The genesis of high-throughput structure-based drug discovery using protein crystallography

A microfluidic approach for screening submicroliter volumes against multiple reagents by using preformed arrays of nanoliter plugs in a three-phase liquid/liquid/gas flow

Plugging a gap in screening

Arrays of nanoliter-sized plugs of different compositions can be preformed in a three-phase liquid/liquid/gas flow. The arrays can be transported into a microfluidic channel to test against a target (see schematic representation), as demonstrated in protein crystallization and an enzymatic assay.

Preparative parallel protein purification (P4)

In state of the art drug discovery, it is essential to gain structural information of pharmacologically relevant proteins. Increasing the output of novel protein structures requires improved preparative methods for high throughput (HT) protein purification. Currently, most HT platforms are limited to small-scale and available technology for increasing throughput at larger scales is scarce. We have adapted a 10-channel parallel flash chromatography system for protein purification applications. The system enables us to perform 10 different purifications in parallel with individual gradients and UV monitoring. Typical protein purification applications were set up. Methods for ion exchange chromatography were developed for different sample proteins and columns. Affinity chromatography was optimized for His-tagged proteins using metal chelating media and buffer exchange by gel filtration was also tested. The results from the present system were comparable, with respect to resolution and reproducibility, with those from control experiments on an AKTA purifier system. Finally, lysates from 10 E. coli cultures expressing different His-tagged proteins were subjected to a three-step parallel purification procedure, combining the above-mentioned procedures. Nine proteins were successfully purified whereas one failed probably due to lack of expression.

An efficient strategy for high-throughput expression screening of recombinant integral membrane proteins

The recombinant expression of integral membrane proteins is considered a major challenge, and together with the crystallization step, the major hurdle toward routine structure determination of membrane proteins. Basic methodologies for high-throughput (HTP) expression optimization of soluble proteins have recently emerged, providing statistically significant success rates for producing such proteins. Experimental procedures for handling integral membrane proteins are generally more challenging, and there have been no previous comprehensive reports of HTP technology for membrane protein production. Here, we present a generic and integrated parallel HTP strategy for cloning and expression screening of membrane proteins in their detergent solubilized form. Based on this strategy, we provide overall success rates for membrane protein production in Escherichia coli, as well as initial benchmarking statistics of parameters such as expression vectors, strains, and solubilizing detergents. The technologies were applied to 49 E. coli integral membrane proteins with human homologs and revealed that 71% of these proteins could be produced at sufficient levels to allow milligram amounts of protein to be relatively easily purified, which is a significantly higher success rate than anticipated. We attribute the high success rate to the quality and robustness of the methodology used, and to introducing multiple parameters such as different vectors, strains, and detergents. The presented strategy demonstrates the usefulness of HTP technologies for membrane protein production, and the feasibility of large-scale programs for elucidation of structure and function of bacterial integral membrane proteins.

Turning protein crystallisation from an art into a science

Protein crystallisation has gained a new strategic and commercial relevance in the post-genomic era because of its pivotal role in structural genomics. Producing high-quality crystals has always been a bottleneck to structure determination and, with the advent of proteomics, this problem is becoming increasingly acute. The task of producing suitable crystals may be tackled using two approaches. The first relies on empirical techniques that are based mainly on trial and error, and what is perceived to be the 'art' of crystallisation. The second approach is aimed at gaining an understanding of the fundamental principles that govern crystallisation; this knowledge may be applied to design experimental methodology for producing high-quality crystals of medical and industrial interest.

Robotic nanolitre protein crystallisation at the MRC Laboratory of Molecular Biology

We have set up high-throughput robotic systems to screen and optimise crystallisation conditions of biological macromolecules with the aim to make difficult structural biology projects easier. The initial screening involves two robots. A Tecan Genesis liquid handler is used to transfer commercially available crystallisation reagents from 15 ml test tubes into the reservoirs of 96-well crystallisation plates. This step is fully automated and includes a carousel for intermediate plate storage, a Beckman plate sealer and a robotic arm, which transfers plates in between steps. For adding the sample, we use a second robot, a 17-tip Cartesian Technologies PixSys 4200 SynQuad liquid handler, which uses a syringe/solenoid valve combination to dispense small quantities of liquid (typically 100 nl) without touching the surface of the plate. Sixteen of the tips are used to transfer the reservoir solution to the crystallisation wells, while the 17th tip is used to dispense the protein. The screening of our standard set of 1440 conditions takes about 3 h and requires 300 microl of protein solution. Once crystallisation conditions have been found, they are optimised using a second Tecan Genesis liquid handler, which is programmed to pipette gradients from four different corner solutions into a wide range of crystallisation plate formats. For 96-well plates, the Cartesian robot can be used to add the sample. The methods described are now used almost exclusively for obtaining diffraction quality crystals in our laboratory with a throughput of several thousand plates per year. Our set-up has been copied in many institutions worldwide.

