ES2349573T3 - METHODS OF GENERATING PROTEIN EXPRESSION MATRICES AND THE USE OF THE SAME IN QUICK DETECTION. - Google Patents

Abstract

A method of generating a protein selection consisting of: (a) providing a set of protein-encoding DNA molecules that are well labeled in the N or C terminal with a portion of the marker molecule; (b) express the labeled individual proteins in a specially separated format; (c) purify and immobilize each labeled protein in a spatially defined format to produce a selection of proteins on a solid support, and thus the specific interactions between the label and the solid support are used both in purification and in immobilization.

Description

The present invention refers to new methods for generating protein expression selections, as well as the use of such selections in rapid medical reviews.

Projects related to genomic mapping are revolutionizing the process of discovering therapeutic objectives and with it the process of drug discovery. As new therapeutic objectives are identified, the analysis of a high production of existing and combinatorial chemical libraries will suggest many potential major compounds that are active against these objectives. Clearly, it is not economical to pursue all major compounds even through clinical trials in the first phase; However, there are currently no rapid methods to evaluate such major compounds in terms of their profiles of possible activity against all proteins in the body. If available, this method would allow potential toxicology profiles of all major compounds to be analyzed in a first phase and this information would greatly improve the process of deciding which compounds to look for and which compounds to set aside.

There is a complementary need in the pharmaceutical industry to identify all existing drug targets (already available in the market or still in development) and, based on that, define their mechanism of action. The availability of this information will greatly facilitate the process of obtaining regulatory approval for new drugs since it is very clear that regulatory bodies now consider that a knowledge of the mechanism of action is of paramount importance. In addition, this type of information allowed to design better second generation drugs. This follows because most drugs have at least minor side effects, which are probably the result of combining the drug or a metabolite of the drug with unwanted targets; All of these objective proteins need to be identified in order to define the criteria needed to design improved drugs. However, at present, there are no simple methods to generate this information and a large number of potential drugs that cost millions of dollars stay on the road simply due to lack of knowledge of the objective of action.

Protein-protein interactions are increasingly recognized as being vitally important in determining cellular responses to both internal and external stresses. Therefore, specific protein-protein interactions represent potential targets for drug-mediated intervention in infections and other disease states. Currently, the yeast double hybrid assay is the only reliable method to evaluate protein-protein interactions but in vivo assays of this type will not be easily compatible even in a low productivity format with the identification of specific agonists or antagonists of interactions. protein-protein Functional proteomic expression selections, or "proteome chips", will allow the specificity of protein-protein interactions and the specificity of any drug-mediated effect to be determined in an in vitro format. Therefore, they will have enormous potential because they will simply revolutionize this area of research.

One way in which functional proteomic selections could be generated is to individually clone, express, purify and immobilize all proteins expressed in the specific proteome. However, here an important initial consideration concerns the absolute size of the genome of interest along with considerations about the availability of sequential data for the entire genome. By way of illustration of these points, a usual bacterial genome has ~ 5Mbp and now a small number have been completely sequenced (for example Helicobacter pylori, Escherichia coli and Mycobacterium tuberculosis); fungal genomes normally have ~ 40Mbp, mammalian genomes in ~ 3Gbp and plant genomes in ~ 10GBP. Current estimates establish that the human genome sequence will end around 2003, although it is questioned that much of this information reaches the public domain. Clearly, it is not practical to expect that genomes of elements other than organisms that serve as a representative model will be available in a real time frame, since from the perspective of functional proteomics, model organisms have only limited value. So, while in principle over the next four years it may be possible to design and synthesize primers to clone each of the ~ 100,000 genes in the human genome from DNA libraries, in practice it will be very expensive (the cost of only the primers are around several million dollars) and an enormously laborious process, even if the necessary sequential data is available.

But what happens to those pharmaceutically relevant organisms for which complete sequential data will not be available? Functional proteomics cannot simply ignore them so what are the alternatives? In principle, cDNA expression libraries can be used in conjunction with non-specific immobilization to create a protein selection, but this technology is significantly limited due to the fact that non-specific immobilization is normally associated with loss of function because it affects the protein fold. In addition, all host cell proteins will also be immobilized which, at best, will significantly reduce signal-to-noise radii, and in the worst case, will result in a confusion of positive results. Therefore, the ability to create a functional proteome selection in which individual proteins are specifically immobilized and purified through a motif or label without affecting function and without requiring knowledge of the complete genome sequence will represent a breakthrough. in the field of functional proteomics.

We have currently developed a new technique that solves the problems described above by providing a methodology that allows each protein in a proteome to be labeled with a common marker at a defined position in the protein without requiring any prior knowledge about the DNA sequence of The corresponding genes. This "tag" can then be used to give similarity and specificity to the immobilization that is carried out downwards and to the purification procedures, which in turn allows the creation of spatially defined selections in which hundreds of proteins of a certain proteome.

Special consideration should be given to the fact related to the price positioning of the "label". If the tag is inserted within the frame in a gene but in an undefined and arbitrary position, there is a risk that the resulting labeled protein will be truncated in an undefined way and in most cases the correct fold will be destroyed, and by It the function. The methodology described here saves this problem by inserting the label immediately after the start codon and immediately before the stop codon of a particular gene so that the labeled and full-length individual proteins fold correctly and thus can retain function when they are specifically frozen in the selection.

Because each protein in the selection will be fully functional, then the selections can be analyzed directly to identify the objectives of the drugs or other biologically relevant molecules. The spatial definition of the selections will allow the phenotype of each protein to be directly related to its genotype, thus enabling the identification of “targets” or “impacts”.

WO 00/04382 describes protein selections, which include proteins that are immobilized on a solid support by means of a label that can be located at the N or C terminal. The proteins are purified and then printed on an organic film. WO 99/39210 describes the generation of a secondary selection of antibodies from a primary protein selection. The proteins are purified and then immobilized in the selection by means of a tag. WO 99/45130 describes a baculovirus expression system, which includes a system for expressing labeled proteins.

Therefore, in a first aspect, the present invention provides a method of generating a protein selection consisting of:

(a) provide a set of coding DNA molecules

of proteins that are well labeled in terminal N or C with

a portion of marker molecule;

(b) express the individual proteins labeled in a format

especially separated;

(c) purify and immobilize in each single step each protein labeled in a spatially defined format to produce a selection of proteins on a solid support, and thus the specific interactions between the label and the solid support are used in both purification and in the immobilization. In some embodiments, the method includes cloning and

expression of one or more proteins as full-length proteins that are labeled individually at the N or C terminal with a marker molecule.

The marker molecule can be a peptide sequence, for example, a hexa-histidine tag, an antibody epitope or a biotin mimic, or even a complete protein, or protein domain, for example the protein domain that binds to maltose. The marker molecule itself can be modified posttranslationally, e.g. eg, by the addition of a biotin or lipid molecule. In a preferred embodiment, the marker molecule will also allow purification of "tagged" proteins.

Therefore, the methods of the present invention allow the specific modification, in a container, of each member of a DNA library so that no knowledge about the sequence of individual genes is necessary. Instead, it is based on the common codon of start and end in all genes. The modification will be carried out in the form of a precise insertion, within the structure, of additional and known sequential DNA either immediately after the start codon or immediately preceding the termination codon of each cDNA as required.

The additional DNA will encode a marker fraction, which will be in the same reading frame as each individual cDNA product. Therefore, each genetically produced cDNA according to the methods of the present invention will encode an individual protein that now has a common fraction, e.g. eg, a polypeptide, a "tag" fused precisely with its N or C terminals. Because all members of the cDNA library will be modified in precisely the same way, the result will be that all proteins encoded by the cDNA library they will now be labeled with a common fraction at terminals N or C.

The selections formed by the methods described herein are described.

