Selection methods Number:7,067,245 from the United States Patent and Trademark Office (PTO) owispatent A rational method for obtaining a novel molecule capable of a desired interaction with a substrate of interest comprising selecting hosts or replicators which encode said novel molecules based upon cell or replicator growth caused by the desired interaction of the novel molecule and a selection molecule expressed by said host.<H Selection, methods, Selection, metho, 7067245.html, Patent, pto, 7067245

Title: Selection methods

Abstract: A rational method for obtaining a novel molecule capable of a desired interaction with a substrate of interest comprising selecting hosts or replicators which encode said novel molecules based upon cell or replicator growth caused by the desired interaction of the novel molecule and a selection molecule expressed by said host.

Patent Number: 7,067,245 Issued on 06/27/2006 to Wohlstadter

Inventors: Wohlstadter; Jacob Nathaniel (Andover, MA)
Appl. No.: 876343

Filed: June 23, 2004

Related U.S. Patent Documents


Application Number	Filing Date	Patent Number	Issue Date
09573830	May., 2000	6846628
08235437	Apr., 1994	6087177

Current U.S. Class: 435/4 ; 435/235.1; 435/440

Current International Class: C12Q 1/00 (20060101); C12N 15/00 (20060101); C12N 7/00 (20060101)

Other References

Yin et al. Evolution of Bacteriophage T7 in a Growing Plaque. J. Bacteriology vol. 175, pp. 1272-1277 (1993). cited by examiner.

Primary Examiner: Brusca; John S.
Attorney, Agent or Firm: Kramer Levin Naftalis Frankel LLP Evans, Esq.; Barry

Parent Case Text

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. application Ser. No. 09/573,830, filed May 18, 2000 now U.S. Pat. No. 6,846,628, which is a continuation of U.S. application Ser. No. 08/235,437, filed Apr. 29, 1994 now U.S. Pat. No. 6,087,177, the disclosures of which are incorporated herein by reference.

Claims

I claim

1. A method for selecting a virus expressing a molecule having an activity of interest comprising: a) growing a population of viruses in a high selection pressure strain of bacteria where said activity of interest confers a growth advantage; b) expanding the virus population produced in step (a) by growing said population in a low selection pressure strain of bacteria; c) cycling the virus population produced in step (b) through steps (a) and (b) one or more times.

2. The method of claim 1 further comprising exposing said population of viruses to mutagenizing conditions.

3. The method of claim 2, wherein exposure to said mutagenizing conditions occurs prior to growth in said high selection pressure strain.

4. The method of claim 2, wherein exposure to said mutagenizing conditions occurs after growth in said high selection pressure strain.

5. The method of claim 1, wherein the virus is a lytic virus.

6. The method of claim 1, wherein periodic addition of bacteria occurs during step a.

7. The method of claim 1, wherein periodic addition of bacteria occurs during step b.

8. The method of claim 1, wherein periodic addition of bacteria occurs during steps a and b.

Description

TECHNICAL FIELD

The invention relates broadly to rational methods using recombinant genetic techniques and selection to isolate, create or direct the evolution of genes which express novel molecules having a desired interaction with substrates of interest.

More specifically, the invention relates to methods for isolating, creating or evolving novel molecules including organic, inorganic and biomolecules such as proteins, peptides, nucleic acids, oligonucleotides, lipids and polysaccharides for use as reactants, catalysts, enzymatic cofactors, repressors, enhancers, hormones and binders for a wide variety of substrates in industrial and therapeutic products.

Even more specifically, the invention relates to methods wherein host cells and/or viruses, which express a modulated growth factor for the host or for the virus functionally associated with a substrate of interest or analog thereof, and multiple copies of a putative novel molecule or a multiplicity of putative novel molecules which may interact with the substrate of interest or analog to alter the activity of the growth factor, are subjected to selection conditions or evolutionary selection conditions to select for hosts or viruses, or mutations thereof carrying the gene which expresses the novel molecule of interest.

The methods of the invention can be used to rationally create molecules having a wide range of interesting properties including catalysts, e.g., proteases, binding peptides, enzymatic cofactors, enhancers, repressors, and hormones, among others, for a variety of industrial, research or therapeutic uses.

Several publications are referenced in this application by Arabic numerals within parentheses. Full citation for these references are found at the end of the specification immediately preceding the claims. The references more fully describe the state of the art to which this invention pertains as well as certain aspects of the invention itself.

BACKGROUND OF THE INVENTION

In general, there are three ways in which a molecule with novel properties may be obtained. A first method, e.g., protein engineering, relies on known properties of a general type of molecule and upon theoretical models which attempt to define the conformation of molecules most likely to have the desired properties. No models have proved general enough or exact enough to reproducibly design appropriate molecules.

A second method is screening. Screening requires that multiple permutations of molecules be tested for a given property. The current status of screening technology and the vast number of different permutations limits the usefulness of this technique. For example, a peptide sequence of twenty amino acids has 20.sup.20 different permutations. To screen bacteria producing different permutations of peptides of significant length, billions upon billions of petri dishes, each on the order of a thousand colonies, would be needed. To screen such large populations to find those few members, if any, which have the desired characteristics is extremely inefficient. Screening techniques are not adequate for the realistic performance of such tasks.

A third method employs natural selection in specific non-generalizable ways. For example, if a unicellular organism is missing an enzyme in a critical metabolic pathway, one can try to select for a molecule with the same function as that lost by the mutant. This technique is limited, however, by the reactions that are encoded in the genome of the organism and that may be complemented within the cell. Moreover, for each different complementation experiment, a new mutant strain is needed.

The Prior Art

Methods for selecting organisms are well known in the art. These methods include growing host cells in the absence of an essential nutrient, on organic compounds which cannot be utilized by parental strains or in the presence of toxic analogs in order to select for organisms which, for example, express molecules essential for cell growth. Such techniques are primitive because growth in the absence of an essential nutrient does not permit the researcher to rationally design procedures for the selection of molecules for any specific type of reaction or for any particular targeted region within the substrate. Selection pressure based on growth in the absence of an essential nutrient is crude in that no rationally defined selection pressure through which a growth advantage or disadvantage is conferred is imposed and therefore hosts may be selected which achieve survival by expressing molecules having a range of functions. This limits the usefulness of such methods since it reduces the ability of the hosts to isolate or create molecules with specific desired capabilities.

