Display library process

Abstract

Disclosed are methods for identifying desired members from a display libraries, including bacteriophage display libraries. Display library members can be amplified in the presence of a target compound so that cycles of selection can be rapidly completed.

Claims

1. A method of selecting phage that encode a target binding protein from a plurality of phage particles, the method comprising: a) contacting a plurality of diverse phage particles with a first target to form phage-target complexes, wherein each phage of the plurality displays a heterologous protein component on its surface and each phage includes a nucleic acid encoding the heterologous protein component, and wherein each phage-target complex comprises a phage from the plurality, the heterologous protein displayed on which binds the first target; b) separating the phage-target complexes from unbound phage particles; c) contacting host cells with the separated phage-target complexes in a suitable growth medium so that the host cells are infected by the phage particles of the phage-target complexes to yield a first population of infected cells; d) incubating the first population of infected cells in the presence of a second target under conditions allowing for replication of the phage particles, thereby forming phage-target complexes that include replicate phage particles that bind to the second target; e) separating replicate phage particles that do not bind to the second target from the replicate phage-target complexes; and f) contacting host cells with the replicate phage target complexes so that host cells are infected with the replicate phage particles of the complexes to yield a second population of infected cells, wherein the second population of infected cells produce phage particles that encode heterologous proteins binding to both the first target and the second target; wherein the first target and the second target are different.

2. The method of claim 1, wherein the first target is immobilized to a support prior to step (a).

3. The method of claim 1, wherein the first target is immobilized to a support during step (a).

4. The method of claim 1, wherein the second target is immobilized to a support prior to step (d).

5. The method of claim 1, wherein the second target is immobilized during step (d).

Description

DETAILED DESCRIPTION

(1) In one aspect, the invention provides a variety of accelerated methods for identifying members having a desired binding property from display libraries, e.g., phage display libraries. Phage display libraries are used to provide a physical association between a protein to be displayed and the nucleic acid that encodes it. This nucleic acid, or “display gene”, can include a secretion signal sequence, a segment that encodes the protein to be displayed and a segment that encodes an attachment moiety (of one or more amino acid residues) that causes the encoded protein to become attached to the phage particle. The segments that encode the attachment moiety are typically constant within the library. In one embodiment, the attachment moiety is an anchor domain, such as a phage coat protein or a fragment thereof. Other methods of attachment are also available, see, e.g., Crameri et al. (1994) Eur J Biochem 226(1):53-8 and WO 01/05950.

(2) The protein to be displayed segment may be picked at will. For example, one can display linear or cyclic peptides, Kunitz domains, Kazal domains, Fabs (antibodies), conotoxins, and the like. It is possible to build phage libraries in which a very high fraction of phage particles contain a display gene according to the design of the library. In some libraries, a large fraction of the phage particles actually display the protein encoded by the insert of that phage. In phagemid libraries, it is often found that only a small fraction, e.g. 10%, of phage particles actually display the encoded library member. Nevertheless, many useful binding molecules have been selected from such libraries because the selection acts on a population that is 100 to 1000 times larger than the diversity of the library. Thus, each protein member that is genotypically represented is actually present on 10 to 100 phage particles (10% of 100 to 1000).

(3) In one embodiment, the method includes amplifying a display library member in the presence of a target compound. The method can be used to identify members of the library that interact with the target compound. In another embodiment, the method restricts amplification to a limited interval or a burst so that the selection process rapidly advances, e.g., through successive cycles. Still other features are disclosed that enable improved selection of the display library. As noted above, the methods are applicable at least to bacteriophage and to phagemids, as well as other types of display libraries.

(4) One exemplary method includes the following: a. Contacting a plurality of diverse display phage to a target compound; b. Separating phage that bind to the target compound from unbound phage; c. Infecting host cells with the bound phage; d. Producing replicate phage from the infected cells in the presence of the target compound (“phage production”); e. Separating phage that bind the target compound from the unbound phage and infected cells; f. Repeating c. to e. one or more times, e.g., one to six times; g. Recovering the bound phage, e.g., for individual characterization.

(5) In the above example, each round of replicate phage production is carried out in the presence of target compound. The host cells are typically F.sup.+ E. coli. The E. coli are maintained in the presence of the target compound during infection and during phage production by the infected cells. The cells can be maintained, e.g., for less than three division cycles, e.g., less than an hour. As soon as replicate phage emerge from the infected cells, they can bind to the target compound since the target compound is present during phage production. Moreover, it is not essential to purify or harvest the emerging phage during each cycle.

(6) The time interval for step (d) can be, for example, between 30 to 120 minutes or between 45 to 60 minutes. Experimental observations of filamentous phage have demonstrated that most phage are internalized within 15 minutes of contact between phage and E. coli. Progeny start to appear about 30 minutes after contact in a burst of about 1000 phage per cell. Progeny production drops significantly after the first 45-60 minutes, although phage may still be produced at about 100-200 per cell per generation if cells are well nourished and aerated.

(7) During the interval for phage production, the method can include a change in temperature. In an example of a phage library that displays immunoglobulins (e.g., Fab's or scFv's), the first 20 minutes can be carried out at 37° C. and a subsequent 25 minutes at 30° C. The lower temperature can improve the folding of some protein domains, e.g., an immunoglobulin domain. Also during this interval, an agent that alters a property of the host cell can be added. For example, in cases wherein production of the displayed protein is under control of an inducible promoter, then an inducer of that promoter can be introduced at an appropriate time during the incubation. In an example in which production of the displayed protein is controlled by a lacI regulated promoter, e.g., the lacZ promoter, a lac inducer, e.g., IPTG, can be added, e.g., at the tenth minute of the propagation step.

(8) For many rapid cycling methods, it is not necessary to add antibiotic during phage production. However, in some implementations, particularly those in which phage production is greater than 60 minutes (e.g., between 60 and 120 minutes), an antibiotic can be added, e.g., after the initial 10 or 15 minutes during which infection occurs. The antibiotic is used to select for cells that are infected by phage that include a nucleic acid encoding resistance to the antibiotic.

