PCR amplification of homologous genes using degenerate primers followed by direct sequencing is a standard approach to generate a data set for evolutionary analysis. In our laboratory, we are systematically designing and testing primers that can amplify by reverse transcription PCR (RT-PCR) a wide array of messenger RNA (mRNA) sequences across extant Panarthropoda, a diverse and ancient assemblage of taxa (i.e., tardigrades, onychophorans, crustaceans, hexapods, myriapods, and chelicerates) whose phylogenetic relationships are of interest (1). Primers (17–24 nucleotides in length) are made to protein coding-invariant regions of Basic Local Alignment Search Tool (BLAST)-aligned orthologs from human, fly, and roundworm with complete degeneracy added (typically≤256-fold) wherever synonymous changes might occur (Regier, Cunningham, Ganley, Shi, and Ball, unpublished data). Since before 1995 (2), we have used bipartite PCR primers that include a 5′ tail of nondegenerate, nonhomologous M13REV or M13(-21) sequence to facilitate direct sequencing of PCR products. In this report, we demonstrate that M13 or other nonhomologous sequences attached to the 5′ ends of degenerate primers can also greatly improve the yield of PCR amplification. This finding expands on that of Rose et al. (3,4), who used bipartite primers with degenerate 3′ ends (approximately 12 nucleotides in length) and nondegenerate but homologous 5′ ends (approximately 20 nucleotides in length) to amplify homologous sequences.

Total nucleic acids were extracted using the SV Total RNA Isolation kit, with the DNase step omitted (Promega, Madison, WI, USA) and used as template for reverse transcription. Ten microliters reverse transcription reactions occurred at 42°C for 35 min and included 5 mM MgCl2, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2 mM of each dNTP (optimal dNTP concentration varies with batch), 1 U/µL RNase inhibitor (Applied Biosystems, Foster City, CA, USA), 2.5 U/µL murine leukemia virus (MuLV) RTase (Applied Biosystems), and 4.0 pmol/µL gene-specific reverse primer or 2.5 pmol/µL dT_12–18. Fifty-microliter PCRs occurred in the same tube and included final concentrations of 2.5 mM MgCl2, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 0.4 mM of each dNTP, 0.025 U/µL AmpliTaq® DNA polymerase (Applied Biosystems), 0.035 µM Taq antibody (BD Biosciences Clontech, Palo Alto, CA, USA), 1.0 pmol/µL forward primer, and 0.4 pmol/µL gene-specific reverse primer (in addition to that already present) or 1.0 pmol/µL gene-specific reverse primer if dT_12–18 had been used for reverse transcription. In one set of experiments, the gene-specific primer concentrations were increased 3.9-fold. Gene-specific primers were made both with and without 5′ tails ((Table 1) and (Table 2)).

PTC-200™ thermal cycler (MJ Research, Waltham, MA, USA) conditions for PCR were set for Touchdown (5) with the annealing temperature set to decrease uniformly from 55° to 45°C over 25 cycles, followed by 13 more cyles at 45°C. To visualize the PCR products, equal volume aliquots of the RT-PCRs were fractionated by electrophoresis on a 1.4% agarose gel containing 0.5 µg/mL ethidium bromide and 1× TAE buffer (0.04 M Tris-acetate, 0.001 M EDTA, approximate pH 8.3). Semiquantification was performed by comparing band intensities over a range of photographic exposures.

(Figure 1) is illustrative of our results. Addition of M13REV/M13(-21) 5′ tails onto degenerate primers can increase the product yield well over 10-fold, making otherwise invisible bands visible [c.f., +M tails and -tails for 63fin2F_3R ((Figure 1), upper panel), 44fin2F_3R ((Figure 1), lower panel), and 226fin1F_4R ((Figure 1), lower panel)]. This effect is not specific to M13 sequences, as their substitution with T3/T7 tails yields similar results [c.f., +M tails and +T tails for 63fin2F_3R ((Figure 1), upper panel)]. Part of the 5′ tail effect can be accounted for by the increased length of primers [a correlate of increased melting temperature (T_m)], as evidenced by the approximately 2-fold greater PCR product yield using nondegenerate and perfectly matched gene-specific primers with 5′ tails than without [c.f., +M tails and -tails for 44Lpofin2F_5R ((Figure 1), lower panel)]. However, a comparable comparison of degenerate primers that bind to the exact same region displays a greater than 10-fold difference [c.f., +M tails and -tails for 44fin4F_5R ((Figure 1), lower panel)]. A reasonable, though as yet untested, hypothesis for this additional effect is that 5′ tails enhance amplification through incorporation of degenerate primers of greater mismatch than in the absence of 5′ tails; that is, following the initial PCR cycle, the nondegenerate 5′ tails increase the overall T_m of the bipartite primer in subsequent cycles and thereby allow a more diverse array of mismatched gene-specific primers to amplify.

Figure 1.

Increasing the overall primer concentration (with or without 5′ tails) might be another means of increasing PCR product yield. Indeed, a 3.9-fold increase in primer concentration does result in an approximate 2-fold increase in product yield when using primers with 5′ tails [c.f., +M tails and +M tails (3.9×) ((Figure 1), middle panel)], although any increase with primers lacking 5′ tails remains below the limit of detection [c.f., -tails and -tails (3.9×) ((Figure 1), middle panel)]. This illustrates that adding 5′ tails is likely to be a more effective means of increasing product yield than simply increasing primer concentration. Finally, the 5′ tail effect is likely to be specific to the PCR, since the use of dT_12–18 as a primer in the reverse transcription reaction (rather than the usual gene-specific degenerate primer) does not dampen the differential amplification during the subsequent PCR phase [c.f., dT/+M tails and dT/-tails ((Figure 1), middle panel)].

Although this report illustrates the 5′ tail effect with four degenerate primer pairs, we have performed equivalent direct comparisons between ± 5′ tails for 49 primer pairs total, representing 25 genes. Thirty-seven of the 49 primer pairs with 5′ tails amplified appropriately sized visible bands for a larger fraction of the six test taxa (see (Figure 1) legend) than the same primers without 5′ tails. For 11 primer pairs, the same number of taxa were amplified. Only one primer pair amplified more taxa without 5′ tails. Averaged across all taxa, primers with 5′ tails amplified 50% more specific bands than the same primers without tails. Furthermore, in most cases, the intensity and purity of the PCR products were greater with 5′ tails than without. We conclude that the addition of nondegenerate, nonhomologous 5′ tails to degenerate gene-specific primers is a useful tool for improving the yield of PCR products and has the added benefit of allowing for convenient direct sequencing.