vol 687 Chapter 2 采用DNA聚合酶融合技术进行长片段PCR扩增 （精华）
丁香园
PART II ：Cloning and Sequencing
Long-Range PCR with a DNA Polymerase Fusion
Holly H. Hogrefe and Michael C. Borns
Abstract
Proofreading DNA polymerase fusions offer several advantages for long-range PCR, including faster runtimes and higher fidelity compared with Taq-based enzymes. However, their use so far has been limited to amplification of small to mid-range targets. In this article, we present a modified protocol for using aDNA polymerase fusion to amplify genomic targets exceeding 20 kb in length. This procedure overcomesseveral limitations of Taq blends, which up until recently, were the only option for long-rangePCR. With a proofreading DNA polymerase fusion, high-molecular-weight amplicon can be generatedand analyzed in a single day, and a significant proportion is expected to be error-free.
Key words: Long PCR, Long-range PCR, Fusion DNA polymerase, Uracil
1. Introduction
Long-range PCR is used to prepare specific high-molecular-weightDNA fragments for a variety of applications, including cloning,genome mapping and sequencing, and contig construction.Routine amplification of large genomic targets remains challenging,despite the availability of specialized long-range PCR enzymes.The majority of researchers use Taq blended with a small amountof proofreading enzyme (Taq blend) to amplify targets greater than10 kb in length (1). Such blends, however, exhibit several disadvantages,including: (1) low fidelity which leads to a high percentageof error-containing clones (e.g., 100% of 10 kb fragmentsare expected to contain at least one error; Table 1); and (2) longPCR extension times that can damage DNA and lead to run timesexceeding 12 h.In this chapter, we describe an improved method for longrangePCR that provides faster, more accurate amplification oflarge genomic targets. The protocol features a proofreading DNA polymerase fusion (2) rather than a Taq blend, a modified cyclingprotocol designed to minimize PCR run time, and a single PCRadditive (DMSO) to facilitate replication of difficult GC-richsequences. In the example provided, we use Herculase II fusionDNA polymerase (Herculase II), which consists of Pfu DNApolymerase fused to a double-stranded DNA binding protein(Pfu fusion) and a thermostable dUTPase. Compared with wildtypeenzymes, Pfu fusion exhibits higher processivity (185 basesversus 10–15 bases for Pfu or 10–42 bases for Taq (3, 4)), whichallows researchers to use shorter extension times and performlong-range PCR in half the time required for Taq blends (Table 1).In addition to speed, the Pfu-only Herculase II enzyme providessuperior fidelity compared with Taq blends (4–6), ensuring that asignificant proportion of amplified DNA is error-free (Table 1).
Table 1
Comparison of long-range PCR enzymes
Herculase II is recommended over other DNA polymerasefusions because the formulation includes a thermostable dUTPasethat enhances length capability. Decxyuracil is a potent inhibitorof archaeal DNA polymerases (e.g., Pfu, Vent, and DeepVent)that arises during PCR by deamination of dCTP to dUTP, andsubsequent incorporation of dUTP into amplicon (7). ArchaealDNA polymerases possess a unique “read-ahead” function, andstall when decxyuracil is encountered in the template strand (8).Uracil poisoning of archaeal DNA polymerases (alone or in Taqblends) has significant implications for long-range PCR due tothe use of long extension times. Prolonged exposure of nucleotidesto heat leads to increased formation of dUTP, incorporation ofuracil in early PCR cycles, and reduced amplification efficiency (9).In the Herculase II formulation, dUTPase is added to eliminatedUTP by conversion to dUMP and PPi. In the absence of dUTPase,all long-range PCR enzymes suffer from uracil poisoning,which can be largely overcome in Taq blends by reducing theproportion of proofreading enzyme, at a cost to fidelity (errorrate comparable to Taq alone (5)).
The protocol provided below has been used to amplifygenomic targets up to 23 kb in length. We have included instructions for amplification of a broad range of target sizes, including bothmid-range (1–10 kb) and long (>10 kb) DNA fragments. Weprovide tips for isolating suitable quality genomic DNA, as well asPCR primer sequences for an 18.4 kb b-globin fragment that canbe used to optimize long-range PCR procedures in individuallaboratories (Fig. 1).
Fig. 1. Optimization parameters for long PCR with a DNA polymerase fusion. An 18.4 kb b-globin fragment was amplifiedin duplicate reactions from human genomic DNA (Promega G3041) using recommended reaction conditions and cyclingparameters (see Subheading 3). Adjustments in DNA template, primer, nucleotide, and DMSO concentrations are indicated.Primer sequences are: (Forward) 5? CACAAGGGCTACTGGTTGCCGATT and (Reverse) 5' CCTGCATTTGTGGGGTGAATTCCTTGCC.In this example, highest product yields were achieved under two different sets of conditions: (lanes 7–8) 250 ng DNA,100 ng each primer, 250 mM each dNTP, and no DMSO; and (lanes 17–18) 250 ng DNA, 200 ng each primer, 500 mMeach dNTP, and 3% DMSO.
