The Successful Partnership of Biotechnology Based Screen Development with High Throughput Screening
Patricia Rose, Jessica Gorman, Steven Kurtz, Pramathesh Patel and Prabhavathi Fernandes*
Bristol-Myers Squibb
Pharmaceutical Research Institute
P.O. Box 4000
Princeton, NJ. 08543-4000
http://www.netsci.org/Science/Screening/feature06.html
Introduction
Screening has been the source of new drugs for the pharmaceutical industry for several decades. Many drugs currently on the market were developed from leads identified through natural product screening. These include antibacterial agents, such as the penicillins and cephalosporins; anticancer agents, such as actinomycin and paclitaxel; immunosuppressants, such as cyclosporin and FK-506; cholesterol lowering agents such as mevinolin and antiparasitic agents such as Avermectin. In the late 1970's and 1980's, however, screening fell out of favor. Using traditional methods, the number of novel selective leads generated did not make this approach cost effective. Screening was not based on rationally designed approaches to target specific molecules of therapeutic interest. For example, bacteria and yeast were used simply to screen for cytocidal or cytostatic antimicrobial agents while mammalian tissue culture cells were used to screen for cytotoxic anticancer agents. Whole animals, organs and tissues were used to look for agents active at certain cardiovascular or central nervous system targets. For the most part, the specific molecule affected by a lead compound or the mechanism of action of a drug was unknown and specificity was lacking. Many companies tended to develop "me-too" programs based on the preparation of analogs to the prototypic compound originally isolated.
The last decade has seen an enormous advance in the understanding of critical cellular processes. This has led to rationally designed approaches in drug discovery. The application of molecular genetics and recombinant DNA technology has led to the isolation of many genes encoding proteins which show promise as targets for new drugs. The availability of such cloned genes for use in targeted screens to identify new small molecule therapeutics has led to revitalization of screening. Targets are now often recombinant proteins, produced from cloned genes which are heterologously expressed in a number of ways. Matched with the revolution in biology has been a revolution in chemistry. During the last five years, combinatorial chemistry has made large numbers of compounds available for testing. These combinatorial libraries complement the enormous numbers of synthetic compounds available from synthetic collections. Advances in biotechnology have also led to the development of many new detection systems facilitating the design and implementation of novel, inexpensive screens. In addition, the development and use of robotics and automation have made it possible to test large numbers of samples in a short period of time. Finally, computerized data analysis and handling systems have become routine in order to efficiently handle the huge amount of data being generated. In short, the combination of all these factors has re-established high throughput screening as the fundamental step in drug discovery in the pharmaceutical industry.
The success of screening depends first upon the selection of the target and, second, the design of screens compatible with the demand for high throughput. This review will focus on the use of biotechnology to design novel high throughput screens using cellular systems.
The Identification of New Targets and Choice of Screening Systems
Molecular cloning techniques have enormously increased the number of targets available for screening and permitted the development of screens focused on specific proteins. The advantages of employing recombinant methodologies are many. Often receptors and enzymes exist in alternative forms, subtypes or isoforms. Using a cloned target focuses the primary screen on the subtype appropriate for the disease. Agonists or antagonists can be identified and their selectivity can then be tested against the other known subtypes.
Once a target protein is identified mammalian cells, insect cells, yeast and bacteria can be used as expression systems for screen development. Mammalian tissue culture cells such as Chinese hamster ovary (CHO) cells, NIH-3T3 and HEK-293 cells are frequently employed. They have the advantage of providing an environment that is similar to the milieu of the natural human cells. To a great degree permeability, post-translational processing, signaling and coupling to other cellular factors in these cells are similar to these processes in most mammalian cells. A major disadvantage of using tissue culture cells for screening lies in their expense with respect to reagents and labor - although one way to reduce some of this cost is to prepare large quantities of membranes that can be aliquoted and frozen for future use.
Baculovirus infected insect cells are often a useful alternative to mammalian cells. In many cases they do not contain a protein homologous to the mammalian target protein and thus are an appropriate "null" cell line. In addition, proteins are frequently expressed at higher levels in insect cells than in mammalian cells and post-translational modifications are usually similar in the two systems. Finally, insect cells are easier to care for and cheaper to grow than mammalian cells. Here, also, large quantities of membranes can be prepared and frozen to increase the efficiency of the screening effort.
Bacteria such as Escherichia coli, Bacillus sp. and Staphylococcus aureus as well as yeast such as Saccharomyces cerevisiae, Schizosaccharomyces pombe and Pichia pastoris provide alternative expression systems for cloned mammalian genes as well as microbial targets. Bacteria are genetically well characterized, have a short generation time, are easy to manipulate and inexpensive to grow. The disadvantages of producing mammalian targets in them are that many post-translational modifications such as glycosylation do not occur, the intracellular signaling pathways are often different, and that they are not as permeable as mammalian cells. In some cases these limitations are not significant or can be circumvented. For example, the permeability problem in bacteria can be overcome with specialized mutations such as the imp mutation (Sampson et al., 1989) or by using gram-positive organisms.
