Volume-rendered 3-D dSTORM super-resolution microscopic image. RyR clusters [red] are distributed along T-tubules [green]. Caveolin-3 immuno-staining is used to determine T-tubule structure. Estimated resolution of this image is ~30nm, which is approximately 10 times better than thtat of conventional light microscopy.
The rhythmic contraction of heart muscle, i.e., the heart beat, is controlled by a uniquely developed regulatory unit made of membranes, which translates the electric excitation of the plasma membrane to dynamic fluxes of calcium ions, which then spread throughout the cell and induce robust cell contractions. The malfunction of this regulatory machinery is known to cause heart failure and loss of coordination among units, and results in arrhythmias.
A UCSD laboratory led by Masa Hoshijima and the biophysics laboratory directed by Christian Soeller at the University of Auckland have had a shared interest in determining the structure and patho-physiological function of this muscle regulatory unit. However, this task has been extremely challenging, mainly due to the fact that the size of this unit is smaller than the resolution limit of conventional light microscopes. Hoshijima and Soeller have each taken completely different approaches to solve the problem. While Hoshijima uses various 3-D electron microscopic tools, Soeller has adapted a novel super-resolution light microscopy method, namely direct stochastic optical reconstruction microscopy (dSTORM). Neither uses direct visualization.
Both rely on extensive computational data processing. Hoshijima and Soeller decided to bring their technologies together and assigned me to work in Soeller’s laboratory to apply dSTORM to heart samples, which were prepared by Hoshijima and shipped to Auckland. The samples were studied in 3-D electron microscopy at UCSD, in parallel.
With technical support provided by colleagues in the Soeller lab, I was able to successfully visualize calcium flux regulatory units as nanometer-scale clustering of ryanodine receptor (RyR) in normal and disease model mouse cardiac myocytes, using dSTORM. RyR clusters were three-dimensionally mapped along with tubular membrane invaginations of surface cell membrane. The geometry of the clusters was remarkably heterogeneous; yet, they were densely assembled at the enlarged bifurcation loci of branches. This was entirely an unprecedented finding, but is well-supported by Hoshijima’s 3-D electron microscopy.
This project is significant, as it is not limited to descriptive observation. The dSTORM data are readily useful for a variety of simulation work, combined with geometry determined by electron microscopy.
PARTICIPATING RESEARCHERS: University of Auckland: Christian Soeller, Vijay Rajagopal; UCSD: Masa Hoshijima