SPR-based Fragment Screening: Advantages and...

1630 Current Topics in Medicinal Chemistry, 2007, 7, 1630-1642

1568-0266/07 $50.00+.00 © 2007 Bentham Science Publishers Ltd.

SPR-based Fragment Screening: Advantages and Applications

T. Neumann, H-D. Junker, K. Schmidt and R. Sekul*

Graffinity Pharmaceuticals, Im Neuenheimer Feld 518-519, 69120 Heidelberg, Germany

Abstract: Fragment-based screening has recently evolved into a promising strategy in drug discovery, and a range of

biophysical methods can be employed for fragment library screening. Relevant approaches, such as X-ray, NMR and

tethering are briefly introduced focussing on their suitability for fragment-based drug discovery. In particular the

application of surface plasmon resonance (SPR) techniques to the primary screening of large libraries comprising small

molecules is discussed in detail. SPR is known to be a powerful tool for studying biomolecular interactions in a sensitive

and label-free detection format. Advantages of SPR methods over more traditional assay formats are discussed and the

application of available channel and array based SPR systems to biosensing are reviewed. Today, SPR protocols have

been applied to secondary screening of compound libraries and hit conformation, but primary screening of large fragment

libraries for drug discovery is often hampered by the throughput of available systems. Chemical microarrays, in

combination with SPR imaging, can simultaneously generate affinity data for protein targets with up to 9,216 immobilized

fragments per array. This approach has proven to be suitable for screening fragment libraries of up to 110,000 compounds

in a high throughput fashion. The design of fragment libraries and appropriate immobilization chemistries are discussed,

as well as suitable follow-up strategies for fragment hit optimization. Finally, described case studies demonstrate the

successful identification of selective low molecular weight inhibitors for pharmacologically relevant drug targets through

the SPR screening of fragment libraries.

Keywords: Fragments, screening, surface plasmon imaging.

INTRODUCTION

Over the last decade fragment-based screening has evolved to be a promising strategy in drug discovery. Fragments are defined as molecules with a molecular weight between 100 and 300 Daltons (Da) and a limited number of functionalities in comparison to larger molecules [1]. They are considered as substructures of drug-like substances that interact with individual protein epitopes or subsites. Nume-rous case studies have been published demonstrating the successful applicability of the fragment discovery concept for the identification of new lead structures in medicinal chemistry [2].

An efficient sampling of fragment chemical space in comparison to lead-like libraries is feasible due to the better relative coverage of imaginable organic structures in this molecular weight range. Reymond et al. have estimated the number of synthetically addressable molecules below 160 Da to be 13.9 million different structures in comparison to a total number of organic molecules of 10

18-10

200 [3]. Also

low-complexity compounds are more likely to match a given binding site, resulting in a larger hit probability [4]. Despite weaker protein-ligand interactions, fragments bind with superior ligand efficiencies – free binding energy G per heavy atom - to their protein targets [5]. These properties increase the chance of identifying new chemotypes usually missed with conventional technologies. Furthermore, small and simple molecules open up more opportunities for further optimization.

*Address correspondence to this author at the Graffinity Pharmaceuticals, Im Neuenheimer Feld 518-519, 69120 Heidelberg, Germany; E-mail:

[email protected]

Fragments typically possess affinities from the micro-molar to millimolar range. Sensitive detection methods are required to identify such weak protein-ligand interactions. For this, several biophysical techniques are available. Most commonly used are: nuclear magnetic resonance (NMR), X-ray crystallography, mass spectrometry, and surface plasmon resonance (SPR).

SPR screening of chemical microarrays is ideally suited for fragment-based discovery, because it combines sensi-tivity with high throughput and low material consumption.

This article describes the key features and principal strategies of SPR-based fragment screening.

FRAGMENT LIBRARIES AND FOLLOW-UP STRA-

TEGIES

Fragment Library Design

Fragment based library design adopts strategies estab-lished for the synthesis of lead-like compound collections. To reduce the complexity of the molecules, compound selection is performed according to the ‘rule-of-three’, which defines fragment properties with molecular weights of <300 Da, a cLogP of <3, the number of H-bond donors <3, the number of H-bond acceptors <3, and the number of rotatable bonds <3 [1,6]. Besides using such biophysical property rules it is also advisable to apply medicinal chemistry filters to exclude undesired functional groups which are toxic or reactive [7]. To increase the hit-identification likelihood the application of diversity selection criteria and the inclusion of biological information into the library design is useful [3,8,9]. Size and properties of fragment libraries are largely influenced by the biophysical method used for the detection of interactions between fragments and their biochemical

SPR-based Fragment Screening Current Topics in Medicinal Chemistry, 2007, Vol. 7, No. 16 1631

targets. X-ray and NMR screening are adaptive for library sizes in the range of 10 to 10 entities by using compound mixtures. In this case library members must be specially designed to allow deconvolution and may sometimes contain functionalities which simplify their detection with the applied technology [10,11]. Approaches like SPR imaging allow the simultaneous detection of up to 9,216 target-ligand interactions on a single array within a few hours and are therefore suited to libraries sizes of up to 10

5.

It is also evident that the identified weak binders must be evaluated further to explore possible structure-activity relationships. The discovery of low affinity fragments by biophysical methods alone will not convince medicinal chemistry teams to work on them if no follow-up strategy is apparent. Therefore, potential optimization strategies should also be considered during the library design.

Follow-up Strategies

There are several distinct strategies for the conversion of fragments into leads. Most frequently, initial hits are opti-mized by fragment expansion or evolution. In this approach the fragment is modified by the addition of substituents to identify favorable interactions with the protein active site. Structure-activity relationships are often established by a combination of structure based design and parallel che-mistry which subsequently enables an efficient compound optimization [12]. The potency of a newly discovered fragment can also be improved by its combination with known motifs. In many cases, this strategy leads to a boost in activity and simultaneously generates novelty [13]. A more demanding concept is the linking of two or more fragments which address adjacent subpockets. A major challenge of this strategy is to identify a suitable linker and to assemble the molecules in a proper orientation. Structural information can support the identification of the optimal fragment connection but there are technology driven approaches such as tethering, as well [14,15]. Fragment assembly is particularly suited for protein targets with extended active sites, e.g. proteases or protein-protein interactions [14,15]. Dynamic combinatorial chemistry or click chemistry are variants of the fragment linking approach. Ligands with intrinsic reactive functionalities are assembled in situ in the target binding site, which serves as a template. This concept is limited to the number of applicable chemistries, which must be compatible with the protein environment [16,17].

An elegant solution to simplify subsequent optimization is the incorporation of chemical handles, like amides or amines, as structural features into the screened fragments. These functionalities can be used for parallel chemistry and rapid optimization. Key to the success of this approach is that the newly synthesized derivatives maintain relevant interaction patterns appearing in the initial hit [18].

Fragment and Lead-Like Libraries for Chemical

Microarrays

Fragment libraries used for SPR imaging offer additional features in comparison to libraries designed for other methods. Compounds are covalently immobilized to the array surface via a flexible, hydrophilic spacer molecule (ChemTag). The preparation of tagged compound libraries is

independent of the array production and includes several quality control steps [19]. Once attained, the synthesis scheme for all substances is known and facilitates efficient follow-up optimization by replacing the spacer moiety with a set of diverse substituents. The miniaturization and paralle-lisation of low affinity microarray screening minimizes protein consumption and is faster in relation to sequential approaches.

Therefore, larger screening sets and more diverse number of scaffolds can be presented on the array surface. In addition, one core motif can be immobilized in several orientations. The linker attachment site might limit the number of possible binding modes, but delivers additional information about positions suitable for further exploration. Additional information for follow-up optimization can be obtained by coordinating the design of one-point diversity fragment and two-point diversity lead-like libraries. A typical example for the combination of fragment and lead-like array compounds on chemical microarrays are libraries based on benzimidazole derivatives. Due to the established chemistry for the construction of this scaffold, the attach-ment position of the spacer molecule and the relative orientation of the substituents to each other, can be easily varied [20,21,22]. The resulting libraries can be divided into linear fragment-like subsets, which combine scaffold and substituents at one position of the core, and two-point diversity subsets (Fig. (1)). The linearity of the fragment libraries allows the use of many similar substituents for a fine sampling of the chemical space in this direction. For the lead-like libraries, a representative number of substitution pattern combinations can be synthesized to understand their possible cooperativeness (Fig. (2)). This methodology pro-motes assembly strategies by integrating the precise frag-ment substitution patterns into binary follow-up compounds.

