Experimental validation of a fragment library for lead discovery using SPR biosensor technology

A new fragment library for lead discovery has been designed and experimentally validated for use in surface plasmon resonance (SPR) biosensor-based screening. The 930 compounds in the library were selected from 4.6 million commercially available compounds using a series of physicochemical and medicinal chemistry filters. They were screened against 3 prototypical drug targets: HIV-1 protease, thrombin and carbonic anhydrase, and a nontarget: human serum albumin. Compound solubility was not a problem under the conditions used for screening. The high sensitivity of the sensor surfaces allowed the detection of interactions for 35% to 97% of the fragments, depending on the target protein. None of the fragments was promiscuous (i.e., interacted with a stoichiometry ≥5:1 with all 4 proteins), and only 2 compounds dissociated slowly from all 4 proteins. The use of several targets proved valuable since several compounds would have been disqualified from the library on the grounds of promiscuity if fewer target proteins had been used. The experimental procedure allowed an efficient evaluation and exploration of the new fragment library and confirmed that the new library is suitable for SPR biosensor-based screening.     

Identification of peptides that bind to the constant region of a humanized IgG1 monoclonal antibody using phage display

The pFc' fragments of a humanized IgG1 monoclonal antibody were generated by digestion with immobilized pepsin. These pFc' fragments were separated from F(ab')2 fragments by affinity chromatography. The pFc' fragments corresponding to the constant region of the humanized IgG1 monoclonal antibody were used as targets for phage display using variable-length peptide libraries. Interacting phage-displayed peptides were selected by repetitious cycles of target screening and phage amplification. Peptide sequences, deduced by sequencing DNA from isolated phage, were aligned and analyzed for amino acid motifs against each other and protein A. These results indicated that an amino acid motif has been identified using phage display technology that is sufficient for pFc' binding. Furthermore, the peptides derived from this study may prove useful in the development of peptidomimetic alternatives to protein A for use in affinity chromatography.     

HHfrag: HMM-based fragment detection using HHpred

MOTIVATION: Over the last decade, both static and dynamic fragment libraries for protein structure prediction have been introduced. The former are built from clusters in either sequence or structure space and aim to extract a universal structural alphabet. The latter are tailored for a particular query protein sequence and aim to provide local structural templates that need to be assembled in order to build the full-length structure. RESULTS: Here, we introduce HHfrag, a dynamic HMM-based fragment search method built on the profile-profile comparison tool HHpred. We show that HHfrag provides advantages over existing fragment assignment methods in that it: (i) improves the precision of the fragments at the expense of a minor loss in sequence coverage; (ii) detects fragments of variable length (6-21 amino acid residues); (iii) allows for gapped fragments and (iv) does not assign fragments to regions where there is no clear sequence conservation. We illustrate the usefulness of fragments detected by HHfrag on targets from most recent CASP.     

Comparative proteomic expression profile in all-trans retinoic acid differentiated neuroblastoma cell line

Neuroblastoma (NB) is an infant tumor which frequently differentiates into neurons. We used two-dimensional differential in-gel electrophoresis (2D-DIGE) to analyze the cytosolic and nuclear protein expression patterns of LAN-5 cells following neuronal differentiating agent all-trans-retinoic acid treatment. We identified several candidate proteins, from which G beta2 and Prefoldin 3 may have a role on NB development. These results strength the use of proteomics to discover new putative protein targets in cancer.     

A novel approach to fold recognition using sequence-derived properties from sets of structurally similar local fragments of proteins

Comparative modeling methods can consistently produce reliable structural models for protein sequences with more than 25% sequence identity to proteins with known structure. However, there is a good chance that also sequences with lower sequence identity have their structural components represented in structural databases. To this end, we present a novel fragment-based method using sets of structurally similar local fragments of proteins. The approach differs from other fragment-based methods that use only single backbone fragments. Instead, we use a library of groups containing sets of sequence fragments with geometrically similar local structures and extract sequence related properties to assign these specific geometrical conformations to target sequences. We test the ability of the approach to recognize correct SCOP folds for 273 sequences from the 49 most popular folds. 49% of these sequences have the correct fold as their top prediction, while 82% have the correct fold in one of the top five predictions. Moreover, the approach shows no performance reduction on a subset of sequence targets with less than 10% sequence identity to any protein used to build the library.     

