Complementarity of PALM and SOFI for super-resolution live-cell imaging of focal adhesions

Abstract

Live-cell imaging of focal adhesions requires a sufficiently high temporal resolution, which remains a challenge for super-resolution microscopy. Here we address this important issue by combining photoactivated localization microscopy (PALM) with super-resolution optical fluctuation imaging (SOFI). Using simulations and fixed-cell focal adhesion images, we investigate the complementarity between PALM and SOFI in terms of spatial and temporal resolution. This PALM-SOFI framework is used to image focal adhesions in living cells, while obtaining a temporal resolution below 10 s. We visualize the dynamics of focal adhesions, and reveal local mean velocities around 190 nm min−1. The complementarity of PALM and SOFI is assessed in detail with a methodology that integrates a resolution and signal-to-noise metric. This PALM and SOFI concept provides an enlarged quantitative imaging framework, allowing unprecedented functional exploration of focal adhesions through the estimation of molecular parameters such as fluorophore densities and photoactivation or photoswitching kinetics.

Combining PALM and SOFI for quantitative imaging of focal adhesions in living cells

“…The live cell experiments were described in our previous work 13 . Briefly, we used a custom built microscope equipped with a temperature and CO 2 controlled incubator 11 .…”

“…The simulations were carried out according to a procedure similar to the one we reported previously 13 . We simulated a square region of 1×1 μm that was randomly filled with single emitters.…”

“…The PALM analysis was performed as described in our previous work 13 . Localization of the single emitters in the camera frames was done using a custom written algorithm (Matlab, The Mathworks) that was adapted from an earlier published algorithm 14 .…”

“…A second selection was made by applying a distance threshold. To account for the localization precision, the distance threshold was based on the Hellinger distance 13 . We used a value of 0.9 for this threshold.…”

“…Additionally, an extension called balanced SOFI (bSOFI) can be used to determine the fluorophore on-time ratio, which allows to estimate of the fluorophore density 12 . In previous work, we investigated the complementarity of PALM and SOFI by applying both techniques to the same sequence of camera frames 13 . PALM was found to give reliable estimates of fluorophore numbers in low-density areas, while SOFI was capable of extracting fluorophore densities in high-density regions.…”