home > ebr > autumn 2010 > systematic drug reprofile
European Biopharmaceutical Review

Systematic Drug Reprofile

Historically, novel drugs were often discovered by screening natural extracts on cell-based assays and observing changes of phenotype. With advances in genomics and proteomics, a shift towards a target-centric approach has occurred, where isolated targets are used to screen libraries of defined chemical substances, with the aim of identifying small molecules that inhibit the activity of a particular target. A consequence of this approach is a reliance on the ‘one drug, one target’ model, where the therapeutic effect of a drug is reduced to its action on a single target. However, recent approaches to systematically profiling the target spectrum of drugs have revealed that many drugs act on more than one target. The drug Glivec (imatinib), for example, initially developed by Novartis as a highly selective inhibitor of the Bcr-Abl oncogene, was shown in later studies to inhibit other kinase targets, such as the receptor tyrosine kinase c-KIT. Based on these findings, the treatment indication for Glivec has now been expanded from chronic myeloid leukaemia to gastrointestinal stromal tumours (1). This example shows that even drugs which were originally designed to be highly specific inhibitors of one particular target may well act on additional targets, and that these actions can be used to reposition such drugs towards additional indications.

Drug profiling approaches are now being frequently used to identify all targets for a given drug in a systematic way, and have demonstrated that many drugs interact with a surprising number of targets (2). In addition, systematic reprofiling may also be used to reveal alternative indications for drugs already on the market and for drug candidates whose development has been stopped due to a lack of efficacy. Such drugs may be given a ‘second chance’ by repurposing them to alternative indications.


A variety of technologies are available today for target identification, including affinity chromatography based (chemical proteomics) methods, three-hybrid based methods, phage display and protein microarray approaches, as well as technologies that provide more global information, such as drug-induced changes in mRNA levels, and protein or metabolite expression profiles (3). Each of these approaches has its own advantages and disadvantages. Generally, methods that supply direct drug-target interaction information require an immobilisation or derivatization step of the drug to be screened, which may alter its target binding spectrum. Chemical proteomics and three-hybrid methods employ target proteins from a native environment (such as cells or tissue), whereas phage displays or microarrays present target proteins in a relatively unphysiological environment and, consequently, some targets may be missed due to folding or aggregation problems with these methods.

In recent years, both chemical proteomics and three-hybrid approaches have been frequently used to profile the target spectrum of many kinase inhibitors (4,5). Protein kinases are prime targets for anticancer therapies, but their close structural similarities make the design of specific inhibitors difficult. As a result, the majority of kinase inhibitors currently in clinical development show affinities of varying degrees to other kinases besides the target they were originally designed for. Chemical proteomics and three-hybrid methods have both been used to great advantage to profile kinase inhibitors and to uncover their respective target spectra. In the following paragraphs, we will briefly discuss those two technologies and highlight some of the findings obtained by drug profiling of kinase inhibitors.


In recent years, the field of proteomics has seen a rapid development of innovative methods for the identification and quantification of protein interactions and post-translational protein modifications. Combining cell biology, biochemistry and mass spectrometry tools now allows indepth analysis of large protein complexes (functional proteomics), post-translational protein modification states (for example phosphoproteomics) and even drug-target interactions (chemical proteomics).

With the chemical proteomics approach, a drug under investigation is bound to a solid support using an appropriate flexible linker and is subsequently incubated with cell or tissue lysates to capture its specific interacting protein targets. The drug may either be immobilised covalently to the solid support, or via a biotin-avidin interaction; the chemical linker is usually an alkyl or polyethylene glycol chain. Following incubation of the lysate with the immobilised drug, non-specifically bound proteins are removed by repeated washing steps and, finally, the specifically bound targets are eluted from the drug and analysed using, for example, liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS).

There are two important considerations to this approach: the length of the linker must be chosen so that it does not interfere with drug-target binding and in such a way that it provides sufficient flexibility to avoid steric hindrance; and the linker must be attached to the drug in such a way that it still allows target binding (the derivatised drug should maintain its original activity). Often, several derivatives are synthesised (with the linker being attached at different sites of the drug) in order to obtain at least one derivative with acceptable activity towards its target(s).

One of the earliest examples of the power of chemical proteomics was the profiling of the kinase inhibitor Purvalanol B. Purvalanol derivatives were known to bind to and inhibit cyclin dependent kinases, but the exact target spectrum was unknown. By screening an immobilised form of Purvalanol B against tissue extracts from different species, Knockaert and colleagues were able to identify several kinases which bind Purvalanol B with high affinity, among them CDK5, ERK1, ERK2 and S6 kinase (6). Chemical proteomics has also been used extensively to characterise the target spectrum of the blockbuster drugs Glivec (imatinib), Tasigna (nilotinib) and Sprycel (dasatinib). Initial studies showed substantial differences in the specificity of these drugs, despite the fact that all three have been primarily developed as inhibitors of Bcr-Abl. For instance, dasatinib appears to be more promiscuous than either imatinib or nilotinib and interacted with more than 20 kinases expressed in K562 cells (7). Apart from identifying novel targets for a drug, and hence potentially novel indications, these studies also highlight the potential of drug profiling approaches for identifying possible side effects associated with drug treatment.