For example, growth on organic compounds which cannot be utilized by parental strains is limited because the hosts are selected only on the basis of their capability of utilizing the organic compound. Use of the organic compound may be accomplished through any of a number of different reactions. There is no rational method to isolate, create or direct the evolution of a molecule capable of a specific reaction with a targeted region within a specific substrate.

Dube et al., Biochemistry, Vol. 28, No. 14, Jul. 11, 1989, disclose the remodeling of genes coding for .beta.-lactamase, by replacing DNA at the active site with random nucleotide sequences. The oligonucleotide replacement preserves certain codons critical for activity but contains base pairs of chemically synthesized random sequences that code for more than a million amino acid substitutions. A population of E. coli were infected with plasmids containing these random inserts and the populations were incubated in the presence of carbenicillin and certain related analogs of carbenicillin. Seven new active-site mutants that rendered the E. coli host resistant to carbenicillin were selected, each containing multinucleotide substitutions that code for different amino acids. Each of the mutants exhibited a temperature-sensitive, .beta.-lactamase activity. Dube et al. is thus limited to enhancing the already known function of a class of enzymes.

A process for producing novel molecules and DNA and RNA sequences through recombinant techniques and selection is disclosed in Kauffman et al., U.K., Patent Application No. GB 2183661A, filed Jun. 17, 1985. Mutated genes are introduced into host cells, the modified hosts are grown so that the mutated genes are cloned, thereby promoting production of the proteins expressed by said genes, the modified host cells are screened and/or selected so as to identify the strains of host cells producing novel proteins with a desired property, and the identified strains are grown so as to produce a novel molecule having the desired property. The techniques taught in Kauffman et al. like those in Dube et al. are limited to methods for modifying the known function of certain classes of molecules.

Schatz et al., Cell, Vol. 53, pp. 107 115 (1988) describe a method for the identification of a fibroblast cell line capable of expressing a gene which encodes an enzyme having known recombinase activity. The method is based upon a process of somatic recombination in which widely separated gene segments are ligated together to form a complete variable region (the variable region being assembled from V (variable), J (joining) and in some cases D (diversity) gene segments in an ordered and highly regulated fashion). Gene transfer is used to stably confer on a fibroblast the ability to carry out V(D)J rearrangements.

Retrovirus-based DNA recombination substrates that comprise a library of genes, some of which encode the recombinase gene, i.e., the gene which expresses the enzyme(s) which play a role in V(D)J recombination, were transfected into host cells which contain a gene expressing a growth factor flanked by the recombinase recognition sequences. Initially, the gene expressing the growth factor was not transcribed or translated. However, transcription and translation of the growth factor was activated when recombinase activity was expressed through the interaction of recombinase with the recombinase recognition sequences.

Bock et al., Nature, Vol. 355, pp. 564 567 (1992), report efforts to select DNA molecules with novel functions. Aptamers, stochastically generated oligonucleotides capable of binding specific molecular targets, were selected in cell-free selection procedures. Single-stranded DNA can be screened for aptamers that bind human thrombin, a protein with no known nucleic acid-binding function. These processes, which actually constitute cell-free screening procedures, include the screening and the amplification of some members of a sub-population. The other members are discarded.

Curtiss, PCT Application No. WO89/03427, discloses methods and techniques for expressing recombinant genes in host cells. Curtiss discloses genetically engineered host cells which express desired gene products because they are maintained in a genetically stable population. The genetically engineered cells are characterized by: (1) the lack of a gene encoding an enzyme essential for cell wall growth, i.e., the inability to catalyze a step in the biosynthesis of an essential cell wall structural component; (2) a first recombinant gene encoding an enzyme which is the functional replacement of the enzyme essential for cell wall growth; and, (3) a second recombinant gene encoding a desired polypeptide which is physically linked to the first recombinant gene. Loss of the first recombinant gene causes the cells to lyse when the cells are in an environment where a product expressed by the first recombinant gene is absent, and where the cells are grown in an environment such that the absence of the first recombinant gene causes the cells to lyse.

Baum et al., Proc. Natl. Acad. Sci., (USA), Vol. 87, pp. 10023 10027 (1990), relates to a method for monitoring cleavage interactions by a variety of proteases. A fusion construct is created by inserting a protease cleavage site e.g., decapeptide human immunodeficiency virus ("HIV") protease recognition sequence, into specific locations of .beta.-galactosidase in E. coli. Those construct genes, which retain their enzymic activity despite insertion of the cleavage site, are subcloned into plasmids which encode wild type and mutant HIV protease, respectively. The fusion construct was found to be cleaved by wild type HIV protease and not mutant HIV protease in both in vivo and in vitro experiments.

Upon cleavage by HIV protease, the altered .beta.-galactosidase is inactivated. The cleavage reaction is inhibited by pepstatin A, a known inhibitor of HIV protease. An analogous construct was developed using a polio protease cleavage site, which was cleaved by polio protease.

Paoletti et al., U.S. Pat. No. 4,769,330, disclose methods for modifying the genome of vaccinia virus in order to produce vaccinia mutants, particularly by the introduction into the vaccinia genome of exogenous DNA. DNA sequences and unmodified and genetically modified microorganisms involved as intermediates are disclosed as are methods for infecting cells and host animals with the vaccinia mutants in order to amplify the exogenous DNA and proteins encoded by the exogenous DNA. This reference is representative of art-known recombinant techniques used to modify both viruses and host cell microorganisms.

Murphy, U.S. Pat. No. 5,080,898, relates to the use of recombinant DNA techniques to make analogs of toxin molecules and to the use of such molecules to treat medical disorders. The toxin molecules can be linked to any specific-binding ligand, whether or not it is a peptide, at a position which is predeterminedly the same for every toxin molecule.

