Using mechanical homogenization to isolate microglia from mouse brain tissue to preserve transcriptomic integrity

Abstract

Numerous approaches have been developed to isolate microglia from the brain, but procedures using enzymatic dissociation at 37°C can introduce drastic transcriptomic changes and confound results from gene expression assays. Here, we present an optimized protocol for microglia isolation using mechanical homogenization. We use Dounce homogenization to homogenize mouse brain tissue into single-cell suspension. We then isolate microglia through Percoll gradient and flow cytometry. Isolated microglia exhibit a gene expression pattern without the changes induced by heated enzymatic digestion. For complete details on the use and execution of this protocol, please refer to Clayton et al. (2021).

Keywords: Cell isolation; Cell separation/fractionation; Classification Description: Cell biology; Flow cytometry/Mass cytometry; Gene expression; Molecular biology; Neuroscience; RNAseq.

Graphical abstract

Figure 1

Representative images of the homogenization solution after completion of pestles A and B (A) Once the homogenization mixture reaches the consistency shown in this image, the suspension is ready to begin homogenization with pestle B. (B) Once the suspension reaches the consistency shown in panel B, the suspension is ready to be poured into a 15 mL tube for centrifugation.

Figure 2

Representative image of the suspension solution after centrifugation (A) Representative image of what the suspension will look like following the short 300 g spin down. A relatively clear supernatant with a robust pellet should be visible.

Figure 3

Representative images of the Percoll layer (A) Image of the 37% Percoll layer with cells suspended within it layered on top of the 70% Percoll layer. It is important to prevent mixing of these two layers to ensure a proper interphase layer in the Percoll centrifugation step.

Figure 4

Representative images of the microglia layer on top of the 70% Percoll layer following the long centrifugation step (A) Image of a microglia layer highlighted with a red box, which appears as an opaque thin layer at 5 mL, which corresponds to the amount of 70% Percoll in the tube. (B) Image of the same layer highlighted with a red box (A).

Figure 5

FACS plot depicting the gating strategy used to sort microglia (A) Side Scatter (SSC)-Area, Forward Scatter (FSC)-Area to remove debris (B) SSC-Width, SSC-Height for exclusion of doublets (C) FSC-Width, FSC-Height to remove debris (D) SSC-A, LIVE/DEAD-UV to exclude dead cells (E) Ly6C-PercpCy5.5-A, CD11b-PE-Cy7-A to exclude peripheral macrophages and include microglia (F) FCRLS-APC-A, CD11b-PE-Cy7-A to confirm microglia specificity. (G) Details the number and percentage of cells gathered from each gate.

Figure 6

RNAseq data from microglia isolated using the described protocol Transcripts per million (TPM) for microglia specific genes including Csf1r, P2ry12 and Tmem119 were enriched, while genes for stress induced by warm dissociation including Fos, Jun, Hspa1a, and Zfp36 were minimally expressed. Markers for other CNS cell types such as astrocytes (GFAP, Aldh1l1), Neurons (Map2, Nsg2) and Oligodendrocytes (Mog, Olig2) were negligible (N=9 mice).

Figure 7

Representative image of properly perfused mouse brain and dissection in preparation for homogenization (A) Image of properly perused brain removed from mouse carcass. (B) Representative image of brain segmented into smaller pieces for more efficient homogenization. Scale bar=1 cm.