October 10, 2012 | 2
Brian Kobilka (Stanford) and Robert Lefkowitz (Duke) have won the 2012 Nobel Prize in Chemistry for their work on one of the most important classes of proteins in living organisms, the G Protein Coupled Receptors (GPCRs). A lot of us had predicted this prize based on groundbreaking work done during the last decade on several kinds of GPCRs, although personally I thought it would take a few more years for the prize to be awarded. Nevertheless, the class of proteins and their physiological importance makes this prize extremely well-deserved and this has been one of the fastest discovery-to-prize transitions in recent decades. From a chemical standpoint, the GPCRs provide a superb example of a molecular machine whose subtle workings we have only started to understand. This Nobel Prize continues the proud tradition of recognizing crystallographers who are among the most persistent and fearless of all scientists. It’s a tradition that goes back to the 1962 Nobel Prize awarded to Max Perutz and John Kendrew for their studies of hemoglobin and myoglobin.
GPCRs are essentially both the gatekeepers and molecular messengers of the cell, transmitting signals from inside to outside. The signal can consist of an astonishing variety of stimuli, from photons (light) to neurotransmitters to hormones. They mediate virtually every important physiological process, from immune system function to taste and smell to the fight-or-flight response in humans. GPCRs are also immensely important in medicine and are the target of about 30% of all drugs. Naturally occurring small molecules which bind to GPCRs include adrenaline, prostaglandins, dopamine, somatostatin and adenosine. Drug-like small molecules which bind to GPCRs include caffeine, morphine, heroin and histamine. The range of stimuli and molecules that GPCRs respond to is remarkable and their role in the workings of life is unquestioned.
Unfortunately for a long time, it wasn’t possible to study the detailed structure of GPCRs because of the great difficulties in crystallizing them. GPCRs are membrane bound proteins that span the cell membrane in the form of seven transmembrane helices which are connected by three loops, three on the intracellular side and three on the extracellular side. Most importantly, these proteins are bound on the inside to a small subunit of a G protein, a key signaling molecule that acts as a sensor and messenger to ferry signals inside the cell. Because GPCRs are membrane proteins, any attempt to take them out of the membrane would rapidly destroy their integrity; it’s like trying to study a delicate embryo by taking it out of the womb. The six loops on a GPCR are particularly floppy and it’s challenging to pin them down; they also turn out to be particularly important determinants of molecular binding. GPCRs can also exist in two states, an active state and an inactive state, and crystallizing the active state turns out to be a fearsome task because of its instability.
Molecules that bind to these proteins come in three flavors. Agonists activate the receptor. Inverse agonists completely shut it down. Antagonists prevent agonist binding but don’t shut the protein down. A small molecule like a drug binds to one of the GPCR helices on the outside and this results in a complex series of motions of the helices that results in the dissociation of one of the G protein subunits. After dissociating, the G protein can interact with a variety of other proteins, including proteins called kinases which attach and detach phosphate groups and control cell signaling. The end result of this process is usually the activation of a so-called second messenger, a small molecule like cyclic GMP (cGMP) which goes into the nucleus of the cell and brings about specific gene expression and attendant physiological responses. There’s a variety of ways in which GPCRs are regulated, activated and inactivated and G protein binding is just one of them. A recently highlighted class of proteins called arrestins for instance can deactivate GPCRs when they are overstimulated and internalize them for degradation.
However as mentioned before, studying GPCRs especially in their crystallized state was immensely challenging. Membrane proteins have always been the bane of crystallographers and GPCRs were especially recalcitrant. But beginning in the 90s, a series of breakthroughs allowed the detailed structural characterization of these proteins, and both of this year’s Nobel Laureates have made key contributions to this development. The “model organism” for GPCRs has been the beta-adrenergic receptor (BAR) which binds to adrenaline and brings about the “fight-or-flight” response among other things. Beginning in the 80s, Robert Lefkowitz pioneered the modern study of GPCRs by first cloning and sequencing the genes for the BARs. In doing this Lefkowitz made the crucial observation that all the genes were similar to those for rhodopsins, GPCRs which sense light. This established the common transmembrane structure of all GPCRs. In addition his group also established a variety of techniques for handling the BAR. Lefkowitz was also the first to identify arrestins which are responsible for GPCR deregulation and control. In a broad sense he can be considered the father of modern GPCR research.