2. Materials
1. PCR components. Commercial long-range PCR enzymestypically come with reaction buffer and one or more enhancersor additives to facilitate amplification of GC-rich targetsequences. Herculase II fusion DNA polymerase (AgilentTechnologies – Stratagene Products) includes 5× reactionbuffer (provides a final 1× Mg2+ concentration of 2 mM) anddimethylsulfoxide (DMSO). Researchers need only providedeoxynucleotides, primers, and DNA template of sufficientquality for long-range PCR.
2. DNA template. The most critical component of long-rangePCR is the DNA template. Great care should be exercised toisolate intact, high-molecular-weight (>50 kb) genomic DNA.Suitable quality templates are routinely isolated using the DNAExtraction Kit or RecoverEase DNA Isolation Kit (AgilentTechnologies – Stratagene Products). Potential shearing ofgenomic DNA template is minimized by using wide-bore tipsfor pipetting or mixing of the template. High-molecular-weighttemplates should be stored at 4°C, and not frozen.
3. PCR primers. Primers for long-range PCR should be at least23 bp in length, and ideally gel- or HPLC-purified. Primerpairs should be designed with balanced melting temperatures(Tm) of at least 60°C. The resulting high-annealingtemperature promotes specificity and discourages secondarystructure formation. Standard primer design rulesshould be applied, including analyzing primer sequencesfor potential duplex and hairpin formation as well as falsepriming sites, to ensure the highest yield of specific PCRproducts.
4. Thermocycling and analysis of PCR products. Cycling conditionsfor long- range PCR (Subheading 3.2) were developedfor single-block temperature cyclers, including the DNAEngine PTC-200 (BioRad), and GeneAmp PCR System9700 (ABI). Cycling parameters are not necessarily transferablebetween thermal cyclers designed by different manufacturers,and although any thermocycler is suitable forlong-range PCR, further optimization may be required forother instruments. After cycling, PCR products up to 20 kbin length can be analyzed by gel electrophoresis using 0.6–0.7% agarose gels, stained with ethidium bromide and visualizedon a suitable UV imaging system.
3. Methods
3.1. PCR Set-Up
The following protocol is recommended for amplifying targetsgreater than 10 kb in length. See Note 1 to modify protocol foramplification of targets <10 kb.
Prepare 50 ml PCRs in sterile thin-walled PCR tubes. Add reagentsin the order listed (see Note 2). Vortex gently.
3.2. PCR Cycling
If the temperature cycler lacks a heated cover, or if extensiontimes are >15 min, overlay each reaction mixture with 50 ml mineraloil. Perform PCR using the cycling conditions provided below, which were developed for single-block temperature cyclers(see Note 3 and Subheading 2, item 4).
3.3. Analysis of PCR Products
PCR products are analyzed by gel electrophoresis using an appropriatepercentage agarose gel. According to Maniatis, the effectiverange of separation of linear DNA molecules is 0.8–10 kb fora 0.7% agarose gel, and 1–20 kb for a 0.6% agarose gel (10). Formaximum separation and resolution, we recommend pulse fieldgel electrophoresis with a 1.0% agarose gel.
4. Notes
1. Modified protocol for shorter targets. Herculase II can also beused to amplify targets less than 10 kb in length. Reactionsshould be set up according to the following protocol:
Cycle shorter targets as follows:
2. Optimization of reaction components for long-range PCR. AllPCR amplification reactions require optimization to achieve the highest product yield and specificity. Critical optimizationparameters for successful amplification of long targets areoutlined below, in order of priority. As shown in Fig. 1, optimizationof DNA template, primer, and DMSO concentrationscan have a significant impact on PCR product yield.
(a) DNA template. Successful amplification of long targets isdependent on the purity, integrity, molecular weight, andconcentration of the DNA template. As discussed inSubheading 2, item 2 above, great care should be takento isolate intact, high-molecular-weight (>50 kb) genomicDNA. Yields are generally improved by increasing theamount of genomic DNA, although excess DNA templatecan be inhibitory. We recommend titrating genomicDNA over the range of 150–400 ng for targets greaterthan 10 kb in length. When amplifying low-complexitytargets (e.g., lambda or plasmid DNA), input DNAshould be titrated from 15 to 60 ng in a 50 ml reactionvolume.
(b) Primers. PCR primers should be purchased from a reputablevendor, and for best results, gel- or HPLC-purified. Product yield is generally improved by adjusting the ratioof primer versus template. As a starting point, we recommendusing 0.5 mM each primer for targets >10 kb,which is equivalent to approximately 200 ng of a 25 baseoligonucleotide primer in a 50 ml reaction volume.