BIOPHYSICAL TECHNIQUES

Various biophysical methods have been shown to be suitable for detecting weak protein-ligand interactions. The most relevant technologies applied for fragment-based screening are described briefly in the following section.

Crystallography

X-ray crystallography can deliver detailed structural information on protein-ligand interactions. Consequently, it has found wide application in pharmaceutical research, guid-ing the lead optimization process [23]. Technical improve-ments have allowed this approach to become a more widely used tool for the primary screening of fragment libraries.

Crystallography requires 10-50 mg of target protein with a purity of >95 % [24]. To detect weak interactions, the test compounds need to be soluble at millimolar concentrations (at least ten times of the IC50 or KD) to occupy the binding site. The current work-flow starts with the evaluation of the crystallization conditions, followed by an optimization step to obtain diffraction quality crystals. The screening process is conducted by: soaking mixtures of shape-diverse compounds (n = 2-10) into the crystal, collecting X-ray data, and determining structures [25,26].

1632 Current Topics in Medicinal Chemistry, 2007, Vol. 7, No. 16 Sekul et al.

Not all targets are amenable for X-ray analysis, because a significant part of the crystal species cannot be optimized to diffract at a useful resolution [27,23].

However, numerous publications illustrate the successful application of crystallography-based fragment screening for targets of different protein families, e.g. kinases, proteases and phosphodiesterases [12,28,24].

Tethering

A quite unique strategy, named “tethering”, has been developed which relies on the formation of a disulfide bridge between a ligand and a cysteine residue of the protein. The molecules carry thiol groups, which are conjugated to a

small, hydrophilic thiol linker, as disulfides. The screening of fragment cocktails (n = 10) is performed under partially reducing conditions to enable the formation of new disulfide bonds via disulfide exchange. Compounds binding to sites adjacent to the cysteine residue are enriched in the protein-ligand complex mixture and can be identified by mass spectrometry. The technology is highly sensitive, because tethering converts weak interacting fragments to more potent molecules. Furthermore, the covalent attachment of the ligands facilitates subsequent X-ray analysis. The method requires a prior knowledge of the binding site, because the cysteine has to be located in or close to the active site. The demand to generate cysteine containing mutant proteins could have potential limitations for the application.

Fig. (1). Interplay of linear fragment libraries and lead-like libraries with a benzimidazole core. The axes represent fragments with one point

of diversity (X-axis: Diversity at positon 2, Y-axis: Diversity at position 1, Z-axis: Diversity at position 5). The planes represent lead-like

compounds wit two points of diversity (e.g. X,Y plane, diversity at position 2 and 1 synthesized in a combinatorial fashion).

Fig. (2). (a) A small set of structurally diverse substituents is used for the two-point combinatorial synthesis resulting in compounds

represented in the planes of Fig. 1 (b) One-point diversity libraries (axis of figure 1) allow the synthesis of a greater number of related

compounds due to the lack of the multiplicative combinatorial effect.

R

N

N

TAG

N

N

R

N

O

TAG

R2

N

N

TAG

R

R

N

N

R

N

O

TAG

N

O

RN

N

TAG

X

Y

Z

two-point diversity library

linear fragment library

1

2

5

O O O O O

R =

R =

A B

R =

R =

R =

R =


Tethering was used to screen a variety of proteins like caspases and difficult targets like protein tyrosine phos-phatase 1B and interleukin 2, yielding fragments for a subse-quent optimization [29,30]

NMR

NMR-based fragment screening was first introduced in 1996 by Fesik and co-workers at Abbott laboratories [31]. Since then, two principal strategies have emerged, one measuring the protein, the other the ligand chemical shifts.

Target-based NMR screening monitors the changes in 1H,13C or 1H15 N correlation signals for a labeled protein, when binding to a test compound. The method can detect nM to mM interactions and provides structural information of the binding site. However, it normally requires large quantities of isotopically labeled protein, between 50 – 200 mg are reported, with protein solubilities between 0.1 and 1 mM [32]. The technique is restricted to small targets, typically <50 kDa, for best performance. Target-based NMR is well suited for determining the structure of fragment binding sites. Excellent results have been achieved with challenging targets such as protein-protein interaction domains [14,33].

Ligand-based NMR screening relies on changes in the ligand signals when binding to the target. In contrast to target-based NMR, the protein concentration is in the low

M range, therefore the protein consumption is substantially lower. Furthermore, no isotopic label is needed. Fragments are typically screened in mixtures followed by a decon-volution step. Ligand affinities from approximately 100 nM to 10 mM can be detected.

The method is commonly used to identify binding fragments in order to perform KD determinations and to conduct competition NMR experiments to search for second site binders. Detailed descriptions of various ligand-based NMR techniques and applications are beyond the scope of this publication and are given by Lepre et al. [11] and Pellecchia et al. [34].

SPR IMAGING TECHNOLOGY FOR FRAGMENT

SCREENING

Surface Plasmon Resonance (SPR) is known to be a versatile and powerful tool for the study of biomolecular interactions and can generate affinity data, as well as kinetic information, in a highly sensitive and label-free detection format. Comprehensive reviews of SPR biosensing can be commonly found in the literature. The following paragraphs will focus on the application of SPR imaging technology to high throughput screening of large collections of small molecules for fragment based discovery. After a short intro-duction to the SPR effect and its advantages over more traditional assay formats, requirements for SPR drug screen-ing are discussed. Finally, currently available SPR systems are compared in the context of their suitability for primary fragment screening.

Surface Plasmon Resonance (SPR)

Surface plasmons (SPs) are collective oscillations of electrons at a metal surface and can be seen as electro-magnetic waves confined to the interfacial plane between a

metal and a dielectric. They can be resonantly excited by coupling polarized light to the surface modes of the metal surface (e.g. gold or silver) in a variety of optical setups such as grating or prism coupling. In the most commonly applied Kretschmann configuration (prism coupling), incident light is reflected at the base plate of a prism covered with a thin metal layer. At incidence angles larger than the critical angle (total reflection conditions), incoming light waves and surface bound plasmon waves can be tuned to resonate by varying either the light wavelength or incidence angle while keeping the other parameter constant. At resonance condi-tions, photon energy is transferred to the surface plasmon mode and hence the reflected light exhibits a sharp attenuation (SPR minimum). The angle and wavelength position of this reflectance minimum is, among other parameters, strongly dependent on the refractive index of the medium adjacent to the metal surface. The latter is sensed by the surface plasmon field exponentially decaying into the dielectric medium resulting in a sensing depth of a couple of hundred nanometers. Processes which alter the local refractive index such as the adsorption of (bio)molecules onto the sensor layer can therefore be monitored in a surface sensitive fashion by recording the shift of the resonance minimum. Generally, both angle-scanning and wavelength scanning systems can be found in the literature, but an overall gain in sensitivity for either method was not clearly demonstrated.

From the shift in the resonance position, thermodynamic equilibrium binding data or kinetic constants (Kd, kon, koff) can be deduced. For kinetic measurements, time dependent changes in the local refractive index on the sensor surface can be monitored by applying different measurement modes: some instruments allow to record the entire resonance curve around the SPR minimum periodically during the time course of the experiment. The time dependent position of the SPR minimum is then followed on-line by software fitting routines. Alternatively, a single detection wavelength or incidence angle in the near-linear range close to the SPR minimum can be chosen for monitoring changes in the reflectivity curve with time.

A typical SPR binding experiment consists of two phases: the binding of the soluble analyte to the sensor and the dissociation of the analyte upon rinsing with analyte free solution. By fitting kinetics from the the association and dissociation phase to appropriate binding models (e.g. Langmuir) the corresponding kinetic rate constants kon and koff (and hence Kd= koff/kon) can be calculated. Another approach is the direct determination of Kd by analyzing equilibrium binding data generated at different analyte concentrations.

