Isolation of nuclei and downstream processing of cell-type-specific nuclei from micro-dissected mouse brain regions – techniques and caveats

Abstract

The mammalian brain consists of several structurally and functionally distinct regions equipped with an equally complex cell-type system. Due to its relevance in uncovering disease mechanisms, the study of cell-type-specific molecular signatures of different brain regions has increased. The rapid evolution of newer and cheaper sequencing techniques has also boosted the interest in cell-type-specific epigenetic studies. In fact, the nucleus holds most of the cell’s epigenetic information and is quite resistant to tissue dissociation processes as compared to cells. As such, nuclei are continually preferred over cells for epigenetic studies. However, the isolation of nuclei from cells is still a biochemically complex process, with every step affecting downstream results. Therefore, it is necessary to use protocols that fit the experimental design to yield nuclei of high quality and quantity. However, the current protocols are not suitable for nuclei isolation of small volumes of micro-dissected brain regions from individual mouse brains.Additionally, the caveats associated with centrifugation steps of nuclei extraction and the effects of different buffers have not been thoroughly investigated. Therefore, in this study, we describe an iodixanol based density gradient ultracentrifugation protocol suitable for micro-dissected brain regions from individual mice using ArccreERT2 (TG/WT).R26CAG-Sun1-sfGFP-Myc (M/WT or M/M). This mouse model shows sfGFP expression (sfGFP+) in the nuclear membrane of specific stimulus activated cells, thereby providing a good basis for the study - nuclei isolation and separation of cell-type-specific nuclei. The study also introduces new tools for rapid visualization and assessment of quality and quantity of nascent extracted nuclei. These tools were then used to examine critical morphological features of nuclei derived from different centrifugation methods and the use of different buffers to uncover underlying effects. Finally, to obtain cell-type-specific nuclei (sfGFP+ nuclei) from the isolated nuclei pool of high viscosity, an optimized protocol for fluorescence activated nuclei sorting (FANS) was established to speed up sorting. Additionally, we present a 1% PFA protocol for fixation of isolated nuclei for long term microscopic visualization.

INTACT vs. FANS for Cell-Type-Specific Nuclei Sorting: A Comprehensive Qualitative and Quantitative Comparison

“…In our experiments, we used the “social interaction test” (SI, [ 30 ]) as the intended behavioral stimulus to obtain populations of sfGFP-positive (sfGFP+) nuclei, following TAM injection. Nuclei were isolated from the neocortices of 13–16 week old transgenic mice using the protocol from Chongtham et al [ 28 ] ( Figure 2 A). Following isolation, nuclei were diluted with wash buffer and aliquoted into equal volumes to isolate the sfGFP+ nuclei using either INTACT or FANS ( Figure 2 B,C).…”

“…The observations remain similar to that in phase-contrast microscopy. Thereafter, the increase in the size of FANS-nuclei was validated over a larger sample size using a macroscript [ 28 ] for faster processing ( p < 0.0001, n = 300–400 nuclei, Supplementary Materials Figure S3D , <3 h after sorting (A)). Already, 4–6 h after sorting the size of FANS-nuclei again decreased as shown in the Supplementary Materials Figure S3D , p < 0.001.…”

“…In our results, FANS-nuclei showed a significant increase in size compared to the INPUT nuclei, while INTACT-nuclei did not change significantly within 3 h of sorting. We hypothesize that the increase in FANS-nuclei size could be due to the high hydrodynamic pressure, shearing stress [ 8 ], or PBS-induced destabilization of membrane integrity during the sort [ 28 , 44 ]. Destabilized membrane integrity has been associated with flow cytometry as stated in previous studies [ 8 , 13 ].…”

“…Following behavioral experiments, the brain regions—neocortex, pre-frontal cortex, hippocampus, hypothalamus, pituitary, and nucleus accumbens—were dissected according to region specifications in the Allen Brain Atlas for mouse. Nuclei were isolated using the iodixanol-based gradient ultracentrifugation protocol for isolation of micro-dissected brain regions as described in Chongtham et al [ 28 ]. The duration for high-speed ultracentrifugation (7820 rpm) for the neocortex was 12 min, while for the other brain regions, it was kept at 18 min.…”

“…In this study, we used a transgenic mouse model (Arc creERT2(TG/WT) .R26 CAG-Sun1-sfGFP-Myc(M/WT or M/M) ) to establish and compare the two techniques—INTACT and FANS—for the isolation of activated neuronal nuclei from small as well as bulk tissue samples. Using a modified protocol for nuclei isolation from small tissues [ 28 ], first, we examined the sorting processes’ speed and efficiency using different tissue types and sfGFP+ nuclei percentage. Then, we selected the neocortex (i.e., bulk tissue) to compare the physical (i.e., nuclei shape and intensity) and molecular (i.e., transcriptional profile and chromatin accessibility) aspects using low-input nuclei from the different sorting methods.…”