacillus thuringien- sis; or a Streptomyces cell, e. g. , Streptomyces lividans or Streptomyces murinus, or gram negative bacteria such as E. coli and Pseudomonas sp . The transformation of a bacterial variant or host cell may, for instance, be effected by protoplast transformation

(see, e . g. , Chang and Cohen, 1979, Molecular General Genetics

168:111-115), by using competent cells (see, e . g. , Young and

Spizizin, 1961, Journal of Bacteriology 81:823-829, or Dubnar and Davidoff-Abelson, 1971, Journal of Molecular Biology 56:209-221), by electroporation (see, e . g. , Shigekawa and Dower, 1988, Biotechniques 6:742-751), or by conjugation (see, e . g. , Koehler and Thorne, 1987, Journal of Bacteriology 169:5771-5278) . The variant or host cell may be a eukaryote, such as a mammalian cell, an insect cell, a plant cell or a fungal cell.

Useful mammalian cells include Chinese hamster ovary (CHO) cells, HeLa cells, baby hamster kidney (BHK) cells, COS cells, or any number of other immortalized cell lines avail- able, e . g. , from the American Type Culture Collection.

Examples of suitable mammalian cell lines are the COS

(ATCC CRL 1650 and 1651), BHK (ATCC CRL 1632, 10314 and 1573,

ATCC CCL 10), CHL (ATCC CCL39) or CHO (ATCC CCL 61) cell lines.

Methods of transfecting mammalian cells and expressing DNA se- quences introduced in the cells are described in e.g. Kaufman and Sharp, J. Mol. Biol. 159 (1982), 601 - 621; Southern and Berg, J. Mol. Appl . Genet. 1 (1982), 327 - 341; Loyter et al . , Proc. Natl. Acad. Sci. USA 79 (1982), 422 - 426; Wigler et al . , Cell 14 (1978), 725; Corsaro and Pearson, Somatic Cell Genetics 7 (1981), 603, Ausubel et al . , Current Protocols in Molecular Biology, John Wiley and Sons, Inc., N.Y., 1987, Hawley-Nelson et al., Focus 15 (1993), 73; Ciccarone et al . , Focus 15 (1993), 80; Graham and van der Eb, Virology 52 (1973), 456; and Neumann et al., EMBO J. 1 (1982), 841 - 845.

The variant or host cell may be a fungal cell. "Fungi" as used herein includes the phyla Ascomycota, Basidiomycota, Chytridiomycota, and Zygomycota (as defined by Hawksworth et al . , In , Ainsworth and Bi sby' s Di ctionary of The Fungi , 8th edition, 1995, CAB International, University Press, Cambridge, UK) as well as the Oomycota (as cited in Hawksworth et al . , 1995, supra, page 171) and all mitosporic fungi (Hawksworth et al . , 1995, supra) . Representative groups of Ascomycota include, e . g . , Neuroεpora , Eupeni ci lli um ( = Penicillium) , Emeri cella ( =Aspergillus) , Eurotium (= spergillus) , and the true yeasts listed above. Examples of Basidiomycota include mushrooms, rusts, and smuts. Representative groups of Chytridiomycota include, e . g. , Allomyces, Blastocladiella , Coelo- momyces, and aquatic fungi. Representative groups of Oomycota include, e . g. , Saprolegniomycetous aquatic fungi (water molds) such as Achlya . Examples of mitosporic fungi include Aspergil - lus, Penicilli um, Candida , and Al ternaria . Representative groups of Zygomycota include, e . g. , Rhizopus and Mucor.

A fungal variant or host cell may also be a yeast cell. "Yeast" as used herein includes ascosporogenous yeast (Endomy- cetales) , basidiosporogenous yeast, and yeast belonging to the Fungi Imperfecti (Blastomycetes) . The ascosporogenous yeasts are divided into the families Spermophthoraceae and Saccharomy- cetaceae. The latter is comprised of four subfamilies, Schizosaccharomycoideae ( e . g. , genus Schizosaccharomyces) , Nad- sonioideae, Lipomycoideae, and Saccharomycoideae (e.g., genera Pichia , Kluyveromyces and Saccharomyces) . The basidiosporogenous yeasts include the genera Leucospori dim, Rhodosporidi um, Sporidiobolus, Filobasidium, and Filobasidiella . Yeasts belonging to the Fungi Imperfecti are divided into two families, Sporobolomycetaceae (e.g., genera Sorobolomyces and Bullera) and Cryptococcaceae (e.g., genus Candida) . Since the classification of yeast may change in the future, for the purposes of this invention, yeast shall be defined as described in Biology and Activi ties of Yeast (Skinner, F.A., Passmore, S.M., and Davenport, R.R., eds, Soc. App . Bacteriol . Symposium Series No. 9, 1980. The biology of yeast and manipulation of yeast genet- 5 ics are well known in the art (see, e.g., Biochemistry and Genetics of Yeast, Bacil, M., Horecker, B.J., and Stopani, A.O.M., editors, 2nd edition, 1987; The Yeasts , Rose, A.H., and Harrison, J.S., editors, 2nd edition, 1987; and The Molecular Biology of the Yeast Saccharomyces, Strathern et al . , editors,

10 1981) .

The yeast variant or host cell may be selected from a cell of a species of Candida , Kluyveromyces, Saccharomyces , Schizosac- charomyces, Candida , Pichia , Hansehula , or Yarrowia . Useful yeast host cells are Saccharomyces carlsbergensis, Saccharomy-

15 ces cerevisiae, Saccharomyces diastaticus, Saccharomyces doug- lasii , Saccharomyces kluyveri , Saccharomyces norbensis or Saccharomyces oviformis cell. Other useful yeast host cells are a Kluyveromyces lactis Kluyveromyces fragilis Hansehula polymor- pha, Pichia pastoris Yarrowia lipolytica , Schizosaccharomyces

20 pombe, Ustilgo maylis, Candida mal tose, Pichia guillermondii and Pichia methanolio cell (cf. Gleeson et al . , J. Gen. Micro- biol. 132, 1986, pp. 3459-3465; US 4,882,279 and US 4,879,231).

The fungal variant or host cell may be a filamentous fungal cell. "Filamentous fungi" include all filamentous forms of

25 the subdivision Eumycota and Oomycota (as defined by Hawksworth et al . , 1995, supra) . The filamentous fungi are characterized by a vegetative mycelium composed of chitin, cellulose, glucan, chitosan, mannan, and other complex polysaccharides. Vegetative growth is by hyphal elongation and carbon catabolism is obliga- 0 tely aerobic. In contrast, vegetative growth by yeasts such as Saccharomyces cerevisiae is by budding of a unicellular thallus and carbon catabolism may be fermentative. The filamentous fungal host cell can be a cell of a species of, but not limited to, Acremonium, Aspergillus, Fusarium, Humicola , Mucor, My- 5 celiophthora , Neurospora , Penicillium, Thielavia , Tolypocla - dium, and Trichoderma or a teleomorph or synonym thereof. Particularly useful filamentous fungal variant or host cells are Aspergillus awamori , Aspergillus foetidus , Aspergil lus japonicus, Aspergillus niger, Aspergillus nidulans or Aspergillus oryzae . The use of Aspergillus spp. for the expres- sion of proteins is described in, e.g., EP 272 277, EP 230 023. Fungal cells may be transformed by a process involving protoplast formation, transformation of the protoplasts, and regeneration of the cell wall in a manner known per se . Suitable procedures for transformation of Aspergillus host cells are described in EP 238 023 and Yelton et al . , 1984, Proceedings of the National Academy of Sciences USA 81:1470-1474. A suitable method of transforming Fusarium species is described by Malardier et al . , 1989, Gene 78:147-156. Examples of other fungal cells are cells of filamentous fungi, e.g. Aspergillus spp., Neurospora spp., Fusarium spp. or Trichoderma spp., in particular strains of A . oryzae, A . nidulans or A . niger. The use of Aspergillus spp. for the expression of proteins is described in, e.g., EP 272 277, EP 230 023.

Yeast may be transformed using the procedures described by Becker and Guarente, In Abelson, J.N. and Simon, M.I., editors, Guide to Yeast Genetics and Molecular Biology, Methods in Enzymology, Volume 194, pp 182-187, Academic Press, Inc., New York; Ito et al . , 1983, Journal of Bacteriology 153:163; and Hinnen et al . , 1978, Proceedings of the Na tional Academy of Sciences USA 75:1920. Mammalian cells may be transformed by direct uptake using the calcium phosphate precipitation method of Graham and Van der Eb (1978, Virology 52:546) .

Transformation of insect cells and production of het- erologous polypeptides therein may be performed as described in US 4,745,051; US 4, 775, 624; US 4,879,236; US 5,155,037; US 5,162,222; EP 397,485) all of which are incorporated herein by reference. The insect cell line used as the host may suitably be a Lepidoptera cell line, such as Spodoptera frugiperda cells or Trichoplusia ni cells (cf. US 5,077,214). Culture conditions may suitably be as described in, for instance, WO 89/01029 or WO 89/01028, or any of the aforementioned references. A recombinant vector into which DNA (coding for a desired polypeptide produced by the variant or host cell) is inserted may be any vector which may conveniently be subjected to recombinant DNA procedures, and the choice of vector will often depend on the host cell into which it is to be introduced. Thus, the vector may be an autonomously replicating vector, i.e. a vector that exists as an extrachromosomal entity, the replication of which is independent of chromosomal replication, e.g. a plasmid. Alternatively, the vector may be one which, when introduced into a host cell, is integrated into the host cell genome and replicated together with the chromosome (s) into which it has been integrated.

The vector is preferably an expression vector in which the DNA sequence encoding the desired polypeptide is operably linked to additional segments required for transcription of the DNA. In general, the expression vector is derived from plasmid or viral DNA, or may contain elements of both. The term, "operably linked" indicates that the segments are arranged so that they function in concert for their intended purposes, e.g. transcription initiates in a promoter and proceeds through the DNA sequence coding for the polypeptide.

The promoter may be any DNA sequence which shows transcriptional activity in the host cell of choice and may be derived from genes encoding proteins either homologous or heterologous to the host cell.

Examples of suitable promoters for directing the transcription of the DNA encoding the polypeptide of the invention in mammalian cells are the SV40 promoter (Subramani et al., Mol. Cell Biol. 1 (1981), 854 -864), the MT-1 (metallothionein gene) promoter (Palmiter et al . , Science 222

(1983), 809 - 814) or the adenovirus 2 major late promoter.

An example of a suitable promoter for use in insect cells is the polyhedrin promoter (US 4,745,051; Vasuvedan et al . , FEBS

Lett. 311, (1992) 7 - 11), the P10 promoter (J.M. Vlak et al . , J. Gen. Virology 69, 1988, pp. 765-776), the Autographa californica polyhedrosis virus basic protein promoter (EP 397 485) , the baculovirus immediate early gene 1 promoter (US 5,155,037; US 5,162,222), or the baculovirus 39K delayed-early gene promoter (US 5,155,037; US 5,162,222).

Examples of suitable promoters for use in yeast host cells include promoters from yeast glycolytic genes (Hitzeman et al., J. Biol. Chem. 255 (1980), 12073 - 12080; Alber and Kawasaki, J. Mol. Appl . Gen. 1 (1982) , 419 - 434) or alcohol dehydrogenase genes (Young et al . , in Genetic Engineering of Microorganisms for Chemicals (Hollaender et al, eds.), Plenum Press, New York, 1982), or the TPI1 (US 4,599,311) or ADH2-4c (Russell et al . , Nature 304 (1983), 652 - 654) promoters.

Examples of suitable promoters for use in filamentous fungus variant or host cells are, for instance, the ADH3 promoter (McKnight et al . , The EMBO J. 4 (1985), 2093 - 2099) or the tpiA promoter. Examples of other useful promoters are those derived from the gene encoding A . oryzae TAKA amylase, Rhizomucor miehei aspartic proteinase, A . niger neutral a- amylase, A . niger acid stable a-amylase, A . niger or A . awamori glucoamylase (gluA) , Rhizomucor miehei lipase, A . oryzae alkaline protease, A . oryzae triose phosphate isomerase or A . nidulans acetamidase . Preferred are the TAKA-amylase and gluA promoters .

Examples of suitable promoters for use in bacterial variant or host cells include the promoter of the Bacillus stearothermophilus maltogenic amylase gene, the Bacillus licheniformis alpha-amylase gene, the Bacillus amyloliquefaciens BAN amylase gene, the Bacillus εubtilis alkaline protease gen, or the Bacillus pumilus xylosidase gene, or by the phage Lambda P_R or P_L promoters or the E. coli lac, trp or tac promoters.

The DNA may also, if necessary, be operably connected to a suitable terminator, such as the human growth hormone terminator (Palmiter et al . , op. cit . ) or (for fungal hosts) the TPI1 (Alber and Kawasaki, op. cit. ) or ADH3 (McKnight et al . , op . cit . ) terminators. The vector may further comprise elements such as polyadenylation signals (e.g. from SV40 or the adenovirus 5 Elb region) , transcriptional enhancer sequences (e.g. the SV40 enhancer) and translational enhancer sequences (e.g. the ones encoding adenovirus VA RNAs) .

The recombinant vector may further comprise a DNA sequence enabling the vector to replicate in the variant or host cell in question. An example of such a sequence (when the variant or host cell is a mammalian cell) is the SV40 origin of replication.

When the variant or host cell is a yeast cell, suitable sequences enabling the vector to replicate are the yeast plasmid 2m replication genes REP 1-3 and origin of replication. When the variant or host cell is a bacterial cell, sequences enabling the vector to replicate are legio in the art The vector may also comprise a selectable marker, e.g. a gene the product of which complements a defect in the host cell, such as the gene coding for dihydrofolate reductase (DHFR) or the Schizosaccharomyces pombe TPI gene (described by P.R. Russell, Gene 40, 1985, pp. 125-130), or one which confers resistance to a drug, e.g. ampicillin, kanamycin, tetracyclin, chloramphenicol, neomycin, hygromycin or methotrexate. For filamentous fungi, selectable markers include amdS, pyrG, argB, niaD, and sC .

To direct a polypeptide encoded by the DNA into the secretory pathway of the variant or host cells, a secretory signal sequence (also known as a leader sequence, prepro sequence or pre sequence) may be provided in the recombinant vector. The secretory signal sequence is joined to the DNA sequence encoding the polypeptide in the correct reading frame. Secretory signal sequences are commonly positioned 5¹ to the DNA sequence encoding the polypeptide. The secretory signal sequence may be that normally associated with the polypeptide or may be from a gene encoding another secreted protein.

For secretion from yeast cells, the secretory signal sequence may encode any signal peptide which ensures efficient direction of the expressed polypeptide into the secretory pathway of the cell. The signal peptide may be naturally occurring signal peptide, or a functional part thereof, or it may be a synthetic peptide. Suitable signal peptides have been found to be the a-factor signal peptide (cf. US 4,870,008), the signal peptide of mouse salivary amylase (cf. 0. Hagenbuchle et 5 al . , Nature 289, 1981, pp. 643-646), a modified carboxypeptidase signal peptide (cf. L.A. Vails et al . , Cell 48, 1987, pp. 887-897), the yeast BAR1 signal peptide (cf. WO 87/02670) , or the yeast aspartic protease 3 (YAP3) signal peptide (cf. M. Egel-Mitani et al . , Yeast 6, 1990, pp. 127-

10 137) .

For efficient secretion in yeast, a sequence encoding a leader peptide may also be inserted downstream of the signal sequence and upstream of the DNA sequence encoding the polypeptide. The function of the leader peptide is to allow the is expressed polypeptide to be directed from the endoplasmic reticulum to the Golgi apparatus and further to a secretory vesicle for secretion into the culture medium (i.e. exportation of the polypeptide across the cell wall or at least through the cellular membrane into the periplasmic space of the yeast

20 cell) . The leader peptide may be the yeast a-factor leader (the use of which is described in e.g. US 4,546,082, EP 16 201, EP 123 294, EP 123 544 and EP 163 529) . Alternatively, the leader peptide may be a synthetic leader peptide, which is to say a leader peptide not found in nature. Synthetic leader peptides

25 may, for instance, be constructed as described in WO 89/02463 or WO 92/11378.

For use in filamentous fungi, the signal peptide may conveniently be derived from a gene encoding an Aspergillus sp.

30 amylase or glucoamylase, a gene encoding a Rhizomucor miehei lipase or protease, or a Humicola lanuginosa lipase. The signal peptide is preferably derived from a gene encoding A . oryzae TAKA amylase, A . niger neutral a-amylase, A . niger acid- stable amylase, or A . niger glucoamylase.

35 For use in insect cells, the signal peptide may conveniently be derived from an insect gene (cf. WO 90/05783), such as the lepidopteran Manduca sexta adipokinetic hormone precursor signal peptide (cf. US 5,023,328).

The procedures used to ligate the DNA, the promoter and optionally the terminator and/or secretory signal sequence, re- spectively, and to insert them into suitable vectors containing the information necessary for replication, are well known to persons skilled in the art (cf., for instance, Sambrook et al . , op. cit.) .

It is also within the scope of the present invention to employ transgenic animal technology to produce the variant cells or the polypeptide or small molecule of interest. A transgenic animal is one in whose genome a heterologous DNA sequence has been introduced. In particular, a polypeptide of the invention may be expressed in the mammary glands of a non- human female mammal, in particular one which is known to produce large quantities of milk. Examples of preferred mammals are livestock animals such as goats, sheep and cattle, although smaller mammals such as mice, rabbits or rats may also be employed. The DNA sequence of interest may be introduced into the animal by any one of the methods previously described for the purpose. For instance, to obtain expression in a mammary gland, a transcription promoter from a milk protein gene is used. Milk protein genes include the genes encoding casein (cf. US 5,304,489), beta-lactoglobulin, alpha-lactalbumin and whey acidic protein. The currently preferred promoter is the beta- lactoglobulin promoter (cf. Whitelaw et al., Biochem J. 286, 1992, pp. 31-39) .

It is generally recognized in the art that DNA sequences lacking introns are poorly expressed in transgenic animals in comparison with those containing introns (cf. Brinster et al . , Proc. Natl. Acad. Sci. USA 85, 1988, pp. 836-840; Palmiter et al., Proc. Natl. Acad. Sci. USA 88, 1991, pp. 478-482; Whitelaw et al . , Transgenic Res. 1, 1991, pp. 3-13; WO 89/01343; WO 91/02318) . For expression in transgenic animals, it is therefore preferred, whenever possible, to use genomic sequences containing all or some of the native introns of the DNA of interest. It may also be preferred to include at least some introns from, e.g. the beta-lactoglobulin gene. One such region is a DNA segment which provides for intron splicing and RNA polyadenylation from the 3' non-coding region of the ovine beta-lactogloblin gene. When substituted for the native 3' non- coding sequences of a gene, this segment may will enhance and stabilize expression levels of the polypeptide of interest. It may also be possible to replace the region surrounding an initiation codon with corresponding sequences of a milk protein gene. Such replacement provides a putative tissue-specific initiation environment to enhance expression.