Yeast offer advantages similar to those of bacteria for the expression of cloned genes in that they are also well characterized, easy to manipulate genetically and fast growing. Additionally, they are capable of many post translational modifications and possess some of the signaling pathways seen in mammalian systems. Their disadvantages include low permeability and the possibility that similar genes may function differently in yeast and mammalian cells. As with bacteria, the permeability problem can be reduced by using strains deleted for efflux pumps such as that encoded by the PDR5 gene (Leonard et al., 1994). In the end, the utility of bacteria or yeast varies with the target under study (see below).
The availability of innumerable genes that can be heterologously expressed in tissue culture cells, insect cells, bacteria and/or yeast, and the development of novel detection systems have enabled researchers to create new types of screens that are specific, sensitive, and often automatable. Several of these will be reviewed here.
Screen Development for Ion Channels
The hunt for drugs with highly selective effects on discrete ion channels has evolved over many years of research. An important observation guiding this pursuit is the effect of compounds like tetrodoxin that block Na+, but not K+ currents (Narahashi et al., 1964). This demonstration of selectivity provided evidence that ion movement occurred through distinct ion selective channel molecules. Many ion channel modulators have since been defined with bioassays that measure their effects on action potentials in isolated tissue fibers or slices. In some cases, compounds with therapeutic value as ion channel blockers were initially identified on the basis of clinical manifestations resulting from an unrelated indication.
The combination of recent advances in electrophysiological techniques and in molecular cloning provides the means and opportunity for drug screening programs to focus on distinct ion channel targets. One such approach couples heterologous expression of the channel encoding gene in either Xenopus laevis oocytes or in mammalian cell lines with patch clamp analysis of currents derived from the encoded channel. These systems provide a powerful method for evaluating the specific effects of test compounds on the channel activity. However, a significant limitation of these methods is capacity; the number of compounds that can be rapidly evaluated is low. Alternative methods for high through put screens are needed. Expression of the target channel in a microorganism provides such an opportunity and can be used in concert with electrophysiological methods to expand the screening capacity and optimize evaluation. An example of this is described below.
In the yeast S. cerevisiae, single genes can be readily over expressed, selectively mutated or deleted from their chromosomal locus. In many instances now documented, yeast mutants have been used for the isolation of functional homologues of genes from other organisms. This approach has been used to define components involved in potassium ion flux from the plant Arabidopsis thaliana. Using a strain of yeast defective in potassium uptake, two groups (Anderson et al., 1992; Sentenac et al., 1992) identified distinct cDNAs from A. thaliana encoding K+ channels that when expressed, complemented the potassium uptake defect in the mutant host. This work opened the possibility of expressing channels from other organisms, and the system has since been extended to include vertebrate inwardly rectifying K+ channels (Tang et al., 1995.). In all cases, strains expressing functional uptake homologues are genetically stable and readily adaptable into high capacity screens for modulators of the ion uptake function. Validation of the sensitivity of this system is evident in the inhibitory effect of micromolar concentrations of cesium on the growth of a yeast mutant complemented for potassium uptake with a guinea pig inwardly rectifying K+ channel (Tang et al., 1995). Growth of a control strain expressing the yeast Trk1 potassium transporter is not affected by cesium addition, indicating that cesium inhibition has selectivity for the ion channel, but not the transporter. Given the diversity of the K+ channel superfamily, it is not clear whether all classes of vertebrate K+ channels will function when expressed in yeast; voltage signals and additional structural or regulatory components are obvious limitations. Nevertheless, this approach enhances the capacity of screening methods that rely on electrophysiological recordings following heterologous expression in Xenopus oocytes.
The utility of yeast for expressing other types of ion channels is also evident in the case of M2, the small ion channel from influenza A virus. Expression of high levels of the influenza virus M2 ion channel results in growth impairment in a wild type yeast strain. This growth phenotype served as the basis for a screen to detect inhibitors of M2 function and uncovered a novel inhibitor, BL-1743, whose anti viral activity was confirmed in a whole virus assay (Kurtz et al., 1995).
The growth inhibitory effect of M2 protein on yeast cells appears to be the result of perturbations to the electrochemical proton gradient established by the proton exporting ATPase (H+ATPase), a central component of the yeast plasma membrane. This proton gradient preserves ion balances and facilitates the uptake of other ions and nutrients through secondary transport systems (Serrano, 1991). Alterations to the proton gradient can be measured by recording extracellular acidification rates through microphysiometry (Hafeman et al., 1988; McConnell et al., 1992). These measurements provide a sensitive means to analyze the function of certain ion channels expressed in yeast cells (Hahnenberger et al., 1996a). A similar analysis has been performed with fungal Na+ antiporters (Hahnenberger et al., 1996b), which opens the possibility of expressing genes encoding vertebrate Na+ exchangers or ATPases to complement relevant yeast mutants. The microphysiometer is also adaptable to a larger scale screening process for compounds that modulate ion channel activity; it is capable of distinguishing functional differences between two M2 inhibitors, amantadine and BL-1743 (Hahnenberger et al. 1996b).
