Myosin-actin interaction in Dictyostelium cells revealed by GFP-based strain sensor and validated linear spectral unmixing

Myosin is an actin-based motor protein that is involved in a wide range of cellular motile processes. Although in vitro properties of the myosin-actin interaction have been extensively studied, the interaction in vivo remains poorly understood. Recently we developed a GFP-based strain sensor termed PriSSM (PRIM-based strain sensor module), by using the proximity imaging (PRIM) technique, which detects spectral changes of two GFP molecules that are in direct contact. Using PriSSM-myosin II fusion proteins, the interaction between myosin II and F-actin can be detected in Dictyostelium cells. In the spectroscopic measurements of PriSSM, to decompose the measured spectra of the cells expressing the sensor proteins into the contributions from the sensor and the background autofluorescence, we applied the linear spectral unmixing approach, which was based on the assumption that the errors at each wavelength were independent. Cellular autofluorescence, however, often includes systematic errors, so that the unmixing procedures might lead to biased estimates. Here, to validate our spectral unmixing procedures, we estimate the possible maximum errors in the fluorescence ratio values that are obtained by unmixing spectra including such systematic errors. This estimation provided a general criterion to validate the results obtained by linear unmixing of spectra including serially correlated error terms. Using the proposed criterion and PriSSM-myosin II fusion proteins, we examined the interaction between myosin II and F-actin in Dictyostelium cells under several different conditions. The spectroscopic results, together with the microscopic observations of the cells expressing the proteins, suggest that the formation of myosin filaments through the tail region has only a slight effect on binding to F-actin but has significant effects on the cortical localization of myosin II.