Effects of the T4 bacteriophage gene 32 product on the efficiency and fidelity of DNA amplification using T4 DNA polymerase

Site size of cooperative single-stranded RNA binding by poliovirus RNA-dependent RNA polymerase

The poliovirus RNA-dependent RNA polymerase binds cooperatively to single-stranded RNA. We have determined the minimal RNA-binding site size of the poliovirus polymerase using binding titration with oligonucleotides of increasing length. A dramatic increase in affinity was observed when the length of the oligo(U) increased from 8 to 10 nucleotides (nt), arguing that the minimal size of RNA for polymerase binding is 10 nt. Another increase in affinity seen as the oligo(U) reached 24 nt suggests that a 24-nucleotide RNA can be occupied by two polymerase molecules. Direct binding of wild-type polymerase to oligo(U)12 and oligo(U)24 RNAs showed differences in affinity and cooperativity consistent with this model. The increase in binding affinity seen for oligo(U)10 suggests either that the RNA-binding determinants are widely spaced on the polymerase structure or that a substantial conformational change in the polymerase occurs upon the filling of its RNA-binding site.

Genotypic selection methods for the direct analysis of point mutations

Genotypic selection enriches a particular DNA sequence relative to another closely-related DNA sequence based only on a change of one or a few bases. This review is a survey of the genotypic selection methods that have the sensitivity to detect rare point mutations. These methods are primarily being used to study mutations caused by environmental mutagens; however, the ability to detect and measure very minor DNA sequence populations is likely to further research efforts in many fields. The approaches for allele-selection have intrinsic strengths and weaknesses, and vary greatly in sensitivity. The most sensitive method is Restriction Fragment Length Polymorphism/Polymerase Chain Reaction (RFLP/PCR) by which mutant fractions as low as 1 mutant allele in 10(8) wild-type alleles can be detected. The RFLP/PCR approach is presented as a prototype genotypic selection method. Genotypic selection methods are categorized in terms of those that (1) selectively destroy the abundant or wild-type allele, (2) selectively amplify the rare or mutant allele, or (3) spatially separate the alleles. Issues relevant to the further development of genotypic selection methods include initial DNA pool size, strategies to eliminate the bulk of extraneous DNA, the use of an internal copy number standard in quantitative PCR, the fidelity of thermostable DNA polymerases, and the effective use of PCR in linking two or more genotypic selection techniques. We conclude that proficient genotypic selection requires more than one allele-enrichment technique with at least one of these preceding a high-fidelity PCR amplification step.

Fidelity and predominant mutations produced by deep vent wild-type and exonuclease-deficient DNA polymerases during in vitro DNA amplification

Denaturing gradient gel electrophoresis (DGGE) was used to examine error rates and mutations induced by native (wt) and exonuclease-deficient (exo-) Deep Vent DNA polymerases during DNA amplification by polymerase chain reaction (PCR), in the presence or absence of the T4 bacteriophage gene 32 protein (gp32).gp32 was found to decrease the error rate of the wt, but not that of the exo-, Deep Vent. The average errors per base duplication for the native form were 8.0 x 10(-5) and 6.0 x 10(-5) in the absence and presence of gp32, respectively. For the exo- form, the error rates were 2.0 x 10(-4) and 2.2 x 10(-4) errors per base duplication in the absence and presence of gp32, respectively. Examination of mutations produced by native Deep Vent showed that A/T to G/C transition predominated, consistent with the results of our earlier studies with DNA polymerases derived from other thermophilic bacteria. These results indicate that PCR with high fidelity can be achieved by using wt Deep Vent in combination with gp32.

Template secondary structure can increase the error frequency of the DNA polymerase from Thermus aquaticus

Amplification of portions of the intergenic spacer between the katE gene and cryptic cel operon of Escherichia coli was accomplished by the polymerase chain reaction using the DNA polymerase from Thermus aquaticus. Nine different segments were amplified and cloned without error, but one 83-bp fragment was amplified with a high error rate such that 32 of 34 selected clones had three or more nucleotide changes from the expected sequence. The changes were all located in two 9-bp segments immediately adjacent to the 3'-ends of the two primers. Moving the end points of the primers to increase the spacing between them resulted in the isolation of significantly fewer error-containing products. It is proposed that stem-loop structures in the template immediately downstream from the primers interfere with an early stage of elongation and cause misincorporation. This is supported by the observation that destabilisation of one of the stem-loop structures reduced the frequency of errors.