Inverse single-strand RACE: an adapter-independent method of 5' RACE.

With the advent of techniques for gene isolation based on polymerase chain reaction (PCR), there is an increasing need for methods to complete missing cDNA parts, especially in the more difficult 5′ side. While library construction and screening is always a possibility, the process is time-consuming and does not promise isolation of complete cDNAs. For this reason, several PCR-based methods have been developed to approach isolation of missing 5′ sequences of cDNAs, collectively known as 5′ rapid amplification of cDNA ends (RACE) and reviewed extensively (1,2,4–7). Some of the published methods use inverse PCR but require the generation of double-stranded (ds)cDNA and therefore involve more steps and are more complicated (6,7). Several other methods are based on single-stranded (ss)cDNA (2,4,8); however, they require either nucleotide tailing or ligation of an artificial sequence (an adapter) to the cDNA as an “anchor” for PCR. This step introduces a complication in the protocol, which may hamper successful isolation of the missing sequences. We have developed a modification of the single-strand 5′ RACE method, which does not involve adapters and is therefore simpler and more prone to success. While this method still suffers from the relatively low efficiency of the T4 RNA ligase (compared to DNA ligase in the dscDNA methods), the protocol presents an improvement to previous protocols due to its simplicity. We use the method routinely to complete missing 5′ cDNA sequences. Isolated lengths varied between 230–1040 bp, and the genes involved include: (i) a citrus-style, S-like RNase cDNA (unpublished results); (ii) a citrus rhamnosyl-transferase cDNA (Frydman, BarPeled, Fluhr, Gressel and Eyal, unpublished results); (iii) a tomato karyopherin-α cDNA (3); and (iv) a putative terpene synthase cDNA from citrus (Naaman and Eyal, unpublished results). A brief overview of the approach and its schematic representation appear in Figure 1. To reverse transcribe single-strand, gene-specific cDNA, mix and bring to a volume of 17 μL: 50–500 ng poly(A) RNA (amount of transcript needed will depend on the representation of the specific mRNA in the population) and 5 pmol phosphorylated gene-specific primer1-antisense (1AS). We phosphorylate the primer using polynucleotide kinase in standard reaction conditions. Heat to 65°C for 10 min and quickly chill on ice. Add and bring to a volume of 29 μL: reverse transcriptase buffer (we use buffer supplied with SuperScript II; Life Technologies, Gaithersburg, MD, USA), 4 μL 10 mM dNTP mixture and 2 μL 100 mM dithiothreitol (DTT). Incubate 2 min at 42°C before adding 200 U reverse transcriptase (we use SuperScript II) and continue incubation for 1 h. Inactivate enzyme at 70°C for 15 min. To hydrolize mRNA and purify sscDNA: first, RNA is hydrolized by adding 2 μL 6 N NaOH to the cDNA tube and incubating at 65°C for 30 min. The pH is neutralized by adding 2 μL 6 N acetic acid, and then sscDNA is purified from hydrolized RNA and primer using silica-gel particles by a standard DNA desalting protocol (we use QIAEX II Gel Extraction System; Qiagen, Valencia, CA, USA). While short DNA and RNA fragments remain bound to the silica-gel, cDNA is eluted in 20 μL water. To self-ligate sscDNA, mix and bring to a volume of 30 μL (diluted conditions favor self-ligation over concatamers): 8 μL sscDNA, 3 μL 10× T4 RNA Ligase Buffer and 20 U T4 RNA Ligase (both from New England Biolabs, Beverly, MA, USA) and ligate at room temperature overnight. Benchmarks