Construction and application of a yeast surface-displayed human cDNA library to identify post-translational modification-dependent protein-protein interactions

Although post-translational modifications such as phosphorylation mediate fundamental biological processes within the cell, relatively few methods exist that allow proteome-wide identification of proteins that interact with these modifications. We constructed a yeast surface-displayed human cDNA library and utilized it to identify protein fragments with affinity for phosphorylated peptides derived from the major tyrosine autophosphorylation sites of the epidermal growth factor receptor or focal adhesion kinase. We identified cDNAs encoding the Src homology 2 domains from adapter protein APS, phosphoinositide 3-kinase regulatory subunit 3, SH2B, and tensin, demonstrating the effectiveness of this approach. Our results suggest that large libraries of functional human protein fragments can be efficiently displayed on the yeast surface. In addition to the analysis of post-translational modifications, yeast surface-displayed human cDNA libraries have many potential applications, including identifying targets and defining potential cross-reactive proteins for small molecules or drugs.     

Reducing the computational complexity of protein folding via fragment folding and assembly

Understanding, and ultimately predicting, how a 1-D protein chain reaches its native 3-D fold has been one of the most challenging problems during the last few decades. Data increasingly indicate that protein folding is a hierarchical process. Hence, the question arises as to whether we can use the hierarchical concept to reduce the practically intractable computational times. For such a scheme to work, the first step is to cut the protein sequence into fragments that form local minima on the polypeptide chain. The conformations of such fragments in solution are likely to be similar to those when the fragments are embedded in the native fold, although alternate conformations may be favored during the mutual stabilization in the combinatorial assembly process. Two elements are needed for such cutting: (1) a library of (clustered) fragments derived from known protein structures and (2) an assignment algorithm that selects optimal combinations to "cover" the protein sequence. The next two steps in hierarchical folding schemes, not addressed here, are the combinatorial assembly of the fragments and finally, optimization of the obtained conformations. Here, we address the first step in a hierarchical protein-folding scheme. The input is a target protein sequence and a library of fragments created by clustering building blocks that were generated by cutting all protein structures. The output is a set of cutout fragments. We briefly outline a graph theoretic algorithm that automatically assigns building blocks to the target sequence, and we describe a sample of the results we have obtained.     

Structural proteomics: methods in deriving protein structural information and issues in data management

Structural proteomics is an emerging paradigm that is gaining importance in the post-genomic era as a valuable discipline to process the protein target information being deciphered. The field plays a crucial role in assigning function to sequenced proteins, defining pathways in which the targets are involved, and understanding structure-function relationships of the protein targets. A key component of this research sector is accessing the three-dimensional structures of protein targets by both experimental and theoretical methods. This then leads to the question of how to store, retrieve, and manipulate vast amounts of sequence (1-D) and structural (3-D) information in a relational format so that extensive data analysis can be achieved. We at SBI have addressed both of these fundamental requirements of structural proteomics. We have developed an extensive collection of three-dimensional protein structures from sequence data and have implemented a relational architecture for data management. In this article we will discuss our approaches to structural proteomics and the tools that life science researchers can use in their discovery efforts.     

A novel subtractive antibody phage display method to discover disease markers

Today's research demands fast identification of potential diagnostic and therapeutic targets. We describe a novel phage display strategy to identify disease-related proteins that are specifically expressed in a certain (diseased) tissue or cells. Phages displaying antibody fragments are selected on complex protein mixtures in a two-step manner combining subtractive selection in solution with further enrichment of specific phages on two-dimensional Western blots. Targets recognized by the resulting recombinant antibodies are immunoaffinity-purified and identified by mass spectrometry. We used antibody fragment libraries from autoimmune patients to discover apoptosis-specific and disease-related targets. One of the three identified targets is the U1-70K protein, a marker for systemic lupus erythematosus overlap disease. Interestingly the epitope on U1-70K recognized by the selected recombinant antibody was shown to be apoptosis-dependent, and such epitopes are believed to be involved in breaking tolerance to self-antigens. The other two proteins were identified as polypyrimidine tract-binding protein-associated splicing factor (PSF)/nuclear RNA- and DNA-binding protein of 54 kDa (p54nrb) and heterogeneous ribonucleoprotein C.     

Role of Slit proteins in the vertebrate brain.

Diffusible chemorepellents play a major role in guiding developing axons towards their correct targets by preventing them from entering or steering them away from certain regions. Genetic studies in Drosophila revealed a novel repulsive guidance system that prevents inappropriate axons from crossing the CNS midline; this repulsive system is mediated by the Roundabout (Robo) receptors and their secreted ligand Slits. Three distinct slit genes (slit1, slit2 and slit3) and three distinct robo genes (robo1, robo2 and rig-1) have been cloned in mammals. In collagen gel co-cultures, Slit1 and Slit2 can repel and collapse olfactory axons. However, there is also some positive effect associated with Slits, as Slit2 stimulates the formation of axon collateral branches by NGF-responsive neurons of the dorsal root ganglia (DRG). Slit2 is a large ECM glycoproteins of about 200 kD, which is proteolytically processed into 140 kD N-terminal and 55-60 kD C-terminal fragments. Slit2 cleavage fragments appear to have different cell association characteristics, with the smaller C-terminal fragment being more diffusible and the larger N-terminal and uncleaved fragments being more tightly cell associated. This suggested that the different fragments might have different functional activities in vivo. We have begun to explore these questions by engineering mutant and truncated versions of hSlit2 representing the two cleavage fragments, N- and C-, and the uncleavable molecule and examining the activities of these mutants in binding and functional assays. We found that an axon's response to Slit2 is not absolute, but rather is reflective of the context in which the protein is encountered.     

