Adrian A. Canutescu

A graph-theory algorithm for rapid protein side-chain prediction

Fast and accurate side‐chain conformation prediction is important for homology modeling, ab initio protein structure prediction, and protein design applications. Many methods have been presented, although only a few computer programs are publicly available. The SCWRL program is one such method and is widely used because of its speed, accuracy, and ease of use. A new algorithm for SCWRL is presented that uses results from graph theory to solve the combinatorial problem encountered in the side‐chain prediction problem. In this method, side chains are represented as vertices in an undirected graph. Any two residues that have rotamers with nonzero interaction energies are considered to have an edge in the graph. The resulting graph can be partitioned into connected subgraphs with no edges between them. These subgraphs can in turn be broken into biconnected components, which are graphs that cannot be disconnected by removal of a single vertex. The combinatorial problem is reduced to finding the minimum energy of these small biconnected components and combining the results to identify the global minimum energy conformation. This algorithm is able to complete predictions on a set of 180 proteins with 34,342 side chains in <7 min of computer time. The total χ1 and χ1 + 2 dihedral angle accuracies are 82.6% and 73.7% using a simple energy function based on the backbone‐dependent rotamer library and a linear repulsive steric energy. The new algorithm will allow for use of SCWRL in more demanding applications such as sequence design and ab initio structure prediction, as well addition of a more complex energy function and conformational flexibility, leading to increased accuracy.

Cyclic coordinate descent: A robotics algorithm for protein loop closure

In protein structure prediction, it is often the case that a protein segment must be adjusted to connect two fixed segments. This occurs during loop structure prediction in homology modeling as well as in ab initio structure prediction. Several algorithms for this purpose are based on the inverse Jacobian of the distance constraints with respect to dihedral angle degrees of freedom. These algorithms are sometimes unstable and fail to converge. We present an algorithm developed originally for inverse kinematics applications in robotics. In robotics, an end effector in the form of a robot hand must reach for an object in space by altering adjustable joint angles and arm lengths. In loop prediction, dihedral angles must be adjusted to move the C‐terminal residue of a segment to superimpose on a fixed anchor residue in the protein structure. The algorithm, referred to as cyclic coordinate descent or CCD, involves adjusting one dihedral angle at a time to minimize the sum of the squared distances between three backbone atoms of the moving C‐terminal anchor and the corresponding atoms in the fixed C‐terminal anchor. The result is an equation in one variable for the proposed change in each dihedral. The algorithm proceeds iteratively through all of the adjustable dihedral angles from the N‐terminal to the C‐terminal end of the loop. CCD is suitable as a component of loop prediction methods that generate large numbers of trial structures. It succeeds in closing loops in a large test set 99.79% of the time, and fails occasionally only for short, highly extended loops. It is very fast, closing loops of length 8 in 0.037 sec on average.

Structure of a human ASF1a-HIRA complex and insights into specificity of histone chaperone complex assembly.

Human HIRA, ASF1a, ASF1b and CAF-1 are evolutionally conserved histone chaperones that form multiple functionally distinct chromatin-assembly complexes, with roles linked to diverse nuclear process, such as DNA replication and formation of heterochromatin in senescent cells. We report the crystal structure of an ASF1a–HIRA heterodimer and a biochemical dissection of ASF1as mutually exclusive interactions with HIRA and the p60 subunit of CAF-1. The HIRA B domain forms an antiparallel β-hairpin that binds perpendicular to the strands of the β-sandwich of ASF1a, via β-sheet, salt bridge and van der Waals contacts. The N- and C-terminal regions of ASF1a and ASF1b determine the different affinities of these two proteins for HIRA, by contacting regions outside the HIRA B domain. CAF-1 p60 also uses B domain–like motifs for binding to ASF1a, thereby competing with HIRA. Together, these studies begin to define the molecular determinants of assembly of functionally diverse macromolecular histone chaperone complexes.

SCWRL and MolIDE: computer programs for side-chain conformation prediction and homology modeling

SCWRL and MolIDE are software applications for prediction of protein structures. SCWRL is designed specifically for the task of prediction of side-chain conformations given a fixed backbone usually obtained from an experimental structure determined by X-ray crystallography or NMR. SCWRL is a command-line program that typically runs in a few seconds. MolIDE provides a graphical interface for basic comparative (homology) modeling using SCWRL and other programs. MolIDE takes an input target sequence and uses PSI-BLAST to identify and align templates for comparative modeling of the target. The sequence alignment to any template can be manually modified within a graphical window of the target–template alignment and visualization of the alignment on the template structure. MolIDE builds the model of the target structure on the basis of the template backbone, predicted side-chain conformations with SCWRL and a loop-modeling program for insertion–deletion regions with user-selected sequence segments. SCWRL and MolIDE can be obtained at http://dunbrack.fccc.edu/Software.php.

