SELEXION. Systematic evolution of ligands by exponential enrichment with integrated optimization by non-linear analysis

SELEX experiments: new prospects, applications and data analysis in inferring regulatory pathways

Systematic Evolution of Ligands by EXponential enrichment (SELEX) is an experimental procedure that allows extraction, from an initially random pool of oligonucleotides, of the oligomers with a desired binding affinity for a given molecular target. The procedure can be used to infer the strongest binders for a given DNA or RNA binding protein, and the highest affinity binding sequences isolated through SELEX can have numerous research, diagnostic and therapeutic applications. Recently, important new modifications of the SELEX protocol have been proposed. In particular, a modification of the standard SELEX procedure allows generating a dataset from which protein-DNA interaction parameters can be determined with unprecedented accuracy. Another variant of SELEX allows investigating interactions of a protein with nucleic-acid fragments derived from the entire genome of an organism. We review here different SELEX-based methods, with particular emphasis on the experimental design and on the applications aimed at inferring protein-DNA interactions. In addition to the experimental issues, we also review relevant methods of data analysis, as well as theoretical modeling of SELEX.

A mathematical analysis of SELEX

Systematic evolution of ligands by exponential enrichment (SELEX) is a procedure by which a mixture of nucleic acids that vary in sequence can be separated into pure components with the goal of isolating those with specific biochemical activities. The basic idea is to combine the mixture with a specific target molecule and then separate the target-NA complex from the resulting reaction. The target-NA complex is then separated by mechanical means (for example by filtration), the NA is then eluted from the complex, amplified by polymerase chain reaction (PCR) and the process repeated. After several rounds, one should be left with a pool of [NA] that consists mostly of the species in the original pool that best binds to the target. In Irvine et al. [Irvine, D., Tuerk, C., Gold, L., 1991. SELEXION, systematic evolution of nucleic acids by exponential enrichment with integrated optimization by non-linear analysis. J. Mol. Biol. 222, 739-761] a mathematical analysis of this process was given. In this paper we revisit Irvine et al. [Ibid]. By rewriting the equations for the SELEX process, we considerably reduce the labor of computing the round to round distribution of nucleic acid fractions. We also establish necessary and sufficient conditions for the SELEX process to converge to a pool consisting solely of the best binding nucleic acid to a fixed target in a manner that maximizes the percentage of bound target. The assumption is that there is a single nucleic acid binding site on the target that permits occupation by not more than one nucleic acid. We analyze the case for which there is no background loss (no support losses and no free [NA] left on the support). We then examine the case in which such there are such losses. The significance of the analysis is that it suggests an experimental approach for the SELEX process as defined in Irvine et al. [Ibid] to converge to a pool consisting of a single best binding nucleic acid without recourse to any a priori information about the nature of the binding constants or the distribution of the individual nucleic acid fragments.

Subtractive SELEX against two heterogeneous target samples: numerical simulations and analysis

Systematic evolution of ligands by exponential (SELEX) is a revolutionary technology that integrates combinatorial chemistry with high throughput screening to generate from synthesized nucleic acid ligand libraries the high affinity nucleic acid ligands (aptamers) for interesting targets. Recently, the SELEX experiments have advanced from targeting the ligand libraries by a single purified target to multiple heterogeneous target samples. Having the potential of bringing enormous technical and economical advantages to drug discovery, the new application suffers from unpredictable performances. To gain an insight of the new method, we develop a computer model to numerically analyze the subtractive SELEX alternatively against two distinct heterogeneous samples of unknown targets. The model features the discretization of ligand library, the ligand-target binding equilibrium equations, and the separation efficiency of bound and unbound ligands in experiments. By computer simulations, we investigate how aptamers for desired targets embedded in undefined target mixtures are generated under different experimental conditions. We find the iterative screening scheme is fundamentally capable of developing desired aptamers. On the other hand, target sample configuration and separation efficiency may all together significantly diversify the screening dynamics and results.

Structural basis for recognition and sequestration of UUU(OH) 3' temini of nascent RNA polymerase III transcripts by La, a rheumatic disease autoantigen

The nuclear phosphoprotein La was identified as an autoantigen in patients with systemic lupus erythematosus and Sjogren's syndrome. La binds to and protects the UUU(OH) 3' terminii of nascent RNA polymerase III transcripts from exonuclease digestion. We report the 1.85 angstroms crystal structure of the N-terminal domain of human La, consisting of La and RRM1 motifs, bound to r(U1-G2-C3-U4-G5-U6-U7-U8-U9OH). The U7-U8-U9OH 3' end, in a splayed-apart orientation, is sequestered within a basic and aromatic amino acid-lined cleft between the La and RRM1 motifs. The specificity-determining U8 residue bridges both motifs, in part through unprecedented targeting of the beta sheet edge, rather than the anticipated face, of the RRM1 motif. Our structural observations, supported by mutation studies of both La and RNA components, illustrate the principles behind RNA sequestration by a rheumatic disease autoantigen, whereby the UUU(OH) 3' ends of nascent RNA transcripts are protected during downstream processing and maturation events.

Quantitative modeling and data analysis of SELEX experiments

SELEX (systematic evolution of ligands by exponential enrichment) is an experimental procedure that allows the extraction, from an initially random pool of DNA, of those oligomers with high affinity for a given DNA-binding protein. We address what is a suitable experimental and computational procedure to infer parameters of transcription factor-DNA interaction from SELEX experiments. To answer this, we use a biophysical model of transcription factor-DNA interactions to quantitatively model SELEX. We show that a standard procedure is unsuitable for obtaining accurate interaction parameters. However, we theoretically show that a modified experiment in which chemical potential is fixed through different rounds of the experiment allows robust generation of an appropriate dataset. Based on our quantitative model, we propose a novel bioinformatic method of data analysis for such a modified experiment and apply it to extract the interaction parameters for a mammalian transcription factor CTF/NFI. From a practical point of view, our method results in a significantly improved false positive/false negative trade-off, as compared to both the standard information theory based method and a widely used empirically formulated procedure.

