PROTEIN-PROTEIN INTERACTION ASSAYS
A number of different assays are used to determine and evaluate protein-protein interactions. Two in vitro assays are coimmunoprecipitation and polyhistidine-tagged glutathione sulfonyl transferase (GST) pull-down. Two in vivo assays are the yeast and mammalian two-site interaction assays.
Coimmunoprecipitation. The commonly used coimmunoprecipitation assay makes use of antisera to specific proteins. In circumstances in which two different proteins, A and B, associate with each other, an antiserum to protein A will immunoprecipitate not only protein A, but also protein B. Likewise, an antiserum to protein B will coimmunoprecipitate proteins B and A. In practice, the proteins under investigation are radiolabeled by synthesis in the presence of radioactive amino acids, either in cell-free transcriptiontranslation systems in vitro, or in cell culture systems in vivo. Coimmunoprecipitated proteins are detected by gel electrophoresis and autoradiography. Alternatively, the proteins so immunoprecipitated or coimmunoprecipitated can be assayed by Western immunoblot techniques.
Glutathione Sulfonyl Transferase Pull-Down. GST is an enzyme that has a high affinity for its substrate, glutathione. This property of high-affinity interactions has been exploited to develop a cloning vector plasmid encoding GST and containing a polylinker site that allows for the insertion of coding sequences for any protein of interest. Thus, if protein A is believed to interact with protein B, the coding sequence for either protein A or protein B can be inserted into the GST vector. The GST–protein A or B fusion protein is synthesized in large amounts by multiplication and expression of the plasmid vector in bacteria. The GST-fusion protein is then incubated with either labeled or unlabeled proteins in extracts of cells or nuclei. Proteins in the extracts bound to protein A or B in the GST-fusion protein are pulled down from the extracts by capturing the GST on glutathione-agarose beads. Proteins are released from the beads and analyzed by either gel electrophoresis and autoradiography (labeled proteins) or by Western immunoblot (unlabeled proteins). Similar methods using polyhistidine tag in place of GST are also used for pull-down experiments.
Far Western Protein Blots. A variation on the Western blotting technique is the Far Western blot. In this technique, a radio-labeled or fluorescence-labeled known protein (instead of an antibody) is applied to a membrane to which proteins from an electrophoretic gel have been transferred. If the known protein binds to any one or more proteins on the membrane, it is detected as a labeled band by autoradiograph or autofluorography. Relatively strong protein-protein interactions are required for this approach to succeed.
Yeast and Mammalian Cell Two-Site Interaction Traps. The two-site interaction traps are useful for demonstrating protein-protein interactions in vivo. The principle of the approach is that, when a specific protein-protein interaction occurs, it reconstitutes an active transcription factor which then activates the transcription of a reporter gene. The cells (yeast or mammalian) are programmed to constitutively express a strong DNA-binding domain, such as Gal-4, fused to the expression sequence of the selected protein, protein A (the bait). The cells are also programmed to express a transcriptional reporter (e.g., CAT or luciferase linked to a promoter) that has binding sites for Gal-4. Thus, protein A anchors to the DNA-binding site of the reporter promoter via the Gal-4 binding domain but does not activate transcription of the reporter gene, and no reporter function is expressed. Protein B, however, is expressed as a fusion protein with a strong transcriptional activator sequence (e.g., the transcriptional transactivation domain of Gal-4 or of VP16). This transcriptional activation domain–protein B fusion protein does not bind DNA, but when, or if, protein B physically interacts with (binds to) protein A, a fully active transcription factor is reconstituted, the promoter reporter gene is transcribed, and the reported function is expressed.
The yeast two-hybrid system can be used to clone proteins that interact with a bait protein such as protein A. In this instance, a cDNA protein expression library is prepared or obtained that has all of the cDNAs of a given tissue fused to a coding sequence for a transactivation domain (e.g., VP16). Further, the reporter consists of a survival factor essential for the growth of the yeast cell. Thus, when a cDNA encodes a protein B (fish) that interacts with the bait protein A, the yeast cell expresses the survival protein and survives, whereas the other yeast cells die.