(9) The infected cells that are removed at step (e) after phage production can be reserved, e.g., for further analysis. For example, such cells can be titred, e.g., to determine the number of bound phage particles. In another example, individual cells are grown for ˜16 hours at 30° C. to produce enough additional phage for analysis by ELISA, e.g., to get a measure of the affinity of display library members for the target compound. In still another example, the cells are plated and the individual display library nucleic acids are recovered from the cells and sequenced. Isolate picking, growing, ELISA assays, sequencing and other operations can be carried out by robots (see, e.g., “Automation,” below). These methods provide a snap-shot of the selection process.

(10) In a final cycle, phage infect host cells. The infected cells are grown, e.g., in liquid culture, on plates or combination thereof, e.g., at a density that results in individual colonies. The display phage vector generally includes an antibiotic-resistance gene such as bla which confers resistance to ampicillin. The corresponding antibiotic (e.g., ampicillin) is included in the growth medium so that only infected cells are grown and the display vector is retained. Since antibiotic resistance does not appear immediately, the antibiotic can be added after a delay, e.g., about 15 to 60 minutes after infection. Phage from the final sample can be evaluated individually, e.g., by ELISA, DNA sequencing, restriction enzyme fingerprinting, and so forth. They can also be analyzed en masse, e.g., by ELISA or by reformatting (see, e.g., U.S. Ser. No. 10/383,902). Identified peptides can be produced synthetically and tested, e.g., for a biological activity or other property. Identified antibodies (e.g., Fab's) can be produced as complete antibodies from mammalian cells, e.g., antibodies including a functional Fc domain, and tested, e.g., for antibody dependent cellular toxicity and/or complement dependent cellular toxicity.

(11) Transfer of Host Cells

(12) Another exemplary method includes transferring the infected cells after infection. This embodiment differs from the first in that cells that have been infected by the phage that are bound to target in a first vessel are transferred to a second vessel before progeny phage are released from the infected cells. The burst of progeny are released in the second vessel. The second vessel can include a fresh supply of the target compound. The target compound can be present in the vessel before transfer of the infected cells, or can be added subsequently.

(13) This method can be used to reduce metabolism of the target compound (e.g., proteolysis, denaturation, or other modification of the target compound) and to provide a supply of the target compound that is free from phage, debris, and other substances that may be remnants from a previous round. The method also enables control of the amount of target compound in each cycle and the use of different target compounds in successive cycles. For example, one could use a peptide as the target compound in one cycle and a cell that expresses a protein that contains the peptide in another cycle.

(14) The method can include: a. Contacting a plurality of diverse display phage to a target compound; b. separating phage that bind to the target compound from unbound phage; c. Infecting host cells with the bound phage in a first vessel; d. Transferring the infected cells to a second vessel, e. Producing replicate phage from the infected cells in the presence of the target compound in the second vessel (“phage production”); f. separating phage that bind the target compound from the unbound phage and infected cells; g. Repeating c. to e. one or more times, e.g., one to six times; h. Recovering the bound phage, e.g., for individual characterization.

(15) To transfer infected cells from one vessel to another, the target compound can be attached to a matrix, e.g., to Sepharose beads. A medium containing the cells is added to the matrix and incubated, e.g., for 10 to 20 or 10 to 15 minutes. Then the medium is separated from the matrix and added to the second vessel. For example, the matrix can be placed in a column and washed to remove the cells. In a related example, the target compound is attached to magnetic particles. A medium containing cells is added to the particles and incubated. A magnet is then used to remove the magnetic particles. A fresh supply of target-bound magnetic particles can be added to the vessel or the cells can be transferred to a second vessel. In still another example, the target compound is attached to the surface of a vessel. A medium containing cells is added to the vessel and incubated. Then magnetic particles that have an attached ligand (e.g., an antibody that binds to an outer membrane protein of E. coli, such as OmpA, OmpT, or LamB) that binds to the cells are added to the vessels. The cells are transferred to a new vessel using a magnet.

(16) It is also possible to use target beads that have been blocked for varying times during successive cycles. The target beads include the target compound on an accessible surface. For example, the beads used in the first cycle can be blocked for one hour, whereas beads used in the second cycle can be blocked for two hours; beads used in the third cycle for three hours, and so on. Increased blocking time may increase the number of high affinity binding phage depending on the concentration of target compound bound to the beads and the temperature (Zhuang et al. 2001 J. Biosci and Bioeng 91:474-481.)

(17) Washing times and conditions can also be varied from cycle to cycle. In one example, both blocking and washing times are decreased with successive cycles to increase the number of binders obtained from the selection process.

(18) Cell-Surface Targets

(19) The target compound can also be presented on the surface of a cell, e.g., a eukaryotic cell. This approach is particularly useful for selections and subsequent screenings to find ligands that bind to a membrane associated protein. Many target proteins are membrane-bound and are difficult to produce in native form (e.g., including glycosylation and other post-translational modifications such as proteolytic processing and accompanying conformational changes), except on cells.

(20) One method for selecting members of a display library that bind to a target compound on the surface of a cell can include: a. Contacting a plurality of diverse display phage to cells that present the target compound on their cell surface; b. Separating phage that bind to the target-presenting cells from unbound phage; c. Infecting host cells with the bound phage (e.g., by contacting the host cells to the target-presenting cells); d. Separating the target-presenting cells from the host cells; e. Producing replicate phage from the infected host cells; f. Contacting the replicate phage to target-presenting cells (e.g., a new sample of such cells) g. Separating phage that bind to the target-presenting cells from unbound phage h. Repeating c. to g. one or more times, e.g., one to six times; i. Recovering the bound phage, e.g., for individual characterization.

(21) The target-presenting cells can be living or fixed. When live target-presenting cells are used, the plurality of diverse phage can be added to a medium that supports the target-presenting cells, e.g., to tissue culture medium. In some cases, agents that alter the metabolism or the behavior of target-presenting cells can also be used. For example, an agent that reduces the rate of endocytosis can be added to reduce internalization of the display phage.

(22) The eukaryotic target-presenting cells can be separated from the host cells by a variety of methods. One exemplary method is differential sedimentation. For example, a brief (and/or low speed) centrifugation can be used to sediment the target-presenting cells while leaving the host cells in solution. (See, e.g., Belshe and Mufson. (1991). Textbook of Human Virology. Mosby Year Book, St. Louis, pp. 388). Another separation method relies on differential adherence. Eukaryotic cells can be adhered to a surface, e.g., a surface that includes a coating of extracellular matrix protein (e.g., fibronectin).