Abrogation of fibroblast activation protein enzymatic activity attenuates tumor growth

Tumor-associated fibroblasts are functionally and phenotypically distinct from normal fibroblasts that are not in the tumor microenvironment. Fibroblast activation protein is a 95 kDa cell surface glycoprotein expressed by tumor stromal fibroblasts, and has been shown to have dipeptidyl peptidase and collagenase activity. Site-directed mutagenesis at the catalytic site of fibroblast activation protein, Ser624 → Ala624, resulted in an ∼100,000-fold loss of fibroblast activation protein dipeptidyl peptidase (DPP) activity. HEK293 cells transfected with wild-type fibroblast activation protein, enzymatic mutant (S624A) fibroblast activation protein, or vector alone, were inoculated subcutaneously into immunodeficient mouse to assess the contribution of fibroblast activation protein enzymatic activity to tumor growth. Overexpression of wild-type fibroblast activation protein showed growth potentiation and enhanced tumorigenicity compared with both fibroblast activation protein S624A and vector-transfected HEK293 xenografts. HEK293 cells transfected with fibroblast activation protein S624A showed tumor growth rates and tumorigenicity potential similar only to vector-transfected HEK293. In vivo assessment of fibroblast activation protein DPP activity of these tumors showed enhanced enzymatic activity of wild-type fibroblast activation protein, with only baseline levels of fibroblast activation protein DPP activity in either fibroblast activation protein S624A or vector-only xenografts. These results indicate that the enzymatic activity of fibroblast activation protein is necessary for fibroblast activation protein–driven tumor growth in the HEK293 xenograft model system. This establishes the proof-of-principle that the enzymatic activity of fibroblast activation protein plays an important role in the promotion of tumor growth, and provides an attractive target for therapeutics designed to alter fibroblast activation protein–induced tumor growth by targeting its enzymatic activity.

Analysis of KIT Mutations in Sporadic and Familial Gastrointestinal Stromal Tumors: Therapeutic Implications through Protein Modeling

Purpose: Gastrointestinal stromal tumors (GIST) are characterized by expressing a gain-of-function mutation in KIT, and to a lesser extent, PDGFR. Imatinib mesylate, a tyrosine kinase inhibitor, has activity against GISTs that contain oncogenic mutations of KIT. In this study, KIT and PDGFRα mutation status was analyzed and protein modeling approaches were used to assess the potential effect of KIT mutations in response to imatinib therapy. Experimental Design: Genomic DNA was isolated from GIST tumors. Exons 9, 11, 13, and 17 of c-KIT and exons 12, 14, and 18 of PDGFRα were evaluated for oncogenic mutations. Protein modeling was used to assess mutations within the juxtamembrane region and the kinase domain of KIT. Results: Mutations in KIT exons 9, 11, and 13 were identified in GISTs with the majority of changes involving the juxtamembrane region of KIT. Molecular modeling indicates that mutations in this region result in disruption of the KIT autoinhibited conformation, and lead to gain-of-function activation of the kinase. Furthermore, a novel germ-line mutation in KIT was identified that is associated with an autosomal dominant predisposition to the development of GIST. Conclusions: We have used protein modeling and structural analyses to elucidate why patients with GIST tumors containing exon 11 mutations are the most responsive to imatinib mesylate treatment. Importantly, mutations detected in this exon and others displayed constitutive activation of KIT. Furthermore, we have found tumors that are both KIT and PDGFRα mutation negative, suggesting that additional, yet unidentified, abnormalities may contribute to GIST tumorigenesis.

