FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2043266732> ?p ?o ?g. }

• W2043266732 endingPage "13584" @default.
• W2043266732 startingPage "13581" @default.
• W2043266732 abstract "Rapid drug discovery is a pressing need in the pharmaceutical industry. As new target proteins are proposed for therapeutic intervention, the methods of random screening or rational drug design may not be adequate to the challenge. SELEX1( 1The abbreviation used is: SELEXsystematic evolution of ligands by exponential enrichment.) -derived oligonucleotides, possessing shapes that interact with target molecules, may provide a surprising alternative to other ligand discovery technologies. In addition, the technology leads to a novel research idea that may enrich our understanding of the mechanisms by which cells and organisms integrate their metabolic activities. systematic evolution of ligands by exponential enrichment. We frequently seek chemicals with properties suitable for a given task. In pharmaceutical, diagnostic, and biological research, these chemicals often must bind tightly and specifically to some target molecule, even though the target may be undefined during the search for the appropriate chemical. Search strategies for novel binding compounds historically depended on screening; the natural products found in fermentation broths, plant extracts, and organisms without tumors or infections have been a source of interesting compounds and new drugs. In addition, medicinal chemists have created derivatives of many natural compounds, seeking improved efficacy. Thus large libraries of natural and synthetic chemicals are extant (perhaps comprising nearly 106 compounds in sum), awaiting ever faster, ever more automated “high throughput” screening methodologies. Since screening begins with libraries made for past screenings, usually unrelated to the present search, library histories constrain the breadth of available compounds. This is an integrated set of methodologies that include structural analysis of target molecules, powerful synthetic chemistries, advanced computational tools, and reiterative structural analyses of compound-target complexes. Rational drug design, when aimed at protein targets, depends critically on the depth of x-ray (or NMR) information and on calculations of electrostatics, hydrophobicities, and solvent accessibility for those targets. Since new potential target proteins are being identified rapidly by human and pathogen genomic sequencing, rational drug design faces a requirement for high throughput compound identification similar to the demands on traditional, random screening. The creation and simultaneous screening of large libraries of synthetic molecules appear to meet the requirements for rapid and successful compound identification for many research, diagnostic, and therapeutic purposes. In this review I focus on SELEX, the combinatorial chemistry technology that uses oligonucleotides as the source of compounds (1Tuerk C. Gold L. Science. 1990; 249: 505-510Crossref PubMed Scopus (7888) Google Scholar). Two very recent reviews describe and evaluate a variety of other combinatorial chemistry methodologies. One review aims to elaborate the various ways in which peptide and protein libraries are constructed and screened and does not deal with oligonucleotide libraries (2Gallop M.A. Barrett R.W. Dower W.J. Fodor S.P.A. Gordon E.M. J. Med. Chem. 1994; 37: 1233-1251Crossref PubMed Scopus (1157) Google Scholar). The other review attempts to be complete and surprisingly does not deal with oligonucleotide libraries (3Janda K.D. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10779-10785Crossref PubMed Scopus (133) Google Scholar), probably because the author and most scientists are unaware of the useful qualities of oligonucleotides. This review of SELEX might be read with the other two reviews in hand for comparative purposes. SELEX (Systematic Evolution of Ligands by EXponential enrichment (1Tuerk C. Gold L. Science. 