Background

Proper skeletal muscle function must be maintained for survival. One component of this maintenance is adjustments in gene expression in response to cellular needs and environmental cues. Nuclear processes modulating gene expression are a critical component in regulating cellular composition and behavior. However, myonuclear proteins involved in these processes are difficult to study because of four technical limitations. First, skeletal muscle is dense, tightly packed with contractile proteins that make up more than 60% of proteins in the tissue (Deshmukh et al., 2015; Cutler et al., 2017). These high abundance contractile proteins eclipse the far less abundant nuclear proteins. Second, the dense fibrous structure of skeletal muscle makes it difficult to dissociate without damaging nuclei, making nuclei difficult to isolate. Third, after centrifugation dense debris cosediments with nuclei, compounding the difficulty of isolating nuclei from the tissue. Fourth, skeletal muscle is comprised of multiple cell types, so nuclei isolated and nuclear proteins detected may be from myonuclei or nuclei of other cell types.

Several approaches have been optimized to enrich myonuclei from different organisms for various downstream applications. Ohkawa et al. presented a detailed protocol for isolating myonuclei from mouse tissue that was developed to maximize access of cross-linking reagent for Chromatin Immunoprecipitation (ChIP) analysis (Ohkawa et al., 2012). Wilkie and Shrimer developed a procedure to isolate the myonuclear envelope and sarcoplasmic reticulum for proteomic comparison (Wilkie and Schirmer, 2008). An approach optimized by Dimauro et al. simultaneously collected mitochondrial, nuclear, and cytoplasmic fractions to compare protein localization among different cellular compartments (Dimauro et al., 2012). While each of these approaches to enrich nuclei from skeletal muscle tissue was effective for the intended subsequent analysis, they did not prioritize isolating intact nuclei and not distinguish between myonuclei and nuclei from other cell types. An affinity-based method to selectively isolate nuclei from specific cell types was developed in Arabidopsis thaliana (Deal and Henikoff, 2011) and is now available for mice (Jankowska et al., 2016). However, this affinity-based approach requires genetic labeling of the cell types of interest, which makes it prohibitively cumbersome to examine myonuclei from multiple mouse models. We optimized an ultracentrifugation sucrose gradient-based fractionation approach that requires relatively small sample sizes, no genetic labeling, and is compatible with downstream analysis by flow cytometry and mass spectrometry. The isolated nuclei are intact, biochemically depleted of proteins from non-nuclear organelles, and 85% of nuclei are myonuclei. To isolate myonuclei more quickly we also optimized affinity-based purification using a myonuclear-specific nuclear envelope protein, Transmembrane Protein 38A (TMEM38A) (Bleunven et al., 2008; Cutler et al., 2017), and magnetic beads. This approach is more rapid and yields nuclei with comparable population purity to nuclear isolation by ultracentrifugation but with lower biochemical purity. These nuclear isolation techniques can be used to purify myonuclei from any mouse model for diverse downstream analyses.