Deconvolution of a complex target using DNA aptamers

In vitro selection of single-stranded nucleic acid aptamers from large random sequence libraries is now a straightforward process particularly when screening with a single target molecule. These libraries contain considerable shape diversity as evident by the successful isolation of aptamers that bind with high affinity and specificity to chemically diverse targets. We propose that aptamer libraries contain sufficient shape diversity to allow deconvolution of a complex mixture of targets. Using unfractionated human plasma as our experimental model, we aim to develop methods to obtain aptamers against as many proteins as possible. To begin, it is critical that we understand how aptamer populations change with increasing rounds of in vitro selection when using complex mixtures. Our results show that sequence representation in the selected population changes dramatically with increasing rounds of selection. Certain aptamer families were apparent after only three selection rounds. Two additional cycles saw a decline in the relative abundance of these families and the emergence of yet another family that accounted for more than 60% of sequences in the pool. To overcome this population convergence, an aptamer-based target depletion method was developed, and the library screen was repeated. The previous dominant family effectively disappeared from the selected populations but was replaced by other aptamer families. Insights gained from these initial experiments are now being applied in the creation of second generation plasma protein screens and also to the analysis of other complex biological targets.

Diverse Evolutionary Trajectories Characterize a Community of RNA-Cleaving Deoxyribozymes: A Case Study into the Population Dynamics of In Vitro Selection

Two parallel in vitro selections (denoted Selection A and Selection B) were conducted under different selection-pressure regimes, yielding a diverse community of RNA-cleaving deoxyribozymes. In Selection A, the reaction time was reduced four times (from 5 h to 5 s) over the course of 24 generations, while in Selection B the reaction time was maintained at 5 h for 30 rounds of selective amplification. Sequence alignment was conducted on more than 800 clones assembled from 18 generations that span both selections. Many prominent catalytic sequence classes, including some that extend across both selections, were identified and used to construct fitness landscapes depicting their rise and fall over time. The landscapes from both selections exhibit similar global trends despite differences in population dynamics. Some deoxyribozymes were predominant in the early rounds of selection but gave way to other species that dominated in the middle rounds. Ultimately, these middle classes disappeared from the landscape in favor of new and presumably more fit deoxyribozyme sequence classes. The shape of these landscapes alludes to the presence of many latent deoxyribozymes in the initial library, which can only be accessed by changes in the selection pressure and/or by adaptive mutations. Basic computer simulations provide theoretical corroboration of the experimentally observed pattern of staggered sequence-class transitions across the fitness landscapes. These simulations model the influence of one or more contributing factors, including catalytic rate, folding efficiency, PCR amplification efficiency, and random mutagenesis. This is the first study which thoroughly documents the topography of a deoxyribozyme fitness landscape over many generations of in vitro selection.

Nonequilibrium capillary electrophoresis of equilibrium mixtures: a universal tool for development of aptamers

Aptamers are DNA (or RNA) ligands selected from large libraries of random DNA sequences and capable of binding different classes of targets with high affinity and selectivity. Both the chances for the aptamer to be selected and the quality of the selected aptamer are largely dependent on the method of selection. Here we introduce selection of aptamers by nonequilibrium capillary electrophoresis of equilibrium mixtures (NECEEM). The new method has a number of advantages over conventional approaches. First, NECEEM-based selection has exceptionally high efficiency, which allows aptamer development with fewer rounds of selection. Second, NECEEM can be equally used for selecting aptamers and finding their binding parameters. Finally, due to its comprehensive kinetic capabilities, the new method can potentially facilitate selection of aptamers with predefined K(d), k(off), and k(on) of the aptamer-target interaction. In this proof-of-principle work, we describe the theoretical bases of the method and demonstrate its application to a one-step selection of DNA aptamers with nanomolar affinity for protein farnesyltransferase.

Base-pairing potential identified byin vitro selection predicts the kinked RNA backbone observed in the crystal structure of the alfalfa mosaic virus RNA-coat protein complex

The three-dimensional structure of the 3' terminus of alfalfa mosaic virus RNA in complex with an amino-terminal coat protein peptide revealed an unusual RNA fold with inter-AUGC basepairing stabilized by key arginine residues (Guogas, et al., 2004). To probe viral RNA interactions with the full-length coat protein, we have used in vitro genetic selection to characterize potential folding patterns among RNAs isolated from a complex randomized pool. Nitrocellulose filter retention, electrophoretic mobility bandshift analysis, and hydroxyl radical footprinting techniques were used to define binding affinities and to localize the potential RNA-protein interaction sites. Minimized binding sites were identified that included both the randomized domain and a portion of the constant regions of the selected RNAs. The selected RNAs, identified by their ability to bind full-length coat protein, have the potential to form the same unusual inter-AUGC Watson-Crick base pairs observed in the crystal structure, although the primary sequences diverge from the wild-type RNA. A constant feature of both the wild-type RNA and the selected RNAs is a G ribonucleotide in the third position of an AUGC-like repeat. Competitive binding assays showed that substituting adenosine for the constant guanosine in either the wild-type or selected RNAs impaired coat protein binding. These data suggest that the interactions observed in the RNA-peptide structure are likely recapitulated when the full-length protein binds. Further, the results underscore the power of in vitro genetic selection for probing RNA-protein structure and function.