Application to Biosensing and Small Molecule-Protein

Interactions

While first used for the analysis of optical properties and the thicknesses of thin (organic) films on metal substrates [35], the potential of SPR for the label-free detection of biologically relevant analytes, using modified sensor sur-faces, was soon recognized. Biosensing applications can be found in numerous publications [36,37,38,39,40]. Numerous studies have been published in which SPR biosensors have been used, for example in studying: DNA-DNA [41,42,43]


DNA-Protein, DNA-PNA, protein-peptide, protein-protein, and antibody-antigen interactions [44,45] as well as enzymatic reactions [46]. Due to the high sensitivity of modern SPR sensors, binding of low molecular weight analytes can be detected and exploited for the study of small molecule interactions with protein targets. For example, Nordin reports on the screening of 22 kinase inhibitors against various kinases using Biacore equipment [47]. Rich used instrumentation from the same manufacturer to successfully study the interaction of a set of 10 compounds with molecular weights between 130 to 800 Da with human serum albumin (HSA) [48].

Advantages/Disadvantages of SPR Biosensing

Labeling of analytes, as required for more traditional assays, is often labor and cost intensive and can interfere with their binding to the target of interest. For SPR readout, the sole requirement is an intrinsic refractive index of the analyte, which is different from that of the buffer surroun-ding. The latter is the case for commonly studied molecules such as proteins, DNA, antibodies etc., which has allowed SPR biosensing to become a popular tool for the label-free detection of biomolecules.

Due to the small surface plasmon penetration depth of the plasmon field into the aqueous media, and the resulting surface sensitivity, bulk effects arising from the analyte solution are almost negligible. Therefore, washing steps are not required before the signal can be read. Measurements can be performed in real time and kinetics, as well as thermo-dynamic data, can be collected. One often discussed issue of SPR based biosensors is the necessity to immobilize one of the binding partners and that tethering of molecules to surfaces may affect the binding constants measured. How-ever, it has been shown that equilibrium, thermodynamic, and kinetic data from surface experiments can mirror those obtained in solution if SPR biosensor experiments are carefully designed [49]. In the case of screening small molecules, tethering of compounds via different attachment points to a sensor surface can be seen as an advantage. In a binding experiment against a soluble protein target, the position of the linker can provide direct information about the orientation of the immobilized compound in the protein binding pocket.

Requirements for High Throughput SPR Drug Screening

For the particular case of using SPR methods for frag-ment based screening against protein targets, the following aspects have to be taken into account.

Sensitivity

Screening of fragment-protein interactions requires technologies capable of detecting and characterizing weak affinities. Among complementary techniques, such as NMR, X-ray, and high concentration biochemical assays, modern SPR instruments exhibit sensitivities that have been shown to be suitable for the detection of low molecular weight analytes and low affinity interactions. Numerous examples concerning the analysis of small molecule inhibitors binding to immobilized protein targets can be found in the literature.

Surface Immobilization Chemistry and Array Preparation

For the study of biomolecular recognition reactions, typically one binding partner of interest is immobilized on the sensor surface using a variety of surface chemistries.

In the design of suitable immobilization strategies, attention has to be paid to: the stability of the surface architecture during the course of the experiment, the precise control over the amount of immobilized binding partner, and the minimization of background signal arising from unspecific binding of the analyte to the surface.

Numerous immobilization schemes for the formation of SPR sensor surfaces and their application can be found in the literature. These include, for example, the use of mediating sensor layers, such as streptavidin layers [50] on gold, and the subsequent attachment of biotinylated molecules, such as antibodies [51] or DNA [52,53]. Other groups have demons-trated the direct immobilization of, for example, DNA sequences on gold via thiol moieties [54]. Commercially available sensor surfaces for use with SPR detection include: Protein A surfaces, carboxymethylated dextran surfaces for covalent coupling to a variety of analytes exhibiting active groups, lipid bilayers [55] for the immobilization of, for example, membrane proteins, and nitrilotriacetic acid (NTA) coupled surfaces to immobilize His-tagged proteins and others [56].

One of the most successful immobilization strategies for biomolecules on sensor surfaces is the formation of self-assembled monolayers (SAMs) on gold or glass surfaces [57,58] and the subsequent covalent attachment of ligands of interest, using appropriate linker chemistry. The highly ordered monolayers allow precise control over the density of immobilized ligands and can be designed to minimize unspecific protein adsorption.

High Throughput SPR Instrumentation

In recent years a wide range of instruments were deve-loped by different companies and label-free screening technology has become a widespread tool for biochemical and pharmaceutical research in academia and industry.

In order to enhance the throughput of single channel systems, approaches were undertaken to increase the number of sensing channels using various excitation schemes such as wavelength or angle scanning [59,60].

Channel based SPR technology has been widely applied to secondary drug screening and hit confirmation for a limited set of hit compounds identified by complimentary HTS technologies such as in-silico screening [61]. Biacore´s A100 system offers five sensor spots each with four chan-nels, and allows the characterization of numerous inter-actions per day when multiple experiments are performed sequentially. However, the number of available channels per system limits the throughput and hampers primary drug screening of large substance collections.

In parallel with the development of channel SPR sensors, studies by Knoll and others [62,63,64] in the 1980s demons-trated that laterally structured coatings on metal films could be analyzed by an imaging technology known as Surface Plasmon Microscopy (SPM). Typically, the technology


utilizes an expanded beam of parallel light to illuminate the entire sensor area covering several square centimetres. The reflected light is then captured by means of a CCD camera. While scanning over a range of wavelengths or incidence angles that cover the SPR resonance conditions, reflection images are recorded stepwise. By subsequent grey-scale analysis of the obtained pictures, SPR resonance curves of individual sensor fields plotted in array format can be obtained [54]. SPM in combination with array type sensors has since been successfully applied to the characterization of almost all kinds of biomolecular interactions [65] such as DNA hybridization [53,66,67,68,69] protein-DNA [65,70], and protein-peptide or antigen-antibodies [71,72] inter-actions.

Researchers at HTS Biosystems have developed an array based SPR imaging system based on grating coupling [73] which allows the interaction of one analyte with up to 400 targets immobilized onto gold chips to be analyzed. The Flexchip system recently acquired by Biacore has since been used for screening and affinity ranking of antibody-arrays [74,75]. Kinetic constants derived from the array approach were found to be similar to those determined by channel based SPR systems [75]. The typical working range for the instrument requires analytes to have molecular weights >5kD, and currently the system has not proved suitable for screening low molecular weight compound libraries.

Lumera´s ProteomicProcessor, also based on an SPR imaging approach, is capable of analyzing binding affinity and kinetics data of up to 10,000 spots per 1.4 cm

2 micro-

array [76] and has been applied to the study of protein-protein, antibody-antigen and protein-oligonucleotide inter-actions.

GWC technologies offers an imaging system operating in the near-IR, the SPRimagerII array system, which for example, could be used for the investigation of antibody-antigen interactions, DNA arrays and protein-protein and DNA-protein interactions, using a variety of surface chemistries. A recent attempt to combine small molecule arrays, produced by photo-cross-linking and SPR imaging, was described by researchers from RIKEN. They studied array based ligand specific interaction of 8 different compounds with human estrogen receptor [77] using an OEM version of a GWC SPRimager instrument.

Currently available SPR imaging systems might be per-fectly suited to the analysis of interactions, such as antibody-protein, protein-protein, DNA-DNA, or protein–small molecule interactions, but fail to demonstrate high-through-put screening of large fragment libraries.

High Throughput SPR Screening of Chemical Microarrays

In our opinion successful primary screening of fragment libraries by SPR requires a symbiotic combination of technology components such as an optimized sensor surface architecture, a suitable fragment library, and a sensitive SPR imaging platform capable of screening thousands of fragment-protein interactions in a highly paralleled set-up.