For expression of the DNA of interest in transgenic animals it is operably linked to additional DNA sequences required for its expression to produce expression units. Such additional sequences include a promoter as indicated above, as well as sequences providing for termination of transcription and polyadenylation of mRNA. The expression unit further includes a DNA sequence encoding a secretory signal sequence operably linked to the sequence encoding the polypeptide. The secretory signal sequence may be one native to the polypeptide or may be that of another protein such as a milk protein (cf. von Heijne et al . , Nucl. Acids Res. 14, 1986, pp. 4683-4690; and US 4,873,316) . Construction of the expression unit for use in transgenic animals may conveniently be done by inserting the DNA sequence of interest into a vector containing the additional DNA sequences, although the expression unit may be constructed by essentially any sequence of ligations. It is particularly convenient to provide a vector containing a DNA sequence encoding a milk protein and to replace the coding region for the milk protein with the DNA sequence of interest, thereby creating a fusion which includes expression control sequences of the milk protein gene. The expression unit is then introduced into fertilized ova or early-stage embryos of the selected host species. Introduction of heterologous DNA may be carried out in a number of ways, including microinjection (cf. US 4,873,191), retroviral infection (cf. Jaenisch, Science 240 , 1988, pp. 1468-1474) or site-directed integration using embryonic stem 5 cells (reviewed by Bradley et al . , Bio/Technology 10, 1992, pp. 534-539) . The ova are then implanted into the oviducts or uteri of pseudopregnant females and allowed to develop to term. Offspring carrying the introduced DNA in their germ line can pass the DNA on to their progeny, allowing the development of lo transgenic herds.

General procedures for producing transgenic animals are known in the art, cf. for instance, Hogan et al . , Manipulating the Mouse Embryo: A Laboratory Manual, Cold Spring Harbor Laboratory, 1986; Simons et al . , Bio/Technology 6, 1988, pp.

15 179-183; Wall et al . , Biol. Reprod. 32, 1985, pp. 645-651; Buhler et al . , Bio/Technology 8 , 1990, pp. 140-143; Ebert et al . , Bio/Technology 6: 179-183, 1988; Krimpenfort et al . , Bio/Technology 9: 844-847, 1991, Wall et al . , J. Cell. Biochem. 49: 113-120, 1992; US 4,873,191, US 4,873,316; WO 88/00239, WO

20 90/05188; WO 92/11757 and GB 87/00458. Techniques for introducing heterologous DNA sequences into mammals and their germ cells were originally developed in the mouse. See e.g. Gordon et al . , Proc. Natl. Acad. Sci. USA 77: 7380-7384, 1980, Gordon and Ruddle, Science 214 : 1244-1246, 1981; Palmiter and

25 Brinster, Cell 41: 343-345, 1985; Brinster et al . , Proc. Natl.

Acad. Sci. USA 82: 4438-4442, 1985; and Hogan et al . (ibid.). These techniques were subsequently adapted for use with larger animals, including livestock species (see e.g., WO 88/00239, WO 90/01588 and WO 92/11757; and Simons et al . , Bio/Technology 6:

30 179-183, 1988). To summarize, in the most efficient route used to date in the generation of transgenic mice or livestock, several hundred linear molecules of the DNA of interest are injected into one of the pro-nuclei of a fertilized egg according to techniques which have become standard in the art .

35 Injection of DNA into the cytoplasm of a zygote can also be employed. Production in transgenic plants may also be employed. It has previously been described to introduce DNA sequences into plants, which sequences code for protein products imparting to the transformed plants certain desirable properties such as increased resistance against pests, pathogens, herbicides or stress conditions (cf. for instance EP 90 033, EP 131 620, EP 205 518, EP 270 355, WO 89/04371 or WO 90/02804), or an improved nutrient value of the plant proteins (cf. for instance EP 90 033, EP 205 518 or WO 89/04371). Furthermore, WO 89/12386 discloses the transformation of plant cells with a gene coding for levansucrase or dextransucrase, regeneration of the plant

(especially a tomato plant) from the cell resulting in fruit products with altered viscosity characteristics.

In the plant cell, the DNA of interest is under the control of a regulatory sequence which directs the expression of the DNA sequence in plant cells and intact plants. The regulatory sequence may be either endogenous or heterologous to the host plant cell.

The regulatory sequence may comprise a promoter capable of directing the transcription of the DNA sequence of interest in plants. Examples of promoters which may be used according to the invention are the 35s RNA promoter from cauliflower mosaic virus (CaMV) , the class I patatin gene B 33 promoter, the ST- LS1 gene promoter, promoters conferring seed-specific expression, e.g. the phaseolin promoter, or promoters which are activated on wounding, such as the promoter of the proteinase inhibitor II gene or the wunl or wun2 genes.

The promoter may be operably connected to an enhancer sequence, the purpose of which is to ensure increased transcription of the DNA of interest. Examples of useful enhancer sequences are enhancers from the 5 ' -upstream region of the 35s RNA of CaMV, the 5 ' -upstream region of the ST-LS1 gene, the 5 ' -upstream region of the Cab gene from wheat, the 5¹- upstream region of the 1'- and 2 ' -genes of the TR-DNA of the Ti plasmid pTi ACH5 , the 5 ' -upstream region of the octopine synthase gene, the 5 ' -upstream region of the leghemoglobin gene, etc.

The regulatory sequence may also comprise a terminator capable of terminating the transcription of the DNA of interest in plants. Examples of suitable terminators are the terminator of the octopine synthase gene of the T-DNA of the Ti-plasmid pTiACH5 of Agrobacterium tumefaciens, of the gene 7 of the T- DNA of the Ti plasmid pTiACH5, of the nopaline synthase gene, of the 35s RNA-coding gene from CaMV or from various plant genes, e.g. the ST-LSl gene, the Cab gene from wheat, class I and class II patatin genes, etc.

The DNA of interest may also be operably connected to a DNA sequence encoding a leader peptide capable of directing the transport of an expressed polypeptide to a specific cellular compartment (e.g. vacuoles) or to extracellular space. Examples of suitable leader peptides are the leader peptide of proteinase inhibitor II from potato, the leader peptide and an additional about 100 amino acid fragments of patatin, or the transit peptide of various nucleus-encoded proteins directed into chloroplasts (e.g. from the St-LSl gene, SS-Rubisco genes, etc.) or into mitochondria (e.g. from the ADP/ATP translocator) .

Furthermore, the DNA of interest may be modified in the 5' non- translated region resulting in enhanced translation of the sequence. Such modifications may, for instance, result in removal of hairpin loops in RNA of the 5¹ non-translated region. Translation enhancement may be provided by suitably modifying the omega sequence of tobacco mosaic virus or the leaders of other plant viruses (e.g. BMV, MSV) or of plant genes expressed at high levels (e.g. SS-Rubisco, class I patatin or proteinase inhibitor II genes from potato) .

The DNA of interest may furthermore be connected to a second DNA sequence encoding another polypeptide or a fragment thereof in such a way that expression of said DNA sequences results in the production of a fusion protein. When the host cell is a potato plant cell, the second DNA sequence may, for instance, encode patatin or a fragment thereof (such as a fragment of about 100 amino acids) .

The plant in which the DNA of interest is introduced may suitably be a dicotyledonous plant, examples of which are is a tobacco, potato, tomato, or leguminous (e.g. bean, pea, soy, alfalfa) plant. It is, however, contemplated that mono- cotyledonous plants, e.g. cereals, may equally well be transformed with the DNA.

Procedures for the genetic manipulation of mono- cotyledonous and dicotyledonous plants are well known. In order to construct foreign genes for their subsequent introduction into higher plants, numerous cloning vectors are available which generally contain a replication system for E. coli and a selectable/screenable marker system permitting the recognition of transformed cells. These vectors include e.g. pBR322, the pUC series, pACYC, M13 mp series etc. The foreign sequence may be cloned into appropriate restriction sites. The recombinant plasmid obtained in this way may subsequently be used for the transformation of E. coli. Transformed E. coli cells may be grown in an appropriate medium, harvested and lysed. The chimeric plasmid may then be reisolated and analyzed. Analysis of the recombinant plasmid may be performed by e.g. determination of the nucleotide sequence, restriction analysis, electrophoresis and other molecular-biochemical methods. After each manipulation the sequence may be cleaved and ligated to another DNA sequence. Each DNA sequence can be cloned on a separate plasmid DNA. Depending on the way used for transferring the foreign DNA into plant cells other DNA sequences might be of importance. In case the Ti-plasmid or the Ri plasmid of Agrobacterium tumefaciens or Agrobacterium rhizogenes, at least the right border of the T-DNA may be used, and often both the right and the left borders of the T-DNA of the Ri or Ti plasmid will be present flanking the DNA sequence to be transferred into plant cells. The use of the T-DNA for transferring foreign DNA into plant cells has been described extensively in the prior literature (cf. Gasser and Fraley, 1989, Science 244, 1293 - 1299 and references cited therein) . After integration of the foreign DNA into the plant genome, this sequence is fairly stable at the original locus and is usually not lost in subsequent mitotic or meiotic divisions. As a general rule, a selectable marker gene will be co-transferred in addition to the gene to be transferred, which marker renders the plant cell resistant to certain antibiotics, e.g. kanamycin, hygromycin, G418 etc. This marker permits the recognition of the transformed cells containing the DNA sequence to be transferred compared to non-transformed cells.

Numerous techniques are available for the introduction of DNA into a plant cell. Examples are the Agro-bacter um mediated transfer, the fusion of protoplasts with liposomes containing the respective DNA, microinjection of foreign DNA, electroporation etc. In case Agrobacterium mediated gene transfer is employed, the DNA to be transferred has to be present in special plasmids which are either of the intermediate type or the binary type. Due to the presence of sequences homologous to T-DNA sequences, intermediate vectors may integrate into the Ri- or Ti-plasmid by homologous recombination. The Ri- or Ti- plasmid additionally contains the vir-region which is necessary for the transfer of the foreign gene into plant cells. Intermediate vectors cannot replicate in Agrobacterium species and are transferred into Agrobacterium by either direct transformation or mobilization by means of helper plasmids (conjugation). (Cf. Gasser and Fraley, op. cit. and references cited therein) .

Binary vectors may replicate in both Agrobacterium species and E. coli. They may contain a selectable marker and a poly-linker region which to the left and right contains the border sequences of the T-DNA of Agrobacterium rhizogenes or Agro-bac erium tumefaciens . Such vectors may be transformed directly into Agrobacterium species. The Agrobacterium cell serving as the host cell has to contain a vir-region on another plasmid. Additional T-DNA sequences may also be contained in the Agrobacterium cell.

The Agrobacterium cell containing the DNA sequences to be transferred into plant cells either on a binary vector or in the form of a co- integrate between the intermediate vector and the T-DNA region may then be used for transforming plant cells. Usually either multicellular explants (e.g. leaf discs, stem segments, roots), single cells (protoplasts) or cell suspensions are co-cultivated with Agrobacterium cells containing the DNA sequence to be transferred into plant cells. The plant cells treated with the Agrobacterium cells are then selected for the co-transferred resistance marker (e.g. kanamycin) and subsequently regenerated to intact plants. These regenerated plants will then be tested for the presence of the DNA sequences to be transferred.

If the DNA is transferred by e.g. electroporation or microinjection, no special requirements are needed to effect transformation. Simple plasmids e.g. of the pUC series may be used to transform plant cells. Regenerated transgenic plants may be grown normally in a greenhouse or under other conditions. They should display a new phenotype (e.g. production of new proteins) due to the transfer of the foreign gene(s) . The transgenic plants may be crossed with other plants which may either be wild-type or transgenic plants transformed with the same or another DNA sequence. Seeds obtained from transgenic plants should be tested to assure that the new genetic trait is inherited in a stable Mendelian fashion. See also Hiatt, Nature 344: 469-479, 1990; Edelbaum et al . , J. Interferon Res. 12: 449-453, 1992; Sijmons et al . , Bio/Technology 8: 217-221, 1990: and EP 255 378.

EXAMPLES

By way of non- limiting examples we illustrate the invention, a HTS-methodology for a microbially produced material of interest. The material of interest could be any microbially produced material, for example polypeptides, enzymes, primary or secondary metabolites, small molecules, antimicrobial molecules, growth promoting/inhibiting molecules, pharmaceutically active molecules etc.

The methodology is based on combining or sequentially performing several different HTS-assays chosen from the group of HTS-assays consisting of FACS-assays, Microtiter Plate based assays, Colony-picking assays, Substrate-replacement assays and

Substrate reloading assays.

Non-limiting examples of each assay type of the above group are shown below.

Preferably at least two assays of different type are performed in a sequential manner in the following order: 1) Substrate Replacement or Substrate Reloading assay; 2) FACS-assay or Colony Picking assay; 3) MTP-based assay. However if more than two assays are performed sequentially two of the assays can be of the same order, e.g. if three are performed, they could be Colony picking followed by FACS followed by an MTP- based assay.

Example 1

This example shows HTS assays based on Substrate Replacement. Substrate Replacement assays are categorized as assays screening for catalysts or enzymes of interest coupled to a substrate by an exchangeable linker pair, X and Y, and a selection method that uses multiple catalytic turnover events to isolate the more active of the catalysts in said library. The assay screens samples comprising a library of catalyst molecules provided in the form of individual units, wherein the individual units comprise a first type individual unit having the following general structure: C-XY-S, wherein C denotes a catalyst molecule, XY an XY exchange pair, and S a substrate which is capable of being catalysed into a product by at least one catalyst comprised within said library of catalyst molecules and thereby providing the possibility of obtaining a second type individual unit comprising the general structure : C-XY-P, wherein C and XY has the meaning defined above and P is the product molecule resulting from the catalytic conversion of the substrate S of the first type individual unit.

The catalytic activity as such is not important for the methodology of the invention, only for the choice of specific setup conditions in the individual assays, e.g. choice of sub- strate.

Substrate Replacement by XY-exchange pairs

Substrate Replacement in this case is based on the use of phage-displaying of the lipase, and a pre-enrichment is done first in order to first select phages display display the protein of interest properly. Non-limiting examples are suggested further below of other functional XY-exchange pairs.

Pre-enrichment of phages displaying His-tagged proteins . Certain proteins are difficult to display on filamentous phage. In particular, large proteins or proteins which have a toxic or growth inhibiting effect on E. coli often have low display efficiency, i.e. the majority of phage particles produced carry no pill-fusion on the surface. Display efficiencies as low as one out of a thousand phages displaying the fusion protein have been reported (Jestin et al . , 1999, Angew. Chemi . Int. Ed., vol. 38, pp. 1124-1127; Demartis et al . , 1999, JMB, vol. 286, pp. 617-633). In such cases, a high non-specific background is expected, because of the large excess of phage particles carrying the DNA encoding the pill fusion, but not displaying the fusion on the surface. To circumvent this potential problem, we inserted a histidine tag between the pill coat protein and the enzyme, allowing the purification of phages displaying His-tagged protein by Ni-NTA column chromatography. Other tags that could have been used in a similar manner as described below for the Histidine tag include the intein- chitin binding domain fusion (Chong et al . , 1997, Gene, vol. 192, pp 271-281), FLAG peptide (Slootstra et al . , 1997, Molecular Diversity, vol. 2, pp. 156-164), and the maltose binding protein (Pryor and Leiting, 1997, Protein expression and Puri- fication, vol. 10, pp. 309-319).

Ni-NTA column purification of phages displaying the lipase- Hisβ-pIII or cellulase-His6-pIII fusion protein.

A Ni-NTA spin column (Qiagen Spin Kit) was equilibrated with 600 μL "50 mM sodium-phosphate buffer pH 8, 300 mM NaCl, 1 mM Imidazole, 0.05% BSA" (centrifuged 2 minutes at 700 G) . To 400 ml phage preparation (approximately 10¹² phage particles) was added 100 mL "250 mM sodium-phosphate buffer pH 8, 1.5 M NaCl, 0.25% BSA" and 4 μL 100 mM Imidazole, the solution loaded onto the pre-equilibrated column, and centrifuged for 4 minutes at 200 G. The column was washed twice with 600 mL "50 mM sodium-phosphate buffer pH 8, 300 mM NaCl, 20 mM Imidazole, 0.05% BSA" (centrifugation 200 G for 4 minutes and 700 G for 2 minutes, respectively) . Then the phages were eluted with 3 x 333 mL "50 mM sodium-phosphate buffer pH 8, 300 mM NaCl, 250 mM Imidazole, 0.05% BSA" (700 G, 2 minutes), and the 999 μL eluate PEG precipitated and re-suspended in 400 mL "50 mM sodium- phosphate buffer pH 8, 300 mM NaCl, 1 mM Imidazole, 0.05% BSA". The solution was loaded on a fresh spin column, and the proce- dure repeated, except that the final PEG precipitate was dissolved in 50 mL TE buffer pH 8. This procedure enriches approximately 500 fold for phage displaying His-tagged protein.

Now that we have effectively isolated phage displaying the lipase, the next step is to isolate the more effective ones.

Enrichment of ildtype lipase in a background of excess, less active lipase variants, using phage-displayed lipase and DNA oligos as the XY exchange pair. Phagemid construction. Phagemid ph8 (wildtype Lipase) and phl8 (Lipase S146A mutant) : DNA oligos (SEQ ID No 1): "Not-His6-sense" (5 ' -ggccgcaccagga- ggaggatcacatcaccatcaccatcactc-3 ' ) and (SEQ ID No 2): "Not-His6- antisense" (5' -ggccgagtgatggtgatggtgatgtgatcctcctggtgc-3 ' ) were annealed, and the double-stranded product ligated into Notl- digested pFab-SP400 (described above) and pFab-SP400-S146A

(identical to pFab-SP400, except that it carries a serine to alanine mutation at position 146, lowering its activity at least 100-fold). The resulting phagemids, ph8 (wildtype Lipase) and phl8 (Lipase S146A mutant) , carry a gene fusion comprising

(reading from the N-terminal end) the pelB signal sequence, the gene encoding mature lipase (wt or S146A mutant) , the insert with the six histidines, and the pill gene from amino acid residue 198.

Production of phage particles.

Phage particles were produced with minor modifications according to ørum et al . Briefly, E. coli XLlblue were transformed with either the phagemid ph8 (Lipase wt) or phagemid phlδ (Li- pase S146A mutant), or phl3 (a negative control; carries a His-tagged cellulase instead of Lipase, but is otherwise identical to the ph8 and phl8 constructs) . The transformed cells were shaken at 37 °C in 2xYT medium containing 100 mg/mL ampicillin, 5 mg/mL tetracyclin and 2 % glucose. At an OD₆₀₀ of 0.5, acid helper phage (a M13K07 derivative, carrying a 30'meric acid peptide extension at the N-terminus of pill, see Pedersen et al . , PNAS (1998), 95, pp. 10523-10528) was added to a final concentration of 1.5 x 10⁸ cfu/mL, and incubated at

37°C for 20 min. The cells were pelleted and resuspended in 2xYT, 5 mM IPTG (100 mM IPTG for phl3), 100 mg/mL ampicillin,

50 mg/mL kanamycin, and shaken for 3 hours at 30 °C . Cells were pelleted and phage particles in the supernatant PEG precipitated three times and re-suspended in 400 mL TE buffer (10 mM Tris-HCl pH 8 ; 1 mM EDTA) . Before covalent coupling of base-linker-X to phage (see below) , a pre-enrichment step for the phages displaying the Lipase variant was performed. The procedure generally led to< a recovery of 0.04-0.2 % of the input phage; we estimate that more than 90 % of the recovered phage display lipase.

Active, histidine-tagged Lipase is displayed on phage.