Such selections comprise the "tagged" protein expression library, immobilized, usually on a solid support. The person skilled in the art will understand that there is a range of possible solid supports commonly used in the area of selections and any of these "substrates" can be used in the production of selections.

As set forth herein, the methods of the present invention allow the labeling of all proteins in a specific proteome specifically in the N or C terminal. While some proteins may not tolerate the extensions of the N terminal and others may not tolerate Terminal C extensions, it is likely that the vast majority of proteins tolerate one or the other of these extensions. However, existing methods for cloning libraries simply cannot address this problem since they clone genes such as full-length and unmodified or random cDNAs and fusions almost inevitably truncated with a protein partner. In comparison with these, the present methods allow precise and full-length cDNA libraries to be created as fusions for, for example, a desired peptide pair. Compared to the former, the method to immobilize proteins in a selection as described herein is through specific rather than non-specific interactions, and these specific interactions are a function of the tag added to the terminals of each cDNA. In addition, the methods described herein can be used to screen purified and immobilized proteins that have been expressed in non-bacterial host organisms and assist in maintaining function through proper folding and post-translational modification, while existing methods such as phage display or λ-cDNA expression libraries are limited to bacterial hosts in which the majority of eukaryotic proteins are synthesized in a non-functional form, due to their incorrect fold or to an incorrect posttranslational modification.

The methods of the present invention have a wide range of potential in vitro applications that can be broadly divided into three main areas. These are the study of protein-ligand interactions, the study of protein-protein interactions, and the study of protein-DNA interactions. Protein-Ligand Interactions

The methods described here will allow a rapid design of the interactions between a new chemical entity and all proteins in a given proteome. This can be easily achieved by probing the appropriate selection of proteome with NCE following several rigors in which a high performance inverse screen can be considered. Reading this screen will be directly useful in many situations, some of which are described below.

High-performance screening programs in which compound libraries are tested against whole cells or organisms often identify initial molecules that give rise to a phenotypic change without the objective being known before the scan. However, the subsequent identification of the primary objective can be a very laborious process. The methods of the present invention can be applied directly to this type of problem since it will be possible to cross a selection of functional proteomes for the species in question and then screen this selection with the main compound to identify which proteins within the proteome it is directed to. This enormously parallel technique to identify protein-ligand interactions will greatly accelerate and simplify the determination of primary NCEs targets, and also allow the identification of weaker secondary interactions that may also be important. In addition, the methods can be applied directly to the issue of cross-reactivity of species, allowing a potential antifungal compound, for example, to be rapidly evaluated in terms of its interactions with, for example, all proteins in a human proteome; This type of information has many possibilities of being very useful in any subsequent optimization of main compounds.

High-performance scanning methods now allow rapid identification of small molecules that bind to a specific protein that has previously been recognized as a potential therapeutic target. However, these methods do not answer the question of how selective an interaction can be, this knowledge being potentially crucial in deciding whether to pursue a particular principal compound.

or not; Popular wisdom states that compounds that target simple proteins are more likely to show fewer side effects than those that also affect a large number of related or unrelated proteins.

There are a number of examples of compounds that have progressed successfully through third-phase clinical trials but have failed to gain regulatory approval because their primary mechanism of action is unknown. The antidepressant drugs mianserin and trazodone and the tenidap drug from the Pfizer company are examples, each representing investments of hundreds of millions of dollars that have not been recovered. The methods described here can potentially be applied to the resurrection of such failed drugs since if the primary objectives of these drugs can be discovered and subsequently verified in terms of the mechanism of action, data from such expensive clinical trials will already be eligible for regulatory approval.

All existing drugs have side effects, to a greater or lesser extent, here being the example, but the attractive clozapine drug to fight schizophrenic. If the molecular origin of these side effects can be determined, the design of future generation drugs with optimized primary effects combined with minimized side effects would be greatly facilitated. Again, the methods described here can be directly applied to these problems since by creating a profile of the interactions between a compound and all the proteins in a proteome, aberrant secondary interactions will be identified and subsequently analyzed to determine if they are related to Known side effects

The methods of the present invention can also be used to identify families of proteins, such as serine proteases, by screening proteome selections with generic inhibitors. This would allow the subsequent development of biochips showing, for example, all human serine proteases or, alternatively, all kinases or all p450 enzymes for a more focused exploration of major compounds. A biochip p450, for example, would be useful in assessing whether a given main compound has a chance of being metabolized or not, since p450-mediated hydroxylation often represents the first step in this process and is probably one of the first sources of variability. patient-to-patient in response to the drug; Now, in fact, one of the goals of drug design is to generate compounds that are not metabolized in the first place, and here, again, a p450 chip would have significant potential utility. Protein-Protein Interactions

Protein-protein interactions and multiprotein complexes are vitally important in cell biology. Signaling pathways, for example, normally begin with an interaction between a cell surface receptor and an external ligand, and this is followed by a cascade of protein-protein interactions that ultimately result in the activation of a specific gene. Protein-protein interactions may depend on the presence of a specific ligand or, alternatively, may be blocked by the action of a specific ligand, while some multiprotein complexes will only be formed in a ligand-dependent manner.

Hundreds of interactions have been identified using dual hybrid technologies. The methods described herein overcome the limitations of such methods and can be used to screen proteome selections with labeled individual proteins in order to identify not only interactive partners but also the relative forces of individual interactions. The methods can also be applied for the identification of complex components with multiproteins, even when their assembly depends on the ligand.

An example of the use of the methods in which it identifies protein-protein interactions would be the identification of the signaling partners of the cytosolic domain of a particular cell surface receptor that has been implicated in a disease state; the identification of such signaling partners would be directly relevant from a pharmaceutical perspective since protein-protein interactions could immediately represent possible therapeutic objectives. Protein-DNA interactions

It has been estimated that approximately 10% of all genes in the human genome encode transcription factors although only a small percentage of these are currently identified. The linking of specific transcription factors with DNA processing elements, often in response to external stimuli, is a prerequisite for the formation of enhancer complexes that then activate gene expression. There are several points where gene expression may in principle be affected by drug administration: a drug could block the binding of a protein or small molecule with a cell surface receptor and, as a result, block the signaling cascade at the beginning; a drug could block a protein-protein interaction or inhibit an enzymatic activity within the signaling cascade; or alternatively, a drug could block the formation of specific protein-DNA or protein-protein interactions in the enhancer complex. Here, by way of example, the NFкB transcription factor participates in various cellular and immune processes and in responses to inflammation, limb development, septic shock, asthma and HIV propeptide production. Most intracellular signaling cascades in the activation of NF-κB are common to all processes so that they do not represent viable targets for intervention. Therefore, the differences between the responses reside in the original ligand-receptor interaction or in the formation of enhancer complexes. It is known that NF-κB binds at least 14 different enhancer elements and, therefore, enhancer complexes represent potential therapeutic targets. However, the delineation of an individual enhancer complex requires knowledge of the number of individual proteins that bind to DNA and also their protein-protein interactions with each other. The present methods can be used to directly answer both questions. A proteomic selection can be screened with specific DNA probes to identify new proteins that bind to DNA. Alternatively, proteomic selection can be screened with the transactivation domain of a particular transcription factor to identify other proteins with which it interacts. The cross relationship of these screens should allow the identification of new components of specific enhancer complexes.

Protein selections generated by the methods of the present invention will also allow the selection of molecules, which recognize each of the proteins shown in the selections. In a preferred embodiment, the selected molecules will be antibody proteins and antibody-like elements and will be exposed in phages or ribosomes or

they will covalently bind with the mRNA encoder.

Therefore, a phage-exposed antibody library can be applied to each of the immobilized proteins in the selection and the antibodies that do not combine are removed by washing. Then, the selected phage can be recovered and used to infect bacteria according to usual procedures. Phage-infected bacteria can either produce phage particles that show the selected antibodies during more rounds of selection, or they can produce soluble antibody fragments for direct use. The terms "antibody" or "antibody fragments" herein refer to Fvs fragments, single chain FABs, light chain individual fragments

or heavy, derived from mouse, human, camel or other organisms.