Screens for G-Protein Coupled Receptors
More than 100 G-protein coupled receptors (GPCR) have been identified from organisms as phylogenetically distant as mammals and yeast. They are all structurally related in that they contain seven hydrophobic regions that are hypothesized to be transmembrane domains. Their ligands include biogenic amines, peptides, proteins, odorants and gustatory molecules. Many of these receptors, such as the adrenergic, angiotensin, serotonin, endothelin, muscarinic, neuropeptide Y and serotonin receptors, are believed to have clinical relevance and thus have become targets for high throughput screens. Perhaps the most common screen is a competition assay using radiolabeled ligand. Two major disadvantages of this screen are the expense incurred with the use of radioactive compounds and the number of steps (washing) required to run the screen.
A few years ago Scintillation Proximity Assays (Amersham) were introduced as a means of circumventing the wash steps required above. No separation of bound and free ligand was needed. The technology was based on the coating of scintillant beads with an acceptor molecule, for example, the target receptor. When the radioligand bound to the receptor, the emitted radiation was captured by the scintillant which then emitted light. If the radioligand/receptor interaction was blocked, radiation energy was dissipated in the media and the signal decreased. This type of assay had great versatility and could be applied to receptor/ligand, other protein/protein interactions, protein/nucleic acid interactions and enzymatic assays. The beads could be coated with membranes expressing the receptor of interest, with antibodies or with other types of linkers such as streptavidin.
A competitive binding screen simply identifies lead compounds
but does not directly determine whether the lead is an agonist or
antagonist. When an agonist for a receptor is desired, a screen
that will immediately identify the hit as such would be very
useful. Yeast transfected with recombinant GPCRs could be developed
into such a system. In yeast, expression of the receptors is
generally linked to the mating factor response pathway which
operates through its own GPCRs (Ste2 and Ste3). For example, the
adrenergic receptor expressed in S. cerevisiae displayed
its characteristic affinities and specificities (King et
al., 1990). When co-expressed with mammalian G
proteins, partial
activation of the mating factor response was achieved. More
recently the mammalian somatostatin receptor (SSTR2) was
co-expressed in yeast with a chimeric mammalian/yeast Gs (Price et
al., 1995). The pathway was linked to a pheromone responsive
HIS3 reporter gene, enabling the cells to grow on medium lacking
histidine when the receptor was activated.
X. laevis melanophores transfected with recombinant
GPCRs have also been used for expression of mammalian GPCRs. The
aggregation or dispersal of pigment is sensitive to cAMP levels: it
aggregates in the presence of low levels of cAMP and disperses when
cAMP or diacylglycerol levels rise (McClintock et al.,
1992). Thus, stimulation of a Gs or Gq linked receptor would cause a darkening of the melanophores while stimulation of a Gi linked receptor would cause them to become lighter.
Other reporter constructs may also be used to identify agonists
for GPCRs. Those commonly used include luciferase, 
-galactosidase,
chloramphenicol acetyltransferase (CAT), soluble alkaline
phosphatase, and green fluorescent protein. Since different GPCRs
link to different G proteins and signaling pathways, customizing
the system is sometimes necessary.
Tyrosine Kinase Receptors
Numerous peptide growth factors and hormones interact with specific cell surface receptors to initiate specific cellular responses. Many of these receptors, such as the EGF, FGF, PDGF and neurotrophin receptors, belong to the superfamily of receptor tyrosine kinases that contain an extracellular ligand binding domain, a single transmembrane domain and a cytoplasmic tyrosine kinase domain. Interaction with ligand causes receptor dimerization, leading to kinase autophosphorylation and activation. A large number of tyrosine kinases have been identified, based on the high degree of conservation of the kinase domain. As with the G-protein coupled receptors, function of the tyrosine kinase receptors is of immense biological importance and many have become targets for drug screening. Methods based on ligand binding have been employed in most cases.
As a means of screening for agents that affect receptor
activation, microbially based systems have recently been developed.
Yeast provide a eucaryotic system where there are no detectable
endogenous receptor tyrosine kinases. Expression of selected
mammalian signaling molecules in a background devoid of host cell
tyrosine kinase linked signaling systems should avoid the problems
of multiple interactions and cross talk encountered with mammalian
cells. Several laboratories have now demonstrated the feasibility
of expressing mammalian tyrosine kinases in yeast. S.
pombe has been used successfully as a host for the expression
of mammalian platelet derived growth factor (PDGF) (Arkinstall
et al., 1995). The receptor is made as a membrane
associated glycosylated protein that is not autophosphorylated.
However, when co-expressed with mammalian phospholipase C
(PLC), the downstream
target of the PDGF receptor, both proteins undergo tyrosine phosphorylation. This activation of PLC leads to a
dramatic increase in the production of inositol 4,5 biphosphate and
inositol 1,4,5 triphosphate. Thus, the heterologous expression of
selected mammalian signaling components in yeast allows the
functional activation of a receptor target. It should be noted,
however, that in this system the activation was ligand-independent, possibly due to receptor dimerization resulting from the overproduction of the receptor molecule.
Similarly, the fibroblast growth factor receptor was shown to be expressed in S. cerevisiae. The presence of this receptor led to an increase in the tyrosine phosphorylation of several cellular proteins( Kinoshita et al., 1995). When co-expressed with the ligand, FGF2, tyrosine phosphorylation increase dramatically if the FGF2 gene contained a signal sequence. However, it was shown that exogenously added ligand did not result in increased tyrosine phosphorylaton, suggesting that the receptor-ligand interaction might be intracellular.
An additional novel method utilizing the yeast two hybrid system
to study receptor ligand interaction was developed by Ozenberger
and Young (1995). Protein-protein interaction was coupled to a
growth phenotype which could be easily scored. Positive interaction
of the extracellular domain of growth hormone (GH) with the
mammalian GH receptor and the extracellular domain of VEGF with
flk1/KDR was demonstrated. To determine if the binding of ligand
resulted in receptor dimerization, a three component system was
employed. The extracellular domain of the receptor was fused to the
GAL4 DNA binding domain in one plasmid, to the GAL4 activating
domain in a second plasmid and the ligand was produced from a third
plasmid. Using this system, GHR showed ligand dependent dimer
formation, as no reporter activation was detected in the absence of
the plasmid expressing GH. Significant readout in absence of ligand
was seen with VEGF/KDR, suggesting some dimerization occurred in
the absence of ligand. However, this increased when ligand was co-expressed. As with the PDGF system, the ligand receptor interactions in these yeast based systems appear to occur in intracellularcompartments. This limits screening to a search for compounds that affect the protein-protein interactions intracellularly. However,
such systems also show promise for identifying novel ligands using
peptide libraries.
Intracellular Receptors
The lipid soluble hormones such as the steroids, retinoids, thyroid hormones and vitamin D3 are potent cell regulators. Given their central role in growth and development, they provide targets of great therapeutic potential. Unlike the water soluble growth factors, which interact at the cell surface, these hormones pass through the cell membrane to interact with intercellular receptors. The receptors comprise a superfamily of hormone responsive trans-active transcriptional regulators. All are characterized by a central DNA binding domain (DBD) which targets the receptor to a specific sequence known as the hormone response element (HRE) found upstream of genes regulated by the hormone. Upon binding of hormone, the receptor dimerizes and acts to transcriptionally activate or repress expression of nuclear target genes. While the steroid receptors are present in a single form and form homodimers, some of the non-steroid receptors, such as the retinoid receptors, are comprised of subfamilies. Within the subfamily, there can be multiple subtypes, each of which may be present as different isoforms ( for a review, see Mangelsdorf et al., 1995). The majority of these receptors (e.g. PPARs, RARs, TRs and VDR) form heterodimers with RXRs in vivo (Mangelsdorf and Evans, 1995).
Given the complexity and multiplicity of receptor subtypes and interactions, there is an advantage to using a heterologous system to examine individual species. By employing a microbial system such as yeast, which lacks any homologous receptors, one can isolate single species and look for compounds specifically interact with one specific receptor subtype. To look at more complex interactions and heterodimers, two subtypes can be expressed, resulting in the formation of heterodimers as well as homodimers (Heery et al., 1993, 1994; Hall et al., 1993). A second advantage of a heterologous system is to circumvent host cell metabolism of the hormone. For example, in the case of retinoids, 9-cis retinoic acid is metabolized to all trans-RA in mammalian cells, but not in yeast (Allegretto et al., 1993). Thus, the specificity of these two compounds could be clearly established in a yeast system.
It has been amply demonstrated that an HRE can confer hormone responsiveness to heterologous genes in yeast expressing a mammalian hormone receptor (Estrogen receptor, Metzger et al., 1988; Glucocorticoid receptor, Schena and Yammamoto, 1988; Vitamin D, McDonnell et al., 1989; Retinoic acid receptors, Heery et al., 1993). A transcriptional assay using a yeast promoter containing a specific HRE fused to a reporter gene such as lacZ or a biosynthetic gene such as URA3 has been commonly employed. The latter has the advantage that it can also be used to examine transcriptional repression, employing the counter-selective agent 5FOA (Pierrat et al., 1992). Yeast strains expressing various nuclear receptors have been widely employed for high throughput screening to isolate receptor specific agonists and antagonists.
The yeast two hybrid system has been used to isolate mammalian proteins that interact with nuclear receptors in a ligand dependent manner to give transactivation or transcriptional silencing (Chen et al., 1995; vomBaur et al., 1996; Lee et al., 1995). This methodology (see below) could readily be adapted for screening. The use of appropriate hybrid receptor molecules might also prove to be an effective strategy for looking for inhibitors of specific functionalities of a nuclear receptor.
Screens to Study Protein/Protein Interactions
The development of cloning technologies that identify genes on the basis of the interaction of their encoded products with a known protein has permitted important mechanistic insights into many fundamental cellular processes. From a therapeutic perspective, the interaction of two components required for an essential biological process provides many attractive targets for drug inhibition or modulation. Genetic systems developed in yeast (Fields and Song, 1989 ) and bacteria (Menzel and Taylor, 1995) use various reporter genes to detect the interaction of protein domains presented as fusion proteins with transcription components that assemble intracellularly. In the yeast 2-hybrid system, protein A is expressed as a fusion protein with the DNA binding domain of Gal4 or lexA and protein B is expressed as a fusion protein with an activation domain, typically a region of acidic amino acids derived from Gal4 or VP16. An interaction between proteins A and B reconstitutes a transactivation function that is observed by expression of a reporter gene (e.g., LacZ, HIS3, URA3). The expression of the reporter gene is regulated by the placement of Gal4 binding sites or lexA operator sites upstream of the reporter coding region. In the bacterial system, fusions are made to the ToxR protein of Vibrio cholerae which has a periplasmic C-terminal domain responsible for dimerization, a single transmembrane domain and an N-terminal domain which can activate the CTX toxin promoter. Tox fusion proteins are constructed such that a domain of choice replaces the C-terminal domain of ToxR. Dimerization of the fusion produces a transcriptional function that is observed by activation of a reporter gene (LacZ or CAT).
These systems are adaptable to drug screens that seek to
identify inhibitors of protein-protein interactions. An underlying
assumption in this approach is that small molecular weight
compounds can inhibit such interactions. Inhibitors may be found by
screening chemical compound banks and identified by their reversal
of a reporter phenotype such as loss of -galactosidase activity or loss of the expression of a gene required for growth (e.g. URA3, HIS3). The diversity of proteins for which interacting proteins have been detected with these methods
encompasses proteins of most classes and known functions. Based on
these results, it can be inferred that a broad range of targets are
serving as screen targets including regulatory components
(transcription factors) and growth modulators (ras, cyclins, MAP
kinases). Modifications to the initial system such as the inclusion
of an additional plasmid expressing an additional interacting
peptide or protein (Osborne et al., 1995), continue to
expand the types of protein interactions that can be surveyed. It
is also possible to adapt microbial systems to the identification
of small molecular weight compounds that promote protein
interactions as illustrated by the binding of FK506 to its cellular
receptor, FKBP-12, and calcineurin (Foor et al., 1992).
Screens to Find Protease Inhibitors
Many proteases have become targets for drug discovery from viral proteases required for the generation of active viral proteins to mammalian proteases that process prohormones to their active mature forms.
Versatile assays have been developed in bacterial systems to screen for compounds that inhibit protease activity (McCall et al., 1994; Block and Grafstrom, 1990; Baum et al., 1990; Smith and Kohorn, 1991). Most of these involve the co-expression of both the protease and a target reporter gene in the same cell. A minimal protease cleavage site is engineered into the middle of the reporter gene. The inserted cleavage site sequence must maintain the reading frame and be situated at a site which does not interfere with the function of the intact protein. This site must also be situated such that it is accessible to the protease. The assay could be designed such that the protease cleavage either activates or deactivates the target protein resulting, in a selectable (e.g. viability) or scorable (e.g. color) phenotype. Deactivation generally involves cleavage of the active target protein. Activation usually is accomplished by cleavage of an inactive precursor to an active protein.
One reporter gene that has been widely used in a deactivation assay is the tetracycline efflux protein, the tetA gene product, that confers resistance to the antibiotic tetracycline. The tetracycline efflux protein, as the name suggests, is a membrane bound tetracycline/H+ antiporter. It consists of twelve transmembrane domains with five cytoplasmic loops and six periplasmic loops. Based on DNA-sequence analysis, it has been suggested that the twelve transmembrane domains may have evolved as a result of gene duplication of a gene coding for a six-transmembrane domain protein. Thus, it has been speculated that the third cytoplasmic loop acts as a hinge between the two major domains of the efflux protein (Allard and Bertrand, 1993; Levy, 1992; Sheridan and Chopra, 1991). This third cytoplasmic loop is relatively large and the least conserved part of the molecule. Splicing a minimal protease cleavage site of up to 15 amino-acid residues into the third cytoplasmic loop does not affect the function of the efflux protein (McCall et al., 1994). E. coli cells expressing this chimeric tet-efflux protein are tetracycline resistant. If now an active protease is expressed in the same cell, the efflux protein is cleaved, resulting in the loss of tetracycline resistance. However, if the cells are grown in the presence of a protease inhibitor, the tetracycline efflux protein is not cleaved and hence the cells are viable in tetracycline containing medium.
Smith and Kohorn (1991) have developed an analogous approach in S. cerevisiae. The system is based on the inactivation of a target protein upon protease cleavage and can be used in high-throughput screening to identify compounds that inhibit protease activity. They make use of the Gal4 transcriptional activator as the target protein. A minimal protease cleavage site is engineered between the DNA binding domain and the transcriptional activator domain of Gal4. The cleavage of the chimeric Gal4 by a protease renders the protein inactive. Cells that express both the chimeric Gal4 and an active protease are unable to metabolize galactose. Inhibition of protease activity can be monitored by assaying the inability of the cells to grow on medium containing galactose as the sole source of carbon or, conversely, the ability of cells to survive in media containing the suicide substrate 2-deoxygalactose.
An example using activation of a target gene is described by Balint and Pooly (1995). They exploit the fact that streptomycin resistance, conferred upon E. coli by mutants of the rpsL gene (coding for the ribosomal protein S12) is recessive. In this particular case, streptomycin resistant cells, by virtue of a mutated chromosomal copy of the rpsL gene, are transformed with plasmid expressing the wild-type ribosomal protein S12 as a fusion protein and containing the relevant protease cleavage site. The intact fusion protein is inactive and does not affect the streptomycin resistance phenotype of the cells. Cleavage of the fusion protein by the specific protease renders the ribosomal protein active. This results in cells becoming streptomycin sensitive and the protease activity can be monitored easily in streptomycin containing medium.
A number of in vitro biochemical assays have also been developed. In most of these cases, a peptide containing the protease cleavage site is labeled at one end using either a radioactive or a fluorescent tag. The other end of the peptide molecule is tethered to a plate or a bead. In the presence of an active protease, the peptide is cleaved and the labeled end is released. The loss of signal from the labeled end of the peptide molecule after washing reflects the activity of the protease and can be easily monitored.
The scintillation proximity assay (SPA from Amersham) described above allows a homogeneous assay for protease inhibitors, avoiding the extensive washing steps. The relevant substrate can be radioactively labeled and attached to the scintillant bead. In the presence of active protease the substrate would be cleaved, releasing the label, and the signal would decrease. Maintenance of the signal would indicate the presence of a protease inhibitor.
Another variation on this theme avoids the use of radioactivity. This is especially useful in high-throughput assays. The modification involves the use of lanthanide chelates in time-resolved fluorometry (Packard). This particular technology takes advantage three different properties of lanthanide chelates. First, the lanthanide chelate decay time exceeds 500 µs as compared to 10 ns for conventional fluorophores. Thus, non-specific background fluorescence can be eliminated by measuring fluorescence after a delay time of 400 ns. Second, the sensitivity is further increased because the Stoke's shift, the difference between the excitation peak and the emission peak, is exceedingly large. In most cases the difference is greater than 300 nm and the emission peak is very sharp. Finally, when the lanthanide chelate europium-cryptate is used in combination with the energy absorbing molecule, allophycocyanin (APC), energy can be transduced over considerable distances (about 10 nm). This allows the development of homogenous assays. For example, a peptide molecule containing the protease cleavage site is labeled at one end with the europium-cryptate and other end is tagged with APC. In the intact peptide, the europium-cryptate absorbs the excitation light. The energy is transmitted to the APC which is in close proximity. The emission peak from APC is then measured. In the presence of an active protease, the peptide is cleaved. This cleavage separates the APC molecule from the europium-cryptate resulting in a loss of fluorescence. Thus the activity of the protease can be measured by monitoring the fluorescence intensity without the need of separating the excitation energy absorbing entity away from the fluorescence emitting moiety.
Future Developments in High Through-Put Screening
It is expected that as many as 50,000 genes will be identified by the Human Genome Project. It is likely that many of the genes will prove to be therapeutically relevant. While the function of many of the genes will not be obvious, many will be pursued based on their homology to proteins from other species. This will greatly expanded pool of new targets for drug discovery.
In addition, an exponential increase in the number of compounds available for screening has occurred in recent years. As mentioned, synthetic libraries from drug companies and natural products traditionally have been the sources of these compounds. Natural products were usually isolated from soil microbes cultured under a wide variety of conditions. The spectrum of organisms now used for isolation of natural products has expanded from actinomycetes and fungi to include plant, marine organisms and insects. Most significantly, the chemistry of creating combinatorial libraries, described in an earlier issue of this journal, is maturing. This is leading towards a vastly increased number of synthetic compounds for testing.
The rapid development of automation has also had a great impact on the methodologies employed in screening. The newest innovation, nano-technology (Stylli, 1995), will further miniaturize the methodology. By using homogeneous assays as well as extremely sensitive reporter systems, it will be possible to screen tens of thousands of compounds in a cost effective manner.
In summary, an ever expanding number of therapeutic targets, coupled with the evolution of natural product and combinatorial chemistry is driving high through-put screening towards ever greater efficiency. All these advances have brought screening to the point that it is setting the pace for drug discovery in the twenty first century.
References
Allard, J. D. and K. P. Bertrand. (1993) Sequence of a class E Tetracycline resistance gene from Escherichia coli and comparison of related Tetracycline efflux proteins. J. Bacteriol., 175: 4554-4560.
Allegretto, E.A., M.R. McClurg, S.B. Lazarchik, D.L. Clemm, S.A. Kerner, M.G. Elgort, M.F. Boehm, S.K. White, J.W. Pike and R.A. Heyman. (1993) Transactivation properties of retinoic acid and retinoid X receptors in mammalian cells and yeast. J. Biol. Chem., 268: 26625-26633.
