Protein Interplays

Published in Lab Times 04-2013.

jpg 1

X-ray crystallography captures the atomic level details of protein-protein interactions. Shown here, ERK5 kinase domain (blue) binds to the MKK5 activator kinase construct comprised of its PB1 domain (orange) and a MAPK binding linear docking motif (yellow)

Attila Reményi has an eye for detail. Equipped with the latest proteomics tools, he enjoys visualizing protein-protein interactions at an atomic scale. When you talk to him, you soon realize that he is, in fact, taking a major stride in drug discovery.

Some 4000 years ago, Mesopotamians who inhabited the region around the Persian Gulf, made their first records in medicine. Their scriptures describe about 1000 plants and plant-derived products of therapeutic potential. Another 1500 years later, ‘practitioners’ of the Indian subcontinent documented plant and animal products with medicinal properties and categorized them into eight classes based on the disorders that the ‘drugs’ alleviated, like gastric, paediatric, reproductive and so on. The same drug was often prescribed for a range of illnesses, in other words, the drugs had a ‘broad-spectrum’. The doctors back then barely had a microscope to test biological specimens for the efficacy of the drugs; let alone customizing drugs for specific ailments. But as eons passed, unnamed and complex diseases crept into our medical diction. Drug discovery research soon leapt forward to face the challenges of increasing ailments.

The advent of techniques such as X-ray crystallography has made it possible to decipher the atomic structure of molecules such as proteins, vitamins and DNA and even watch the interplay of biological molecules in a test-tube. Resolving the action that takes place within cells brings us a big step forward in formulating drugs that can target molecules with a high degree of specificity. Attila Reményi, Professor at the Department of Biochemistry, Eötvös Loránd University, Budapest has been using this technique extensively to study the molecular basis of specific protein-protein interactions. His work on the specificity of MAP kinase (MAPK) signalling, a central signalling pathway in cells, has been recently published in the Journal of Biological Chemistry (J Biol Chem vol.288(12):8596-8609).

Crystallizing his goals

During his PhD years that Attila spent in the lab of Hans Schöler and Matthias Wilmanns at EMBL in Heidelberg, Germany, he mastered the intricacies of X-ray crystallography. In this technique, a beam of X-rays impinge on crystals of molecules such as proteins or DNA, flash-frozen either alone or in complex with other binding partners. The X-rays diffract along different directions as they encounter the crystal’s atoms, much the way a headlight beam is diffracted by dust particles in the air. As the X-rays fly off in definite directions and at definite angles, this information can be used to predict the position of the atoms in three-dimension and to reconstitute the crystal structure of the sample. “Today high quality data collection is feasible even for small crystals and there’s been a huge advance in making software for X-ray structure determination more user-friendly”, the Hungarian speaks high of his favourite technique. But he cautions, “Of course, we are talking about in vitro work here; and of artificial systems. We are crystallizing proteins in milligram amounts, and we are doing it under non-physiological conditions”.

In his graduate years, Attila resolved the structure of transcription factors as they are assembled on regulatory elements on DNA, called enhancers. When he moved to the University of California San Francisco, California for his postdoctoral research, his interests crystallized on MAPK signalling. “MAP kinases are ubiquitous proteins capable of interacting with hundreds of other proteins. By doing so, they engage in functional signalling networks to regulate many aspects of cellular life such as cell growth and division, survival and differentiation”, elaborates Attila.

No promiscuity business…

MAPKs are enzymes that catalyze the transfer of phosphate groups to proteins to convey their signals forward. They fall into several classes – p38s, JNKs, ERK1/2, ERK5, among others, and are activated by upstream MAPK kinases (MKK). The present-day MAPKs have evolved from gene duplication events of an ancestral sequence and hence, have a high degree of homology. “Paradoxically, the MAPKs have varied roles and exhibit very little functional redundancy”, Attila comes to a crucial point. How do paralogous enzymes that have a common descent and similar catalytic properties perform diverse physiological functions?

Like all other enzymes, MAP kinases have an active site which participates in catalysis but Attila’s interests were drawn to a ‘docking groove’ outside the active site that binds docking motifs (D-motifs) on interaction partners. “MAPKs have somewhat similar binding surfaces but it is the nature of the D-motif on the targets that confers specificity to the interactions”, he explains. In an earlier study (Sci Signal vol.5(245):1-13), the Reményi lab looked closely at the D-motifs on target proteins in complex with their specific MAPK partners, p38s, JNKs or ERK1/2.

A typical D-motif has the following sequence: ψ1-3-X2-5-φ-X-φ, where ψ is basic, φ is hydrophobic and X is any amino acid.

“The D-motifs on different MAPK binding partners have some common or ‘consensus’ amino acids on both ends of their sequence (ψ and φ-X-φ) called ‘anchor’ points, which make rather non-specific contacts on MAPKs docking grooves. What then defines the specificity of the interaction is the length of the amino acids that intersperse these anchor points (X2-5) and the changes in conformation that arise thereof”, describes Attila. The different MAPK interactors have different D-motif linker lengths and hence, differ in the three-dimensional folding in this region. This fine-tunes the binding of targets to the cognate MAPKs, and ensures little promiscuity in binding and no redundancy in function.

…even among namesakes

Even though the specificity factor is the D-motif on targets, we were talking about three different MAPKs above, with some distinguishing structural features. But what about ERK1/2’s closest relative ERK5, that has very similar domains?

Attila was intrigued by the fact that even ERK1/2’s closest homolog ERK5 has very diverse roles. He justifies his curiosity, “in contrast to the generic mitogenic role of MAPKs, ERK5 has a very specific function in maintaining blood vessel integrity and so is of clinical importance in cancer metastasis”. Investigating around this topic brought the Reményi lab to publish their recent paper. Using X-ray structure determination, they unveiled that despite its high homology with ERK1/2, ERK5 has a topographically different protein-protein interaction surface. “This extended surface accommodates a second domain together with the linear D-motif in the ERK5 interacting partner MKK5, making it a high-affinity binding”, explains Attila. MKK5 binds to ERK5 more tightly as opposed to ERK1/2 owing to an additional PB1 domain that enhances the binding apart from its D-motif.

New leads for drugs

“Such structural studies come in handy when you want to design peptide-based tools to specifically ablate the activity of a MAPK”, remarks Attila as he discusses the use of his findings in drug preparation. He brings our attention to the large family of kinase inhibitors in the market. Owing to their crucial role in carcinogenesis, besides muscle and renal disorders, kinases are among the most targeted proteins for drug discoveries.

Attila contrasts two strategies to develop drugs that target kinases, “to interfere with the activity of a kinase, one can design inhibitors that compete with ATP – a cofactor needed for all kinases – or you could target its protein-protein interactions which give it a more specific signalling identity”. The challenge here is that the ATP-binding pocket on kinases is deeper and easier to target than the shallow crevices to which interaction proteins bind, but “the second strategy is conceptually better if you need specific inhibitors”. Of course, the whole drug deal is to minimize side-effects, so we are striving towards making drugs as specific as possible no matter what it takes to get there.

Way to go!

At the end of our little chat, Attila Reményi highlights the implications of his research, “focused and mechanistic biochemical studies, which may appear to decipher a seemingly tiny fraction of the molecular world around us, actually helps us to understand how specific systems work in detail and to eventually formulate hypotheses involving bigger systems”. Though the Reményi lab concentrates on minute structural details, they are in fact making a huge point in discerning MAPK signaling pathways – how these kinases have functionally diverged and how specific their activities are. And needless to say, their structured interests can go a long way in shaping drug discovery research.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: