There are a number of factors that should be taken into consideration when designing degenerate oligonucleotide primers. These will be illustrated using the isolation of chick activin type II receptors by degenerate RT-PCR as an example. The amino acid sequences of activin type II receptors from both mouse and Drosophila are shown in Fig. 2 (3). There are a number of regions that are well-conserved between these family members, and the amino acid sequences that were chosen to make primers against are highlighted, and their corresponding DNA sequences shown in Fig. 2. In this case, nested primers were employed, two external and one internal. The utilization of nested primers increases the specificity of the reaction owing to the fact that the substrate for the second and final round of PCR is considerably less complex than a total cDNA sample, since it has already been selectively amplified in the first round of PCR with the external primers. The two external primers were used in the first round of PCR, and then a small aliquot was taken, and in the second round this was subjected to further amplification using the internal primer and the 5'-external primer. The primers are spaced such that the final amplified sequence will be around 500 bp, which is a useful size, because it also allows one to use this fragment for other purposes, such as whole-mount in situ hybridization immediately after cloning it. The sequence of the primers to be used is dependent on the available amino acid sequences. The primers should be as nondegenerate as possible. This in effect means trying to avoid amino acid sequences that are comprised of amino acids encoded by multiple codons, such as arginine, serine, or leucine (Fig. 1). That having been said, an overriding consideration in primer design is that the 3' of the primers end be as close as possible to being unique. This in turn means looking for methionines or tryp-tophans in any peptide sequence, and since each of these amino acids is only encoded by one codon, making the 3'-end of the oligo correspond to these positions. If neither of these amino acids is conveniently located, then the next best option is to scan for amino acids that are only encoded by two codons, both of which will only vary at the third base. Therefore, in the case of the oligo-nucleotide which will prime the replication of the sense strand, to ensure that there is an exact match at the 3'-end the oligonucleotide should terminate at the second base of that codon. This is not a problem for the oligonucleotide that primes the replication of the antisense strand, since the two most 3'-bases of this oligo will be complementary to the first two positions of the codon, which are often invariant. As can be seen from Fig. 2, the activin receptor type II primers are all anchored on methionine and tryptophan codons and they cover sequence encoding six amino acids in all and should therefore be long enough to be stable at 72°C, which is the optimal temperature for Taq polymerase. These oligonucleotides also have restriction enzyme sites added at their 5'-ends—EcoRI in the case of the 5'-external oligo and Xbal for both the 3'-oligos—to aid the subcloning of the amplified fragment. To ensure that the restriction enzymes cut efficiently, two extra bases have been added to the 5'-end. The choice of the enzymes that are used and bases added at the 5'-end to enhance digestion are based on available information, such as that in the appendices of the New England Bioloabs catalog. It is of course possible that the amplified sequence may also contain an internal sites for one of the restriction enzymes that is chosen. If this is the case, then the PCR product can be cloned without prior digestion in the standard manner for blunt-end ligation, or it may be directionally cloned by only digesting with the enzyme that does not cleave internally and using the other end as a blunt end.
Was this article helpful?