Knowledge-based Fragment Binding Prediction

Target-based drug discovery must assess many drug-like compounds for potential activity. Focusing on low-molecular-weight compounds (fragments) can dramatically reduce the chemical search space. However, approaches for determining protein-fragment interactions have limitations. Experimental assays are time-consuming, expensive, and not always applicable. At the same time, computational approaches using physics-based methods have limited accuracy. With increasing high-resolution structural data for protein-ligand complexes, there is now an opportunity for data-driven approaches to fragment binding prediction. We present FragFEATURE, a machine learning approach to predict small molecule fragments preferred by a target protein structure. We first create a knowledge base of protein structural environments annotated with the small molecule substructures they bind. These substructures have low-molecular weight and serve as a proxy for fragments. FragFEATURE then compares the structural environments within a target protein to those in the knowledge base to retrieve statistically preferred fragments. It merges information across diverse ligands with shared substructures to generate predictions. Our results demonstrate FragFEATURE''s ability to rediscover fragments corresponding to the ligand bound with 74% precision and 82% recall on average. For many protein targets, it identifies high scoring fragments that are substructures of known inhibitors. FragFEATURE thus predicts fragments that can serve as inputs to fragment-based drug design or serve as refinement criteria for creating target-specific compound libraries for experimental or computational screening.     

A new anti-mesothelin antibody targets selectively the membrane-associated form

Mesothelin is a glycosylphosphatidylinositol (GPI)-anchored membrane protein that shows promise as a target for antibody-directed cancer therapy. High levels of soluble forms of the antigen represent a barrier to directing therapy to cellular targets. The ability to develop antibodies that can selectively discriminate between membrane-bound and soluble conformations of a specific protein, and thus target only the membrane-associated antigen, is a substantive issue. We show that use of a tolerance protocol provides a route to such discrimination. Mice were tolerized with soluble mesothelin and a second round of immunizations was performed using mesothelin transfected P815 cells. RNA extracted from splenocytes was used in phage display to obtain mesothelin-specific antigen-binding fragments (Fabs) that were subsequently screened by flow cytometry and ELISA. This approach generated 147 different Fabs in 34 VH-CDR3 families. Utilizing competition assays with soluble protein and mesothelin-containing serum obtained from metastatic cancer patients, 10 of these 34 VH-CDR3 families were found to bind exclusively to the membrane-associated form of mesothelin. Epitope mapping performed for the 1H7 clone showed that it does not recognize GPI anchor. VH-CDR3 sequence analysis of all Fabs showed significant differences between Fabs selective for the membrane-associated form of the antigen and those that recognize both membrane bound and soluble forms. This work demonstrates the potential to generate an antibody specific to the membrane-bound form of mesothelin. 1H7 offers potential for therapeutic application against mesothelin-bearing tumors, which would be largely unaffected by the presence of the soluble antigen.     

Going native: Direct high throughput screening of secreted full-length IgG antibodies against cell membrane proteins

Gel microdroplet – fluorescence activated cell sorting (GMD-FACS) is an innovative high throughput screening platform for recombinant protein libraries, and we show here that GMD-FACS can overcome many of the limitations associated with conventional screening methods for antibody libraries. For example, phage and cell surface display benefit from exceptionally high throughput, but generally require high quality, soluble antigen target and necessitate the use of anchored antibody fragments. In contrast, the GMD-FACS assay can screen for soluble, secreted, full-length IgGs at rates of several thousand clones per second, and the technique enables direct screening against membrane protein targets in their native cellular context. In proof-of-concept experiments, rare anti-EGFR antibody clones were efficiently enriched from a 10,000-fold excess of anti-CCR5 clones in just three days. Looking forward, GMD-FACS has the potential to contribute to antibody discovery and engineering for difficult targets, such as ion channels and G protein-coupled receptors.     

