A Novel Optical Tissue Clearing Protocol for Mouse Skeletal Muscle to Visualize Endplates in Their Tissue Context

Figures

• 

  Figure 1. Sequence of staining and hydrogel embedding affects overall imaging depth and quality of muscle samples. All samples were imaged in 3D using a Leica SP8 confocal microscope and images were processed with Leica LAS X software. (A–C) Mouse EDL muscle was PFA-fixed, hydrogel embedded, stained with BGT-AF647, and then RI matched in 88% glycerol before imaging. (D–F) Mouse EDL was PFA-fixed, stained with BGT-AF647, hydrogel embedded, and then RI matched in 88% glycerol before imaging. (A,D) depict overviews of the whole mouse EDL muscles with the boxed region representing zooms shown in (B,E). (C,F) portray cross sections cropped from the center of the EDL imaging data and depth-coded on the z-axis in order to visualize imaging depth and quality for both methods. (G) Graphical display of NMJ-signal SNRs in correspondence to muscle tissue depth and staining / clearing order. Muscle tissue extension in the central muscle region is depicted in the background as reddish round shape. Mean SNR values are shown as horizontal bars with corresponding numbers next to it. Left and right halves correspond to muscles shown in (C,F), respectively.

• 

  Figure 2. Overview of the MYOCLEAR protocol. MYOCLEAR can be divided into three major stages: hydrogel embedding (days 2–7), staining (days 8–15), and RI matching (days 16–17). The panel gives a graphical overview on the detailed descriptions found in Materials and Methods and Supplementary Methods sections. Photographs next to processing days 0, 10, and 17 show the appearance of EDL muscles at the start of the clearing protocol, before staining, and upon RI matching, respectively.

• 

  Figure 3. MYOCLEAR enables imaging of muscle fibers, myonuclei, and NMJs by using red autofluorescence and spectral unmixing of far-red wavelengths dyes. (A,B) Mouse EDL was processed via the MYOCLEAR protocol and stained with BGT-AF555 and Draq5. (A) depicts a confocal section of the EDL, with (B) representing a zoom of the boxed region. Strong autofluorescence of the tissue in the AF555 channel (green) resulted in a poor SNR for NMJ detection (some NMJs are highlighted in B, arrowheads). In contrast, the near-infrared dye Draq5 displayed crisp and clear nuclei. (C–F) Mouse EDL muscle was processed via the MYOCLEAR protocol and stained with BGT-AF647 and Draq5. In order to overcome the auto-fluorescence shown in this figure, the emission windows for each dye were adjusted according to their peak values and acquired separately using the same 633-nm excitation laser on a SP8 confocal microscope. The images were processed using Leica LAS X software and spectrally un-mixed in ImageJ. (C) Maximum-z projection of the whole EDL before applying spectral un-mixing. Draq5, red; BGT-AF647, green. Green autofluorescence of the thread keeping the muscle in place for imaging is visible at the proximal…

• 

  Figure 4. MYOCLEAR is functional with adult mouse diaphragm muscle. Adult mouse diaphragm muscles were processed using the MYOCLEAR protocol and stained with Draq5 and BGT-AF647. (A) Maximum-z projection of a representative tile stack showing fluorescence signals of BGT and Draq5 in green and red, respectively. Note, that NMJs are not fragmented but partially covered by myonuclei. This is evident in the insert, which shows only BGT signals of a small region. (B) Side view to show depth extension of fluorescence signals. The entire depth of around 500 μm of the diaphragm became transparent. (C,D) Depth coded side views of nuclear (C) and NMJ signals (D). Pseudocolor code is explained on the right side.

• 

  Figure 5. Analysis of whole mount NMJ morphology and quantification of NMJ numbers is enabled in wildtype and mdx muscles upon clearing. EDL muscles from wildtype (A,A',C,D) and dystrophic mdx mice (B,B',E,F) were processed using the MYOCLEAR protocol and stained with BGT-AF647. (A,B) Maximum-z projections of all NMJs detected by hand segmentation. Each cyan spot represents a single NMJ. (A',B') High-power images of some representative NMJs from each muscle shown in (A,B). (C,E) Upper panels, maximum-z projections of representative muscles showing BGT-staining signals. Lower panels, high power display of ROIs 1–5 in corresponding upper panel. (D,F) Quantitative analysis of key morphological parameters: area, fragmentation index, perimeter, and bounding rectangle diagonal of NMJs. Depicted are mean ± SD for all en face NMJs detected as a function of ROI number. *p < 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

• 

  Figure 6. Integrity of NMJ presynapse and other muscle structures is maintained upon clearing. Adult mouse EDL muscles were processed using the MYOCLEAR protocol and co-stained with Draq5 and antibodies against either NMJ presynapse (A, vAChT), sarcolemma (B, dystrophin), ECM (C, collagen I), or sarcomere (D, troponin I). Images show maximum-z projections of confocal z-stacks with an interplane interval of 2 μm and depths from muscle surface of 466, 665, 104, and 214 μm for (A–D), respectively.