The concentration of single-stranded DNA-binding proteins is a critical factor in recombinase polymerase amplification (RPA), as revealed by insights from an open-source system

Abstract

Recombinase polymerase amplification (RPA) facilitates rapid, exponential, isothermal nucleic acid amplification without the need for specialized equipment. Since its development in 2006, RPA has been widely applied to detect hundreds of RNA and DNA targets, spanning point-of-care diagnostics and agricultural uses. However, its reliance on pre-assembled commercial kits limits flexibility for customization. In this study, we introduce an open-source alternative to commercial RPA kits, utilizing purified, heterologously expressed proteins to circumvent the fixed molar ratios of proprietary systems. Our method incorporates enzymes from the bacteriophage T4 homologous recombination pathway-single-stranded binding protein (gp32), recombinase (UvsX), and mediator (UvsY)-along with Moloney murine leukemia virus (MMLV) reverse transcriptase with enhanced thermal stability, and Bst and Bsu DNA polymerases. We assessed the impact of buffer composition, reagent concentrations, and reaction temperature using synthetic SARS-CoV-2 genes. Notably, gp32 concentration and buffer composition emerged as critical factors in optimizing RPA performance. Using this tailored system, we demonstrated successful detection of the SARS-CoV-2 N gene on lateral flow devices (LFDs) with cDNA from eight clinical samples, achieving results consistent with RT-PCR. This open-source RPA platform provides an adaptable and cost-effective alternative for researchers, enabling the exploration of diverse experimental conditions and offering a viable solution for those without access to commercial kits.

Keywords: Point of care; Recombinase polymerase amplification; SARS-COV2.

Figure 1. The recombinantly expressed T4 homologous recombination (HR) system successfully assembled a D-loop.

(A) Schematic representation of T4 HR D-loop. During this process, single-stranded DNA (ssDNA) is coated with gp32, while the mediator protein UvsY displaces gp32 from the ssDNA and facilitates the loading of the recombinase UvsX. The loading of UvsX promotes the formation of presynaptic filaments that search for homologous regions within large double-stranded DNA (dsDNA) and execute base pairing with the ssDNA. The 3′-OH of the ssDNA oligonucleotide serves as a primer for DNA polymerases. (Panel created with BioRender.com). (B) A total of 12.5% SDS-PAGE of purified T4 HR proteins is presented. The gel demonstrates the purity and molecular weight of gp32, UvsX, and UvsY, with their relative migration indicated by arrows. (C) D-loop assembly mediated by T4 HR components was compared to that of bacterial RecA. Bacterial RecA, at a concentration of 300 nM, annealed a Cy5-labeled DNA substrate at two nM to a complementary region of a supercoiled pGEM plasmid in the presence of 5 mM ATP (lanes 2 and 3). The Cy5-labeled DNA substrate was unable to self-anneal to the supercoiled template (lanes 4 and 5). D-loop formation was assessed using two concentrations of the individual components of the T4 HR system in the presence of ATP (lanes 4 to 11). Reactions containing gp32, UvsX, and UvsY were performed at concentrations of 600 nM (lanes 4, 6, 8) or six µM (lanes 5, 7, 9, 10, 11, 12).

Figure 2. Open-source RPA system.

(A) SDS-gel showing purified Bsu DNA pol, Bst DNA pol, RT 4M, Chicken creatine kinase, and human APE1 An arrow indicates the identity of each protein. (B) D-loop extension using a supercoiled M13mp18 plasmid at a concentration of 1 nM. Agarose gel showing the use of a D-loop assembled with T4 HR proteins (lane 2) as a primer for Bsu and Bst DNA polymerases (80 nM) (lanes 3 to 6 and 7 to 10, respectively) during a timer course from 0 to 30 min of incubation. The full-length extension product of the M13mp18 plasmid is indicated by an arrow.

Figure 3. In vitro system used to evaluate RPA based on the isothermal amplification of SARS-COV2 genes.