Anderson et al., U.S. Pat. No. 4,403,035, disclose a method for delivery and transfer of genetic information by packaging a hybrid DNA-protein complex into a viral vector, and then transferring this genetic information from the hybrid virus into susceptible microorganisms. An organism having a function or capability desired to be transferred is selected and the DNA thereof is isolated/purified and cleaved to separate the exogenous genes controlling the function desired to be transferred or cloned. These exogenous genes are inserted into the DNA of a virus. The resulting hybrid DNA-protein is introduced into a cell-free in vitro medium, along with a source of viral capsid precursor structure, i.e., proheads, and required accessory viral structural and packaging proteins in order to assemble an infectious hybrid virus encapsidating the hybrid DNA.

The viral capsid precursor structure, and accessory viral structural and packaging proteins, are produced by infecting capable microorganisms with a first viral mutant capable of producing capsid precursor structures without producing at least one packaging protein and infecting compatible microorganisms with a second viral mutant capable of producing accessory viral structural and packaging proteins without producing capsid precursor structures. These infected cells are then mixed and lysed to provide the source of virus components for in vitro packaging for hybrid DNA-protein.

The hybrid virus is then used to infect microorganisms compatible with the virus to program the infected cells to serially reproduce the desired function of the exogenous genes and the genes themselves as nucleic acids.

Dulbecco, U.S. Pat. No. 4,593,002, discloses a method for incorporating DNA fragments into the DNA gene of a virus. The DNA fragments encode for proteins which have specific medical or commercial use. Small segments of an original protein exhibiting desired functions are identified and a DNA fragment, having a nucleotide base sequence encoding that segment of the protein, is isolated/purified from an organism or synthesized chemically. The isolated/purified DNA fragment is inserted into the DNA genome of the virus so that the inserted DNA fragment expresses itself as the foreign segment of a surface viral protein and so that neither the function of the protein segment nor the function of any viral protein critical for viral replication is impaired.

None of the prior art methods offers a rational approach employing selection procedures to the isolation, creation, or creation by directed evolution of novel molecules having a specific function with respect to a chosen substrate of interest. The screening methods are inherently inefficient, wasteful and time consuming. The primitive methods of selection disclosed in the art do not permit the creation, for example, of molecules having high specificity, either as a binder or as catalyst, for a particular recognition sequence. They produce limited numbers of molecules with limited properties. Moreover, none of the prior art references teach methods which are universal in their applicability. There are no prior art methods for the isolation, creation or directed evolution of genes which express different molecules each having a rationally designed activity with respect to a substrate of interest.

OBJECTS OF THE INVENTION

It is thus a primary object of the invention to create novel molecules, capable of interacting with substrates of interest.

It is a related object of the invention to create novel molecules for use as reactants, catalysts, enzymatic cofactors, repressors, enhancers, hormones and binders for substrates, while avoiding the time, effort and failure relatively associated with prior art protein engineering, screening and selection methods.

It is still a further object of the invention to harness the power of selection processes and recombinant genetic techniques to produce molecules not heretofore known and having new functions or improved known functions with respect to a wide array of substrates of interest.

It is still a further and related object of this invention to isolate the genes which express novel molecules from existing gene pools by matching the interactive specificity of these novel molecules to the substrates of interest for which they are specific.

It is still a further and related object of this invention to create novel molecules for interaction with the substrate of interest by rational design of selection based methods which employ specific expressible molecules incorporating the recognition sequences of the substrate of interest.

It is still a further and related object of this invention to harness the rapid replication of cellular and viral genomes in the selection of genes which express novel molecules of interest.

It is still another object of this invention to control and direct evolutionary pressures on cellular and viral systems so as to create and evolve genes which express novel molecules of interest having heretofore unknown physical and/or chemical interactions with substrates of interest.

SUMMARY OF THE INVENTION

The invention is broadly in rational methods for the isolation, creation or directed evolution of a gene which encodes a novel molecule capable of desired interaction with a substrate of interest. The method involves selecting hosts, or replicators in hosts, which encode novel molecules based upon cell or replicator growth caused by the desired interaction of the novel molecule and a selection molecule expressed by the host. The method is performed by expressing multiple copies of a putative novel molecule or a multiplicity of different putative novel molecules in a population of host cells containing a cell growth factor and/or a replicator (e.g., a virus) growth factor, and a substrate of interest or analog thereof functionally associated with said growth factor, and imposing selection conditions on the population of host cells to select for those hosts or those replicators which express a novel molecule which interacts with the substrate of interest or analog to alter the activity of the growth factor.

The invention is also in the modified host cells for use in the invention, in the modified replicators, in certain selection molecules used in carrying out the methods, in the genes and novel molecules produced by the methods of the invention and in systems and kits useful for practicing the invention.

Selection for Host Methods

The isolation, creation or directed evolution of a gene which encodes a novel molecule capable of a desired interaction with a substrate of interest may be performed by the steps of expressing in a population of host cells multiple copies of a putative novel molecule or a multiplicity of putative novel molecules, and adding or expressing a cell growth factor, a substrate of interest or analog thereof having a recognition sequence which represents the substrate of interest and which is functionally associated with the growth factor, and optionally a growth factor modulation moiety, and imposing selection conditions on the population of host cells to select for those hosts containing genes capable of expressing a novel molecule which interacts with the recognition sequence to alter the activity of the growth factor.

The order of expression of the putative novel molecules and the order of expression and/or addition of the growth factor, substrate of interest or analog thereof and modulation moiety relative to one another and to the expression of the putative novel molecules, and the timing of the selection process with respect to any of such steps is a matter of choice. In some embodiments it may be advantageous to impose selection conditions on a population of hosts or replicators prior to modifying the host or replicators to express growth factors, recognition sequences or modulation moieties, or selection molecules incorporating same, so as to develop a desired host or replicator strain for subsequent selection.