Determination and augmentation of RNA sequence specificity of the Nova K-homology domains

The Nova onconeural antigens are implicated in the pathogenesis of paraneoplastic opsoclonus-myoclonus-ataxia (POMA). The Nova antigens are neuron-specific RNA-binding proteins harboring three repeats of the K-homology (KH) motif; they have been implicated in the regulation of alternative splicing of a host of genes involved in inhibitory synaptic transmission. Although the third Nova KH domain (KH3) has been extensively characterized using biochemical and crystallographic techniques, the roles of the KH1 and KH2 domains remain unclear. Furthermore, the specificity determinants that distinguish the Nova KH domains from those of the closely related hnRNP E and hnRNP K proteins are undefined. We demonstrate through the use of RNA selection and biochemical analysis that the sequence specificity of the Nova KH1/2 domains is similar to that of Nova KH3. We also show that the mutagenesis of a Nova KH domain to render it similar to the KH domains of the heterogeneous nuclear ribonucleoprotein E (hnRNP E) and hnRNP K allow it to recognize longer RNA sequences. These data yield important insights into KH domain function and suggest a strategy by which to engineer KH domains with novel sequence preferences.

Nucleosomal locations of dominant DNA sequence motifs for histone-DNA interactions and nucleosome positioning

DNA sequence is an important determinant of the positioning, stability, and activity of nucleosomes, yet the molecular basis of these effects remains elusive. A "consensus DNA sequence" for nucleosome positioning has not been reported and, while certain DNA sequence preferences or motifs for nucleosome positioning have been discovered, how they function is not known. Here, we report that an unexpected observation concerning the reassembly of nucleosomes during salt gradient dialysis has allowed a breakthrough in our efforts to identify the nucleosomal locations of the DNA sequence motifs that dominate histone-DNA interactions and nucleosome positioning. We conclude that a previous selection experiment for high-affinity, nucleosome-forming DNA sequences exerted selective pressure chiefly on the central stretch of the nucleosomal DNA. This observation implies that algorithms for aligning the selected DNA sequences should seek to optimize the alignment over much less than the full 147 bp of nucleosomal DNA. A new alignment calculation implemented these ideas and successfully aligned 19 of the 41 sequences in a non-redundant database of selected high-affinity, nucleosome-positioning sequences. The resulting alignment reveals strong conservation of several stretches within a central 71 bp of the nucleosomal DNA. The alignment further reveals an inherent palindromic symmetry in the selected DNAs; it makes testable predictions of nucleosome positioning on the aligned sequences and for the creation of new positioning sequences, both of which are upheld experimentally; and it suggests new signals that may be important in translational nucleosome positioning.

RNA and DNA aptamers in cytomics analysis

BACKGROUND

The systematic evolution of ligands by exponential enrichment (SELEX) technique is a combinatorial library approach in which DNA or RNA molecules (aptamers) are selected by their ability to bind their protein targets with high affinity and specificity, comparable to that of monoclonal antibodies. In contrast to antibodies conventionally selected in animals, aptamers are generated by an in vitro selection process, and can be directed against almost every target, including antigens like toxins or nonimmunogenic targets, against which conventional antibodies cannot be raised.

METHODS

Aptamers are ideal candidates for cytomics, as they can be attached to fluorescent reporters or nanoparticles in order to study biological function by fluorescence microscopy, by flow cytometry, or to quantify the concentration of their target in biological fluids or cells using ELISA, RIA, and Western blot assays.

RESULTS

We demonstrate the in vitro selection of anti-kinin B1 receptor aptamers that could be used to determine B1 receptor expression during inflammation processes. These aptamers specifically recognize their target in a Northern-Western blot assay, and bind to their target protein whenever they are exposed in the membrane.

CONCLUSIONS

Currently, aptamers are linked to fluorescent reporters. We discuss here the present status and future directions concerning the use of the SELEX technique in cytomics.

RNA binding protein sex-lethal (Sxl) and control of Drosophila sex determination and dosage compensation

In the past two decades, scientists have elucidated the molecular mechanisms behind Drosophila sex determination and dosage compensation. These two processes are controlled essentially by two different sets of genes, which have in common a master regulatory gene, Sex-lethal (Sxl). Sxl encodes one of the best-characterized members of the family of RNA binding proteins. The analysis of different mechanisms involved in the regulation of the three identified Sxl target genes (Sex-lethal itself, transformer, and male specific lethal-2) has contributed to a better understanding of translation repression, as well as constitutive and alternative splicing. Studies using the Drosophila system have identified the features of the protein that contribute to its target specificity and regulatory functions. In this article, we review the existing data concerning Sxl protein, its biological functions, and the regulation of its target genes.

Dynamics of DNA in vitro evolution with Mnt-repressor: simulations and analysis

The dynamics of DNA in vitro evolution with Mnt-repressor has been studied numerically and analytically. Based on experimental data and realistic energy landscape for DNA-Mnt-repressor interaction, we investigated the dynamics of DNA in vitro evolution using stochastic simulations. The binding energy of DNA to Mnt-repressor was considered to consist of two parts: the DNA sequence specific and nonspecific. The crossover observed in real experiments is numerically recovered. We demonstrate that the evolution trajectories are drastically dispersed and no typical evolution passage exists during the evolution. Particularly, Fisher's theorem of natural selection is verified. A theoretical analysis for the evolution is also included.

Isolation of a peptide inhibitor of human rhinovirus

Cell culture-based transdominant genetic techniques provide new methods for discovering peptide/RNA modulators of cellular pathways. We applied this technology to isolate a peptide inhibitor of human rhinovirus. A green fluorescent protein (GFP)-scaffolded library of cDNA fragments was expressed in HeLa cells from a retroviral vector and screened for inhibitors of rhinovirus-mediated cell killing. A DNA clone, I421, increased cell survival in an HRV14 challenge assay from less than 0.5% to greater than 60%. It encodes a 53-amino-acid C-terminal extension of the GFP scaffold. Particular subclones of Hela cells expressing I421 (exemplified by I421dp3) show a delay in virus production and a 50-fold decrease in viral RNA levels at 6-8 h postinfection. HRV2, HRV14, and HRV16 show a dramatic decrease in plaque-forming ability on I421dp3 while Coxsackievirus B3 showed a small reduction. Levels of ICAM-1, the receptor for the main rhinovirus serotype, are not altered in I421dp3.

