Correction to: Nature https://doi.org/10.1038/s41586-018-0016-3, published online 04 April 2018.

In this Letter, there are several errors to the following figures. In Fig. 2e, the y-axis labels should have been ‘0, 50, 100, 150, 200, 250 and 300’ rather than ‘0, 50, 100, 100, 200, 200 and 300’. In Extended Data Fig. 5g, owing to an error during figure preparation, the adult human heart tissue group was duplicated and different regions of the same section were shown for both the ‘adult’ and the ‘intensity’ images. This figure has been corrected online, and the original, incorrect panels are shown as Fig. 1 of this Amendment for transparency. Raw data for the images in Extended Data Fig. 5g and additional contemporary transmission electron microscopy (TEM) images are included in ‘SI ED5.zip’ in Figshare (https://doi.org/10.6084/m9.figshare.5765559). In addition, the same TEM micrograph of mitochondria was used in Fig. 2d and the last panel in Extended Data Fig. 6 (labelled ‘Mitochondria’), which focused on two different areas to illustrate lipid droplets and the ultrastructure of mitochondria. The same micrograph was shown at different magnifications (scale bars in Fig. 2d and Extended Data Fig. 6 are 200 nm and 500 nm, respectively) and orientations. Raw data for these images and additional contemporary TEM images are included in ‘SI ED2’ in Figshare (https://doi.org/10.6084/m9.figshare.5765559).

Fig. 1
figure 1

This figure shows the original, published ‘adult’ and ‘intensity’ images from Extended Data Fig. 5g, and the corrected versions.

Full size image
Fig. 2
figure 2

This figure shows the original, published Extended Data Fig. 8f, g, and the corrected Extended Data Fig. 8f, g.

Full size image

In Extended Data Fig. 8f, g, the boxes indicating the position of the insets were misplaced. This figure has been corrected online, and the original incorrect panels are shown as Fig. 2 to this Amendment for transparency. High-resolution images of Extended Data Fig. 8 and the individual panels are included in ‘SI ED8.zip’ in Figshare (https://doi.org/10.6084/m9.figshare.5765559). All data were analysed (no exclusions), as specified in the Reporting Summary. In addition, in Fig. 1d and Extended Data Fig. 3f data are ‘mean ± s.e.m.’ rather than ‘mean ± s.d’, and the legend and figure, respectively, have been updated accordingly.

In the Methods, the pulsatile electrical field voltage should be ‘3.5–4.0 V cm−1’ rather than ‘4.5 mV’ (as correctly shown in Extended Data Fig. 1e). We also wish to clarify that the immunostainings in Extended Data Fig. 5e, f were performed in two adjacent tissue sections, as specified in the figure legend. Finally, we wish to include two additional references to further document the T-tubule system in adult human heart tissue. These references should appear after the following text: “…dependent on both intracellular calcium reserves and the proximity of CaV1.2 channels and T-tubules…”, and have been included as references 28 and 29, with all subsequent citations renumbered. This Letter has been corrected online.