The method may be performed by introducing a homogeneous population of genes which may express multiple copies of a putative novel molecule or a heterogeneous population of genes which may express a multiplicity of different putative novel molecules or molecules with evolutionary potential into a population of host cells whose genome has been artificially altered to express a cell growth factor and a recognition sequence which represents the substrate of interest and which is functionally associated with the growth factor, imposing selection conditions, e.g., cultivating or incubating the population of host cells under selection conditions to select for those hosts containing genes capable of expressing a novel molecule which interacts with the sequence to alter the activity of the growth factor and isolating/purifying the gene of interest from the selected cell population. The gene of interest may then be used to express additional quantities of the novel molecule.

The growth factor and recognition sequence may be present as individual molecules or groups of molecules, or, may be associated together in molecules which incorporate both of them. The host cells, e.g., E. coli, may be modified by exogenous addition of the growth factor and/or recognition sequence, or, the growth factor and/or recognition sequence may be expressed by the host. By imposing selection conditions on the population of host cells it is possible to select for those hosts containing genes, or mutations thereof, capable of expressing a novel molecule which has the desired interaction with the recognition sequence and which thereby affects the activity of the growth factor.

Selection for Replicator Methods

The isolation, creation or directed evolution of a gene which encodes novel molecule capable of a desired interaction with a substrate of interest may also be performed by expressing multiple copies of a putative novel molecule or multiplicity of different putative novel molecules encoded by a replicator, e.g., a virus, in a population of host cells which contain or express a growth factor for the replicator, a substrate of interest or analog thereof which incorporates a recognition sequence which represents the substrate of interest and which is functionally associated with the growth factor, and optionally a growth factor modulation moiety, and imposing selection conditions on the population of host cells to select for the replicator, e.g., virus, capable of expressing a novel molecule which interacts with the recognition sequence so as to alter the activity of the growth factor.

These methods may be performed, for example, by introducing a replicator, e.g., a virus, into a population of host cells whose genome has been artificially altered to express a growth factor for the virus and a recognition sequence representing the substrate of interest which is functionally associated with the growth factor, cultivating or incubating that population of host cells to select for the viruses capable of expressing the novel molecule which interacts with the recognition sequence so as to alter the activity of the growth factor, and isolating/purifying the gene of interest. As in the host methods, the order of expression and/or addition of the several components of the process and the order of expression and/or addition relative to imposition of selection conditions is a matter of choice.

In such methods, a homogeneous population of viruses which expresses multiple copies of a putative novel molecule or a heterogeneous population of viruses containing a multiplicity of mutant genes, each of which may express a different putative novel molecule, is introduced into a population of modified host cells which contain a functionally down-modulated growth factor necessary for the growth and/or replication of the viruses and a recognition sequence as described above. Those viruses which express novel molecules which interact with the recognition sequence and thereby up-modulate the activity of the growth factor will replicate within the host. Those viruses which express novel molecules which do not have the desired interaction will not be replicated. The host cells can then be incubated or cultivated under selection conditions to select for the population of the viruses which express the novel molecules of interest.

In a preferred method, the genome of the host cells are artificially altered to express a molecule or molecules which include the growth factor and the recognition sequence which represents the substrate of interest and which is functionally associated with the growth factor. The population of cells is then infected by a replicator, e.g., a virus, whose genome is capable of expressing multiple copies of a molecule or a multiplicity of different molecules which may interact with the selection molecule expressed by the recombinant genome of the host cell. Those novel molecules which interact with the recognition sequence so as to alter the function of the growth factor will confer a selective growth advantage on the virus which expresses the novel molecule of choice. The population of host cells can then be cultivated or incubated to create an amplified population of the desired virus.

As in host selection, the genome of the host cells are artificially altered to express a growth factor and a recognition sequence, as individual molecules or as physical or chemical associations or combinations thereof. The recognition sequence represents the substrate of interest and is functionally associated with the replicator growth factor. Desirably the genome is modified by recombinant methods to express a selection molecule e.g., a fusion or deletion protein, which includes both the growth factor and the recognition sequence. The growth factor and recognition sequence may be associated with a selection moiety which modulates the activity of the growth factor. The selection moiety may be an individual molecule(s) or may be part of a selection molecule(s), e.g., fusion or deletion protein, which also includes the growth factor and the recognition sequence.

The novel molecule to be obtained may act through a cascade of events, i.e., it may interact with the recognition sequence to cause the desired effect or that interaction may start a cascade of events with any number of intermediate steps which ultimately affects the activity of the growth factor. Each molecule in the cascade can be a natural or engineered substrate within the host cell or an exogenously supplied substrate or can be, itself, a novel molecule.

The host selection and replicator selection methods of the invention can be used to create a wide range of novel molecules, e.g., novel proteases capable of a desired interaction with a protease recognition sequence. Molecules other than proteases can be produced, e.g., enzymes capable of site specific glycosylation (or phosphorylation, etc.) around an important cellular protein for growth which is particularly sensitive to glycosylation (or phosphorylation, etc.) and is permissive to the insertion of recognition sequences.

The universal selection method links the formation of virtually any product to the growth of a cell. For example, in the reaction A+B.fwdarw.C (catalyzed by X), linking the production of C to the growth ability of a cell, even though C may have no effect on the growth of the cell directly or indirectly, one can select for that member of a population of putative novel molecules which is capable of catalyzing the reaction (whatever the reaction type may be) or is capable of acting as a substrate or is capable of acting as the product C itself--in short--capable of acting in any way so as to contribute to the production of C.

The invention offers significant advantages over the prior art techniques. It offers an inherent efficiency increase over screening and places the burdens on the experimental system rather than on the experimenters. In selection, the environmental conditions determine which members of a population are viable. By properly defining the selection procedures and conditions, those clones with the desired properties can be obtained from a huge population. The selection procedures of the invention have the advantage that they may be used to obtain a vast array of novel molecules each of which is highly specific for a given recognition sequence and interaction. In contrast, the primitive selection methods of the prior art are crude and empirical.

DETAILED DESCRIPTION OF THE INVENTION

Definitions

Certain terms used herein are defined as follows.

The term "isolation" means bringing forth a gene which exists in nature from an existing gene pool.

The term "creation" means bringing forth a gene not found in nature which encodes a novel molecule.

The terms "directed evolution" and "creation by directed evolution" mean bringing forth a gene not found in nature which encodes a novel molecule by mutating genes under rationally designed selection conditions and pressures.

The term "novel molecules of new function" embraces any molecule having a previously unknown structure and/or sequence and/or physical/chemical properties and having a previously unknown or unrealized reaction type, e.g., phosphorylation, proteolysis, binding, etc., and/or specificity.

The term "functionally enhanced novel molecule" embraces molecules having a previously unknown structure and/or sequence and/or physical/chemical properties whose function, i.e., combination of reaction type, e.g., phosphorylation, proteolysis, binding, etc. and specificity is known or realized, but which differ in their degree of known function.

The term "novel molecule" includes any molecule having a previously unknown structure and/or sequence and/or physical/chemical properties or a novel molecule of new function or a functionally enhanced novel molecule and includes organic, inorganic and biomolecules such as nucleic acids, proteins, oligonucleotides, sugars, lipids, peptides and any substituted or modified versions thereof.

The terms "putative novel molecule", "putative functionally enhanced novel molecule" or "putative novel molecule of new function" mean any molecule, known or otherwise, which one skilled in the art considers to be a candidate for isolation, creation or directed evolution according to the methods of the invention.

The term "substrate of interest" includes any naturally occurring molecule or synthetic molecule, whether known or unknown, upon which a chemical and/or physical interaction is desired. The substrate of interest may comprise organic or inorganic molecules or biomolecules, e.g., proteins, oligonucleotides, lipids and polysaccharides. Substrates of interest include, for example, compounds which are known substrates for enzymatic action, peptides, polypeptides and proteins of various descriptions, which are to be reacted, cleaved, linked, modified or substituted, or bound by a novel molecule.

The term "analog of the substrate of interest" means a molecule or a portion thereof which contains a recognition sequence which renders it a functional analog of the substrate of interest.

The term "gene" is used in its broadest commonly understood meaning and includes any nucleic acid, e.g., oligonucleotides (DNA, RNA, etc.) capable of expression and further includes combinations or sets of genes.

The term "genome" refers to the entire complement of genes capable of being expressed, including chromosomal genes, plasmids, transposons and viral DNA.

The term "expression" has its generally understood scientific meaning and includes replication of oligonucleotides, transcription, translation and reverse transcription.

The term "expression by the host" means expression by any part of the host genome.

The terms "interaction" and "capable of interaction" broadly encompass any intermolecular and/or intramolecular operations including chemical reactions, catalysis or physical binding. The novel molecules sought to be created by the methods of the invention may be capable of reaction with the recognition sequence in any kind of reaction including, for example, isomerizations, additions, substitutions, syntheses and fragmentations, etc. Other "interactions" of interest may include (de)phosphorylation, (de)glycosylation, (de)hydroxylation, (de)adenylation, (de)acylation, (de)acetylation and stereo-isomerization. Such reactions may be limited to interactions between the novel molecules and the recognition sequence or may include the addition or deletion of atoms or molecules from the novel molecule and/or recognition sequence, respectively, and/or from the host cell or medium, respectively. Interactions may also include catalyses in which the novel molecule acts catalytically on the recognition sequence to effect, inter alia, proteolysis, esterolysis, hydrolysis, stereoisomerization, or to effect any of the reactions referred to above. The term interaction also includes binding interactions, e.g., as between antibodies and antigens or binding induced allosteric effects.

The term "host cell" includes a living organism which is unicellular, multicellular, procaryotic or eucaryotic, e.g., yeast, COS cells, CHO cells or hybridoma cells.

The term "growth factor" as used herein is a molecule(s) which confers a growth advantage or disadvantage upon a host cell or upon a replicator. Typical growth factors include nutrients, enzymes necessary for metabolism of nutrients, and binding and structural proteins, and proteins involved in replication, metabolism, formation and maintenance of essential structural components, or cellular and subcellular growth.

The term "recognition sequence" is used in its most universal chemical sense and means any chemical atom(s), bond(s), molecule(s), submolecular group(s), combination of any of the foregoing, or any physical or electrical state or configuration thereof, e.g., an amino acid sequence. The recognition sequence may be a discrete molecule or may be a submolecular portion of a complex molecule which includes the growth factor and/or the selection moiety.

It is essential that the recognition sequence has some functional interaction either directly or indirectly, with the growth factor and/or the selection moiety such that when the desired interaction of the novel molecule and the recognition sequence takes place, the function of the growth factor will be affected and has an impact (positive or negative) on cell or replicator growth. The function and/or the structure of the selection moiety and/or the recognition sequence may be combined and the function and/or structure of the recognition sequence and the growth factor may also be combined. As further described below, fusion or deletion proteins are particularly useful in the selection methods of the invention and represent an embodiment in which all three functions are combined in a single molecule.

Modifications to the recognition sequence enable selection of multiple novel molecules. For example, a large protein that is a substrate of interest may be able to tolerate insertion or replacement of a stretch of amino acids without losing its function, (29). It is possible to insert in such a protein a variety of different potential proteolytic recognition sequences in order to obtain multiple proteases of desired specificity.

It is also essential that the recognition sequence represent the substrate of interest. For example, the recognition sequence may be an amino acid sequence from a protein for which no known or satisfactory proteolytic enzyme exists. Incorporating that amino acid sequence or certain analogs thereof as recognition sequences in the methods of the invention, will lead to the creation of novel molecules having specificity and/or high turnover for the proteolysis of that amino acid sequence.

The term "selection moiety" or "modulation moiety" refers to any molecule which, in physical or chemical association with a growth factor, either directly or indirectly, either increases or decreases the function of that growth factor. The selection moiety may be for example a bulky protein which, because of steric hindrance and conformational changes, inactivates or functionally impairs the action of an enzyme necessary for cell or viral growth.

The terms "selection molecule" or "universal selection molecule" refer to a molecule which incorporates a growth factor and a recognition sequence and optionally a modulation moiety.