Protein crystallization for genomics: throughput versus output

Structural genomics involves many steps in order to reach from Gene to structure. This article focuses on the crystallization step in this chain of tasks. It is becoming increasingly evident that the current high throughput procedures for crystallising proteins do not always produce the expected output of high quality crystals required for structure determination by x-ray crystallography. This problem is discussed and suggestions for raising the output are presented.

Robotic grid loading system for a transmission electron microscope

The design, construction, and testing of a robotic grid loading system for a transmission electron microscope is presented. The system, when integrated with automated data collection software, has the potential to carry out large scale multi-grid experiments, as required, for example, by 2D protein crystallization screening trials. We present a detailed description of the system that utilizes a 6 axis articulate robotic arm to load microscope grids into a specimen holder and then load the holder into the microscope. Grids are stored in trays with a capacity of 96 that are based on the format of a standard 96-well microtiter plate. The system has been integrated with an FEI Tecnai microscope and does not require any modification to the microscope and only minor modifications to the specimen holder. Preliminary results regarding cycle time, failure rates, positioning accuracy and a 96 grid screening experiment are presented.

Combinatorial approaches to protein stability and structure

Why do proteins adopt the conformations that they do, and what determines their stabilities? While we have come to some understanding of the forces that underlie protein architecture, a precise, predictive, physicochemical explanation is still elusive. Two obstacles to addressing these questions are the unfathomable vastness of protein sequence space, and the difficulty in making direct physical measurements on large numbers of protein variants. Here, we review combinatorial methods that have been applied to problems in protein biophysics over the last 15 years. The effects of hydrophobic core composition, the most important determinant of structure and stability, are still poorly understood. Particular attention is given to core composition as addressed by library methods. Increasingly useful screens and selections, in combination with modern high-throughput approaches borrowed from genomics and proteomics efforts, are making the empirical, statistical correlation between sequence and structure a tractable problem for the coming years.

Nucleation of protein crystals

This paper introduces nucleation theory applied to crystallizing protein solutions. It is shown that the classical approach explains the available nucleation data under most conditions used for growing protein crystals for structural studies and for industrial crystallization. However, it fails to explain most experimental data on the structure of the critical clusters. It is also shown that for open systems working out of equilibrium, such as hanging-drop and counterdiffusion techniques, the geometry of the Ostwald-Myers protein solubility diagram and the number, size, and quality of the forming crystals depend not only on supersaturation but also on the rate of development of supersaturation.

Ace revisited: a new target for structure-based drug design

Angiotensin-converting enzyme (ACE) is a chloride-dependent metalloenzyme that catalyses the hydrolytic cleavage of dipeptides from the carboxyl terminus of many regulatory oligopeptides. ACE is central to the renin–angiotensin system that regulates blood pressure, fluid homeostasis, and renal and vascular function. It is therefore a major target for cardiovascular therapies.

ACE inhibitors (for example, captopril, enalaprilat and lisinopril) have been on the market for more than 20 years. Side effects of treatment with ACE inhibitors include cough and angioedema.

ACE comprises an N- and a C-domain, each containing an active site with distinct substrates and activation properties. The design of domain-selective inhibitors might produce new drugs with improved safety and efficacy — this endeavour will be facilitated by the recent determination of the three-dimensional structure of ACE.

The C-domain seems to be primarily responsible for the regulation of blood pressure. Data indicate that C-domain-selective inhibitors will have less severe side effects than current-generation inhibitors, which generally target both the N- and C-domains.

In contrast to the C-domain, the N-domain seems to have relatively low affinity for the peptides that control blood pressure. It preferentially hydrolyses at least three other physiologically important peptides, so targeted inhibition of the N-domain might have novel therapeutic applications.