In a preferred embodiment, the protein selection will have a micro-well format so that, after the first selection step, the phage particles can be recovered by adding appropriate bacterial cells to each well where they will be infected by the action of the selected phage particles. Then, the growth medium can be added to each well and the infected bacteria are allowed to grow and express the antibody fragments, while maintaining the physical separation of the selected antibody fragments from each immobilized protein in the selection. If desired, new phage particles produced by infected bacteria can be used in subsequent rounds of selection. These procedures are now routine in the selection of polyclonal and monoclonal antibody fragments in a single purified and immobilized protein. Next, and in practice, the original protein selections presented here will allow the generation of polyclonal and monoclonal antibody fragments for hundreds of correctly folded proteins in a greatly parallel manner whereas, if not, the standard methods of in vitro antibody selection.

The antibody fragments selected and soluble in each well of the original selection can be immobilized themselves in a new spatially defined selection so that the antibody fragments in each position of the new selection were selected against the immobilized proteins in a single position. defined in the original selection. The antibody selections so selected will contain either polyclonal or monoclonal antibody fragments in each position, depending on the number of selection rounds made before immobilization of the soluble antibody fragments.

These antibody selections will have a number of potential uses that include the capture of individual proteins from a crude cell or a cell lysate controlling the differential expression of the relevant proteome. Alternatively, the proteins captured by antibody can be analyzed directly for ligand binding function. In general, any monoclonal antibody could be combined with the target protein to block its function, but another monoclonal antibody could combine but not block the function. In a very parallel technique, it is clearly not practical to evaluate all monoclonal antibodies with all proteins in a proteome individually to check their binding ability without affecting function. In contrast, a polyclonal set of antibodies for all proteins in a proteome has the potential to contain individual antibodies that have the desired ability to combine but not to affect function and, in addition, will contain individual antibodies that recognize all post-translational modifications of a certain protein Therefore, in general, selections of polyclonal antibodies rather than monoclonal antibodies generated as described herein will possibly present advantages for screening proteins captured directly for function.

Compared to the original protein selection, antibody selections created by the methods described herein will have the advantage that all proteins immobilized in the selection will be stable under similar conditions. Proteins captured from the crude cell or tissue lysate will not be recombinant but will be expressed naturally. In addition, the captured proteins can be screened for function or linkage with ligand, etc., directly after the capture of the crude cell or tissue lysate, which should help to maintain the function.

(i)

In some embodiments, the method of the invention further comprises the step of contacting one or more compounds with a selection of proteins as defined herein and measuring in bond of one or more compounds with the proteins in the selection; in this way one or more compounds are analyzed for biological activity.

(ii)

In some embodiments, the method of the invention further comprises the step of contacting one or more proteins, e.g. eg, a cell surface receptor with a selection as defined herein, and measuring the linkage of one or more specific proteins with the selection proteins; in this way a

or more proteins for specific protein-protein interactions.

(iii) In some embodiments, the method of the invention further comprises the step of contacting one or more nucleic acid probes with a selection as defined herein, and measuring the linkage of the probes with the selection proteins; in this way a

or more proteins for specific protein-nucleic acid interactions.

(iv)

The use of a selection as described herein is described for rapid selection of a compound, protein or nucleic acid.

(v)

The use of a selection is described as defined herein to select molecules that recognize each protein in the selection, where the molecules are preferably antibodies.

(saw)

In some embodiments, the method of the invention further comprises the step of contacting a selection of proteins, as defined herein, with

an antibody library, so that one or more proteins in the protein selection bind to at least one antibody in the antibody library, removing all unbound antibodies and immobilizing those antibodies that bind to the proteins in the protein selection, thereby generating a selection of antibodies.

(vii) A method for selecting protein function or abundance is described which comprises the step of contacting a selection of antibodies such as those defined herein with a mixture of one or more proteins.

Methods (i), (ii), (iii) and (vi) may also include the step of first providing the selection according to one or more methods of the present invention.

The preferred features of each aspect of the invention are applicable to each of the aspects, with the corresponding changes.

In the following, the present invention will be described with reference to the following examples, which in no case should be considered as limiting the scope of the invention.

FIGURE 1a: shows the construction of the vector pMM106H;

FIGURE 1b: shows the details of PCR amplification and exonuclease digestion of a gene by way of examples (GST) before labeling;

FIGURE 1c: shows details of the specific ligation and PCR amplification to introduce the label; FIGURE 1d: shows details of the cloning of PCR products; Y

FIGURE 1e: shows the reaction between Glutathione and 1-chloro-2,4-dithitrobenzene catalyzed by GST. Example 1

(a) Vector Construction (see Figure 1a)

We have constructed a pMM106H vector derived from pUC19 that contains a strong hybrid promoter (Ptrc) to drive the expression of the cloned genes to an Nco I point immediately below the promoter sequence. We insert a 676 bp nonsense DNA sequence as a filler fragment between point Nco I and a descending point Hpa I. Hpa I is a cutter with a pointed tip and is located to split the vector so that the descending DNA encodes a polyasparagine , hexahistidine peptide if the reading frame is on the first phase of the dull tip. After the hexahistidine label, we found an amber stop codon (TAG) followed by the green fluorescent protein (GFP) coding gene of the jellyfish Aequorea victoria. The genes cloned in pMM106H as Nco I fragments / blunt tip result in fusions with the His and GFP tag only if the correct reading frame is created at the Hpa I point during cloning. Here, GFP is used as a reporter gene to facilitate visual exploration of clones that express the His tag, while also providing an indication of the correct fold of the fusion protein, since GFP is only active when folded in the correct conformation. . The amber stop codon will result in a small amount of full-length fusion proteins for visualization of green colonies, while most of the fusion proteins will end immediately after the His tag and can be used for subsequent immobilization and for enzyme tests. The construction of pMM106H was confirmed by sequences:

We constructed a second pGSTN vector using a first PCR amplification of Schistosoma japonicum glutathione S transferase (GST) gene from pGEX-2T (Pharmacia) under standard conditions using primers “GSTfwd2” (5 '-ATG CTG CAG ACG TCA ACA GTA TCC ATG GCC CCT ATA CTA GG-3 ') and' GSTHindIII '(5' -GCG AGG AAG CTT GTC AAT CAG TCA CGA TGA ATT CCC G-3 '). These primers introduce an Nco I restriction site in the GST start codon, mutate the second GST residue from serine to alanine, and introduce a stop codon at the 3'-multiple cloning site of the GST gene followed by a place restriction Hin dIII. The PCR product is then cloned under standard conditions as an Nco I / Hin dIII fragment to pTrcHisA (Invitrogen) previously digested with Nco I / Hin dIII to generate pGSTN.

(b) PCR amplification and digestion of exonuclease genes before labeling (see figure 1b)

We amplify the GST gene of the pGSTN construct using a polymerase chain reaction with specific vector primers designed “ST-forward” (5 '-ATG CTG ACG TCA TGA GGC CCA TGG GGC CCG GAT AAC AAT TTC ACA CAG G3') and 'STtrasero' (5'-GCG GAT CCT TGC GGC CGC CAG GCA AAT TCT GTT T-3 ') which are combined with vector 156 bp on the start codon and 84 bp under the stop codon, respectively. 30 cycles of PCR (94 ° C 1 min; 57 ° C 1 min; 72 ° C 2 min) were carried out in four separate 100 µl reactions. Each PCR reaction contained ~ 20 ng of DNA template, 50 pmol of each primer and 2.5 units of Pwo polymerase. Each PCR reaction was performed in a standard buffer (10 mM Tris.HCl pH 8.8, 25 mM KCl, 5 mM (NH4) 2SO4, 2 mM MgSO4, 10% DMSO). In addition, each of the four PCR reactions also contained a non-standard mixture of deoxynucleotide triphosphate, as follows: Reaction 1) 200µM dATP, 200µM dTTP, 200µM dCTP, 150µM dGTP, 50µM α-S-dGTP; Reaction 2) 200µM dATP, 200µM dTTP, 200µM dGTP, 150µM dCTP, 50µM α-S-dCTP; Reaction 3) 200µM dATP, 200µM dGTP, 200µM dCTP, 150µM dTTP, 50µM α-S-dTTP; Reaction 4) 200µM dGTP, 200µM dTTP, 200µM dCTP, 150µM dATP, 50µM α-S-dATP.

