2005-LB: Optimization of Single Nuclei Isolation from Frozen Human Skeletal Muscle and Pancreatic Tissue Samples for Single Nuclei Multiome ATAC + Gene Expression

Background: Dysfunctional pancreatic islets beta cells are a well established hallmark of type 2 diabetes. Despite intense research, we still lack understanding of the underlying gene dysregulation mechanisms, including gene regulatory pathways driving the shift between normal and diabetic cells. Single-nucleus (sn) multiome (joint RNA and ATAC profiling on the same nucleus) sequencing is a powerful technique to acquire multi-modal data from the same nucleus. sn-multiome quality depends on the quality of the input material. Increasing the quality of single nuclei isolated from previously frozen human skeletal muscle and pancreas samples can enhance sn-multiome output.

Goal of the study: Reduce noise in sn-multiome data by isolating high quality nuclei from human samples.

Methods: We compared data from nuclei isolated with citric acid and sucrose gradient protocols to nuclei sorted with the MACS Quant Tyto flow cytometer. The quality of isolated nuclei was visually determined by light microscopy, quantified by ATAC library trace, and RNA integrity number (RIN) score.

Results: We optimized isolation of nuclei from frozen human skeletal muscle samples. Quality controls checks revealed intact nuclei microscopically, all expected peaks on ATAC library traces, and 9.2 RNA RIN. Optimization of nuclei isolation from frozen pancreatic human tissues was more challenging due to the high amount of RNases, which quickly degrade RNA quality and compromise subsequent library preps, even with high-quality nuclei. We optimized nuclei isolation protocol by fine tuning experimental procedures along with adjustment of concentrations of RNase inhibitors.

Conclusions: Optimization of nuclei isolation from frozen human skeletal muscle and pancreatic tissue with MACS Quant Tyto cell sorter demonstrated high quality of sorted nuclei by visual integrity, ATAC trace and high RNA RIN and high sn-multiome data quality.