GENETIC MANIPULATIONS IN ANIMALS IN VIVO
To create transgenic mice, DNA is injected into the male pronucleus of one-cell mammalian embryos (fertilized ova) that are then allowed to develop by insertion into the reproductive tract of pseudopregnant foster mothers (Fig. 2-3A). The transgenic animals that develop from this procedure contain the foreign DNA integrated into one or more of the host chromosomes at an early stage of embryo development. As a consequence, the foreign DNA is generally transmitted to the germline, and, in a number of instances, the foreign genes are expressed. Because the foreign DNA is injected at the one-cell stage, a good chance exists that the DNA will be distributed among all the progeny cells as development proceeds. This situation provides an opportunity to analyze and compare the qualitative and quantitative efficiencies of expression of the genes among various organs. The technique is quite efficient; >50% of postinjection ova produce viable offspring, and, of these, ~10% efficiently carry the foreign genes. In the transgenic animals, the foreign genes can be passed on and expressed at high levels in subsequent generations of progeny.
FIGURE 2-3. Approaches for (A) the integration of a foreign gene into the germline of mice, and (B) disruption or knock-out of a specific gene. A, DNA containing a specific foreign gene is microinjected into the male pronucleus of fertilized ova obtained from the oviduct of a mouse. Ova are then implanted into the uterus of pseudopregnant surrogate mothers. Progeny are analyzed for the presence of foreign genes by hybridization with 32P-labeled DNA probe and DNA prepared from a piece of tail from a mouse, which has been immobilized on a nitrocellulose filter (tail blots). B, To create a knock-out of a gene, pluripotential embryonic stem (Es) cells are used in vitro to introduce an engineered plasmid DNA sequence that will recombine with a homologous gene that is targeted. The recombination excises a portion of the gene in the ES cells, rendering it inactive (no longer expressible). ES cells in which the homologous recombination occurred successfully are selected by a combined positive-negative drug selection. The engineered ES cells are injected into the blastocoele of 3.5-day blastocysts that are then implanted into the uterus of pseudopregnant mice. The offspring are both chimeric and germline for expression of the knock-out gene and must be cross-bred to homozygosity for the genotype of a complete knock-out of the gene that is targeted for disruption.
Transgenic approaches can also be used to prevent the development of the lineage of a particular cell phenotype or to impair the expression of a selected gene. A cell lineage can be ablated by targeting a microinjected DNA containing a subunit of the diphtheria toxin to a particular cell type, using a promoter sequence specifically expressed in that cell type. The diphtheria toxin subunit inhibits protein synthesis when expressed in a cell, thereby killing the cell. The expression of a particular gene can be impaired by similar cell promoter– specific targeting of a DNA expression vector to a cell that produces an antisense mRNA to the mRNA expressed by the gene of interest. The antisense mRNA hybridizes to nuclear transcripts and processed mRNAs; this results in their degradation by double-stranded RNA–specific nucleases, thereby effectively attenuating the functional expression of the gene. The efficacy of the impairment of the mRNA can be enhanced by incorporating a ribozyme hammerhead sequence in the expressed antisense mRNA so as physically to cleave the mRNA to which it hybridizes. Another approach to producing a particular gene loss of function is to direct expression of a dominant negative protein (e.g., a receptor made deficient in intracellular signaling by an appropriate mutation, or a mutant transcription factor deficient in transactivation functions but sufficient for DNA binding). These dominant negative proteins compete for the essential functions of the wild-type proteins, resulting in a net loss of function.
Another approach, termed targeted transgenesis, combines targeted homologous recombination in embryonic stem (ES) cells with gain-of-function transgenic approaches.11,12,17 This method allows for targeted integration of a single-copy transgene to a single desired locus in the genome and thereby avoids problems of random and multiple-copy integrations, which may compromise faithful expression of the transgene in the conventional approach.
