( PfuTurbo ...)Comparing Fidelity and Performance of Proofreading PCR Enzymes

PfuTurbo DNA polymerase is superior to Platinum Pfx DNA polymerase

Michael Borns Janice Cline Holly Hogrefe
Stratagene

We recently evaluated the properties of Platinum Pfx DNA polymerase, a competitors new PCR enzyme formulation consisting of KOD DNA polymerase and polymerase-specific antibodies. Comparisons revealed that Stratagenes PfuTurbo DNA polymerase *^, is superior to Pfx DNA polymerase, with respect to both replication fidelity and PCR performance. The intrinsic error rate of Pfx DNA polymerase was found to be 2.7-fold higher than the average error rate of PfuTurbo DNA polymerase. Moreover, in side-by-side comparisons, PfuTurbo DNA polymerase provided higher product yields, greater sensitivity, and amplification of longer targets than Pfx DNA polymerase.

The development of high-fidelity PCR enzymes for accurate DNA replication has greatly simplified a number of laboratory procedures, including cloning, directed mutagenesis, and mutation detection. The enzyme best suited for high-fidelity PCR applications is Stratagenes Pfu DNA polymerase, the most accurate thermostable DNA polymerase described to date.^1-3 PCR reaction conditions have been optimized for highest fidelity,¹ and further improvements in PCR product yield, sensitivity, rate, and target-length capability have been achieved with the discovery of a novel PCR-enhancing factor**, referred to as the Turbo enhancer.⁴ PCR comparisons have demonstrated that PfuTurbo DNA polymerase amplifies longer targets in higher yield than Taq DNA polymerase, cloned Pfu DNA polymerase (without enhancer), and other Pfu-related proofreading enzymes, including Vent, Deep Vent, KOD, and Pwo DNA polymerases.^4,5

A competitor recently introduced a new proofreading PCR enzyme, called PLATINUM Pfx DNA polymerase. Contrary to its trade name, Pfx was derived from the Pyrococcus sp. strain KOD1 (P. sp. KOD) rather than Pyrococcus furiosus (Pfu), the source of Pfu DNA polymerase. KOD DNA polymerase was first described by Tagaki, et al.⁶, and commercialized for PCR by Toyobo. Phylogenetic analyses have shown that, although reportedly from a Pyrococcus isolate,⁶ KOD DNA polymerase is more closely related to enzymes from Thermococcus species (e.g., Vent, 9N-7) than Pyrococcus species (Pfu, Deep Vent).⁷ According to its manufacturer, the Pfx version of KOD DNA polymerase possesses hot start capability provided by neutralizing polymerase antibodies.

The manufacturers claimed that Pfx DNA polymerase exhibits greater accuracy than Pfu DNA polymerase⁸ (authors cite Tagaki, et al.⁶). Pfx was also reported to provide comparable sensitivity and higher amplification specificity compared to PfuTurbo DNA polymerase.⁸ In this report, we present the results of direct fidelity and PCR performance comparisons between PfuTurbo DNA polymerase and Pfx DNA polymerase. PCR performance is assessed with respect to product yield, sensitivity, specificity, and target-length capability.

PCR Fidelity Comparisons

PCR enzyme fidelity has been measured using a number of different methods, including DGGE analyses³ and monitoring phenotypic changes in mutational target genes (lacI¹ and p53 ²). These studies showed that Pfu DNA polymerase exhibits the lowest intrinsic error rate of any commercial thermostable DNA polymerase. For example, using a PCR mutation assay based upon the lacIOlacZa target gene,¹ Pfu DNA polymerase exhibited an average error rate 2-fold lower than that of Vent and Deep Vent DNA polymerases,¹ 3- to 6-fold lower than those of DNA polymerase mixtures,^1,9 and 6- to 12-fold lower than that of Taq DNA polymerase.^1,10 Using the same lacI PCR mutation assay, PfuTurbo DNA polymerase exhibited the same high fidelity as cloned Pfu DNA polymerase.⁴

DNA polymerase fidelity is expressed as error ratethe mutation frequency per base pair per duplication (MF/bp/d). When determining polymerase error rates in PCR-based assays, mutation frequencies [# mutants/total # clones] must be normalized with respect to the number of template doublings (d), as errors accumulate with each template duplication. The number of template doublings [d; determined as 2^d = (amount of product)/(amount of template)] is influenced by the amount of starting template, the presence of excess reactants, PCR efficiency, the presence of inhibitors, and the number of cycles performed. Therefore, to compare PCR enzyme fidelity accurately, one must compare intrinsic error rates rather than mutation frequencies, which vary from experiment to experiment.