In most SPR studies for screening small molecules, protein targets of interest are immobilized on the sensor surface and interactions with soluble low molecular weight

analytes are monitored. In the case of fragments <300Da, this detection scheme can lead to reduced signal to noise ratios, particularly for image based SPR systems. Further-more, due to the rather weak affinities between protein targets and fragments, the latter need to be screened at high compound concentrations which in turn can lead to artifacts resulting from compound solubility problems.

Complementary to this, in the case of chemical micro-arrays low molecular weight fragments are tethered to the surface and the binding of high molecular weight protein targets from solution is detected. Thus, the sensitivity and detection limit of Graffinity´s SPR imager technology (PlasmonImager), allows even small protein targets to be conveniently screened with a good signal/noise ratio. SPR imaging enables a parallel readout of thousands of fragment ligands against one target simultaneously [78,79].

Graffinity has developed high density chemical micro-arrays for fragment screening consisting of small molecules immobilized onto gold chips based on maleimide-thiol coupling chemistry in combination with high density pintool spotting [78,79]. The array preparation is a three-step pro-cess consisting of the synthesis of tagged fragments, preparation of SAM-Gold chips, and covalent coupling of tagged fragments to the activated SAM surface by pintool spotting. This process allows the sensor surface and syn-thesized fragments to be analyzed and quality controlled independently.

Small molecules are synthesized on long, flexible, and protein-adsorption resistant linkers by solid phase chemistry. Different types of spacer molecules contain a diverse set of terminal functionalities, which enable a large range of chemistries for fragment attachment and library synthesis. The common element of all linkers is a thiol group located at the opposite terminus. This thiol is the attachment point to the solid phase during library production. After synthesis completion, each compound is cleaved off the solid support and stored in separate stock solutions. The terminal thiol is deprotected during cleavage and is used for quality control by a thiol-selective LC-MS method which permits the establishment of a purity and quantity measure for every single compound [19]. The terminal thiol functionality of the library members also enables the subsequent covalent coupling to the activated binary SAM (Fig. (3a)). Synthe-sized fragment libraries are stored in microtiter plates until spotting onto the sensor fields by using a customized high throughput and high precision pintool spotting robot (Fig. (3b)). Currently, Graffinity´s small molecule library consists of 20,000 fragments and 90,000 lead-like compounds immobilized on chemical microarrays.

The first step of array production consists of micro-structuring glass slides with a titanium layer in order to define up to 9,216 spots per chip. Covering the chips with a thin gold layer allows the subsequent formation of a binary SAM composed of two thiols (Fig. (3b)). One of the thiols exhibits a maleimide moiety and serves as an anchor point for the subsequent attachment of molecules containing thiol groups. By controlling the ratio between the two SAM thiols, the surface density of anchor points, and hence the surface concentrations of ligands on the sensor, can be adjusted precisely.


The SAM surface/linker architecture shown in Fig (4) allows fragments to be immobilized in a highly defined manner. It has been proven to be resistant to unspecific protein adsorption, thus reducing background during the SPR experiment.

Fig. (4). Schematic drawing of the sensor surface architecture. The

chip consists of glass substrate covered with a thin gold layer onto

which a SAM was allowed to assemble. The latter contains anchor

groups which covalently react with the chemtag-linked fragments

during spotting in order to obtain a highly ordered sensor surface

structure.

Array screening is performed by floating protein solu-tions over the chemical microarray (Fig (5a)) and following the binding of the target by end-point measurements over 3 hours. Protein binding to individual fragments (Fig (5b)) causes a change in the SPR resonance conditions for the corresponding sensor field and gives rise to a shift in the wavelength dependent SPR minimum. Graffinity´s Plasmon Imager allows the parallel readout for up to 9,216 sensor fields per array (Fig (5c)).

The measured SPR shift can be visualized in false colored 2D plots to obtain affinity fingerprints, such as the ones given in Fig. (6). Here, a matrix metalloprotease (MMP) was screened against a library of small molecules using SPR imaging technology. Colored spots in Fig. (6a) represent sensor fields carrying small molecules to which strong protein binding occurred during the course of the measurement. These spots indicate hit series. As this screening project was aimed at identifying novel non-Zinc-site binders for MPPs, a binding site mapping was performed by on-array competition studies. In an independent fingerprinting experiment, the protein was preincubated with aceto-hydroxamate (a known Zn site binder for MMPs) before the target was applied to the microarray. Blocking of the enzyme´s Zn site prevented binding to several hit series initially identified for the non-blocked protein. The remaining signals were interpreted to represent alternative site binders for the screened MMP. Members of this hit series were synthesized, without the Chemtag, and tested in a secondary solution based assay. Here, the IC50s of the hit compounds were determined to be 1 M for the best binary compounds (MW 420 Da) and 10 μM for the best fragment (MW 250 Da). In addition to this, a Zinc site independent binding mode for these compounds could be confirmed in a biological assay. In further studies, not shown, other MMPs were screened against the same library which resulted in clear differences in binding patterns and hence in identifying MMP subtype selective compounds having low molecular weights and activities in the μM range.

CASE STUDIES

Thrombin Case Study

In the thrombin case study, a combination of low affinity fragment-based SPR screening with information from co-crystal-structures enabled a new strategy to be pursued in an already established research area. The serine protease thrombin catalyses the final step of the blood coagulation cascade. Here, fibrin is proteolytically cleaved to fibrinogen and the factor XIIIa is converted to XIII, which crosslinks the fibrin clot. Furthermore, it is the most potent endogenous platelet activator [80]. Thrombin is regarded as an important target for the treatment of thromboembolic disorders, such as deep vein thrombosis, stroke and myocardial infarction [81]. The majority of published thrombin antagonists like arga-troban and melagatran suffer from poor oral absorption because of highly charged guanidine- or benzamidine

Fig. (3). Preparation of fragment arrays by pintool spotting on microstructured gold substrates. (a) Glass substrates are microstructured by

photolithographic procedures, covered with a thin gold layer and a binary SAM of two thiols is deposited in order to form the activated

sensor surface. (b) Nanoliter droplets of fragment solutions are transferred by pintool spotting and allowed to couple covalently to the

activated SAM surface. (c) The final array consists of 9600 individual sensor fields each being occupied by a different type of fragment.


functionalities binding to the enzyme´s S1 pocket. Therefore, one challenge of thrombin inhibitor design is to identify structurally different molecules with better oral bio-availability [82].

SPR screening of thrombin against a small molecule compound collection immobilized on chemical microarrays enabled the identification of several key motifs responsible for the binding affinity (Fig. (7)). Charged motifs 1 and 2 are typical S1 pocket binders for thrombin, whereas neutral fragment motifs 3 and 4 are more unusual for thrombin inhibitors. On-array competition experiments with a nanomolar active site thrombin inhibitor (Pefablock TH58), resulted in disappearance of the identified motifs. This suggested that active site binding of the compounds was taking place. Based on the analysis of the binding data, a hit derivative comprising of motifs 1 and 4 was synthesized (Fig. (8) Compound 1). This weak inhibitor (IC50 of 200 M) was co-crystallized with thrombin and revealed an unusual binding mode. In the X-ray structure, the chloro-phenyl-thioether motif was binding in the S1 pocket, and the dedicated S1 phenylguanidine pocket binder was exposed to the solvent. This information was used in the design of a new series of thrombin inhibitors with a neutral P1 fragment. The optimization resulted in a compound (Fig. (8) Compound 2) with an IC50 of 2 nM, which delivered a starting point for a

medicinal chemistry program resulting in compounds with efficacy in animal models [13,83].

Phosphodiesterase 5 Case Study

The phosphodiesterase five (PDE5) case study is an excellent example for the optimization of weakly binding fragments by chemistry methods. In this case the analysis of the microarray binding data was used as a guidance for parallel synthesis. Phosphodiesterases (PDEs) are a family of metalloposphohydrolases which play an important role in cell signalling [84]. The substrates of phosphodiesterases are cyclic guanosine monophosphate (cGMP) and cyclic adeno-sine monophosphate (cAMP), which are ubiquitous secon-dary messengers mediating a range of important physio-logical processes. Cellular levels are regulated by cAMP and cGMP synthesis by cyclases, and by hydrolysis of the cyclic phosphodiester bonds into the corresponding 5’-mono-phosphates by phosphodiesterases. The class of human PDEs consists of established targets for therapeutic invention in several disease areas [85]. The most prominent example, PDE5, is a key target for the treatment of erectile dysfunc-tion and possibly plays an important role in other medical conditions like circulatory disorders or cystic fibrosis [86]. The goal of this study was the identification of novel inhi-bitor chemotypes as starting point for medicinal chemistry programs for PDE5. This was achieved by SPR imaging of

Fig. (5). Array Readout by SPR Imaging. (a) The array is incubated with the protein to be screened against the immobilized fragment

collection. (b) Protein binding to individual fragments cause a change in the SPR resonance conditions for the corresponding sensor field and

give rise to a shift in the SPR minimum. (c) All sensor fields per array are read out simultaneously by wavelength dependent SPR Imaging.

Fig. (6). Fingerprinting affinity map of a Metallomatrixprotease screened against a library of small molecules using Graffinity´s SPRImager.

(a) SPR shifts are represented in a false colored interaction map (b) Competition experiments using known Zn site binders of MMPS

identified hit series binding to alternative binding sites.


the PDE5 catalytic domain against a collection of 20,000 fragment and 90,000 lead-like compounds immobilized on chemical microarrays.

(Fig. (9)) gives an overview of four identified PDE5 fragment motifs 5-8, with the R group representing their attachment point towards the polyethylene glycole linker moiety. This spacer is necessary for the covalent thioether linkage to the anchor groups of the self-assembled array surface monolayer (Fig. (4)). Fragment motifs 7 and 8 are abundant in several decorated variations allowing the establishment of initial structure-affinity relationships by analysing the SPR shift signals. The relative ranking of the SPR fragment signal intensities, with different substituents, showed that the chloro-substituted fragment motifs 7a, b and 8b are significant better binders than the motifs 5, 6, 7c and 8a. To study the translation of the observed binding affinity into inhibitory activity in a secondary biochemical assay, a small number of derivatives from each fragment motif was synthesized by replacing the linker moiety with other functional groups (Fig. (10) B-D).

Due to the immobilization of the compounds, the spacer moiety is exposed to the bulk solvent. The linker is designed to avoid any binding with the target. However, follow-up tag replacement is suitable to identify potential interactions at the linker exit point. The chemical handle, used for the connection of the fragment to the spacer moiety, can also be used to introduce diversity at this part of the molecule. The polyethylene glycol units of the linker can be completely removed by leaving the chemical handle intact (Fig. (10), B). An alternative approach is to replace the tag by a small

ethylene glycol ether acting as a spacer substitute (Fig. (10), C). This small ether substituent usually improves the solubility of the resulting compounds, which is particularly important if a high concentration biochemical assay is subsequently applied.

The results from the secondary testing of the synthesized binder derivatives are presented in Table (1). PDE5 inhi-bition was observed in the mM to M range for the different fragment derivatives, as expected from the high sensitivity of the biophysical SPR detection method and the wide dynamic range of the observed SPR signals.

A comparison of the biochemical activities of tag replacement compounds for each motif shows the influence of the different spacer substitutions. For example in motif 6 (Table (1), No 5, 6) there is no significant difference in activity between a fragment just containing the chemical handle and a derivative with the ethyleneglycole linker substitute. This suggests that in this case the interactions between linker and protein is negligable. A slightly different picture is shown in the case of motif 8 derivates. Bioche-mical testing of these compounds revealed a modulation of the enzyme inhibition, depending on the spacer substitute structure (Table (1), No. 13-17).

An example of the generation of relationships between structure and inhibitory activity by tag replacement, is the benzyloxythiophene motif 8. This scaffold appeared in two variations on the arrays and was also found as a substructure of a lead-like binary array compound giving additional hints for further optimization. In accordance with the observed SPR binding affinity derivatives of the p-chlorosubstituted

Fig. (7). Privileged binder motifs for thrombin 1,2 represent typical charged S1 pocket binders, 3 and 4 are non-polar substitutes.

Fig. (8). Initial weak competitive thrombin inhibitor 1 and the optimized lead candidate 2.

N

N

N N

N

N

Cl

S

Cl

O

Cl

Motif 1 Motif 2 Motif 3 Motif 4

1, Ki (Thrombin) = 200 M

N

N N

N

O

N O

S

Cl

H2N

O

N

O

NH

S

Cl

2, Ki (Thrombin = 2 nM)


benzyloxythiophene (Table (1), No 15,17) showed much better PDE5 inhibition than the unsubstituted analogues (Table (1), No 13,14). The variation of linker replacements showed that the methoxyethylene amide of the benzyloxy-thiophene fragment is the most potent compound in this series with an IC50 value of 4 M and a MW of 326 Da (Table (1), No 17).

Array motif No 8 was also identified as a substructure in several lead-like array compounds depicted in Fig. (11). The structure-affinity information from the array screening was subsequently used for the design of a follow-up library of 20 compounds, leading to the racemic “cis”-derivative of motif 9 with an IC50 value of ~1 M. Chiral HPLC allowed the separation of both enantiomers leading to IC50’s of 15 M for motif 9a and of 360 nM for motif 9b.

Fig. (9). PDE5 binder motifs identified by SPR imaging (TAG = see figure 10).

Fig. (10). Tag replacement strategy for fragment validation. A) Fragment-tag attachment (R = motifs figure 9) via amide linkage. B-D)

Typical tag replacements for first generation test compounds. E) Thiol group used for covalent attachment to the array surface.

Fig. (11). Array informed parallel chemistry (R = CONH-TAG). Array motifs 8 c - e are used as a guidance for the synthesis of PDE5

inhibitors leading to compound 9b with nM potency.

NO

R

Z

7 a) X,Y = Cl, Z = H

7 b) X,Z = Cl, Y = H

7 c) X = CF3, Y,Z = H

S

R

O R

O

S

R

X

8 a) X = H

8 b) X = Cl

X

Motif 5

Y

Motif 6NH

O

TAGR=

Motif 7

Motif 8

NH

O

TAGR=NH

O

ONH2

O

OH

O

SHN

O

O

ON

O

OOTAG =

Tag ReplacementA B C D

E

SO

O

HN

HN

HN

R

HN

R

SO

O

HN

O

Motif 9a (IC50 ent-A = 15 M)

Motif 9b (IC50 ent-B = 360 nM)Motif 8c

R

Motif 8d Motif 8e

Motif 8

c-e

R


Motif 9 is an excellent starting point for further optimi-zation, by using the array binding information from the different substitution patterns of the benzyloxythiophene motif.

In summary, the PDE5 case study demonstrates that microarray based SPR screening of a large fragment library delivers structure-affinity information from several deriva-tives of each scaffold, and guides subsequent optimization leading to compounds with nM potency. The incorporation of chemical handles in the fragment motifs enables fast follow-up chemistry and allows an initial high throughput optimization without the need for additional information.

CONCLUSION

SPR detection has become a popular tool for the analysis of virtually any kind of biomolecular interactions and a variety of instruments are available today. Although being perfectly suited for analysing protein arrays and protein-protein interactions, primary screening of large fragment libraries for drug discovery is often encumbered by the limited number of sensing channels of current systems.

The authors demonstrated that chemical microarrays are a promising tool for successful primary screening of immobilized fragment libraries by SPR imaging. The app-

roach requires a combination of an appropriate sensor surface chemistry, a suitable fragment library, and a sensitive SPR imaging technology, being capable of simultaneously screening thousands of fragment-protein interactions.

The potential of SPR imaging for fragment-based dis-covery was exemplified in several case studies. Here, microarray screening led to the identification of weakly binding fragment motifs, the inhibitory activity of which could be confirmed in secondary biochemical assays. Basic structure activity relationships for fragment motifs could be derived by analyzing the SPR signals of the immobilized structures and their analogues. Subsequently, array informed parallel chemistry led to the optimization of initially weak fragments to inhibitors with nM potency in a short time frame of about 4 - 6 months.

REFERENCES

[1] Rees, D. C.; Congreve, M.; Murray, C. W.; Carr, R. Fragment-

Based Lead Discovery. Nat. Rev. Drug Discov., 2004, 3, 660-672. [2] Erlanson, D. A. Fragment-Based Lead Discovery: a Chemical

Update. Curr. Opin. Biotechnol., 2006, 17, 643-652. [3] Fink, T.; Bruggesser, H.; Reymond, J. L. Virtual Exploration of the

Small-Molecule Chemical Universe Below 160 Daltons. Angew. Chem. Int. Ed. Engl., 2005, 44, 1504-1508.

Table 1. Selected Biochemical Data of 1st Generation Motif Derivatives

NO MOTIF R = MW % Inh 1 mM % Inh 0,5 mM % Inh 0,1 mM IC50 M

3 5 CONH2 213 27 - - -

4 5 CONHCH2CH2OCH3 271 - 20 - -

5 6 CONH2 181 34 - - -

6 6 CONHCH2CH2OCH3 239 35 - - -

7 7a CONH2 297 - - 77 -

8 7a CONHCH2CH2OCH3 355 - - 87 4

9 7b CONH2 296 - - - 3

10 7b COOH 297 - - 46 -

11 7b CONHCH2CH2OCH3 355 - - 85 11

12 7c CONH2 296 - 37 - -

13 8a CONH2 233 - - 8 -

14 8a CONHCH2CH2OCH3 291 - - 41 -

15 8b CONH2 267 - - 65 -

16 8b COOH 269 - - 0 -

17 8b CONHCH2CH2OCH3 326 - - 94 4

18 9a H 358 - - 86 15

19 9b H 358 - - 97 0.36

Inhibition of PDE5 activity was monitored in a fluorescence polarisation assay using fluorescein labeled cGMP as substrate. Compounds were preincubated with 25 nM PDE5 for 15

min in assay buffer (50 mM Tris pH 7.5, 150 mM NaCl, 5 mM MgCl2, 0.005 % Tween20, 1 mM DTT). The enzymatic reaction was started by the addition of substrate at a final concentration of 20 nM in total assay volume of 20 μl. After 30 minutes incubation at 25 °C, the fluorescein-cGMP hydrolysis was determined by addition of IMAP detection beads (Molecular Devices), which bind selectively the GMP phosphate group and generate a high polarisation signal. The inhibition values could be determined with a precision of +/- 10 %. Zaprinast was used as reference inhibitor and showed an IC50 of 2 μM.


[4] Hann, M. M.; Leach, A. R.; Harper, G. Molecular Complexity and

Its Impact on the Probability of Finding Leads for Drug Discovery. J. Chem. Inf. Comput. Sci., 2001, 41, 856-864.

[5] Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand Efficiency: a Useful Metric for Lead Selection. Drug Discov. Today, 2004, 9,

430-431. [6] Congreve, M.; Carr, R.; Murray, C.; Jhoti, H. A Rule of Three for

Fragment-Based Lead Discovery? Drug Discov. Today, 2003, 8, 876-877.

[7] Mitchell, T.; Showell, G. A. Design Strategies for Building Drug-Like Chemical Libraries. Curr. Opin. Drug Discov. Devel., 2001,

4, 314-318. [8] Lipinski, C.; Hopkins, A. Navigating Chemical Space for Biology

and Medicine. Nature, 2004, 432, 855-861. [9] Perez, J. J. Managing Molecular Diversity. Chem. Soc. Rev., 2005,

34, 143-152. [10] Baurin, N.; Aboul-Ela, F.; Barril, X.; Davis, B.; Drysdale, M.;

Dymock, B.; Finch, H.; Fromont, C.; Richardson, C.; Simmonite, H.; Hubbard, R. E. Design and Characterization of Libraries of

Molecular Fragments for Use in NMR Screening Against Protein Targets. J. Chem. Inf. Comput. Sci., 2004, 44, 2157-2166.

[11] Lepre, C. A. Library Design for NMR-Based Screening. Drug Discov. Today, 2001, 6, 133-140.

[12] Card, G. L.; Blasdel, L.; England, B. P.; Zhang, C.; Suzuki, Y.; Gillette, S.; Fong, D.; Ibrahim, P. N.; Artis, D. R.; Bollag, G.;

Milburn, M. V.; Kim, S. H.; Schlessinger, J.; Zhang, K. Y. A Family of Phosphodiesterase Inhibitors Discovered by

Cocrystallography and Scaffold-Based Drug Design. Nat. Biotechnol., 2005, 23, 201-207.

[13] Neumann, T.; Junker, H. D.; Keil, O.; Burkert, K.; Ottleben, H.; Gamer, J.; Sekul, R.; Deppe, H.; Feurer, A.; Tomandl, D.; Metz, G.

Discovery of Thrombin Inhibitor Fragments From Chemical Microarray Screening. Lett. Drug Des. Discov., 2005, 2, 590-594.

[14] Petros, A. M.; Dinges, J.; Augeri, D. J.; Baumeister, S. A.; Betebenner, D. A.; Bures, M. G.; Elmore, S. W.; Hajduk, P. J.;

Joseph, M. K.; Landis, S. K.; Nettesheim, D. G.; Rosenberg, S. H.; Shen, W.; Thomas, S.; Wang, X.; Zanze, I.; Zhang, H.; Fesik, S.

W. Discovery of a Potent Inhibitor of the Antiapoptotic Protein Bcl-XL From NMR and Parallel Synthesis. J. Med. Chem., 2006,

49, 656-663. [15] Choong, I. C.; Lew, W.; Lee, D.; Pham, P.; Burdett, M. T.; Lam, J.

W.; Wiesmann, C.; Luong, T. N.; Fahr, B.; DeLano, W. L.; McDowell, R. S.; Allen, D. A.; Erlanson, D. A.; Gordon, E. M.;

O'Brien, T. Identification of Potent and Selective Small-Molecule Inhibitors of Caspase-3 Through the Use of Extended Tethering

and Structure-Based Drug Design. J. Med. Chem., 2002, 45, 5005-5022.

[16] Hochgurtel, M.; Kroth, H.; Piecha, D.; Hofmann, M. W.; Nicolau, C.; Krause, S.; Schaaf, O.; Sonnenmoser, G.; Eliseev, A. V. Target-

Induced Formation of Neuraminidase Inhibitors From in Vitro Virtual Combinatorial Libraries. Proc. Natl. Acad. Sci. USA, 2002,

99, 3382-3387. [17] Whiting, M.; Muldoon, J.; Lin, Y. C.; Silverman, S. M.; Lindstrom,

W.; Olson, A. J.; Kolb, H. C.; Finn, M. G.; Sharpless, K. B.; Elder, J. H.; Fokin, V. V. Inhibitors of HIV-1 Protease by Using in Situ

Click Chemistry. Angew. Chem. Int. Ed. Engl., 2006, 45, 1435-1439.

[18] Schuffenhauer, A.; Ruedisser, S.; Marzinzik, A. L.; Jahnke, W.; Blommers, M.; Selzer, P.; Jacoby, E. Library Design for Fragment

Based Screening. Curr. Top Med. Chem., 2005, 5, 751-762. [19] Maier, S.; Frank, M.; Rau, H.; Lewandrowski, P.; Uhrig, R.; Keil,

O.; Deppe, H.; Muller, N.; Vanier, C.; Mannsperger, H.; Zepter, S.; Junker, H. D. Synthesis and Quality Control of Thiol Tagged

Compound Libraries for Chemical Microarrays. QSAR Comb. Sci., 2006, 25, 1047-1054.

[20] Purandare, A. V.; Gao, A.; Poss, M. A. Solid-Phase Synthesis of 'Diverse' Heterocycles. Tetrahedron Lett., 2002, 43, 3903-3906.

[21] Mayer, J. P.; Lewis, G. S.; McGee, C.; Bankaitis-Davis, D. Solid-Phase Synthesis of Benzimidazoles. Tetrahedron Lett., 1998, 39,

6655-6658. [22] Akamatsu, H.; Fukase, K.; Kusumoto, S. New Efficient Route for

Solid-Phase Synthesis of Benzimidazole Derivatives. J. Comb. Chem., 2002, 4, 475-483.

[23] Blundell, T. L.; Jhoti, H.; Abell, C. High-Throughput Crystallo-graphy for Lead Discovery in Drug Design. Nat. Rev. Drug

Discov., 2002, 1, 45-54.

[24] Gill, A. New Lead Generation Strategies for Protein Kinase

Inhibitors - Fragment Based Screening Approaches. Mini Rev. Med. Chem., 2004, 4, 301-311.

[25] Davies, T. G.; van Montfort, R. L.; Williams G; Jhoti, H. Pyramid: An Integrated Platform for Fragment based Discovery In

Fragment-based Approaches in Drug Discovery; Jahnke, W.; Erlanson, D. A.; Eds.; Wiley-VCH: Weinheim, Germany, 2006;

193-214 [26] Blaney, J.; Nienhaber, V.; Burley, S. K. Fragment-based Lead

Discovery and Optimization Using X-Ray Crystallography, Computational Chemistry, and High-Throughput Organic Synthesis

In Fragment-based Approaches in Drug Discovery; Jahnke, W.; Erlanson, D. A.; Eds.; Wiley-VCH: Weinheim, Germany, 2006;

215-248 [27] Tickle, I.; Sharff, A.; Vinkovic, M.; Yon, J.; Jhoti, H. High-

Throughput Protein Crystallography and Drug Discovery. Chem. Soc. Rev., 2004, 33, 558-565.

[28] Howard, N.; Abell, C.; Blakemore, W.; Chessari, G.; Congreve, M.; Howard, S.; Jhoti, H.; Murray, C. W.; Seavers, L. C.; van

Montfort, R. L. Application of Fragment Screening and Fragment Linking to the Discovery of Novel Thrombin Inhibitors. J. Med.

Chem., 2006, 49, 1346-1355. [29] Oslob, J. D.; Erlanson, D. A. Tethering in Early Target

Assessment. Drug Discov. Today Targets, 2004, 3, 143-150. [30] Erlanson, D. A.; McDowell, R. S.; O'Brien, T. Fragment-Based

Drug Discovery. J. Med. Chem., 2004, 47, 3463-3482. [31] Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.; Fesik, S. W.

Discovering High-Affinity Ligands for Proteins: SAR by NMR. Science, 1996, 274, 1531-1534.

[32] Klages, J.; Coles, M.; Kessler, H. NMR-Based Screening: a Powerful Tool in Fragment-Based Drug Discovery. Mol. Biosyst.,

2006, 2, 319-331. [33] Schade, M.; Oschkinat, H. NMR Fragment Screening: Tackling

Protein-Protein Interaction Targets. Curr. Opin. Drug Discov. Devel., 2005, 8, 365-373.

[34] Pellecchia, M.; Sem, D. S.; Wuthrich, K. NMR in Drug Discovery. Nat. Rev. Drug Discov., 2002, 1, 211-219.

[35] Aust, E. F.; Ito, S. F.; Sadowny, M.; Knoll, W. Investigation of Thin Films Using Surface Plasmon Modes and Optical Waveguide

Modes. Trends Polymer Sci., 1994, 2, 313-323. [36] Schuck, P. Use of Surface Plasmon Resonance to Probe the Equili-

brium and Dynamic Aspects of Interactions Between Biological Macromolecules. Annu. Rev. Biophys. Biomol. Struct., 1997, 26,

541-566. [37] Homola, J.; Yee, S. S.; Gauglitz, G. Surface Plasmon Resonance

Sensors: Review. Sens Actuators B Chem., 1999, 54, 3-15. [38] Homola, J. Present and Future of Surface Plasmon Resonance

Biosensors. Anal. Bioanal. Chem., 2003, 377, 528-539. [39] Rich, R. L.; Myszka, D. G. Why You Should Be Using More SPR

Biosensor Technology. Drug Discov. Today Technol., 2004, 1, 301-308.

[40] Rich, R. L.; Myszka, D. G. Survey of the Year 2004 Commercial Optical Biosensor Literature. J. Mol. Recognit., 2005, 18, 431-478.

[41] Nilsson, P.; Persson, B.; Uhlen, M.; Nygren, P. A. Real-Time Monitoring of DNA Manipulations Using Biosensor Technology.

Anal. Biochem., 1995, 224, 400-408. [42] Bates, P. J.; Dosanjh, H. S.; Kumar, S.; Jenkins, T. C.; Laughton,

C. A.; Neidle, S. Detection and Kinetic Studies of Triplex Formation by Oligodeoxynucleotides Using Real-Time

Biomolecular Interaction Analysis (BIA). Nucleic Acids Res., 1995, 23, 3627-3632.

[43] Peterlinz, K. A.; Georgiadis, R. M.; Herne, T. M.; Tarlov, M. J. Observation of Hybridization and Dehybridization of Thiol-

Tethered DNA Using Two-Color Surface Plasmon Resonance Spectroscopy. J. Am. Chem. Soc., 1997, 119, 3401-3402.

[44] Rahn, J. R.; Hallock, R. B. Antibody-Binding to Antigen-Coated Substrates Studied With Surface-Plasmon Oscillations. Langmuir,

1995, 11, 650-654. [45] Brynda, E.; Homola, J.; Houska, M.; Pfeifer, P.; Skvor, J. Antibody

Networks for Surface Plasmon Resonance Immunosensors. Sens Actuators B Chem., 1999, 54, 132-136.

[46] Rich, R. L.; Myszka, D. G. Survey of the Year 2001 Commercial Optical Biosensor Literature. J. Mol. Recognit., 2002, 15, 352-376.

[47] Nordin, H.; Jungnelius, M.; Karlsson, R.; Karlsson, O. P. Kinetic Studies of Small Molecule Interactions With Protein Kinases Using

Biosensor Technology. Anal. Biochem., 2005, 340, 359-368.


[48] Rich, R. L.; Day, Y. S.; Morton, T. A.; Myszka, D. G. High-

Resolution and High-Throughput Protocols for Measuring Drug/Human Serum Albumin Interactions Using BIACORE. Anal.

Biochem., 2001, 296, 197-207. [49] Day, Y. S.; Baird, C. L.; Rich, R. L.; Myszka, D. G. Direct

Comparison of Binding Equilibrium, Thermodynamic, and Rate Constants Determined by Surface- and Solution-Based Biophysical

Methods. Protein Sci., 2002, 11, 1017-1025. [50] Knoll, W.; Zizlsperger, M.; Liebermann, T.; Arnold, S.; Badia, A.;

Liley, M.; Piscevic, D.; Schmitt, F. J.; Spinke, J. Streptavidin Arrays As Supramolecular Architectures in Surface-Plasmon

Optical Sensor Formats. Colloids Surf. A Physicochem. Eng. Asp., 2000, 161, 115-137.

[51] Morgan, H.; Taylor, D. M. A Surface Plasmon Resonance Immunosensor Based on the Streptavidin-Biotin Complex. Biosens.

Bioelectron, 1992, 7, 405-410. [52] Watts, H. J.; Yeung, D.; Parkes, H. Real-Time Detection and

Quantification of DNA Hybridization by an Optical Biosensor. Anal. Chem., 1995, 67, 4283-4289.

[53] Jordan, C. E.; Frutos, A. G.; Thiel, A. J.; Corn, R. M. Surface Plasmon Resonance Imaging Measurements of DNA Hybridization

Adsorption and Streptavidin/DNA Multilayer Formation at Chemically Modified Gold Surfaces. Anal. Chem., 1997, 69, 4939-

4947. [54] Zizlsperger, M.; Knoll, W. Multispot Parallel on-Line Monitoring

of Interfacial Binding Reactions by Surface Plasmon Microscopy. Progr. Colloid. Polym. Sci., 1998, 109, 244-253.

[55] Duschl, C.; Liley, M.; Lang, H.; Ghandi, A.; Zakeeruddin, S. M.; Stahlberg, H.; Dubochet, J.; Nemetz, A.; Knoll, W.; Vogel, H.

Sulphur-Bearing Lipids for the Covalent Attachment of Supported Lipid Bilayers to Gold Surfaces: A Detailed Characterisation and

Analysis. Mat. Sci. Eng. C-Bio., 1996, 4, 7-18. [56] BIACORE, www.biacore.com 2007

[57] Flink, S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Sensor Functionalities in Self-Assembled Monolayers. Adv. Mater.

Deerfield, 2000, 12, 1315-1328. [58] Wink, T.; vanZuilen, S. J.; Bult, A.; vanBennekom, W. P. Self-

Assembled Monolayers for Biosensors. Analyst, 1997, 122, R43-R50.

[59] Nenninger, G. G.; Clendenning, J. B.; Furlong, C. E.; Yee, S. S. Reference-Compensated Biosensing Using a Dual-Channel Surface

Plasmon Resonance Sensor System Based on a Planar Lightpipe Configuration. Sens. Actuators B Chem., 1998, 51, 38-45.

[60] Lofas, S.; Malmqvist, M.; Ronnenberg, I.; Stenberg, E.; Liedberg, B.; Lundstrom, I. Bioanalysis With Surface-Plasmon Resonance.

Sens. Actuators B Chem., 1991, 5, 79-84. [61] Huber, W. A New Strategy for Improved Secondary Screening and

Lead Optimization Using High-Resolution SPR Characterization of Compound-Target Interactions. J. Mol. Recognit., 2005, 18, 273-

281. [62] Yeatman, E.; ASH, E. A. Surface-Plasmon Microscopy. Electron

Lett., 1987, 23, 1091-1092. [63] Yeatman, E.; Ash E.A. SPIE Scanning. Microsoc. Technol. Appl.,

1988, 897, 100-107. [64] Rothenhausler, B.; Knoll, W. Surface-Plasmon Microscopy.

Nature, 1988, 332, 615-617. [65] Brockman, J. M.; Nelson, B. P.; Corn, R. M. Surface Plasmon

Resonance Imaging Measurements of Ultrathin Organic Films. Annu. Rev. Phys. Chem., 2000, 51, 41-63.

[66] Silin, V. I.; Balchytis, G. A.; Yakolev, V. A. Study on Surface-Plasmon Line Broadening Due to Protein Adsorption on a

Modified Silver Surface. Opt. Commun., 1993, 97, 19-24. [67] Piscevic, D.; Knoll, W.; Tarlov, M. J. Surface Plasmon Microscopy

of Biotin-Streptavidin Binding Reactions on UV-Photopatterned Alkanethiol Self-Assembled Monolayers. Supramol. Sci., 1995, 2,

99-106.

[68] Thiel, A. J.; Frutos, A. G.; Jordan, C. E.; Corn, R. M.; Smith, L. M.

In Situ Surface Plasmon Resonance Imaging Detection of DNA Hybridization to Oligonucleotide Arrays on Gold Surfaces. Anal.

Chem., 1997, 69, 4948-4956. [69] Frutos, A. G.; Corn, R. M. SPR of Ultrathin Organic Films. Anal.

Chem., 1998, 70, 449A-455A. [70] Nelson, B. P.; Frutos, A. G.; Brockman, J. M.; Corn, R. M. Near-

Infrared Surface Plasmon Resonance Measurements of Ultrathin Films. 1. Angle Shift and SPR Imaging Experiments. Anal. Chem.,

1999, 71, 3928-3934. [71] Berger, C. E. H.; Beumer, T. A. M.; Kooyman, R. P. H.; Greve, J.

Surface Plasmon Resonance Multisensing. Anal. Chem., 1998, 70, 703-706.

[72] Kyo, M.; Usui-Aoki, K.; Koga, H. Label-Free Detection of Proteins in Crude Cell Lysate With Antibody Arrays by a Surface Plasmon

Resonance Imaging Technique. Anal. Chem., 2005, 77, 7115-7121. [73] Unfricht, D. W.; Colpitts, S. L.; Fernandez, S. M.; Lynes, M. A.

Grating-Coupled Surface Plasmon Resonance: a Cell and Protein Microarray Platform. Proteomics, 2005 , 5, 4432-4442.

[74] Usui-Aoki, K.; Shimada, K.; Nagano, M.; Kawai, M.; Koga, H. A Novel Approach to Protein Expression Profiling Using Antibody

Microarrays Combined With Surface Plasmon Resonance Technology. Proteomics, 2005, 5, 2396-2401.

[75] Wassaf, D.; Kuang, G.; Kopacz, K.; Wu, Q. L.; Nguyen, Q.; Toews, M.; Cosic, J.; Jacques, J.; Wiltshire, S.; Lambert, J.;

Pazmany, C. C.; Hogan, S.; Ladner, R. C.; Nixon, A. E.; Sexton, D. J. High-Throughput Affinity Ranking of Antibodies Using Surface

Plasmon Resonance Microarrays. Anal. Biochem., 2006, 351, 241-253.

[76] Boozer, C.; Kim, G.; Cong, S.; Guan, H.; Londergan, T. Looking Towards Label-Free Biomolecular Interaction Analysis in a High-

Throughput Format: a Review of New Surface Plasmon Resonance Technologies. Curr. Opin. Biotechnol., 2006, 17, 400-405.

[77] Kanoh, N.; Kyo, M.; Inamori, K.; Ando, A.; Asami, A.; Nakao, A.; Osada, H. SPR Imaging of Photo-Cross-Linked Small-Molecule

Arrays on Gold. Anal. Chem., 2006, 78, 2226-2230. [78] Vetter, D. Chemical Microarrays, Fragment Diversity, Label-Free

Imaging by Plasmon Resonance--a Chemical Genomics Approach. J. Cell. Biochem. Suppl., 2002, 39, 79-84.

[79] Dickopf, S.; Frank, M.; Junker, H. D.; Maier, S.; Metz, G.; Ottleben, H.; Rau, H.; Schellhaas, N.; Schmidt, K.; Sekul, R.;

Vanier, C.; Vetter, D.; Czech, J.; Lorenz, M.; Matter, H.; Schudok, M.; Schreuder, H.; Will, D. W.; Nestler, H. P. Custom Chemical

Microarray Production and Affinity Fingerprinting for the S1 Pocket of Factor VIIa. Anal. Biochem., 2004, 335, 50-57.

[80] Mann, K. G.; Butenas, S.; Brummel, K. The Dynamics of Thrombin Formation. Arterioscler Thromb. Vasc. Biol., 2003, 23,

17-25. [81] Steinmetzer, T.; Sturzebecher, J. Progress in the Development of

Synthetic Thrombin Inhibitors As New Orally Active Anticoagulants. Curr. Med. Chem., 2004, 11, 2297-2321.

[82] Schwienhorst, A. Direct Thrombin Inhibitors - a Survey of Recent Developments. Cell. Mol. Life Sci., 2006, 63, 2773-2791.

[83] Bulat, S.; Bosio, S.; Papadopoulos, M. A.; Cerezo-Galvez, S.; Grabowski, E.; Rosenbaum, C.; Matassa, V. G.; Ott, I.; Metz, G.;

Schamberger, J.; Sekul, R.; Feurer, A. Design and Discovery of Novel, Potent Pyrazinone-Based Thrombin Inhibitors With a

Solubilizing Amino P-1-P-2-Linker. Lett. Drug Des. Discov., 2006, 3, 289-292.

[84] Francis, S. H.; Turko, I. V.; Corbin, J. D. Cyclic Nucleotide Phosphodiesterases: Relating Structure and Function. Prog. Nucleic

Acid Res. Mol. Biol., 2001, 65, 1-52. [85] Lugnier, C. Cyclic Nucleotide Phosphodiesterase (PDE) Super-

family: a New Target for the Development of Specific Therapeutic Agents. Pharmacol. Ther., 2006, 109, 366-398.

[86] Rotella, D. P. Phosphodiesterase 5 Inhibitors: Current Status and Potential Applications. Nat. Rev. Drug Discov., 2002, 1, 674-682.

Received: March 6, 2007 Accepted: April 6, 2007

SPR-based Fragment Screening: Advantages and...

Documents

Transcript of SPR-based Fragment Screening: Advantages and...