GENE ABLATION (KNOCK-OUTS)
A major advance beyond the gain-of-function transgenic mouse technique has been the development of methods for producing loss of function by targeted disruption or replacement of genes. This approach uses the techniques of homologous recombination in cultured pluripotential ES cells, which are then injected into mouse blastocysts and implanted into the uteri of pseudopregnant mice (Fig. 2-3B). The targeting vector contains a core replacement sequence consisting of an expressed-cell lethal-drug resistance gene (selectable marker) (e.g., neomycin [Pgk-neo]) flanked by sequences homologous to the targeted cellular gene, and a second selectable marker gene (e.g., thymidine kinase [pgk-tk]). The ES cells are transfected with the gene-specific targeting vector. Cells that take up vector DNA and in which homologous recombination occurs are selected by their resistance to neomycin (positive selection). To select against random integration, a susceptibility to killing by thymidine kinase (negative selection) is used; only homologous recombination in which the thymidine kinase gene has been lost will confer survival benefit. Because the ES cells are injected into multicellular 3.5-day blastocysts, many of the offspring are mosaics, but some are germline heterozygous for the recombined gene. F1 generation mice are then bred to homozygosity so as to manifest the phenotype of the gene knock-out. Using this approach of targeted gene disruption, literally thousands of knock-out mice have been created. Many of these knock-out mice are models for human genetic disorders (e.g., those of endocrine systems such as pancreatic agenesis [homeodomain protein IDX-1], familial hypocalciuric hypercalcemia [calcium receptor], intrauterine growth retardation [insulinlike growth factor-II receptor], salt-sensitive hypertension [atrial natriuretic peptide], and obesity [a3-adrenergic receptor]).
CONDITIONAL (DEVELOPMENTAL) INTERRUPTION OF GENE EXPRESSION
Although targeted transgenesis using chosen site integration and targeted disruption of genes has proven helpful in analyses of the functions of genes, conditionally to induce expression of transgenes or conditionally to inactivate a specific gene is useful. Early on, randomly integrated vectors for the expression of transgenes used the metallothionein promoter that is readily inducible by the administration of heavy metals to transgenic mice. Now techniques have been developed to conditionally inactivate targeted genes in a defined spatial and temporal pattern. Several approaches to achieve conditional gene inactivation have been developed. Two of these approaches are (a) the Cre recombinase–loxP system (Fig. 2-4)18 and (b) the tetracycline-inducible transactivator vector (tTA) system (Fig. 2-5).19 Occasionally, both of these systems have been used effectively to knock out (Cre-loxP) or to attenuate (reverse tTA) the expression of specific genes. Both the Cre-loxP and reverse tTA systems require the creation of two independent strains of transgenic mice, which are then crossed to produce double transgenic mice.
FIGURE 2-4. Schema of the Cre-loxP approach to conditionally knock out a specifically targeted gene in mice. A, The approach requires the creation of two separate strains of transgenic mice that are crossed to produce double transgenic mice to effect the conditional gene knock-outs. One mouse strain is created so as to replace the gene of interest by one that has been flanked by loxP recombination sequences (floxed), using targeted recombinational gene replacement in embryonic stem cells as illustrated in Figure 2-3B. The other mouse strain is a transgenic mouse in which the Cre recombinase enzyme expression vector is targeted to the tissue of interest using a tissue-specific promoter, such as the proinsulin gene promoter, to target and restrict expression to pancreatic B cells. B, A more detailed depiction of the strategy for preparation of the gene replacement by homologous recombination to generate mice with a floxed gene. This approach is similar to that described in Figure 2-3B to create knock-out mice.18
FIGURE 2-5. Diagram showing the approach to reversible conditional expression of a gene in mice, using a tetracycline-inducible gene expression system. A, As in the Cre-loxP system (see Fig. 2-4A), the tetracyclineinducible gene system requires the creation of two independent strains of transgenic mice. One strain of mice targets the expression of a specially engineered transcription factor (rtTA) to the tissue of interest, using a tissue-specific promoter (TSP). B, the rtTA transcription factor consists of a modification of the bacterial tetracycline-responsive repressor that has been genetically engineered so as to convert it into a transcriptional transactivator when exposed to tetracycline or one of its analogs. The other mouse strain is one in which a gene of interest is introduced, usually driven by a ubiquitous promoter such as a viral promoter (CMV, RSV) or an actin promoter. The gene of interest could be one encoding an antisense RNA to a messenger RNA of a protein that is to be knocked out. The creation of double transgenic mice then allows for the expression of the gene of interest in a specific tissue under the control of the induced tetracycline. (See text for more detailed description.57) (tet op, tetracycline resistance operon; P, promoter; As, antisense; TPE, tissue promoter element.)
