Using confocal microscopy and image processing to measure properties of sub-resolution cell structures: Application to cardiac transverse tubules.

Confocal microscopes increase the resolution of images by rejecting unwanted data from the out of focus regions of the specimen. This incurs a penalty in the recorded signal to noise ratio that may be partially ameliorated by inputting additional information during image processing. This is commonly seen in digital deconvolution techniques, where the point spread function (PSF) of the microscope is used to refine the image. By accounting for the interaction between the microscope PSF and labelled objects, information about sub-resolution structures can be quantified. In the case of cardiac t-tubules, we show the detailed 3D organisation and estimates of the t-tubule diameter and approximately 50% of t-tubules are below the classical optical resolution limit. By obtaining local t-tubule volume (V) and surface area (SA) by using dual labelling, together with control of the PSF, a simple relationship between t-tubule radius and the V:SA ratio can be derived. Solute movement can also be analysed by accounting for the shape of the illumination field during fluorescence recovery after photobleaching (FRAP). Using these methods, we show that the properties of murine and lagomorph ventricular t-tubules are quite different.