Two ZBP1 KH domains facilitate beta-actin mRNA localization, granule formation, and cytoskeletal attachment

Chicken embryo fibroblasts (CEFs) localize beta-actin mRNA to their lamellae, a process important for the maintenance of cell polarity and motility. The localization of beta-actin mRNA requires a cis localization element (zipcode) and involves zipcode binding protein 1 (ZBP1), a protein that specifically binds to the zipcode. Both localize to the lamellipodia of polarized CEFs. ZBP1 and its homologues contain two NH2-terminal RNA recognition motifs (RRMs) and four COOH-terminal hnRNP K homology (KH) domains. By using ZBP1 truncations fused to GFP in conjunction with in situ hybridization analysis, we have determined that KH domains three and four were responsible for granule formation and cytoskeletal association. When the NH2 terminus was deleted, granules formed by the KH domains alone did not accumulate at the leading edge, suggesting a role for the NH2 terminus in targeting transport granules to their destination. RNA binding studies were used to show that the third and fourth KH domains, not the RRM domains, bind the zipcode of beta-actin mRNA. Overexpression of the four KH domains or certain subsets of these domains delocalized beta-actin mRNA in CEFs and inhibited fibroblast motility, demonstrating the importance of ZBP1 function in both beta-actin mRNA localization and cell motility.

Tenascin-C aptamers are generated using tumor cells and purified protein

Tenascin-C (TN-C) is an extracellular matrix protein that is overexpressed during tissue remodeling processes, including tumor growth. To identify an aptamer for testing as a tumor-selective ligand, SELEX (systematic evolution of ligands by exponential enrichment) procedures were performed using both TN-C and TN-C-expressing U251 glioblastoma cells. The different selection techniques yielded TN-C aptamers that are related in sequence. In addition, a crossover procedure that switched from tumor cell to purified protein selections was effective in isolating two high-affinity TN-C aptamers. When targeting tumor cells in vitro, the observed propensity of naive oligonucleotide pools to evolve TN-C aptamers may be due to the abundance of this protein. In vivo, TN-C abundance may also be well suited for aptamer accumulation in the tumor milieu. A size-minimized and nuclease-stabilized aptamer, TTA1, binds to the fibrinogen-like domain of TN-C with an equilibrium dissociation constant (K(d)) of 5 x 10(-9) m. At 13 kDa, this aptamer is intermediate in size between peptides and single chain antibody fragments, both of which are superior to antibodies for tumor targeting because of their smaller size. TTA1 defines a new class of ligands that are intended for targeted delivery of radioisotopes or chemical agents to diseased tissues.

SERF: in vitro election of random RNA fragments to identify protein binding sites within large RNAs

In vitro selection experiments have various goals depending on the composition of the initial pool and the selection method applied. We developed an in vitro selection variant (SERF, selection of random RNA fragments) that is useful for the identification of short RNA fragments originating from large RNAs that bind specifically to a protein. A pool of randomly fragmented RNA is constructed from a large RNA, which is the natural binding partner for a protein. Such a pool contains all the potential binding sites and is therefore used as starting material for affinity selection with the purified protein to find its natural target. Here we provide a detailed experimental protocol of the method. SERF has been developed for ribosomal systems and is a general approach providing a basis for functional and structural characterization of RNA-protein interactions in large ribonucleoprotein particles.

In vitro selection of allosteric ribozymes: theory and experimental validation

In vitro selection techniques offer powerful and versatile methods to isolate nucleic acid sequences with specific activities from huge libraries. We describe an in vitro selection strategy for the de novo selection of allosteric self-cleaving ribozymes responding to pefloxacin and other quinolone derivatives. Within 16 selection cycles, highly sensitive clones responding to drug levels in the sub-micromolar range were obtained. The morpholine moiety of the quinolone derivatives was required for inhibition of the self-cleavage of the selected ribozymes: modifications of the aromatic system were tolerated better than modifications of the morpholine ring. We also present a theoretical model that analyzes the predicted fraction of ribozymes with a given binding constant and cleavage rate recovered after each selection cycle. This model precisely predicts the actual experimental values obtained with the selection procedure. It can thus be used to determine the optimal conditions for an in vitro selection of an allosteric ribozyme with a desired dissociation constant and cleavage rate for a given application.

Quantitative analysis of receptors for adenosine nucleotides obtained via in vitro selection from a library incorporating a cationic nucleotide analog

5-(3"-Aminopropynyl)-2'-deoxyuridine (dJ), a modified nucleoside with a side chain carrying a cationic functional group, was incorporated into an oligonucleotide library, which was amplified using the Vent DNA polymerase in a polymerase chain reaction (PCR). When coupled to an in vitro selection procedure, PCR amplification generated receptors that bind ATP. This is the first example of an in vitro selection generating oligonucleotide receptors where the oligonucleotide library has incorporated a cationic nucleotide functionality. The selection yielded functionalized receptors having sequences differing from a motif known to arise in a standard selection experiment using only natural nucleotides. Surprisingly, both the natural and the functionalized motifs convergently evolved to bind not one, but two ATP molecules cooperatively. Likewise, the affinity of the receptors for ATP had converged; in both cases, the receptors are half saturated at the 3 mM concentrations of ATP presented during the selection. The convergence of phenotype suggests that the outcome of this selection experiment was determined by features of the environment during which selection occurs, in particular, a highly loaded affinity resin used in the selection step. Further, the convergence of phenotype suggests that the optimal molecular phenotype has been achieved by both selections for the selection conditions. This interplay between environmental conditions demanding a function of a biopolymer and the ability of the biopolymer to deliver that function is strictly analogous to that observed during natural selection, illustrating the nature of life as a self-sustaining chemical system capable of Darwinian evolution.

