The Influence of Sarcoplasmic Reticulum Ca²⁺ Concentration on Ca²⁺ Sparks and Spontaneous Transient Outward Currents in Single Smooth Muscle Cells

Dr. Fay died on 18 March 1997.

Address correspondence to John V. Walsh, Jr., Department of Physiology, Biomedical Imaging Group, University of Massachusetts Medical Center, Worcester, MA 01605. Fax: 508-856-5997; E-mail: [email protected]

The ability of the ultrafast microscope to resolve and measure highly localized calcium signals was examined using a computer simulation of Ca²⁺ sparks of known peak [Ca²⁺] as imaged inside a model cell. Fluorescence ratios (ΔF/F₀) were calculated from simulated images of a range of spark [Ca²⁺] amplitudes, both in and out of focus. From these simulations, we estimated that an observed average spark amplitude of 10% (ΔF/F₀) is consistent with a peak spark [Ca²⁺] of 200 nM, or 100 nM above resting [Ca²⁺]. This estimate was made in the following way.

First, the fluorescence intensity distribution of a typical Ca²⁺ spark inside a smooth muscle cell was simulated. Custom software was used to calculate the three-dimensional image of a model smooth muscle cell filled with 50 μM fluo-3 (K_d = 390 nM) in equilibrium with a resting [Ca²⁺] of 100 nM. The cell was modeled as a cylinder with cross-sectional diameters of 10 μm in the transverse direction and 6 μm in the axial direction, the direction of focus in the microscope, and of infinite length with respect to the imaging. These dimensions were previously derived from three-dimensional reconstructions of toad gastric smooth muscle cells (our unpublished data). The three-dimensional fluorescence intensity distribution was calculated assuming bound fluo-3 was 100× as fluorescent as the free species. At resting [Ca²⁺], ∼25% of the fluo-3 was bound to Ca²⁺ and the fluorescence signal at rest was ∼20% of the maximum attainable with saturating Ca²⁺.

Second, the spatial [Ca²⁺] profile of a Ca²⁺ spark was added to the simulated resting cell. A Ca²⁺ spark was modeled as a stationary, Gaussian spot of calcium with a known peak [Ca²⁺] and a spatial full width at half-maximum amplitude of 1.7 μm. This simulates the spark at a single point in time, corresponding to the observed images of sparks at the time of peak fluorescence intensity. The Ca²⁺ spark was added to the resting cell model with the spark peak at the center of the cell, and the corresponding three-dimensional image of fluo-3 distribution was calculated as for the resting cell alone. The center of the cell was used to avoid having to simulate the effect of the cell membrane on diffusion. Since the fluorescence background due to resting [Ca²⁺] is likely highest at the center, where the cell is thickest, this is probably the worst case, for our purposes, for measuring ΔF/F₀.

Next, the image formation and acquisition was simulated. The three-dimensional fluorescence image of the cell and spark was blurred with the three-dimensional image of a theoretical, wide-field, point spread function, for a 1.3 NA objective lens (Nikon Inc.) calculated at 530 nm wavelength (Tella, 1985). The image resolution was decreased to 300-nm pixels, by adding three-by-three groups of pixels, in order to simulate the image formation (point spread function) and acquisition (camera pixelization) process. The resulting three- dimensional image contained images of the spark in and out of focus, as seen against the fluorescence background arising from the global resting [Ca²⁺].

Lastly, using the blurred images of the cell with and without the spark, the fluorescence ratios (ΔF/F₀) were calculated at the pixel corresponding to the spark center, at 200-nm focus steps through the 6-μm depth of the cell. The effect of uncertainty in focus was examined by weighting the ΔF/F₀ calculated at each depth through the cell by the probability of a spark occurring at that depth. Although the modeled spark was located in the cell center, the model used for spark spatial distribution assumed that sparks were constrained to occur at the outer edge of the cell, adjacent to the plasma membrane, and were equally likely to occur anywhere along the plasma membrane. A spark with a known peak [Ca²⁺] of 200 nM (100 nM above resting [Ca²⁺]) yielded a ΔF/F₀ of 18% when in focus (centered in depth) and 5.5% when 3 μm out of focus (top or bottom of cell). After accounting for the effects of spark location on focus, the average ΔF/F₀ was 10%, a value equivalent to the average observed spark peak amplitude described in this report.