The inclusion of a single α-thio deoxynucleotide triphosphate in each specific PCR mixture results in an arbitrary but statistical incorporation of α-S-dNTP in the specific final PCR product. Next, the four individual PCR mixtures were pooled and purified using a QIAquick PCR cleaning kit (Qiagen), under standard conditions, and digested to completion with the restriction enzyme Aat II. Subsequently, the resulting ~ 1000bp PCR products were gel purified.

5 µg of digested PCR product was then incubated with 375 units of Exonuclease III for 45 minutes in a 50 µl reaction. Exo III digestion was carried out in a standard reaction buffer (66mM Tris.HCl pH 8.0, 6.6mM MgCl2, 5mM DTT, 50µg / ml bovine serum albumin). These conditions ensure that digestion by Exo III comes to an end. The enzyme was then deactivated by heating it at 75 ° C for 15 minutes. The Exo III digestion product is a nested set of deletions from the 3 ’end of the PCR product; the 5 ’end of the PCR product is protected from digestion since restriction with Aat II leaves a protrusion at point 3’ that is also resistant to Exonuclease III activity.

Exonouclease III is a non-processing 3'-a 5'-exonuclease that is incapable of hydrolyzing nucleotides containing α-thio so in the present protocol, every time Exo III reaches an α-thio-deoxynucleotide base, truncation The progressive 3 'end of the PCR product stops. Therefore, the network result is a nested set of deletions as a result of the arbitrary incorporation of each α-S-dNTP in the early phase. The radius of α-S-dNTP with dNTP used in the original PCR amplifications was determined empirically so that the envelope of nested deletions encompassed a 4000bp window of centered sizes approximately 100bp shorter than the original full-length PCR product. We confirm this by taking a portion of the Exo II mixture and treating the nested deletions with mung bean nuclease. This process removed the 5 'and 3' protrusions to produce products with a pointed tip that were then measured on 1% agarose / TBE gels, using a 100bp DNA chain as standard.

Clearly, a number of different nuclease activities 3 ’and 5’ could be used to generate the necessary set of eliminations embedded in 3 ’in the procedure described above; they are included but not limited to Exonuclease III, E. coli DNA polymerase I, T4 DNA polymerase, T7 DNA polymerase.

(c) Specific ligation and PCR amplification to introduce labels (see figure 1c)

5 µl of the Exo III reaction mixture was diluted in T4 DNA ligase buffer in the presence of a molar excess multiplied by 25 of an "oligo mixture". The "oligo mixture" consists of any of the 2 subsets of oligonucleotides. The first subset, "oligomezclaA", contains 12 oligonucleotides in which each of the three possible stop codons are represented at the 5 'end, followed immediately by a degenerate base. The remaining region of each of the oligos is the same in the 12 cases and is indeed an arbitrary sequence except for two restriction enzyme sites Type IIS (Sap I and Bam I), followed by a complementary recognition sequence for primer "LMB2" (5'-GTA AAA CGA CGG CCA GT-3 ') at the 3' end. The

15 sequences of the 12 oligos are the following:

image 1

35 Sap I Bam I binding site LMB2

The second subconjutno, "oligomezclaB", consists of 3 sets of oligonucleotides, each of which with one of the 3 stop codons represented at the 5 'end, followed immediately by 6 degenerated residues. The remaining region of each of the oligos is the same in the 3 sets and contains two Type IIS restriction enzyme sites (Bpm I and Bse RII), followed by a complementary recognition sequence for the "LMB2" primer in the 3 'end. The sequences of the 3 sets of oligos are as follows:

image2

Bse RI Bpm I binding site LMB2

The Exo III mixture plus the oligomix A or oligomix B was aligned for 30 minutes at 16 ° C and then 400 Units of T4 DNA ligase were added, and the reaction was then incubated overnight at 16 ° C. Ligation products were purified using a QIAquick PCR cleaning kit (Qiagen) under standard conditions and used as a template in a standard PCR reaction using Pwo polymerase with "STdelantero" and "LMB2" primers. 30 cycles of PCR (94 ° C, 1 min; 57 ° C 1 min; 72 ° C 2 min) were performed to generate PCR products that range from 1000bp.

Both oligomix A and oligomix B are capable of competitively aligning with DNA regions with a single chain of the original template exposed by Exo II hydrolysis of a duplex chain performed in the price step. After alignment, a correct ligation between an oligo and a template requires that the oligo be brought in line with absolute complementarity at its 5 'end and, in addition, that the residue embedded in 3' of the duplex template is directly adjacent to residue 5 'of the specifically aligned oligo. PCR that uses the primer that binds with the newly ligated oligo and a second primer that binds at the 5 ’end of the duplex template selectively and specifically amplifies then only those duplexes that have undergone such ligation. The 5 ′ end of each of the 12 oligos in the oligomexclaA corresponds to a stop codon and therefore, of the 12 oligos contained in this mixture, only one can align with absolute complementarity at its 5 ′ end with the sequence of recognition with four base pairs consisting of the first structural stop codon of GST and base immediately 3 of the stop codon, as shown in Figure 1c. The rest of the 11 oligos can be perfectly aligned at their ends 5 in all places within the nested set of eliminations but these other specific facts to align can only occur in stop codons outside the frame in the GST gene or in stop codons by under the first stop codon inside the frame. Whenever a newly aligned oligo directly collides with the residue embedded in 3 ’of the duplex template, ligation takes place. Therefore, PCR in this phase will amplify not only the exact and full-length gene but also a set of truncated and extended products. Oligos in oligomixB are expected to react in the same way. The 5 ’end of each of the 3 sets of oligos in the subset corresponds to a stop codon, followed by 6 residues that are indeed arbitrary. Therefore, the subset will contain a permutation in which the stop codon and the next 6 residues will coincide perfectly with those of the lower part of the gene of interest. This oligo will bind to a higher specificity compared to the corresponding oligo of the subset of 12 oligos, since the complementarity extends over 9 nucleotides instead of 4.

The theoretical difference between oligomezclaA and oligmezclaB lies in the sequence that immediately goes after the stop codon at the 5 ’end of each oligo. In oligomix A, a simple degenerate base is then followed by a simple nonsense but defined DNA sequence so that it is possible that this defined region can influence when aligning the oligomix in favor of individual "stop" codons ( whether within or outside the frame) within any given gene providing complementarity with the involuntary but additional base pair beyond the four interactions designed with base matching at the 5 'end of each oligo. Any partiality in the alignment can be manifested in the downward direction in a deviation towards clones in which the specific excision and replacement of an individual stop codon has occurred (either inside or outside the frame). Such a deviation in the frequency of modification in different stop codons may not be adequate and oligomixB is designed to overcome this difficulty as follows. Any stop codon in an individual gene or library will be immediately followed by a defined but unknown sequence. The three stop codons are represented in oligomezlcaB and each of them is immediately followed by all possible hexanucleotide sequences (i.e., by a hexameric sequence) so that at any given stop codon in a gene or library, there will be an oligo in oligomezclaB that will correspond to the stop codon and, exactly, to its precise and inferior sequence, resulting in 9 base pairs of total complementarity. Because this will be true for all stop codons, oligomixB should not suffer any deviation of the type that could be possible with oligomezlcaA.

