3D insights into the molecular teamwork in biomembranes
For chemists, cellular biomembranes are hard nuts to crack. It is difficult to analyze proteins that are firmly anchored in biomembranes using standard biochemical methods; and it is even more difficult to investigate their three-dimensional structure and interaction with other proteins. A group of researchers led by Prof. Dr. Anne S. Ulrich at the Karlsruhe Institute of Technology (KIT) have developed a method that enables them to take a close look at individual atoms, and even at the atoms’ natural environment in the lipid bilayer. How do antimicrobial peptides act on the membrane of a bacterium? How does a virus manage to penetrate the envelope of the host cell? How is the information from hormonal signals translated into the cell’s interior? Can this knowledge be used to produce new substances whose molecular properties can be accurately defined, for example drugs for the treatment of cancer and infections?
Around thirty per cent of all cellular proteins are in some way or other associated with a biomembrane – some proteins cross the lipid bilayer once and some even several times while others are bound to membranes and protrude into the extra- or intracellular space. The pharmaceutical industry has long known that cellular membranes are important drug targets. Aspirin acts on a specific membrane-associated protein, thereby preventing the transmission of pain signals, to cite just one example. Researchers are currently particularly interested in the structure-function relationship of membrane proteins and membrane-active peptides. Their targets include ion channels, signal receptors, peptide antibiotics as well as cell penetrating peptides that are used to shuttle drugs across the cell membrane. “In all these cases, we are not only interested in the three-dimensional structure of these molecules; we are also interested in how these molecules interact with each other,” said Prof. Dr. Anne S. Ulrich from the Institute of Biological Interfaces 2 (IBG-2) at the Karlsruhe Institute of Technology (KIT). “We are looking at the molecules when they bind to membranes; we specifically manipulate the molecular contact areas and try to deduce their mechanisms of action from this information.”
Research group led by Prof. Dr. Anne S. Ulrich (in the front, wearing a white jacket). (© Prof. Dr. Anne S. Ulrich)
A special trick
Molecules located on or in a membrane are difficult to analyze. Intractable forces prevail in lipid bilayers, which chemists who want to isolate individual molecules need to overcome. However, the three-dimensional structure of a membrane protein that is investigated in an aqueous solution provides little information about its structure in the natural environment where it interacts with lipids and other proteins in a way that is difficult to predict. Crystallographic methods only provide results for molecules that are removed from the lipid bilayer and are present as crystals. “This is why around 15 years ago I started developing a different method that was specifically applicable to membrane-active peptides and membrane proteins,” said Ulrich. “This method, which is known as solid-state nuclear magnetic resonance spectroscopy (NMR), enables us to look at the accurate three-dimensional structure of proteins in their natural membrane environment.” The researchers use a special trick, namely fluorine substitutes, to do this. Fluorine is not found in biological tissue, which is why the NMR analysis cannot be mislead by natural abundance backround signals. The sensitivity of the method is extremely high and the excellent resolution enables the specific visualization of the fluorine-labelled molecular segments.
Ulrich and her team now use this method routinely. If everything goes to plan, they are able to resolve the three-dimensional structure of a peptide molecule in a lipid bilayer within six weeks, compared to the early days when it took them several years to obtain a result. The fluorine-NMR approach is very useful for analyzing the structure-function relationships of membrane-active peptides. For example, Ulrich’s team managed to find out how antimicrobial peptides, which are present in the skin of frogs and in human sweat, attack the cell membrane of bacteria. In contrast to the majority of antibiotics developed by the pharmaceutical industry, the antimicrobial peptides examined by Ulrich and her group accumulate on the cellular membrane of a pathogen and form a pore – the bacterium is perforated and its contents leak out. This mechanism targets the physical properties of the membrane, which prevents the bacterium from adapting to the altered situation by evolving mutations and resistance to the antimicrobial agent. This mechanism appears to be a promising new approach in the development of new antibiotics. Using solid-state NMR, Ulrich’s team was able to determine the molecular architecture of different peptides. This now enables them to explain how the molecules group together and form pores. The peptides are highly selective and do not attack the host’s own cell membrane.
Antimicrobial peptides (e.g. gramicidin S) form pores in bacterial membranes and ensure that the cellular content of the microorganisms flows out. (© Prof. Dr. Anne S. Ulrich)
It is interesting to note that the chemical composition of some antimicrobial peptides is similar to that of so-called fusion peptides. Viruses that attack a host cell catapult these fusion peptides into the membrane of the host cell where they lead to the fusion of the two membrane envelopes. Comparative NMR measurements have shown that both types of peptides can induce similar mechanical disturbances in the lipid bilayer when they engane in unspecific interactions. However, nature has optimized the antimicrobial molecules in a way that does not normally happen; a few molecular “missiles” can induce the controlled formation of pores. A fusion machinery, on the other hand, activates these weapons by way of a multi-tier cascade as they would otherwise stick to each other in an uncontrolled way. If these fusogenic peptides were given free rein they would form fibrils similar to those found in the tissue of Alzheimer’s patients, and immediately destroy the membrane.
Signal transduction caught in the act
The model shows a dimer of the E5-oncoprotein from papillomavirus as it is embedded in a lipid membrane. The E5 dimer is able to bind to a transmembrane hormone receptor, activate it and thereby lead to uncontrolled cell profileration (i.e. cancer), which is usually strictly controlled by hormones. (© Prof. Dr. Anne S. Ulrich)
Another example from the field of signalling is focussed on receptor proteins that control the transmission of information between cells. These receptor proteins are anchored in the cellular membrane by way of one or several transmembrane segments. If an exogenous hormone docks to the cell membrane, two receptor proteins assemble in such a way that the regions that protrude into the cytosol are able to interact with cellular signalling molecules which then transfer the signal to the cell nucleus or to other molecular control centres. What are the structural interactions that occur in the transmembrane regions of the receptor pair when a ligand has bound and the information is transmitted across the membrane? Ulrich and her team are currently investigating a receptor that besides being switched on by a hormone, but it can also be switched on by the small papillomavirus oncoprotein E5. The interaction with E5 activates a signalling cascade that causes a cell to grow in an uncontrolled manner, as in cancer. “Crystallographic methods are not suitable for finding out which areas on E5 and on the receptor interact with one other and what happens spatially,” said Ulrich. Using a combination of different NMR methods, the researchers succeeded in clarifying the molecular composition of E5 and of the receptor in its membrane-bound state. Each is preferentially present as a dimer. The next step will be to catch the mixed proteins in the act, i.e. the point at which E5 binds to the membrane segments of the receptor, thus imitating the situation in which an exogenous hormone docks to the membrane.
In addition to solid-state NMR, Ulrich and her KIT group have recently installed a circular dichroism beamline in the KIT’s ANKA synchrotron facility. This particle accelerator generates high energy radiation in the UV range, which is used for the qualitative clarification of molecular structures. This device leads to far better results than any other commercial laboratory device. “We use the device for very basic experiments in order to investigate whether and how a protein segment is folded in a membrane, or whether it has aggregated and has lost its function,” said Ulrich. The subsequent solid-state NMR measurements of the same samples are highly accurate; individual atoms and even the atoms’ movements can be visualized. Solid-state NMR is the method of choice for researchers who want to find out step by step what is actually happening in and around a membrane.
A contribution from:
Prof. Dr. Anne S. Ulrich
Institute of Biological Interfaces (IBG-2)
Tel.: +49 (0)721/ 608 23 201
Fax: +49 (0)721/ 608 24 842