1990; 249: 505-510Crossref PubMed Scopus (7888) Google Scholar)) is a combinatorial chemistry methodology in which vast numbers of oligonucleotides (DNA, RNA, or unnatural compounds) are screened rapidly for specific sequences that have appropriate binding affinities and specificities toward any target. SELEX also has been used to identify new ribozymes and deoxyribozymes (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar, 5Breaker R.R. Joyce G.F. Chem. & Biol. 1994; 1: 223-229Abstract Full Text PDF PubMed Scopus (1101) Google Scholar). The basic strategy is to prepare a library of nucleic acid sequences that can be amplified, to challenge the library for either binding or catalytic properties, to selectively amplify those members of the library that were best at the challenge, and then (if necessary) to rechallenge the amplified subset and continue the process. One cycle of challenge and selective amplification is a SELEX “round”; selective amplification usually is preceded by physical partitioning of the better oligonucleotides prior to amplification but can be accomplished with “partitionless” amplification that is dependent upon a positive response to the challenge (for example, the ligation of an oligonucleotide sequence that is used directly for amplification (6Bartel D.P. Szostak J.W. Science. 1993; 261: 1411-1418Crossref PubMed Scopus (698) Google Scholar)). Virtually all SELEX experiments have been reviewed in more depth than is possible here (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar). Janda (3Janda K.D. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10779-10785Crossref PubMed Scopus (133) Google Scholar) concludes his review by noting the following. “It appears that the next wave of combinatorial research will be directed at the design of libraries that are devoid of the repetitive backbone linkage found within peptides or nucleotides. It is here that structural/stereoelectronic variation and unconstrained motifs will be allowed to expand to unparalleled combinatorial chemical diversity.” And yet no binding challenge thus far has been beyond oligonucleotides (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar), in spite of the repetitive backbone linkage. Single-stranded nucleic acids fold into a bewildering number of stable structures, available in the vast starting libraries of SELEX. SELEX provides “unparalleled combinatorial chemical diversity” now, and modified nucleotides that are compatible with SELEX (7Eaton B.E. Pieken W.A. Annu. Rev. Biochem. 1995; 64: 837-863Crossref PubMed Scopus (91) Google Scholar) add features beyond what may be already “unparalleled.” Nucleic acids can be shapes, not tapes (8Judson H.F. The Eighth Day of Creation. Simon & Schuster, New York1979Google Scholar); biology has utilized nucleic acids as tapes for the creation of genomes and for some (no longer obvious (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar, 9Gold L. Tuerk C. Allen P. Binkley J. Brown D. Green L. MacDougal S. Schneider D. Tasset D. Eddy S.R. Gesteland R. Atkins J. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1993: 497-509Google Scholar)) reason utilized proteins for most recognition (and catalytic) phenomena. Nucleic acids are synthetically randomized by condensing a mixture of activated monomers over some specified number of positions (n), where the sequence diversity for a nucleic acid with four different bases is n4; the availability of novel, additional base pairs could allow even more random sequences over the same length (10Switzer C.Y. Moroney S.E. Benner S.A. Biochemistry. 1993; 32: 10489-10496Crossref PubMed Scopus (243) Google Scholar). Commercially available DNA synthesis machines are approaching millimole scale, that is on a single column in a single run (with 15% yield) about 1020 molecules can be made. For SELEX experiments in the literature the number of molecules in the starting library has been 1014-1015, but this number could be higher if necessary. Large lengths of randomness quickly provide sequence diversity potential that dwarfs even the number 1020, so two libraries made at different times with, say, 100 randomized nucleotide positions, will share essentially no exact sequences longer than 25-35 nucleotides. Thus SELEX will always provide winners that reflect the best oligonucleotides in a specific library, and the same sequences will be found in repeat selections only when the best sequences have modest information content (11Schneider T.D. Stormo G.D. Gold L. Ehrenfeucht A. J. Mol. Biol. 1986; 188: 415-431Crossref PubMed Scopus (704) Google Scholar). In the SELEX experiments reviewed (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar) the winning ligands were most often found to represent between 1 in 109 and 1 in 1013 of the starting library. Random regions of roughly 30 nucleotides were used in many of the first experiments. The known simple single-stranded oligonucleotide motifs can be built from 30 nucleotides (9Gold L. Tuerk C. Allen P. Binkley J. Brown D. Green L. MacDougal S. Schneider D. Tasset D. Eddy S.R. Gesteland R. Atkins J. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1993: 497-509Google Scholar): hairpins, bulges within helices, pseudoknots, and G-quartets. Thus the major structural frameworks for single-stranded oligonucleotides are reached easily with randomization over the lengths that can be searched thoroughly; the experimentalist need not specify a motif in building the initial library. The same targets have been used for a few SELEX experiments in which the libraries had different lengths of randomized sequences; while the same motifs and sequences emerged, it seems likely that large and very precise structural motifs will emerge idiosynchratically if long randomized domains are used. In some instances the experimental goal is to identify a rather short oligonucleotide drug candidate, and so very long random regions have been avoided. The first SELEX experiments were done with RNA (1Tuerk C. Gold L. Science. 1990; 249: 505-510Crossref PubMed Scopus (7888) Google Scholar, 12Ellington A.D. Szostak J.W. Nature. 1990; 346: 818-822Crossref PubMed Scopus (7468) Google Scholar). In the 50 SELEX experiments reviewed in Gold et al.(4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar), 36 used RNA, 5 used DNA, and 9 used libraries containing modified nucleotides (7Eaton B.E. Pieken W.A. Annu. Rev. Biochem. 1995; 64: 837-863Crossref PubMed Scopus (91) Google Scholar). The number of experiments is small, and other variables (including partitioning methods and the length of the randomized region) are numerous. At this moment I can see no obvious pattern with respect to binding affinity or specificity that correlates with the choice of oligonucleotide chemistry. Binding affinity and specificity do seem correlated with the number of SELEX rounds performed, but even for this parameter (13Irvine D. Tuerk C. Gold L. J. Mol. Biol. 1991; 222: 739-761Crossref PubMed Scopus (203) Google Scholar) the correlation is driven by the small number of experiments done in which weakly bound ligands were identified after a small number of SELEX rounds. For many purposes the winning ligands should be nuclease-resistant, and modifications toward that end can be incorporated into the entire library (14Lin Y. Qui Q. Gill S.C. Jayasena S.D. Nucleic Acids Res. 1994; 22: 5229-5234Crossref PubMed Scopus (159) Google Scholar). High affinity, high specificity ligands have been found no matter what monomers have been used to make the library. When a high affinity ligand has been identified using a particular set of monomers (ribonucleotides, deoxyribonucleotides, 2′-aminopyrimidine ribonucleotides, or 5-iodopyrimidine ribonucleotides), full replacement of the monomers in the winners by other nucleotide monomers degrades the affinity and specificity of the ligands (winning RNA ligands are not potent and specific ligands when made as DNA and vice versa). We usually front-load SELEX with the modifications intended for final use, but we also perform post-SELEX modifications (15Green L. Waugh S. Binkley J.P. Hostomska Z. Hostomsky Z. Tuerk C. J. Mol. Biol. 1995; 247: 60-68Crossref PubMed Scopus (48) Google Scholar). Substitutions that enhance the ligand performance can be made at some positions in SELEX-derived ligands. Again, there are too few experiments and not enough data to form rules for acceptable or enhancing post-SELEX modifications. At first glance I have described a technology that yields binding reagents and not bioactive compounds. While this would be interesting in its own right and while SELEX thus contributes to discussions about an early RNA world (9Gold L. Tuerk C. Allen P. Binkley J. Brown D. Green L. MacDougal S. Schneider D. Tasset D. Eddy S.R. Gesteland R. Atkins J. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1993: 497-509Google Scholar), in fact antagonists of a variety of protein activities have been found without difficulty. Intracellular protein targets that function naturally by interacting with nucleic acids have been used to derive ligands; when studied in vitro or in vivo those ligands are antagonists. The delivery issues for ligands aimed at intracellular targets in a therapeutic setting include all of the problems faced by antisense compounds. Extracellular targets, including growth factors and secreted enzymes, have yielded antagonists, although not every selected oligonucleotide is an inhibitor of the target protein (nor would one expect that the dominant SELEX epitope for every protein would overlap the active site). Rapid screening for inhibitory ligands is routinely part of the technology, and that screening (of 100 or so oligonucleotides) has been successful (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar). Oligonucleotide agonists have not been reported thus far (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar), but such compounds are likely to emerge. The targets that have been used for SELEX include nucleic acid-binding proteins, other proteins (including growth factors, proteases, antibodies, and small peptides), and small molecules (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar). The protein targets that do not function in vivo as nucleic acid-binding proteins include large and small proteins, glycosylated and non-glycosylated proteins, enzymes and structural proteins, and complex mixtures of target proteins found as aggregates or on the outside of cells. The small molecule targets include biologically interesting redox-sensitive molecules (16Louhon C.T. Szostak J.W. J. Am. Chem. Soc. 1995; 117: 1246-1257Crossref PubMed Scopus (182) Google Scholar), dyes (12Ellington A.D. Szostak J.W. Nature. 1990; 346: 818-822Crossref PubMed Scopus (7468) Google Scholar), drugs (17Jenison R.D. Gill S.C. Pardi A. Polisky B. Science. 1994; 263: 1425-1429Crossref PubMed Scopus (925) Google Scholar), amino acids (18Famulok M. Szostak J. J. Am. Chem. Soc. 1992; 114: 3990-3991Crossref Scopus (130) Google Scholar, 19Famulok M. J. Am. Chem. Soc. 1994; 116: 1698-1706Crossref Scopus (247) Google Scholar, 20Connell G.J. Illangeskare M. Yarus M. Science. 1994; 264: 1137-1141Crossref PubMed Google Scholar), and nucleotides (21Sassanfar M. Szostak J. Nature. 1993; 364: 550-553Crossref PubMed Scopus (503) Google Scholar, 22Connell G.J. Yarus M. Science. 1994; 264: 1137-1141Crossref PubMed Scopus (107) Google Scholar). Nucleic acid-binding proteins have yielded ligands after SELEX with high specificity and dissociation constants in the low nanomolar range. With respect to specificity, oligonucleotides selected to bind to one retroviral reverse transcriptase do not bind with high affinity to other reverse transcriptases (23Tuerk C. MacDougal S. Gold L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6988-6992Crossref PubMed Scopus (368) Google Scholar, 24Chen H. Gold L. Biochemistry. 1994; 33: 8746-8756Crossref PubMed Scopus (75) Google Scholar). SELEX has been used for proteins whose biological targets are known; the SELEX-derived ligands usually bind more tightly than the natural sequences. Often the selected ligands have significant structural or sequence relations to the natural ligands (25Schneider D. Tuerk C. Gold L. J. Mol. Biol. 1992; 228: 862-869Crossref PubMed Scopus (115) Google Scholar). No extremely high affinity SELEX-derived ligands (low picomolar, for example) have been reported thus far for known nucleic acid-binding proteins. Protein targets that are not thought to bind nucleic acids naturally give ligands after SELEX with high specificity and dissociation constants in the low nanomolar range and, frequently, in the picomolar range. SELEX against members of the fibroblast growth factor family gives ligands with extraordinary specificity for the exact protein used as the target during SELEX (26Jellinek D. Lynott C.K. Rifkin D.B. Janjic N. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11227-11231Crossref PubMed Scopus (146) Google Scholar). Ligands identified for vascular endothelial growth factor bind within the polyanionic recognition site that is functionally aimed at heparin in vivo; again, those ligands do not bind to the other heparin-binding proteins that were tested (27Jellinek D. Green L.S. Bell C. Janjic N. Biochemistry. 1994; 33: 10450-10456Crossref PubMed Scopus (159) Google Scholar). Non-nucleic acid-binding proteins appear to give extremely high affinity ligands more frequently than known nucleic acid-binding proteins. Antagonists are found without difficulty. It was unexpected that ordinary proteins would be suitable targets for SELEX (28Abelson J. Science. 