Introduction

Development of on-site diagnostic methods has recently attracted attention for the timely detection and treatment of diseases. Sensitive and specific diagnostic results achieved by an on-site diagnostic method would help to provide more appropriate treatment to the patient at an earlier stage of the disease, providing considerable socioeconomic benefits.

Among diagnostic methods, DNA-based molecular diagnostic methods such as polymerase chain reaction (PCR) are more accurate than other serology-based diagnostic methods [1,2,3]. Antibody detection methods have limitations; for example, they cannot discriminate previously infected patients from currently infected patients. Hence, the use of DNA-based molecular diagnostic methods in various areas including human health, animal health, and food safety has become more popular [2,3,4,5]. However, standard laboratory reagents and equipment have limitations for on-site use due to the need for a “cold chain” to ensure the stability of reagents containing protein-based biomolecules and the need for electricity for device operation.

Proteins are large molecules with a complex secondary and tertiary structure prone to chemical and physical degradation in solution, which can result in a loss of protein activity [6, 7]. The use of a protein reducing agent is crucial to disrupt disulfide bonds in various protein preparations [8, 9]. Tris(2-carboxyethyl)phosphine (TCEP) is a stable and effective reducing agent because it does not react with other functional groups found in proteins [10,11,12]. Freeze-drying technology has been applied to protein molecules to ensure the retention of stable forms, which is critical to their safety and efficacy [13, 14]. Freeze-drying of the biomolecules can be one of the prerequisites for the decentralized use of the biomolecule-containing reagent in the field. To obtain optimal conditions for freeze-drying, various stabilizing additives and lyoprotectants are added in the freeze-drying formulation and process [7, 15, 16]. Tertiary butyl alcohol (TBA) is frequently used as a freeze-drying medium because of its physico-chemical properties of low toxicity, high vapor pressure, high melting temperature, and miscibility with water [17, 18].

Convection PCR (cPCR) is a new technology [19, 20] that involves the use of three heating plates for denaturation, annealing, and polymerization to generate convection in the PCR tubes. This method does not require ramping between temperatures as in conventional thermocyclers, and thus the reaction time is dramatically reduced. Recently, several articles regarding the applications of cPCR technology have been published [4, 5].

In this study, to move closer to decentralized on-site molecular diagnostics, we chose two chemicals are representative of the antioxidant and freeze-drying solvent chemical groups which are essential to be used during freeze-drying of biological molecules. And we investigated the effect of TCEP, a reducing agent and TBA, freeze-drying medium on the performance of cPCR using a battery-operated hand-held device.

Materials and methods

Human HEK 293 cell culture and lyophilization chemical treatment

Human HEK 293 cells (ATCC® CRL-1573) were maintained in the logarithmic phase of growth in Dulbecco’s modified Eagle’s medium (DMEM; Welgene, USA) supplemented with 5% fetal bovine serum (FBS; Gibco BRL, USA), 5% horse serum, 2 mM l-glutamine, and antibiotics. Cultures were maintained at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Logarithmically-growing HEK 293 cells were used in all experiments. We have tested treatment of TCEP and TBA in cPCR mixture. TCEP and TBA were obtained from Sigma-Aldrich (646547 and 471712, St. Louis, MO, USA). Chemicals were stored at − 20 °C. When used, chemicals were diluted in distilled water and used at room temperature.

Preparation of genomic DNA samples

Genomic DNA (gDNA) from HEK 293 cells was extracted using the DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. After extraction, gDNA concentration was measured using a NanoDrop spectrophotometer (Thermo Scientific, Waltham, MA, USA). gDNA samples were serially diluted to prepare samples with designated gDNA concentrations. Copy numbers of gDNA in the samples were calculated from 1 ng of DNA, based on the molecular weight of double-stranded DNA and chromosomal DNA size (http://scienceprimer.com/copy-number-calculator-for-realtime-pcr), as 2.9 × 102 copies/ng HEK 293 cells.