A drawback of the chemical proteomics approach is the high number of potential drug targets found in those screens. Extensive lists of dozens or even hundreds of co-purified proteins, many of them weak or non-specific interactors, often obscure targets of high value and render a straightforward analysis rather difficult. The combination of chemical proteomics and quantitative mass spectrometry using labelling technologies such as SILAC or iTRAQ (reviewed in (8)) may offer a solution to this problem. This approach, termed ‘quantitative chemical proteomics’ accurately measures relative affinities of a drug under investigation in a competition assay, which allows not only the identification of targets for a particular drug, but also the determination of affinities for each drug-target interaction. This information is subsequently used to prioritise drug targets for further downstream studies. Quantitative chemical proteomics has recently been used to study the mechanism of action for several clinical Bcr-Abl inhibitors (9).


The three-hybrid based drug profiling technology is built upon a workhorse of modern biology – the yeast two-hybrid system (10). Originally conceived by Stanley Fields and Ok-kyu Song in 1989, the yeast two-hybrid system has been in constant use for the past 20 years to discover novel protein-protein interactions, and its success is underscored by the fact that the majority of protein interactions recorded in public databases have been identified by its use. Three-hybrid based drug profiling takes advantage of the yeast two-hybrid mechanism by replacing one of the protein partners by a small molecule drug, thereby extending the application from protein-protein to drug-protein interactions. The drug under investigation is derivatized by attaching a linker and a so-called ‘anchor’ part, which is needed to couple the drug derivative to the DNA binding domain of a transcription factor. The cognate target of the drug in turn is expressed as a fusion to the activation domain of a transcription factor. Interactions between a drug and its target are detected by the activity of a transcription factor, which is reconstituted from the two separate domains by the drug-target interaction. The reconstituted (‘hybrid’) transcription factor activates one or several reporter genes integrated into the yeast genome, whose activity is then measured. Typical reporter genes are either growth markers which allow direct selection of colonies bearing a drug-target interaction, or colorimetric markers, such as lacZ. An advantage of three-hybrid systems when compared to chemical proteomics approaches is the fact that drug-target interactions are measured in situ inside intact cells, which results in an added level of specificity and the ability to detect weak and transient interactions. Several three-hybrid systems exist, which differ mainly in their reporter genes and the anchors used for immobilisation of the drug (11,12).

Like chemical proteomics, three-hybridbased drug profiling has also been used to explore the target spectrum of kinase inhibitors. Extensive screening of the drug Purvalanol B against several cDNA libraries identified known targets, such as CDK2 and ERK2, but also a whole range of new kinases inhibited by Purvalanol B (13). Since the cDNA libraries used for expressing the targets inside yeast cells require only minute quantities of tissue, three-hybrid approaches could in principle also be used to screen patient tissues for the presence or absence of particular targets of a drug under development.

A drawback of the three-hybrid approach is the relative inaccessibility of integral membrane proteins as targets, since drugtarget interactions must take place in the nucleus of the yeast cell in order to be detectable. To circumvent this limitation, other genetic screening systems have been adapted to the three-hybrid approach, such as the split-ubiquitin system and the MappIT system (14,15). These systems allow greater flexibility in the choice of detectable targets and also operate at cellular membranes or organelles – thereby greatly expanding the target space which can be searched with a drug. However, neither of the systems has seen widespread use so far and their potential for drug profiling thus remains to be demonstrated.


With the looming patent cliff and the shrinking development pipelines of many pharmaceutical companies, a central task is the investigation of alternative and more innovative approaches to drug development. A promising route is indication expansion for drugs under development or even the repositioning of marketed drugs to alternative indications. Here, drug profiling may be of great benefit since it allows quick and unbiased screening for unknown drug targets. Drug profiling technologies have seen rapid development in the past few years and many bottlenecks formerly associated with both chemical proteomics and three-hybrid technologies have been overcome, opening the possibility of at least mediumthroughput screening for future systematic reprofiling or repositioning efforts.