The term "artificial selection molecule" means a contrived selection molecule containing a recognition sequence which is not an inherent part of a natural or synthetic growth factor.

The term "replicator" refers to subcellular entities capable of replication, e.g., plasmids, viruses, bacteriophage, self-replicating oligonucleotides such as RNA molecules which have recognition sequences for a replicase(s), mycoplasma, etc. Replicator expression refers to expression directed by a replicator through the use of host and/or replicator components.

The terms "selection", "selection conditions" and "selection pressure" refer to generally known as well as novel procedures for growing a population of host cells in the absence of an essential nutrient, on compounds which cannot be utilized by parental strains, in the presence of toxins, in various peculiar environmental conditions, e.g., temperature, light, pH, or in the presence of mixed cultures, as may cause some but not all members of the population to survive and replicate.

A "nucleotide" is one of the five bases: adenine, cytosine, guanine, thymine and uracil, plus a sugar, deoxyribose or ribose, plus a phosphate.

An "oligonucleotide" is a sequence formed of at least two nucleotides, and a "polynucleotide" is a long oligonucleotide and may be either RNA or DNA with or without modified bases. While the term oligonucleotide is generally used in the art to denote smaller nucleic acid chains, and "polynucleotide" is generally used in the art to denote larger nucleic acid chains including DNA or RNA chromosomes or fragments thereof, the use of one or the other term herein is not a limitation or description of size unless expressly stated to be.

The term "nucleic acid" refers to a polynucleotide of any length, including DNA or RNA genomes or fragments thereof, with or without modified bases as described above.

The term "isolation/purification" refers to techniques for isolating, purifying or extracting, as these terms are conventionally used, to describe methods for recovering a gene or a molecule from a cell and/or replicator and/or a medium.

The term "mutagenesis" refers to techniques for the creation of heterogeneous population of genes, e.g., by irradiation, chemical treatment, low fidelity replication, etc.

IN THE DRAWINGS

FIGS. 1A and 1B are schematic representations of a host-selection method embodiment of the invention for the creation or directed evolution of a gene which encodes a novel molecule.

FIGS. 2A, 2B and 2C are schematic representations of a viral replicator embodiment of the invention for the creation or directed evolution of a gene which encodes a novel molecule.

FIGS. 3A and 3B are schematic representations of a further embodiment of the invention for the creation or directed evolution of a novel hydroxylase based upon a binding interaction.

FIGS. 4A and 4B are schematic representations of a cell-free embodiment of the invention.

With reference to FIGS. 1A and 1B, reference numeral 10 refers to a host cell (E. coli) having chromosome 12. The cell is engineered, as described elsewhere, to be a deletion mutant, 14, lacking the ability to express an essential growth factor. E. coil deletion mutant 14 is further engineered as shown at reference numeral 16 to encode a selection molecule which incorporates the essential growth factor for the host cell in a down-modulated form and a recognition sequence.

Reference numeral 18 refers to a plasmid which is engineered as shown at reference numeral 20 to encode T7 bacteriophage origin of replication and low fidelity T7 replication machinery. Plasmid 20 is further engineered to encode a heterogeneous population of genes which express putative novel molecules. The so engineered plasmid is shown at reference numeral 22.

Plasmid 22 is introduced into transformed deletion mutant host 16 and cultivated in a suitable environment in a nutrient-rich, non-limiting medium as shown at reference numeral 24. Plasmids which express a novel molecule having the desired functional interaction with a selection molecule expressed by deletion mutant host 16 are shown at reference numeral 26.

Incubation in the nutrient-rich, non-limiting medium in incubator 24 results in the growth of a population of host cells 16 containing plasmids 22 and plasmid 26. The population of transformed host cells are shown at reference numeral 30.

The population of incubated host cells is then selected in a chemostat in nutrient-limiting medium. This confers a growth advantage upon those transformed host cells 16 harboring plasmids 26 which encode novel molecules having the desired function of interacting with the selection molecule and up-modulating the growth factor. The selected population of host cells is shown at reference numeral 32. As can be seen, those cells which harbor a plasmid which encodes a novel molecule having the desired function have had a preferential growth advantage.

The plasmids 26 are then isolated/purified from the selected population of cells 32. The novel molecule genes are cloned, sequenced and functionally characterized.

Referring to FIGS. 2A, 2B and 2C, reference numeral 50 refers to a plasmid carrying functional T7 bacteriophage genes. Plasmids 50 are introduced into E. coli host cell 52 having chromosome 54. The transformed host is shown at reference numeral 56.

Wild type T7 bacteriophage, is shown at reference numeral 58. A population, 60, of deletion mutant bacteriophage which do not encode a growth factor essential for replicator growth or replication, are engineered to encode putative novel molecules.

The heterogeneous population of T7 deletion mutants is introduced into transformed host cell 56 which complements the function of the T7 deletion mutants and a population of those deletion mutants are incubated. The population of deletion mutants is shown at reference numeral 62. The one T7 mutant within the population which carries the gene which expresses a novel molecule which is capable of the desired interaction is shown at reference numeral 64.

Reference numeral 66 refers to a plasmid carrying the genes which express a selection molecule containing the growth factor deleted from phage 60 and 64, a recognition sequence and a modulation moiety. Plasmid 66 is introduced into E. coli host cell 68 containing chromosome 70 thereby forming a second population of transformed E. coli hosts as shown at reference numeral 72.

The T7 deletion mutant population grown up in the first population of transformed host cells which complement their deletions, are then allowed to infect cells in the second population which expresses the selection molecules. The infection step is shown at reference numeral 74 and the incubation of the infected second population of host cells is shown at reference numeral 76. Viral replication occurs only in those host cells in which the novel molecule of desired function is expressed. The process can be carried out batchwise or additional amounts of mutant T7 bacteriophage can be added to the second population of transformed host cells in a continuous fashion, as shown, until cell lysis is monitored. The lysis of a cell in the incubated second population of transformed host cells is shown at reference numeral 78. As can be seen the population of T7 deletion mutants 64 which encodes the novel molecule of desired function has been substantially amplified as shown at reference numeral 80.