(c) DMSO. DMSO facilitates amplification of extra-long orGC-rich targets by destabilizing secondary structures inthe DNA template that impede polymerization or primerannealing. DMSO may increase DNA polymerase errorrate slightly (<50% increase with 3% DMSO), so its useshould be avoided in cases where there is no benefit to yieldor specificity. The DMSO concentration should be titratedfor each primer-template system, as the degree of improvementwill vary according to target length, complexity, andGC content. As a guideline, we recommend titratingDMSO in 1% increments between 0 and 3% for targets upto 20 kb, and between 3 and 6% for targets >20 kb. DMSOconcentrations up to 8% may enhance amplification of targetscontaining GC-rich sequences. DMSO reduces primerTm, and reoptimization of PCR annealing temperature maybe required for maximum yield.
(d) Nucleotides and magnesium. Nucleotide concentrationsof 250 mM each are sufficient for amplifying targets up to20 kb in length, and further optimization is not usuallyrequired. The Herculase II reaction buffer provides themagnesium ion concentration that is optimal for theenzyme (final 1× Mg2+ concentration of 2 mM). Increasing the magnesium concentration may lower fidelity (11) orspecificity, and should be avoided.
3. Optimization of cycling conditions for long-range PCR. Key modificationsto standard protocols include lowering the denaturation(from 95 to 92°C) and extension (from 72 to 68°C)temperatures to minimize thermal damage to the DNA template.With DNA polymerase fusions, short extension times of30 s per kb are maintained for the first ten cycles, after which20 s are added to the total extension time at each of cycles11–30 (e.g., cycle 11, 30 s/kb + 20 s; cycle 12, 30 s/kb + 40 s,and so on). Modifying extension parameters in this fashionprovides significant improvements in the yield of long targets,while overall run times are increased by not more than 25%.
Cycling parameters provided in Subheading 3.2 may requirefurther optimization depending on the primer-template set andthermal cycler used. For example, raising or lowering the annealingtemperature may provide further improvements in specificity oryield, respectively. GC-rich targets, which may be difficult to melt,typically require the use of more stringent denaturation conditions(e.g., 98°C for 40 s) and higher extension temperatures(72°C). Finally, the duration of the denaturation, annealing, andextension steps may require further adjustment depending on theramp rate of the thermal cycler used.
References
1. Barnes, W.M. (1994) PCR amplification of upto 35-kb DNA with high fidelity and high yield from l bacteriophage templates. Proc. Natl. Acad. Sci. USA 91, 2216–20.
2. Wang, Y., Prosen, D.E., Mei, L., Sullivan, J.C.,Finney, M., and Vander Horn, P.B. (2004) A novel strategy to engineer DNA polymerases forenhanced processivity and improved performancein vitro. Nucleic Acids Res. 32, 1197–207.
3. Hogrefe, H.H., Cline, J., Lovejoy, A., andNielson, K.B. (2001) DNA Polymerases from Hyperthermophiles. Methods Enzymol. 343, 91–116.
4. (2006) Herculase II fusion DNA polymerase product literature. Strateg. Newsl. 19, 4–5,34–35.
5. Hogrefe, H.H. and Borns, M.C. (2003) High Fidelity PCR Enzymes, PCR Primer, Dieffenbach, C.W., and Dveksler, G.S., eds., Cold Spring Harbor Laboratory Press, Plainview, NY.
6. Borns, M. and Hogrefe, H. (2000) Unique DNApolymerase formulation excels in a broad rangeof PCR applications. Strateg. Newsl. 13, 1–3.
7. Hogrefe, H.H., Hansen, C.J., Scott, B.R.,and Nielson, K.B. (2002) Archaeal dUTPase enhances PCR amplifications with archaeal DNA polymerases by preventing dUTP incorporation.Proc. Natl. Acad. Sci. USA 99,596–601.
8. Greagg, M.A., Fogg, M.J., Panayotou, G.,Evans, S.J., Connolly, B.A., and Pearl, L.H. (1999) A read-ahead function in archaeal DNA polymerases detects promutagenic template-strand uracil. Proc. Natl. Acad. Sci. USA 96, 9045–50.
9. Arezi, B., Xing, W., Sorge, J.A., and Hogrefe,H.H. (2003) Amplification efficiency of thermostableDNA polymerases. Anal. Biochem.321, 226–35.
10. Maniatis, T., Fritsch, E.F., and Sambrook, J.(eds.) (1982) Molecular Cloning: A LaboratoryManual. Cold Spring Harbor Laboratory.Cold Spring Harbor, NY.
11. Cline, J., Braman, J.C., and Hogrefe, H.H.(1996) PCR fidelity of Pfu DNA polymeraseand other thermostable DNA polymerases.Nucleic Acids Res. 24, 3546–51.
