Backscattering Interferometry - Shining Light on Protein-Ligand Interactions
By: Carol A. Rouzer, VICB Communications
Published: March 28, 2011
VICB investigator Darryl Bornhop demonstrates a new method for monitoring the interactions of membrane proteins with their environment.
The vast majority of modern drugs act by stimulating or inhibiting the activity of cellular proteins. These proteins may be enzymes, which control the biochemical reactions required for life, or receptors, which bind and transmit the messages of signaling molecules that cells use to communicate with their environment. Over 70% of proteins that serve as drug targets are associated with cellular membranes. However, investigation of the interaction of drug candidates with proteins in the complex membrane environment can be particularly challenging. The presence of the multitude of proteins and lipids in cell membranes can make data interpretation difficult, while attempts to simplify the system by removing the protein from the membrane can lead to alterations of protein function. Nearly all assays of protein-drug interactions require that the drug be labeled in some fashion, which further complicates the process. Now, VICB member Darryl Bornhop and his laboratory, in collaboration with the Michael Baksh of the Scripps Research Institute (La Jolla, CA) and his laboratory report a new method, called backscattering interferometry (BSI), for assessing the interaction between membrane-associated proteins and other molecules [Baksh et al. (2011) Nat. Biotech. published online March 13, DOI: 10.1038/nbt1790].
Figure 1. (a) Diagram of BSI instrumentation. Light from an HeNe laser is reflected by a mirror to the sample in the microfluidic chip. The pattern of light resulting from passage through the sample in the chip is reflected back to the mirror and then to a CCD camera for recording. (b) Binding studies are carried out using small unillammelar vesicles (SUVs) comprised of a lipid bilayer membrane to which the protein of interest is associated. Reproduced by permission from Macmillan publishers from Baksh et al. (2011) Nat. Biotech. published online March 13, DOI: 10.1038/nbt.1790. Copyright 2011.
BSI is based on the principle of light interference. Whenever light passes through matter, its speed is slowed, leading to a “bending” of the light. The refractive index of a medium, the ratio of the speed of light in a vacuum to its speed in the medium, is one measure of light interference. Refractive index is a unique quality of any form of matter, indicating that interference is highly dependent on the composition of a substance. The Bornhop lab has found that changes in the association of particles or formation of a new species in solution can have a significant effect on the interference properties of that solution. They have exploited this observation by designing a microfluidic chip containing a narrow channel that allows a laser beam to be reflected multiple times as it passes through a sample in the channel. The long path length through the channel magnifies the effects of any interference changes. A HeNe laser (similar to those found in barcode scanners) serves as the source of light, which is directed to the chip by use of a mirror. After passing through the sample, the light leaves the channel in the form of a scattered fan pattern. The pattern of the reflected light is transferred by the mirror to a high-resolution linear charge-coupled device (CCD) similar to that found in a digital camera (Figure 1a). The pattern of light consists of a set of high contrast “interference fringes”. Mathematical analysis of these fringes reveals changes in the interference properties of the sample that result from alterations in the components of the sample - in this case, the interaction of the membrane protein with a ligand.
To assess the versatility of BSI, the collaborators tested its ability to detect protein-ligand interactions in a number of different environments. GM1 is a glycosphingolipid (Figure 2) that is found on the cell surface. Although well recognized for its role in modulating nerve cell function, GM1 also serves as the point of attachment for cholera toxin. This protein from the bacteria Vibrio cholerae causes the debilitating diarrhea that occurs in cholera. The Baksh lab incorporated GM1 into synthetic membranes made from a mixture of phosphatidylserine and phosphatidylcholine. Dispersal of the membranes in the form of small unilamellar vesicles (SUVs) facilitated analysis by BSI (Figure 1B). Indeed, the investigators found that BSI readily detected the interaction between cholera toxin and GM1 in the synthetic membranes, while tetanus toxin, which does not bind to GM1 produced no signal in the assay.
Figure 2. Structure of the GM1 ganglioside. Image courtesy of Wikimedia Commons under the GNU Free Documentation License.
The enzyme, fatty acid amide hydrolase (FAAH, Figure 3), hydrolyzes the lipid arachidonoyl ethanolamide (anandamide). Anandamide is an endogenous ligand for the cannabinoid receptors, which mediate the pharmacologic effects of marijuana. FAAH ends the activity of anandamide in vivo, and there is currently considerable interest in discovering inhibitors of FAAH in order to prolong anandamide’s anti-inflammatory and analgesic effects. Following incorporation of FAAH into synthetic membranes, the investigators could easily detect the binding of three candidate FAAH inhibitors to the enzyme. Interestingly, the data suggested that binding occurs at a lower concentration than actual enzyme inhibition, a finding that reflects the fact that inhibition requires steps beyond binding alone. Relatively few methods are available that directly assess inhibitor binding in the absence of inhibition, so BSI offers a valuable new tool to fill that gap in our knowledge.
Figure 3. Crystal structure of FAAH. Image courtesy of Wikimedia Commons and the European Bioinformatics Institute under the GNU Free Documentation License.
CXCL12 is a chemoattractant peptide for lymphocytes. Its receptor, CXCR4 (Figure 4), also serves as a binding site by which the human immunodeficiency virus enters and infects lymphocytes. Antagonists of CXCR4 are of interest medically, because they increase the number of stem cells in the blood stream. Thus, such inhibitors are valuable as a pretreatment for donors in the case of hematopoietic stem cell transplants. In order to study the CXCR4-CXCL12 interaction, the investigators used SUVs derived from the membranes of SUP-T1 lymphoma cells, which naturally express the receptor. Despite the complexity of this membrane preparation, binding between CXCR4 and added CXCL12 was readily detected. Membranes from control cells that did not express CXCR4 failed to exhibit any interaction with CXCL12, verifying that the observed binding was specific for CXCR4.
Figure 4. Crystal structure of CXCR4. Image courtesy of Wikimedia Commons under the GNU Free Documentation License.
Encouraged by their success with CXCR4, the investigators turned to an even more complex receptor system. Gamma-aminobutyric acid (GABA) is the primary inhibitory neurotransmitter in the central nervous system. There are two receptors for GABA. The GABAB receptor is composed of two subunits, B(1b) and B2. The investigators expressed both of these subunits in CHO cells, and then prepared SUVs from the CHO cell membranes. They were able to demonstrate binding of GABA, two receptor agonists, and one receptor antagonist in this complex system. No binding occurred to membranes of cells that did not express the receptor.
Together the results clearly demonstrate the power and versatility of BSI for detecting protein-ligand interactions in a complex membrane environment. BSI is highly sensitive and requires only minute amounts of material. The equipment required is relatively simple and inexpensive. We look forward to seeing additional applications of BSI to the study of protein-ligand interactions in the near future!