Statistical analysis of interface similarity in crystals of homologous proteins

Many proteins function as homo-oligomers and are regulated via their oligomeric state. For some proteins, the stoichiometry of homo-oligomeric states under various conditions has been studied using gel filtration or analytical ultracentrifugation experiments. The interfaces involved in these assemblies may be identified using cross-linking and mass spectrometry, solution-state NMR, and other experiments. However, for most proteins, the actual interfaces that are involved in oligomerization are inferred from X-ray crystallographic structures using assumptions about interface surface areas and physical properties. Examination of interfaces across different Protein Data Bank (PDB) entries in a protein family reveals several important features. First, similarities in space group, asymmetric unit size, and cell dimensions and angles (within 1%) do not guarantee that two crystals are actually the same crystal form, containing similar relative orientations and interactions within the crystal. Conversely, two crystals in different space groups may be quite similar in terms of all the interfaces within each crystal. Second, NMR structures and an existing benchmark of PDB crystallographic entries consisting of 126 dimers as well as larger structures and 132 monomers were used to determine whether the existence or lack of common interfaces across multiple crystal forms can be used to predict whether a protein is an oligomer or not. Monomeric proteins tend to have common interfaces across only a minority of crystal forms, whereas higher-order structures exhibit common interfaces across a majority of available crystal forms. The data can be used to estimate the probability that an interface is biological if two or more crystal forms are available. Finally, the Protein Interfaces, Surfaces, and Assemblies (PISA) database available from the European Bioinformatics Institute is more consistent in identifying interfaces observed in many crystal forms compared with the PDB and the European Bioinformatics Institutes Protein Quaternary Server (PQS). The PDB, in particular, is missing highly likely biological interfaces in its biological unit files for about 10% of PDB entries.

Oligomerization of BAK by p53 Utilizes Conserved Residues of the p53 DNA Binding Domain

Genotoxic stress triggers a rapid translocation of p53 to the mitochondria, contributing to apoptosis in a transcription-independent manner. Using immunopurification protocols and mass spectrometry, we previously identified the proapoptotic protein BAK as a mitochondrial p53-binding protein and showed that recombinant p53 directly binds to BAK and can induce its oligomerization, leading to cytochrome c release. In this work we describe a combination of molecular modeling, electrostatic analysis, and site-directed mutagenesis to define contact residues between BAK and p53. Our data indicate that three regions within the core DNA binding domain of p53 make contact with BAK; these are the conserved H2 α-helix and the L1 and L3 loop. Notably, point mutations in these regions markedly impair the ability of p53 to oligomerize BAK and to induce transcription-independent cell death. We present a model whereby positively charged residues within the H2 helix and L1 loop of p53 interact with an electronegative domain on the N-terminal α-helix of BAK; the latter is known to undergo conformational changes upon BAK activation. We show that mutation of acidic residues in the N-terminal helix impair the ability of BAK to bind to p53. Interestingly, many of the p53 contact residues predicted by our model are also direct DNA contact residues, suggesting that p53 interacts with BAK in a manner analogous to DNA. The combined data point to the H2 helix and L1 and L3 loops of p53 as novel functional domains contributing to transcription-independent apoptosis by this tumor suppressor protein.

A Novel Cas Family Member, HEPL, Regulates FAK and Cell Spreading

For over a decade, p130Cas/BCAR1, HEF1/NEDD9/Cas-L, and Efs/Sin have defined the Cas (Crk-associated substrate) scaffolding protein family. Cas proteins mediate integrin-dependent signals at focal adhesions, regulating cell invasion and survival; at least one family member, HEF1, regulates mitosis. We here report a previously undescribed novel branch of the Cas protein family, designated HEPL (for HEF1-Efs-p130Cas-like). The HEPL branch is evolutionarily conserved through jawed vertebrates, and HEPL is found in some species lacking other members of the Cas family. The human HEPL mRNA and protein are selectively expressed in specific primary tissues and cancer cell lines, and HEPL maintains Cas family function in localization to focal adhesions, as well as regulation of FAK activity, focal adhesion integrity, and cell spreading. It has recently been demonstrated that upregulation of HEF1 expression marks and induces metastasis, whereas high endogenous levels of p130Cas are associated with poor prognosis in breast cancer, emphasizing the clinical relevance of Cas proteins. Better understanding of the complete protein family should help inform prediction of cancer incidence and prognosis.

MollDE: a homology modeling framework you can click with

UNLABELLED Molecular Integrated Development Environment (MolIDE) is an integrated application designed to provide homology modeling tools and protocols under a uniform, user-friendly graphical interface. Its main purpose is to combine the most frequent modeling steps in a semi-automatic, interactive way, guiding the user from the target protein sequence to the final three-dimensional protein structure. The typical basic homology modeling process is composed of building sequence profiles of the target sequence family, secondary structure prediction, sequence alignment with PDB structures, assisted alignment editing, side-chain prediction and loop building. All of these steps are available through a graphical user interface. MolIDEs user-friendly and streamlined interactive modeling protocol allows the user to focus on the important modeling questions, hiding from the user the raw data generation and conversion steps. MolIDE was designed from the ground up as an open-source, cross-platform, extensible framework. This allows developers to integrate additional third-party programs to MolIDE. AVAILABILITY http://dunbrack.fccc.edu/molide/molide.php CONTACT rl_dunbrack@fccc.edu.