GEN Exclusives

More »

Feature Articles

More »
Mar 15, 2014 (Vol. 34, No. 6)

Protein-Specific Discovery Strategies

  • Click Image To Enlarge +
    Researchers at Stanford University used NMR to study the beta-2 adrenergic receptor. This image, which was produced using the program MacPyMOL, shows the receptor embedded in a lipid bilayer and in complex with the G protein Gs.

    One of the most widely used experimental approaches in virtually every life sciences laboratory, protein expression has witnessed, particularly in recent years, transformative changes that were greatly catalyzed by advances in biotechnology.

    Depending on several factors, including individual characteristics and downstream applications, proteins can be generated in various settings, most frequently involving bacterial, eukaryotic, and cell-free expression systems.

    Proteins represent one of the major therapeutic targets. Accordingly, the strategies that are used to characterize these targets influence drug discovery and development efforts.

    “It is beneficial to use a number of different biophysical approaches to characterize protein structure to be confident that one can obtain relevant structural information that is not adversely influenced by a particular method,” says Brian K. Kobilka, M.D., professor of molecular and cellular physiology and medicine at Stanford University School of Medicine and co-recipient of the 2012 Nobel Prize in Chemistry.

    In a recent study, Dr. Kobilka and colleagues used NMR to study the beta 2-adrenergic receptor, a prototypical G-protein-coupled receptor. Approximately 40–50% of the existing therapeutic agents target GPCRs, the most diverse group of eukaryotic membrane receptors, making them occupy a particularly important position in the therapeutic arena.

    By using NMR spectroscopy to examine the effect of different drugs on receptor structure, Dr. Kobilka and colleagues unveiled a significant conformational flexibility that exists particularly in the agonist-bound receptor, pointing toward the importance of studying protein dynamics for a better understanding of the events shaping signal transduction. “The most challenging aspect is finding ways to express and prepare labeled G-protein-coupled receptors for NMR studies, so that we can learn about the timescales of conformational changes,” notes Dr. Kobilka.

    Structure-function relationship studies are particularly difficult for membrane proteins, which have lagged in terms of X-ray crystallographic characterization, and for which NMR approaches also open challenges. “We learned quite a bit about proteins by using fluorescent techniques even before crystal structures became available, and these studies ultimately informed our approach to generating crystals, but having the crystal structures now helps us interpret fluorescence, NMR, and other biophysical experiments, and they are all powerfully complementing each other,” remarks Dr. Kobilka.

    Part of the difficulty in understanding the structure-function relationships of GPCRs stems from their highly dynamic behavior, the complex environment within the phospholipid membrane bilayer where their biology has to be captured, and the associated signaling proteins that are also important for their activity.

    “We continue to explore not only NMR but also other approaches, including fluorescence, single-molecule methods, and electron paramagnetic resonance spectroscopy, and one aspect that we are really interested in is understanding the role of lipids,” concludes Dr. Kobilka.

  • Preserving Folds

    “Our particular interest is focusing on the cannabinoid receptor in the context of drug abuse, but this particular receptor is also involved in inflammatory disorders, spinal cord injury, brain cancer, and trauma, opening a large array of possible applications,” says Alexei Yeliseev, Ph.D., staff scientist at the NIH. The cannabinoid receptor, a GPCR superfamily member, is expressed at low levels in mammalian tissues.

    “For functional studies we are using bacterial expression systems,” adds Dr. Yeliseev. “We are now able to produce several milligrams of pure protein at a time.”

    Dr. Yeliseev and colleagues are relying on nuclear magnetic resonance, fluorescence microscopy, surface plasmon resonance, and labeling techniques for functionally characterizing the cannabinoid receptor. One of their recent findings is that the lipid bilayer affects the function of the protein. This result suggests that the receptor will have to be maintained in an environment that is as close as possible to its natural environment, which is the lipid membrane.

    “We need to incorporate these proteins into the membrane, and use particular combinations of buffers and detergents, in addition to lipids and cholesterol, to maintain them properly folded,” observes Dr. Yeliseev. Once the receptor unfolds, it is extremely difficult and sometimes almost impossible to properly refold it again. “This is why it is important to preserve the conformation as much as one can,” concludes Dr. Yeliseev.

  • Emulating Native Expression

    Click Image To Enlarge +
    Lab technician using a pipette in preparation for carrying out a protein expression experiment. [Horsche/iStock]

    “Our work has shown that the way we manufacture a specific protein that is used for screening assays has a big impact on the activity of that protein,” says Jeffrey R. Peterson, Ph.D., associate professor at the Fox Chase Cancer Center. Producing proteins in their most physiologically relevant context is essential when screening for drugs intended to target full-length or native proteins.

    Even though bacterial systems are well established, they are not ideal for expressing many eukaryotic proteins. Also, with bacterial systems, protein yield and activity are often suboptimal. However, insect or mammalian expression systems allow native or native-like proteins to be expressed.

    “A significant challenge, when expressing proteins in nonnative heterologous systems, is that the regulatory aspects may not be recapitulated as in their native source,” comments Dr. Peterson. For example, certain post-translational modifications that are critical for function do not take place in bacterial expression systems.

    Proteins may also have different degrees of functionality, depending on the expression system and the source they were isolated from. “This can happen either because the protein does not fold properly to its native state, or it folds but it lacks a cofactor that in the native environment promotes its activity,” notes Dr. Peterson.

    Even proteins purified from the same expression system may show differences in their activities, as revealed by a recent project in Dr. Peterson’s lab that involved two recombinant proteins, the insulin receptor and the related insulin-like growth factor 1 receptor, obtained from two commercial sources. The proteins obtained from the two sources showed significantly different degrees of activity.

    “We ultimately understood that the protein produced by one source was phosphorylated, promoting catalytic activity, while the protein from the other source was not phosphorylated and initially had much weaker activity,” reports Dr. Peterson. Using both forms to screen for inhibitors, Dr. Peterson and colleagues found an inhibitor that was selective for the nonphosphorylated form. This compound bound the ATP-binding pocket of the unphosphorylated, inactive receptors, and stabilized an activation loop in the inactive conformation.

    “Had we only used phosphorylated forms of these proteins, we would never have identified that inhibitor,” emphasizes Dr. Peterson. Taking the point even further, Dr. Peterson and colleagues realized that their work pointed to a new concept. They showed that the same compound binds the ATP-binding pocket of the ERK serine/threonine kinase in a different conformation, providing an example of a single compound targeting distinct conformations on several unrelated kinases.

Related content


GEN Jobs powered by connects you directly to employers in pharma, biotech, and the life sciences. View 40 to 50 fresh job postings daily or search for employment opportunities including those in R&D, clinical research, QA/QC, biomanufacturing, and regulatory affairs.
More »

Be sure to take the GEN Poll

Disclosing Clinical Trial Data

Do you believe that conducting performance audits for new drugs will incentivize pharma firms to become more transparent?

More »