Non-independence of Mnt repressor-operator interaction determined by a new quantitative multiple fluorescence relative affinity (QuMFRA) assay

Salmonella bacteriophage repressor Mnt belongs to the ribbon-helix-helix class of transcription factors. Previous SELEX results suggested that interactions of Mnt with positions 16 and 17 of the operator DNA are not independent. Using a newly developed high-throughput quantitative multiple fluorescence relative affinity (QuMFRA) assay, we directly quantified the relative equilibrium binding constants (K(ref)) of Mnt to operators carrying all the possible dinucleotide combinations at these two positions. Results show that Mnt prefers binding to C, instead of wild-type A, at position 16 when wild-type C at position 17 is changed to other bases. The measured K(ref) values of double mutants were also higher than the values predicted from single mutants, demonstrating the non-independence of these two positions. The ability to produce a large number of quantitative binding data simultaneously and the potential to scale up makes QuMFRA a valuable tool for the large-scale study of macromolecular interaction.

The hairpin loop but not the bulged C of the iron responsive element is essential for high affinity binding to iron regulatory protein-1

Vertebrates control intracellular iron concentration principally through the interaction of iron regulatory proteins with mRNAs that contain an iron responsive element, a small hairpin with a bulged C. The hairpin loop and bulged C have previously been assumed to be critical for binding and have been proposed to make direct contact with the iron regulatory proteins. However, we show here that a U or G can be substituted for the bulged C provided that specific nucleotides are also present within internal loops. The K(d), IC(50) and chemical modifications of the iron responsive element variants are similar to the wild-type. Results are more consistent with a role in which the C-bulge functions to orient the hairpin for optimal protein binding rather than to directly contact the protein. Characterization of these novel iron responsive element variants may facilitate the identification of additional mRNAs whose expression is controlled by iron regulatory proteins, as well as provide insight into the nature of a critical RNA-protein interaction.

Rational evolutionary design: the theory of in vitro protein evolution

Directed evolution uses a combination of powerful search techniques to generate proteins with improved properties. Part of the success is due to the stochastic element of random mutagenesis; improvements can be made without a detailed description of the complex interactions that constitute function or stability. However, optimization is not a conglomeration of random processes. Rather, it requires both knowledge of the system that is being optimized and a logical series of techniques that best explores the pathways of evolution (Eigen et al., 1988). The weighing of parameters associated with mutation, recombination, and screening to achieve the maximum fitness improvement is the beginning of rational evolutionary design. The optimal mutation rate is strongly influenced by the finite number of mutants that can be screened. A smooth fitness landscape implies that many mutations can be accumulated without disrupting the fitness. This has the effect of lowering the required library size to sample a higher mutation rate. As the sequence ascends the fitness landscape, the optimal mutation rate decreases as the probability of discovering improved mutations also decreases. Highly coupled regions require that many mutations be simultaneously made to generate a positive mutant. Therefore, positive mutations are discovered at uncoupled positions as the fitness of the parent increases. The benefit of recombination is twofold: it combines good mutations and searches more sequence space in a meaningful way. Recombination is most beneficial when the number of mutants that can be screened is limited and the landscape is of an intermediate ruggedness. The structure of schema in proteins leads to the conclusion that many cut points are required. The number of parents and their sequence identity are determined by the balance between exploration and exploitation. Many disparate parents can explore more space, but at the risk of losing information. The required screening effort is related to the number of uphill paths, which decreases more rapidly for rugged landscapes. Noise in the fitness measurements causes a dramatic increase in the required mutant library size, thus implying a smaller optimal mutation rate. Because of strict limitations on the number of mutants that can be screened, there is motivation to optimize the content of the mutant library. By restricting mutations to regions of the gene that are expected to show improvement, a greater return can be made with the same number of mutants. Initial studies with subtilisin E have shown that structurally tolerant positions tend to be where positive activity mutants are made during directed evolution. Mutant fitness information is produced by the screening step that has the potential to provide insight into the structure of the fitness landscape, thus aiding the setting of experimental parameters. By analyzing the mutant fitness distribution and targeting specific regions of the sequence, in vitro evolution can be accelerated. However, when expediting the search, there is a trade-off between rapid improvement and the quality of the long-term solution. The benefit of neutrality has yet to be captured with in vitro protein evolution. Neutral theory predicts the punctuated emergence of novel structure and function, however, with current methods, the required time scale is not feasible. Utilizing neutral evolution to accelerate the discovery of new functional and structural solutions requires a theory that predicts the behavior of mutational pathways between networks. Because the transition from neutral to adaptive evolution requires a multi-mutational switch, increasing the mutation rate decreases the time required for a punctuated change to occur. By limiting the search to the less coupled region of the sequence (smooth portion of the fitness landscape), the required larger mutation rate can be tolerated. Advances in directed evolution will be achieved when the driving forces behind such proce

Solution structure of the two N-terminal RNA-binding domains of nucleolin and NMR study of the interaction with its RNA target

Nucleolin is an abundant 70 kDa nucleolar protein involved in many aspects of ribosomal RNA biogenesis. The central region of nucleolin contains four tandem consensus RNA-binding domains (RBD). The two most N-terminal domains (RBD12) bind with nanomolar affinity to an RNA stem-loop containing the consensus sequence UCCCGA in the loop. We have determined the solution structure of nucleolin RBD12 in its free form and have studied its interaction with a 22 nt RNA stem-loop using multidimensional NMR spectroscopy. The two RBDs adopt the expected beta alpha beta beta alpha beta fold, but the position of the beta 2 strand in both domains differs from what was predicted from sequence alignments. RBD1 and RBD2 are significantly different from each others and this is likely important in their sequence specific recognition of the RNA. RBD1 has a longer alpha-helix 1 and a shorter beta 2-beta 3 loop than RBD2, and differs from most other RBDs in these respects. The two RBDs are separated by a 12 amino acid flexible linker and do not interact with one another in the free protein. This linker becomes ordered when RBD12 binds to the RNA. Analysis of the observed NOEs between the protein and the RNA indicates that both RBDs interact with the RNA loop via their beta-sheet. Each domain binds residues on one side of the loop; specifically, RBD2 contacts the 5' side and RBD1 contacts the 3'.

