TechNotes 7(3): Get the 5'-End

Most cDNA synthesis reactions performed for cDNA library construction are initiated by oligo (dT) primed reverse transcription. This insures the inclusion of the 3'-end of most mRNA species in the library, and helps to orient the clones during subsequent analysis. However the majority of clones are truncated at their 5'-ends. This is due partly because it is difficult for reverse transcriptase to negotiate long RNA templates and stable secondary structure often encountered in RNA molecules. The predominant method for synthesizing double-stranded cDNA for cloning also contributes to under representation of the 5'-ends of mRNAs in cDNA libraries. Typically, RNase H is used to fragment the mRNA template and the fragments serve as primers for second strand synthesis. (1,2). Statistically, the further towards the 5'-end a sequence lies, the less likely a successful priming event will occur upstream from that end to make the region double-stranded and cloneable. In other words, you get only one chance to prime at the extreme 5'-end of the cDNA and make the 5'-end double-stranded. One the other hand, priming anywhere along the length of the cDNA produced a double-stranded 3'-end. Taken together, these factors make retrieving the 5'-end of a transcript from a cDNA library difficult and sometimes impossible.

PCR can be used to facilitate isolation of 5'-ends of mRNA by several similar methods collectively termed Rapid Amplification of cDNA Ends, or RACE. RACE involves performing a randomly-primed reverse transcription (RT) reaction, adding an adapter to the 3'-end of the synthesized cDNA (the 5'-end of the gene sequence) by ligation or polymerase extension, and amplifying by PCR with a gene specific primer and a primer that recognizes the adapter sequence. While RACE can produce results in a relatively short time, the procedure frequently yields sequence information exclusively from truncated RT products. This occurs because RACE has no mechanism to prevent premature termination of cDNA synthesis. In fact, because RACE is PCR-based, the technique is very efficient at selectively amplifying the shortest targets in a mixed population.

In order to add selectivity to RACE, several variations to the basic procedure have been developed. The most promising is a method of positive selection for amplification products that contain the true 5'-end of the desired mRNA. The best of these second-generation RACE techniques is RNA-Ligase-Mediated RACE, or RLM-RACE (3), and it is offered in kit form as the FirstChoice™ RLM-RACE Kit from Ambion. As illustrated in Figure 1, an RNA sample containing the target of interest is treated with Calf Intestinal Phosphatase (CIP) to remove the 5'-phosphate from all RNA species except those that have a cap structure (present on all Pol II transcripts). Molecules that are dephosphorylated in the CIP reaction include rRNA, tRNA, and most importantly, fragmented mRNA that does not contain the 5'-end. Tobacco Acid Pyrophosphatase (TAP) is then used to remove the cap structure from mRNA, leaving a 5'-phosphate. Next a synthetic RNA oligonulceotide is added and ligated to the CIP/TAP-treated RNA. This adds an adapter only to the decapped mRNA — no ligation occurs to dephosphorylated molecules. Ambion's optimization of both the ligation reaction and the adapter sequence yields an extraordinarily high efficiency RNA ligation reaction. The chimeric RNA is then reverse transcribed using random decamers as primers. If the RT reaction extends to the natural 5'-end of an RNA, it will incorporate the adapter sequence into the first strand cDNA. Reactions that do not incorporate the adapter sequence (e.g. due to incomplete synthesis or lack of an adapter on the template RNA) will not serve as template for PCR amplification. Nested PCR is performed using gene specific primers and adapter primers. Figure 2 shows the amplification of the 5'-end of the mouse CXCR4 mRNA and Xenopus TGF-ß related mRNA from mouse liver and Xenopus stage 41 embryos using Ambion's RLM-RACE procedure.

Figure 2. FirstChoice™ RLM-RACE for Mouse CXCR4 Gene and Xenopus TGF-ß Related Gene. Total RNA from mouse liver and Xenopus embryos (stage 41) were analyzed using Ambion's RLM-RACE Kit. Initial PCR (A) and nested PCR (B) products were analyzed on a 2% agarose gel with EtBr staining. The results indicate that a product of the expected size is amplified after nested PCR. Note that CXCR4 is a moderately expressed message while the TGF-ß related gene is an extremely rare message.

Most often, the nested reaction produces a single discrete band that is easily cloned. Occasionally, for very rare transcripts, Ambion suggests starting with oligo dT-selected RNA to increase the percentage of target in the sample. Additionally, if the distance between your primers and the 5'-end of the sequence is more than 1.5 kb, Ambion recommends using SuperTaq™ Plus DNA polymerase in order to maximize efficiency of the PCR reactions.

Ambion's FirstChoice RLM-RACE Kit contains everything you need for RLM-RACE, and the entire procedure can be completed in a single day. No optimization of the CIP, TAP, or ligation reactions is necessary. However optimization of annealing temperatures in the nested PCR reactions is suggested. The kit includes reagents for 100 nested PCR reactions to allow for this optimization. In addition, mouse thymus control RNA and primers for mouse CXCR4 are included to validate kit reagents and protocols.

SuperTaq and SuperTaq Plus are a trademark of and are manufactured by Enzyme Technologies, Ltd in the UK and is distributed in the USA by Ambion, Inc. See inside front cover for licensing information regarding PCR.

An adage that holds true in life as well as in science is that the intrinsic worth of an item is often proportional to the difficulty involved in obtaining it. Although difficult to achieve, isolation and analysis of the 5' end of a cDNA provides extremely valuable and essential information. First, the transcription start site of an mRNA relative to the corresponding genomic locus can be mapped. This information is essential for the analysis of the promoter of a gene, part of which usually lies immediately upstream of the transcription unit. Second, determining the transcription start site of a gene facilitates investigation of heterologous transcription start sites and alternative first exon usage. Finally, the first exon of an mRNA typically includes both a non-coding region upstream of the initiator codon and the translation start site itself. These facts are often required to determine the protein sequence encoded by an mRNA. The value of locating the putative start site for translation is obvious, but without knowledge of the 5' untranslated sequence, it is difficult to be certain that 'an' AUG in the cDNA sequence is 'the' AUG start site. The upstream, untranslated region will typically contain stop codons in-frame with the genuine open reading frame, confirming their non-coding identity and supporting the role of a downstream AUG as the genuine translation start site.