Primer design and cPCR

Human gDNA specific primers were designed using two different gene segments. The 249 primers (forward: 5′-CCA AGC TGG AGT TCT CTA TTT ACC-3′, reverse: 5′-GGG CTC CAT CAA ATC TCA GG-3′) were designed to detect human chromosome DNA within the TUBA1A gene (tubulin alpha 1a, GenBank accession number NC_000012.12) and the 500 primers (forward: 5′-GGA CAC ATG CTC ACA TAC GG-3′, reverse: 5′-AGG TCC TGT AAT GGG TCC TG-3′) were designed to detect human chromosome DNA, HBE1 and OR51B4 genes (hemoglobin subunit epsilon and olfactory receptor 51B4, GenBank accession number NC_000011.10). The primers were chosen to have the size of approximately 250 bp and 500 bp. The cPCR mixture (20 µl) contained 1⋅ PalmTaq HS buffer (including 1.5 mM MgCl2), 0.2 mM dNTPs, 0.4 U PalmTaq High-speed DNA polymerase (Ahram Biosystems, Seoul, Korea), and 10 µM each of primers for single or multiple PCR detection. The mixture was treated with TCEP and/or TBA at various concentrations. Unless otherwise stated, 10 ng (2.9 × 103 copies) of human gDNA was used as a template. PCR was performed with a battery-operated convection thermal cycler Palm PCR device (G2-12, Ahram Biosystems, Seoul, Korea). The speed level was set to T1 to amplify the 249 bp amplicons or F3 to amplify the 500 bp amplicons. The annealing temperature was set to 56 °C. PCR reactions were run for 25 cycles in 18 min at T1 speed mode or 20 min at F3 speed mode, unless stated otherwise. All experiments were performed at least in triplicate. Upon completion of the cPCR, an aliquot of the PCR mixture was analyzed by 1.5% agarose gel electrophoresis (25 min at 135 V). PCR products were visualized with ethidium bromide staining and an imaging system (Ultra-Lum, Carson, CA, USA). The intensity of each PCR product on the agarose gel was analyzed by Image J software.

Results

Effect of single and combined treatment with TCEP and TBA on convection PCR

First, we observed the effect of TCEP on cPCR performance. Here, we define 50% cPCR performance when the intensity of amplicon acquired from an experimental cPCR was 50% of the intensity of the amplicon obtained from control no-treatment cPCR. TCEP was added to the PCR reaction mixture to final concentrations ranging from 0 to 2 mM and PCR was performed with 25 cycles (18 min). Human gDNA was used as the template and primer sets were designed to detect 249 bp of the α-tubulin gene. Amplicons were analyzed using agarose gels and the relative intensity of each amplicon was measured and plotted (Fig. 1a). There was no significant effect of TCEP on cPCR performance, although amplicon intensity was slightly lower at a higher concentration of TCEP (1–2%). Similarly, TBA was added to the same cPCR reaction mixture described above to final concentrations ranging from 0 to 5%, and the cPCR performance was analyzed by agarose gel electrophoresis and the intensity of each amplicon were compared (Fig. 1b). A dose-dependent decrease in amplicon intensity was observed. At 2% TBA, a significant decrease in amplicon intensity was observed.

Fig. 1
figure 1

Effect of TCEP and TBA addition on cPCR performance. Dose-dependent effect of TCEP (a) or TBA (b) on cPCR performance was observed. TCEP or TBA was added to the reaction mixture to final concentrations ranging from 0 to 2 mM or 0–5%, respectively. Or combination of TCEP (0.1 mM or 1 mM) and TBA (0–5%) (c) was added and the effect of combined chemical treatment on the performance of cPCR was measured. After cPCR for 25 cycles, the results were analyzed by agarose gel electrophoresis, and the relative intensities of each amplicon were plotted as means ± SD of three independent experiments

Full size image

Next, a combination of TCEP and TBA was added to the PCR reaction mixture and the cPCR performance was tested. The concentration of TCEP was fixed to either 0.1 mM or 1 mM, and the concentration of TBA was varied from 0 to 5%. The amplicon intensity after 25 cycles of cPCR was analyzed (Fig. 1c). In both cases, a gradual decrease in DNA amplicon intensity was observed. A less pronounced decrease was observed in the group treated with a lower concentration of TCEP (0.1 mM) than in the group with a higher concentration of TCEP (1 mM). At 0.1 mM of TCEP and 1% TBA, approximately 82% of performance was observed. At 1 mM TCEP and 1% TBA, the cPCR performance was approximately 64%. We used 0.1 mM TCEP and 1% TBA in the cPCR reaction mixture for the subsequent experiments.

Effect of combined treatment of TCEP and TBA on the sensitivity of singleplex cPCR

To investigate the effect of TCEP and TBA treatment on the sensitivity and amplification efficiency of the cPCR, we tested for changes in the limit of detection (LOD) in the chemical-treated group with human gDNA as template and primer sets for 249 bp amplicons (Fig. 2a). Unexpectedly, both untreated control and chemical-treated groups showed a similar LOD of approximately 3 × 10° copies. The efficiency of PCR amplification for the chemical-treated group was lower in the human gDNA concentration ranges used in this study. We then tested the appearance of detectable DNA amplicon bands of 249 bp in the untreated control and chemical-treated groups after 10, 15, 20, and 25 cycles of cPCR (Fig. 2b). The DNA amplicons were detectable by agarose gel electrophoresis at 15 cycles in the untreated control group and at 20 cycles in the chemical-treated group.