We have discovered that the overall process of digestion, annealing and ligation of Exo III oligomixes and the specific PCR amplification have been reproducible to a high degree. As controls in this process, we have shown that if Exonucleas II, T45 DNA ligase or any of the oligomixes is omitted, we obtain absolutely no PCR products. This shows that this process is highly selective.

It will also be appreciated that the T4 DNA ligase could be replaced by a number of different DNA ligases, for example Taq DNA ligase or Tsc DNA ligase, which could show different specificities.

(d) Variation in the procedure

Clearly, a number of different 3 ’and 5’ activities could be used to generate the necessary set of built-in and nested 3 ’eliminations in the procedure described above; these include but are not restricted to Exonuclease III, E. coli DNA polymerase I, T4 DNA polymerase, T7 DNA polymerase.

An obvious variation in the original procedure for producing full length inserts with the stop codon or start codon removed accurately involves the production of a nested set of 5 'embedded eliminations in a PCR product that encompasses the coding sequence. Potentially, this could be accomplished using nucleases such as λ exonuclease, E. coli DNA polymerase I, Taq DNA polymerase, T7 gene exonuclease 6. Therefore, the original PCR product, for example, could be digested with λ exonuclease, which removes the nucleotides from the dsDNA chain in a 5 ′ to 3 ′ direction. Because the enzyme only recognizes 5 ’phosphate groups as substrates, one of the two chains can be protected by incorporating a hydroxyl group at the end of the appropriate primer in the initial PCR amplification. Therefore, a nested set of deletions could be created, as described above, for exonuclease III, unless the opposite chain was digested to leave 5 'embedded terminals. A mixture of oligos in which the complement of the stop codon is represented at the 3 'end of the oligo, is immediately preceded by 6 random residues and, before them, by a defined sequence encoding a Type IIS restriction point that can then align with the exposed regions of the single chain. An oligo of the mixture will specifically align with each exposed stop codon and serve as a primer for polymerase I DNA E. coli, which polymerizes in the 5 'and 3 direction, digesting the chain in front of it with its exonuclease activity in 5 'and 3'. This new DNA duplex fragment can be specifically amplified using a primer that binds to the 5 ′ end of the original PCR product and one that specifically binds to the newly aligned and extended oligo. In this procedure, it is not necessary that the 3 ’end of the aligned oligo directly collides with the 5’ recessed residue of the duplex template. The oligos of the mixture can be aligned with any complementary region on the exposed DNA template with a single chain in a manner similar to the arbitrary preparation of hexamers, but it is important that the chain extension will only occur where the 3 'end it has been aligned with absolute complementarity and the 3 'ends of the oligos in the mix are designed to be only complementary to the stop codons. Therefore, only when the 3 ’end of an oligo has been specifically linked to the corresponding stop codon, will the primer extension and subsequent PCR amplification take place. The link with stop codons outside the frame and those below the true stop codon of the gene of interest will also take place, resulting in the PCR amplification of not only the exact full length gene, but also a set of truncated products and extended.

(e) Cloning and analysis of PCR products (see figure 1d)

PCR products (~ 5µg) ranging in size from 800 to 1000 bp were cleaned using a QIAquick (Qiagen) PCR purification kit and then digested to completion with the Bpm I restriction enzyme of Type IS. This enzyme cuts remotely but specifically 14 bases of its recognition sequence in one chain and 16bp in the other chain leaving a 3 'embedded end. Therefore, this modification in the restriction specifically removes from the PCR product the stop codon in which the 5 ’end of an individual oligo of the“ oligo mixture ”was aligned and ligated successfully in the previous step.

The 3 'embedded ends of the digested products were then removed using mung bean nuclease under standard conditions to generate a dull end at the 3' end of the PCR products. The DNA was purified using a QIAquick (Qiagen) PCR purification kit under standard conditions and then digested to the end with restriction enzyme Nco I. Restricted DNA fragments ranging between 800bp and 1000bp were then purified on a 1% agarose / TBE gel using a QIAquick gel extraction kit (Qiagen). The vector pMM106H (3 µg) was digested until its completion with the enzymes Nco I and Hpa I and the 2870bp axis fragment was gel purified. The vector DNA and the restricted PCR products were subsequently ligated under standard conditions and the ligation mixture was used to transform E. coli DH5α cells that were then recovered and placed on LB plates containing 100 µg / ml carbenicillin.

This cloning procedure was carried out on the complete set of PCR products obtained in the previous step. However, only PCR products derived from specifically aligning and ligating an oligo with the first stop codon within the frame should be able to cause structural fusions with the hexahistidine and GFP tag after cloning steps through it. process; The rest of PCR products cloned in this way should only result in fusions outside the frame labeled with hexahistidine and GFP. This occurs because ligation of the dull end of the PCR product with the dull end of the vector results in a genetic fusion in which the translation reading frame of the descending DNA vector is dictated by the original reading frame of the suppressed stop codon. . If the stop codon were outside the frame with respect to the GST gene, the newly added hexahistidine coding sequence would also be outside the frame with respect to the GST gene, while if the stop codon were within the frame with respect to the GST gene , the newly added hexahistidine coding sequence would also be found within the framework with respect to the GST gene. However, only those products in which the specifically suppressed stop codon was the first stop codon within the GST framework can cause a hexahistidine (and GFP) tagged GST fusion protein when the aDN is transcribed and transferred. The only hexahistidine (and GFP) tagged proteins that may arise from the specific global process described above will be consequently and necessarily full length GST fusions with the poliasparagine, hexahistidine tag.

Colonies obtained from the cloning procedure described above were visualized at 365 nm to identify fluorescent green colonies. Ninety colonies (white and green) were collected arbitrarily, subjected to plaque replication and analyzed by Western blotting of colonies under standard conditions using anti-His tag antibodies and anti-GST antibodies. The anti-His tag antibody will only bind to colonies that express the hexahistidine-tagged protein so that the Western blot give direct information on the number of colonies that express fusions within the framework with the hexahistidine tag. On the other hand, the anti-GST antibody binds near the C-terminal of the GSTE protein and therefore only recognizes colonies that express full-length or almost full-length GST proteins. We identified those colonies that contained a protein that was positively recognized by both the anti-His tag antibodies and the anti-GST antibodies. The DNA of these colonies was amplified, purified and sequenced. Sequencing data confirmed the presence of two perfect fusions within the framework with full-length GST, that is, clones in which the first stop codon within the framework of the original GST gene had been specifically suppressed and replaced by a polyasparagine within of the frame, hexahistidine label. Therefore, the rate of successful modification we obtained through this global procedure is approximately 2.2%. The two positive clones turned out to be fluorescent green after exposure to long-wave ultraviolet light, due to the expression of sufficient amounts of the GSThexahistidine-GFP full-length fusion. Therefore, in subsequent experiments, only fluorescent green colonies were collected for analysis with Western blot. We have discovered that approximately 70-80% of the fluorescent colonies express proteins recognized by the anti-His tag antibody. It is possible that in the rest of 20-30% of cases, the translation begins in the first ATG of the GFP gene, regardless of the hexahistidine tag, possibly with the help of a pseudo-ribosome binding point introduced by the cloned insert .

We have amplified, purified and sequenced plasmid DNA from fluorescent green colonies that express protein recognized by both anti-His tag antibodies and anti-GST antibodies. For insertions prepared using both oligomezclaA and oligomezclaB, the rate of successful modifications was approximately 25% of all green colonies. Therefore, the use of the GFP gene as a marker for fusions within the framework increases the efficiency, approximately 10 times, in detecting the correct clones.