The viral population expressing the desired novel molecule expands and infects other cells upon cell lysis. No new T7 is added to the culture. The expansion and infection of other cells is shown at reference numeral 82. The multiple infections give rise to more virions which carry the desired novel molecule as well as those that do not carry the novel molecule. This is shown at reference numeral 84.

The selected and amplified viral population encoding the novel molecule of desired function shown at reference numeral 86 is isolated/purified from the cultivated host cells and then grown up on cells at low dilution which express the selection molecule. This separates out single viral clones carrying the genes which encode the novel molecule with desired function. This isolation/purification is shown at reference numeral 88.

With reference to FIGS. 3A and 3B, the method depicted there is creation or directed evolution of a site-specific hydroxylase for the substrate represented by the formula R, i.e., a hydroxylase which can convert R to R--OH. Reference numeral 110 refers to a host cell (E. coli) having cell wall 112, periplasmic space 114, cytoplasm 116 and bacterial chromosome 118. Cell 110 contains substrate R in its periplasmic space as well as an antibody specific to the compound R--OH identified by reference numeral 120 to which is bound a modulated growth factor 122. Also in periplasmic space 114 is a protease 124 which is specific for the bound conformation of antibody 122.

The population of host cells 110 is transformed with plasmids which encode a heterogeneous population of hydroxylases which is replicated with low fidelity replication machinery. The step of infection is shown generally at reference numeral 126. The infected host cells, shown at reference numeral 128 contain multiple plasmids 130 in cytoplasm 116. These plasmids express a heterogeneous population of hydroxylases into periplasmic space 114. Those hydroxylases having the desired site-specific hydroxylase activity with respect to substrate R are shown at reference numeral 132 and non-desired hydroxylases are shown at reference numeral 134.

The population of transformed host cells are then incubated under selection conditions to select for and/or direct the evolution of the desired hydroxylase. This step is shown generally at reference numeral 136. The selected population of transformed host cells contains the hydroxylated substrate R, i.e., the compound R--OH as shown at reference numeral 138. In turn, antibody 120 binds the compound R--OH as shown at reference numeral 140. Protease 124, which is specific for the bound conformation of antibody 140, cleaves the modulated growth factor 122 from antibody 140 leaving the cleaved antibody 142 and the up-modulated growth factor 144 which confers a growth advantage on transformed host cell 128.

The selected population of transformed host cells 128 is then cultivated and the DNA isolated/purified. The DNA for the desired hydroxylase is cloned and the desired hydroxylase expressed and characterized.

DETAILED DESCRIPTION

The preferred embodiments of the invention are further described below with respect to the several particular features of the invention.

I. The Selection Molecule and its Component Parts

Selection molecules are used to direct selection pressure so as to obtain a desired gene. Selection molecules include growth factors and recognition sequences, and optionally modulation moieties. Some are capable of being used to select multiple different novel genes by utilizing the recognition sequence in a cassette-like fashion. Once a rational configuration of the desired components is established, steps are taken to prevent the mutation of the selection molecule. This provides a constant target for the population of novel molecules and serves to direct the selection or evolutionary process.

A. The Growth Factor

Growth factors are any factors capable of conferring a growth advantage or disadvantage upon a cell or replicator. Growth factors include nutrients such as carbon sources, nitrogen sources, energy sources, phosphate sources, inorganic ions, nucleic acids, amino acids, etc.; toxins such as antibiotics, inhibitors of enzymes critical for replication, detergents, etc.; enzymes which are essential for cell or replicator growth or which confer an advantage or disadvantage upon cellular growth, such as polymerases, ligases, topoisomerases, enzymes catalyzing reactions in the biosynthesis of proteins, etc.; molecules whose function is not catalytic but rather is structural or based on the binding capabilities of the molecule, such as actin, lipids, nucleosomes, receptors, hormones, cyclic AMP, etc.; and coenzymes or cofactors such as water, inorganic ions, NADPH, coenzyme A, etc.

B. The Recognition Sequence

The recognition sequence is a molecule or a portion of a molecule which interacts with the novel molecule. As such, recognition sequences may be a variety of different structures such as a sequence of amino acids or nucleic acids. This sequence may represent a unique sequence or it may represent a class of related sequences. The recognition sequence may also be a particular conformation or class of conformations of various molecules, e.g., a particular three dimensional structure of a protein, inorganic molecule, lipid, oligosaccharide, etc. In addition the recognition sequence may be an analog of any of the foregoing.

In addition, depending upon the selection system, the potential recognition sequences may be limited to a very specific region, conformation, sequence, etc., or may be a broad set of potential recognition sequences. For example, by using specific, multiple, redundant sequences in common to a plurality of selection molecules, with which the desired novel molecule may interact so as to modulate growth, the true recognition sequence is limited to that specific sequence common to all of the selection molecules. On the other hand, by using only one selection molecule or by using multiple selection molecules with large regions common to all, the potential recognition sequence may be a variety of regions, conformations, sequences, etc., within the one selection molecule or within the large regions common to multiple selection molecules. Recognition sequences thus may be highly specific to a region, conformation, sequence, etc., or be specific to a broader, yet defined set of regions, conformations, sequences, etc.

C. The Modulation Moiety

Central to the invention is the concept of modulation of the activity of the growth factor. There are many ways to modulate biological activity and nature has provided a number of precedents. Modulation of activity may be carried out through mechanisms as complicated and intricate as allosteric induced quaternary change to simple presence/absence, e.g., expression/degradation, systems. Indeed, the repression/activation of expression of many biological molecules is itself mediated by molecules whose activities are capable of being modulated through a variety of mechanisms.

A table of chemical modifications to bacterial proteins appears in (2), p. 73. As is noted in the table, some modifications are involved in proper assembly and other modifications are not, but in either case such modifications are capable of causing modulation of function.