A trans-acting factor, isolated by the three-hybrid system, that influences alternative splicing of the amyloid precursor protein minigene

Two clones were isolated in a three-hybrid screen of a rat fetal brain P5 cDNA library with an intronic splicing enhancer of the amyloid precursor protein (APP) gene as RNA bait. These clones represent the rat homologues of the previously described genes CUG-binding protein (CUG-BP) and Siah-binding protein (Siah-BP). Both interact in a sequence-specific manner with the RNA bait used for library screening as well as with the CUG repeat. In contrast, no interactions were observed in the three-hybrid assay with other baits tested. In two-hybrid assays, Siah-BP interacts with U2AF65 as well as with itself. EWS, an RGG-type RNA-binding protein associated with Ewing sarcoma, was identified as an interacting partner for the CUG-BP homologue in a two-hybrid assay for protein-protein interactions performed with various factors involved in RNA metabolism. Splicing assays performed by RT-PCR from cells cotransfected with certain cDNAs and an APP minigene, used as a reporter, indicate exclusion of exon 8 if the CUG-BP homologue is present. We conclude that clone AF169013 and its counterpart in human CUG-BP could be the trans-acting factors that interact with the splicing enhancer downstream of exon 8, and in this way influence alternative splicing of the APP minigene.

The tetranucleotide UCAY directs the specific recognition of RNA by the Nova K-homology 3 domain

The Nova family of proteins are target antigens in the autoimmune disorder paraneoplastic opsoclonus-myoclonus ataxia and contain K-homology (KH)-type RNA binding domains. The Nova-1 protein has recently been shown to regulate alternative splicing of the alpha2 glycine receptor subunit pre-mRNA by binding to an intronic element containing repeats of the tetranucleotide UCAU. Here, we have used selection-amplification to demonstrate that the KH3 domain of Nova recognizes a single UCAY element in the context of a 20-base hairpin RNA; the UCAY tetranucleotide is optimally presented as a loop element of the hairpin scaffold and requires protein residues C-terminal to the previously defined KH domain. These results suggest that KH domains in general recognize tetranucleotide motifs and that biological RNA targets of KH domains may use either RNA secondary structure or repeated sequence elements to achieve high affinity and specificity of protein binding.

Selecting rRNA binding sites for the ribosomal proteins L4 and L6 from randomly fragmented rRNA: application of a method called SERF

Two-thirds of the 54 proteins of the Escherichia coli ribosome interact directly with the rRNAs, but the rRNA binding sites of only a very few proteins are known. We present a method (selection of random RNA fragments; SERF) that can identify the minimal binding region for proteins within ribonucleo-protein complexes such as the ribosome. The power of the method is exemplified with the ribosomal proteins L4 and L6. Binding sequences are identified for both proteins and characterized by phosphorothioate footprinting. Surprisingly, the binding region of L4, a 53-nt rRNA fragment of domain I of 23S rRNA, can simultaneously and independently bind L24, one of the two assembly initiator proteins of the large subunit.

Biotechnology match making: screening orphan ligands and receptors

To date there has been a considerable amount of interest and success in the pharmaceutical industry in the discovery of drug targets and diagnostics via genomic technologies, namely DNA sequencing, mutation/polymorphism detection and expression monitoring of mRNA. As the ultimate targets for the majority of these methods are actually proteins, more and more emphasis has been placed upon protein-based methods in an effort to define the function of proteins discovered by genomic technologies. One of the most challenging areas of drug target discovery facing researchers today is the search for novel receptor-ligand pairs. Database mining techniques in conjunction with other computational methods are able to identify many novel sequences of putative receptors, but the ability to similarly identify the receptor's natural ligand is not possible by these methods. The past few years have seen an increase in methodology and instrumentation focused on the ability to discover and characterize protein-protein interactions, as well as receptor-ligand pairs. Significant advances have been made in the areas of instrumentation (biosensors and fluorescent plate readers) as well as methodologies relating to phage/ribosome display and library construction.

In vitro selection of DNA aptamers to anthrax spores with electrochemiluminescence detection

Systematic evolution of ligands by exponential enrichment (SELEX) was used to select and PCR amplify DNA sequences (aptamers) capable of binding to and detecting nonpathogenic Sterne strain Bacillus anthracis spores. A simplified affinity separation approach was employed, in which autoclaved anthrax spores were used as the separation matrix. An aptamer-magnetic bead-electrochemiluminescence (AM-ECL) sandwich assay scheme was devised for detecting anthrax spores. Using a low SELEX DNA to spore ratio (154 ng DNA/10(6) spores), at least three distinct populations of single-stranded DNA aptamers, having varied affinities for anthrax spores, were noted by the AM-ECL assay. Results reflect detection of spore components with a dynamic range equivalent to < 10- > 6 x 10(6) anthrax spores. In the low DNA to spore ratio experiments, aptamers could be liberated from spore pellets by heating at 96 degrees C for 5 min after each round of SELEX. When a much higher DNA to spore ratio (10,256 ng DNA/10(6) spores) was used for SELEX development, a higher affinity set of aptamers was selected that could not be heat-eluted even at 99 degrees C for 5 min following round four of SELEX. However, high affinity spore surface bound aptamers were detectable via their 5'-biotinylated tails using labeled avidin and could be eluted in deionized water. Aptamers have potential for use as inexpensive, in vitro-generated receptors for biosensors in biological warfare detection and other areas.

