Looking at variations in membrane composition is of interest to structural biologists. However, investigating lateral variations caused by interactions of membrane components — which are believed to play a crucial role in a range of cellular events — has proved challenging, as achieving the necessary length scales of tens to hundreds of nanometers can be difficult with the technologies currently used for such probes. These lengths of interest are well below the diffraction limit of light microscopy. Researchers often use fluorescence microscopy to explore labeled components, but the labeling process can alter significantly the physical properties of molecules (especially those of small molecules, such as lipids, that interact weakly). And although atomic force microscopy can provide much higher resolution, it does not shed any light on the chemical composition of the membranes. The lateral variations posed a problem to a group at Stanford University in California: How do you image and acquire compositional information about these variations with a higher spatial resolution than is available with conventional light microscopy, with high sensitivity and in a nonperturbative way? This led the researchers to various kinds of imaging mass spectrometers and, specifically, to secondary ion mass spectrometry, principal investigator Steven G. Boxer said. The scientists reported their use of the latter technique to analyze phase separation of lipid membranes in the Sept. 29 issue of Science. Secondary ion mass spectrometry could help shed light on a variety of questions; e.g., whether reorganization of membrane components associated with one another occurs collectively. In addition,researchers at Stanford University hope to improve the sensitivity of the technique to investigate membrane-anchored proteins and, perhaps, to detect lateral variations of individual labeled proteins in real cell membranes trapped on substrate surfaces. With secondary ion mass spectrometry, a focused primary ion beam is rastered across a sample. Sputtering produces secondary ions. These are collected, and the respective charge-to-mass ratios are analyzed. Users can acquire secondary ion images characteristic of individual membrane components by incorporating distinctive stable isotopes into them. From these they can derive a component-specific compositional map of the sample. The technique was developed — and is typically used — for the study of hard materials such as minerals (as opposed to soft materials such as cellular membranes). Generally, it uses a number of types of beams — often pulsed ones, as are used for time-of-flight secondary ion mass spectrometry. For this study, the researchers used a NanoSims 50 secondary ion mass spectrometry system made by Cameca Instruments of Courbevoie, France. “The NanoSims is quite special,” Boxer said, “as it uses a continuous ion beam made of cesium ions that can be very tightly focused.” This provides very high spatial resolution and enables rather complete destruction of the molecules on the surface of the sample because of the energy and nature of the beam. Conversion to secondary ions with high yield enables high sensitivity. Lawrence Livermore National Laboratory in Livermore, Calif., maintains the secondary ion mass spectrometry system, which is used by a variety of labs for study of dust collected from comets, among other applications. Fortunately, Boxer said, some scientists who otherwise focus on hard materials were willing to take a gamble with respect to collaborations on investigations of soft materials and biology, setting the stage for broad application of the existing technology to a whole new class of problems. They used the spectrometry system to measure the chemical composition of a phase-separated lipid bilayer, relying on isotopic labels to identify each molecular species. The bilayer contained gel and fluid phases, made up primarily of 13C18-DSPC (a phospholipid with long stiff chains) and 15N-DLPC (a phospholipid with shorter chains), respectively. They used 13C1H— and 12C15N— secondary ion signals to determine the distribution of each of these, enabling construction of a component-specific compositional map of the bilayer, which showed that it was not homogeneous. To validate their findings, they characterized the lipid bilayer with atomic force microscopy prior to analysis with secondary ion mass spectrometry. Comparison of the images revealed that the geometries were almost the same. In fact, the secondary ion images included phase-separated domains with complex structures and domains, on the order of approximately 100 nm in diameter according to the atomic force microscopy images. Beyond this, the study showed that the scientists could quantify the lipid composition within small areas of the bilayer — helping to determine whether the variations in 15N-DLPC distribution, for example, were statistically significant. Having demonstrated the feasibility of measuring lateral variations in membrane composition with secondary ion mass spectrometry, the researchers plan to return to the question that led them to the technique: whether membrane components that are associated with one another — in lipid rafts, for example — reorganize collectively. “We’re sort of looking up the ladder of complexity to see how far we can push the sensitivity issue,” Boxer said. They plan to look at membrane-anchored proteins next. “But, conceivably, we could detect down to the single protein level with appropriate labeling and move on to real cell membranes and interfaces trapped on our surfaces,” he added.