Anderson, J. A., S.S. Huprikar, L.V. Kochian, W.J. Lucas and R.F. Gaber. (1992) Functional expression of a probable Arabidopsis thaliana potassium channel in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U.S.A., 89: 3736-3740.
Arkinstall, S., M. Payton and K. Maundrell. (1995) Activation of
phospholipase C in Schizosaccharomyces pombe by co-expression
of receptor or nonreceptor tyrosine kinases. Mol. Cell.
Biol., 15: 1431-1438.
Balint, R. F. and I. Plooy. (1995) Protease-dependent Streptomycin sensitivity in E. coli -- a system for protease inhibitor selection. Bio/Technology, 13: 506-510.
Baum, E. Z., G. A. Bebernitz and Y. Gluzman. (1990) -galactosidase
containing a human immunodeficiency virus protease cleavage site is
cleaved and inactivated by human immunodeficiency virus protease.
Proc. Natl. Acad. Sci. USA, 87: 10023-10027.
Baum, E. Z, G. A. Bebernitz and Y. Gluzman. (1990) Isolation of mutants of immunodeficiency virus protease based on the toxicity of the enzyme in Escherichia coli. Proc. Natl. Acad. Sci. USA, 87: 5573-5577.
Block, T. M. and R. H. Grafstrom. (1990) Novel bacteriological assay for detection of potential antiviral agents. Antimicrob. Agents Chemother., 34: 2337-2341.
Chen, D.J. and R.M. Evans. (1995) A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature, 377: 454-457.
Fields, S. and O. Song. (1989) A novel genetic system to detect protein-protein interactions. Nature, 340: 245-246.
Foor, F., S.A. Parent, N. Morin, A.M. Dahl, N. Ramadan, G. Chrebet, K. Bostian and J.B. Nielsen. (1992) Calcineurin mediates inhibition by FK506 and cyclosporin of recovery from a-factor arrest in yeast. Nature, 360: 682-684.
Hafeman, D.G., J.W. Parce and H.M. McConnell. (1988) Light-addressable potentiometric sensor based biochemical systems. Science, 240: 1182-1185.
Hahnenberger, K.M., K. Esposito, W. Tang, M. Krystal and S. Kurtz. (1996a) Use of microphysiometry for analysis of heterologous ion channels expressed in yeast: Correlation of changes in transmembrane proton flux with modulation of ion channel activity. Nature Biotechnology, in press.
Hahnenberger, K.M., Z. Jia and P. Young. (1996b) Functional expression of the Schizosaccharomyces pombe Na+/H+ antiporter gene, sod2 in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci., U.S.A., in press.
Hall, B.L., Z.Smit-McBride and M.L. Privalsky. (1993) Reconstitution of retinoid X receptor function and combinatorial regulation of other nuclear hormone receptors in the yeast Saccharomyces cerevisiae. Proc Natl. Acad. Sci USA, 90: 6929-6933.
Heery, D.M., T. Zacharewski, B.Pierrat, H. Gronemeyer, P. Chambon and R. Losson (1993) Efficient transactivation by retinoic acid receptors in yeast requires retinoid X receptors. Proc Natl. Acad. Sci USA, 90: 4281-4285.
Heery, D.M. B. Pierrat, H. Gronemeyer, P. Chambon and R.Lossin. (1994) Homo-and heterodimers of the retinoid X receptor RXR activate transcription in yeast. Nucl. Acids Res., 22: 726-731.
King, K., H.G. Dohlman, J. Thorner, M.G. Caron and R.J.
Lefkowitz. (1990) Control of yeast mating signal transduction by a
mammalian 2-adrenergic receptor and Gs subunit.
Science, 250: 121-123.
Kinoshita, N., J. Minshall and M.W. Kirschner. (1995) The identification of two novel ligands of the FGF receptor by a yeast screening method and their activity in Xenopus development. Cell 83: 621-630.
Kurtz, S., G. Luo, K.M. Hahnenberger, C. Brooks, O. Gecha, K. Ingalls, K. Numata and M. Krystal. (1995) Growth impairment resulting from expression of influenza virus M2 protein in yeast: identification of an inhibitor of M2. AntiMicrob. Agents and Chemotherapy, 39: 2204-2209.
Lee, J.W., F. Ryan, J.C. Swaffield, S.A. Johnston and D.D. Moore. (1995) Interaction of thyroid hormone receptor with a conserved transcriptional mediator. Nature 374: 91-94.
Leonard, P.J., P.K. Rathod and J. Golin. (1994) Loss of function mutation in the yeast multiple drug resistance gene PDR5 causes a reduction in chloramphenicol efflux. Antimicrobial Agents and Chemotherapy, 38: 2492-2494.
Levy, S. B. (1992) Active efflux mechanism for antimicrobial resistance. Antimicrob. Agents and Chemother., 36: 695-703.