  1. Druker BJ, STI571 (Gleevec) as a paradigm for cancer therapy, Trends in Molecular Medicine 8: S14-S18, 2002
  2. Hart CP, Finding the target after screening the phenotype, Drug Discovery Today 10: pp513-519, 2005
  3. Terstappen GC, Schlüpen C, Raggiaschi R and Gaviraghi G, Target deconvolution strategies in drug discovery, Nature Reviews Drug Discovery 6: pp891-903, 2007
  4. Kruse U, Bantscheff M, Drewes G and Hopf C, Chemical and pathway proteomics: powerful tools for oncology drug discovery and personalized healthcare, Mol Cell Proteomics 7: pp1,887-1,901, 2008
  5. Lefurgy S and Cornish V, Finding Cinderella after the ball: a threehybrid approach to drug target identification, Chem Biol 11: pp151-153, 2004
  6. Knockaert M, Gray N, Damiens E, Chang YT, Grellier P, Grant K, Fergusson D, Mottram J, Soete M, Dubremetz JF, Le Roch K, Doerig C, Schultz P and Meijer L, Intracellular targets of cyclin-dependent kinase inhibitors: identification by affinity chromatography using immobilised inhibitors, Chem Biol 7: pp411-422, 2000
  7. Rix U, Hantschel O, Dürnberger G, Remsing Rix LL, Planyavsky M, Fernbach NV, Kaupe I, Bennett KL, Valent P, Colinge J, Köcher T and Superti-Furga G, Chemical proteomic profiles of the BCR-ABL inhibitors imatinib, nilotinib, and dasatinib reveal novel kinase and nonkinase targets, Blood 110: pp4,055-4,063, 2007
  8. Gingras AC, Gstaiger M, Raught B and Aebersold R, Analysis of protein complexes using mass spectrometry, Nat Rev Mol Cell Biol 8: pp645-654, 2007
  9. Bantscheff M et al, Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors, Nat Biotechnol 25: pp1,035-1,044, 2007
  10. Fields S and Song O, A novel genetic system to detect protein-protein interactions, Nature 340: pp245-246, 1989
  11. Licitra EJ and Liu JO, A three-hybrid system for detecting small ligandprotein receptor interactions, PNAS 93: pp12,817-12,821, 1996
  12. Baker K, Sengupta D, Salazar- Jimenez G and Cornish VW, An optimized dexamethasonemethotrexate yeast three-hybrid system for high-throughput screening of small molecule-protein interactions, Anal Biochem 315: pp134-137, 2003
  13. Becker F, Murthi K, Smith C, Come J, Costa-Roldán N, Kaufmann C, Hanke U, Degenhart C, Baumann S, Wallner W, Huber A, Dedier S, Dill S, Kinsman D, Hediger M, Bockovich N, Meier-Ewert S, Kluge AF and Kley N, A three-hybrid approach to scanning the proteome for targets of small molecule kinase inhibitors, Chem Biol 11: pp211-223, 2004
  14. Dirnberger D, Unsin G, Schlenker S and Reichel C, A small-moleculeprotein interaction system with splitubiquitin as sensor, Chem Bio Chem 7: pp936-942, 2006
  15. Caligiuri M, Molz L, Liu Q, Kaplan F, Xu JP, Majeti JZ, Ramos-Kelsey R, Murthi K, Lievens S, Tavernier J and Kley N, MASPIT: three-hybrid trap for quantitative proteome fingerprinting of small molecule-protein interactions in mammalian cells, Chem Biol 13: pp711-722, 2006

Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:

There are no comments in regards to this article.


Mandana Rezwan joined Dualsystems Biotech AG in 2007 where she has established the yeast three-hybrid platform. Since 2009 she has been responsible for the research and development department at Dualsystems. She gained her PhD at the University of Zurich where she investigated the disease tuberculosis. She studied Applied Biology at the ETH Zurich where she also carried out her diploma thesis focusing on screening assays.

Lukas Baumann, a research associate at Dualsystems Biotech AG, has been with the company for two years supporting various projects in the research and development department. Lukas Baumann graduated from the Zurich University of Applied Sciences in 2008 as an Engineer in Biotechnology, focusing on the expression of soluble- and membrane-proteins in Pichia pastoris during his diploma thesis. Before his studies, Lukas did an apprenticeship as laboratory technician.

Daniel Auerbach joined Dualsystems Biotech in 2001 and was responsible for the development of several proprietary screening technologies, including the DUALhybrid, DUALmembrane and DUALhunter platforms. In 2006, he became CEO of Dualsystems Biotech. He obtained his education as a molecular biologist and biochemist from the Federal Institute of Technology (ETH) in Zurich, Switzerland, where he also performed his PhD studies on the topic of cardiac muscle biology.

Mandana Rezwan
Lukas Baumann
Daniel Auerbach
Print this page
Send to a friend
Privacy statement
News and Press Releases

Datwyler to Double Capacity at its Indian Facility

Datwyler is increasing capacity for all three of its production standards in Pune, India. The expansion will nearly double capacity and add 400 jobs.
More info >>

White Papers

Another Successful Visit Against the Odds

Medical Research Network (MRN) Limited

Our dedicated team of nurses, couriers and project managers deliver, come rain or shine. Making sure a Clinical Trial visit can go ahead in the home can be a complex process, and last month bad weather nearly stopped play on an important Home Trial Support visit. Thanks to the commitment and dedication from the entire team it went ahead against the odds.
More info >>




©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement