Source: Pablo Sanchez Bosch², Sean Corcoran² and Katja Brückner^1,2,3
1Eli and Edythe Broad Center of Regeneration Medicine and Stem Cell Research
²Department of Cell and Tissue Biology, 
³Cardiovascular Research Institute, University of California San Francisco, San Francisco, CA, USA    

Rapid Amplification of cDNA Ends (RACE) is a technique that allows amplification of full-length cDNA from mRNA by extending to the 3’ or 5’ end, even without prior knowledge of the sequence (Frohman et al., 1988). In contrast to regular PCR, it uses only one specific PCR primer and a second non-specific primer that will indiscriminately bind to most mRNAs. Depending on whether the area to be amplified is on the 3’ or 5’ end of the mRNA, the second primer is chosen to bind the polyA tail (3’ end) or a synthetic linker added to all transcripts (5’ end). These primer combinations are known to yield “one-sided” or “anchored” PCR due to PCR amplification using the known sequence of one side- 3’ or 5’(Ohara et al., 1989). The approach allows the capture of gene-specific rare mRNAs that otherwise would be hard to detect, e.g. because of their relatively low expression or unknown complete sequence (Frohman et al., 1988).

RACE is initiated with a reverse transcription step (RT) to synthetize single stranded complimentary DNA (cDNA) from mRNA. This is followed by two consecutive PCRs that amplify the cDNA fragments from a gene of interest. To perform RACE, the sequence of the gene of interest must be at least partially known, as it is needed to design the gene-specific primers (GSPs; Frohman, 1994; Liu and Gorovsky, 1993). As the second primer pair is a generic primer that will anneal to all transcripts present in the sample, the specificity of the PCR reactions is reduced. RACE is therefore ideally performed with two nested PCRs to lower the chances of amplifying non-specific transcripts (Figure 1). Depending on the direction of amplification relative to the GSP, the technique is categorized as 5’ or 3’ RACE (Figure 1). 

3’ RACE (Figure 1A) takes advantage of the poly(A) tail present in mRNAs as a generic binding site for the non-specific primer. This simplifies the technique, as one can synthetize cDNA by using oligo (dT) primers. GSPs are located in the 5’ region of the transcripts. This RACE variant allows the detection of transcript variants and different 3’ UTRs (Scotto Lavino et al., 2007a). In 5’ RACE (Figure 1B), GSPs are located in the 3’ region of the gene. To be able to use a non-specific primer that binds to all transcripts, an adapter that serves as generic primer binding site is attached to the 5’ RNA ends. To attach the adapter to the mRNA, the 5’ cap that protects mRNAs against exonucleases and promotes translation must be removed (Bird et al., 2016). Opposed to 3’ RACE, 5’ RACE helps finding differential 5’ splice variants and alternative 5’ UTRs (Scotto Lavino et al., 2007b).

Here we will perform 3’ RACE to identify and isolate different transcript variants using the known 5’ sequence (Figure 2A) encoded by the Drosophila dSmad2 (Smox) gene. dSmad2, a fly Smad protein, is an important transducer of Activin-β signaling, a pathway of the TGF-β superfamily. dSmad2 regulates multiple cellular processes, such as cell proliferation, apoptosis and differentiation (Upadhyay et al., 2017). As differentially spliced transcripts of dSmad2 may have different functions in the adult fly, a first step in exploring these potential functions is to assess all transcript variants of dSmad2 at this developmental stage.

1. Experiment set up for 3’ RACE

1. Design a 5’ specific primer for a gene of interest (gene specific primer 1, GSP-1). It must be highly specific, as it is the only primer that introduces specificity. Therefore, a relatively long primer would be preferable (typical length around 24 nucleotides, Tm ranging from 55-65˚C). 
2. Design a second, nested primer (GSP-2), i.e. the primer should be located 3’ of GSP-1. This step is optional, but performing a second PCR with a nested pri

There are two annotated transcripts for Drosophila dSmad2 in FlyBase (Fig. 2A). Our results, however, reveal three different transcripts for dSmad2, ranging in size from 750 bp to 1400 bp (Fig. 2B). Differences in the intensity of the bands indicate their relative expression levels. Among the predicted products, one transcript is predominant (Fig. 2B, black arrow A), and one is expressed at a lower level (Fig. 2B, black arrow B). In addition, a previously undescribed smaller product was detected (Fig. 2

RACE provides a quick, inexpensive and powerful tool to obtain cDNAs from sequences that are only partially known, or from rare transcripts that are otherwise harder to amplify. It can be used to either find the sequence of unknown transcripts from an already known gene or to clone such transcripts for further studies. Following RACE, such transcripts can be overexpressed in cell-based systems or model organisms to investigate their function in vivo.