Fig. 2
figure 2

Limit of detection (LOD) (a, c) and PCR cycle-dependent appearance of DNA amplicon (b, d) in the presence of 0.1 mM TCEP and 1% TBA. Human genomic DNA was used as a template and either 249 bp primer set (a, b) or 500 bp primer set (c, d) were used for amplicon detection. The cPCR results were analyzed by agarose gel electrophoresis and compared with those obtained for the untreated control group. The relative intensities of each amplicon were plotted as means ± SD of three independent experiments

Full size image

Next, we investigated the effect of TCEP and TBA on cPCR with the larger amplicon. Overall, the efficiency of cPCR for the 500 bp target in the presence of 0.1 mM TCEP and 1% TBA was ~ 78% (Fig. 2c, d). The LOD was approximately 3 × 10° copies and 3 × 101 copies for the control and chemical-treated groups, respectively (Fig. 2c). In both untreated control and chemical-treated groups, the amplified DNA bands began to appear after 20 cycles of cPCR (Fig. 2d). In some cases the amplicon of 500 bp appeared to be a little retarded on the agarose gel in multiplex cPCR.

Effects of combined treatment of TCEP and TBA on the sensitivity of multiplex cPCR

We then investigated the effect of combined chemical treatment on multiplex cPCR. Human gDNA was used as the template, and primer sets for 249 and 500 bp amplicons were mixed in the PCR reaction mixture. The LOD values for both 249 and 500 bp amplicons were slightly lower than the LOD values obtained from the singleplex cPCR (Fig. 3a). For the untreated control group, a LOD value of approximately 3 × 101 copies was obtained for both 249 and 500 bp amplicons. For the chemical-treated group, the LOD values for 249 and 500 bp amplicons were approximately 3 × 101 copies and 3 × 102 copies, respectively.

Fig. 3
figure 3

Effect of combined chemical treatment on the limit of detection (LOD) (a) and PCR cycle-dependent appearance of DNA amplicon (b) in multiplex cPCR. Human genomic DNA was used as template and 249 and 500 bp primer sets were used for amplicon detection. The cPCR was run in the absence or presence of 0.1 mM TCEP and 1% TBA. The results were analyzed by agarose gel electrophoresis and compared with those obtained for the untreated control group. The relative intensities of each amplicon were plotted as means ± SD of three independent experiments

Full size image

The appearance of detectable DNA bands on the agarose gel was compared in multiplex cPCR format (Fig. 3b). Both the 249 and 500 bp amplicons became detectable after 20 cycles of cPCR amplification.

Discussion

To perform on-site molecular diagnostics, certain conditions are required, including equipment that can be used despite limited electricity and protein-based reagents that do not require a cold chain for stability. Freeze-drying is the most popular method to preserve protein-based biomolecules for prolonged storage at ambient temperature. Freeze-drying technology has been developed greatly, and the formulation and additives have been studied extensively [21].

Here, we chose two chemicals and evaluated the effect on the performance of cPCR; one is a protein antioxidant, TCEP, which is used for most protein preparations [9, 22] and the second is TBA, which is used as a freeze-drying solvent [17, 18, 23]. While there are a number of papers that that assess the use of freeze dried reagents in on-site applications [24, 25], the effect of these two representative antioxidant and freeze-drying solvent on the performance of PCR never been tested before. We used a battery-operable rapid cPCR device, which can therefore be used on-site.

The working concentrations of TCEP and TBA are higher than the concentrations used in this study [9, 17, 18, 22]. Although these chemicals are diluted and evaporated during the subsequent and freeze-drying processes, residual amounts would remain in the reconstructed solution to be used on-site.

PCR is an amplification process and reaches saturation point after a certain number of cycles. Thus, we tried to evaluate the effect of chemicals on the performance of cPCR before it reaches the saturation point. In most of the experiments, we ran cPCR up to 25 cycles (18 min), so it does not reach the saturation point. In most molecular diagnostics study, amplicon sizes between 200 and 500 bp were used [4, 5]. We used amplicons of 249 bp and 500 bp to cover a broader range.

The effect of TCEP and TBA on the performance of cPCR was not significant under the experimental conditions used in this study. The performance was approximately 80% at 0.1 mM TCEP and 1% TBA. The LOD value of the chemical-treated group was almost the same as or slightly lower than that of the untreated control group. However, when we ran the PCR for more cycles (e.g., 30–40 cycles), the amplification reached near saturation and the difference between the groups was not detectable (data not shown).

Conclusion

In the current study, we investigated the effect of two chemicals, TCEP and TBA, on the performance of cPCR. We used the cPCR condition at the early stage of cPCR amplification, which is distant to the saturation point, so that the difference of amplification in untreated control and chemical-treated groups was easily observed. Treatment with 0.1 mM TCEP and 1% TBA resulted in approximately 82% performance for a relatively small amplicon (249 bp) and 78% for a relatively large amplicon (500 bp) at 25 cycles with 3 × 103 copies of template. Experiments with multiplex cPCR revealed similar results. Our data suggested that relatively low concentrations of TCEP and TBA do not interfere with cPCR performance significantly. We suggest using more than 1 × 101 copies of the template and running more than 25 cycles of PCR for on-site use to ensure the diagnostic results.