(23) Generally target-presenting cells and host cells can also be separated by distinctive cell surface markers. For example, the target-presenting cells can be biotinylated and captured on streptavidin (Sv) beads. After contact with the diverse phage, the target-presenting cells on the beads are washed to remove non-binding phage. In another example, an antibody that binds to a cell-surface marker other than the target is used to attach the target-presenting cells to an insoluble support, e.g., beads.

(24) Still other methods for separating the target-presenting cells from host cells include:

(25) The host cells (e.g., E. coli) can be labeled with biotin and captured on Sv beads.

(26) The host cells (e.g., E. coli) can include an accessible surface tag, e.g., a peptide sequence that binds a known partner. The peptide sequence can be a component (e.g., a heterologous component) of an outer surface protein. For example, a peptide having the amino acid sequence of “HPQ” (Devlin et al. 1990 Science 249(4967) pp. 404-406.) can be positioned in an outer loop of OmpA, LamB, or OmpT cell surface protein. Peptides having the amino acid sequence of “HPQF” (SEQ ID NO:1) frequently bind to Sv. Koebnik (J. Bact. (1999) 181(12)3688-3694) shows a schematic of the structure of Escherichia coli OmpA. Residues that are exposed on E. coli outer membrane comprise:

(27) TABLE-US-00001 (SEQ ID NO: 2) L1 = H.sub.19DTGFINNNGPTHE.sub.32, (SEQ ID NO: 3) L2 = P.sub.62YKGSVENGAYK.sub.73, (SEQ ID NO: 4) L3 = D.sub.105TKSNVYGKNHDTG.sub.118,, and (SEQ ID NO: 5) L4 = I.sub.147GDAHYIGTRPD.sub.158.

(28) Koebnik also reports that OmpA folds and assembles into the outer membrane when each of the loops is drastically shortened. Thus, the sequence and conformation of these loops are not essential to the transport, folding, or assembly of OmpA into the outer membrane of E. coli. The base sequence of E. coli OmpA is disclosed in GenBank entry: LOCUS ECOMPA, Accession V00307 J01654, Version V00307.1 GI:42159.

(29) The amino-acid sequence WHPQFSS (SEQ ID NO:6) has been reported to bind Sv (Cwirla et al. (1990) Proc Natl Acad Sci USA, 87:6378-6382.). Cwirla et al. reported several other Sv-binding sequences and any of these could be used. The subsequence HPQ is the most important part of the motif. Thus one could, for example, replace the DNA encoding L1 with DNA that encodes the sequence L1A=H.sub.19DTG W.sub.23HPQFSS.sub.29 THE.sub.32 (SEQ ID NO:7) using strains and vectors similar to those reported by Koebnik. Note that the novel amino acids are shown bold. It is not essential to include all of WHPQFSS (SEQ ID NO:6); the loop L1B=H.sub.19DTGF HPQF GPTHE.sub.32 (SEQ ID NO:8)

(30) is likely to make cells that express this variant of OmpA bind Sv.

(31) Alternatively, L2 could be replace with L2A=P.sub.62YKGS.sub.66 W.sub.67HP.sub.69Q.sub.69aF.sub.69bS.sub.69cS.sub.69d G.sub.70AYK.sub.73 (SEQ ID NO:9)

(32) in which the loop has been lengthened. Note that residues 62-66 and 70-73 have not been altered and the Sv-binding sequence is bounded by flexible sequence (GS and GA). In and alternative example, L3 could be replaced by L3A=D.sub.105TKS.sub.108 W.sub.109HPQFSS.sub.114a H.sub.115DTG.sub.118, (SEQ ID NO:10)

(33) in which the loop is lengthened by one residue. A further example would be to replace L4 with L4A=I.sub.147GDA.sub.150 W.sub.151HPQ.sub.154F.sub.154aS.sub.154b T.sub.155RPD.sub.158 (SEQ ID NO:11) in which residues 151-154 have been replaced with six residues including the HPQ motif.

(34) Sousa et al. (J. Bact. (1998) 180(9)2280-2284.) disclose that large insertions can be made into E. coli LamB at position 153. For example, one could insert DNA encoding the amino-acid sequence GPCHPQFPRCYI (SEQ ID NO:12) at codon 153. Other binding sequences and epitopes can be used in addition to the examples provided here. Preferably such sequences can be specifically recognized and placed in an accessible position on the cell surface.

(35) The host cells (e.g., E. coli) can be captured with an antibody that binds to the host cells (E. coli).

(36) The target-presenting cells can be biotinylated.

(37) The target-presenting cells can present a marker in addition to the target compound. The target-presenting cells can be captured using a binding partner (e.g. an antibody) that recognizes the marker. In one example, the marker is a peptide, such as HPQF (SEQ ID NO:1), that binds a known protein, such as Sv.

(38) The target-presenting cells can be captured with beads known to bind adherent cells (e.g., “Microcarrier Beads” from Amersham-Pharmacia).

(39) Phage that bind to the surface of the cell in the vicinity of a specific cell surface marker can be tagged (e.g., biotinylated) and then captured, e.g., using Sv. See, e.g., Osbourn et al. (1998) Immunotechnology 3:293.

(40) In some implementations, the target-binding phage are separated from the target-presenting cells, e.g., prior to infection of the host cells. In one example, acid or base is used to dissociate the target-binding phage from the target-bearing cells. In another example in which the phage include a specific protease cleavage site between the displayed protein and the bacteriophage particle, the complex of target cells and bound phage can be treated with a protease that specifically cleaves the site. Cleavage releases the phage from the cells and enables recovery of the nucleic acids that encode proteins that bind to the target.

(41) Specificity

(42) The methods can also be used to identify proteins that bind to a particular part of the target compound (i.e. an epitope of interest) or that bind selectively, e.g., relative to a related non-target compound (such as a protein homolog). Criteria for selection may also include a particular minimum affinity, or affinity range.