1990; 249: 488-489Crossref PubMed Scopus (30) Google Scholar), and it is quite shocking that those ordinary proteins could be among the better protein targets. The dissociation constants for oligonucleotides aimed at low molecular weight targets are usually in the high micromolar range, with an occasional target yielding a ligand with a Kd of about 100 nM. The specificity of these oligonucleotides can be startling. Oligonucleotides have been found that bind to AMP and not to simple variants (21Sassanfar M. Szostak J. Nature. 1993; 364: 550-553Crossref PubMed Scopus (503) Google Scholar). Oligonucleotides have been found that bind to theophylline and not to caffeine (17Jenison R.D. Gill S.C. Pardi A. Polisky B. Science. 1994; 263: 1425-1429Crossref PubMed Scopus (925) Google Scholar). Oligonucleotides have been found that bind to an amino acid and not its enantiomer (18Famulok M. Szostak J. J. Am. Chem. Soc. 1992; 114: 3990-3991Crossref Scopus (130) Google Scholar). Recently, oligonucleotides have been found that distinguish between NAD and NADH (16Louhon C.T. Szostak J.W. J. Am. Chem. Soc. 1995; 117: 1246-1257Crossref PubMed Scopus (182) Google Scholar). Oligonucleotides identified through SELEX are as potent as antibodies with respect to affinities and specificities (29Griffiths A.D. Williams S.C. Hartley O. Tomlinson I.M. Waterhouse P. Crosby W.L. Kontermann R.E. Jones P.T. Low N.M. Allison T.J. Prospero T.D. Hoogenboom H.R. Nissim A. Cox J.P.L. Harrison J.L. Zaccolo M. Gherardi E. Winter G. EMBO J. 1994; 13: 3245-3260Crossref PubMed Scopus (875) Google Scholar). SELEX bypasses the obvious issues of immune tolerance and thus seems suitable for many experiments in which antibodies are inappropriate. For peptide libraries, constrained or not, and the early versions of small molecule combinatorial libraries, the affinities and specificities of the winners do not approach the affinities and specificities of molecules identified through SELEX (2Gallop M.A. Barrett R.W. Dower W.J. Fodor S.P.A. Gordon E.M. J. Med. Chem. 1994; 37: 1233-1251Crossref PubMed Scopus (1157) Google Scholar, 3Janda K.D. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10779-10785Crossref PubMed Scopus (133) Google Scholar). SELEX may work as well as it does because of the stable structural motifs and restrained “loops” (9Gold L. Tuerk C. Allen P. Binkley J. Brown D. Green L. MacDougal S. Schneider D. Tasset D. Eddy S.R. Gesteland R. Atkins J. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1993: 497-509Google Scholar) that are formed by many random, single-stranded nucleic acid sequences. The perfect binding partner for any target object would complement its molecular shape, ionic character, and hydrophobicity; such molecules could provide astounding affinities and specificities, even for rather small contact surfaces. Biotin-avidin springs to mind as an example of a ligand-target pair that may approach what is possible for the atoms of biology. Perfect binding partners are not found often in biology, probably because finite dissociation rates are demanded in organismic evolution. Proteins, which are macromolecules of distinct and exquisite shapes, have surprising dynamic properties (30Careaga C.L. Sutherland J. Sabeti J. Falke J.J. Biochemistry. 1995; 34: 3048-3055Crossref PubMed Scopus (66) Google Scholar, 31Schoichet B.K. Baase W.A. Kuroki R. Matthews B.W. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 452-456Crossref PubMed Scopus (571) Google Scholar, 32Frauenfelder H. Sligar S.G. Wolynes P.G. Science. 1991; 254: 1598-1603Crossref PubMed Scopus (2596) Google Scholar). We know little of the dynamics of oligonucleotides and nothing about the dynamics of those ordered oligonucleotides identified through SELEX as potent and specific binding partners. Structural studies now under way in many laboratories on these molecules will be informative and will suggest if the entropic gains achieved by stable oligonucleotide ligands rationalize the surprising potency and specificity of these compounds. SELEX-derived oligonucleotides are an alternative to antibodies for research purposes. The common step of preparing antibodies to a new protein for intracellular localization experiments could be complemented or replaced with SELEX-derived reagents; one could modify oligonucleotides with visualization-enhancing adducts and reporters. (Care should be taken to utilize target proteins in the same structural state as those proteins will be during the localization experiments, since SELEX ligands often bind to structural epitopes that are lost upon protein denaturation (26Jellinek D. Lynott C.K. Rifkin D.B. Janjic N. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11227-11231Crossref PubMed Scopus (146) Google Scholar, 27Jellinek D. Green L.S. Bell C. Janjic N. Biochemistry. 1994; 33: 10450-10456Crossref PubMed Scopus (159) Google Scholar).) Similarly, SELEX-derived reagents could be used for affinity purification of proteins. We have been thinking about genomic SELEX as a research tool to identify biological regulatory loops that must exist. Imagine that the oligonucleotide library has been made for SELEX by fragmentation of the DNA from an organism of interest (with further modifications such that the library can be amplified and used to seek RNA or DNA ligands). In an obvious use of genomic SELEX a regulatory protein (the lac repressor, for example) is used to find those sequences from the organism that are the most likely in vivo targets. Genomic SELEX allows the natural targets for known nucleic acid-binding proteins to be uncovered exhaustively, independent of the particular genetic and biochemical experiments that identified a specific target for that regulatory protein. Regulation of gene expression might be even more intricate than we suspect. More interestingly, if every protein can be a SELEX target in vitro, isn't it reasonable to wonder if many (even all) proteins regulate cell activity by binding to some specific nucleic acid in vivo? Any protein, housekeeping or not, may contact a nucleic acid for a meaningful biological purpose. Genomic SELEX experiments are under way and are formally similar to the beautiful work of Fields and his colleagues aimed at uncovering the protein-protein linkage map for an organism (33Bartel P.L. Chien C.-T. Sternglanz R. Fields S. Hartley D.A. Cellular Interactions in Development: A Practical Approach. Oxford University Press, Oxford1993: 153-179Google Scholar). Genomic SELEX offers the opportunity to identify the protein-oligonucleotide linkage map for an organism. We have no idea how common this phenomenon might be, but the accidental discoveries of such regulatory loops are provocative (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar, 34Klausner R.D. Roualt T.A. Hartford J.B. Cell. 1993; 72: 19-28Abstract Full Text PDF PubMed Scopus (1051) Google Scholar, 35He J. Furmanski P. Nature. 1995; 373: 721-724Crossref PubMed Scopus (321) Google Scholar). Oligonucleotides could be viewed as alternatives to antibodies for in vitro and in vivo diagnostic reagents. The strength of these ideas depends on low nonspecific oligonucleotide binding to other proteins (which is already reasonably well established) and to components found in biological specimens or animals. Data are being accumulated that high concentrations of non-target proteins do not degrade affinities or specificities.2( 2K. B. Jensen, B. L. Atkinson, M. C. Willis, T. H. Koch, and L. Gold, submitted for publication.) In vivo imaging will depend on some of the same issues raised by the potential therapeutic uses of oligonucleotides. One major effort since the invention of SELEX has been aimed at direct use of oligonucleotides as drugs. For example, a target for oncology has been angiogenesis, the process by which most tumors recruit new vasculature for the required supply of nutrients (26Jellinek D. Lynott C.K. Rifkin D.B. Janjic N. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11227-11231Crossref PubMed Scopus (146) Google Scholar, 27Jellinek D. Green L.S. Bell C. Janjic N. Biochemistry. 1994; 33: 10450-10456Crossref PubMed Scopus (159) Google Scholar). Tumors secrete a number of angiogenic factors, and the responsive endothelial cells express some key receptors at elevated levels (37Brooks P.C. Montgomery A.M.P. Rosenfeld M. Reisfeld R.A. Hu T. Klier G. Cheresh D.A. Cell. 1994; 79: 1157-1164Abstract Full Text PDF PubMed Scopus (2170) Google Scholar, 38Dvorak H.F. Sioussat T.M. Brown L.F. Berse B. Nagy J.A. Sotrel A. Manseau E.J. Van De Water L. Senger D. J. Exp. Med. 1991; 174: 1275-1278Crossref PubMed Scopus (466) Google Scholar). SELEX has been used to make antagonists that block angiogenesis. SELEX has also been used to identify antagonists of blood clotting, some antibodies, some essential viral proteins, and growth factors (4Gold L. Polisky B. Uhlenbeck O. Yarus M. Annu. Rev. Biochem. 1995; 64: 763-797Crossref PubMed Scopus (742) Google Scholar). In all cases therapeutic uses of these fascinating molecules will depend on in vivo data for various formulations of the drug candidates. Finally, for those situations in which permanent antagonists (analogous to suicide enzyme substrates) are required, the SELEX technology has been modified so as to allow direct selection of oligonucleotides that bind covalently to their targets.2 With these goals in mind I define a fully successful combinatorial chemistry methodology as providing: 1) enough compounds to provide rapidly very high affinity individual molecules for any target, enough compounds to provide molecules aimed at several epitopes on that target, and enough compounds to saturate shape space so thoroughly that the library history is not relevant; 2) high specificity of the emergent ligands toward their cognate targets and weak binding to either related targets or other potentially interactive molecules in the environment of intended use; 3) rapid, cost-appropriate, generic methods to produce and use the ligands in the intended environment; 4) rapid, generic formulation steps that retain ligand activity with high bioavailability and low toxicity; and 5) immediate testing in appropriate animal disease models if the intended uses are therapeutic. A fully successful combinatorial chemistry methodology for therapeutics yields drug candidates rather than compounds that require extensive further experimentation and development, including reiterative medicinal chemistry accompanied by tedious functional and toxicity assays. That is, combinatorial chemistry should not seek lead compounds, “hits” for structural analysis and further development, and compounds that with expensive and slow research may lead to a definitive experiment in an animal. Combinatorial chemistry technologies should provide, if possible, compounds for direct animal testing because in animals, and finally in humans, the clinical outcome is all that matters. SELEX may meet most, if not all, of the criteria for a robust, generic drug discovery technology. Biologists often think they have seen a set of biological regulatory loops that, when perturbed, will lead to a desired therapeutic outcome. Because of an incomplete biology data base, target selection for such perturbation is a guess, sometimes informed yet always biased by accidental observations that appear to have been made just in time for the experimental scientist. Jacob understood evolution as “tinkerer” (36Jacob F. Science. 1977; 196: 1161-1166Crossref PubMed Scopus (1602) Google Scholar), leaving the unavoidable reality of the human organism as a Rube Goldberg machine; those machines are not predictable in detail from even substantial sets of data. The deep value of combinatorial chemistry paradigms is that many therapeutic interventions can be tried inexpensively in animal models for the diseases of interest. For those of us concerned with drug development, only robust combinatorial chemistry methodologies can keep up with the rapid extension of our knowledge afforded by the understanding of genes in pathogens, normal cells, and their altered disease counterparts. I thank my many colleagues at the University of Colorado and NeXstar Pharmaceuticals Inc. for useful and stimulating conversations." @default.
• W2043266732 created "2016-06-24" @default.
• W2043266732 creator A5052497198 @default.
• W2043266732 date "1995-06-01" @default.
• W2043266732 modified "2023-10-15" @default.
• W2043266732 title "Oligonucleotides as Research, Diagnostic, and Therapeutic Agents" @default.
• W2043266732 cites W1585699712 @default.
• W2043266732 cites W1606431180 @default.
• W2043266732 cites W1978202265 @default.
• W2043266732 cites W1980461160 @default.
• W2043266732 cites W1991082568 @default.
• W2043266732 cites W1994947678 @default.
• W2043266732 cites W2001266427 @default.
• W2043266732 cites W2001753204 @default.
• W2043266732 cites W2011267456 @default.
• W2043266732 cites W2022108432 @default.
• W2043266732 cites W2022942432 @default.
• W2043266732 cites W2026237349 @default.
• W2043266732 cites W2026368455 @default.
• W2043266732 cites W2030503112 @default.
• W2043266732 cites W2031415145 @default.
• W2043266732 cites W2033231386 @default.
• W2043266732 cites W2046880909 @default.
• W2043266732 cites W2050749062 @default.
• W2043266732 cites W2056600479 @default.
• W2043266732 cites W2061305204 @default.
• W2043266732 cites W2065127848 @default.
• W2043266732 cites W2066743380 @default.
• W2043266732 cites W2069688986 @default.
• W2043266732 cites W2074703985 @default.
• W2043266732 cites W2079321730 @default.
• W2043266732 cites W2090167866 @default.
• W2043266732 cites W2094795801 @default.
• W2043266732 cites W2094876954 @default.
• W2043266732 cites W2104525153 @default.
• W2043266732 cites W2116265145 @default.
• W2043266732 cites W2120190211 @default.
• W2043266732 cites W2134672977 @default.
• W2043266732 cites W2152049155 @default.
• W2043266732 cites W2171610577 @default.
• W2043266732 doi "https://doi.org/10.1074/jbc.270.23.13581" @default.
• W2043266732 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/7775406" @default.
• W2043266732 hasPublicationYear "1995" @default.
• W2043266732 type Work @default.
• W2043266732 sameAs 2043266732 @default.
• W2043266732 citedByCount "249" @default.
• W2043266732 countsByYear W20432667322012 @default.
• W2043266732 countsByYear W20432667322013 @default.
• W2043266732 countsByYear W20432667322014 @default.
• W2043266732 countsByYear W20432667322015 @default.
• W2043266732 countsByYear W20432667322016 @default.
• W2043266732 countsByYear W20432667322017 @default.
• W2043266732 countsByYear W20432667322018 @default.
• W2043266732 countsByYear W20432667322019 @default.
• W2043266732 countsByYear W20432667322020 @default.
• W2043266732 countsByYear W20432667322021 @default.
• W2043266732 countsByYear W20432667322022 @default.
• W2043266732 countsByYear W20432667322023 @default.
• W2043266732 crossrefType "journal-article" @default.
• W2043266732 hasAuthorship W2043266732A5052497198 @default.
• W2043266732 hasBestOaLocation W20432667321 @default.
• W2043266732 hasConcept C129312508 @default.
• W2043266732 hasConcept C185592680 @default.
• W2043266732 hasConcept C552990157 @default.
• W2043266732 hasConcept C55493867 @default.
• W2043266732 hasConcept C70721500 @default.
• W2043266732 hasConcept C71924100 @default.
• W2043266732 hasConcept C86803240 @default.
• W2043266732 hasConcept C98274493 @default.
• W2043266732 hasConceptScore W2043266732C129312508 @default.
• W2043266732 hasConceptScore W2043266732C185592680 @default.
• W2043266732 hasConceptScore W2043266732C552990157 @default.
• W2043266732 hasConceptScore W2043266732C55493867 @default.
• W2043266732 hasConceptScore W2043266732C70721500 @default.
• W2043266732 hasConceptScore W2043266732C71924100 @default.
• W2043266732 hasConceptScore W2043266732C86803240 @default.
• W2043266732 hasConceptScore W2043266732C98274493 @default.
• W2043266732 hasIssue "23" @default.
• W2043266732 hasLocation W20432667321 @default.
• W2043266732 hasOpenAccess W2043266732 @default.
• W2043266732 hasPrimaryLocation W20432667321 @default.
• W2043266732 hasRelatedWork W1533184902 @default.
• W2043266732 hasRelatedWork W2038780389 @default.
• W2043266732 hasRelatedWork W2076712430 @default.
• W2043266732 hasRelatedWork W2101090140 @default.
• W2043266732 hasRelatedWork W2439422421 @default.
• W2043266732 hasRelatedWork W2748952813 @default.
• W2043266732 hasRelatedWork W2899084033 @default.
• W2043266732 hasRelatedWork W2950980724 @default.
• W2043266732 hasRelatedWork W2411539162 @default.
• W2043266732 hasRelatedWork W2992156405 @default.
• W2043266732 hasVolume "270" @default.
• W2043266732 isParatext "false" @default.
• W2043266732 isRetracted "false" @default.
• W2043266732 magId "2043266732" @default.
• W2043266732 workType "article" @default.