Dynamic changes in sarcoplasmic reticulum structure in ventricular myocytes

Abstract

The fidelity of excitation-contraction (EC) coupling in ventricular myocytes is remarkable, with each action potential evoking a [Ca²⁺](i) transient. The prevalent model is that the consistency in EC coupling in ventricular myocytes is due to the formation of fixed, tight junctions between the sarcoplasmic reticulum (SR) and the sarcolemma where Ca²⁺ release is activated. Here, we tested the hypothesis that the SR is a structurally inert organelle in ventricular myocytes. Our data suggest that rather than being static, the SR undergoes frequent dynamic structural changes. SR boutons expressing functional ryanodine receptors moved throughout the cell, approaching or moving away from the sarcolemma of ventricular myocytes. These changes in SR structure occurred in the absence of changes in [Ca²⁺](i) during EC coupling. Microtubules and the molecular motors dynein and kinesin 1(Kif5b) were important regulators of SR motility. These findings support a model in which the SR is a motile organelle capable of molecular motor protein-driven structural changes.

Figure 1

Beat-to-beat fidelity of EC coupling in ventricular myocytes. (a) Average peak amplitude of action potential evoked (1 Hz) global Ca²⁺ transients (nM) from adult (closed circles) and neonatal (open circles) ventricular myocytes measured at 1 minute interval for five minutes and corresponding coefficient of variation among the adult and neonatal myocyte population. Dashed line represents average peak [Ca²⁺]_i signal over 5 minutes. (b) Representative [Ca²⁺]_i transient of adult and neonatal myocytes. (c) Representative [Ca²⁺]_i signal variance of adult and neonatal ventricular myocytes and the distribution of peak amplitude [Ca²⁺]_i variance (nM²) of adult ventricular myocytes in the presence and absence of the SR Ca²⁺ pump inhibitor thapsigargin (1 μM).

Figure 2

Imaging of network, junctional, and corbular SR boutons in living and fixed ventricular myocytes. (a) A cartoon of the tRFP-SR construct. (b) Confocal image of representative living neonatal ventricular myocyte expressing tRFP-SR. (c) Confocal images of representative living adult ventricular myocyte. Insets show zooms of a corbular SR bouton (red box) and SR network (green box) in these cells. Graphs show fluorescence intensity profile in the region in the image marked by arrows. (d) Electron microscopy image of non-permeabilized neonatal ventricular myocytes infected with tRFP-SR (scale bar = 1 μm; N = nucleus). Insets depict invaginations of the endoplasmic/sarcoplasmic reticulum (scale bar = 100 nm). Panel Dii shows an example of how the dimensions of SR boutons were made. (e) Immunogold labeling of tRFP-SR (arrows of insets) from a permeabilized neonatal ventricular myocyte infected with tRFP-SR depicting invaginations of the endoplasmic/sarcoplasmic reticulum (scale bar = 500 nm and 100 nm (insets); N = nucleus).

Figure 3

Motile corbular SR boutons in neonatal and adult ventricular myocytes. (a) Time-lapse confocal images of a neonatal ventricular myocytes expressing tRFP-SR. Images shown were obtained 0, 150, and 300 seconds after the initiation of the experiment. Motile corbular SR boutons are labeled with colored circles. (b) Plot of the velocity of corbular SR boutons i and ii as a function of time. (c) Time-course of corbular SR bouton motility in a representative adult ventricular myocyte expressing tRFP-SR. The insets in the lower right corner of each image show an expansion of the region in the cell marked by the red box. (d) Plot of velocity of this corbular SR bouton as a function of time. (e) Bar plot of the mean velocity of corbular SR boutons in neonatal and adult ventricular myocytes (*P < 0.05). (f) Histogram of the total distance traveled over a period of 300 seconds of 79 corbular SR boutons in adult and neonatal ventricular myocytes. (g) Histogram of the step distance traveled by corbular SR boutons. The inset shows an expanded view of the region of the histogram from 1.8 to 4.2 μm.