In order to verify that the prepared phages display the Lipase- pIII fusion proteins, separate wells of a microtiter plate were coated with antibodies against the histidine tag (Penta-His antibody, Qiagen) , antibodies against the Lipase protein, or, as a negative control, an antibody against an unrelated amylase enzyme. Approximately 10⁹ phages displaying non-His-tagged wild- type Lipase (ph3), His-tagged wildtype Lipase (ph8) or His- tagged Lipase S146A mutant (phl8) were added to separate wells. The results showed that ph8 and phl8 both display the His tag as expected; the non-His-tagged wildtype Lipase is not significantly bound to the anti-His antibody. Both mutant and wildtype Lipase are immobilized on anti-Lipase antibody, as expected. Finally, none of the Lipase variants are immobilized on the unrelated antibody. It is therefore concluded that the ph8 and phl8 phagemids encode Lipase-Histidine tag-pill fusions that are folded properly on the surface of phage.

The His-tagged phages (ph8 and phlδ) , were taken through the Ni-NTA column purification step described above. The enrichment step is expected to yield a phage population almost entirely consisting of phages displaying the His-tagged protein. The procedure involves two consecutive Ni-NTA column purifications; the first run reproducibly recovered 0.2-0.3% of the input, the second run recovered 10-20% of the input. These numbers are taken as an indication that the resulting phage population have been enriched dramatically with regard to displayed protein, and that the final phage population consists of phages that nearly all display the protein. Finally, in a Brilliant Green plate assay, cells containing the wildtype Lipase-His-pIII fusion but not cells containing the Lipase S146A-pIII fusion, ex- hibited Lipase activity. It is therefore tentatively concluded that ph8 display active, properly folded wildtype Lipase; the phl8-phages display properly folded Lipase S146A mutant with decreased Lipase activity.

Synthesis of base-linker-DNA conjugate ("base-linker-X"). The base-linker peptide (SEQ ID NO 3): C (GGS) ₄AQLKKKLQALKKKNAQ- LKWKLQALKKKLAQGGC was conjugated to the 5' -thiol of the DNA oligos "X-26mer" (SEQ ID NO 4) (HS-5' -attaaattagcgcaatgaa- gggcaac-3') and photodeprotected, following the protocol described in Pedersen et al . , PNAS (1998), 95, 10523-10528. The underlined sequence constitutes the X moiety. The resulting conjugate is called "base-X-26mer" .

Synthesis of Y-substrate-biotin conjugates.

The hetero-functional molecule containing a maleimide moiety at one end, a biotin at the other, and an ester that serves as a substrate for the Lipase in the middle, was prepared, ω- Aminododecanoic acid was first protected as its methyl ester which was coupled with biotin-NHS followed by hydrolysis to give biotin-acid. Convergently, Maleimide alcohol was prepared by reacting maleimide-NHS with 6-hydroxyhexylamine . Esterifica- tion afforded the target substrate. Finally the compound was conjugated to either Yl DNA oligo (SEQ ID No. 5) (5'-SH- aataaataaacgggttgcccttcatt-3 ' ) or Y2 DNA oligo (SEQ ID No. 6) (5' -ttcattgcgcttcggcaaataaataa-SH-3 ' ) . The underlined sequences constitute the Yl and Y2 exchange moieties.

Covalent attachment of base-linker-DNA conjugate to phage.

The X-26mer conjugate (see above) is covalently attached to the

Ni-NTA purified phages (see above) , for example by following the guidelines above. The phages from the coupling reaction may optionally be taken through another purification step, in order to assure a high degree of coupling. This involves annealing of the phages to streptavidin coated beads, to which a biotinylated oligo, complementary to the X-26mer, has been immobi- lized. Following several washes to remove phages that have not been covalently attached to the X-26mer, the temperature is in- creased to melt the DNA duplexes and release the coupled phages .

Isolation of the more active lipase. The Yl-substrate-biotin and Y2-substrate-biotin conjugates are mixed with a streptavidin-derivatized matrix (for example streptavidin immobilized on 4% agarose, Sigma) , at a concentration of 1-10 mM, and incubated at room temperature 1-2 hours. The column is washed, and phages to which the X-26mer conjugate has been coupled (see above) are added in a buffer that allows Lipase activity as well as efficient annealing (contains MgCl₂ and CaCl₂) at 20-35°C. Alternatively, the coupling step is performed directly on the column. The X-26mer DNA coupled to phage will anneal to the Yl- or Y2-substrate-biotin molecules and be- come immobilized on the matrix through the substrate. Phages displaying catalytically active Lipase will cleave the substrate and continue the migration through the column; upon interaction with another Yl- or Y2-substrate, an exchange reaction may take place, which will immobilize the phage again. A less catalytic Lipase will spend more time bound to a given substrate. Therefore, the catalytically more active phages will migrate faster than the less active phages, and can therefore be collected first at the bottom of the column.

DNA oligos as XY exchange units: Measurement of exchange rates by fluorescense polarization spectroscopy.

A fundamental feature of the substrate replacement protocol above is the XY exchange unit that links the substrate and enzyme. Ideally, the XY unit should provide a means for the dynamic, fast and efficient substrate reloading on the enzyme.

We wanted to design nucleic acids that would fulfill (at least partly) the requirements of the ideal XY unit : fast exchange, yet intrinsically stable XY complexes. Therefore, two sets of oligos, Set#l and Set#2, were designed. The XY com- plexes of the two sets have expected melting temperatures around 60 and 40°C, respectively. To provide a dynamic exchange, the DNA oligos were designed so that the two DNA oligos, Yl and Y2, bound to different but overlapping targets on X (see Materials and Methods) . With the expectation that the overlap would provide a faster exchange between the X-Yl and X-Y2 complexes. The exchange rates of the two sets of oligos were analyzed by fluorescense polarization spectroscopy at various temperatures. The dynamic range (the temperature range at which the oligos exchange relatively fast yet where X is complexed) was slightly lower than the predicted melting temperature of the XY du- plexes. The time required to obtain a 90% exchange of free Y for complex bound Y ( "t (_ex_cange) " ) varied between 30 and 500 seconds. It is expected that with other designs of oligos and optimized conditions (in particular Mg⁺⁺ concentration and temperature) , it should be possible to obtain exchange rates for nucleic acids faster than one per second, possibly 10-100 per second.

Materials and Methods.

DNA oligo sequences.

Set #1:

X#l (SEQ ID 7): 3 ' -tgctagcatggcccaacgggaagtaacgcgaagccgatgctag catgc-5' Y1-F1#1 (SEQ ID 8): 5 ' -Fam-cgggttgcccttcatt-3 ' Yl#l (SEQ ID 9): 5 ' -cgggttgcccttcatt-3 ' Y2#l (SEQ ID 10): 5 ' -ttcattgcgcttcggc-3 '

Set#II:

X#2 (SEQ ID 11): 3 ' -acgggaagtaacgcga-5 ' Y1-F1#2 (SEQ ID 12): 5 ' -Fam-tgcccttcatt-3 '

Yl#2 (SEQ ID 13): 5 ' -tgcccttcatt-3 '

Y2#2 (SEQ ID 14): 5 ' - ttcattgcgct-3 '

Sequences that are complementary in X and Y are in bold; the region of Yl and Y2 that overlap is underlined. FAM denotes the fluoresccent moiety. Flourescense polarization spectroscopy.

The measurements were done on a Perkin Elmer LS50 spectrophotometer. Excitation was at 485 nm, emission was recorded at 525 nm. All measurements were done in Buffer A (10 mM Tris-HCl pH 9, 100 mM NaCl, and 1 mM MgCl₂ unless noted otherwise) . A thin tubing was connected to the sample cuvette; therefore, it was not necessary to open the lid during addition of extra material. With this set-up, it is not possible to measure rates of reactions that proceed to near-completion within 30 seconds.

Results .

Design of XY exchange units based on nucleic acids.

For the final application, the XY exchange unit in the enzyme- linker XY linker-substrate structure should be in rapid equilibrium with an excess of Y-substrate molecules in the buffer, thus facilitating a rapid exchange of product (or substrate) "attached" to the enzyme, through an exchange of Y-product (or Y-substrate) with Y-substrate in the buffer. In order to accomplish this rapid equilibrium, we designed two oligos, Yl and Y2 , with overlapping (but not identical) binding sites on the X oligo (see materials and methods) . Thus, Y2 would be able to transiently interact with X in the X:Y1 complex through the portion of X that does not anneal to Yl, and vice versa. When annealing to the full complementary sequence on X, Yl and Y2 form the same number of AT and GC base pairs, and are therefore expected to have near- identical affinity for X, and presumably similar on- and off -rates. Therefore, Yl should replace Y2 from the X: Y2 complex as fast as Y2 replaces Yl.

We designed two sets of oligos, with different length of annealing site and different relative length of annealing and overlapping regions . The sequence of the overlapping region (5' -TTCATT-3 ' ) was chosen so as to avoid triplex formation; moreover, in the experiments very low concentrations of X and Y are used. Therefore, triplex formation is highly unlikely. Exchange of Y2 with Yl of the X-Yl complex.

Fluorescense polarization was used to analyze the exchange rates of each of the two sets of oligos. Fluorescense polarization of a fluorescently labelled molecule in solution is pro- portional to the molecule's rotational relaxation time. If viscosity and temperature is held constant, the fluorescense polarization value is directly proportional to the molecular volume. Changes in molecular volume may result from binding or dissociation of two molecules, as used in this study. First oligo Set#l was analyzed. Fluorescein-labelled Yl oligo (Yl-Fl#l, see Materials and Methods) at a concentration of 5 nM and temperature 46 °C gives a fluorescense polarization value of 0.028. Upon addition of a 200-fold excess (1 mM) of oligo X#l at time t=660sec, the polarization rapidly increases to a pla- teau at about 0.037, indicating the formation of the X:Y1-F1#1 complex. When Y2#l is added in a 20-fold excess to X (20mM) at t=2400, the polarization rapidly drops, indicating the release of Y1-F1#1 from the X: Yl-Fl#l complex. Most likely, the X:Y2#1 complex is formed, displacing Yl-Fl#l from X#l . The displacement of Y1-F1#1 from X#l by Y2#l is a relatively fast process; within 30 seconds, 90% equilibrium is observed. Formation of the X#1:Y1-F1#1 complex is a slower process (approximately 600 seconds for 90 % equilibrium to be obtained) . However, this is explained by the fact that the concentration of the oligo in excess is 20 fold lower.

Next, the same binding reactions were analyzed at 50°C. Less time (300 sec) is required for formation of the X: Yl complex; on the other hand, the exchange goes to 90% completion a little slower (40 sec, see table 1) . We wanted to challenge the idea that overlapping targets can speed up the exchange rate. Therefore, the X#1:Y-F1#1 complex was formed under identical conditions (46°C) , but now oligo Yl#l (instead of Y2#l) was added in a twenty-fold excess to X#l. The displacement of Yl-Fl#l by un-labelled Yl#l is slow. About 500 seconds are now required to obtain 90% equilibrium. We conclude that the principle of overlapping targets for Yl and Y2 on X in this case accelerates the exchange by a factor of approximately 17.

Oligo Set#2 was analyzed in the same way (see table 1) . At ref- erence conditions of 24°C and 1 mM MgCl₂, the exchange reaction goes to 90% completion within 200 seconds. Increasing the temperature to 30°C speeds up the exchange 4-fold; likewise, increasing the Mg⁺⁺ concentration to 10 mM increases the exchange 4-fold. The present set-up has a response time of about 30 sec- onds. Therefore, we did not attempt to measure the exchange rate at 10 mM MgCl₂ and 30 °C.

Finally, the principle of overlapping targets was again challenged. Under less than optimal conditions (24°C, 1 mM MgCl₂) , the exchange of Yl for Yl-Fl complexed to X goes to 90% comple- tion within 500 seconds; this is 2.5-fold slower than for the exchange of Y2 for Yl . Therefore, it is advantageous to include two rather than one "Y-unit", eventhough under these conditions it speeds up the exchange only 2.5-fold. The fluorescense polarization signal does not come down to the base line (the signal level for free Yl-Fl) . This observation was done several times, for both Set#l and Set#2 oligos, for exchange of Yl for Yl and Y2 for Yl, as well as with different MgCl concentrations. We have no explanation for this phenomenon; however, it did not seem to influence the measured ex- change rates. Addition of a large excess of Y2 to free Yl-Fl in the absence of X has no effect on the fluorescense polarization signal .

Discussion The exchange rates of both sets of oligos showed a strong dependence on the temperature; the temperature at which the selection experiment is performed should be held within a relatively narrow window of about +/- 5°C around the optimal temperature, in order for the exchange to be efficient. The MgCl₂ concentration had a strong effect on the exchange rate. A similar effect may have been observed by (Shimayama et al., (1995), FEBS Letters 368, 304-306), for a so-called "DNA- armed hammerhead ribozyme" . They found that the catalytic ac- tivity of the hammerhead ribozyme, in which the hybridizing arms had been replaced with deoxyribonucleotides, depended strongly on the Mg⁺⁺ concentration, even at high magnesium concentrations where the active site (requiring Mg⁺⁺ for activity) is believed to be saturated with Mg⁺⁺ . Their results might therefore be interpreted in terms of high Mg⁺⁺ concentrations speeding up the exchange of the nucleic acids arms on the RNA target .

Design of XY exchange units based on nucleic acids should be a very general way to produce fast and efficient exchange units. Appropriate choice of length and composition of annealing sites and conditions under which the selection is performed, should allow the use of DNA oligos as XY exchange units under different conditions of pH, salt, temperature, pressure etc. In the present study it was shown that optimization of either Mg++ concentration or temperature could bring the exchange rate down to the limit of the apparatus (tens of seconds) . A combination of these conditions, potentially combined with optimization of other conditions, should bring the exchange rate down to ap- proximately 1 per second. Finally, tuning of the relative length and composition of the annealing sites and overlapping regions of the Yl and Y2 oligos should provide further improvements .

The design of Yl and Y2 oligos with overlapping binding sites on the X oligo accelerated the exchange rate. Presumably the overlapping targets mediate active displacement of one oligo by the other. More sophisticated designs of XY exchange units, based on this concept and on the structural and mechanistic features of antisense RNA, should improve the dynamics of the system even more . Enrichment of cells producing the more active proteases in a background of cells producing less active protease variants.

In this example, the screening does not involve phage- display, rather the assay is based on a microbial cell producing the enzyme, a substrate attached to the surface of that cell through an XY exchange unit, and enzymes produced and secreted by the cell. Free Y-substrate in the surrounding media continuously replace substrate attached to the cell surface. Consequently, as the local concentration of secreted enzyme is much higher near the substrate attached to the cell from which it was secreted than near any of the other cells' attached substrates, there will be more product attached on a cell secreting an active enzyme than on a cell secreting a less active enzyme.

The substrate can be attached to the cell surface in many ways. For example, phospholipids, fatty acids, sterols, choles- teryl esters may be derivatized with the substrate of the target reaction. When incubated with cells, these molecules read- ily localize in the membrane interior, and expose the substrate on the surface of the cell. Alternatively, the substrate may be derivatized with crosslinking reagents that react with the surface constituents. Finally, the substrate may be derivatized with structures (e.g. proteins, antibodies) that bind to mem- brane components such as polysaccharides or membrane proteins .

The principle is here examplified in the case where the individual unit consists of a cell (for example Bacillus) , attached substrate (peptide) , and secreted enzyme (protease) . The His6-metal-IDA or His6-metal-NTA complex is used as XY exchange unit. The selection is performed in the column format. The column matrix is coated with peptides carrying a target sequence for the protease. The peptides are attached to the column matrix at one end, and carry a polyhistidine (His6) tag at the other. The experiment is performed as follows. Bacillus cells, secreting the protease of interest (for example c-component from Bacillus licheniformis or the commercial Savinase protease), are harvested in the exponential growth phase, resuspended in appropriate buffer and incubated with a bi- functional molecule that will cross-link to cell surface components. The bi-functional molecule may be a N- hydroxysuccinimide (NHS) moiety, linked to an iminodiacetic acid (IDA) or nitrilotriacetic acid (NTA) moiety. NHS reacts with primary amines on the cell surface, which covalently anchors the IDA or NTA moiety to the cell surface. A suitable column matrix for the separation of cells (for example Sepharose or Sephadex) is coated or derivatized with a peptide target for the protease of interest (for example, in the case of the c-component, the sequence IELSEPIGNTVCHHHHHH) . At one end the peptide carries a polyhistidine extension. At the other end it is attached to the column matrix (for example through reaction of the N-terminal amine with NHS-activated Sepharose, Pharmacia Biotech) .

The IDA- or NTA-modified cells are loaded on the peptide- modified column, in appropriate buffer (for example 2xYT or LB medium, with added Ni⁺⁺, Zn⁺⁺, Cu⁺⁺, or Co⁺⁺; temperature 35- 50°C) . The flow-rate is kept below 0.5 mL/min.

The complex His6-metal-IDA (or His6-metal-NTA) forms, and thus the protease substrate becomes attached to the cell. If the cell secretes active proteases, these cleave the target, and release the cell from the column matrix; also, the His6- metal-IDA (or His6-metal-NTA) complexes are in rapid equilibrium, resulting in continuous replacement of the substrate or product with new substrate. Consequently, the cells secreting the more efficient enzymes, or cells that secrete more of an active enzyme, elute first at the bottom of the column.

The principle is tested in the following way. Cells secreting a number of protease variants, or that secretes the same protease in varying amounts, are treated as described above, and the substrate reloading protocol performed as described. Cells secreting the more efficient protease, or secreting most of the protease, will elute first.

The stringency of the selection is controlled by the density of substrate immobilized on the column matrix, and on the IDA- .(or NTA-) coupling density on the cell surface. Variants with improved expression or activity at different conditions, such as salt concentration, pH or temperature may be isolated by this method.

Table 1. Time of 90% complex formation and 90% exchange, at various conditions of temperature and MgCl₂. First oligo (Yl-Fl) was added to 5 nM; second oligo (X) was added to 1 mM; third oligo (Y2 or Yl) was added to 20 mM. Example 2

This example shows non- limiting examples of HTS assays based on Substrate Reloading. Substrate Reloading is based on using a sample comprising a number of individual units in said in vi tro selection method and further wherein said selection method is characterised by the use of one or more reagent (s) which are capable of converting a product generated by a catalyst or enzyme molecule of interest back into the substrate for said catalyst of interest. The assay is performed by screening samples comprising a number of individual units wherein said samples comprise a library of catalyst molecules provided in the form of individual units, wherein the individual units comprise a first type individual unit having the following general structure: C-S, wherein C denotes a catalyst molecule and S a substrate which is capable of being catalysed into a product by at least one catalyst comprised within said library of catalyst molecules and thereby providing the possibility of obtaining a second type individual unit comprising the general structure: C-P, wherein C has the meaning defined above and P is the product molecule resulting from the catalytic conversion of the substrate S of the first type individual unit. The substrate S is attached to the catalyst in a configuration that allows catalytic reaction between the catalyst and the substrate within said individual unit; and

(a) the nature of said attachment of the substrate and the catalyst provides the possibility, by means of a characteristic of the product, of isolating an entity comprising information allowing the unambiguous identification of the catalyst molecule which has been capable of catalysing the reaction substrate molecule to product molecule; (b) under suitable conditions where a catalyst molecule of interest performs its catalytic activity of interest and where said method is characterised by that said sample is further under conditions wherein the product generated by a catalyst of interest are in contact with one or more reagent (s) which convert it back into the substrate S.

Substrate Reloading assay for a SNase

The catalyst of interest is a SNase; substrate is a single stranded oligonucleotide (ssDNNA) ; and product is the ssDNA cleaved by a SNase of interest.