Combining evolutionary and structural information for local protein structure prediction

We study the effects of various factors in representing and combining evolutionary and structural information for local protein structural prediction based on fragment selection. We prepare databases of fragments from a set of non-redundant protein domains. For each fragment, evolutionary information is derived from homologous sequences and represented as estimated effective counts and frequencies of amino acids (evolutionary frequencies) at each position. Position-specific amino acid preferences called structural frequencies are derived from statistical analysis of discrete local structural environments in database structures. Our method for local structure prediction is based on ranking and selecting database fragments that are most similar to a target fragment. Using secondary structure type as a local structural property, we test our method in a number of settings. The major findings are: (1) the COMPASS-type scoring function for fragment similarity comparison gives better prediction accuracy than three other tested scoring functions for profile-profile comparison. We show that the COMPASS-type scoring function can be derived both in the probabilistic framework and in the framework of statistical potentials. (2) Using the evolutionary frequencies of database fragments gives better prediction accuracy than using structural frequencies. (3) Finer definition of local environments, such as including more side-chain solvent accessibility classes and considering the backbone conformations of neighboring residues, gives increasingly better prediction accuracy using structural frequencies. (4) Combining evolutionary and structural frequencies of database fragments, either in a linear fashion or using a pseudocount mixture formula, results in improvement of prediction accuracy. Combination at the log-odds score level is not as effective as combination at the frequency level. This suggests that there might be better ways of combining sequence and structural information than the commonly used linear combination of log-odds scores. Our method of fragment selection and frequency combination gives reasonable results of secondary structure prediction tested on 56 CASP5 targets (average SOV score 0.77), suggesting that it is a valid method for local protein structure prediction. Mixture of predicted structural frequencies and evolutionary frequencies improve the quality of local profile-to-profile alignment by COMPASS.     

Structural basis of GC-1 selectivity for thyroid hormone receptor isoforms

Background

Multiple protein templates are commonly used in manual protein structure prediction. However, few automated algorithms of selecting and combining multiple templates are available.

Results

Here we develop an effective multi-template combination algorithm for protein comparative modeling. The algorithm selects templates according to the similarity significance of the alignments between template and target proteins. It combines the whole template-target alignments whose similarity significance score is close to that of the top template-target alignment within a threshold, whereas it only takes alignment fragments from a less similar template-target alignment that align with a sizable uncovered region of the target. We compare the algorithm with the traditional method of using a single top template on the 45 comparative modeling targets (i.e. easy template-based modeling targets) used in the seventh edition of Critical Assessment of Techniques for Protein Structure Prediction (CASP7). The multi-template combination algorithm improves the GDT-TS scores of predicted models by 6.8% on average. The statistical analysis shows that the improvement is significant (p-value < 10^-4). Compared with the ideal approach that always uses the best template, the multi-template approach yields only slightly better performance. During the CASP7 experiment, the preliminary implementation of the multi-template combination algorithm (FOLDpro) was ranked second among 67 servers in the category of high-accuracy structure prediction in terms of GDT-TS measure.

Conclusion

We have developed a novel multi-template algorithm to improve protein comparative modeling.     

Single domain intracellular antibodies: a minimal fragment for direct in vivo selection of antigen-specific intrabodies

There is a major need in target validation and therapeutic applications for molecules that can interfere with protein function inside cells. Intracellular antibodies (intrabodies) can bind to specific targets in cells but isolation of intrabodies is currently difficult. Intrabodies are normally single chain Fv fragments comprising variable domains of the immunoglobulin heavy (VH) and light chains (VL). We now demonstrate that single VH domains have excellent intracellular properties of solubility, stability and expression within the cells of higher organisms and can exhibit specific antigen recognition in vivo. We have used this intracellular single variable domain (IDab) format, based on a previously characterised intrabody consensus scaffold, to generate diverse intrabody libraries for direct in vivo screening. IDabs were isolated using two distinct antigens and affinities of isolated IDabs ranged between 20 nM and 200 nM. Moreover, IDabs selected for binding to the RAS protein could inhibit RAS-dependent oncogenic transformation of NIH3T3 cells. The IDab format is therefore ideal for in vivo intrabody use. This approach to intrabodies obviates the need for phage antibody libraries, avoids the requirement for production of antigen in vitro and allows for direct selection of intrabodies in vivo.     

A novel approach to local similarity of protein binding sites substantially improves computational drug design results

We present a novel notion of binding site local similarity based on the analysis of complete protein environments of ligand fragments. Comparison of a query protein binding site (target) against the 3D structure of another protein (analog) in complex with a ligand enables ligand fragments from the analog complex to be transferred to positions in the target site, so that the complete protein environments of the fragment and its image are similar. The revealed environments are similarity regions and the fragments transferred to the target site are considered as binding patterns. The set of such binding patterns derived from a database of analog complexes forms a cloud-like structure (fragment cloud), which is a powerful tool for computational drug design. It has been shown on independent test sets that the combined use of a traditional energy-based score together with the cloud-based score responsible for the quality of embedding of a ligand into the fragment cloud improves the self-docking and screening results dramatically. The usage of a fragment cloud as a source of positioned molecular fragments fitting the binding protein environment has been validated by reproduction of experimental ligand optimization results.