(43) The following example provides a method for isolating a protein that binds to an epitope that is defined by a given ligand. a. Contacting a plurality of diverse display phage to a target compound; b. Separating phage that bind to the target compound from unbound phage; c. Providing the given ligand; d. Separating phage that dissociate from the target compound in the presence of the given ligand; e. Infecting host cells with the dissociated phage; f. Producing replicate phage from the infected cells, e.g., in the presence of the target compound (“phage production”); g. Separating phage that bind the target compound from the unbound phage and the infected cells; h. Repeating c. to g. one or more times, e.g., one to six times; i. Recovering the bound phage or, again, phage dissociated by the given ligand. Recovered phage can be individually characterized.

(44) The given ligand can be added in excess of its IQ for the target so that any phage that bind the epitope recognized by the given ligand are likely to be prevented from re-binding to the epitope after the given ligand is added. For example, a peptide that has K.sub.d=40 nM might be added at a concentration of 5 μM. By competing and preventing rebinding of phage to the epitope, phage that display proteins that bind the epitope are maintained in the supernatant. The supernatant is removed from the immobilized target and added to growing F.sup.+ E. coli. The supernatant also includes the given ligand as it is present in excess during the competition reaction. Phage in the supernatant infect the E. coli and more E. coli are produced.

(45) To remove the given ligand from the medium so that replicates of the infecting phage are produced in a medium without the given ligand, the infected E. coli cells are sedimented and then resuspended in fresh growth medium. The target compound can be included or added to the fresh growth medium. The process can then be repeated two to five more times.

(46) In other embodiments, it is also possible to include a competing compound, e.g., during phage production. The competing compound can be a compound that differs from the target compound, but is structurally similar to the target compound. In one embodiment, a model the structure of the competing compound can be superimposed on a model of the structure of the target compound with a root-mean-squared standard deviation of less than 2.5, 2, 1.5, or 1 Angstroms. In another embodiment, the target compound is a protein, and the competing compound is a related protein, e.g., a protein with at least 25, 35, 50, 60, 70, 80, 90, 95, or 99% identity to the target compound. The comparison of sequences and determination of percent identity between two sequences can be accomplished using a mathematical algorithm. In particular, the percent identity between two amino acid sequences is determined using the Needleman and Wunsch ((1970) J. Mol. Biol. 48:444-453) algorithm which has been incorporated into the GAP program in the GCG software package, using a Blossum 62 scoring matrix with a gap penalty of 12, a gap extend penalty of 4, and a frameshift gap penalty of 5. In still another embodiment, the target compound differs from the competing compound by less than 5, 4, 3, or 2 amino acids.

(47) Phagemids

(48) In some implementations, the diverse phage include phagemid particles. These implementations can include the addition of helper phage or other methods that supply bacteriophage genes necessary for producing replicate particles. An exemplary method includes: a. Contacting a plurality of diverse display phagemids to a target compound; b. Separating phagemids that bind to the target compound from unbound phagemids; c. Infecting host cells with the bound phagemids; d. Adding helper phage; e. Producing replicate phagemids from the infected cells in the presence of the target compound (“phagemid production”); f. Separating phagemids that bind the target compound from the unbound phagemid, helper phage, and infected cells; g. Repeating c. to e. one or more times, e.g., one to six times; h. Recovering the bound phagemids, e.g., for individual characterization.
Immobilization

(49) In many methods, the target compound becomes fixed to an immobilized support, e.g., before or after addition of the diverse phage.

(50) Non-covalent attachment. A target compound can be attached to an insoluble support by a non-covalent interaction. This interaction can occur either before, during or after the target compound is contacted to a display library.

(51) The target compound can be attached to one member of a binding pair. The insoluble support includes the other member of the binding pair. When the target compound and the insoluble support are mixed, the complementary binding pairs interact and attach the target compound to the support. The binding interaction between biotin and Sv is an exemplary high affinity non-covalent binding interaction. Other exemplary non-covalent interactions that can be used for binding pairs include: glutathione-S-transferase and glutathione (see, e.g., U.S. Pat. No. 5,654,176), hexa-histidine and Ni.sup.2+ (see, e.g., German Patent No. DE 19507 166), and an antibody and a peptide epitope (see, e.g., Kolodziej and Young (1991) Methods Enz. 194:508-519 for general methods of providing an epitope tag).

(52) Covalent Attachment. A target compound can also be covalently attached to an insoluble support. For example, the insoluble support might include a chemically reactive group, e.g., a thiol reactive, amine-reactive, or carboxylate-reactive group. The reactive group can be used to attach to a functional group on the target compound. For example, a thiol-reactive reagent such as iodoacetamide can be used to bond to a free cysteine on a target compound. Another exemplary reagent is NHS-sepharose affinity chromatography resin.

(53) The target compound can be immobilized to any insoluble support, e.g., a matrix, bead, resin, planar surface, or immunotube. The insoluble support can be treated, e.g., coated with a blocking agent to prevent non-specific binding to the support. The target compound can be in solution (i.e., not immobilized) while phage are being produced. For example, the target compound can be biotinylated and contacted to immobilized Sv, e.g., after binding to phage in solution. In another example, the target compound is a chimeric protein, e.g., a GST-fusion protein. As a chimeric protein, the target compound can also be immobilized to an appropriate insoluble support after solution phase binding to phage.

(54) In one embodiment, the display library is depleted of members that interact with the insoluble support in the absence of the target compound prior to selecting members of the library that bind the target using a support that includes the target compound. For example, the display library can be passed over Sv coated beads to remove members that bind to Sv. Then the library can be contacted to Sv coated beads that are bound by a biotinylated target compound to select library members that interact with the target compound.

(55) It is also possible to add additional target compound during the phage production (step (d)). For example, if the target compound is bound to beads, additional target beads can be added. The addition of target compound can counter possible loss of functional target compound during exposure to the host cells.

(56) Selection Methods

(57) In one embodiment, magnetically responsive particles (e.g., paramagnetic beads) are used to capture display library members. The particles can include a capture reagent, e.g., which binds to the target compound or the target compound can be directly attached to the particle. In one example, Sv coated magnetic particles are used. The target compound is covalently linked to a biotin moiety. The target compound can be attached (before or after binding a display library member) to the magnetic particle. Magnetic particle separation can be automated, e.g., using a magnetic particle processor such as the KingFisher™ system of Thermo LabSystems, e.g., as described below.