CRE RECOMBINASE–LOXP SYSTEM
The Cre-loxP approach is based on the Cre-loxP recombination system of bacteriophage P1 (see Fig. 2-4). This system is capable of mediating loxP sitespecific recombination in embryonal stem cells and in transgenic mice. Conditional targeting requires the generation of two mouse strains. One transgenic strain expresses the Cre recombinase under control of a promoter that is cell-type specific or developmental stage specific. The other strain is prepared by using ES cells to effect a replacement of the targeted gene with an exact copy that is flanked by loxP sequences required for recombination by the Cre recombinase. The recombined gene is said to be floxed. The presence of the loxP sites does not interfere with the functional expression of the gene and will be normally expressed in all of its usual tissues not coexpressing the Cre recombinase. In those tissues in which the Cre is expressed by virtue of its tissue-specific promoter, the target gene will be deleted by homologous recombination. Thus, the Cre-loxP system acts like a timer in which the events that are to take place are predetermined by the prior reprogramming of the genes: the target gene will be ablated during development where and when the promoter chosen to drive the expression of Cre is activated. Thus, a disadvantage of the Cre-loxP system is the lack of control over when the gene knock-out will take place, because it is preprogrammed in the system. Newer genetically engineered Cre derivatives allow for pharmacologic activation of the recombinant event. A potential advantage of the Cre-loxP system is that one can theoretically generate extensive collections of mice expressing the Cre recombinase specifically and individually in many different tissues so that these mice could be made commercially available to investigators.
CONDITIONAL TETRACYCLINE-INDUCIBLE FORWARD AND REVERSE TRANSACTIVATOR VECTOR SYSTEMS
The Cre-loxP system leads to the irreversible targeted disruption of a particular gene at the time that the promoter encoding the Cre recombinase is activated during development. Having available a system that can be reversibly activated at any time would be desirable. A system that holds promise in this regard is the tetracycline-inducible transactivator vector (forward or reverse tTA), which, in response to tetracycline, switches on a specific gene bearing a promoter containing the tetracycline-responsive operon (see Fig. 2-5). This system allows any recombinant gene marked by the presence of the tet operon to be turned on or off at will simply by the administration of a potent tetracycline analog to the transgenic mice. The vectors were engineered from the sequences of the E. coli bacterial tetracycline resistance operon (tet op), in which a repressor sits on the operon, keeping the resistance gene off. When tetracycline binds to the repressor, it is deactivated, falls off of the operon, and turns on the gene. First, the repressor was converted into an activator by fusing the DNA-binding domain to the potent activator sequence (VP16) of the herpes simplex virus. In this system, tetracycline turned off the activator (tet-off) and thereby caused failure of expression of target genes containing the tet operon binding sites for the repressor turned into an activator. This tTA system required the continued presence of tetracycline to keep the gene off and withdrawal of the tetracycline to turn on the gene, raising problems of long and variable clearance times for the drugs. Turning the gene on by administration of tetracycline (tet-on) would be preferable. Therefore, the tTA vector was reengineered to reverse the action of tetracycline: in the current vector system, the binding of tetracycline to the reverse tTA enhances its binding to the tet operon. Theoretically, as the reverse tTA system now works, any gene can be reversibly turned on by the administration of tetracycline or one if its more potent analogs in the double transgenic mouse, which consists of a cross between a mouse that has the reverse tTA targeted to express in a specific tissue and a mouse that has a ubiquitously expressed transgene for any gene X under the control of the tet operon. The equivalent of gene knock-outs can be accomplished by constructing gene X in a context to express an antisense RNA containing a ribozyme sequence. When induced by tetracycline, antisense-ribozyme RNA binds to the mRNA expressed by gene X, cleaves it, and thereby functionally inactivates the gene.
PROSPECTS FOR THE FUTURE FOR CONDITIONAL TRANSGENE EXPRESSION
The availability of the Cre-loxP and the forward and reverse tTA systems now makes it feasible to combine their key features in the creation of triple transgenic mice so that a targeted recombinational disruption of a gene can be accomplished by the administration of tetracycline. The Cre recombinase could be placed under the control of a tissue-specific promoter containing the tet operon uniquely responsive to the presence of tTA and targeted to a specific tissue by standard pronuclear injection targeted transgenesis. A second transgenic mouse is created with a ubiquitously expressed promoter during the expression of the reverse tTA. In the third mouse, the gene desired to be deleted would be replaced with an appropriately floxed gene. The latter mouse would be prepared by implantation of recombinantly engineered ES cells into blastocysts. The administration of tetracycline to the triple transgenic mouse would induce the Cre recombinase in a tissue-specific manner, thus allowing temporal and spatial control of gene knock-outs.