In some instances modulation of functional usefulness may be mediated simply through the proper/improper localization of the molecule. Molecules may function to provide a growth advantage or disadvantage only if they are targeted to a particular location. For example, starch is a macromolecule which is typically not taken up by bacteria, so it is necessary to secrete enzymes responsible for its degradation, e.g., amylases, so that it may be converted into useable energy forms. Thus, production and retention of amylases within the bacteria down-modulates its functional usefulness when the bacteria is grown in a starch limiting media. It is only when the amylases are excreted that they are capable of conferring a growth advantage to the bacteria. The inherent enzymatic capabilities of the amylase may be the same inside or outside of the bacteria, but its functional usefulness is drastically down-modulated when it is targeted intra-cellularly relative to being targeted extra-cellularly.

Localization targeting of proteins carried out through cleavage of signal peptides is one way in which modulation of functional usefulness through molecular targeting is used within the invention. In this case, selection for a specific endoprotease catalytic activity is selected.

The functional usefulness of enzymes may also be modulated by altering their capability of catalyzing a reaction. Such a modulation may be carried out by differential localization (i.e., permissive local environment vs. non-permissive), but this need not be the mechanism. Illustrative examples of modulated molecules are zymogens, formation/disassociation of multi-subunit functional complexes, RNA virus poly-protein chains, allosteric interactions, general steric hindrance (covalent and non-covalent) and a variety of chemical modifications such as phosphorylation, methylation, acetylation, adenylation, and uridenylation ((2), p. 73, 315).

Zymogens are examples of naturally occurring protein fusions which cause modulation of enzymatic activity. Zymogens are one class of proteins which are converted into their active state through limited proteolysis ((3) p. 54). Nature has developed a mechanism of down-modulating the activity of certain enzymes, such as trypsin, by expressing these enzymes with additional "leader" peptide sequences at their amino termini. With the extra peptide sequence the enzyme is in the inactive zymogen state. Upon cleavage of this sequence the zymogen is converted to its enzymatically active state. The overall reaction rates of the zymogen are "about 10.sup.5 10.sup.6 times lower than those of the corresponding enzyme" ((3) p. 54).

It is therefore possible to down-modulate the function of certain enzymes simply by the addition of a peptide sequence to one of its termini. For example, this property may be used within the invention to select for endoproteases with desired characteristics.

The formation or disassociation of multi-subunit enzymes is another way through which modulation may occur. Different mechanisms may be responsible for the modulation of activity upon formation or disassociation of multi-subunit enzymes. Two mechanisms are illustrative.

Tryptophan synthetase is composed of two different subunits, alpha and beta, in an alpha-beta-alpha-beta tetramer. The tetramer can disassociate into two alpha subunits and a beta-beta subunit each of which exhibit catalytic activity, however, the independent subunits are substantially less efficient than the tetrameric holoenzyme. The efficiency increase of the holoenzyme is thought to be due in part to the formation of a tunnel between the alpha and beta active sites (4). Through the determination of the three dimensional crystal structure of this enzyme it appears that the tunnel prevents the loss of the intermediate product of the alpha catalyzed reaction to the solvent by channelling it directly to the beta subunit active site thus increasing efficiency.

Modulation of activity upon formation of the holoenzyme for aspartate transcarbamoylase occurs through a different mechanism. In the aspartate transcarbamoylase holoenzyme the active sites are formed at the interface of catalytic subunits. In both aspartate transcarbamoylase and tryptophan synthetase the proper specific interaction of different subunits is critical for efficient activity of the holoenzyme. Therefore, sterically hindering the proper specific subunit interactions will down-modulate the catalytic activity. Such complexes could be used within the invention for the selection of a variety of molecules.

Other examples of mechanisms through which modulation of function may occur are RNA virus poly-proteins, allosteric effects, and general covalent and non-covalent steric hindrance. The HIV virus is a well studied example of an RNA virus which expresses non-functional poly-protein constructs. In the HIV virus "the gag, pol, and env poly-proteins are processed to yield, respectively, the viral structural proteins p17, p24, and p15--reverse transcriptase and integrase--and the two envelope proteins gp41 and gp120" (5). The proper cleavage of the poly-proteins is crucial for replication of the virus, and virions carrying inactive mutant HIV protease are non-infectious (5). This is another example of the fusion of proteins down-modulating their activity. Thus, it is possible to construct recombinant viruses which require sequence dependent endoproteases for proper replication.

Certain enzyme inhibitors afford good examples of functional down-modulation through covalent steric hindrance or modification. Suicide substrates which irreversibly bind to the active site of an enzyme at a catalytically important amino acid in the active site are examples of covalent modifications which sterically block the enzymatic active site. An example of a suicide substrate is TPCK for chymotrypsin (6). This type of modulation may be used in embodiments of the invention to select for compounds capable of covalently binding to catalytically active sites or cleaving moieties from a non-active catalytic site thereby converting it into a catalytically active one.

There are also examples of non-covalent steric hindrance including many repressor molecules. Lambda repressor is of interest since it simultaneously down-modulates the expression of other phage genes such as cro while up-modulating its own expression. It accomplishes this by non-covalently binding to DNA sequences and sterically hindering the interaction of these sequences with RNA polymerase thereby preventing RNA polymerase from transcribing towards the cro genes while simultaneously stimulating the RNA polymerase to transcribe in the opposite direction. Thus the repressor molecules are capable of sterically hindering and thus down-modulating the function of the DNA sequences by preventing particular DNA-RNA polymerase interactions.

The selection of non-covalent binding compounds offers possibilities and advantages because binding molecules can be created based on their ability to modify the activities of various substrates of interest.

Allosteric effects are another way through which modulation is carried out in some biological systems. Aspartate transcarbamoylase is a well characterized allosteric enzyme. Interacting with the catalytic subunits are regulatory domains. Upon binding to CTP or UTP the regulatory subunits are capable of inducing a quaternary structural change in the holoenzyme causing down-modulation of catalytic activity. In contrast, binding of ATP to the regulatory subunits is capable of causing up-modulation of catalytic activity (7). Using methods of the invention, molecules are selected which are capable of binding and causing modulatory quaternary or tertiary changes.

In addition, a variety of chemical modifications, e.g., phosphorylation, methylation, acetylation