Intracellular RNA-tracking methods

Abstract

RNA tracking allows researchers to visualize RNA molecules in cells and tissues, providing important spatio-temporal information regarding RNA dynamics and function. Methods such as fluorescent in situ hybridization (FISH) and molecular beacons rely on complementary oligonucleotides to label and view endogenous transcripts. Other methods create artificial chimeric transcripts coupled with bacteriophage-derived coat proteins (e.g. MS2, λN) to tag molecules in live cells. In other approaches, endogenous RNAs are recognized by complementary RNAs complexed with noncatalytic Cas proteins. Each technique has its own set of strengths and limitations that must be considered when planning an experiment. Here, we discuss the mechanisms, advantages, and weaknesses of in situ hybridization, molecular beacons, MS2 tagging and Cas-derived systems, as well as how RNA tracking can be employed to study various aspects of molecular biology.

Keywords: RNA labelling; mRNAs; noncoding RNAs; post-transcriptional gene regulation; ribonucleoprotein complexes.

Figure 1.

RNA FISH. (a) Short oligomers are synthesized with a targeting sequence that is antisense to the RNA of interest. Antisense oligomers (ASOs) are conjugated with fluorescent dyes that can be excited with confocal or fluorescent microscopy. Presently, RNA FISH oligomers are provided as a ‘cocktail’ of different ASOs that can tile the target RNA. (b) ASOs hybridize to the target RNA with the antisense targeting sequence. By tiling the RNA of interest with ASOs, the signal is amplified to facilitate detection of low-copy or diffuse transcripts in the cell. (c) Example of RNA FISH in HeLa cells imaged with confocal microscopy (unpublished image from the authors). GAPDH mRNA (light green), tagged with Quasar 670 ASOs, is detected in the cytoplasm. MALAT1 (red), tagged with Quasar 570 ASOs, is localized in nuclear speckles.

Figure 2.

Molecular beacons. (a) The most common molecular beacon structure is a stem-loop style probe. The stem-loop structure brings the fluorescent dye close to the quencher molecule. This quencher absorbs the energy emitted by the dye and releases it as heat, reducing fluorescent background noise. The loop contains the targeting sequence that will hybridize with the RNA of interest. (b) The hairpin dissociates when the targeting sequence hybridizes to the target RNA. This separation moves the fluorescent dye out of range of the quencher, allowing emission of detectable fluorescence.

Figure 3.

RNAscope probe design. Two Z-shaped probes bind adjacent sequences on the RNA of interest, forming a platform-like structure across the tail sequences. Here, the preamplifier binds, and can only bind when two Z probes are bound. Amplifiers bind the preamplifier, serving as scaffolds for the fluorescent probes to bind. Each set of Z probes provides binding sites for hundreds of fluorescent probes, dramatically amplifying fluorescence on a single molecule. Z probes can be tiled along an RNA molecule to further improve detection.

Figure 4.

MS2 RNA tagging. (a) MS2 experiments require generation of two constructs. One construct encodes the RNA of interest (or a certain region, such as only the 3′UTR) with MS2 hairpins encoded downstream. When transcribed, the MS2 hairpins will form in the 3′ end of the transcript and be recognized by the MS2 binding proteins (BPs). MS2-BPs are expressed from the other construct and generally include a fluorescent protein such as GFP. (b) Following transfection, both constructs are transcribed and the MS2-BPs are translated. The ectopic transcript of interest is detected and bound by the fluorescent MS2-BPs. The number of hairpins encoded in the plasmid will determine the number of MS2-BP binding sites; cloning more hairpins will amplify the signal and improve detection. Since unbound MS2-BPs will still fluoresce in the cell, determining proper plasmid transfection ratios is essential for maximizing signal-to-noise ratios.

Figure 5.

dCas RNA tracking methods. (a) Two plasmids are required for dCas experiments. One plasmid encodes the catalytically inactive Cas protein, with mutations to inhibit any nuclease activity. Cas is also modified with two nuclear localization signals and a fluorescence domain, such as GFP or mCherry. The other plasmid contains the sgRNA scaffold, and will synthesize the sgRNA when transcribed. The targeting sequence in determined by the sequence of the sgRNA. If dCas9 from S. pyogenes is used, a separate oligomer called a PAMmer must also be transfected into the cell. The PAMmer is not necessary when using dLwaCas13a protein. (b) Structure of a fully assembled dCas9 complex. The target RNA is identified by the sgRNA targeting sequence. dCas9 requires a PAM sequence to be present on the off-target strand. Since RNA is single stranded, the PAMmer binds downstream to provide a PAM sequence for dCas9 to recognize and bind. (c) The NLSs improve signal-to-noise ratio by sequestering unbound dCas proteins. When dCas is free, or only bound to the sgRNA, the protein is held in the nucleus due to the double-NLS tag. Only fully assembled complexes, which consist of dCas, the sgRNA, the target RNA and the PAMmer (if necessary), are exported to the cytoplasm. In principle, any fluorescence observed in the cytoplasm is true signal.

Figure 6.

Fluorescent RNA aptamers. (a) Structure of an RNA aptamer. Following SELEX for an aptamer that binds the desired dye, this sequence is cloned into the 3′UTR of the transcript of interest. (b) Most dye molecules, such as DFHBI (Spinach), do not fluoresce when not bound to the aptamer. Energy is released through bond rotation and other molecular motion. When the dye is bound by the aptamer, motion is restricted. Therefore, energy must be emitted as light, improving signal-to-noise ratio and confidence in molecular detection.