It will also be appreciated that Mungo Bean Nuclease could be replaced by other different single stranded nucleases, for example, S1 or RNasaT nuclease, which could show different specificities and that a number of different properly positioned Type IIS restriction enzymes could be used in BpmI place, for example SapI.

(f) Immobilization and functional analysis of labeled proteins (see figure 1e)

E. coli DH5α cells were transformed with one of the full length GST plasmids and with hexahistidine tag, created from the above-mentioned methodology. A single carbenicillin-resistant colony grew in a logarithmic or exponential phase in 10ml of liquid culture and was then supplemented with 100µM IPTG to cause GST expression with hexahistidine tagging. After growth for an additional 4 hours, the cells were harvested and lysed by freezing defrosting / lysozyme. SDS-PAGE of crude lysate showed an overexpressed protein of the expected size (27kDA), which represented approximately 20% of the total soluble protein, as well as a small amount of the 54 kDa GSThexahistidine-GFP fusion, generated through amber suppression . The crude lysate (500 µl; 100 µg) was then mixed with magnetic drops Nickel-NTA (50 µl; 15 µg binding capacity protein labeled with hexahistidine) and the drops were recovered by sedimentation under a magnetic field. The supernatant was discarded and the drops were washed and subsequently resuspended in a test buffer with glutathione S transferase containing 1 mM respectively of glutathione and 1-chloro-2,4-dithitrobenzene. At the time of the end of the test, the data was collected after 30 minutes at room temperature by measuring the absorbance at 340 nm; This wavelength corresponds to λmax of the reaction product catalyzed with GST.

As controls, cultures of DH5α that contained either the matrix vector (pMM106H) or a plasmid encoding an unrelated His tag protein (alanine racemase) were cultured, induced, harvested, lysed and tested in parallel. The GST activity was only detected in the drops that had been mixed with the crude lysate containing His tag GST, clearly demonstrating that the observed GST activity was specifically due to the immobilized His tag GST and also that the protein retained the activity on the specific immobilization.

After completion of the enzymatic assay, the protein was eluted from the magnetic drops by adding buffer containing 100mM imidazole and analyzed with SDS-PAGE. This showed that the sample that resulted in a positive activity in the assay contained a single immobilized protein of the exact size expected for glutathione S transferase (27kDa), thus confirming that the activity observed in the drops was due solely to this recombinant protein with His tag. Example 2

(a) Modification, immobilization, and GST assay using two different labels

Following the procedure described in Example 1 to modify glutathione-S-transferase with hexahistidine tag, we have demonstrated that this procedure is independent of the precise nature of the label being added.

First, two vectors that were identical to pMM106H were constructed except that the non-sense DNA filler fragment Nco I / Hpa I 676bp was replaced by an Nco I / Hpa I 300bp fragment derived from the Escherichia coli gdhA gene and the hexahistidine tag it was replaced by the FLAG peptide (an epitope tag) or the Strep II tag (which specifically binds and has a high affinity with streptavidin). These vectors have been designated pMM104F and pMM104S, respectively. These vectors (3 µg each) were digested separately until their completion with Nco I and Hpa I and the 2870bp axis fragments were gel purified and ligated with the GST fragments generated from oligomezl A as described in the Example 1. Using a combination of anti-FLAG and anti-GST antibodies, which was followed by sequencing, the clones were identified to contain perfect fusions within the framework of the full-length GST gene labeled FLAG. Similarly, using a combination of anti-GST antibodies and a streptavidin-horseradish peroxidase conjugate that was subsequently sequenced, clones containing perfect fusions were identified within the framework of the full-length GST gene labeled Strep II. In both examples, the frequency with which full length fusions were found within the framework was the same (within the experimental error) as determined in Example I.

These clones were used in immobilization experiments, essentially as described in Example 1, except that the immobilization substrates were magnetic drops in 96-well plates covered by anti-FLAG antibody and streptavidin, respectively. As in Example 1, we were able to demonstrate the specific immobilization of fusion proteins by means of these tags, and we were also able to demonstrate that GST fusion retains its activity when immobilized by means of the FLAG or Strep tag . Example 3

(a) Modification of a second protein using the hexahistidine tag

Following the procedure described in Example 1 for glutathione-S-transferase, we have shown that this procedure is independent of the precise gene being manipulated.

Therefore, starting with a plasmid encoding the human transcription factor NF-кB p50 and continuing with exactly the same procedure described in Example 1 (using oligomix A) unless otherwise specified, we were able to demonstrate the modification of NF-κB p50 so that the first stop codon within the frame has been specifically suppressed and replaced by a structural fusion with DNA encoding a polyasparagine, hexahistidine tag. Western blots with colonies that used an antibody with anti-His tag allowed the identification of clones expressing hexahistidine-tagged protein. The DNA of these colonies was amplified, purified and sequenced. Sequential data confirmed that several clones encoded perfect fusions within the framework with full-length NF-κB, that is, clones in which the stop codon has been specifically suppressed and replaced by a hexahistidine tag within the structural framework. The frequency with which full-length fusions were found within the frame in the case of NF-кB p50 was 1.1%, which is close to, and within the experimental error, what was determined in Example 1 for GST.

(b) Immobilization and functional analysis of NF-кB p50 labeled with hexahistidine

E. coli DH5α cells were transformed with one of the full-length NF-κB plasmids and hexahistidine labeled, created from the above-mentioned methodology. A single carbenicillin-resistant colony grew in a logarithmic or exponential phase in 10ml of liquid culture and was then supplemented with 100µM IPTG to cause expression of NFкB p50 with hexahistidine tag. After growth for an additional 4 hours, the cells were harvested and lysed by sonication. SDS-PAGE of crude lysate showed an overexpressed protein of the expected size (38kDA), which represented approximately 5% of the total soluble protein, as well as a small amount of the 65 kDa fusion NF-кB p50-hexahistidine-GFP, generated at Through amber suppression.

image3

Link point NF- кB P50

A duplex oligonucleotide, "κB motif", which contains a palindromic binding point for NF-κB50, was labeled in bases 3 ’with digoxigenin using terminal 3 transferase under standard conditions (Boehringer Mannheim).

Protein lysates were prepared using the freeze-thaw / lysozyme method in PBS (phosphate buffer saline pH 7.5) containing 5 mM β-mercaptoethanol. 200 µl of the soluble protein lysate of each clone was applied to the Ni-NTA-coated micro well and incubated at room temperature for 45 minutes. At the end of the incubation period, the wells were washed three times with PBST (PBS containing 0.02% Triton X-100) to remove all unbound proteins. The wells were washed three times with a buffer that binds to DNA (10 mM Tris.HCl pH 7.4, 75 mM KCl containing 5 mM β-mercaptoethanol with a soak time of 1 minute. The кB motif labeled 3 'was added Digoxigenin (2 pmol) to the wells in 200 µl of the buffer that binds to DNA containing 1 µg of non-specific poly (dI-dC) DNA After another 30 minutes of incubation, unbound DNA was removed by washing the wells three times with 10 mM Tris.HCl pH 7.4, 25 mM KCl containing 0.02% Triton X-100. An alkaline phosphatase conjugate of anti-digoxigenin antibody was diluted with 150mU / ml in "antibody dilution buffer" (10mM Tris .HCl pH 7.4, 25mM potassium chloride) supplemented with 0.2% bovine serum albumin The diluted antibody (200 µl) was then applied to the microwells After 30 minutes at room temperature, the unbound antibody was removed by washing the micro wells with "antibody dilution buffer" (3x350 µl) comple Mentioned with 0.02% Triton X-100. Next, 200µl of a buffer (100mM Tris.HCl pH9.5, 100mM NaCl, 50mM MgCl2) containing 250µM p-nitrophenyl phosphate (pNPP), an alkaline phosphate substrate was added they were added to the wells and the reaction was allowed to stand overnight at room temperature, and then the yellow coloration in each well (corresponding to the formation of a product, p-nitrophenol) was quantified at 405 nm. The background rate of hydrolysis of the pNPP substrate was low so that a positive test result was clear from the appearance of yellow in the wells. As controls in this test we omit the crude lysate, or the labeled oligonucleotide, or the antibody, or add an excess multiplied by 20 of unlabeled duplex oligo or replace the NF-кB p50 labeled with hexahistidine containing crude lysates with equivalent amounts of a crude cell lysate of DH5α cells expressing GST with hexahistidine tag on the same vector background.