(A) Diagram indicating the synthetic SARS-CoV-2 DNA cloned in a pET28b plasmid and the number of amplified base pairs at a specific pair of RPA oligonucleotide would amplify for each gene. The diagram also depicts the RNA promoter sequences that facilitate the generation of SARS-CoV-2-like transcripts. (B) PCR amplification results for the P, E, and N genes utilizing both free oligonucleotides and 5′-biotinylated oligonucleotides. (C) RPA reactions targeting the P, E, and N genes were conducted both in the absence and presence of 40 mM potassium acetate, using 5′-OH and 5′-biotinylated oligonucleotides. The amplified RPA products are marked with a red asterisk. (D) Assessment of the impact of varying concentrations of T4-HR proteins on RPA efficiency. RPA reactions were performed with increasing concentrations of UvsY (0, 0.47, 0.94, 1.88, 2.82, 3.76 µM), UvsX (0, 0.6, 1.35, 2.7, 4, 5.4 µM), and gp32 proteins (0, 3.33, 6.6, 13.33, 20, 26.6 µM). The relative migration of the RPA products and primer-dimers is indicated by arrows. In each standard RPA reaction, the concentration of one component of the T4-HR system was increased while maintaining the other two proteins at constant levels.

Figure 4. Effect of temperature and additives on RPA efficiency.

(A) RPA reactions assayed at three different temperatures and pHs using Bsu or Bst DNA polymerases and visualized by agarose gel electrophoresis (B) and (C) effect of different additives on RPA reactions for genes P and N. Reactions were executed at 42 °C in the presence of Bsu DNAP and 80 mM of potassium acetate. Where indicated, reactions were incubated in the presence of DMSO 5%, Trehalose 2%, Sorbitol 2%, or Triton X-100 0.1%.

Figure 5. Serial dilutions of RPA reactions.

RPA reactions using 10-fold serial dilution from one ng to 1 × 10⁻⁷ ng of plasmid DNA for genes P, E, and N respectively and their corresponding negative controls. The RPA amplifications products were run on a 1.5% agarose gel electrophoresis and visualized with ethidium bromide. A red asterisk indicates the presence of a sharp well-defined band that we interpreted as the detection limit by this method.

Figure 6. Evaluation of transcription, reverse-transcription, and selectivity test for of SARS-CoV-2 detection by RPA.

(A) Transcription reactions of SARS-CoV-2 genes RpRd, E, and N by home-made T7RNAP on linear and supercoiled plasmids (lanes 3 to 5 and 7 to 9, respectively). (B) Purification of the transcript of gene N from SARS-CoV-2 transcripts. Initial transcript of gene N and its DNAse I treatment (lanes 3 and 4). Amplification band showing cDNA synthesis of gene N of S SARS-CoV-2 by MMLV RT-4M (lane 5). (C) APE1 nuclease processing and extension products by Bsu DNA polymerase. Untreated probe harboring an abasic site (lane 1) treated with T. thermophilus Endo IV that cleaves the probe when is hybridized to a PCR or RPA amplification product of gene N (lanes 2 and 4). In the presence of Bsu DNA polymerase, the cleaved probe can be used as a primer for DNA amplification and produced amplification products that are longer than the initial probe (lanes 3 and 5).

Figure 7. Lateral flow immunoassay (LFD) using the open source T4-HR system.

(A) Graduated lateral flow strip to detect amplified SARS-CoV-2. The sample is deposited on the sample pad containing a nitrocellulose membrane. The sample migrates by capillary action, and in our case the double-stranded DNA has a biotic molecule attached at its 5′ end and the probe a fluorescein molecule also at the 5′ end. The lateral flow graduated strip contains colloidal gold-conjugated anti-fluorescein antibodies and control antibodies. The presence or absence of amplified DNA is made visually in the control band or line. (B) RPA reaction from SARS-CoV-2. The first four strips correspond to samples from positive patients (previously identified by end-point RT-PCR) and the last two are negative controls. In the positive strips there is an amplification of the SARS-Cov2 N gene that is observed in as a line in the corresponding position.