(58) In another embodiment, the display library is panned against the surface of a vessel, e.g., the bottom of a well of a microtitre plate. The method can be automated by a liquid handling device that can be used to add, remove, and transfer liquid to and from the vessel.

(59) In still another embodiment, the display library is contacted to the target compound in a flow chamber, e.g., a chromatography column. Liquid can be flowed through the chamber at appropriate times, e.g., to separate binders from non-binders, and to deliver and/or remove host cells, etc. The flow chamber can also be closed, e.g., to prevent flow and allow for agitation of the chamber.

(60) Automation

(61) Various devices can used to automate the methods described herein. These devices include multi-well plate conveyance systems, magnetic bead particle processors, liquid handling units, colony picking units, and other robotics. These devices can be built on custom specifications or purchased from commercial sources, such as Autogen (Framingham Mass.), Beckman Coulter (USA), Biorobotics (Woburn Mass.), Genetix (New Milton, Hampshire UK), Hamilton (Reno Nev.), Hudson (Springfield N.J.), Labsystems (Helsinki, Finland), Packard Bioscience (Meriden Conn.), and Tecan (Mannedorf, Switzerland).

(62) In some cases, the methods described herein include an automated process for handling magnetic particles. The target compound is immobilized on the magnetic particles. The KingFisher™ system, a magnetic particle processor from Thermo LabSystems (Helsinki, Finland), for example, can be used to select display library members against the target. The display library is contacted to the magnetic particles in a tube. The beads and library are mixed. Then a magnetic pin, covered by a disposable sheath, retrieves the magnetic particles and transfers them to another tube that includes a wash solution. The particles are mixed with the was solution. In this manner, the magnetic particle processor can be used to serially transfer the magnetic particles to multiple tubes to wash non-specifically or weakly bound library members from the particles. After washing, the particles can be transferred to a vessel that includes a medium that supports display library member amplification. In the case of phage display the vessel may also include host cells.

(63) In some cases, e.g., for phage display, the processor can also separate infected host cells from the previously-used particles. The processor can also add a new supply of magnetic particles for an additional round of selection.

(64) The use of automation to perform the selection can increase the reproducibility of the selection process as well as the through-put.

(65) An exemplary magnetically responsive particle is the Dynabead® available from Dynal Biotech (Oslo, Norway). Dynabeads® provide a spherical surface of uniform size, e.g., 2 μm, 4.5 μm, and 5.0 μm diameter. The beads include gamma Fe.sub.2O.sub.3 and Fe.sub.3O.sub.4 as magnetic material. The particles are superparamagnetic as they have magnetic properties in a magnetic field, but lack residual magnetism outside the field. The particles are available with a variety of surfaces, e.g., hydrophilic with a carboxylated surface and hydrophobic with a tosyl-activated surface. Particles can also be blocked with a blocking agent, such as BSA or casein to reduce non-specific binding and coupling of compounds other than the target to the particle.

(66) The target is attached to the paramagnetic particle directly or indirectly. A variety of target molecules can be purchased in a form linked to paramagnetic particles. In one example, a target is chemically coupled to a particle that includes a reactive group, e.g., a crosslinker (e.g., N-hydroxy-succinimidyl ester) or a thiol.

(67) In another example, the target is linked to the particle using a member of a specific binding pair. For example, the target can be coupled to biotin. The target is then bound to paramagnetic particles that are coated with Sv (e.g., M-270 and M-280 Streptavidin Dynaparticles® available from Dynal Biotech, Oslo, Norway). In one embodiment, the target is contacted to the sample prior to attachment of the target to the paramagnetic particles.

(68) In some implementations, automation is also used to analyze display library members identified in the selection process. From the final sample, individual clones of each display member can be obtained. Each member can be individually analyzed, e.g., to assess a functional property. Exemplary functional properties include: a kinetic parameter (e.g., for binding to the target compound), an equilibrium parameter (e.g., avidity, affinity, and so forth, e.g., for binding to the target compound), a structural or biochemical property (e.g., thermal stability, oligomerization state, solubility and so forth), and a physiological property (e.g., renal clearance, toxicity, target tissue specificity, and so forth) and so forth. Methods for analyzing binding parameters include ELISA, homogenous binding assays, and SPR. For example, ELISAs on a displayed protein can be performed directly, e.g., in the context of the phage or other display vehicle, or the displayed protein removed from the context of the phage or other display vehicle.

(69) Each member can also be sequenced, e.g., to determine the nucleic acid sequence of the encoded protein that is displayed.

(70) Additional exemplary methods of automating a library screens and selections are described in U.S. 2003-0129659.

(71) Target Compounds

(72) Generally, any molecular species can be used as a target compound. The target compound can be of a small molecule (e.g., a small organic or inorganic molecule), a protein, a nucleic acid, cells, and so forth. By way of example, a number of examples and configurations are described for targets. Of course, target compounds other than, or having properties other, than those listed below can also be used.

(73) One class of target compounds includes polypeptides. Examples of such target compounds include small peptides (e.g., about 3 to 30 amino acids in length), single polypeptide chains, and multimeric polypeptides (e.g., protein complexes).

(74) A protein target can be modified, e.g., glycosylated, phosphorylated, ubiquitinated, methylated, cleaved, disulfide bonded and so forth. Preferably, the polypeptide has a specific conformation, e.g., a native state or a non-native state. In one embodiment, the polypeptide has more than one specific conformation. For example, prions can adopt more than one conformation. Either the native or the diseased conformation can be a desirable target, e.g., to isolate agents that stabilize the native conformation or that identify or target the diseased conformation. In some cases, however, the target compound is unstructured, e.g., adopts a random coil conformation or lacks a single stable conformation. Agents that bind to an unstructured protein can be used to identify the protein when it is denatured, e.g., in a denaturing SDS-PAGE gel, or to separate unstructured isoforms of the protein for correctly folded isoforms, e.g., in a preparative purification process.

(75) A few exemplary protein targets include: cell surface proteins (e.g., glycosylated surface proteins or hypoglycosylated variants), cancer-associated proteins, cytokines, chemokines, peptide hormones, neurotransmitters, cell surface receptors (e.g., cell surface receptor kinases, seven transmembrane receptors, virus receptors and co-receptors, extracellular matrix binding proteins, or a cell surface protein (e.g., of a mammalian cancer cell or a pathogen). In some embodiments, the polypeptide is associated with a disease, e.g., cancer.