Further, a filamentous phage is used as a carrier system and an acid/base linker is used as a flexible linker.

Accordingly, the individual units in this example has following general structure: SNase - fil. Phage - acid/base link. - ssDNA

Catalyst - Carrier system - flexible linker - substrate.

In this example the "selection characteristic" of the product (i.e. cleaved ssDNA) is that said product does not bind to a matrix and the substrate (ssDNA) does bind to a matrix. Accordingly, in this example a SNase molecule of interest is isolated by selecting for individual units which are released from said matrix.

MATERIALS AND METHODS Synthesis of compounds. Fmoc-S- (2-nitro-4 , 5-dimethoxybenzyl) - L-cysteine 1 was synthesized by a variation of the method of Merrifield (6) . Briefly, 605 mg L-cysteine (5 mmol) was suspended in 100 mL degassed ethanol/water (2:1), and 1.39 mL triethylamine (10 mmol) and 1.39g 1- (bromomethyl) -2-nitro-4 , 5- dimethoxybenzene (5 mmol) were added. The mixture was stirred for 10 h at 23 °C in the dark under nitrogen and filtered. The filter cake was washed with ethanol and recrystallized from ethanol/water to provide 0.95 g S- (2-nitro-4 , 5- dimethoxybenzyl) -L-cysteine (3 mmol) . The recrystallized product (0.8 g) was suspended in 20 ml water; 0.53 ml triethylamine (3.8 mmol) was added followed by a solution of 0.9 g 9- fluorenyl-methoxycarbonyl succinate ester (2.7 mmol) in 12 mL acetonitrile and the mixture stirred for 10 h at 25°C under nitrogen. The product precipitated upon acidification to pH 2-3 with 1 M HCl and evaporation of the acetonitrile. The precipi- tate was collected on a frit and washed with water and ethy- lacetate to remove excess HCl and reagent. The resulting crude product 1 (1.13 g) was extensively dried under vacuum, and used directly in the synthesis of the base-linker peptide (SEQ ID No. 3): C (GGS) ₄AQLKKKLQALKKKNAQLKWKLQALKK-KLAQGGC (base se- quence underlined, photoprotected cysteine in bold) . Compounds 2, 3 and 4 were synthesized on an ABI DNA synthesizer on a 1 mmole scale with a 3 ' -biotin group (BiotinTEG CPG, Glen Research) and a 5' -thiol (5 ' -Thiol-Modifier C6, Glen Research) and purified by reverse phase HPLC following removal from the resin (Rainin Microsorb C18 column, flow 1 mL/min.; solvent A: 50 mM triethylammonium acetate (TEAA) , pH 7, solvent B: acetonitrile, linear gradient from 5 to 50 % solvent B over 40 min) ; the trityl protecting group on the thiol was removed according to the protocol of Glen Research. The products were ly- ophilized and dissolved in water (1.0 mM final concentration). The conjugate of 2 with the base-linker peptide was prepared as follows: 2 mg (415 nmole) base-linker peptide was reacted with a 20 fold molar excess of N,N' -bis (3-maleimidopropionyl) -2- hydroxy-1, 3-propanediamine (3.2 mg) in 1 mL of 50 mM sodium phosphate buffer, pH 5.5, for 10 h under nitrogen at 4 °C. Compound 5 was purified from the reaction mixture by reverse phase HPLC (Vydac RP-18 column, flow 2 mL/min; solvent A: 0.1 % TFA in water, solvent B: 0.1 % TFA in acetonitrile; linear gradient from 10 to 55 % solvent B over 35 minutes) , and the product fractions concentrated to approximately 0.3 mL (OD_28O = 6, compound 5 should not be concentrated to dryness) . To 100 mL (138 nmoles) of this solution was added 75 ul water, 75 mL of aqueous 1 M aqueous sodium phosphate, pH 7, 30 mL of aqueous 5 M NaCl, and 22 mL (22 nmoles) of compound 2, and the reaction in- cubated for 10 h under nitrogen at 23 °C (to avoid precipitation the reagents should be added in this order) . The product was purified by anion exchange FPLC (Mono Q HR 5/5 column (Pharmacia), flow 0.75 mL/min solvent A: 20 mM Tris-HCl, pH 7, solvent B: 20 mM Tris-HCl, pH 7 , 2 M NaCl; linear gradient from 20 to 60 % B in 7.5 min); on a 10 % denaturing polyacryiamide gel the product ran as a single band. Fractions of OD_26O = 0.3-1 were used directly for the photo-deprotection step ( vide infra) . The conjugates of 3 and 4 with the base-linker-peptide were prepared as follows: approximately 200 nmoles of either 3 or 4, and a 20 fold excess of bismaleimide were incubated in 1 mL of aqueous 50 mM phosphate buffer, pH 5.5, at 4 °C for 15 hours. After purification by reverse phase HPLC and lyophiliza- tion, the identity of compounds 6 and 7 was verified by Maldi- ToF MS. Either 6 or 7 (150 nmoles) was then incubated with 100 nmoles base-linker-peptide in 100 mL of 10 mM TEAA, pH 6.5, 100 mM NaCl for 15 hours at 4°C. The products were purified by reverse phase HPLC (Vydac RP-18 column, conditions as described above) , lyophilized and analyzed by Maldi-ToF MS (7) . The 2- nitro-4 , 5-dimethoxybenzyl protecting group on the C-terminal cysteine of the three conjugates was removed by photolysis to afford compounds 8, 9 and 10 as follows: for compound 8, 100 mL of the FPLC purified fraction containing the protected conjugate ( vide supra) was degassed thoroughly with argon for 15 min, and then exposed to a mercury lamp (450 W high pressure mercury lamp, Ace-Hanovia; Pyrex TM filter, cutoff = 300 nm) in a septum capped glass vial for 30 min (8) . For compounds 9 and 10, 10 nmole of the conjugate was dissolved in 100 mL of 10 mM DTT, degassed and photolyzed as described above. After 30 min of irradiation no remaining starting material could be detected by MALDI-ToF MS. The reaction mixture was separated by HPLC (Vydac RP-18 column, conditions as described above) and the product fractions were lyophilized. The conjugates were stored frozen, and used within a week after photo-deprotection, to ensure efficient attachment to phage.

Construction of acid helper phage A Narl restriction site was introduced between the third and fourth codon of mature pill protein of M13K07 helper phage (Promega) by Kunkel mutagenesis (9) with the primer (SEQ ID No.15): "K07-NarI-prim" (5 ' -acaactttcaacggcgccagtttcagcgg-3 ' ) to give Narl-helper phage.

DNA encoding the amino acids (SEQ ID No. 16) : GAAQLEKELQALEKENAQLEWELQALEKELAQGGCPAGA (acid peptide sequence underlined, GGC motif in bold) with a Narl restriction site at both ends, was produced by polymerase chain reaction (PCR) with the plasmid pCRII acid (Ellis L. Reinherz, Dana Farber Cancer Institute, Boston) with the primers (SEQ ID No. 17) : "Narl-fwd" 5 ' -actacaaattggcgccgctcagctcgaaaaagagc-3 ' ) and (SEQ ID No. 18): "Narl-bck" 5 ' -aattataggcgccagccgggcaaccgccctgagccagttccttttcc- 3 ' . The PCR product was digested with Narl and inserted into Narl digested Narl-helper phage to afford acid helper phage.

Construction of phagemids encoding the staphylococcal nuclease- pIII fusion and 39-All Fab-pIII fusions

To make the SNase-pIII fusion, PCR was performed on the plasmid pONFl (10) , carrying the gene encoding SNase, with primers (SEQ ID No. 19): 5 ' -cgcgaattggcccagccggccat- ggccgcaacttcaactaaa-3 ' (Sfil restriction site underlined) and (SEQ ID No. 20): 5 ' -gcgaattggtgcggccgcttgacctgaatcagcgttg-3 ' (Notl restriction site underlined) . The product was digested with Sfil and Notl and inserted into Sfil-Notl digested pFAB- 5c. His, a derivative of plasmid pFAB-5c (11), to give phagemid pII78-6. As a negative control the phagemid pComb3H.DA was employed. This phagemid (12) carries the 39-All Fab antibody (13) fused to the pill protein. The expression of both the SNase and control protein is driven by the lac promoter.

Production of phage particles

Phage particles were produced with minor modifications according to ørum et al . (11) . Briefly, E. coli XLl-blue was transformed with pII78-6 or pComb3H.DA, and shaken at 37 °C in 2xYT broth and 100 mg/mL ampicillin. At an OD₆₀₀ of 0.5, acid helper phage was added to a final concentration of 1.5 x 10⁸ cfu/ L, and incubated at 37°C for 20 min. The cells were pelleted and re-suspended in 2xYT, 100 mM IPTG, 100 mg/mL ampicillin, 50 mg/mL kanamycin, and shaken for 14 hours at RT. Cells were pelleted and phage particles in the supernatant were PEG precipitated, followed by re-suspension in TBS (25 mM Tris-HCl, pH 7.4, 140 mM NaCl, 2.5 mM KC1) . Phage titrations were performed with E. coli XLl-blue using standard procedures (14) .

Covalent attachment of base-linker-substrate conjugates to phage

Approximately 10⁸ phage particles were incubated in 40 mL buffer A (TBS, 10 mM EDTA, 0.1 % BSA), supplemented with 1 mM mercaptoethylamine (MEA) and 1 nmole of either base-linker- oligodeoxynucleotide (8) , base-linker-pTp (9) or base-linker- pTpTp (10) , at 37 °C for 60 minutes, then PEG precipitated twice and re-suspended in buffer A.

Phage immobilization and release from solid support Approximately 10⁸ phage particles, covalently attached to the base-linker-substrate conjugates, were incubated with 50 mL magnetic streptavidin beads (Boehringer Mannheim, biotin binding capacity: 1.5 nmole/mL) in 1 mL buffer A for 15 minutes at 23 °C; eight 1 min washes were performed in buffer A with 0.1 % Tween 20, followed by two 1 min washes in buffer A. The number of phage immobilized on the beads was determined by suspending the beads in buffer A, and then either directly infecting E. coli XLl-blue with the bead suspension and titering or alternatively, infecting after treatment of the beads with DNase 1 (lunit/mL DNase 1, 10 mM MgCl₂, 20 mM Tris-HCl, pH 8, 23 °C for 15 min) . Calcium-dependent release (cleavage) from solid support was examined by suspending beads in buffer B (TBS, 10 mM CaCl₂, 0.1% BSA), incubating at 23°C for 5 min, and titering the supernatant. Calcium-independent release from the beads (leakage) was determined by re-suspending the beads in buffer A, incubating for five minutes at 23 °C, and titering the supernatant .

Enrichment of active enzymes from a library-like ensemble Phage particles displaying SNase or 39-All Fab were mixed in a 1:100 ratio and the base-linker-oligodeoxynucleotide conjugate (8) was covalently attached. Phage were then immobilized on magnetic streptavidin beads, washed in buffer A, and incubated in buffer B as described above, E. coli XLl-blue were in- fected with the supernatant and the cells plated on a LA plate containing 100 mg/mL ampicillin. Randomly picked colonies were identified as SNase- or control clones by PCR or restriction enzyme digestion.

RESULTS & DISCUSSION

Selection scheme. To test the above strategy for directed- enzyme evolution in a phage-display format, it was first necessary to develop a general method for selectively attaching a given substrate to or near a phage-displayed enzyme. Impor- tantly, the substrate must be attached so that it can bind productively in the active site of the conjugated enzyme. Moreover, the substrate should be covalently linked to the phage to ensure that there is no crossover of reaction product between members of the library. One possible strategy involves selec- tive chemical modification of the enzyme or a nearby phage coat protein (e.g., pill protein) with substrate by a disulfide exchange reaction. For example, a cysteine residue introduced near the active site of staphylococcal nuclease through site- directed mutagenesis has been used to selectively introduce unique chemical functionality by a disulfide exchange reaction

(15) . To apply this method to proteins expressed on filamentous phage, the three single cysteines of the pVI, pVII and pIX coat proteins were first mutagenized to alanine. The eight buried cysteine residues in the pill protein were left unchanged, as they likely form structurally important disulfide bridges

(16) . Unfortunately, repeated attempts to selectively modify unique cysteine residues introduced near the active site of several enzymes displayed on phage, by either disulfide exchange, maleimide addition or alkylation reactions, resulted in significant nonspecific labelling of phage coat proteins. No conditions or reagents were found that made possible selective labelling of the pill fusion protein containing the unique surface cysteine residue. It is likely that the thousands of proteins constituting the phage coat make the specificity requirement for a chemical reaction too great; also, the probability of cysteine misincorporation due to the intrinsic error rate in protein biosynthesis becomes significant for such a large ensemble of proteins. Alternatively, the cysteine residues in the pill protein may be accessible to cross-linking reagents.

To circumvent these problems, a two-step process was developed in which chemical cross-linking is preceded by the selective formation of a noncovalent complex at the site of modification. The complex is a heterodimeric coiled-coil consisting of a synthetic basic peptide B (SEQ ID No. 21) : C (GGS) ₄AQLKKKLQALKKKNAQLKWKLQALKKKLAQGGC, to which substrate is covalently coupled before heterodimerization, and an acidic peptide A (SEQ ID No. 22) : GAAQLEKELQALEKENAQLEWELQAL- EKELAQGGCPAGA that is expressed as an N-terminal fusion to the pill coat protein of filamentous phage. The acid and base peptides (underlined) were chosen as dimerization domains because of their small size (thirty amino acids) and high tendency to form stable, parallel heterodimeric coiled-coil structures -- the acid-acid and base-base homodimers form 10⁵ fold less efficiently than the heterodimer (17) . Heterodimerization of the synthetic (B) and phage-encoded (A) peptides should bring the substrate into close proximity of the displayed enzyme, and lead to spontaneous disulfide bond formation between cysteines on each of the peptides. The tripeptide Gly-Gly-Cys was added to the C-termini of the acid and base peptides to facilitate formation of a disulfide bridge between the two helices (17) . The substrate is covalently linked to the basic peptide B through a flexible linker to facilitate productive binding of substrate to enzyme. The acidic peptide A is fused to the pill protein of the phage rather than to the displayed enzyme itself for the following reasons: (i) insertion of the acid peptide sequence into an enzyme might interfere with enzyme function; (ii) the flexible linker of the base-linker-peptide as well as hinges in the pill protein and a peptide linker inserted between pill and the displayed enzyme, should allow many possible orientations of the substrate relative to the enzyme active site,- and (iii) it should be possible to use a single helper phage bearing the acid peptide extension to display many enzyme-substrate pairs, rather than having to engineer into each enzyme a functional conjugation site.

Generation of the acid helper phage and base-linker-substrate conjugate

To attach the base-linker-substrate conjugate to phage we introduced the acidic peptide A at the N-terminus of pill protein in the M13K07 helper phage. The enzyme library is fused to the N-terminus of the pill coat protein; this construct is car- ried in the phagemid. Upon superinfection by helper phage, phage particles are produced that contain the phagemid DNA but whose coat consists (with one exception) of proteins encoded by the helper phage genome. The one exception is the pill protein, present in 4-5 copies at one tip of the phage. During packag- ing of the phage, both enzyme-pill fusions and acid peptide A- pIII fusions are produced; the phage particles obtained from a typical preparation carry either one or zero enzyme-pill fusions plus three to five copies of acid peptide A-pIII fusion.

To generate phages bearing an acid peptide-pIII fusion, DNA encoding the acidic peptide A with a C-terminal extension containing a cysteine residue, was introduced into the 5 '-end of gene III of the M13K07 helper phage. The resulting acid helper phage particles were immobilized more than hundred fold more efficiently than M13K07 on an ELISA-plate coated with ba- sic peptide B, indicating that the mutant helper phage carry accessible acid peptide extensions on their pill proteins. Likewise, when E. coli containing a phagemid encoding a pill fusion protein was superinfected with the acid helper phage, the resulting phage particles displayed modified pill extensions in addition to the pill fusion protein. The insertion of the acid peptide did not appear to change the titer or rescue efficiency of the helper phage significantly.

The synthetic base-linker-peptide (B) to which substrate is attached consists of the twelve residue (GlyGlySer)₄ linker followed by the thirty amino acids constituting the base se- quence . The base-linker peptide also contains cysteine residues at the N-and C-termini that allow efficient, selective coupling of the peptide to substrates and disulfide bond formation to phage, respectively. The C-terminal cysteine of the synthetic peptide is initially protected with the photochemically remov- able 2-nitro-4 , 5-dimethoxybenzyl protecting group. This allows substrate to be selectively conjugated by a thiol specific reaction (e.g., by disulfide exchange, alkylation, or Michael addition reactions) to the free thiol group of the N-terminal cysteine. After substrate conjugation, the C-terminal cysteine is photochemically deprotected in high yield to generate a free thiol available for cross-linking to the acid peptide extension on phage. Because the chemical conjugation of substrate and base-linker peptide, and the cross-linking of this conjugate to phage are carried out separately, many different chemistries and reaction conditions can be used to couple the base-linker peptide and substrate. Moreover, the composition of the conjugate can be purified and characterized (e.g., by mass spec- trometry) before it is cross-linked to phage.

Staphylococcal nuclease as a model system

The enzyme staphylococcal nuclease is a well- characterized enzyme consisting of single polypeptide chain 149 amino acids in length (18) . The enzyme preferentially hydro- lyzes the phosphodiester bonds of single-stranded RNA (ssRNA) , ssDNA, and duplex DNA at A,U- or A,T- rich regions to generate

3 ' -phosphate and 5 ' -hydroxyl termini (18). Ca is required for enzymatic activity, providing a mechanism for modulating enzyme action. In addition, SNase has successfully been displayed as a pill fusion protein on phage (19) .

Because no reagent, antibody or receptor is available that can easily distinguish between a single-stranded oligode- oxynucleotide substrate and its cleavage product (a complementary oligonucleotide would be degraded) , a selection scheme was developed in which enzymatic cleavage of ssDNA substrate results in release of phage from solid support. In this scheme, one round of selection involves the following steps: i) attachment of phage displaying SNase to solid support through a single-stranded oligodeoxynucleotide (in the absence of Ca²⁺ to inactivate SNase) ; ii) removal of unbound phage by washing; iii) initiation of the cleavage reaction by addition of Ca²⁺, and iv) isolation of eluted phage. In later rounds of selection, elution can be done under increasingly stringent conditions, e.g., shorter reaction time, lower temperature and altered pH. Attachment of phage to solid support is carried out by coiled-coil formation between 5 ' -biotinylated oligodeoxynu- cleotide-peptide B conjugates and acid peptide A extensions on phage, followed by disulfide cross-linking of the two peptides and immobilization on streptavidin beads. This scheme, in which the phage is attached to solid support through the substrate, requires that the enzyme or substrate be maintained in an inac- tive state during attachment to phage, and then be activated by a change in reaction conditions. Such changes can include modulation of pH, addition of cofactors or co-substrates, and photochemical or chemical activation of the substrate. In the case of biomolecular condensation reactions in which bond formation results in phage immobilization on solid support, it is not necessary to initiate the reaction; the same is true if capture of active enzymes is by a product-specific reagent, antibody or receptor.