The mutation frequency of KOD DNA polymerase was determined previously by Tagaki, et al.⁶, using both gap-filling (single primer extension) and PCR-based forward mutation assays (lacZ target gene). In the gap-filling M13 assay, the mutation frequency of KOD DNA polymerase was, as Tagaki, et al. described, similar to that of Pfu DNA polymerase.⁶ Unfortunately, PCR error rates could not be calculated from these studies as only mutation frequencies were reported.

Fig.1

To assess the fidelity of Pfx DNA polymerase, we measured the error rates of Pfx, PfuTurbo, and Taq DNA polymerases in the same assay (Figure 1). PCR fidelity measurements were performed in each manufacturers recommended PCR buffer. The mean error rate of Pfx DNA polymerase was 2.7-fold higher than the mean error rate of PfuTurbo DNA polymerase (Figure 1). Moreover, we observed no significant differences in error rate when Pfx DNA polymerase was used in cloned Pfu PCR buffer (data not show), indicating that relatively poor fidelity is an intrinsic property of KOD DNA polymerase.

PCR Yield and Target-Length Comparisons

The PCR performance of Pfx DNA polymerase was then evaluated with respect to product yield and target-length capability. PfuTurbo DNA polymerase and Pfx DNA polymerase were used to amplify a panel of complex, genomic DNA targets, ranging in length from 0.9 kb to 19 kb. Standard reaction conditions were employed, including the use of 100 to 200 ng of template, 30 PCR cycles, 1 minute per kb extension times, and room-temperature reaction assemblies (see below). Amplifications were performed under identical conditions (Methods), with the following exceptions: As per manufacturers recommendations, Pfx PCRs were carried out with 300 M each dNTP, 1.25 U of enzyme and 68C extension temperatures, while PfuTurbo DNA polymerase amplifications employed 200 M (<10 kb) or 500 M (>10 kb) each dNTP, 2.5 U (<12 kb) or 5 U (>12 kb) of enzyme, and 72C extension temperatures.

Fig.2

In all comparisons, PfuTurbo DNA polymerase produced higher product yields than Pfx DNA polymerase (Figure 2). Moreover, synthesis of the four longest 9.3-kb to 19-kb targets was achieved with PfuTurbo DNA polymerase, but not with Pfx DNA polymerase. As described , amplification of long genomic targets (>9 kb) by PfuTurbo DNA polymerase is limited by buffer components, rather than the robustness of the enzyme. Increasing dNTPs (to 500 M) and PCR buffer concentration (to 1.5X) was sufficient to allow PfuTurbo DNA polymerase to synthesize complex genomic targets up to 19 kb in length. In comparison, Pfx DNA polymerase could not successfully amplify the 17-kb and 19-kb targets in the presence of 500 M each dNTP and 0.5X to 2X concentrations of Pfx PCR buffer .

Sensitivity Comparisons

Fig.3

Next, we compared PfuTurbo DNA polymerase and Pfx DNA polymerase with respect to sensitivity. Using the standard PCR reaction conditions described above, PfuTurbo DNA polymerase synthesized a 4-kb product from as little as 10 fg of lambda DNA (Figure 3). In contrast, at least 5 ng of lambda DNA was consistently required for amplifications carried out using Pfx DNA polymerase (Figure 3, data not shown). Despite the inclusion of polymerase-neutralizing antibodies, Pfx DNA polymerase produced additional background bands in this amplification system, which were absent or greatly reduced in PCRs carried out with PfuTurbo DNA polymerase (Figure 3).

Fig.4

The superior sensitivity of PfuTurbo DNA polymerase is further demonstrated in Figure 4, where PfuTurbo DNA polymerase synthesized high yields of a 4-kb product from as little as 25 ng of genomic DNA. In contrast, Pfx DNA polymerase required at least 50 ng of genomic DNA, and product yields were significantly lower than those produced by PfuTurbo DNA polymerase (Figure 4). In studies employing a 6-kb genomic target, PfuTurbo DNA polymerase again exhibited superior sensitivity, compared to Pfx DNA polymerase (Figure 4).

Rate Comparisons

Fig.5

Finally, we compared product yields produced with PfuTurbo and Pfx DNA polymerases using shortened PCR extension times. In Figure 5, PCR reactions were carried out as described above, except that extension times of 15 seconds/kb were used. Under these conditions, PfuTurbo DNA polymerase was found to synthesize significantly higher yield of a 1.9-kb product than Pfx DNA polymerase. Thus, despite the higher processivity and polymerization rate reported for KOD DNA polymerase,⁶ Pfu DNA polymerase can produce higher product yields, in a shorter time, when combined with the Turbo PCR enhancing factor.