Current-generation angiotensin-converting enzyme (ACE) inhibitors are widely used for cardiovascular diseases, including high blood pressure, heart failure, heart attack and kidney failure, and have combined annual sales in excess of US $6 billion. However, the use of these ACE inhibitors, which were developed in the late 1970s and early 1980s, is hampered by common side effects. Moreover, we now know that ACE actually consists of two parts (called the N- and C-domains) that have different functions. Therefore, the design of specific domain-selective ACE inhibitors is expected to produce next-generation drugs that might be safer and more effective. Here we discuss the structural features of current inhibitors and outline how next-generation ACE inhibitors could be designed by using the three-dimensional molecular structure of human testis ACE. The ACE structure provides a unique opportunity for rational drug design, based on a combination of in silico modelling using existing inhibitors as scaffolds and iterative lead optimization to drive the synthetic chemistry.

Scaffold Hopping in De Novo Design. Ligand Generation in the Absence of Receptor Information

We report here the de novo generation of chemotypes and scaffolds for the estrogen receptor, without use of the receptor structure in the assembly phase. Through use of ligand superpositions or a single bound conformation of a known active, a pseudoreceptor can be generated as a design envelope, within which novel structures are readily assembled. Many of these structures have high similarity to known chemotypes. Scaffold hopping is readily achieved within this pseudoreceptor, indicating the advantages of such an approach in discovery research.

High-throughput crystallography to enhance drug discovery

Protein crystallography has traditionally been regarded as a resource-intensive, time-consuming technique that, with some notable exceptions, has not made a significant impact on drug discovery. However, inspired by successes in the genome-sequencing initiatives, recent years have seen major changes in X-ray crystallography methodologies and the concept of high-throughput crystallography has emerged. Advances have been made in all phases of the process, including improved molecular biology, protein expression, crystallization and structure determination. This transformation has allowed X-ray crystallography to impact more broadly in the drug-discovery process, extending its utility from structure-based lead optimisation to novel fragment-based lead generation approaches.

A fully integrated protein crystallization platform for small-molecule drug discovery

Structure-based drug discovery in the pharmaceutical industry benefits from cost-efficient methodologies that quickly assess the feasibility of specific, often refractory, protein targets to form well-diffracting crystals. By tightly coupling construct and purification diversity with nanovolume crystallization, the Structural Biology Group at Syrrx has developed such a platform to support its small-molecule drug-discovery program. During the past 18 months of operation at Syrrx, the Structural Biology Group has executed several million crystallization and imaging trials on over 400 unique drug-discovery targets. Here, key components of the platform, as well as an analysis of some experimental results that allowed for platform optimization, will be described.

A deliberate approach to screening for initial crystallization conditions of biological macromolecules

A method to rationally predict crystallization conditions for a previously uncrystallized macromolecule has not yet been developed. One way around this problem is to determine initial crystallization conditions by casting a wide net, surveying a large number of chemical and physical conditions to locate crystallization leads. A facility that executes the rapid survey of crystallization lead conditions is described in detail. Results and guidelines for the initial screening of crystallization conditions, applicable to both manual and robotic setups, are discussed.

Efficient protein crystallization

High-throughput molecular biology and crystallography advances have placed an increasing demand on crystallization, the one remaining bottleneck in macromolecular crystallography. This paper describes three experimental approaches, an incomplete factorial crystallization screen, a high-throughput nanoliter crystallization system, and the use of a neural net to predict crystallization conditions via a small sample (approximately 0.1%) of screening results. The use of these technologies has the potential to reduce time and sample requirements. Initial experimental results indicate that the incomplete factorial design detects initial crystallization conditions not previously discovered using commercial screens. This may be due to the ability of the incomplete factorial screen to sample a broader portion of "crystallization space," using a multidimensional set of components, concentrations, and physical conditions. The incomplete factorial screen is complemented by a neural network program used to model crystallization. This capability is used to help predict new crystallization conditions. An automated, nanoliter crystallization system, with a throughput of up to 400 conditions/h in 40-nl droplets (total volume), accommodates microbatch or traditional "sitting-drop" vapor diffusion experiments. The goal of this research is to develop a fully-automated high-throughput crystallization system that integrates incomplete factorial screen and neural net capabilities.