Figure 4

Dynamic remodeling of the SR network of neonatal and adult ventricular myocyte. (a) Confocal image of a neonatal ventricular myocytes expressing tRFP-SR. Images to the left show expanded views of sites i and ii at various time points. The red arrow in site ii marks a corbular SR bouton. Redlines delineate a portion of the SR network and shows elongation and subsequent fusion of this SR bouton to an adjacent SR tubule. (b) Time-lapse confocal imaging of a region of an adult ventricular myocyte expressing tRFP-SR. Redlines delineate a portion of the SR network and shows elongation and subsequent fusion of SR tubule to an adjacent SR tubule. (c) Confocal image of an adult ventricular myocyte. Surface plots at multiple time points of sites i and ii are shown to the right of this image.

Figure 5

TIRF imaging of submembrane SR motility in neonatal and adult ventricular myocytes. (a) TIRF images of representative neonatal ventricular myocyte expressing tRFP-SR. (b) TIRF image of representative adult ventricular myocyte expressing tRFP-SR. Images to the right are two dimensional images (top row) and surface plots of the region delineated by the white.

Figure 6

Expression of functional ryanodine receptors in motile corbular SR boutons. (a) Confocal image of a neonatal ventricular myocyte expressing tRFP-SR (left), labeled with BODIPY FL-X Ryanodine (center) and merged image (right). (b) Time-lapse confocal images of neonatal ventricular myocyte from panel A expressing tRFP-SR and labeled with BODIPY FL-X Ryanodine. Images shown were obtained 0, 150, and 300 seconds after the initiation of the experiment. Motile corbular SR boutons are labeled with colored circles. (c) Confocal image of adult ventricular myocyte expressing tRFP-SR (left), labeled with BODIPY FL-X Ryanodine (center) and merged image (right). (d) TIRF image (left) of a representative neonatal myocyte expressing tRFP-SR loaded with fluo-4. Green dot represents corbular SR bouton of interest. The plot in the center shows the time course of Ca²⁺ spark activity in this corbular SR bouton. The plot to the left shows the velocity of the corbular SR bouton marked by the green dot as a function of time. (e) Confocal image (left) of a representative adult ventricular myocyte expressing tRFP-SR loaded with fluo-4 AM. Colored dots mark corbular SR boutons in this cell. The two plots in this panel show the time-course of spontaneous Ca²⁺ sparks in corbular SR boutons i and ii.

Figure 7

Microtubules regulate SR structure in neonatal and adult ventricular myocytes. (a) Confocal images of a neonatal ventricular myocyte expressing tRFP-SR in the absence (top; control) or presence of 10 μM nocodazole (bottom). (b) Confocal image of an adult ventricular myocyte expressing tRFP-SR. The inset shows a zoomed-in region of the cell delineated by the white box. The image to the left is a 2D Fourier transformation of the confocal image. (c) Histograms of SR_Power values in control and nocodazole-treated adult ventricular myocytes. Dashed green line shows global shift in SR_Power distributions in nocodazole-treated myocytes from the center (1.7 μm) of control myocytes. Solid red lines represent fits to the data with single (control; center = 1.7 μm) and dual (nocodazole; center peak 1 = 0.88 μm, center peak 2 = 1.53 μm) Gaussian functions.

Figure 8

The microtubule-associated motors dynein and Kif5b regulate SR motility in ventricular myocytes. (a) Confocal image of neonatal ventricular myocyte expressing tRFP-SR under control conditions and after 120 minutes of 30 μM EHNA perfusion. Corbular SR boutons are identified with colored circles. (b) Confocal images of neonatal ventricular myocyte expressing tRFP-SR (left) and Kif5b-WT-BFP (top/center) or Kif5b-DN-BFP (bottom/center). The images to the right were generated by merging the tRFP-SR and Kif5b-BFP images. (c) Bar plot of the mean total distance traveled by corbular SR boutons before and after EHNA perfusion and myocytes infected with tRFP-SR (control) and myocytes infected with tRFP-SR and Kif5B-DN-BFP. (d) Confocal images showing tRFP-SR fluorescence of adult ventricular myocytes under control conditions (top) and incubated in the presence of 30 μM EHNA (bottom). Surface plots show the time-course of fluorescence intensity in the region delineated by the white box. (e) Cartoon depicting a hypothetical model for the regulation of SR structure and motility in ventricular myocytes (LTCC = L-type Ca²⁺ channel; RyR = ryanodine receptor).