(76) Target compounds that are proteins and that can be produced by a bacterial source can be produced by the host cells that support phage replication. For example, the host cells can be transformed with a recombinant nucleic acid that encodes the target compound. Expression of the recombinant nucleic acid results in secretion of the target compound into the periplasm and extracellular milieu where it can bind to replicates of the display phage. Such target compounds can then be immobilized, e.g., by a ligand (such as an antibody) that binds to a region of the target compound for which binding proteins are not being sought or by a member of a binding pair which binds to a corresponding member that is attached to the target compound. For example, the target compound can be produced as a fusion protein to a tag (e.g., a 6HIS or GST fusion protein). After secretion, the target compound can be immobilized to an insoluble matrix that includes a binding member that binds to the tag. Examples of such tags and binding members include: glutathione-S-transferase, and glutathione; chitin binding protein and chitin; Cellulase (CBD) and cellulose; maltose binding protein and amylose, or maltose; dihydrofolate reductases and methotrexate; FKBP and FK506.

(77) Types of Display Libraries

(78) May aspects of the method described here can be applied to any type of display library. Exemplary display libraries include phage display, cell display, ribosome display, and mRNA-protein fusions.

(79) Phage Display. One format utilizes viruses, particularly bacteriophages. This format is termed “phage display.” The protein component is typically covalently linked to a bacteriophage coat protein. The linkage results form translation of a nucleic acid encoding the protein component fused to the coat protein. The linkage can include a flexible peptide linker, a protease site, or an amino acid incorporated as a result of suppression of a stop codon. Phage display is described, for example, in Ladner et al., U.S. Pat. No. 5,223,409; Smith (1985) Science 228:1315-1317; WO 92/18619; WO 91/17271; WO 92/20791; WO 92/15679; WO 93/01288; WO 92/01047; WO 92/09690; WO 90/02809; de Haard et al. (1999) J. Biol. Chem 274:18218-30; Hoogenboom et al. (1998) Immunotechnology 4:1-20; Hoogenboom et al. (2000) Immunol Today 2:371-8; Fuchs et al. (1991) Bio/Technology 9:1370-1372; Hay et al. (1992) Hum Antibod Hybridomas 3:81-85; Huse et al. (1989) Science 246:1275-1281; Griffiths et al. (1993) EMBO J 12:725-734; Hawkins et al. (1992) J Mol Biol 226:889-896; Clackson et al. (1991) Nature 352:624-628; Gram et al. (1992) PNAS 89:3576-3580; Garrard et al. (1991) Bio/Technology 9:1373-1377; Rebar et al. (1996) Methods Enzymol. 267:129-49; Hoogenboom et al. (1991) Nuc Acid Res 19:4133-4137; and Barbas et al. (1991) PNAS 88:7978-7982.

(80) Phage display systems have been developed for Ff filamentous phage (phage fl, fd, and M13) as well as other bacteriophage (e.g. T7 bacteriophage and lambdoid phages; see, e.g., Santini (1998) J. Mol. Biol. 282:125-135; Rosenberg et al. (1996) Innovations 6:1-6; Houshm et. al. (1999) Anal Biochem 268:363-370). The filamentous phage display systems typically use fusions to a minor coat protein, such as gene III protein, and gene VIII protein, a major coat protein, but fusions to other coat proteins such as gene VI protein, gene VII protein, gene IX protein, or domains thereof can also been used (see, e.g., WO 00/71694). In a preferred embodiment, the fusion is to a domain of the gene III protein, e.g., the anchor domain or “stump,” (see, e.g., U.S. Pat. No. 5,658,727 for a description of the gene III protein anchor domain). It is also possible to physically associate the protein being displayed to the coat using a non-peptide linkage, e.g., a non-covalent bond or a non-peptide covalent bond. For example, a disulfide bond and/or c-fos and c-jun coiled-coils can be used for physically association (see, e.g., Crameri et al. (1993) Gene 137:69 and WO 01/05950).

(81) The valency of the protein component can also be controlled. Cloning of the sequence encoding the protein component into the complete phage genome results in multivariant display since all replicates of the gene III protein are fused to the protein component. For reduced valency, a phagemid system can be utilized. In this system, the nucleic acid encoding the protein component fused to gene III is provided on a plasmid, typically of length less than 7000 nucleotides. The plasmid includes a phage origin of replication so that the plasmid is incorporated into bacteriophage particles when bacterial cells bearing the plasmid are infected with helper phage, e.g. M13K07. The helper phage provides an intact copy of gene III and other phage genes required for phage replication and assembly. The helper phage has a defective origin such that the helper phage genome is not efficiently incorporated into phage particles relative to the plasmid that has a wild type origin.

(82) Bacteriophage displaying the protein component can be grown and harvested using standard phage preparatory methods, e.g. PEG precipitation from growth media.

(83) After selection of individual display phages, the nucleic acid encoding the selected protein components, by infecting cells using the selected phages. Individual colonies or plaques can be picked, the nucleic acid isolated and sequenced.

(84) Cell-Based Display. In still another format the library is a cell-display library. Proteins are displayed on the surface of a cell, e.g., a eukaryotic or prokaryotic cell. Exemplary prokaryotic cells include E. coli cells, B. subtilis cells, spores (see, e.g., Lu et al. (1995) Biotechnology 13:366). Exemplary eukaryotic cells include yeast (e.g., Saccharomyces cerevisiae, Schizosaccharomyces pombe, Hanseula, or Pichia pastoris). Yeast surface display is described, e.g., in Boder and Wittrup (1997) Nat. Biotechnol. 15:553-557.

(85) In one embodiment, variegated nucleic acid sequences are cloned into a vector for yeast display. The cloning joins the variegated sequence with a domain (or complete) yeast cell surface protein, e.g., Aga2, Aga1, Flo1, or Gas1. A domain of these proteins can anchor the polypeptide encoded by the variegated nucleic acid sequence by a transmembrane domain (e.g., Flo1) or by covalent linkage to the phospholipid bilayer (e.g., Gas1). The vector can be configured to express two polypeptide chains on the cell surface such that one of the chains is linked to the yeast cell surface protein. For example, the two chains can be immunoglobulin chains.

