From convolution to clarity: effect of different point spread functions for deconvolution in CLSM and STED microscopy images of the nuclear lamina

Abstract Blurring and noise (convolution) significantly affect the quality of microscopy images. To reverse this process, deconvolution can be applied. The Point Spread Function (PSF) of the microscope, which can be theoretically modelled or retrieved experimentally by using sub-resolution fluorescent beads, is indispensable for performing deconvolution. The study aimed to optimize the deconvolution approach for both Confocal Laser Scanning Microscopy (CLSM) and Stimulated Emission Depletion (STED) microscopy images of immunofluorescently labelled lamins A/C or lamin B1, important components of the nuclear lamina. For that purpose, the impact of using different PSFs on deconvolution performance was assessed visually and by determining the lamina thickness. For CLSM images, performing deconvolution with the experimental PSF led to a higher resolution compared to the theoretical PSF. For STED, the experimental PSF introduced a ringing artifact. Therefore, a new PSF for STED microscopy is introduced, the experimentally supported theoretical (EST) PSF. This uses parameters from the experimental PSF to create a theoretical PSF, which solves issues arising from noisy PSF recordings due to the very small size and thereby low fluorescence intensity of the sub-resolution fluorescent beads. In conclusion, for CLSM the experimental PSF was best for deconvolution, while for STED, the EST PSF was optimal.