Calcium wave propagation varies with changes in endoplasmic reticulum parameters: A computer model

Calcium (Ca2+) waves provide a complement to neuronal electrical signaling, forming a key part of a neuron's second messenger system. We developed a reaction-diffusion model of an apical dendrite in the NEURON simulator, which included diffusible inositol triphosphate (IP3), diffusible Ca2+, IP3 receptors (IP3Rs), endoplasmic reticulum (ER) Ca2+ leak, and ER pump (SERCA) on ER. Ca2+ is released from ER stores via IP3Rs in response to binding of both IP3 and Ca2+. This results in Ca2+-induced Ca2+ release (CICR), which increases the efficiency of spread of Ca2+ waves. We adjusted the parameters in order to replicate Ca2+ wave initiation and rate of spread observed in vitro. At least two modes of Ca2+ wave spread have been suggested: a continuous mode that depends on continuous underlying substrate that can regenerate the Ca2+ wave, presumed based on relative homogeneity of ER within the cell; and a saltatory model where Ca2+ regeneration occurs at discrete points with diffusion between them. In order to explore similarities and differences between these modes, we compared the effects of 3 patterns of hypothesized IP3R distribution: 1. continuous homogeneous ER, 2. areas of increased ER density (ER stacks); 3. hotspots with increased IP3R density (IP3R hotspots). All 3 models could produce spread of Ca2+ waves at a velocity similar to that measured experimentally ( 50 $μ$m /sec). However, sensitivity to differences in ER density differed greatly across the 3 situations. Continuous ER showed far greater sensitivity to IP3R density increase than did the other patterns: time to onset was reduced from 80 to 10 ms, speed increased from 30 to 80 $μ$m/sec, duration at one location increased from 0.8 to 1.1 s. Increases in SERCA density resulted in opposite effects: time to onset decreased from 10 to 40 ms, speed decreased from 60 to 35 $μ$m/sec, and duration was reduced from 1.2 to 0.6 s. By contrast, our measures were generally insensitive to changes in density of IP3R hotspots or stacks, although time to onset showed a reduction in both cases from 10 to 5 ms. We also tested the effects of varying distance between hotspots between 10 and 100 $μ$m. Speed decreased from 75 to 55 $μ$m/sec with increasing distance between hotspots, with other measures insensitive to this distance variation. Our modeling suggests that pharmacological targeting of IP3Rs and SERCA could allow modulation of Ca2+ wave propagation in diseases where Ca2+ dysregulation has been implicated.