Evaluation of a modular strategy for the construction of novel polydactyl zinc finger DNA-binding proteins

In previous studies, we have developed a technology for the rapid construction of novel DNA-binding proteins with the potential to recognize any unique site in a given genome. This technology relies on the modular assembly of modified zinc finger DNA-binding domains, each of which recognizes a three bp subsite of DNA. A complete set of 64 domains would provide comprehensive recognition of any desired DNA sequence, and new proteins could be assembled by any laboratory in a matter of hours. However, a critical parameter for this approach is the extent to which each domain functions as an independent, modular unit, without influence or dependence on its neighboring domains. We therefore examined the detailed binding behavior of several modularly assembled polydactyl zinc finger proteins. We first demonstrated that 80 modularly assembled 3-finger proteins can recognize their DNA target with very high specificity using a multitarget ELISA-based specificity assay. A more detailed analysis of DNA binding specificity for eight 3-finger proteins and two 6-finger proteins was performed using a target site selection assay. Results showed that the specificity of these proteins was as good or better than that of zinc finger proteins constructed using methods that allow for interdependency. In some cases, near perfect specificity was achieved. Complications due to target site overlap were found to be restricted to only one particular amino acid interaction (involving an aspartate in position 2 of the alpha-helix) that occurs in a minority of cases. As this is the first report of target site selection for designed, well characterized 6-finger proteins, unique insights are discussed concerning the relationship of protein length and specificity. These results have important implications for the design of proteins that can recognize extended DNA sequences, as well as provide insights into the general rules of recognition for naturally occurring zinc finger proteins.     

Computer-aided design of modular protein devices: Boolean AND gene activation

Many potentially useful synthetic gene networks require the expression of an engineered gene if and only if two different DNA-binding proteins exist in sufficient concentration. While some natural and engineered systems activate gene expression according to a logical AND-like behavior, they often utilize allosteric or cooperative protein-protein interactions, rendering their components unsuitable for a toolbox of modular parts for use in multiple applications. Here, we develop a quantitative model to demonstrate that a small system of interacting fusion proteins, called a protein device, can activate an engineered gene according to the Boolean AND behavior while using only modular protein domains and DNA sites. The fusion proteins are created from transactivating, DNA-binding, non-DNA binding, and protein-protein interaction domains along with the corresponding peptide ligands. Using a combined kinetic and thermodynamic model, we identify the characteristics of the molecular components and their rates of constitutive production that maximize the fidelity of AND behavior. These AND protein devices facilitate the creation of complex genetic programs and may be used to create gene therapies, biosensors and other biomedical and biotechnological applications that turn on gene expression only when multiple DNA-binding proteins are simultaneously present.     

Dissection of reverse gyrase activities: insight into the evolution of a thermostable molecular machine

Reverse gyrase is a peculiar DNA topoisomerase, specific of thermophilic microorganisms, which induces positive supercoiling into DNA molecules in an ATP-dependent reaction. It is a modular enzyme and comprises an N-terminal helicase-like module fused to a C-terminal topoisomerase IA-like domain. The exact molecular mechanism of this unique reaction is not understood, and a fundamental mechanistic question is how its distinct steps are coordinated. We studied the cross-talk between the components of this molecular motor and probed communication between the DNA-binding sites and the different activities (DNA relaxation, ATP hydrolysis and positive supercoiling). We show that the isolated ATPase and topoisomerase domains of reverse gyrase form specific physical interactions, retain their own DNA binding and enzymatic activities, and when combined cooperate to achieve the unique ATP-dependent positive supercoiling activity. Our results indicate a mutual effect of both domains on all individual steps of the reaction. The C-terminal domain shows ATP-independent topoisomerase activity, which is repressed by the N-terminal domain in the full-length enzyme; experiments with the isolated domains showed that the C-terminal domain has stimulatory influence on the ATPase activity of the N-terminal domain. In addition, the two domains showed a striking reciprocal thermostabilization effect.     

Determinants of high-affinity DNA binding by the glucocorticoid receptor: evaluation of receptor domains outside the DNA-binding domain.

In this study, we have investigated the influence of regions outside the DNA-binding domain of the human glucocorticoid receptor on high-affinity DNA binding. We find that the DNA-binding domain shows a 10-fold lower affinity for a palindromic DNA-binding site than the intact receptor. The N-terminal part of the receptor protein does not influence its DNA-binding affinity, while the C-terminal steroid-binding domain increases the DNA-binding affinity of the receptor molecule. It has previously been shown that both the intact glucocorticoid receptor and the glucocorticoid receptor DNA-binding domain bind to a palindromic glucocorticoid response element on DNA as dimers. It is likely that differences in DNA-binding affinity observed result from protein-protein interactions outside the DNA-binding domain between receptor monomers, as has been shown for the estrogen receptor. We have previously identified a segment involved in protein-protein interactions between DNA-binding domains of glucocorticoid receptors. This, in combination with results presented in this study, suggests that there are at least two sites of contact between receptor monomers bound to DNA. We suggest that the interaction between the DNA-binding domains may act primarily to restrict DNA binding to binding sites with appropriate half-site spacing and that additional stability of the receptor dimer is provided by the interactions between the steroid-binding domains.     

Systematic method for the detection of potential lambda Cro-like DNA-binding regions in proteins

We have developed and tested a systematic method for the location and statistical evaluation of potential DNA-binding regions of the lambda Cro type in protein sequences. Using this approach to examine proteins expected to contain such regions, we have been able to compile a statistically homogeneous master set of 37 lambda Cro-like DNA-binding domains. Examination of a protein database revealed other prokaryotic proteins that are similar to this lambda Cro-like group. There are also many DNA-binding proteins that are not found to be significantly similar to the lambda Cro group, consistent with previous suggestions that different types of protein sequence may be able to achieve a similar mode of binding and that there exist other modes of sequence-specific DNA-binding. A useful feature of the method is that it can be applied without a computer.     

Building an automated classification of DNA-binding protein domains

Intensive growth in 3D structure data on DNA-protein complexes as reflected in the Protein Data Bank (PDB) demands new approaches to the annotation and characterization of these data and will lead to a new understanding of critical biological processes involving these data. These data and those from other protein structure classifications will become increasingly important for the modeling of complete proteomes. We propose a fully automated classification of DNA-binding protein domains based on existing 3D-structures from the PDB. The classification, by domain, relies on the Protein Domain Parser (PDP) and the Combinatorial Extension (CE) algorithm for structural alignment. The approach involves the analysis of 3D-interaction patterns in DNA-protein interfaces, assignment of structural domains interacting with DNA, clustering of domains based on structural similarity and DNA-interacting patterns. Comparison with existing resources on describing structural and functional classifications of DNA-binding proteins was used to validate and improve the approach proposed here. In the course of our study we defined a set of criteria and heuristics allowing us to automatically build a biologically meaningful classification and define classes of functionally related protein domains. It was shown that taking into consideration interactions between protein domains and DNA considerably improves the classification accuracy. Our approach provides a high-throughput and up-to-date annotation of DNA-binding protein families which can be found at http://spdc.sdsc.edu.     

DomPep--a general method for predicting modular domain-mediated protein-protein interactions

Protein-protein interactions (PPIs) are frequently mediated by the binding of a modular domain in one protein to a short, linear peptide motif in its partner. The advent of proteomic methods such as peptide and protein arrays has led to the accumulation of a wealth of interaction data for modular interaction domains. Although several computational programs have been developed to predict modular domain-mediated PPI events, they are often restricted to a given domain type. We describe DomPep, a method that can potentially be used to predict PPIs mediated by any modular domains. DomPep combines proteomic data with sequence information to achieve high accuracy and high coverage in PPI prediction. Proteomic binding data were employed to determine a simple yet novel parameter Ligand-Binding Similarity which, in turn, is used to calibrate Domain Sequence Identity and Position-Weighted-Matrix distance, two parameters that are used in constructing prediction models. Moreover, DomPep can be used to predict PPIs for both domains with experimental binding data and those without. Using the PDZ and SH2 domain families as test cases, we show that DomPep can predict PPIs with accuracies superior to existing methods. To evaluate DomPep as a discovery tool, we deployed DomPep to identify interactions mediated by three human PDZ domains. Subsequent in-solution binding assays validated the high accuracy of DomPep in predicting authentic PPIs at the proteome scale. Because DomPep makes use of only interaction data and the primary sequence of a domain, it can be readily expanded to include other types of modular domains.