Room-Temperature PCR Reaction Assembly

All PCR amplifications with PfuTurbo DNA polymerase, including those described here and in other Strategies articles,^4,5 are assembled at room temperature. Previously, we showed that Pfu DNA polymerase exhibits only 0.7% and 3.9% maximal polymerase activity at 25C and 45C, respectively.¹¹ In contrast, Taq DNA polymerase exhibits 2.6% activity at 25C (room temperature) and 44% activity at 45C.¹¹ Reduced activity at low temperatures encountered during PCR is thought to contribute to higher amplification specificity achieved with Pfu DNA polymerase, as compared to Taq DNA polymerase.¹¹ Various hot start procedures have been developed for Taq-based PCR to reduce mispaired primer extension that results in the synthesis of nonspecific background. The benefits of hot start modifications (e.g., neutralizing antibodies) have not been directly demonstrated for PCR enzymes derived from hyperthermophilic archaea (e.g. Pfu DNA polymerase, KOD, Deep Vent). Although Pfx DNA polymerase is blended with neutralizing antibody, the results shown here for 10 different amplification systems, of varying lengths and complexities, indicate that PfuTurbo DNA polymerase provides comparable, if not superior (Figure 3), specificity compared to Platinum Pfx DNA polymerase.

Conclusions

Stratagenes PfuTurbo DNA polymerase is the superior choice for all high-fidelity PCR applications requiring the highest performance possible. It exhibits unparalleled PCR fidelity and performance compared to a competitors Platinum Pfx DNA polymerase. The intrinsic error rate of PfuTurbo DNA polymerase is significantly lower than the error rate of Pfx DNA polymerase. Side-by-side comparisons show that PfuTurbo DNA polymerase produces higher product yields, while exhibiting greater sensitivity and target-length capability, compared to Pfx DNA polymerase.

Methods

The PCR fidelity assay was carried out as described,¹ using cloned PfuTurbo DNA polymerase (Stratagene), Platinum Pfx DNA polymerase (LTI), and Taq2000 DNA polymerase (Stratagene). For fidelity assays, PCR amplifications were carried out using identical conditions, except that PCRs employed each enzymes recommended PCR buffer.

For PCR comparisons, amplifications employed 100 ng (<12 kb) or 200 ng (>12 kb) of genomic DNA template and 100 ng (<12 kb) or 200 ng (>12 kb) of each primer. PfuTurbo DNA polymerase amplifications employed 200 M (<10 kb) or 500 M (>10-kb) each dNTP, 2.5 U (<12 kb) or 5 U (>12-kb) of PfuTurbo DNA polymerase, and 1X (<9 kb) or 1.5X (>9 kb) PfuTurbo PCR buffer, as recommended. Pfx PCRs employed 300 M each dNTP, 1.25 U of Pfx DNA polymerase, and 1X Pfx PCR buffer, as recommended by the manufacturer. PCR reactions were conducted in a RoboCycler Gradient 96 temperature cycler (Stratagene), fitted with a Hot Top assembly, using 200-l thin-walled PCR tubes. Except as noted, the temperature cycling parameters were as follows: 1 cycle at 95C for 1 minute, followed by 30 cycles at 95C for 1 minute (denaturation); 58 to 65C for 1 minute (annealing); and 72C for 1 minute per kb of target; and 1 final extension cycle of 72C for 10 minutes. Identical cycling conditions were used for Pfx, except that a 68C extension temperature was used as recommended by the manufacturer. PCR products were electrophoresed on 1% agarose/1XTBE gels, stained with ethidium bromide and imaged using the Eagle Eye II still video system. Lanes labeled M contained the kb-DNA markers.

REFERENCES

1. Cline, J., Braman, J.C., and Hogrefe, H.H. (1996) Nucleic Acids Res. 24: 3546-3551.

2. Flaman, J-.M. Frebourg, et al. (1994) Nucleic Acids Res. 22: 3259-3260.

3. Andre, P., et al. (1997) Genome Res. 7: 843-852.

4. Hogrefe, H., et al. (1997) Strategies 10: 93-96.

5. Hogrefe, H., Bai, F., and Cline, J. (1998) Strategies 11: 36-37.

6. Tagaki, M., et al. (1997) Appl. Environ. Microbiol. 63: 4504-4510.

7. Perler, F.B., Kumar, H., and Kong, H. (96) Adv. Protein Chem. 48: 377-435.

8. Westfall, B., et al. (1999) Focus 21: 46.

9. Guide to Pfu DNA Polymerase, Stratagene, 1996.

10. Lundberg, K.S., et al. (1991) Gene 180: 1-6.

11. Nielson, K.B., Cline, J. and Hogrefe, H. (1997) Strategies 10: 40-43.

* U.S. Patent Nos. 5,545,552, 5,866,395, and 5,948,663 and patents pending
** Patents pending