Limb regeneration in the Mexican axolotl requires the dedifferentiation of mature limb cells into progenitor cells known as blastema cells. Blastema cells become responsive to the signals that define the blueprint, or pattern, of the regenerating cells organizing into complex limb structures. Signaling downstream of the limb nerves plays a significant role in inducing patterning competency; however, the foundations of this process remain poorly understood. Studying patterning competency in amputated axolotl limbs is challenging because it occurs within the context of an intricate network of signals and tissue interactions that are activated simultaneously during regeneration. Here, we develop a simplified limb injury model assay called the competency accessory limb model (CALM), which we used as a platform to characterize the specific timing of competency induction and maintenance and identify the H3K27me3 chromatin signatures associated with this cellular state. We also discovered that a combination of FGF and BMP ligands is sufficient to induce patterning competency in limb wound cells and identified the ErBB pathway among the regulatory epigenetic targets downstream of these signals.

Tissues were harvested, fixed overnight in 4% formaldehyde, decalcified in 10% EDTA (Amresco) for 5 days on an orbital shaker, and then moved into 30% sucrose (Fisher Scientific) for 3 days on an orbital shaker at 4 °C. Samples were then flash frozen in OCT (Sakura) and cut into 10 μm longitudinal tissue sections for immunodetection. After sectioning, the samples were warmed and then washed twice in PBS and once in 1X PBS-Tween to remove OCT. Samples for H3K27me3 detection were then pressure cooked for 2 minutes in sodium citrate buffer. Samples are then washed with diH₂O. After blocking with 0.5% BSA, the samples were incubated overnight in a 1:300 solution of rabbit anti-histone H3K27me3 polyclonal antibody (A4039050, Epigentek) in PBS-Tween at 4 °C. Slides were washed and blocked for an additional 10 minutes before being incubated for 2 hours at 4 °C in a 1:200 dilution of goat anti-rabbit IgG H&L secondary antibody (ab150080, Abcam). Samples were then stained with DAPI in PBS-Tween for 10 minutes prior to being washed in PBS-Tween and mounted with VECTASHIELD Antifade Mounting Medium.

Tissue sections were imaged on an Axio Observer Z1 inverted fluorescence microscope equipped with an EXCELITAS Technologies X-Cite 120 LED lamp, a 63X oil immersion objective with a Hamamatsu digital camera C11440 and an Apotome2 unit. The imaging was controlled by the Zen Pro 2.0 software package. Z-stacks were imaged using 0.24 μm intervals. Raw z-stacks were then converted into apotome calculation stacks using Zen software and exported to TIFF format files for analysis. Images were preprocessed using the custom ImageJ plug-in Crop&Snip as previously reported in Sosnik et al. and uploaded here as the raw images.

In the paper, these images are then run through H3K_CbyC, a new algorithm and MATLAB code that calculates the relative amount of fluorescently tagged protein present within a cell nucleus in large 3-dimensional image datasets (Supplemental Figure 3). This works by Gaussian blurring the individual images within the DAPI channel, binarizing, setting the image to a black background, filling holes, and then recombining the individual layers back into a matrix. Small objects and under segmented nuclei are removed by volume filtering, and the number of nuclei is detected using a connected component. From the red channel, (specific to AB detection) pixels that overlap with the identified nucleus locations are counted, and the total fluorescence per nucleus within an image stack is calculated. Outlier z-stacks were removed from subsequent analysis. 50-100 nuclei were included in the calculation of averages for each biological sample.  Analyzed data are saved as a single csv format file for the entire data set.

References:

Sosnik, J., Vieira, W. A., Webster, K. A., Siegfried, K. R. & McCusker, C. D. A new and improved algorithm for the quantification of chromatin condensation from microscopic data shows decreased chromatin condensation in regenerating axolotl limb cells. PLoS One 12, e0185292 (2017).