Single-molecule fluorescence microscopy offers numerous approaches for measuring the spatial arrangement of biological structures. For example, single-molecule Forster resonance energy transfer (smFRET) can read out absolute and relative distances, and advanced localization microscopy methods offer spatial resolution of around 5 nm for examining structural organization. Despite the power of these approaches, better methods are still needed for measuring distances beyond the ~8-nm limit of smFRET that do not require very sophisticated microscopes.

To address this challenge, Ronald Vale and Nico Stuurman at the University of California, San Francisco, along with graduate student Stefan Niekamp and colleagues, developed an optimized approach for determining the distance between two different-colored fluorophores using images acquired on a total internal reflectance (TIRF) microscope. The first step involves dual labeling of the molecule of interest, followed by immobilization on a surface and imaging in both colors. In addition to sample imaging, beads are imaged as fiducial markers to generate a nonuniform local transformation map, which allows for sub-nanometer color registration. This registration step is crucial for accurate determination of the distance between the fluorophores.

From there, the authors developed two methods for accurate determination of distances that involve optimized fitting of the distance measurements. The first, Sigma-P2D, incorporates information about localization and imaging errors, and the second, Vector-P2D, makes use of averages of multiple images of the same molecule. The latter gives meaningful population statistics for heterogeneous samples. They validated their approaches using Monte Carlo simulations and experiments on motor proteins, in which they analyzed the head-to-head distance of kinesin-1 homodimers and were within error of the expected distance. They also showed that they could measure the stalk length of dynein in multiple nucleotide binding states—measurements that are outside the range of smFRET. These new methods fill an important gap in nanometer distance measurements.