Femtosecond laser microdissection isolates regenerating C. elegans neurons for single cell RNA-sequencing

The complexity of tissue is shaped by profound molecular diversity at a single cell level. Isolating cells of interest based on specific physiology is a crucial step to analyze its molecular underpinnings. Many cell isolation methods have been developed towards such goal. However, none is optimal when highly specific, especially rare cells are desired. For example, cell isolation by fluorescence activated cell sorting (FACS) entails tissue dissociation, which leads to a low yield (only 1 - 10% of labeled cells are collected by FACS), widespread transcriptional artifacts in isolated cells, and loss of the cells’ contextual information. Alternative methods such as laser capture microdissection (LCM) and Patch-seq have been proposed to bypass tissue dissociation, but they also suffer from numerous drawbacks, notably the degradation of cellular content. I developed a new single cell isolation method, femtosecond laser microdissection (fs-LM), to isolate intact single cells directly from living tissue or model organisms. fs-LM resects a single cell from its surrounding tissue by a series of micron-scale ablation spots, which are created by fs-laser ablation in a spherical pattern encasing the cell. I first demonstrated feasibility of fs-LM by isolating neurons of Caenorhabditis elegans and performing single cell RNA-sequencing (scRNA-seq). I achieved a yield of 32% (n = 384) and detected 2,261 ± 132 genes in single C. elegans neuron. In comparison with neurons isolated by the dissociation-FACS method, neurons isolated by fs-LM displayed reduced transcriptional artifacts induced by tissue dissociation, including up-regulation of genes involved in heat shock response and mitochondrial unfolded protein response. Existing clinical interventions to spinal cord injury have been unsatisfactory, and future development requires comprehensive knowledge of the genetic activities driving nerve regeneration. Therefore, I isolated and sequenced regenerating C. elegans neurons following fs-laser axotomy. I revealed transcriptional programs leading to successful regeneration in wild-type animals and regeneration failure in animals lacking DLK-1/p38 kinase. I further investigated the molecular basis of regeneration heterogeneity displayed by neurons of the same type, which has remained understudied despite its clinical importance. In total, 6 distinct gene modules were found to play a potential regulatory role in nerve regeneration.