EXPRESSED SEQUENCE TAGS
A very informative database of expressed sequence tags (ESTs) is being generated and placed in GenBank. Expressed sequence tags are prepared by random, single-pass sequencing of mRNAs from a repertoire of different tissues, mostly embryonic tissues (e.g., brain, eye, placenta, liver). Currently, the EST database contains ~50% of the estimated expressed genes in humans and mammals (70,000–80,000). The EST database will become extremely valuable when the sequences of the human, rat, and mouse genomes are completed.
DNA ARRAYS FOR THE PROFILING OF GENE EXPRESSION
Two variants of DNA-array chip design exist.20,21 The first consists of cDNA (sequences unknown) immobilized to a solid surface such as glass and exposed to a set of labeled probes of known sequences, either separately or in a mixture of the probes. The second is an array of oligonucleotide probes (sequences known, based on either known genes in GenBank or ESTs) that are synthesized either in situ or by conventional synthesis followed by on-chip immobilization (Fig. 2-6). The array is exposed to labeled sample DNA (unknown sequence) and hybridized, and complementary sequences are determined.
FIGURE 2-6. Sample preparation and hybridization for oligonucleotide assay. A complementary DNA (cDNA) is transcribed in vitro to RNA, and then reverse-transcribed to cRNA. This material is fragmented and tagged with a fluorescent tag molecule (F). The fragments are hybridized to an array of oligonucleotides representing portions of DNA sequences of interest. After washing, hybridization of the cRNA probe is detected by localization of the fluorescent signals. (PCR, polymerase chain reaction.)
In cDNA chips, immobilized targets of single-stranded cDNAs prepared from a specific tissue are hybridized to single-stranded DNA fluorescent probes produced from total mRNAs to evaluate the expression levels of target genes.
OLIGONUCLEOTIDE ARRAYS (GENOMIC AND EXPRESSED SEQUENCE TAGS)
The oligonucleotide gene chip (1.28 × 1.28 cm2) consists of a solid-phase template (glass wafer) to which high-density arrays of oligonucleotides (distance between oligonucleotides of 100 Å) are attached, with each probe in a predefined position in the array. Each gene EST is represented by 20 pairs of 25 base oligonucleotides from different parts of the gene (5′ end, middle, and 3′ end).
The specificity of the detection method is controlled by the presence of single-base mismatch probes. Pairs of perfect and single-base mismatch probes corresponding to each target gene are synthesized on adjacent areas on the arrays. This is done to identify and subtract nonspecific background signals. The gene chip is sensitive enough to detect one to five transcripts per cell and is much more sensitive than the Northern blot technique.
COMPLEMENTARY DNA ARRAYS (SPECIFIC TISSUES)
Poly (A) mRNA is isolated from cells or tissue of interest, and synthesis of double-stranded cDNA is accomplished by reverse transcription of cDNA, followed by synthesis of double-stranded cDNA using DNA polymerase I. In vitro transcription of double-stranded cDNA to cRNA is accomplished using biotin-16-UTP and biotin-11-CTP for labeling and a T7 RNA polymerase as enzyme. This cRNA is used for hybridization with the gene chip. The gene chip is stained with R-physoerythrin streptavidin to detect biotin-labeled nucleotides, and different wash cycles are performed. Thereafter the gene chip is scanned digitally and analyzed by special software. (A grid is automatically placed over the scanned image of the probe array chip to demarcate the probe cells.) After grid alignment, the average intensity of each probe cell is calculated by the software, which then analyzes the patterns and generates a report.
The applications of the gene chip include:
Simultaneous analysis of temporal changes in gene expression of all known genes and ESTs.
Sequencing of DNA.
Large-scale detection of mutations and polymorphisms in specific genes (i.e., BRCA1, HIV-1, cystic fibrosis CFTR, b-thalassemia).
Gene mapping by determining the order of overlapping clones.
Expensive equipment for generating and analyzing the data using genechips is required. When the cloning of all genes is completed (Human Genome Project), the gene chip will allow monitoring of the expression of all known genes in various situations.