Synergizing superresolution optical fluctuation imaging with single molecule localization microscopy

Abstract

Single-molecule-localization-microscopy (SMLM) and superresolution-optical-fluctuation-imaging (SOFI) enable imaging biological samples well beyond the diffraction-limit of light. SOFI imaging is typically faster, yet has lower resolution than SMLM. Since the same (or similar) data format is acquired for both methods, their algorithms could presumably be combined synergistically for reconstruction and improvement of overall imaging performance. For that, we first defined a measure of the acquired-SNR for each method. This measure was ∼x10 to x100 higher for SOFI as compared to SMLM, indicating faster recognition and acquisition of features by SOFI. This measure also allowed fluorophore-specific optimization of SOFI reconstruction over its time-window and time-lag. We show that SOFI-assisted SMLM imaging can improve image reconstruction by rejecting common sources of background (e.g. out-of-focus emission and auto-fluorescence), especially under low signal-to-noise ratio conditions, by efficient optical sectioning and by shortening image reconstruction time. The performance and utility of our approach was evaluated by realistic simulations and by SOFI-assisted SMLM imaging of the plasma membrane of activated fixed and live T-cells (in isolation or in conjugation to antigen presenting cells). Our approach enhances SMLM performance under demanding imaging conditions and could set an example for synergizing additional imaging techniques.