Apparently Lefkowitz’s enthusiasm must have rubbed off on his postdoc Brian Kobilka. After beginning his independent career at Stanford, Kobilka began work that resulted in a series of tour-de-forces that gazed into the inner structure and workings of the BAR in exquisite and unprecedented detail. The structural story had started to unfold in the late 90s. In 2000, Krzysztof Palczewski became the first to obtain crystals of rhodopsin. This was a major achievement, but the structure was relatively low resolution and could not shed light on the details of GPCR activation. This task was left to Kobilka’s group.
In the last decade beginning with the first crystal in 2007, Kobilka’s group has provided an astonishing set of insights into GPCR structure through crystallography. Their workhorse has been the BAR. In 2007 they obtained a crystal structure of the BAR in its inactive state. The breakthrough became possible because of detailed manipulation, trial and error and sheer persistence (a trait shared by many protein crystallographers). Crystallography is still very much of an art and there isn’t always a rational path to crystallizing a particular protein. Kobilka’s group tried countless combinations of amino acid sequence variants of the protein with a small section deleted along with several detergents, stabilizing small molecules and proteins. The winning combination turned out to be an antibody which stabilized the GPCR when it was lifted out of the membrane. A second structure was stabilized by attaching a bacteriophage virus to it. Once these tricks were figured out Kobilka’s group ploughed ahead. Another major breakthrough came with a structure of the GPCR in its active state, a state which had previously thought to be too unstable to be isolated. The comparison of the structures in the active and inactive state provided valuable insights into the activation process, including the identity of the specific helices that move and affect G protein binding. Finally, Kobilka’s group dealt the coup-de-grace in 2011 when they published the first ever structure of a GPCR bound to a G protein. That structure has again led to unprecedented, atomic-level understanding of GPCR function including the movement of the crucial helices and loops. Kobilka’s work was also accompanied by pioneering work from Raymond Stevens’s lab at Scripps which also developed special techniques to crystallize GPCRs. All these projects are technical tour-de-forces, involving the testing of thousands of conditions, countless hours of manual labor and the delicate handling of temperamental proteins. Together Kobilka and Stevens’s labs have turned into the world’s top GPCR crystallography destinations.
These breakthroughs have opened the way to not only a detailed dissection of the mechanism of GPCR function but also to structure-based drug design. In addition they have struck a blow to the belief that GPCRs cannot be crystallized. In the last 12 years more than a dozen GPCR crystal structures have been resolved. These include the dopamine receptor, the CXCR4 chemokine receptor which is thought to be involved in HIV infection, the adenosine A2 receptor which binds to caffeine and provides our daily morning high and most recently, the key opioid receptors which bind to morphine. In fact most crystal structures of the GPCRs come with drugs attached to them and therefore they are invaluable starting points for designing more potent drugs with better safety profiles. The morphine-bound structures are particularly promising in this regard since finding a safe alternative to morphine has for decades been a kind of medical crusade with many corpses but no holy grail in sight.
But most importantly, as with any other major discovery, Kobilka and Lefkowitz’s work asks more questions than it answers and points to exciting, uncharted territory. GPCR function has turned out to be much more subtle than we thought. The most significant discovery in this context has been the observation that similar small molecules – two agonists for instance – can nonetheless activate the proteins in different ways and lead to very different physiological responses, a phenomenon called “functional selectivity”. This results from the differential interaction of GPCRs with G proteins, arrestins and other proteins in the interior of the cell. Remarkably, functional selectivity can even lead to the exact same molecule acting alternatively as an agonist or antagonist, depending on the physiological response that is being mediated by the protein. Thus, GPCR function is more like Nature’s Bach symphony than a set of independent notes where it’s combinations of interactions rather than the binding of a single molecule that result in a complex physiological response. The work done by Kobilka, Lefkowitz and others opens the door to understanding how this symphony is exactly conducted.
Addendum: I have to say that the whole “But is this chemistry?!” meme is getting quite boring. Binding of a small molecule to a GPCR is as much of a molecular interaction as anything in chemistry. Plus, think about the downstream chemistry that GPCRs do, including phosphorylation and salt-bridge breakage. I thought chemists were supposed to rub their hands with glee at the reduction of biology to chemistry while biologists fret and fume. But I see the opposite, biologists being quite sanguine about proteins being awarded medicine Nobels while chemistry continue to complain about proteins (chemicals!) being awarded chemistry Nobels. As I have said before though, this very bickering shows the astonishing reach and diversity of the field. If you can’t even agree on a definition for your field, well, that means your field is truly omnipresent.