Covalent attachment of the substrate to phage Phage displaying either SNase or a control protein (antibody 39-All Fab fragment) was prepared by superinfection with the acid helper phage. To evaluate the efficiency of the attachment of base-linker-substrate conjugates to phage, an ex- cess of a control conjugate, "pTp" -peptide B (compound 9), was incubated with the phage. The base-linker-pTp conjugate consists of a biotin moiety, followed by deoxythymidine-3 ' , 5- diphosphate (pTp) , the flexible peptide linker and base peptide sequence, and a C-terminal cysteine. The base-linker-pTp con- jugate is not a substrate for wildtype SNase in solution (pTp is a potent inhibitor of SNase) (20) . Phage and the substrate- peptide B conjugate were first incubated with the reducing agent mercaptoethylamine (MEA) to reduce disulfide bonds between cysteines on the phage acid peptide or the synthetic pep- tide. Then, MEA and free base-linker-pTp were removed by PEG precipitation, and magnetic streptavidin beads were added. After ten washes, the number of phage that was immobilized was determined by infection of E. coli XLl-blue with the beads, and titering phage. When measured this way, the efficiency of phage immobilization was approximately 10%, for both phage displaying SNase and 39-All Fab.

Next it was determined whether an oligodeoxynucleotide substrate attached to phage displaying SNase would be stable in the absence of Ca²⁺. The base-linker-oligodeoxynucleotide con- jugate was attached to phage displaying SNase (in the presence of EDTA) , and the immobilization efficiency determined as above. The efficiency of immobilization was again approximately 10% indicating that the tethered oligodeoxynucleotide substrate is not cleaved by SNase in the absence of Ca²⁺. It is possible that the true immobilization efficiency is higher than observed if some of the phage are rendered non-infective when attached to the beads. This notion was tested by addition of DNase I, which should cleave the tethered oligodeoxynucleotide substrate and release the immobilized phage. As can be seen in most of the immobilized phage are non-infective, but become infective upon addition of DNase I, indicating that the true im- mobilization efficiency is about 80 %. If the oligodeoxynucleo- tide-peptide B conjugate is not included, less than 0.01% of the phage become immobilized; if the wildtype M13K07 helper phage is used to superinfect, about 0.3% of phage are immobi- lized. It thus appears that the two-step protocol for attachment of substrate to phage pill protein is efficient and highly site-specific .

Enzyme dependent cleavage of phage from solid support and en- richment

To determine whether phage-displayed SNase is capable of specifically cleaving the tethered oligodeoxynucleotide sub- state in an intramolecular reaction, Ca²⁺ was added to the immobilized phage to activate the enzyme. Approximately 15% of the phage was released in contrast to release of only 0.2% of the control phage displaying Fab 39-All. This experiment demonstrates that SNase cleaves and releases phage from the solid support much more efficiently than the control protein, as expected. However, it appears that a small but significant frac- tion of the phage leak off the support during the assay (this background leakage is observed without Ca²⁺, for both the base- linker-oligodeoxynucleotide and base-linker-pTp conjugates, and for both displayed proteins. Addition of Ca²⁺ leads to an initial burst of phage release from support; however, the release of phage quickly declines to a level corresponding to the leakage observed without Ca²⁺. This result demonstrates that phage released into solution by intramolecular cleavage events do not release other phage from support as a result of intermolecular cleavage reaction. Cross-reactivity therefore does not appear to be significant, even with a very active enzyme like SNase.

The above analysis suggests that it should be possible to enrich phage-displaying SNase from a library-like ensemble of phage displaying catalytically inactive proteins. To test this, phage displaying SNase and the Fab 39A-11 control protein were mixed in a ratio of 1:100, cross-linked to the oligodeoxynu- cleotide-peptide B conjugate and immobilized. After incubation with Ca ⁺, the ratio of recovered phage was 22:18, which corresponds to an enrichment factor slightly higher than 100. This degree of enrichment should be sufficient to isolate an active catalyst from a library of 10¹⁰ members after five rounds of selection and amplification.

The enrichment factor can likely be increased by minimizing background leakage of phage from support. This leakage may result from release of streptavidin from support, or alterna- tively, reduction or incorrect formation of the disulfide bridge between the synthetic and phage encoded peptides. We are currently exploring these possibilities. Alternatively, the enrichment factor can be raised by increasing the extent of the enzyme-catalyzed cleavage reaction. Under the conditions of phage production, the ratio of pill expressed from the helper phage relative to the pill fusion protein expressed from the phagemid is such that most of the phage carry only wildtype pill proteins; only a minor fraction of the phage carry the protein-pill fusion. The number of phage that can cleave them- selves off can be increased simply by increasing the number of phage that display the enzyme. For the phagemid/helper phage combination described here, we estimate that only about 15% of the phage is monovalent. By appropriate vector design and phage preparation, it should be possible to increase the average dis- play to about one protein per phage. This should increase the cleavage to leakage ratio 7 fold, and hence, increase the enrichment factor of active versus inactive enzymes from the present -100 to about 700.

To examine whether the selection scheme described here can be used for reactions that involve small molecule substrates, a pTpTp-peptide B conjugate (compound 10) was attached to phage displaying SNase or the control protein. Phage were carried through the enrichment routine described above, and again SNase displaying phages were enriched. MALDI-ToF mass spectrometry was used to show that the pTpTp substrate was cleaved at the phosphodiester bond between the two thymidines; no side products were detected. It thus appears that the methodology is applicable to both macromolecular and small molecule substrates. We are currently exploring the possibilities for isolating novel catalysts from libraries of enzyme or antibody origin.

Most enzyme libraries displayed on phage require superin- fection by a helper phage like M13K07. The selection protocol described here can therefore be applied directly to these libraries - one simply needs to prepare phage after superinfec- tion of the phagemid encoded library with the acid peptide helper phage, and conjugate the substrate of choice to the basic peptide B. Likewise, this methodology can be applied to populations of structurally diverse proteins. The collection of proteins encoded by a genome is one such population. For exam- pie, it should be possible to isolate natural kinases with predefined substrate specificity from a genomic protein library using this selection scheme. This type of functional cloning in which a natural enzyme (and the gene that encodes it) is isolated on the basis of its catalytic activity should be applica- ble to many reactions catalyzed by natural enzymes.

References and Notes used above

1. Schultz, P.G. & Lerner, R.A. (1995) Science 269, 1835-

1842. 2. (a) Marks, J.D., Hoogenboom, H.R., Bonnert, T.P., McCaf- ferty, J., Griffiths, A.D. & Winter, G. (1991) J. Mol .

Biol . 222, 581-597. (b) Barbas, C.F., III, Bain, J.D.,

Hoekstra, D.M. & Lerner, R.A. (1992) Proc . Na tl . Acad .

Sci . U. S . A . 89, 4457-4461. (c) Griffiths, A.D., et al . (1994) EMBO J. 13, 3245-3260.

3. Janda, K.D., Lo, C-H.L., Li, T., Barbas, C.F., III, Wirsching, P. & Lerner, R.A. (1994) Proc . Natl . Acad. Sci . U. S . A . 91, 2532-2536.

4. (a) Soumillion, P., Jespers, L., Bouchet, M., Marchand- Brynaert, J., Winter, G. & Fastrez, J. (1994) J. Mol .

Biol . 237, 415-422. (b) Janda, K.D., Lo, L-C, Lo, C- H.L., Sim, M.M., Wang, R., Wong, C-H. & Lerner, R.A. (1997) Science 275, 945.

5. Gao, C, Lin, C.H., Lo, C-H.L., Mao, S., Wirsching, P., Lerner, R.A. & Janda, K.D. (1997) Proc . Na tl . Acad. Sci .

5 U. S . A . 94, 11777-11782.

6. Erickson, B.W. & Merrifield, R.B. (1973) J. Am . Chem . Soc . 95, 3750-3756.

7. Piles, U., Zϋrcher, W. , Schar, M. & Moser, H.E. (1993) Nucl . Acids . Res . 21, 3191-3196. lo 8. Marriott, G. & Heidecker, M. Biochem . (1994) 33, 9092- 9097.

9. Kunkel, T.A., Roberts, J.D. & Zakour, R.A. (1987) Methods in Enzymology 154, 369.

10. Hibler, D.W., Barr, P.J., Gerlt, J.A. & Inouye, M. (1985) is J. Biol . Chem . 260, 2670-2674.

11. ørum, H., Andersen, P.S., øster, A., Johansen, L. K. , Ruse, E., Bjørnvad, M. Svendsen, I. & Engberg. J. (1993) Nucl . Acids Res . 21, 4491-4498.

12. Schultz, P.G. & Romesberg, F.E. unpublished results.

20 13. Romesberg, F.E., Spiller, B., Schultz, P.G. & Stevens, R.C. (1998) Science 279, 1929-1933. 14. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) "Molecular Cloning: A Laboratory Manual," 2nd Ed. Cold Spring Harbor Lab., Cold Spring Harbor, NY.

25 15. Pei, D., Corey, D. R. & Schultz, P.G. (1990) Proc. Na tl . Acad. Sci . 87, 9858-9862.

16. (a) Lubkowski, J., Hennecke, F., Plϋckthun, A. & Wlodawer, A. (1998) Nature Structural Biology 5, 140-147. (b) Kremser, A. & Rasched, I. (1994) Biochemistry 33, 13954-

30 13958.

17. (a) O'Shea, E.K., Rutkowski, R., Kim, P.S. (1989) Science 243, 538-542. (b) O'Shea, E.K., Rutkowski, R. , Stafford, W.F. Ill & Kim, P.S. (1989) Science 245, 646-648. (c) O'Shea, E.K., Klemm, J.D., Kim, P.S. & Alber, T. (1991) Science 254, 539-544. (d) xZhou, N.E., Kay, C. M. & Hodges, R.S. (1993) Biochemistry 32, 3178-3187.

18. (a) Cotton, F.A., Hazen, E.E., Jr., & egg, M.J. (1979) Proc . Na tl . Acad . Sci . U. S . A . 76, 2551-2555. (b) Tucker, P.W., Hazen, E.E., & Cotton, F.A. (1978) Mol . Cell . Biochem . 22, 67-77. (c) Sondek, J. & Shortle, D. (1990) Proteins 7, 299-305. (d) Hale, S.P., Poole, L.B. & Gerlt, J.A. (1993) Biochemistry, 32, 7479-7487. (e) Hynes, T.R. & Fox, R.O. (1991) Proteins 10, 92-105. (f) Loll, P.J., Quirk, S., Lattman, E.E. & Gravito, R.M. (1995) Biochem. 34, 4316-4324. (g) Judice, K. , Gamble, T.R., Murphy, E.C., de Vos, A.M. & Schultz, P.G. (1993) Science 261, 1578- 1581.

19. Ku, J. & Schultz, P.G. (1994) Biomed . Chem . Lett . 2,1413- 1415.

20. Tucker, P.W., Hazen, E.E., Jr. & Cotton, F.A. (1979) Mol . & Cell . Biochem . 23, 3-16.

Optimization of a secreted enzyme's activity. In Substrate Reloading assays the individual unit may consist of a cell, a substrate attached to the surface of that cell, and enzymes produced and secreted by the cell. The substrate is attached to the surface of an enzyme-secreting cell through a linker. Reagent (s) in the media continuously turn product (produced by the secreted enzymes) into substrate. Consequently, as the local concentration of secreted enzyme is much higher near the substrate attached to the cell from which it was secreted than near any of the other cells' attached substrates, there will be more product attached on a cell secret- ing an active enzyme than on a cell secreting a less active enzyme .

The substrate can be attached to the cell surface in many ways. For example, phospholipids, fatty acids, sterols or cho- lesteryl esters may be derivatized with the substrate of the target reaction. When incubated with cells, these carbon chains readily localize in the membrane interior, and expose the sub- strate on the surface of the cell. Alternatively, the substrate may be derivatized with cross-linking reagents that react with the surface constituents. Finally, the substrate may be conjugated to structures (e.g. proteins, antibodies) that bind to membrane components such as polysaccharides or membrane proteins .

The principle is here examplified in the case where the individual unit consists of a cell (bacteria or yeast) , attached substrate (double stranded DNA with 5 ' -overhang) , and secreted enzyme (ligase; for example in a recombinant form that allows its secretion) . A restriction enzyme (for example EcoRI) is used as the reagent. The selection is performed in the column format. The column matrix is coated with double stranded DNA with 5' -overhangs that are complementary to the overhangs exposed on the surface of the cell, and that create an EcoRI restriction site upon ligation of the two DNA fragments.

DNA substrate is attached to the cells as follows. A PCR reaction is performed, using two primers one of which is 5'- derivatized with N-hydroxysuccinimide, and DNA containing an EcoRI restriction site as template. Approximately 50 mg DNA product is digested with EcoRI and purified, to yield a 10-100 base pair double stranded DNA with EcoRI 5' -overhangs at one end, and the N-hydroxysuccinimide moiety at the other. The DNA is mixed with cells, harvested at the exponential growth phase and resuspended in appropriate buffer. This results in the reaction of primary amines on the surface of the cell with the N- hydroxysuccinimide moiety of the DNA. As a result, the DNA becomes covalently attached to the cell surface.

A DNA column is made as follows. A PCR reaction is per- formed, using a biotinylated and a non-biotinylated primer, and a DNA containing an EcoRI restriction site is used as template. The PCR product is cleaved with EcoRI, to produce a fragment of 10-100 base pairs. The fragment is mixed with streptavidin- coated Sepharose or Sephadex column material . The DNA coated cells are loaded on the DNA column, together with the restriction enzyme (EcoRI) , in a suitable buffer that allows activity of both the restriction enzyme and the ligase (i.e. contains ATP, pH 7-9), as well as allows efficient secretion of the ligase. The temperature is 24-37°C; the flow is kept below 0.5 mL/min. Cells secreting an active ligase become immobilized on the column, through ligation of the complementary DNA overhangs of the cell surface and the column. Upon ligation, EcoRI will cleave the DNA and thereby release the cell from the column matrix. Consequently, cells secreting active ligase will migrate slower through the column; cells secreting very active ligases may not migrate through the column at all. Therefore, cells secreting the more active ligase are isolated from the later column fractions, or may be eluted from the column matrix, for example by addition of high concentrations of DNase. The principle is tested by isolating the cells that secrete the most active ligase/highest expressed ligase from a model library of well-characterized cell clones. The stringency of the selection is controlled by the density of DNA fragments on the surface of the cell and solid support, as well as by the concentration of the restriction enzyme in the buffer. Ligase variants with improved characteristics at desired pH, salt and temperature conditions may be evolved by this method. Also, the inverse experiment may be performed; using ligase and ATP as the reloading reagents, and conditions similar to those described above, and cells secreting restriction enzymes, it may be possible to isolate restriction enzymes that cleave novel targets, or that cleave under different conditions

Example 3

This example shows non-limiting examples of HTS assays based on Fluorescence Activated Cell Sorter (FACS) based assays. FACS-based assays are used to screen for products secreted from cells, where a means to establish a correlation be- tween the activity of the secreted product and the secreting cell has been established first as part of the screening protocol .

More specifically in the current context a FACS-based assay is defined as a screening method for DNA of interest com- prising the steps of a) creating host cells comprising the DNA library, b) generating samples each comprising a host cell of step a) , c) establishing a means for correlating host cell secretion of a material of interest in a sample with the fluorescence of the sample, d) determining which intensity interval of fluorescence indicates secretion in the sample when the correlating means of step c) is used, e) cultivating the samples under suitable conditions, and f) selecting the samples exhibiting fluorescence within the intensity interval of step d) using a fluorescence analyzer; wherein the host cell comprises DNA of interest .

Below are given a non-limiting number of examples of a FACS-based assay as defined in this invention.

FACS-based Antibody quenching assay for a protease This example illustrates the FACS screening of a gene library for protease activity based on antibody quenching of fluorescein labeled GMD's. The library was constructed in Es- cherichia coli as described in PCT DK99/00495 (not published at filing date) . The library was amplified in E. coli and plasmid was extracted and transformed into Bacillus subtilis Sha273 (WO95/10603) .

Protease assay principle

The fluorescence of agarose beads labeled with fluo- rescein is quenched by an anti-fluorescein antibody. The action of a protease will degrade the anti-fluorescein antibody whereupon fluorescein will be de-quenched, thus allowing protease positive GMD's to be sorted on a FACS on the basis of increased fluorescence .

Quenching / de-quenching in microtiter plates In 96-well plates, lOμl of anti-fluorescein IgG (Molecular probes, Eugene; Or) was added to 250 μl fluorescein solution in each well and fluorescence intensities were measured on a PolarStar (BMG Labtechnologies, Germany) with and without Ab addition (see table 2) . A 53 -fold reduction in fluorescence was shown with Ab addition. After addition of lμl protease (alca- lase, Novo Nordisk, Denmark) a 33-fold de-quenching was measured, showing a high increase in fluorescence when the antibody was degraded by protease (see table 2) .

Table 2. Fluorescence of Fluorescein before and after quenching with an antibody (Ab) , and after degradation of the quenching Ab (de-quenching) .

FACS sorting of GMD's

Fluorescein labeling of agarose (F-agarose) was accomplished by dissolving agarose (2.65g Ultra-low Gelling temperature agarose cat #A2576 Sigma Aldrich) in water (70 ml) at 65°C, pH was adjusted to 10.5 using IN NaOH-solution. The temperature was lowered to 35°C. 25 mg DTAF (Dichlorotriazino-5-amino- fluorescein) was added and the reaction mixture was stirred in darkness overnight. The labeled polymer was precipitated in EtOH (600 ml) , filtered, and washed (EtOH) .

Fluorescein labeled GMD's (F-GMD's) were prepared by first mixing lOμl of a 0.5% F-agarose in 400μl 4% agarose (Ultra-low Gelling temperature agarose cat # A2576 Sigma Aldrich) at 80°C, and then generating the F-GMD's on a CelSys 100 Micro- Drop Maker (OneCell Systems) using the following protocol:

1) 15 ml emulsion oil (Dimethylpolysiloxane DMPS2C, Sigma Aldrich) in a scintillation vial and the melted agarose mix were equilibrated in a 40°C water bath.

2) Add 100 μl Pluronic acid (Pluronic F68 solution, Sigma Aldrich) to the agarose and let equilibrate at 40°C (3min) .

3) Harvest cells by centrifugation and re-suspend in lOOμl PBS (pH7.5)

4) Add 100 μl cells (adjust concentrations of cells to give less than one cell/GMD) or lOOμl PBS to the molten agarose; equilibrate at 40°C (3min) .

5) Add .the agarose-cell mixture dropwise to the warmed emulsion oil avoiding air bubbles.

6) Emulsify using the following settings:

2100 RPM for 1 min at room temperature 2100 RPM for 2 min in ice bath

1100 RPM for 10 min in ice bath

7) Harvest GMD's by dividing the encapsulation mixture in two 15ml conical tubes and carefully overlay the emulsion with 5 ml PBS and centrifuge for 10 min at 1500 RPM in a Megafuge 1. OR (Heraeus Instruments) centrifuge. A pellet should be visible

8) Remove the oil phase and the overlaying PBS. Wash the pellet by adding the concentrated GMD suspension to a new 15 ml conical tube with 10 ml PBS. Spin for 10 min at 1500RPM in a Megafuge 1. OR (Heraeus Instruments) centrifuge. Wash step can be repeated if oil still is present .

9) Decant supernatant and re-suspend GMD's according to application protocol .

The protease assay was set up in the following way: Assay mix consisted of 20 μl F-GMD solution mixed with 10 μl 50mM

Hepes (pH 8) , optional lOμl anti-fluorescein antibody (Molecular Probes, anti-fluorescein, rabbit IgG fraction; catalog # A889;), optional protease (alcalase, Novo Nordisk A/S) ; the volume was brought to 50 μl using milliQ water. The assay mix was incubated over night .

The result of the FACS analysis (using a FACSCalibur, Becton Dickinson, USA) showed the distribution of fluorescence 5 of F-GMD's. Quenching was achieved by addition of anti- fluorescein Ab . Addition of 5μl alcalase showed that fluorescence was recovered almost totally. The FACS-analysis of a mixture of 99% quenched GMD's and 1% protease de-quenched GMD's showed that 0.7% of the GMD's were sorted as being fluorescent, o showing that even a low number of F-GMD's can be recovered.

Encapsulation of the B . subtilis library expressing protease was accomplished by adding around 10⁶ cells (transfor- mants) in PBS to the agarose as stated in step 4 in the protocol above. F-GMD's with cells were harvested as described s above. The anti-fluorescein antibody was dialysed and lOμl Ab was added to a 1:1 mixture of GMD solution and TY containing the appropriate antibiotics to maintain the B . subtilis library. The GMD cultures were grown until appropriate protease expression was obtained. Alternatively the anti-fluorescein an- o tibody can be added after an appropriate growth period, whereafter a sufficient incubation/growth period is included to allow for degradation of the antibodies by the clones producing protease .

After the growth phase the GMD cultures were har- 5 vested and re-suspended in 0.2 mM Tris buffer pH7 , and sorted on a FACScalibur flow cytometer (Becton Dickinson, USA) . For isolation of protease active GMD's the gates were set allowing only the most fluorescent GMD's to be sorted (i.e. GMD's de- quenched by action of the produced protease) . GMD's were sorted 0 at a rate corresponding to around 1000 events per second, and deposited on to a filter. Isolated GMD's were distributed directly into microtiter plates or plated onto indicator plates for protease activity (e.g. LB plates containing 1% skim milk) . Optionally agarase can be added in order to aid outgrowth of 5 the sorted clones from the GMD's.

The sorted cells can be re-grown in TY medium con- taining appropriate antibiotics, harvested, re-embedded in F- GMD's and subjected to a second round of FACS screening as mentioned above. This screening resulted in a significant enrichment for protease positive clones.

FACS-based substrate quenching assay for protease

Protease assay principle This example illustrates the FACS screening of a gene library for protease activity based on a substrate quenching principle in GMD's. The gene library was prepared as described in above .

When hemoglobin is labeled with flourescein, the fluores- cence is quenched by the hemoglobin. Therefore coupling of hemoglobin to agarose via DTAF (Dichlorotriazino-5- aminofluorescein; a bi-functional fluorescein) will allow the degradation of hemoglobin by a protease to be detected as an increase in GMD associated fluorescence.

Hemoglobin de-quenching assay

Hemoglobin was labeled with DTAF by dissolving 5,66 g hemoglobin in 600 ml miliQ water. pH was adjusted to 10,0 using 4N NaOH. 95,2 mg DTAF dissolved in 4 mL DMF was added and the reaction mixture was stirred in the dark at room temperature for 24 hours. The mixture was transferred to an Amicon RA-2000 unit, equipped with a Filtron 10-kD. filter, and was dialysed against miliQ water until no fluorescence could be found in the filtrate . To show degradation of hemoglobin by protease, lOOμl F-

Hemoglobin at a concentration of 10 μg/ml in 50 mM bicine

(Sigma) pH 9 was incubated with and without 1 μl protease (Sav- inase , Novo Nordisk A/S) for 60 min in 384-well microtiter plates (Nunc, Denmark) . An approximate 10-fold increase in fluorescence intensity was measured on a PolarStar Galaxy (BMG LabTechnologies, Germany) in the sample with protease (Table 3). This clearly demonstrates the de-quenching of fluorescein when hemoglobin is degraded by protease.

Table 3.

In order to cross-link the hemoglobin molecule to agarose using DTAF as a bi-functional linker, the agarose was first labeled with DTAF under mild conditions. Agarose was dissolved in milliQ water and the pH was adjusted to 9,0 using 4 N NaOH. DTAF dissolved in 4 ml DMF was added, and the reaction mixture was stirred in the dark at room temperature for 2-4 hours. The labeled agarose was precipitated in ETOH (96%) and washed 3 times in ETOH (96%) and once in cold water and then freeze dried. The DTAF-agarose was dissolved in milliQ water and the pH was adjusted to between 7.5-8.5, hemoglobin was added, and the reaction mixture was stirred at room temperature for up to 24 hours . Hemoglobin-fluorescein GMD's were prepared as described above.

Encapsulation of a B . subtilis library expressing prote- ase in Hemoglobin-fluorescein GMD's was accomplished by adding around 10⁶ cells (transformants) in PBS to the agarose as described above. Hemoglobin-fluorescein GMD's with cells were harvested as described above.

A 1:1 mixture of Hemoglobin-fluorescein GMD containing cells and TY containing the appropriate antibiotics to maintain the B . subtilis library was grown until appropriate protease expression was obtained. After the growth phase the GMD cultures were harvested and re-suspended in 0.2 mM Tris buffer pH7 and GMD population was sorted on a FACSCalibur flow cytometer (Becton Dickinson, USA). For isolation of protease active GMD's the gates were set to allow only the most fluorescent GMD's to be sorted (i.e. strong fluorescence is present in GMD's where protease activity has degraded the hemoglobin and the fluorescein is thus de-quenched) . GMD's were sorted at a rate corresponding to around 1000 events per second, and deposited on to a filter. Isolated GMD's were plated directly onto indicator plates for protease activity (eg. LB plates containing 1% skim milk) . Optionally agarase can be added in order to aid outgrowth of the sorted clones from the GMD's. The sorted cells can be re-grown in TY medium containing antibiotics, harvested, re-embedded in Hemoglobin-fluorescein GMD's and subjected to a second round of FACS screening as mentioned above. This screening resulted in a significant enrichment for protease positive clones .

FACS-based biotin reloading assay for protease

This example illustrates the FACS screening of a gene library for protease activity based on the biotin reloading principle in GMD's. The gene library was prepared as described above .

Protease assay principle

In biotinylated agarose beads avidin is bound to all biotin molecules. Activity of a protease will remove avidin and the amount of free biotin can be measured by addition of fluo- rescein labeled avidin. Only GMD's containing cells with protease activity will be fluorescent.

GMD's were prepared as described above using biotinylated agarose (CelBioGel encapsulation matrix, OneCell Systems) , then harvested and re-suspended in PBS. Assay reactions were set up in 50 mM Tris pH 9 buffer by adding the appropriate amount of buffer until 50μl volumes. When 20μl Biotin-GMD' s and lOμl of fluorescein conjugated Avidin (Molecular Probes; Eugene Or. Cat # A821) were mixed in a total volume of 50 μl, the beads turned fluorescent. In con- trast, when F-avidin was incubated with 2μl protease (Savinase , Novo Nordisk A/S) at 37°C in Tris pH 9.0 and 20μl of the protease treated F-avidin was added to Biotin-GMD's the fluorescence was observed to be at background levels. This shows that protease treated avidin does not bind to Biotin-GMD's, thus demon- strating that the basis for the assay is sound.

Encapsulation of a B . subtilis library expressing protease in Biotinylated GMD's (B-GMD's) was accomplished by adding around 10⁶ cells (transformants) in PBS to the agarose. B-GMD's with cells were harvested, and 20μl Avidin (Molecular Probes; Eugene Or. Cat # A887) was added to a 1:1 mixture of B-GMD solution and TY containing the appropriate antibiotics to maintain the B . subtilis library. The GMD-cultures were grown until appropriate protease expression was obtained. Alternatively the avidin can be added after an appropriate growth period, where- after a sufficient incubation period is included to allow for degradation of avidin in the GMD's where protease is expressed.

After the growth phase the GMD cultures were harvested and re-suspended in 0.2 mM Tris buffer pH7 , and lOμl F- avidin was added in order to label the biotin liberated by the action of the protease. After a second wash, the GMD's were sorted on a FACScalibur flow cytometer (Becton Dickinson, USA) . For isolation of protease active GMD's, the gates were set to allow only the most fluorescent GMD's to be sorted (i.e. F- avidin is primarily bound to GMD's where unlabeled avidin is degraded by the produced protease) . GMD's were sorted at a rate corresponding to around 1000 events per second, and deposited on to a filter. Isolated GMD's was plated directly onto indicator plates for protease activity (e.g. LB plates containing 1% skim milk) . Optionally agarase can be added in order to aid outgrowth of the sorted clones from the GMD's. The sorted cells can be re-grown in TY medium containing antibiotics, harvested, re-embedded in B-GMD's and subjected to a second round of FACS screening as mentioned above. The screening resulted in a significant enrichment for protease positive clones.

FACS-based dye quenching/FRET assay for amylase This example illustrates the FACS-screening for amylase activity based on a dye quenching or a FRET assay principle where the fluorescence in retained in GMD's. The gene library was prepared as described above.

Quenching or FRET assay principle

The substrate is designed to exhibit quenching or FRET (fluorescence resonance energy transfer) in its native (non- hydrolyzed) state. Furthermore the substrate is bound to the GMD using a bi-functional fluorescent dye. Upon degradation of the substrate, the quenching or FRET is relieved, and since the fluorescent dye is bound to the agarose only GMD's containing enzyme activity will be fluorescent, or in the FRET case GMD's will fluoresce at the lower wavelength. The substrate for this assay, an oligosaccharide (size may vary between DP 2 and DP 20) or a polysaccharide, is labeled with a fluorescent group (optionally fluorescein) and a quenching group (optionally tetramethylrhodamine) . The distance between dyes is between 10-75 A in order for efficient FRET to occur. An oligosaccharide may be labeled with tetramethylrhodamine at the reducing end and with fluorescein at free OH groups preferably on the C6-carbon. A double functionalized fluorescent dye such as DTAF is used to couple the substrate to the agarose of the GMD.

Quenching assay based screening

In a quenching assay the substrate is designed so that fluorescein emission is quenched by a tetramethylrhodamine quencher group. When bonds in the oligosaccharide or polysac- charide are broken by enzymatic hydrolysis originating from the cells in the GMD, the fluorescein group will be de-quenched and the fluorescence will be retained in the GMD because the fluorescein also is linked to the agarose. The library will be encapsulated in GMD's, GMD harvested and incubated in LB or TY medium and grown until sufficient enzyme activity is present. The library will be screened by FACS, and for isolation of en- zymatically active GMD's the gates will be set to allow only the most fluorescent GMD's to be sorted. GMD's will be sorted at a rate corresponding to around 1000 events per second, and deposited on to a filter or directly into microtiter plates. Optionally agarase can be added in order to aid outgrowth of the sorted clones from the GMD's. The sorted cells can be re- grown in TY medium containing antibiotics, harvested, re- embedded in GMD's and subjected to a second round of FACS screening.

FRET-assay based screening

In the FRET-based assay the substrate is designed so that excitation of fluorescein will result in energy transfer to and emission from tetramethylrhodamine. When bonds in the oligosac- charide or polysaccharide are broken by enzymatic activity originating from the cells in the GMD, the tetramethylrhodamine emission will diminish whereas the fluorescein emission will increase. The library will be encapsulated in GMD's, GMD's will be harvested and incubated in LB or TY medium and grown until detectable enzyme activity is present. The encapsulated library will be screened by flow cytometry. During FACS sorting the gates will be set taking changes in two colors into account; for fluorescein emission the high fluorescent population is gated and for tetramethylrhodamine emission the low fluorescent population is gated. Enzymatically active GMD's will then be sorted based on the increase in fluorescein emission or preferably on the increase in the ratio between fluorescein emission and tetramethylrhodamine emission. GMD's will be sorted at a rate corresponding to around 1000 events per second, and de- posited on to a filter or directly into microtiter plates. Optionally agarase can be added in order to aid outgrowth of the sorted clones from the GMD's. The sorted cells can be re-grown in TY or LB medium containing antibiotics, harvested, re- embedded in GMD's and subjected to a second round of FACS screening. FACS-based substrate assay for amylase

This example illustrates the FACS-screening for amylase activity based on generating microspheres of substrate incorporated into GMD's. The gene library was prepared as described above .

Hydrolase assay principle

Fluorescently labeled substrate (e.g. starch) is incorporated in GMD's by using cross-linked microspheres of such a size that they are retained in the GMD's when the substrate is intact. When substrate hydrolysis occurs the fluorescence is lost from the GMD by diffusion, and non-fluorescent GMD's are sorted.

Starch microspheres were generated as described by G. Hamdi et al . (G. Hamdi , G. Ponchel, and D. Duchene . An original method for studying in vitro the enzymatic degradation of cross-linked starch microspheres. J. Control . Release 55 (2-3): 193-201, 1998.) by epichlorohydrin crosslinking of fluorescein labeled starch (F8387, Sigma Aldrich) . After cross-linking mi- crospheres were fluorescent as observed by epi-fluorescence microscopy, and the diameters were between 1 to 12 μm. The average diameter of GMD's generated with the emulsification setting used above are approximately 25-35 μm. Accordingly the concentration of starch microspheres was adjusted so that each GMD contained between 1 and 5 microspheres . To incorporate the starch microspheres in the GMD's, the GMD's are generated as described above except that also 30 μl microspheres are added to the molten agarose and only 70μl cells are added. GMD's containing microspheres and cells are harvested and washed as de- scribed above.

The library will be encapsulated in GMD's containing labeled starch microspheres as described above, GMD's can be harvested and incubated in LB or TY medium containing appropriate antibiotics and grown until detectable enzyme activity is pre- sent. The encapsulated library can be screened by flow cytome- try. During FACS sorting the gates can be set to select the fraction of GMD's which have lost their fluorescence (i.e. the least fluorescent population is sorted) . Since positive clones are expected to lose fluorescence it could be advantageous to include a cell stain in the assay in order to stain GMD's with microcolonies . The cell stain should emit at a wavelength different than fluorescein, and could be a membrane probe such as the red fluorescent Dil (Molecular Probes) . Gates can in this case be set to identify GMD's with microcolonies (red fluores- cent) which have lost their green fluorescence, indicative of breakdown and leakage of fluorescein labeled starch. GMD's can be sorted at a rate corresponding to around 1000 events per second, and deposited on to a filter or directly into microtiter plates. Optionally agarase can be added in order to aid outgrowth of the sorted clones from the GMD's. The sorted cells can be re-grown in TY or LB medium containing antibiotics, harvested, re-embedded in GMD's and subjected to a second round of FACS screening.

Example 4

This example shows HTS assays based on Colony Picking. Colony Picking assays are categorized as assays screening a large population of variant cells or cell colonies present on a first surface, which cells or colonies may be capable of pro- ducing a useful polypeptide, RNA or small molecule, which method comprises the steps of i) on the first surface, subjecting the cells or cell colonies to an assay correlated to a property of the useful polypeptide, RNA or small molecule; ii) by means of a colony picker, selecting cells having the prop- erty from the first surface; and iii) transferring the selected cells to a second surface.

By using the methodology of the present invention it has now been made feasible to pre-screen very large populations of variant cells or cell colonies for material of interest, e.g. for polypeptides/proteins, RNA or small molecules which are produced by the cells or colonies and which has useful and desirable properties. An example is prescreening of DNA libraries for an industrially useful enzyme.

In the present context, the term "colony picker" denotes an apparatus capable of a) detecting desirable cells or colonies present on a first surface by means of an automated visual analysis of the cells or colonies which is based on predetermined criteria; and b) transferring the desired cells or colonies from the first surface to a second surface. The first and second surfaces are preferably a plate, optionally a sub- strate plate including substrate for growth or non-growth of the desired variant cells or colonies. The plate may be a well plate or a plate with variant cells or colonies arranged in a spatial array. The colony picker produces a digital picture of the plate including the cells or colonies and analyses the pic- ture in order to locate the desired cells or colonies which may be detected as being e.g. coloured, growing, growth-inhibited, colourless or fluorescent. Based on the detection of location of the desired variant cells or colonies, the colony picker touches each desired variant cell or colony with a needle in order to transfer material from the colony to the needle and the needle then transfers the material to a second surface, for example to inoculate growth media in a microtiter well . Before the next colony is picked the needle is sterilized, for example in a bath containing ethanol or another conventional steriliz- ing chemical compound or composition.

A useful colony picker is the ^λQ' Pix Colony Picker which is an automated benchtop Colony Picker, Gridder and MicroAr- rayer manufactured by Genetix. The ^λQ' Pix colony picker can pick and re-array over 3.500 colonies per hour into 384 or 96 well plates. It may be set to pick based on absorption at a given wavelength. It picks out of 22x22 Bioassay trays, Petri Dishes or Omnitrays. When picking colonies, the needles are sterilized in sterilization baths after a colony has been picked, and before the next colony is picked, by forcing ster- ilization solutions (e.g. ethanol) across the pins at high pressure. The colony picker may also be used to pick from 384 or 96 well plates into agar plates (called gridding) . Arrays are done with a 16-Pin Head, using either for example 0.15 um diameter Genetix solid pins, or much thicker needles, for example with a diameter of 1 mm.

5 Other useful colony pickers are devices capable of picking based on any visual detection of desirable variant cells or colonies, i.e. light emission or light absorption including fluorescence .

A number of non-limiting examples of Colony Picking as- lo says are described below to exemplify the invention.

Colony Picking of colonies producing clearing-zones on skim- milk agar plates

150 ml 15-fold diluted LB-medium (J. Sambrook, E.F. Fritsch is and T. Maniatis (1989), A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory Press) containing skim-milk and 6 mg/ml Chloramphenicol is poured into a 22x22 cm plate (Genetix) . Bacillus cells expressing various mutants of Savinase, suspended in 0.5-1.0 ml LB-medium, are dispensed onto the agar 20 at a density of 1000-5000 and the cells are spread with glass- beads by shaking. The diluted media ensures a homogenous size distribution of the colonies, which are otherwise of too varied size to be efficiently and accurately picked based on the appearance of a clearing zone. The plates are incubated at 37 or

25 52°C over night. Growth at 52° gives a surprisingly high level of protease activity early in the growth process. Transformants forming colonies surrounded by clearing zones in the agar are recognized and picked by a fully automated QPix from Genetix.

30 Colony Picking of cells producing an anti-fungal activity, thereby generating clearing-zones on plates covered with a layer of fungi

A large number (more than 10⁵) of fungal cells or spores suspended in 1 ml YPD are plated on a 22x22 cm plate containing

35 YPD or other solid media supporting the growth of the fungi, and incubated at 30°C for 2-5 days. Approximately 500-5000 Bacillus cells, each producing a separate variant of an anti- fungal peptide, in 40 ml 2xYT are added to the plate, and the Bacillus cells allowed to sink down on the agar. The liquid is poured off the plate, and the plates incubated at 30°C for 1-5 days. A colony picker is used to pick Bacillus colonies around which a clearing zone has formed. The clearing zone is an area of less dense growth of the fungal cells.

In an alternative format, 500-5000 Bacillus cells each pro- ducing a separate variant of an anti-fungal peptide are suspended in approximately 1 ml LB-medium. The cells are dispensed onto a 22x22 cm agar plate and are spread with glassbeads by shaking. The Bacilli are incubated ON/for 1-5 days allowing for colonies to form. A large number (10⁴-10⁵) of fungal (e.g. Bo- trytis cinerae) spores suspended in 5 ml of YPD containing low melting agarose are spread over the plate and incubated for 2-5 days at 30°C. A colony picker is used to pick Bacillus colonies around which a clearing zone has formed. The clearing zone is an area of less dense growth of the fungal cells. To allow bet- ter identification of clearing zones, the fungal and/or bacterial cells may be stained with strain-specific stains or dyes, or by using other techniques known in the art.

Colony Picking of cells producing a substance promoting fungal growth

Approximately 1000-5000 Bacillus cells, each producing a separate variant of a growth promoting substance, and a large number (more than 10⁵) of fungal cells, are plated on a 22x22 cm plate containing YPD or other solid media supporting the growth of the fungi, and incubated at 30°C for 2-5 days. A colony picker is used to pick Bacillus colonies surrounded by a dense layer of fungal cells. Fungal and/or bacterial cells may be stained with strain-specific chromophores, in order to better detect fungal growth. Colony Picking of cells producing stable protease on agar plates containing urea and/or SDS

Approximately 1000-5000 B. subtilis cells expressing a library of protease variants are spread out on a 22x22 cm LB agar plate (Genetix) , containing 1% skimmed milk, and up to approximately 0.02% SDS and/or up to approximately 0.75 M urea, and incubated overnight at desired temperature (between 30°C and

50°C) . Bacillus cells producing heat and detergent stable Savinase variants can be recognized and picked by the colony picker, as colonies producing clearing zones.

Liquid Colony Picking: Liquid phase re-array of clones expressing active substance for high throughput screening using automated workstations and robotic arms Any colorimetric assay which can be performed reading between 340nM and 800nM is adaptable to this liquid mode colony picking method with the described set-up. The assay should be in a range determined for the particular system and assay, but generally the reading should be between 0.2 and 2.0 for the wavelength used. For example, once background OD is determined, the following enzyme activities could be measured with, for example, but not limited to, the following substrates, (a) Lipase using para nitophenyl butyrate, (b) Laccase using 2,2'-azino- bis (3-ethylbenz-thiazoline-6-sulfonic acid (ABTS) (c) alkaline protease using N-succinyl-Ala-Ala-pNitoranilide or N-succinyl- Ala-Ala-Pro-Phe-pNitoranilide, (d) peroxidase using ABTS and hydrogen peroxide, (e) Novamyl using a coupled substrate consisting of peroxidase, ABTS, glucose oxidase, the dye Direct Violet 51, maltotriose. The details of this screen simplifies and optimizes the automation of high throughput screens: i) diluted media can be used in small volumes (20ml - 40ml) in 96 well plates whereby the substrates (e.g. 200 ml) can be directly applied to the growth well, ii) viable cultures can be recovered from the growth well to which substrate has been added without the need for a duplicate plate, iii) it is possible to do more manipulations of a sample in liquid phase than in solid phase (advantage over colony picking from e.g., agar-plate) . For example, if the treatment (e.g., heat treatment) or the assay (e.g., as- say at low pH) kills the cells, it is possible in the liquid colony picking approach to split the master plate (growth plate) into two plates, an assay plate and a growth plate. The reaction plate can be exposed to a treatment (e.g., high temperature) , and based on the selected criteria, the correspond- ing positive clone can be picked from the growth plate.

Colony Picking in a Protease selection assay / Antimicrobial scavenging system

In this approach, positive protease producing clones is enriched on the basis of their ability to detoxify growth media containing peptide-based antimicrobials; in the present example Protamine . A protease depleted B . subtilis strain (SHA273) was transformed with either no DNA or a protease (Savinase) expressing plasmid (pSX222) .

An overnight culture of SHA273 and pSX222/SHA273 was individually diluted to a density of lxlO³ cells/ml and plated on agar plates with increasing concentrations of the antimicrobial peptide Protamine (0, 50, 75, 100, 125. 150, 175, 200 μg/ml).

Agar plates were incubated overnight at 37 °C and the number of colonies was scored. Numbers are shown in table 4 below and plotted in figure 1. The result shows that Protamine inhibits the growth of cells with increasing concentrations. In addition, cells expressing Savinase has a higher survival rate compared to cells expressing no Savinase, in the presence of Protamine. These results show that applying an AMP as selective agent, al- lows preferential growth of protease producing cells.

Table 4. ^DMicro colonies

Though the concept above was exemplified with the antimicrobial peptide Protamine, it is not limited to this antimicrobial alone. A number of other protein-based antimicrobials exist and have been characterized that could be employed. Exam- pies of a-helical antimicrobial peptides are Magainin 1 and 2, Cecropin A, B and PI, CAP18; Andropin; Clavanin A, Styelin C and D, and Buforin II. Examples of cysteine-rich peptides are a-defensins HNP-1, HNP-2 and HNP-3, b-Defensin-12 , Drosomycin, gl-Pyrothionin and insect Defensin A. Examples of b-sheet pep- tides are Lactoferricin B, Tachyplesin I, and Protegrin PGl-5. Examples of peptides with unusual amino acid composition are Indolicidin, PR39, Bac5, Bac7 and Histatin. Examples of peptides with unusual amino acids are Nisin, Gramicidin A and Alamethicin. Another example is the anti-fungal peptide (AFP) from Aspergillus giganteus . Furthermore, to select/enrich for a protease with a given substrate specificity, the nature of the antimicrobial agent would reflect this preference. For example, synthetic antimicrobial peptides in all-D configuration would enrich cells expressing a protease capable of degrading peptides in this configuration. Other examples are histidine-rich or proline/argenine-rich antimicrobial peptides, and the antim- icrobial peptides mentioned above which contain posttransla- tionally modified amino acids.

The described approach is not limited to antimicrobial peptides, but can also utilize antimicrobial enzymes including lysozymes, phospholipases, oxidoreductases, laccases, chiti- nases, glucanases or cellulases.

•The enzyme employed to quench the toxicity of the protein- based antimicrobial is not limited to a protease. Many antimicrobials contain essential sulfur-bridges, and their activity is destroyed or diminished by breaking or re-arrangement of these sulfur bridges. This could be mediated by for example re- ductases or protein disulfide isomerases. Other enzymes that could quench the activity of protein-based antimicrobials could be trans-glutaminases (coupling glutamate and lysine) . If the target protein does not contain lysine and glutamate residues available for cross-linking, another protein or peptide could be added that would supply this missing residue and correspondingly allow the inactivation of the antimicrobial through cross-linking.

Yet other enzymes, e.g. oxidoreductases, e.g. laccases, peroxidases and haloperoxidases, could be employed to destroy the antimicrobial activity.

Finally, the enrichment could be strongly enhanced if the proteinaceous antimicrobial agent also served as the only available carbon and/or nitrogen source, cf. the described ra- tionale of in vivo selection.

Colony Picking of cells producing protease by growth on an antimicrobial peptide as Iturin

Iturins are antibiotics produced by some strains of Bacil - lus subtilis . The structure consists of a heptapeptide sequence closed in a ring with lipophilic β-amino acid with typically 14-16 carbon atoms. This antibiotic is a potent anti-fungal agent and correspondingly can be employed in the present invention against enzyme-producing fungi and yeasts. It has been shown that Iturins interact with the membranes of sensitive or- 5 ganisms, most likely creating pores leading to efflux of essential cellular compounds and/or compromising the membrane integrity.

Again, microbes producing an enzyme activity that would allow for the inactivation of the antimicrobial agent, Iturin, o would selectively proliferate at the expense of fungi or yeast not producing this activity. Essential determinants for the antimicrobial activity of the Iturin are the heptapeptide itself, which is in the invariant chiral configuration LDDLLDL, an invariant Tyrosine residue within these seven α-amino acids, as s well as a lipophilic β-amino acid. Due to the peptide-based backbone of the Iturins, cells secreting a protease with the desired specificity would selectively proliferate.

Cells expressing an enzyme activity capable of 0- methylating the invariant Tyrosine would also selectively pro- 0 liferate, as this O-methylation has been shown to dramatically decrease the antibiotic activity of the Iturins.

Finally, the third major structure in the Iturins, the lipophilic β-amino acid, is of strong interest. The lipophilic nature of the β-amino acid is essential, whereas the exact bond- 5 ing of the lipophilic structure to the hepta-ring is less critical. This allows for the synthetic or semi-synthetic synthesis of Iturin derivatives that contain a specific chemical bonding between the hepta-ring and the lipophilic structure. The breaking of this artificial bond would separate the lipo- o philic structure and the hepta-ring and, correspondingly, inactivate the antimicrobial. Designing and synthesizing Iturin derivatives with a given type of bonding would allow the selection of a secreted enzyme that were active on this specific type of bond. Other examples of pairs of chemical bonds and the 5 corresponding enzyme are esters and esterases, nitrils and a nitrilases, phosphates and phosphatases, disulfide bridges and reductases, and various glucosidic bonds and enzymes such as glucanases, amylases, pullulanases and glucosidases .

Colony Picking via two-component antimicrobial systems Another example is a two-component system; one component is a molecule able to penetrate or mediate a molecular transfer into the cell in question; the other component is an antimicrobial activity that on its own is unable to penetrate the cell and hence exert its antimicrobial activity. The covalent link- ing of these two components creates a potent antimicrobial. The exact physical linkage between the components is of no or little importance to the antimicrobial activity of these two connected components, and allows for the creation of chimeras physically linked by a substrate of interest. A cell producing an enzyme that would break/split this linkage would inactivate the two-component antimicrobial and hence allow the cell to selectively proliferate in the presence of the two-component antimicrobial .

As a specific example, one component would be a peptide or protein able to penetrate the cell of interest without being toxic on its own. Examples of this could be specific amphipatic α-helices, or scavenging proteins transported over the cellular membrane and into the cell (e.g. siderophores) . The toxic component could be an antisense molecule, preferentially the arti- ficial nucleic acid mimic denoted PNA. PNA has superior antisense properties when compared to both DNA and RNA, but is unable to penetrate cells. Other candidates of an antimicrobial agent that is unable to penetrate the cell is the CcdB Gyrase inhibitor.

Colony Picking of Aspergillus clones secreting active enzyme

It is a difficult task to make protein libraries in filamentous fungi . One problem is the different morphology and growth rate of different clones. This gives rise to large and small clones, as well as clones that might be outcompeted by other clones. Assaying fungal transformants of different sizes is also a difficult task due to differences in amounts of expressed enzymes.

These problems can be overcome by growing the individual clones in alginate balls with a high molecular weight (HMW) dextran as carbon source. The HMW dextran lets the slow-growing transformants outgrow, without cross contamination from faster growing transformants. Then the balls are spread out on a flat surface, substrate may be added that give rise to for example coloring of clones producing active substance. Beads containing active colonies are picked on the basis of this signal, and transferred to a spatial array, for example a microtiter plate.

Experimental :

A Polymerase chain reaction, using pAHL (carrying a lipase gene) as the template and 2 pmol/ml of each primer: oligo 115120 (SEQ ID No. 23) : gctttgtgcagggtaaatc, and oligo 134532 (SEQ ID No. 24): gagcaatatcaggccgcgcacg is run under the following conditions: 94°C, 5 min.; 30 cycles of (94°C, 30 sec; 50°C, 30 sec; 72°C, 1 min.), and 72°C, 5 min. and a commercial Taq polymerase such as AmpliTaq, (Perkin-Elmer Corp., Branchburg NJ, USA) . For example, the PCR conditions may result in a high rate of error, and therefore a library of Lipase mutants are generated. Protoplasts of the filamentous fungi strain JaL250 is trans- formed using 2 mg of pENI1298, which had been digested with Ball and SgrAl to remove most of the lipase encoding sequence, and 5 mg of the above PCR product. The vector and the PCR fragment are allowed to recombine in vivo as described in WO 97/07205. The protoplasts are washed twice with ST (0.6M sorbitol, lOOmM TRIS pH7.0) in order to remove CaCl₂. The protoplasts are then re-suspended in an alginate solution (1.5% alginate, 2 % high molecular weight dextran (5-40*10⁶ kd) , 1.2 M sorbitol, 10 mM TRIS pH7.5) . Using a pump the suspension is pumped through a tube ending in a small hole, where small suspension droplets are made. These drops fall down (15 cm) into a shake flask (500ml) containing 0.2M CaCl₂, 1.2 M sorbitol, 10 mM TRIS pH7.5. Droplets of an alginate solution (typically 1-2 %) turn into hard balls when they encounter a CaCl₂ solution (such as a 0.2 M solution) . Therefore, small alginate ball of the size 2.5 mm in diameter are generated by this procedure. The protoplast suspension should be made so that approximately 1 out of 5 balls contains a transformed protoplast, in order to avoid multiple clones in the same ball.

The protoplast-containing alginate balls are grown over- night at 30°C degrees in STC (10 mM CaCl₂, 1.2 M sorbitol, 10 mM TRIS pH7.5) in order to regenerate the cell wall. After a couple of washes in sterile water to remove sorbitol, the balls are transferred to 1* Vogel media and grown 2-3 days at 30°C degrees. The sole carbon is the dextran in the balls. This pre- vents cross-contamination from ball to ball, and allows slow- growing transformants to gain reasonable biomass.

The alginate balls are spread out on an agar plate. Then, the balls are soaked in 0.01% para nitrophenol-butyrate, 0.1% Triton X-100, 10 mM CaCl₂, 50 mM Tris-HCl pH 7.5, and incubated at room temperature for approximately 5 minutes. Balls containing active lipases will appear yellow, and can be picked by a colony picker, for example a fully automated QPix from Genetix, based on absorbance at 405 nm.

Example 5

This example shows non- limiting examples of array based HTS assays. Array based assays are categorized as HTS assays for production of a molecule of interest in a large population of host cells, the assays comprising the steps of: a) arranging the host cells in a spatial array so each position in the spatial array is occupied by one cell, b) cultivating the host cells under growth conditions suitable for HTS, c) assaying each array position for production of the mole- cule of interest, and d) selecting the cells from those positions where the molecule was produced, as determined in step c) . Array based assays can be performed in many ways, the spatial array of step a) can take on any physical form whatso- ever, that enables the assaying of several samples at once, without one sample contaminating another. Examples of preferred spatial arrays are different kinds of microtiter-plates with any number of wells, such as 96 or 384, and of any kind of material, as well as positions in a High Performance Liquid Chro- matography (HPLC) autosampler device. Any kind of physical arrangement which allows the unambiguous identification of the samples by a number or a position in the array. Even samples placed as drops on a surface in a specific recorded pattern, the surface being of a solid material or of more complex nature such as a textile or a tissue, e.g. cotton, wool, paper, or cellulose. A way of carrying out step c) of the array based assay above could be to take a sample from each position of the spatial array, e.g. from a supernatant or cell, and transfer this to another spatial array for further testing or assaying for production of the molecule of interest. The second spatial array may be identical to the first one used in the specific method, but may also be of any other kind that fulfills the above mentioned criteria, such as a microtiter plate, a solid surface, a textile, any material etc.

Array assay for detergent α-amylase

Coating of swatches with starch

Unmodified starch from a natural source is mixed with small amounts of fluorescently labelled starch and coated on a solid phase. The natural starch source can be flour derived from e.g. potato, corn, or rice. Especially rice flour has been observed to provide a good correlation to higher scale wash trials. The solid phase may be twill, as twill has been found to provide good correlation to larger scale washes and has good handling properties. The overall starch concentration as well as the ratio between labelled and unlabeled starch can be var- ied over a wide range, but we have found 5% (w/v) unmodified starch and 0.025% (w/v) fluorescin 5-isothiocyanate (FITC) labelled potato starch (50-300 glucose units per FITC molecule) to provide the best compromise between sensitivity and response level (FITC was obtained from Sigma) .

The starches are suspended in water and boiled for 10 min, alternatively jet-cooked for 5 min at 105°C and 2 bar. After cooling, a selected textile is soaked with the cooked suspension, excess slurry removed by rolling, and the textile is dried either overnight at ambient temperature or for a few minutes at high temperature, e.g. 70°C, at high air velocity, e.g. 12 m/s. We have found that automation of this coating procedure to significantly dampen coating variation. By use of a coating machine to continuously coat several hundred meters at a time, subsequently performed assays with repeated dose-response curves exhibit considerably less variation compared to non- continuous coatings.

Growth of Bacillus cells secreting α-amylase variants Bacillus cells secreting α-amylase variants are grown in

2xYT and 6 μg/ml Chloramphenicol for three days at 37°C and 240 rpm, in a culture box (see example 1) .

High throughput screen for detergent α-amylases with better wash performance

Dry starch coated textile is punched into wells of a polystyrene microtiter plate, preferentially of the 96-well format. Assays are performed by first applying a detergent solution, e.g. 150μl, to each well. A range of detergents can be used such as liquid detergent or powder detergent dissolved in water. Water hardness is controlled by the addition of a de- sired amount of calcium and magnesium ions. The assay itself is insensitive to water hardness over a wide range, e.g. 0-30°dH.

If a detergent that contains enzymes is used, these enzymes can be inactivated by e.g. heating the detergent to 85°C for 5 min prior to the assay. Furthermore, the detergent can be centri- fuged and/or filtered before use to minimise particulate matter. However, no assay interference due to un-removed particles has been observed with detergents used in the range of dosages recommended by the manufacturers. The intrinsic pH of detergents can be enforced with buffering capacity by adding e.g. glycine or 3- [cyclohexylamino] -1- propane-sulfonic acid (CAPS) . This is an important quality enhancing measure, since the high throughput screen uses culture supernatants as the source of expressed enzyme. Growth media normally contain buffer components to ensure the optimal growth pH, which is rarely equal to the high pH (often pH 9-11) present in dissolved detergents.

Culture medium from above is added immediately after dispensing the detergent. We prefer to use a small amount of en- zyme solution compared to detergent/buffer, in order to minimise artefacts caused by e.g. a rich growth medium. For example, we routinely apply 15 μl of enzyme solution to 150 μl detergent .

Following addition of detergent and enzyme (culture me- dium) , the simulated "wash" takes place. Application of mechanic motion such as vigorous shaking (e.g., 240 rpm) at this point strongly promotes the effect of tested enzymes, thus creating a good dynamic range for ranking candidates. Incubation times at this step from 5 min to 2 h have been tested. When screening for amylases at low-temperature conditions, we have found 15 min "washing" to provide a good window of difference between "good" or "bad" candidates .

Heating can be applied during incubation to simulate the heating of water during machine washing used in many parts of the world. For example, simulation of European washing conditions could include heating up to 40 or 60°C. This heating can be introduced gradually by for example placing the ambient temperature microtiter tray in a shaking incubator set to the appropriate temperature. By this approach, a heat-up profile is generated. Alternatively, if constant high temperature is de- sired, heated buffers can be added or insulated tubings and thermostated surroundings can be employed.

The assay responses can be read by transferring the wash solution to another microtiter plate and measure released fluo- rescence in this solution. Alternatively, the wash solution can be completely removed and the response measured as residual fluorescence of the swatch.

By using the conditions described above, the two approaches generate almost exact mirror image data, meaning that a high degree of released fluorescence in a well is reflected by a low degree of residual fluorescence on the corresponding swatch and vice versa . We prefer the last method, as this only requires one well per assay, thus ensuring minimal amounts of operations leading to maximal throughput in e.g. a robotic set- up.

The growth conditions used ensure a reasonably uniform growth of identical clones in different wells of the plate. However, large differences in expression level between different variants still exist. We have found that amylase variants with different performance in the final application release different amounts of fluorescent stain from the swatch in the wash performance assay, even when the enzyme is used in saturating concentrations. Therefore, the screen has been set up using very high concentrations of enzyme. Under these condi- tions the stain removal efficiency depends only minimally on the enzyme concentration within a broad range; consequently, differences in expression levels between variants have little effect on the performance of the individual variant in the high throughput screening assay. When applying the assay to libraries of amylase producing clones, hits are defined as wells of the plate wherein the measurements exceed a given assay response relative to a selected benchmark amylase.

Results .

Correlation with the more application-relevant mini-wash test Orange coloured rice starch textile was shaken vigorously in 60 mL Asia-Pacific model detergent containing an amylase variant, and the reflectance measured. Table 5 below shows the correlation between the performance of three different amylases (1, 2, and 3) in the swatch assay (the High Throughput Screening set-up) and the conventional mini-wash test. The improvement factor in a given assay is given by the performance of the variant amylase divided by the performance of a given benchmark amylase. Obviously, the improved amylase (amylase 3) performs better in both the swatch assay and the mini-wash.

Table 5 : Relative residual fluorescence intensity on washed swathes compared to relative reflectance of orange coloured rice starch textile in mini wash. Application dosage of 0.2 μg/ml amylase was used for the mini wash, while approx. 2 μg/ml final dosage was used in the swatch assay (the average concentration from culture supernatants in the High Throughput Assay) .

Array assay for anti-fungal activity

An anti-fungal agent can either inhibit the outgrowth of spores, vegetative cells, or both. To identify an anti-fungal agent inhibiting vegetative cells, 50 μl samples of a liquid culture of a tester strain, e.g. Botrytis cinerae, are distributed into each well of a microtiter plate; 50 μl of samples of sterile culture supernatants from cultures of bacterial cells, e.g. Bacillus, to be screened for secretion of anti-fungal ac- tivities (e.g. variants of an antimicrobial peptide or enzyme), are then added to each well of the microtiter plate. Optionally, instead of sterilizing the culture supernatants, a tissue culture insert (e.g., Nunc TC Insert) may be inserted, to prevent contact between the secreting cells and cells of the tester strain. The insert may contain a membrane non-permeable to proteins or other macromolecular components, which allows the passage of for example small antimicrobial peptides. The microtiter plate is incubated at 30°C for 2-5 days whereafter anti-fungal activity is analysed by optical density measurements in each well at a suitable wavelength. Low optical density indicates the presence of an anti-fungal activity in the well.

Array assay for enzyme-secreting fungi for improved expression using very dilute media

General HTS set up for yield mutant screening. Mutant libraries of Aspergillus oryzae or Aspergillus niger strains are screened by HTS to identify mutants having increased expression, i.e. improved yield of an enzyme. Better yield mutants expressing for example peroxidase, laccase, lipase, glucose oxidase, amylase, xylanase, phytase, and aminopeptidase have been identified this way.

For 96-well plate screens, very dilute medium containing 100 - 300 mg/1 carbon-source, ensures carbon-limitation without oxygen limitation.

Primary 96-well plate screens involve the dilution of spores from distinct mutant pools into fermentation substrate so that one spore in average is inoculated per well when lOOμl of medium is dispensed into each well. After inoculation, the plates are incubated for 3-4 days at 30-34°C under static conditions in a culture box (see example 1) . The wells are then assayed for enzyme activity, for example by the addition of a substrate solution directly to the growth wells.

Mutants of interest are isolated and single colonies transferred to agar slants. Spores from agar slants are inoculated into 96-well plates with approx. 10³ spores per well and fermented under static conditions for 4 days at 34°C to retest isolates. Subsequently, selected isolates are tested in shake flask fermentations.

Isolation of mutants having increased Shearzyme expression 5 An Aspergillus oryzae transformant MT2181 expressing recombinant Shearzyme , was mutated by using UV-irradiation. The MT2181 mutant library was fermented in SH-CDY substrate (pH 6.5) in 96 wells plates as described above. The SH-CDY substrate contained pr litre: 0.3 g MgS0₄,7H₂0, 0.3 g K₂S0₄, 5.0 g lo KH₂P0₄, 0.1 ml tracemetal solution, 0.013 g urea, 0.010 g yeast extract, and 0.1 g maltose. The tracemetal solution contains pr litre: 13.8 g FeS0₄, 7H₂0, 8.5 g MnS0₄, H₂0, 14.3 g ZnS0₄, 7 H₂0, CuS0₄, 5 H₂0, NiCl₂, 6 H₂0, 3 g citric acid.

Cultures were then assayed for increased yield, i.e. in-

15 creased Shearzyme activity. Shearzyme , which is an endoxy- lanase, could be quantitated by using a microtiter plate assay and p-nitrophenyl-β-D-xylopyranoside (Sigma N-2132) as substrate, since the enzyme possesses some exo activity, and yellow p-nitrophenol was generated. A 100 mg/ml p-nitrophenyl-β-D-

20 xylopyranoside stock solution was prepared in DMSO . Subsequently, working substrate was prepared by diluting the stock 10 folds in 0.1 M phosphate buffer (pH 6.0). Standard Shearzyme^" (BioFeed^* (S) , Novo Nordisk A/S, Bagsvaerd, Denmark) was prepared so as to contain 20 FXU/ml in 0.1 M phosphate

25 buffer, pH 6.0. The standard was stored at -20°C until use. Shearzyme stock was diluted appropriately to obtain a standard series ranging from 0.1 to 5.0 FXU/ml just before use. 25 μl samples were dispensed to wells in 96-well plates followed by 120 μl of substrate by using a robotic set up. Using a plate

30 reader, the absorbance at 405 nm was recorded as the difference of two readings taken at approximately 4 minutes intervals. Shearzyme units/ml (FXU/ml) were calculated relative to the Shearzyme standards .

In the 96-well primary screen followed by the re-test in

35 96-well plates, 9 mutants from the "MT2181 mutant library" were selected. These mutants produced higher yields of Shearzyme than the control strain Aspergillus oryzae MT2181. The 4 Shearzyme producmg mutants with highest yields were evaluated in shake flasks. The results obtained are shown in table 6 below where the Shearzyme yield of A . oryzae MT2181 as a control is normalized to 100.

Ta e . earzyme express on y MT 1 Mutants

As shown in table 6, the 4 selected mutants produce ap- proximately 20-30 % more Shearzyme than the control strain A . oryzae MT2181, when fermented in shake flasks.

Array assay for cellulase activity using Dichlorotriazino-5- a inofluorescein (DTAF) labelled cotton Swatches

Preparation of Cotton swatches

Style 400 cotton (Center for Testmaterials, PO Box 120, 3130 AC Vlowdingen, Holland) was cut to appropriate sizes. The cotton swatches were swollen in 0.1 N NaOH solution for 24 hrs. Prior to the labelling in order to enhance the accessibility of the hydroxylic groups to the probe. DTAF dissolved in 0.1 N NaOH was added to the swollen swatches, and the reaction mixture was allowed to react for 24 hrs. at room temperature in the dark on a celloshaker-board. The labelled swatches were washed several times using 0.1 N NaOH, water, MeOH, and finally water again. After the last wash-cycle no probe could be found in the supernatant. The swatches were finally dried. The labelled cotton swatches have a clear yellow colour. Further the DTAF-probe makes the swatches highly fluorescent at the 485/518 (excitation/emission) wavelength bands.

Similar fluorescence measurements were obtained whether cotton was labelled in swatches of 5-11 g or whether it had been cut into swatches of approx. 3 mg, suitable for subsequent microtiterplate analysis.

Specific release of fluorescence by incubation of DTAF labelled cottonswatches with purified cellulases Labelled swatches were distributed in each well of a 96- well microtiterplate, washed in deionized water, and incubated with 50 μl cellulase enzyme solution and 150 μl buffer for 2h, at 40°C, 400 rpm. The microtiter plates were centrifuged at 2800 rpm and fluorescence intensity (485/518 nm) was measured on 150 μl of the supernatant, using a spectrophotometer. The cellulases used were endoglucanase (EG) V, EGV-core, EG I, and CBH II from Humicola insolens (prepared as described in M. Schϋlein, J. Biotechnology, vol 57, pp. 71-81, 1997) .

Table 4 shows that EG V performs much better per endocel- lulase activity unit than EG I, EG V-Core or CBH II. This is consistent with the performance of these enzymes in multi-wash experiments to determine their effect. Hence, DTAF-swatches can be used to assay cellulases and the results correlate with the performance of the cellulase in performing a full-scale wash experiments.

Further, the signal-to-noise ratio is improved by pre- washing the cotton twenty times in a laundry machine, using either water or a model detergent, before labelling. Also, the signal-to-noise ratio is improved by increasing the number of labelling cycles.

Table 7. Performance of 4 cellulases in the swatch assay expressed as fluorescence intensity in arbitrary units (example 7) .

High throughput screening for cellulase activities in cell culture supernatants

Recombinant H. insolens cellulase VI (WO9901544) was expressed in yeast cells (S. cerevisiae, YNG318) using standard methodologies (Sambrook et al . : "Molecular Cloning. A Laboratory Manual", Cold Spring Harbor, NY, 1989). Transformed cells were grown in SC-ura^" media for 3 days at 30°C. 100 μL culture supernatant was transferred to an assay plate containing labelled cellulose swatches as described above. After an appropriate time of incubation the well supernatants were transferred to an assay plate, and the fluorescence was measured. A control experiment was performed in 0,2M tris buffer at pH 7 employing

100 μl supernatant from yeast cultures expressing either recombinant active cellulase EGVI, or an inactive variant of EGVI . Clones expressing active cellulases gave rise to higher levels of fluorescence than clones expressing the inactive variant :

EGVI (active) : 47.8 +/- 1.1 (n=6) EGVI (inactive) : 77.7 +/- 1.9 (n=6)

The DTAF labelled cotton swatches can also be used in a cell culture-based screen for active cellulases in a high- throughput format: Yeast cells expressing cellulases are suspended in minimal media, and aliquoted into 96 well plates at an average of about 30 wells with growth per 96 well plate (to obtain predominantly single cell isolates) . After 3 days of growth at 30°C, lOOμL culture supernatant is transferred to an assay plate containing cellulose swatches as described above. Optionally, a well insert (e.g., Nunc TC Insert) may be inserted, to separate yeast cells and the cellulose swatch. The insert should contain a membrane permeable to the cellulase en- 5 zyme but not to the cells. After an appropriate time of incubation, the well supernatants (without any prior centrifugation) are transferred to an empty assay plate, and the fluorescence is measured for each well. Clones expressing active cellulases give rise to higher levels of fluorescence. 0

Array assay for of proteases using His-tagging/purification to achieve equal protein input in each array position from differentially expressing clones s The DNA sequence encoding the protease Savinase (Novo Nordisk A/S, Denmark) is translationally fused to a sequence encoding a His6 tag and libraries of Savinase^" -His6 variants are produced and introduced into Bacillus . After standard growth, a limited number of Savinase enzymes of each variant (about 10% o of what is secreted by Bacillus carrying the wildtype Savinase gene) are immobilized in the wells of Ni-NTA microtiter plates. The unbound fraction including cells and excess Savinase^" is removed, the plate washed once or twice in a buffer containing 5- 20 mM Imidazole. The His-tagged Savinase^" variants are released 5 from the solid support by the addition of 250 mM Imidazole, and aliquots of the supernatants from each well are used as input in a wash performance assay as described in the previous examples .

0

Array assay for determination of protease concentration by fluorescence polarization measurements

The CI-2 protease inhibitor is labeled by standard means with a fluorescent probe. After growth as described above, the 5 amount of a protease (Savinase , Novo Nordisk, Denmark) variant in a given well may be measured directly in the wells by addi- tion of fluorescence-labeled inhibitor which upon binding to the protease changes rotational speed. The rotational speed may be monitored by fluorescence polarization analysis or by other means which monitors diffusion (e.g. fluorescence correlation 5 spectroscopy) . Since the amount of Savinase variant may vary significantly between the individual wells, the fluorescently labeled CI-2 inhibitor is added in two or three steps of defined amounts and the fluorescence polarisation is measured after each CI-2 addition. The determined concentration of Savi- lo nase variant in the individual well may be used to adjust the input volume from this well into the activity assay, or can be used to correct the obtained activity data, in order to determine the specific activity of the Savinase variant of that well .

15

Array assays for filamentous fungi

It is a difficult task to make protein libraries in filamentous fungi. One problem is the different morphology and

20 growth rate of different clones. This gives rise to large and small clones, as well as clones that might be outcompeted by other clones. Assaying fungal transformants of different sizes is also a difficult task due to differences in amounts of expressed enzymes. Another problem is the handling of all the

25 clones on the plate.

These problems can be overcome by growing the individual clones in alginate balls with a high molecular weigth (HMW) dextran as carbon source, and then using the balls as the source of clonal cell cultures in a microtiter plate. The HMW

30 dextran lets the slow-growing transformants outgrow, without cross contamination from faster growing transformants .

A Polymerase chain reaction, using pAHL (carrying a Lipase gene) as the template and 2 pmol/ml of each primer: oligo 115120 and oligo 134532 is run under the following conditions:

35 94°C, 5 min.; 30 cycles of (94°C, 30 sec.; 50°C, 30 sec; 72°C, 1 min.), and 72°C, 5 min. and a commercial Taq polymerase, such as AmpliTaq, (Perkin-Elmer Corp., Branchburg NJ, USA). For example, the PCR conditions may result in a high rate of error, and therefore a library of Lipase mutants are generated.

Protoplasts of the filamentous fungi strain JaL250 is transformed using 2μg of pENI1298, which had been digested with Ball and SgrAl to remove most of the lipase encoding sequence, and 5μg of the above PCR product . The vector and the PCR fragment are allowed to recombine in vivo as described in WO 97/07205. The protoplasts are washed twice with ST (0.6M sorbitol, lOOmM TRIS pH7.0) in order to remove CaCl₂. The protoplasts are then resuspended in an alginate solution (1.5% alginate, 2 % high molecular weight dextran (5-40*10⁶ kd) , 1.2 M sorbitol, 10 mM TRIS pH7.5). Using a pump, the suspension is pumped through a small nozzle where small suspension droplets are made. These drops fall down (15 cm) into a shake flask (500ml) containing 0.2M CaCl₂, 1.2 M sorbitol, 10 mM TRIS pH7.5. Droplets of an alginate solution (typically 1-2 %) turn into hard balls when they encounter a CaCl₂ solution (such as a 0.2 M solution) . Small alginate balls of the size 2.5 mm in diameter are generated by this procedure. The protoplast suspension should be made so that approximately 1 out of 5 balls contains a transformed protoplast, in order to avoid multiple clones in the same ball. The protoplast-containing alginate balls are grown overnight at 30°C degrees in STC (10 mM CaCl₂, 1.2 M sorbitol, 10 mM TRIS pH7.5) in order to regenerate the cell wall. After a couple of washes in sterile water to remove sorbitol, the balls are transferred to 1* Vogel media and grown 2-3 days at 30°C de- grees. The sole carbon source is the dextran in the balls. This prevents cross-contamination from ball to ball, and allows slow-growing transformants to gain reasonable biomass.

The alginate balls are transferred to a microtiter plate; one ball into each microtiter well. Liquid growth media is added to the wells and after a period of growth incubation, the growth media is analyzed for lipase activity under the desired conditions .

Claims

Claims

1. A High Troughput Screening (HTS) Method for a microbially produced material of interest, comprising sequentially performing at least two different HTS-assays chosen from the group of HTS-assays consisting of FACS-assays, Array based assays, Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays.

2. A High Troughput Screening (HTS) Method for a microbially produced material of interest, the method comprising the steps of: a) choosing at least two different HTS assays from the group of HTS-assays consisting of FACS-assays, Array based assays, Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays; and b) performing the assays chosen in step a) in a sequential manner in the following prioritized order: 1) Substrate Replacement assays or Substrate Reloading assays; 2) FACS- assays or Colony Picking assays; and 3) Array based assays.

3. A High Troughput Screening (HTS) Method for a microbially produced material of interest, the method comprising the steps of: a) choosing at least two different HTS assays from the group of HTS-assays consisting of FACS-assays, Array based assays,

Colony Picking assays, Substrate Replacement assays, and Substrate Reloading assays; and b) performing the assays chosen in step a) in a sequential manner in the following prioritized order: 1) FACS-assays or Colony Picking assays; 2) Substrate Replacement assays or Substrate Reloading assays; and 3) Array based assays.

4. The method according to claims 2 or 3 , wherein at least two assays chosen in step a) belong to the same order of priority in step b) and where those assays belonging to the same order of step b) are performed sequentially in no particular order.

5. The method according to any of claims 1-4, wherein an assay chosen is performed multiple times before another different chosen assay is performed.

6. The method according to any of claims 1-5, wherein the material of interest is a polynucleotide, preferably DNA, more preferably cDNA.

7. The method according to any of claims 1-5, wherein the mate- rial of interest is a peptide or a polypeptide.

8. The method according to claim 7, wherein the peptide is an antimicrobial peptide, a growth promoting peptide, a neuropep- tide, or a pharmaceutical peptide.

9. The method according to claim 7, wherein the polypeptide is an enzyme .

10. The method according to claim 9, wherein the enzyme is selected from the group consisting of proteases, cellulases

(endoglucanases) , β-glucanases, hemicellulases, lipases, peroxidases, laccases, α-amylases, glucoamylases, cutinases, pectinases, reductases, oxidases, phenoloxidases, ligninases, pullulanases, pectate lyases, xyloglucanases, xylanases, pectin acetyl esterases, polygalacturonases, rhamnogalacturonases, pectin lyases, mannanases, pectin methylesterases, cellobiohydrolases , transglutaminases and phytases .

11. The method according to any of claims 1-5, wherein the ma- terial of interest is a small molecule, a pharmaceutical com- pound, a primary or secondary metabolite, or a cellular component .

12. The method according to any of claims 1-11, wherein the ma- terial of interest originates from or is produced in bacterial cells .

13. The method according to claim 12, where the bacterial cells belong to a strain selected from the group consisting of the species Bacillus alkalophilus, Bacillus agaradhaerens, Bacillus amyloliquefaciens, Bacillus brevis, Bacillus clausii , Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacillus lentus, Bacillus licheniformis, Bacillus megaterium, Bacillus stearothermophilus, Bacillus subtilis, Bacillus thuringiensis, Streptomyces lividans and Streptomyces murinus .

14. The method according to any of claims 1-11, wherein the material of interest originates from or is produced in fungal cells .

15. The method according to claim 14, where the fungal cells belong to a strain selected from the group consisting of the genera Acremonium, Aspergillus, Fusarium, Humicola, My- celiophthora, Neurospora, Penicillium, Thielavia, Tolypocla- dium, Trichoderma, Eupenici Ilium, Emericella, Eurotium, Allomyces, Blastocladiella, Coelomomyces , Achlya, Candida, Al ter- naria, Rhizopus and Mucor; preferably the species Aspergillus awamori , Aspergillus foetidus, Aspergillus japonicus, Aspergil lus niger, Aspergillus nidulans or Aspergillus oryzae.

16. The method according to any of claims 1-15, where at least a Substrate Reloading assay and an Array based assay is chosen.

17. The method according to claim 16, where the material of in- terest is an amylase, and the Array based assay is an Array assay for detergent α-amylase.

18. The method according to any of claims 1-15, where at least a Colony Picking assay and an Array based assay is chosen.

19. The method according to claim 18, where the material of interest is an amylase, and the Array based assay is an Array assay for detergent α-amylase.