OverFlap PCR: A reliable approach for generating plasmid DNA libraries containing random sequences without a template bias

Abstract

Over the decades, practical biotechnology researchers have aimed to improve naturally occurring proteins and create novel ones. It is widely recognized that coupling protein sequence randomization with various effect screening methodologies is one of the most powerful techniques for quickly, efficiently, and purposefully acquiring these desired improvements. Over the years, considerable advancements have been made in this field. However, developing PCR-based or template-guided methodologies has been hampered by resultant template sequence biases. Here, we present a novel whole plasmid amplification-based approach, which we named OverFlap PCR, for randomizing virtually any region of plasmid DNA without introducing a template sequence bias.

Fig 1. Randomization was carried out by employing a modified whole plasmid amplification (WPA) strategy, where we utilized 1) one long reverse strand oligonucleotide (starting from the 3’ end, it contained a sequence complementary to plasmid DNA, a randomized region (represented by the red line), a stop codon, and additional sequence where the first thymidine was replaced with deoxyuridine (represented by U)), 2) one short forward strand oligonucleotide (starting from the 3’ end, it contained a sequence complementary to plasmid DNA and a sequence complementary to the reverse primer’s 5’ nucleotides (here, also, the first thymidine was replaced with deoxyuridine)), and 3) high fidelity polymerase (Phusion U) which tolerates uridine. Following amplification, the reaction mixture was treated with DpnI restriction enzyme, which cleaves only methylated (bacterial origin) DNA, and USER (Uracil-Specific Excision Reagent) enzyme mix, which excises the uridine base from amplification products and cleaves abasic sites, thus forming a ‘sticky end’.

The amplification product was then circularized and transformed into competent E.coli Dh5α strain cells, which were seeded onto an ampicillin-supplemented LB media petri dish and incubated overnight at 37°C. The next day, transformation positive colonies were quantified, washed off, inoculated in ampicillin-containing liquid media, and following overnight culturing, used to extract randomized plasmid DNA. The grey, dotted, and dashed lines represent the synthesized chain, DNA synthesis, and template DNA degradation by DpnI restriction endonuclease, respectively.

Fig 2. The principal scheme of the OverFlap strategy for whole plasmid amplification-based randomization of the selected plasmid DNA region.

The methodological approach following circularization is identical to the one described in Fig 1. However, here the plasmid is linearized in a way that the random region of the longer primer does not interact with the template, and an additional stage of asymmetric PCR with only a random region containing a primer is performed to increase the number of sequence variants within the final library. U represents deoxyuridine within the employed primers. The red, black, grey, dotted, and dashed lines represent the randomized region, template DNA, the synthesized chain, DNA synthesis, and template DNA degradation by DpnI restriction endonuclease, respectively.

Fig 3. Visualization of whole plasmid amplification (WPA) products in agarose gel electrophoresis.

a) Prior to the DpnI and USER enzyme mix treatment, and b) after treatment, indicating that the WPA process was successful and the size of the amplification product was not affected by treatment, i.e., no degradation was observed. The first line in each gel contains 2 μl of GeneRuler 1 kb DNA Ladder (Thermo Fisher Scientific, Lithuania), the uppermost band is 10 000 bp, and the three brightest bands are 6 000 bp, 3 000 bp, and 1 000 bp long. The second lane in each gel contains a 5 μl sample of the acquired reaction products.

Fig 4. Photographic pictures of the 8.8 mm diameter petri dish with the selective Amp+ media containing colonies of the E.coli Dh5α strain transformed with the circularized product.

a) Whole circular plasmid amplification (WPA) (3 802 colonies), b) OverFlapPCR-based whole plasmid amplification (OverFlapWPA) (4 534 colonies), and c) OverFlapPCR-based asymmetric whole plasmid amplification (OverFlapAsymWPA) (4 865 colonies). All the procedures were completed as described in the Materials and methods section.

Fig 5. Assembling whole plasmid amplification (WPA) Sanger sequencing capillary electrophoresis chromatograms.

Sequencing was performed using an a-Factor-Fw primer for forward strand sequencing and an M13-Fw primer for reverse strand sequencing. The fragment containing the sequence template (α-MSH (red font)) was used as a reference for this analysis.

Fig 6. Chromatograms of the sequencing library fragment sizing analysis carried out on an Agilent Bioanalyzer 2100.

WPA: whole circular plasmid amplification, OverFlapWPA: OverFlapPCR-based whole plasmid amplification, and OverFlapAsymWPA: OverFlapPCR based asymmetric whole plasmid amplification. The 35 bp (green) and 10 380 bp (purple) peaks represent the upper and lower markers employed by data analysis software for the internal calibration of each analysis.

Fig 7. Quantification of randomized region encoded peptides according to size.

As the Materials and methods section describe, in addition to randomized region the data also includes the last aa of the α-factor secretion signal CDS and stop codon. In the case of ‘1’, last aa of α-factor secretion signal CDS was missing.