In this assay, NF-κB p50 first binds to the labeled oligonucleotide through the specific binding point. The protein-DNA complex is then immobilized in the micro wells by means of the hexahistidine tag and the rest of the proteins (including complexes between the labeled oligo and other DNA binding proteins present in the crude lysate) together with other unbound oligos. and tagged wash. Because the antibody conjugate recognizes the label in oligo, not the hexahistidine-labeled protein, a positive signal can only be observed in the assay if the NF-κB p50-DNA interaction is maintained during the immobilization of NF-κB50 by middle of the label; If this interaction is not maintained, the oligo will be lost during the washing steps and no color change will be observed.

We found that the yellow product was only detected in the microwells that had contained the crude lysate of NF-κB p50 with the hexahistidine label and the oligonucleotide labeled with digoxigenin and to which the alkaline phosphatase conjugate of anti-digoxigenin antibody had been added. This showed that the color change observed was specifically due to the immobilized complex NF-κB p50-oligonucleotide and also that NF-κB50 maintained activity during specific immobilization.

Example 4

An alternative method for producing full-length genes or cDNAs with the stop codon accurately removed involves the production of a specific chain set of 5 ′ embedded deletions in a PCR product that encompasses the coding sequence. Potentially, this could be accomplished using any 5'-a 3'-exonuclease such as λ exonuclease, E. coli DNA polymerase I, Taq DNA polymerase, or T7 gene exonucleus 6. Once the nested set of removals embedded in 5’s has been created, an oligix can be aligned with the exposed regions of the single chain; this mixture of oligos consists of a set of oligos in which the complement of each stop codon is represented at the 3 'end, immediately preceded by 6 arbitrary residues and, before that, by a defined sequence encoding a restriction point Type IIS and a complementary recognition sequence for the "LMB2" primer (see Example 1). Therefore, the sequence of the set of oligos is as follows:

image2

An oligo of the mixture will specifically align each stop codon exposed in the sense chain and serve as a primer for DNA polymerase that has good activity to displace the chain, such as Taq polymerase, or 5'-o 3 exonuclease activity '- as polymerase I from DNA E. coli. The newly generated DNA duplex fragment can then be specifically amplified using a primer that binds to the 5'-end of the original PCR product and one that specifically binds to the newly aligned and extended oligo. In this procedure, the 3 'end of the aligned oligo is not necessary to directly collide with the residue embedded in 5’-of the duplex template. The oligos of the mixture can be aligned with a complementary region on the exposed DNA template with a single chain in a manner similar to the arbitrary preparation of hexamers, but as an important fact, the extension of the chain will only occur in the places where the 3 'end has aligned with absolute complementarity. Because the 3 'ends of the oligos in the mixture are designed to be complementary to the stop codons, only when the 3'-end of an oligo has specifically joined the corresponding stop codon will the extension of the primer take place and subsequent PCR amplification. The union with the stop codons outside the frame and those located under the true stop codon of the gene of interest will also take place, resulting in PCR amplification of not only the exact full length gene, but also a set of products truncated and extended. However, these products can be easily discarded in successive steps because they will not cause structural fusions with the peptide tag.

Performing the alignment and extension procedure on a nested set of 5 'deletions as described has significant advantages over carrying out the alignment and extension procedure on a DNA or RNA molecule with only one chain that encompasses a complete coding region. This is because the number of points with which the primers can specifically align before extension is greatly restricted by the presence of the double chain portion of the nested set of eliminations. The single chain portion of deletions will primarily cover the non-translated 3'-region of the genes and will have the effect of greatly influencing extension products in favor of stop codons external to the coding sequence and also in favor of longer extension products; These factors will act greatly increasing the frequency at which the first stop codon within the frame is specifically removed by the overall procedure. In fact, we have tried to align and extend the oligos described in step (b) below using a single chain coding region template, and we have been able to identify the full length and correctly modified clones from the experiment. In comparison, the results of the experiment described in detail below clearly demonstrates that the use of the nested set of 5 'deletions as the template for alignment and extension steps is both effective and efficient to facilitate the specific withdrawal of the first stop codon. within the frame of the coding sequence to produce fusions with the full length polypeptide tag and within the frame.

(a) PCR amplification and gene exonuclease digestion before labeling

Therefore, we first perform an initial PCR amplification of a GST gene exactly as described in Example 1, steps (a) and (b), except that the "STinverse" primer in the amplification is now 5 ' -phosphorized. The purified PCR product was then digested with 2.5 units of λ exonuclease (Novagen Strandase kit) for 40 minutes at 37 ° C in a standard reaction buffer (67mM glycine-KOH pH9.4, 2.5mM MgCl2, 50µg / ml serum albumin bovine). The enzyme was then deactivated by heating it at 75 ° C for 15 minutes. Because the λ exonuclease only recognizes 5 'phosphate groups as substrates, the sense chain is protected from digestion and the digestion product is therefore a nested set of deletions of the 5' end of the antisense chain of the PCR product.

(b) Specific binding, extension and amplification to introduce the label

5 µl of the λ exonuclease reaction mixture was diluted in reaction buffer with E. coli DNA polymerase I in the presence of 250 µM dNTPs and a molar excess multiplied approximately 25 by the following set of degenerate oligo:

5’-GTA AAA CGA CGG CCA GTC TGG AGG AGG AGA NNN NNN TCA-3 ’

The digested fragments and oligos were aligned for 30 minutes at 37 ° C and then 5 units of E. coli DNA polymerase I were added and subsequently the reaction was incubated for 3 hours at 37 ° C. The extended products were purified using a QIAquick PCR cleaning kit (Qiagen) under standard conditions and used as a template in a standard PCR reaction using Pwo polymerase with "STdelantero" and "LMB2" primers. 30 PCR cycles (94 ° C, 1 min; 57 ° C, 1 min; 72 ° C, 2 min) were performed to generate PCR products that ranged up to 1000bp.

The PCR products generated from the procedure described above were digested and cloned into the vector pMM106H as described in step (e) of Example 1. The colonies obtained from the cloning process were visualized at 365 nm to identify colonies. fluorescent greens 73 colonies were collected, subjected to plaque replication technique and analyzed by means of Wester blot colonies under standard conditions using anti-His and anti-GST labeled antibodies. DNA from 15 colonies that were positive with anti-His and anti-GST was amplified, purified and sequenced. Sequential data confirmed the presence of 10 perfect structural fusions with full-length GST, that is, clones in which the first stop codon within the framework of the original GST gene has been specifically suppressed and replaced by a polyasparagine within the framework, label of hexahistidine. Therefore, the rate of successful modification we obtained through this global procedure is estimated at 39% of the total number of green fluorescent colonies. Example 5

(a) Identification of a protein from a well of 11 genes

We have applied exactly the same process as in Example 1 except in the well specifications of 11 different genes listed in the table below. We have generated selections of specifically modified resulting motorcycle proteins that each position in the selection corresponds to a single recombinant protein immobilized by means of the tag added as a result of this procedure. Next, we have analyzed the selection through a functional assay and have successfully identified individual protein components of the well.

image2

Initially, the eleven genes were subcloned in the same pTrcHisA vector axis since, among other things, this fact mimics the situation in which the DNA library is located. The "STdelantero" and "STtrasero" primers described in Example 1 were designed to be universal primers for the amplification of genes encoded within the pTrcHisA vector axis.