(86) In one embodiment, yeast display cells are amplified in the presence of a target compound. For example, a library of display cells can be contacted to a target compound. Cells that bind can be separated from ones that do not, e.g., under conditions that do not support cell division. After separation, the binding cells can be amplified, e.g., by providing conditions that do support cell division. After or during amplification, an insoluble support that includes the target compound can be washed to remove cells that do not bind or that do not bind stringently.

(87) In one example, the condition that is altered is temperature. For example, binding may be at 4° C., whereas amplification may be 30° C. In another example, the yeast display cells may include a temperature sensitive mutation, e.g., a mutation that prevents cell division at high temperature such as CDC28, CDC37, or CDC14. Cells can be maintained at the restrictive temperature during binding and then at the permissive temperature for amplification.

(88) Ribosome Display. RNA and the polypeptide encoded by the RNA can be physically associated by stabilizing ribosomes that are translating the RNA and have the nascent polypeptide still attached. Typically, high divalent Mg.sup.2+ concentrations and low temperature are used. See, e.g., Mattheakis et al. (1994) Proc. Natl. Acad. Sci. USA 91:9022 and Hanes et al. (2000) Nat Biotechnol. 18:1287-92; Hanes et al. (2000) Methods Enzymol. 328:404-30. and Schaffitzel et al. (1999)J Immunol Methods. 231(1-2):119-35. The in vitro translation process can be effected in the presence of the target compound. RNA amplification, e.g., using Qβ polymerase, can also be effected in the presence of the target compound.

(89) Protein-Nucleic Acid Fusions. Another format utilizes protein-nucleic acid fusions. Protein-nucleic acid fusions can be generated by the in vitro translation of mRNA that include a covalently attached puromycin group, e.g., as described in Roberts and Szostak (1997) Proc. Natl. Acad. Sci. USA 94:12297-12302, and U.S. Pat. No. 6,207,446. The mRNA can then be reverse transcribed into DNA and crosslinked to the protein.

(90) Again, the in vitro translation process can be effected in the presence of the target compound. Nucleic acid amplification, particularly amplification methods at 42° or less (e.g., an isothermal amplification method, such as rolling circle amplification), can also be effected in the presence of the target compound.

(91) A variety of proteins can be displayed by a display library. The methods described herein are applicable to a library that displays any type of protein, including libraries that display linear and cyclic peptides, small proteins, large proteins, and antibodies. The libraries can be constructed by diversifying a protein scaffold, by derivation from cDNA or genomic nucleic acid, or by providing a diverse set of synthetic sequences (e.g., random peptides).

(92) Scaffolds. Exemplary scaffolds for display can include: antibodies (e.g., Fab fragments, single chain Fv molecules (scFv), single domain antibodies, camelid antibodies, and camelized antibodies); T-cell receptors; MHC proteins; extracellular domains (e.g., fibronectin Type III repeats, EGF repeats); protease inhibitors (e.g., Kunitz domains, ecotin, CMTI (Cucurbida maxima trypsin inhibitor) and so forth); TPR repeats; trifoil structures; zinc finger domains; DNA-binding proteins; particularly monomeric DNA binding proteins; RNA binding proteins; enzymes, e.g., proteases (particularly inactivated proteases), RNase; chaperones, e.g., thioredoxin, and heat shock proteins; and intracellular signaling domains (such as SH2 and SH3 domains). Another useful type of scaffolding domain is the immunoglobulin (Ig) domain. Methods using immunoglobulin domains for display are also known (see, e.g., Haard et al. (1999) J. Biol. Chem 274:18218-30; Hoogenboom et al. (1998) Immunotechnology 4:1-20. and Hoogenboom et al. (2000) Immunol Today 21:371-8).

(93) Artificial Peptides. The binding ligand can include an artificial peptide, e.g., an artificial peptide of 30 amino acids or less. The artificial peptide can include one or more disulfide bonds. Other artificial peptides, so-called “linear peptides,” are devoid of cysteines. artificial peptides may have little or no structure in solution (e.g., unstructured), heterogeneous structures (e.g., alternative conformations or “loosely structured), or a singular native structure (e.g., cooperatively folded). Some artificial peptides adopt a particular structure when bound to a target molecule. Some exemplary artificial peptides are so-called “cyclic peptides” that have at least disulfide bond, and, for example, a loop of about 4 to 12 non-cysteine residues.

(94) Peptide sequences that bind a molecular target are selected from a phage-display library. After identification, such peptides can be produced synthetically or by recombinant means. The sequences can be incorporated (e.g., inserted, appended, or attached) into longer sequences.

(95) Flow Chamber

(96) In one embodiment, a flow chamber is used to amplify a display library member. For example, the flow chamber includes an inlet for flowing in a medium for cell growth and wash solutions, and an outlet for disposal of the medium and for collecting samples of interest, e.g., from a elution. The flow chamber may also include an aperture into which a magnetic field inducer can be inserted into the flow chamber in order to capture magnetically responsive particles in a zone surrounding the sheath. The flow chamber can further include a stirrer, e.g., which is rotated in order to circulate the growth medium and wash solutions through the chamber. When the magnetic field inducer is removed, the stirrer can be used to agitate the magnetically responsive particles. The flow chamber can also include an oxygenation system, a temperature controller, and glucose, pH and CO.sub.2 monitors.

(97) The flow chamber can be used to select members of a cell-display library. For example, a yeast cell display library that displays immunoglobulin molecules is mixed in the flow chamber with magnetically responsive particles that include a target compound. The magnetic field inducer is inserted into the sheath to capture the particles. PBS (phosphate buffered-saline) or another wash solution is flushed through the chamber. Flow is arrested, and the magnetic field is released to agitate the particles. Then the magnetic field is reapplied. This wash method can be repeated for a number of cycles.

(98) After washing, media, e.g., YEPD or minimal media, can be flowed through the chamber under conditions that allow the cells to grow and multiply. After an interval sufficient for a desired number of cell divisions elapses, the wash method can be repeated, e.g., with a more stringent wash solution.

