RNA in situ hybridisation

Every cell in the body carries the same genes, but each cell type reads (expresses) a different selection of them. When a gene is expressed, its DNA is copied into messenger RNA (mRNA), which leaves the nucleus and is translated into protein. In situ hybridisation (ISH) shows which cells contain a given mRNA, in the tissue itself (in situ, "in its place"). A probe, a piece of nucleic acid whose sequence is complementary to the mRNA (an antisense probe), is applied to a thin section; it pairs (hybridises) with the mRNA, and a label on the probe is then made visible. The principle of detecting RNA–DNA hybrids in cytological preparations was described by Gall & Pardue (1969), and ISH has since been used to map the expression of almost every gene in the mouse brain (Lein et al., 2007).

Neurons can be told apart by the genes they use to make and package their neurotransmitter. Glutamatergic (excitatory) neurons express a vesicular glutamate transporter: VGLUT1 (gene Slc17a7) in most cortical neurons and VGLUT2 (gene Slc17a6) in many subcortical and spinal neurons (Fremeau et al., 2001; Oliveira et al., 2003). GABAergic (inhibitory) neurons express glutamic acid decarboxylase, encoded by Gad1 (GAD67) and Gad2 (GAD65) (Esclapez et al., 1994). Single-cell RNA sequencing of the mouse cortex separates neurons into distinct glutamatergic and GABAergic types (Tasic et al., 2016); ISH shows where those cells sit in the tissue.

Single-molecule fluorescent ISH makes each mRNA molecule visible as a separate fluorescent dot, so expression can be counted cell by cell (Femino et al., 1998; Raj et al., 2008). In the branched-DNA method used here (RNAscope), pairs of short probes must bind side by side on the target before a tree of amplifier molecules is built on them; this amplifies the specific signal while keeping background low, and each dot corresponds to one mRNA molecule (Wang et al., 2012). The region is chosen on a real reference atlas: the Allen Mouse Brain Common Coordinate Framework (Wang et al., 2020), the Allen Reference Atlas of the mouse spinal cord, or the BigBrain histological model of a human brain (Amunts et al., 2013). Image credit: Allen Institute; BigBrain Project.

Learning objectives

• Explain how an antisense probe detects an mRNA in a tissue section, why a sense probe or a bacterial probe gives no signal, and what positive and negative control probes show
• Choose between fresh-frozen and fixed tissue, keep RNA intact with RNase-free handling, and judge human post-mortem tissue by its post-mortem interval, pH and fixation
• Find a target region on a reference atlas of the mouse brain, the mouse spinal cord or the human brain
• Optimise the pretreatment and the protease digestion, and recognise under-digestion and over-digestion in the images
• Image single mRNA dots with suitable filter sets, objective, exposure and z-stack, and assign dots to nuclei with a spot threshold, a cell expansion and a positivity cut-off set from the negative control
• Analyse the proportions of marker-positive neurons with the animal or donor as the experimental unit, explain why pooling cells is pseudoreplication, and explain why a carefully run but small study can still miss a real difference (statistical power)

Ethics and safety

Work with animals needs approval by an Institutional Animal Ethics Committee (in India, the CCSEA under the Prevention of Cruelty to Animals Act 1960; elsewhere, frameworks such as EU Directive 2010/63/EU, the US Animal Welfare Act and PHS Policy, and the UK Animals (Scientific Procedures) Act 1986), an ethical review that applies the 3Rs (replace, reduce, refine), and trained, authorised staff; tissue for this experiment is taken after humane killing or under deep terminal anaesthesia, and no experimental procedure is carried out on the living animal. Report the study following the ARRIVE guidelines, developed by the NC3Rs in the United Kingdom and endorsed by journals internationally (Percie du Sert et al., 2020). Human brain tissue comes only from brain banks, from donors who gave informed consent (or whose next of kin did) for research; An ethics committee (in India, an Institutional Ethics Committee following the ICMR National Ethical Guidelines of 2017; elsewhere, a research ethics committee or, in the United States, an institutional review board) must approve the project, donors are known to the researcher only by a code, and their diagnoses and circumstances, not any experimental treatment, define the groups. Formaldehyde and paraformaldehyde are toxic and are handled in a fume hood with gloves and eye protection; human tissue is handled as potentially infectious, following the laboratory's biosafety rules.

Key terms

• Messenger RNA (mRNA): The copy of a gene that carries its instructions from the nucleus to the ribosomes, where protein is made. Cells that express a gene contain its mRNA.
• In situ hybridisation (ISH): A method that detects a chosen RNA (or DNA) sequence in cells or tissue sections with a labelled complementary probe, showing which cells contain it.
• Antisense probe: A probe whose sequence is complementary to the mRNA, so it can pair with it. A sense probe has the same sequence as the mRNA, cannot pair with it, and is used as a negative control.
• RNase: An enzyme that cuts RNA. RNases are present on skin, in dust and in many solutions and are very stable, so RNA work uses gloves, cleaned surfaces and RNase-free solutions.
• Fresh-frozen tissue: Tissue frozen within minutes of removal, without fixative, and fixed briefly only after sectioning; RNA is well preserved.
• Post-mortem interval (PMI): The time between a donor's death and the freezing or fixation of the brain.
• Branched-DNA amplification: A signal amplification in which a pair of target probes binds the mRNA and then preamplifier, amplifier and label molecules build a tree carrying many dye molecules on each mRNA.
• Protease digestion: A pretreatment with a protein-cutting enzyme that opens the tissue so that probes can reach the mRNA; too little gives no signal, too much destroys the tissue.
• Housekeeping gene: A gene expressed by almost every cell at a fairly constant level, such as Ppib or Ubc; its probe is the positive control for RNA quality and for the assay.
• dapB: A gene of the bacterium Bacillus subtilis that is absent from animal tissue; its probe is the negative control and shows the background.
• Lipofuscin: A pigment that builds up with age inside neurons, especially in human brain, and glows over a broad range of colours (autofluorescence).
• Z-stack: A series of images taken at different focus depths through the section, often combined into a maximum-intensity projection.
• Cell expansion: In image analysis, enlarging each detected nucleus by a fixed distance to approximate the cell body, so that dots in the cytoplasm are assigned to that cell.
• Pseudoreplication: Treating many measurements from the same animal or donor (for example many cells) as if they were independent replicates.