Abstract
When the polymerase chain reaction (PCR) is used to amplify complex templates such as metagenomic DNA using single or degenerate primers, preferential amplification of templates (PCR bias) leads to a distorted representation of the original templates in the final amplicon pool. This bias can be influenced by mismatches between primers and templates, the locations of mismatches, and the nucleotide pairing of mismatches. Many studies have examined primer-template interactions through interrogation of the final products of PCR amplification with controlled input templates. Direct measurement of primer-template interactions, however, has not been possible, leading to uncertainty when optimizing PCR reactions and degenerate primer pools. In this study, we employed a method developed to reduce PCR bias (i.e., Deconstructed PCR, or DePCR) that also provides empirical data regarding primer-template interactions during the first two cycles of PCR amplification. We systematically examined interactions between primers and templates using synthetic DNA templates and varying primer pools, amplified using standard PCR and DePCR protocols. We observed that in simple primer-template systems, perfect match primer-template interactions are favored, particularly when mismatches are close to the 3' end of the primer. In more complex primer-template systems that better represent natural samples, mismatch amplifications can dominate, and heavily degenerate primer pools can improve representation of input templates. When employing the DePCR methodology, mismatched primer-template annealing led to amplification of source templates with significantly lower distortion relative to standard PCR. We establish here a quantitative experimental system for interrogating primer-template interactions and demonstrate the efficacy of DePCR for amplification of complex template mixtures with complex primer pools.