(99) The system can include repeatedly washing and growing of the cells as required. The use of the flow chamber for cell growth and multiplication allows for immediate amplification of cells that bind to the magnetically responsive particles and can obviate the need for some of the external manipulations that are required for multiple selection cycles. As a result the cells are amplified in the presence of the target which is attached to the magnetically responsive particles. Similarly other replicable entities can be grown in the flow chamber (or any flow chamber herein) so that the replicable entities can be amplified in the presence of a target compound. For example, phage display library members can be amplified by addition of host cells (and helper phage for phagemids) for sufficient time that a burst is formed.

(100) Fractions of the solutions (e.g., media or wash solutions) that are emerging from the flow chamber can be collected. In particular, fractions can be taken during any wash step, and during the most stringent washing step, i.e., an elution step. The fraction can be plated, e.g., onto agar plates with yeast media, in order to form colonies from the individual cells that are present in the fraction. The fraction can also be diluted into liquid media to amplify the cells outside of the growth chamber.

(101) In another embodiment, the flow chamber has a single port, e.g., a flask with an upper opening that is sealable. Fluid is removed and provided from the single port. In still other embodiments, the flow chamber has multiple ports, e.g., three or more ports.

Example 1

(102) Approximately 100 μl of Dynal Sv coated magnetic beads (Dynal M280) were blocked with 500 μl of 2% milk in PBS (MPBS) for 30 minutes. Following a wash step to remove the excess milk, 3.Math.10.sup.11 Fab-displaying phage from a Fab-fragment phage display library diluted in MPBS was incubated with the blocked Sv beads in a total volume of 500 μl for 1 hour at room temperature. The Sv magnetic beads were collected and the unbound phage were removed by aspiration. The beads were washed three times in 1×PBS followed by the addition of 500 μl of XL1Blue-MRF′ cells (Stratagene) at an OD.sub.600 of 0.50 in 2×YT. The mixture was incubated at 37° C. for 15 minutes at which time 5 μl of 100 mM IPTG was added to achieve a final concentration of 1 mM IPTG. At 20 minutes, the bacteria were transferred to a 30° C. air shaker for an additional 25 minutes for a total incubation time of 45 minutes. The bacteria were removed and the beads were washed three times with 500 μl of 0.01% Tween-20 PBS. An additional 500 μl of XL1Blue-MRF′ cells were supplemented to the beads and the process of incubating and washing was repeated for a total of 3 rounds. The 500 μl of phage infected bacteria from each round were titered on ampicillin-containing plates as well as grown overnight in 10 mL of 2×YT containing 1 mM IPTG at 30° C. The resulting phage were purified by standard PEG precipitation.

(103) Three parallel experiments were carried out. In the first, the temperature was held at 37° C. for 20 minutes and then dropped to 30° C. for 25 minutes. In the second, the temperature of incubation was held constant at 37° C. for 45 minutes. In the third, the temperature was held constant at 30° C.

(104) The titres of the various rounds are shown in Table 1. In the headings, “37×20+30×25” denotes the experiment in which the first 20 minutes of incubation was at 37° C. and the final 25 minutes was at 30° C., “37×45” denotes the experiment in which 37° C. was used for 45 minutes, and “30×45” denotes the experiment in which 30° C. was used for 45 minutes.

(105) TABLE-US-00002 TABLE 1 37 × 20 + 30 × 25 37 × 45 30 × 45 cfu round 1/foi 5.8e7/1.95E−4   .sup. 8e7/2.7E−4  .sup. 5.1e7/1.7E−4 cfu round 2/foi 5.1e5/0.009 .sup.  1.2e6/0.015 7.5e5/0.012 cfu round 3/foi 7.3e4/0.14  .sup.  5.4e4/0.045 4.8e4/0.063

Example 2

(106) A target protein that includes a nickel chelating tag was contacted to a phage display library that displays Fabs. The mixture was then bound to nickel magnetic beads. After washing the beads three times, XL1 Blue MRF′ cells were contacted to the beads. Phage produced by these cells were allowed to bind to the target protein on the beads. The XL1 Blue MRF′ cells were removed. The phage-target-bead complexes were washed to remove unbound phage. For a second round of amplification, fresh XL1 Blue MRF′ cells were contacted to the beads. This was repeated one more time, thus providing a total of three rounds of amplification.

(107) The output FAb-displaying phage from both rounds two and three from either of the selection campaigns were screened by ELISA to determine whether the isolated phage bound to the target protein. The target protein was passively immobilized on Immulon 2 HB 96-well ELISA plates (Thermo Labsystems) overnight at 4° C. The plates were blocked for thirty minutes with phosphate buffered saline containing 3% (w/v) bovine serum albumin (BSA) and 0.05% (v/v) Tween-20. Overnight bacterial growths of FAb-displaying phage were then incubated with the target protein for 1 hour at room temperature. FAb-displaying phage were detected with an anti-M13 HRP-conjugated antibody. In this particular example, the hit rate for both rounds 2 and 3 was about 69%.

Example 3

(108) The following is an example of a method of amplifying a display library of phagemids in the presence of a target compound. The method includes the following steps: 1) Mix phagemid library with biotinylated target 2) Capture target and binding phage on Sv beads 3) Wash away non-binding phagemid (cold target can be used for a limited time to elute weak binders; as many washes as needed can be performed) 4) Add F+ cells and growth medium 5) Incubate for time T1 (between 30 min to 120 minutes, optionally with antibiotic which could be added after time T2 to select for infected cells) 6) Aliquot cells into empty vials (one per target) and plate as round 1 7) Set round counter R=1 8) Into each vial, add helper phage and target on Sv beads (if necessary additional target could be added after the helper phage, burst of phagemid is expected ˜30-45 minutes after addition of helper phage) 9) At time T3, wash away cells and non-binding phage (optionally can use cold target-wash for a limited time to elute weak binders) 10) Add cells and GM 11) Incubate for time T4 (e.g., 30 min to 120 minutes; probably=T1) (antibiotic can be added after time T5 (probably=T2) to select for infected cells) 12) Make round counter R=R+1 13) Transfer an aliquot of cells to new counter 14) Plate an aliquot of cells for colonies as round “R” (2, 3, . . . ) 15) Go back to step 8 as needed.

(109) Other embodiments are within the following claims.