Mangelsdorf, D.J., C. Thummel, M.Beato, P. Herrlich, G. Schutz, K.Umesono, B. Blumberg, P. Kastner, M. Mark, P. Chambon and R.M.Evans. (1995) The nuclear receptor superfamily: The second decade. Cell, 83: 835-839.
Mangelsdorf, D.J. and R.M. Evans. (1995b) The RXR heterodimers and orphan receptors. Cell 83: 841-850.
McCall, J. O., S. Kadam and L. Katz. (1994) A high capacity microbial screen for inhibitors of human Rhinovirus protease 3c. Bio/Technology, 12: 1016.
McClintock, T.S., G.F. Graminski and M.R. Lerner. (1992) A
method for evaluating the affects of ligands upon Gs protein coupled
receptors using a recombinant melanophore-based bioassay.
Analytical Biochem., 206: 315-322.
McConnell, H.M., J.C. Owicki, J.W. Parce, D.L. Miller, G.T. Baxter, H.G. Wada and S. Pitchford. (1992) The sytosensor microphysiometer: biological application of silicon technology. Science 257: 1906-1912.
McDonnell, D.P., J.W. Pike, D.J.Drutz, T.R. Butt and B.W. O'Mallet. (1989) Reconstitution of the vitamin responsive osteocalcin transcription unit in Saccharomyces cerevisiae. Mol. Cell. Biol., 9: 3517-3523.
Menzel, R. and S. Taylor. (1995) US Patent 08/121201.
Metzger,D.J, H. White and P. Chambon. (1988) The human estrogen receptor functions in yeast. Nature, 344: 31-36.
Narahashi, T.J., J.W. Moore and W.R. Scott. (1964) Tetrotoxin blockage of sodium conductance increase in lobster giant axons. J. Gen. Physiol., 47: 965-974.
Osborne, M. A., S. Dalton, and J.P. Kochan. (1995) The Yeast Tribrid System-genetic detection of trans-phosphorylated ITAM-SH2-Interactions. Bio/technology, 13: 1474-1478.
Osenberger, B.A. and K.H. Young. (1995) Functional interaction of ligands and receptors of the hematopoietic superfamily in yeast. Mol. Endocrinol., 9: 1321-1329.
Pierrat, B. D.M. Heer, Y. Lemoine and R.Losson. (1992) Functional analysis of the human estrogen receptor using a phenotypic transactivation assay in yeast. Gene, 119: 237-245.
Price, L., K. Kajkowski, J. Hadcock, B. Ozenberger and M. Pausch. (1995) Yeast cell growth in response to agonist-dependent activation of a mammalian somatostatin receptor. FASEB J. 9: A1450.
Sampson, B. A., R. Misra and S. A. Benson. (1989) Identification and characterization of a new gene of Escherichia coli K-12 involved in outer membrane permeability. Genetics 122: 491-501
Schena, M. and K.R. Yamamoto. (1988) Mammalian glucocorticoid receptor derivatives enhance transcription in yeast. Science, 241: 965-967.
Sentenac, H., N. Bonneaud, M. Minet, F. Lacroute, J.M. Salmon, F. Gaymard and C. Grignon. (1992) Cloning and expression in yeast of a plant potassium ion transport system. Science, 256: 663-666.
Serrano, R. (1991) Transport across yeast vacuolar and plasma membranes. In The Molecular Biology of the Yeast Saccharomyces; Broach, J. R. et al., eds. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. pp 523-585.
Sheridan R. P. and I. Chopra. (1991) Origin of Tetracycline efflux proteins: conclusions from nucleotide sequence analysis. Mol. Microbiol., 5: 895-900.
Smith, T. A. and B. D. Kohorn. (1991) Direct selection for sequences encoding proteases of known specificity. Proc. Natl. Acad. Sci. USA. 88: 5159-5162.
Stylli, (1995) Soc. for Biomolecular Screening, 1st Annual Conference, Philadelphia.
Tang, W., A. Ruknudin, W.-P. Yang, S.-Y. Shaw, A. Knickerbocker and S. Kurtz. (1995) Functional expression of a vertebrate inwardly rectifying K+ channel in yeast. Mol. Biol. Cell 6: 1231-1240.
vomBaur, E., C. Zechel, D. Heery, M.J.S.Heine, J.M. Garnier, V. Vivat, B. Le Douarin, H. Gronemeyer, P. Chambon and R. Losson. (1996) Differential ligand-dependent interactions between the AF-2 activating domain of nuclear receptors and the putative transcriptional intermediary factors mSUG1 and TIF1. EMBO J. 15: 110-124.
NetSci, ISSN 1092-7360, is published by Network Science Corporation. Except where expressly stated, content at this site is copyright (© 1995 - 2010) by Network Science Corporation and is for your personal use only. No redistribution is allowed without written permission from Network Science Corporation. This web site is managed by:
- Network Science Corporation
- 4411 Connecticut Avenue NW, STE 514
- Washington, DC 20008
- Tel: (828) 817-9811
- E-mail: TheEditors@netsci.org
- Website Hosted by Total Choice