Amplification of GC-rich DNA sequences is still a difficult task worldwide. Two frequently seen and inexpensive reagents—ethylene glycol and 1,2-propanediol—were found to be more effective than betaine in the amplification of 104 randomly selected GC-rich human DNA sequences with GC contents of 60–80% and lengths of 700–800 bp.

Researchers have put much effort into how to conveniently amplify GC-rich sequences using additives (1,2,3,4,5,6). Novel thermal cycling patterns such as ‘slowdown PCR’ can also work efficiently for some GC-rich DNA amplification (7). Mechanisms for overcoming difficulty in the PCR amplification of GC-rich DNA templates have been developed (8,9), but amplification of GC-rich DNA sequences is still not an easy task worldwide.

Preliminary bioinformatics analysis from the Ensembl database (www.ensembl.org) indicated that the number of intragenic GC-rich DNA fragments (>65%) was 773. About half of these GC-rich sequences were <1 kb in length. 506 fragments were <2 kb, 686 fragments were <5 kb, and only 87 fragments were >5 kb (Figure 1).

Figure 1. The number, GC%, and length of intragenic GC-rich sequences in the 25 human chromosomes. (Click to enlarge)

In this report, two inexpensive reagents, ethylene glycol and 1,2-propanediol, were found to be effective for amplification of GC-rich DNA. These additives are potentially better than betaine (Figure 2).

Figure 2. Ethylene glycol and 1,2-propanediol as enhancers for GC-rich PCR. (Click to enlarge)

One hundred and four GC-rich human genomic amplicons were randomly selected (as parts of the sequences of 104 genes, Table 1) with lengths of 700–800 bp (bp) and GC% of 60–80%. The final concentrations for 25 µL PCR components were 0.3 mM dNTP, 0.16 µM each primer, 0.09 mM MgCl₂, 22.4 ng/µL human genomic DNA, and 2.5 U Taq DNA polymerase in 1× PCR buffer (Cat. no. B55; Sangon, Shanghai, China). PCR additives: 10 M betaine 5.5 µL (final 2.2 M), 1.5 µL ethylene glycol (final 1.075 M) and 1.5 µL 1,2-propanediol (final 0.816 M). The PCR cycles were performed on a TGradient Thermal Block (Biometra, Goettingen, Germany) and/or a Speed-Cycler 36 (Analytik Jena, Jena, Germany) with a program of denaturation at 94°C for 2 min; 30 cycles of 95°C denaturation for 30 s, 57°C annealing for 50 s, and 72°C extension for 50 s; and extension at 72°C for 7 min. All reactions were repeated 3–4 times to get similar results and one of the results was presented. Fifty microliters of the PCR reaction was employed in the fidelity assay.

Table 1. PCR information for 104 amplicons (700–800 bp) (Click to enlarge)

Among all 104 amplicons, 14 (13%) were successfully amplified without any additive; 75 (72%) were successfully amplified only with betaine; 94 (90%) were successfully amplied only with 1,2-propanediol; and 91 (87%) were successfully amplified only with ethylene glycol (see Supplementary Materials).

Among 104 amplicons, 101 were successfully amplified with one of the following: betaine, 1,2-propanediol, or ethylene glycol. Only three amplicons successfully amplified with betaine were not amplified well when betaine was replaced with either 1,2-propanediol or ethylene glycol individually (nos. 22, 72, 85; Table 1). Though 1,2-propanediol and ethylene glycol were not capable of replacing betaine completely, they did display better general performance. Interestingly, three pairs of primers (nos. 7, 36, 43; Table 1) failed with betaine, but succeeded with 1,2-propanediol, and failed again with both betaine and 1,2-propanediol. Three pairs of primers (nos. 29, 31, 100) failed with betaine, but succeeded with ethylene glycol, and failed again with both betaine and ethylene glycol. One pair of primers (no. 3) failed with betaine, but succeeded with either 1,2-propanediol or ethylene glycol, and failed again when betaine, 1,2-propanediol, and ethylene glycol were used together.

To test for potential interactions between the two cosolutes and Taq DNA polymerase, the amplification fidelity was assayed. The error rate of Taq DNA polymerase was reported as a magnitude between 10^-5 and 10^-6 (10). The fidelity assay indicated that betaine, ethylene glycol, and 1,2-propanediol had similar error rates to Taq DNA polymerase alone (Supplementary Table 1), indicating that the PCR fidelity was not compromised.

Previous studies have indicated that both betaine and ethylene glycol as cosol-vents were able to decrease the melting temperatures of DNA, but their effects on DNA melting enthalpy were different (11). Also, it has been shown that betaine and ethylene glycol have different affinities to ssDNA and dsDNA (12). It has also been demonstrated that ethylene glycol destabilizes the high melting region of polypeptide-bound DNA and reduces the extent of higher-ordered structure in model complexes and chromatin (13). It appears that ethylene glycol and 1,2-propanediol are able to modulate the annealing process of GC-rich templates, and the ways in which they interact with DNA are different from betaine. Further investigations are needed to understand the detailed mechanisms.