The "Stdelantero" primer was designed in such a way as to encode a number of restriction points as indicated below:

image2

Therefore, any of the restriction enzymes Aat II or Sfi I can be used to generate 3 'protrusions for exonuclease protection purposes. For the purposes of directional cloning at the end of the modification process, in this Example we choose to use Bsp HI since, although it will statistically cut more frequently inside the library, it generates cohesive ends that are compatible with the cloning point of Nco I in the vector labeled pMM106H used here and does not cut any of the 11 genes in the present well. Clearly, in principle any of the encoded restriction points of the primer could be used as long as the tag vector contains an equivalent cloning site below the promoter; Sfi I would have many advantages in this regard in a larger library format because it has an 8bp recognition sequence so that the frequency of arbitrary incidence of a Sfi I site in a given gene will be much lower (1 in 6.5x104) than for a 6bp recognition sequence such as Bsp HI (1 in 4,096).

The vector of the pMM106H tag is an "ATG" vector, that is, the 5 'cloning point (Nco I) partly coincides with the ATG start codon positioned under a ribosome binding point (RBS) for the expression of native proteins However, in the procedure described here we do not depend on cloned genes that have a common restriction point in their start codons. Instead, we simply depend on the promoter encoded by the vector that initiates transcription to produce mRNA, with the signals necessary for translational initiation provided by the cloned genes themselves. Therefore, in this Example all genes in the original well have a start codon immediately preceded by an RBS, regardless of the presence or absence of an ATG cloning point. Because the "STdelantero" primer is linked under RBS in the initial eleven clones, subsequent post-modification cloning using any of the restriction points encoded by the primer will introduce the newly modified genes into the tag vector along with your original RBS or ATG so that the start of the translation will be assured. In a cDNA library format, the same situation applies in which all full-length cDNAs will have their own 5 ’non-translated regions (UTRs) that contain eukaryotic translational initiation signals. All this is necessary to obtain an adequate translational initiation, and in this case the modified cDNA is then cloned together with its 5’-UTR to a eukaryotic vector which produces signals of translational initiation once again; therefore, a universal set of PCR primers equivalent to those used in this Example can be used.

The modification process was carried out as described in Example 1 with the following modifications. An equimolar well of eleven genes was used as a template for the initial PCR amplification using "STdelantero" and "STtrasero" primers, and subsequently the fragments were digested with Sfi to protect the 5 'end since it has an 8bp recognition sequence and does not cut any of the 11 genes. The oligonucleotide (see Example 1) was used in the alignment step. After the second PCR amplification, the modified fragments were divided into 2 backgrounds that were digested separately with Bpm I and Sap I. Statistically, a small fraction of genes in the libraries will contain either a Sap I point (7bp recognition sequence; probability of arbitrary incidence = 1 in 16,384) or a Bpm I point (6bp recognition sequence; probability of arbitrary incidence = 1 in 4,096) but only a very small fraction will contain both (probability of arbitrary incidence = 1 in 6.7x107). Therefore, the two Type II restriction enzymes were used separately to effectively ensure that the specific modification of all full-length genes was not excluded by the presence of one or another restriction point within the gene.

The digested fragments of the two wells were joined for treatment with mung bean nuclease and digestion with Bsp HI. The resulting fragments were gel purified in 4 different size ranges and ligated separately to the pMM106H vector (digested to completion with Nco I and Hpa I gel purified) in order to avoid a preferential ligation of smaller inserts . The transformed cells were visualized under UV light (365 nm) and the green fluorescent colonies were visually selected for Western blot analysis.

Approximately 30% of the total number of transformed colonies were fluorescent green and, of that, 73% expressed proteins that are recognized by antibodies with anti-His tag. 190 green and His positive colonies were inoculated in 1.5 ml of liquid medium in blocks with 96 deep wells and grew overnight. The cells were collected by centrifugation and lysed by freezing defrosting / lysozyme. Then, the crude lysates were applied to individual wells of a 96-well plate coated with Nickel-NTA and unbound proteins were removed by washing, leaving the His-tagged recombinant proteins immobilized in the wells. Subsequently, immobilized proteins were tested for NF-κB activities using the assay described in Example 2 and wells containing positive “hits” were identified by the appearance of yellowing. Three clones showed positive "κB-motif" DNA binding activity. An additional characterization of the positive clones showed that one encoded full-length and precise fusions within the p50 NF-κB gene with the hexahistidine tag as expected. It was found that the other two clones encode related DNA binding proteins that are known to share the same DNA specificity as p-50 NF-κB, although with lower binding affinities.

Therefore, this result demonstrates that the functional interrogation of the selections generated by this procedure can identify both specific interactions and weaker interactions that are however specific and biologically relevant. As a consequence, we have used this procedure to create functional protein selections in a micro-well format and, using these selections, we have successfully identified individual proteins from a small subset based on a specific protein-ligand interaction.

Claims (1)

1st. - A method to generate a protein selection consisting of: (a) providing a set of protein-encoding DNA molecules that are well labeled at the N or C terminal with a portion of the marker molecule; (b) express the labeled individual proteins in a specially separated format; (c) purify and immobilize each labeled protein in a spatially defined format to produce a selection of proteins on a solid support, and thus the specific interactions between the label and the solid support are used both in purification and in immobilization. 2nd.- The method of claim 1, wherein the marker molecule portion is a peptide sequence selected from the group consisting of: (a) a hexa-histidine tag; (b) a complete protein or protein domain; (c) an antibody epitope; (d) a biotin mimic; Y (e) protein domain that binds with maltose. 3.- The method of the previous claims, wherein the portion The marker molecule is modified post-translationally. 4th.- The method of claim 3, where the posttranslational modification is done includes the addition of a biotin or lipid molecule. 5th. - The method of the preceding claims, wherein the labeled proteins are correctly folded to maintain function. 6th.- The method of the previous claims, wherein the labeled proteins have full length. 7th. - The method of any of the previous claims, wherein the cDNA molecules encoding labeled proteins are produced by a method consisting of the insertion of an additional and known sequence of DNA encoding the portion of the marker molecule either immediately after a stop codon. or either immediately before a stop codon present in said DNA molecules consisting of the following steps: (a) generate a set of nested deletions from said DNA molecules; (b) align and link with the set of nested deletions a first and a second set of oligonucleotides; in which

(i)

the sequence or complementary sequence of one or more of the three possible stop codons is represented in the mixture of the first set of oligonucleotides, and

(ii)

the sequence or complementary sequence of one or more of the three main stop codons is represented in the mixture of the second set of oligonucleotides

(c) specifically amplifying bound products by a primer extension reaction using an oligonucleotide primer chosen to be complementary to the first and second set of oligonucleotides. 8th. - The method of any one of claims 1 to 6, further comprising the steps of contacting one more test compounds with the protein selection and measuring the bond of one or more compounds with the proteins in the selection, screening thus one or more compounds for biological activity. 9th. - The method of any one of claims 1 to 6, further comprising the step of contacting one or more proteins, e.g. ex. A cell surface receptor, with protein selection and measuring the linkage of one or more specific proteins with the selection proteins, thereby screening one or more proteins for specific protein-protein interaction. 10th. - The method of any of claims 1 to 6, further comprising the step of contacting acidic probes with protein selection, and measuring the linkage of the probes with the proteins, thereby screening one or more proteins for interactions specific protein-nucleic acid. 11th.- The method of any one of claims 1 to 6, further comprising the step of contacting the protein selection with an antibody library, so that one or more proteins in the protein selection is linked with at least a 5 antibody from the antibody library, removing those antibodies that are not bound, and immobilizing those antibodies linked to the proteins in the protein selection, thereby producing a selection of antibodies.