The solution structure of an N-terminally truncated version of the yeast CDC24p PB1 domain shows a different beta-sheet topology.

Phox and Bem1 (PB1) domains mediate protein-protein interactions via the formation of homo- or hetero-dimers. The C-terminal PB1 domain of yeast cell division cycle 24 (CDC24p), a guanine-nucleotide exchange factor involved in cell polarity establishment, is known to interact with the PB1 domain occurring in bud emergence MSB1 interacting 1 (BEM1p) during the regulation of the yeast budding process via its OPR/PC/AID (OPCA) motif. Here, we present the structure of an N-terminally truncated version of the Sc CDC24p PB1 domain. It shows a different topology of the beta-sheet than the long form. However, the C-terminal part of the structure shows the conserved PB1 domain features including the OPCA motif with a slight rearrangement of helix alpha1. Residues which are important for the heterodimerization with BEM1p are structurally preserved.

Quantification of PDZ domain specificity, prediction of ligand affinity and rational design of super-binding peptides.

Transient macromolecular complexes are often formed by protein-protein interaction domains (e.g. PDZ, SH2, SH3, WW) which recognize linear sequence motifs with in vitro affinities typically in the micromolar range. The analysis of the resulting interaction networks requires a quantification of domain specificity and selectivity towards all possible ligands with physiologically relevant affinity. As representative examples, we determined specificity as a function of ligand sequence-dependent affinity contributions by statistical analysis of peptide library screens for the AF6, ERBIN and SNA1 (alpha-1-syntrophin) PDZ domains. For this purpose, the three PDZ domains were first screened for binding with a peptide library comprising 6223 human C termini created by SPOT synthesis. Based on the detected ligand preferences, we designed focused peptide libraries (profile libraries). These libraries were used to quantify the affinity contributions of the four C-terminal ligand residues by means of ANOVA models (analysis of variance) relating the C-terminal ligand sequences to the corresponding dissociation constants. Our models agreed well with experimentally determined dissociation constants and allowed us to design super binding peptides. The latter were shown experimentally to bind to their cognate PDZ domains with the highest affinity. In addition, we determined structure-activity relationships and thereby rationalized the position-specific affinity contributions. Furthermore, we used the statistical models to predict the dissociation constants for the complete ligand sequence space and thus determined the specificity overlap for the three investigated PDZ domains (). Altogether, we present an efficient method for profiling protein-protein interaction domains that provides a biophysical picture of specificity and selectivity. This approach allows the rational design of functional experiments and provides a basis for simulating interaction networks in the field of systems biology.

Statistically significant dependence of the Xaa-Pro peptide bond conformation on secondary structure and amino acid sequence.

BACKGROUND: A reliable prediction of the Xaa-Pro peptide bond conformation would be a useful tool for many protein structure calculation methods. We have analyzed the Protein Data Bank and show that the combined use of sequential and structural information has a predictive value for the assessment of the cis versus trans peptide bond conformation of Xaa-Pro within proteins. For the analysis of the data sets different statistical methods such as the calculation of the Chou-Fasman parameters and occurrence matrices were used. Furthermore we analyzed the relationship between the relative solvent accessibility and the relative occurrence of prolines in the cis and in the trans conformation. RESULTS: One of the main results of the statistical investigations is the ranking of the secondary structure and sequence information with respect to the prediction of the Xaa-Pro peptide bond conformation. We observed a significant impact of secondary structure information on the occurrence of the Xaa-Pro peptide bond conformation, while the sequence information of amino acids neighboring proline is of little predictive value for the conformation of this bond. CONCLUSION: In this work, we present an extensive analysis of the occurrence of the cis and trans proline conformation in proteins. Based on the data set, we derived patterns and rules for a possible prediction of the proline conformation. Upon adoption of the Chou-Fasman parameters, we are able to derive statistically relevant correlations between the secondary structure of amino acid fragments and the Xaa-Pro peptide bond conformation.

The solution structure of the SODD BAG domain reveals additional electrostatic interactions in the HSP70 complexes of SODD subfamily BAG domains.

The solution structure of an N-terminally extended construct of the SODD BAG domain was determined by nuclear magnetic resonance spectroscopy. A homology model of the SODD-BAG/HSP70 complex reveals additional possible interactions that are specific for the SODD subfamily of BAG domains while the overall geometry of the complex remains the same. Relaxation rate measurements show that amino acids N358-S375 of SODD which were previously assigned to its BAG domain are not structured in our construct. The SODD BAG domain is thus indeed smaller than the homologous domain in Bag1 defining a new subfamily of BAG domains.

The SEP domain of p47 acts as a reversible competitive inhibitor of cathepsin L.

The solution structure of the human p47 SEP domain in a construct comprising residues G1-S2-p47(171-270) was determined by NMR spectroscopy. A structure-derived hypothesis about the domains' function was formulated and pursued in binding experiments with cysteine proteases. The SEP domain was found to be a reversible competitive inhibitor of cathepsin L with a Ki of 1.5 microM. The binding of G1-S2-p47(171-270) to cathepsin L was mapped by biochemical assays and the binding interface was investigated by NMR chemical shift perturbation experiments.

Fast identification of folded human protein domains expressed in E. coli suitable for structural analysis.

BACKGROUND: High-throughput protein structure analysis of individual protein domains requires analysis of large numbers of expression clones to identify suitable constructs for structure determination. For this purpose, methods need to be implemented for fast and reliable screening of the expressed proteins as early as possible in the overall process from cloning to structure determination. RESULTS: 88 different E. coli expression constructs for 17 human protein domains were analysed using high-throughput cloning, purification and folding analysis to obtain candidates suitable for structural analysis. After 96 deep-well microplate expression and automated protein purification, protein domains were directly analysed using 1D 1H-NMR spectroscopy. In addition, analytical hydrophobic interaction chromatography (HIC) was used to detect natively folded protein. With these two analytical methods, six constructs (representing two domains) were quickly identified as being well folded and suitable for structural analysis. CONCLUSION: The described approach facilitates high-throughput structural analysis. Clones expressing natively folded proteins suitable for NMR structure determination were quickly identified upon small scale expression screening using 1D 1H-NMR and/or analytical HIC. This procedure is especially effective as a fast and inexpensive screen for the 'low hanging fruits' in structural genomics.

COPS--cis/trans peptide bond conformation prediction of amino acids on the basis of secondary structure information.

UNLABELLED: COPS predicts for all 20 naturally occurring amino acids whether the peptide bond in a protein is in cis or trans conformation. The algorithm is based only on secondary structure information of amino acid triplets without considering the amino acid sequence information. Conformation parameters are derived from solved 3D structures deposited in the PDB and led to propensities based on modified Chou-Fasman parameters. COPS analyses amino acid triplets taking only their respective secondary structure into consideration and upon application of a set of rules utilizing the conformation parameters, the N-terminal peptide bond conformation of the middle residue is predicted. COPS was tested on a random selection of protein datasets. AVAILABILITY: The COPS program and further information are freely available from the FMP website at http://www.fmp-berlin.de/nmr/cops CONTACT: labudde@fmp-berlin.de.

Influence of chemical shift tolerances on NMR structure calculations using ARIA protocols for assigning NOE data.

Large-scale protein structure determination by NMR via automatic assignment of NOESY spectra requires the adjustment of several parameters for optimal performance. Among those are the chemical shift tolerance windows (delta), which allow for the compensation of badly matching chemical shifts in the assignment-list and peak-lists, and the maximum number of assignment possibilities allowed per peak (n(max)). Here, we test the influence of different values for Delta and n(max) on the performance of automated assignment of NOESY spectra by ARIA. Using Cesta.py (a Python script available from http://pasteur.fr/binfs/), we analyse the number of rejected peaks and the average number of assignments as a function of Delta and derive criteria for optimising delta and n(max) prior to structure calculation. The analysis also makes it possible to detect inconsistencies in the dataset, e.g., badly matching frequencies in the NOESY peak-lists and in the provided assignment-list. We show that ARIA can deal with a large number of assignment possibilities for each peak, provided the correct option is present, and that consequently narrow tolerances should be avoided.

A modified strategy for sequence specific assignment of protein NMR spectra based on amino acid type selective experiments.

The determination of the three-dimensional structure of a protein or the study of protein-ligand interactions requires the assignment of all relevant nuclei as an initial step. This is nowadays almost exclusively performed using triple-resonance experiments. The conventional strategy utilizes one or more pairs of three dimensional spectra to obtain redundant information and thus reliable assignments. Here, a modified strategy for obtaining sequence specific assignments based on two dimensional amino acid type selective triple-resonance experiments is proposed. These experiments can be recorded with good resolution in a relatively short time. They provide very specific and redundant information, in particular on sequential connectivities, that drastically increases the ease and reliability of the assignment procedure, done either manually or in an automated fashion. The new strategy is demonstrated with the protein domain PB1 from yeast CDC24p.

Solution structure, backbone dynamics, and association behavior of the C-terminal BRCT domain from the breast cancer-associated protein BRCA1.

BRCA1 is a tumor suppressor protein associated with breast and ovarian cancer. The C-terminal region of BRCA1 consists of two closely spaced BRCT domains which mediate essential biological functions, including regulation of transcription and control of cell-cycle progression by their interaction with phosphorylated effector proteins. Here we report the NMR structure of the isolated C-terminal BRCT domain (BRCT-c) from human BRCA1. BRCT-c is well-structured in solution, folding independently in the absence of its BRCT-n counterpart. Ultracentrifugation experiments and size exclusion chromatography reveal that BRCT-c exists as a monomer under near-physiological conditions. Dynamics measurements from NMR data show three loops which coincide with the most variable sequence regions in BRCT domains, to be genuinely flexible in solution. The solution structure of BRCT-c shows subtle conformational changes when compared to the crystal structure of BRCT-c in the tandem repeat of BRCA1. These affect sites involved in formation of the BRCT-n-BRCT-c interface and the binding to phosphoserine-containing peptides. The results suggest that the presence of native BRCT-n and a properly aligned BRCT-n-BRCT-c interface are essential if BRCT-c is to adopt a biologically active conformation. Structural consequences of cancer-associated mutations and biological implications of the dynamic behavior are discussed.