Modeling DNA shuffling

In vitro evolution is a new, important laboratory method to evolve molecules with desired properties. It has been used in a variety of biological studies and drug development. In this paper, we study one important mutagenesis method used in in vitro evolution experiments called DNA shuffling. We construct a mathematical model for DNA shuffling and study the properties of molecules after DNA shuffling experiments based on this model. The model for DNA shuffling consists of two parts. First we apply the Lander-Waterman model for physical mapping by fingerprinting random clones to model the distribution of regions that can be reassembled through DNA shuffling. Then we present a model for recombination between different DNA species with different mutations. We compare our theoretical results with experimental data. Finally we propose novel applications of the theoretical results to the optimal design of DNA shuffling experiments and to physical mapping using DNA shuffling.

Ribosome display efficiently selects and evolves high-affinity antibodies in vitro from immune libraries

Ribosome display was applied for affinity selection of antibody single-chain fragments (scFv) from a diverse library generated from mice immunized with a variant peptide of the transcription factor GCN4 dimerization domain. After three rounds of ribosome display, positive scFvs were isolated and characterized. Several different scFvs were selected, but those in the largest group were closely related to each other and differed in 0 to 5 amino acid residues with respect to their consensus sequence, the likely common progenitor. The best scFv had a dissociation constant of (4 +/- 1) x 10(-11) M, measured in solution. One amino acid residue in complementarity determining region L1 was found to be responsible for a 65-fold higher affinity than the likely progenitor. It appears that this high-affinity scFv was selected from the mutations occurring during ribosome display in vitro, and that this constitutes an affinity maturation inherent in this method. The in vitro-selected scFvs could be functionally expressed in the Escherichia coli periplasm with good yields or prepared by in vitro refolding. Thus, ribosome display can be a powerful methodology for in vitro library screening and simultaneous sequence evolution.

Identification of functional exonic splicing enhancer motifs recognized by individual SR proteins

Using an in vitro randomization and functional selection procedure, we have identified three novel classes of exonic splicing enhancers (ESEs) recognized by human SF2/ASF, SRp40, and SRp55, respectively. These ESEs are functional in splicing and are highly specific. For SF2/ASF and SRp55, in most cases, only the cognate SR protein can efficiently recognize an ESE and activate splicing. In contrast, the SRp40-selected ESEs can function with either SRp40 or SRp55, but not with SF2/ASF. UV cross-linking/competition and immunoprecipitation experiments showed that SR proteins recognize their cognate ESEs in nuclear extract by direct and specific binding. A motif search algorithm was used to derive consensus sequences for ESEs recognized by these SR proteins. Each SR protein yielded a distinct 5- to 7-nucleotide degenerate consensus. These three consensus sequences occur at higher frequencies in exons than in introns and may thus help define exon-intron boundaries. They occur in clusters within regions corresponding to naturally occurring, mapped ESEs. We conclude that a remarkably diverse set of sequences can function as ESEs. The degeneracy of these motifs is consistent with the fact that exonic enhancers evolved within extremely diverse protein coding sequences and are recognized by a small number of SR proteins that bind RNA with limited sequence specificity.

DNA ligands that bind tightly and selectively to cellobiose

Cell surface oligosaccharides have been shown to play essential biological roles in such diverse biological phenomena as cellular adhesion, molecular recognition, and inflammatory response. The development of high-affinity ligands capable of selectively recognizing a variety of small motifs in different oligosaccharides would be of significant interest as experimental and diagnostic tools. As a step toward this goal we have developed DNA ligands that recognize the disaccharide cellobiose, whether in soluble form or as the repeating unit of the polymer, cellulose. These DNA "aptamers" bind with high selectivity to cellobiose with little or no affinity for the related disaccharides lactose, maltose, and gentiobiose. Thus, the DNA ligands can discriminate sugar epimers, anomers, and disaccharide linkages.

The mathematics of SELEX against complex targets

We have developed a computer model for the simulation of simultaneous SELEX against multiple targets. The model assumes equilibrium behavior for the formation of binary ligand:target complexes, and that there is no ligand:ligand or target:target interaction. Target concentrations, ligand concentrations, and affinity distributions of the initial ligand pool for each individual target may be set by the user. We have used this program to gain an understanding of how the presence of multiple targets affects the selection process. In most cases, we find that SELEX is capable of generating different ligands for the different targets in a heterogeneous mixture, regardless of large variations in target concentrations and ligand:target affinities. A low relative partitioning efficiency (the efficiency with which ligands complexed with a target are separated from free ligands) for a target in a mixture gives a greatly reduced rate of selection of high-affinity ligands to that target. The ratio of each high-affinity ligand to its individual target within a pool of ligands selected for binding against a mixture of targets is approximately proportional to the concentration of the target multiplied by the ligand:target partitioning efficiency.

Stochastic modeling and optimization of phage display

Phage display, SELEX and other methods of combinatorial chemistry have become very popular means of finding ligands with high affinities to given targets. Despite their success, they suffer from numerous sources of error and bias, such as very low initial concentrations of species, non-specific binding, and the sampling of only a tiny fraction of the library at the end of an experiment. To understand the interaction of these errors and to better devise molecular search strategies that take the errors into account, I devise and analyze a highly detailed model of phage display. The model is specifically designed to study the influence of the stochastic nature of each laboratory step. The model includes phage multivalency, multiple classes of targets, and solid-phase equilibrium and washing, yet it is amenable to analytic results and rapid computer simulation. With both analytic and simulation approaches, I: (1) describe the effects of target concentration, phage valency, degree of background binding and other laboratory parameters on the probabilities of phage binding and of being selected; (2) show the effects of an increasing selection stringency strategy and how it results in a tradeoff between rapid library enrichment and high probability of sampling the best ligands; and (3) show how the number of phage sampled for detailed study at the end of a search alters search success. The work concludes with several practical suggestions for the control of selection stringency.

High affinity ligands from in vitro selection: complex targets

Human red blood cell membranes were used as a model system to determine if the systematic evolution of ligands by exponential enrichment (SELEX) methodology, an in vitro protocol for isolating high-affinity oligonucleotides that bind specifically to virtually any single protein, could be used with a complex mixture of potential targets. Ligands to multiple targets were generated simultaneously during the selection process, and the binding affinities of these ligands for their targets are comparable to those found in similar experiments against pure targets. A secondary selection scheme, deconvolution-SELEX, facilitates rapid isolation of the ligands to targets of special interest within the mixture. SELEX provides high-affinity compounds for multiple targets in a mixture and might allow a means for dissecting complex biological systems.

New DNA sequence rules for high affinity binding to histone octamer and sequence-directed nucleosome positioning

DNA sequences that position nucleosomes are of increasing interest because of their relationship to gene regulation in vivo and because of their utility in studies of nucleosome structure and function in vitro. However, at present our understanding of the rules for DNA sequence-directed nucleosome positioning is fragmentary, and existing positioning sequences have many limitations. We carried out a SELEX experiment starting with a large pool of chemically synthetic random. DNA molecules to identify those individuals having the highest affinity for histone octamer. A set of highest-affinity molecules were selected, cloned, and sequenced, their affinities (free energies) for histone octamer in nucleosome reconstitution measured, and their ability to position nucleosomes in vitro assessed by native gel electrophoresis. The selected sequences have higher affinity than previously known natural or non-natural sequences, and have a correspondingly strong nucleosome positioning ability. A variety of analyses including Fourier transform, real-space correlation, and direct counting computations were carried out to assess non-random features in the selected sequences. The results reveal sequence rules that were already identified in earlier studies of natural nucleosomal DNA, together with a large set of new rules having even stronger statistical significance. Possible physical origins of the selected molecules' high affinities are discussed. The sequences isolated in this study should prove valuable for studies of chromatin structure and function in vitro and, potentially, for studies in vivo.

Quantitative specificity of the Mnt repressor

The Mnt protein of Salmonella phage P22 binds site-specifically to its operator. To better understand this binding we used dideoxy DNA sequencing in a quantitative manner to determine the relative binding constants, and hence the relative free energies, of wild-type Mnt protein to a substantial number of variants of its operator. These measurements were supported by experiments which used the SELEX procedure to generate a set of operators from an initially randomized population. In the Discussion we show that the present model of Mnt protein/operator binding, due to Sauer and co-workers, along with the assumption of an independent contribution of each position in the operator to the total binding, provides a reasonably accurate description of the system. We also discuss the use of information content as a measure of DNA-protein binding specificity with the Mnt protein/operator system serving as an example and show again that the assumption of independence supports the current view of this case of site-specific binding.

Interlocking structural motifs mediate molecular discrimination by a theophylline-binding RNA

To visualize the interplay of RNA structural interactions in a ligand binding site, we have determined the solution structure of a high affinity RNA-theophylline complex using NMR spectroscopy. The structure provides insight into the ability of this in vitro selected RNA to discriminate theophylline from the structurally similar molecule caffeine. Numerous RNA structural motifs combine to form a well-ordered binding pocket where an intricate network of hydrogen bonds and stacking interactions lock the theophylline into the complex. Two internal loops interact to form the binding site which consists of a sandwich of three base triples. The complex also contains novel base-zipper and 1-3-2 stacking motifs, in addition to an adenosine platform and a reversed sugar. An important feature of the RNA is that many of the conserved core residues participate in multiple overlapping tertiary interactions. This complex illustrates how interlocking structural motifs can be assembled into a highly specific ligand-binding site that possesses high levels of affinity and molecular discrimination.

In vitro selection of RNAs that bind to the human immunodeficiency virus type-1 gag polyprotein

RNA ligands that bind to the human immunodeficiency virus type-1 (HIV-1) gag polyprotein with 10(-9) M affinity were isolated from a complex pool of RNAs using an in vitro selection method. The ligands bind to two different regions within gag, either to the matrix protein or to the nucleocapsid protein. Binding of a matrix ligand to gag did not interfere with the binding of a nucleocapsid ligand, and binding of a nucleocapsid ligand to gag did not interfere with the binding of a matrix ligand. However, binding of a nucleocapsid ligand to gag did interfere with binding of an RNA containing the HIV-1 RNA packaging element (psi), even though the sequence of the nucleocapsid ligand is not similar topsi. The minimal sequences required for the ligands to bind to matrix or nucleocapsid were determined. Minimal nucleocapsid ligands are predicted to form a stem-loop structure that has a self-complementary sequence at one end. Minimal matrix ligands are predicted to form a different stem-loop structure that has a CAARU loop sequence. The properties of these RNA ligands may provide tools for studying RNA interactions with matrix and nucleocapsid, and a novel method for inhibiting HIV replication.

The role of peptide modules in protein evolution

Protein evolution shows interesting strategies to be used in protein design. During evolution the creation of new proteins has been accomplished by combining different peptide modules, i.e. evolutionary successful stable folding units. Thereby, the evolution of proteins has been greatly enhanced. Today this mechanism of recombining optimized building blocks to design new proteins has been introduced into applied molecular evolution.

Post-SELEX combinatorial optimization of aptamers

In vitro selection techniques provide a means of isolating nucleic acid ligands for binding to particular protein targets. Although most aptamers have quite high affinities for their target proteins, it has been shown that post-SELEX modification can result in further enhancement of binding affinity, as well as other desired properties. This has led to the current development of a more systematic approach to aptamer optimization using a combinatorial screening methodology.