We use cookies to improve your experience. By continuing to browse this site, you accept our cookie policy.×
ReportsOpen Accesscc iconby iconnc iconnd icon

PCR-based gene synthesis with overlapping unisense-oligomers asymmetric extension supported by a simulator for oligonucleotide extension achieved 1 kbp dsDNA

    Yasunori Nishida

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    ,
    Kotetsu Kayama

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    ,
    Taichi Endoh

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    ,
    Kiwamu Hanazono

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    ,
    Gerry Amor Camer

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    College of Veterinary Medicine, University of Eastern Philippines, Catarman, Northern Samar, 6400, Philippines

    &
    Daiji Endoh

    *Author for correspondence:

    E-mail Address: dendoh@rakuno.ac.jp

    Department of Radiation Biology, School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, 069-8501, Japan

    Published Online:https://doi.org/10.2144/btn-2022-0127

    Abstract

    We formulated a method to synthesize 1 kbp DNA fragments using ‘oligomer unidirectional joining method’ via asymmetric extension supported by a simulator for oligonucleotide extension (AESOE). In this study, trials were conducted on 41 sets of different genomic pieces of ten flaviviral genomes, and 31 bacterial 16s rRNA fragments with sizes ranging from 500 bases to 1.0 kbp. Synthetic gene production was found to be successful in all those sets. The synthesis method has three steps: the first step is a seven-linked AESOE, the second step is the linking of the 400-base fragments from the first step, and the third step is the final amplification. Our present approach is highly reproducible and may no longer require optimization of oligomer design.

    METHOD SUMMARY

    Using asymmetric extension supported by a simulator for oligonucleotide extension (AESOE), an enhanced method was accomplished to synthesize ≥1 kbp dsDNA products. Done on 41 sets of different genomic pieces of ten flaviviral genomes, and 31 bacterial 16s rRNA fragments with sizes ranging from 500 bases to 1.0 kbp, our current method consisted of three steps: the first step is a seven-linked AESOE, the second step is the linking of the 400-base fragments from the first step, and the third step is the final amplification. Effective and efficient synthetic gene production, including troubleshooting of failure-prone processes, was reproducibly attained.

    Graphical abstract

    The production of synthetic genes based on chemically synthesized DNA oligomers has become a major technology in synthetic biology [1]. The need for synthetic genes is currently growing, including the need for a DNA fragment as a positive standard for PCR or loop-mediated isothermal amplification (LAMP) [2–4]. Moreover, despite the need for expanding assembly of synthetic genes, a design must be made in such a way that the movement of genetic materials must be restricted. In the development of virus-detection methods, many standard DNA fragments are required. At the same time, it must be done in a way that avoids the risk of inadvertently contracting an infection. This is likewise true for genome detection in wildlife animals. Accordingly, genomic DNAs of endangered animals are prohibited by the Washington Convention from crossing national borders. Moreover, this is about protecting the samples from international trading exploitation. These are some major concerns in processing genes [2].

    Numerous articles were published on synthetic genes based on chemically synthesized DNA oligomers [5–7]. As genetic information for genomes and other genetic material progresses, methods have been improved and widely used [8,9].

    Overlap extension PCR (OE-PCR) can be utilized, in which oligomers are overlapped and DNA fragments can be synthesized by DNA polymerization from overlapped oligomers [7–9]. Simultaneous synthesis of several hundred artificial genes has taken place [10]. This technology is expected to be used not only as an artificial gene but also as a computational and data storage device [11–13]. Thus, artificial gene synthesis technology is expected to be used for new applications using artificial life that collaboratively uses these molecules [14]. For example, in wild animal conservations where preliminary experiments are required to determine whether correct PCR occurs in hairs and feces, limitations in obtaining DNAs from them can be resolved using synthetic gene technology [15]. Hence, a larger-than-usual field of synthetic biology is evolving [16].

    DNA synthesis technology is now widely supported by companies wherein they are asked to synthesize DNA at considerably high-rate pricing [17]. Moreover, safeguards must be in place such as ensuring the continuing publication of technological enhancements.

    We have been working on an effective and efficient synthetic gene technology with the intent to lower commercial gene synthesis costs [17,18]. We published an artificial gene synthesis method by using oligomer elongation reactions, which enables sequential synthesis by placing oligomers in one direction in succession [18]. Asymmetric extension supported by a simulator for oligonucleotide extension (AESOE) views DNA synthesis as the 3′ extension of oligomers forming partial duplexes. By setting the direction of the oligomers as one in the forward direction and the other three in the reverse direction, such extensions could occur in sequence: when the oligomers are arranged as F1-R1-R2-R3, the binding strand is first synthesized from F1-R1 with both 3′ ends forming a double strand. Subsequently, the extended F1 of the F1-R1-partial dimer includes a joining region with R2, from which the next extension occurs. The order of the synthetic reactions in the tube can also be controlled, as these reactions proceed sequentially with each temperature cycle (AESOE reaction). As a result, there is no need to consider the independence of the junctions for the 3′ end of all oligomers. AESOE requires a concentration gradient between oligomers, and there was a problem in that synthesis could not occur if the minimum concentration was below the desired limit [18]. Therefore, the synthesis size was limited to 230 bp or less with four oligomers linked together. When synthetic genes are used as templates for examining the conditions of PCR and LAMP methods, longer sizes are required, but linking five or more oligomers at once was unachievable. To synthesize more than 1 kb, improvements were needed in the reaction conditions and oligomer linking method.

    This study aimed to synthesize beyond the synthetic length limit of AESOE. First, it increases the number of oligomers that can be ligated in the AESOE reaction, since the synthetic length limit of AESOE arises from the number of ligated oligomers. Second, by ligating the dsDNA produced by AESOE reaction, it is possible to synthesize a dsDNA of 1 kbp or more. The first improvement is made possible by performing PCR reaction at the end of AESOE reaction, and the second improvement is made possible by applying OE-PCR to join dsDNA. In addition, the redundancy of AESOE reactions that is allowed for full-length synthesis is done by replacing one of the oligomers if one of the AESOE reactions is not successful; otherwise, the same oligomers are used. The method is designed to assume usage of low-cost oligomer (https://eurofinsgenomics.jp) lengths, thus enabling low-cost synthesis of 1 kbp or above.

    Materials & methods

    Target sequence & oligomer design method

    Those encoding structural proteins (mainly envelope protein E) from ten virus species of the family Flaviviridae were obtained for 1000 bp each and used as design targets (Supplementary Table 1). To confirm the success rate of oligomer ligation synthesis, we also synthesized the 3′-terminal region of 16s rRNA from 31 bacterial species (Supplementary Table 2).

    Because synthesizing many oligomers at once results in uncontrolled byproducts due to nonspecific reactions, we used AESOE strategy to construct DNA fragments. Since the number of oligomers linked in the first step of gene synthesis determines the overall synthetic efficiency, we investigated the optimal number of oligomers in the AESOE design. Since the original AESOE [18] is limited to four oligonucleotides (230 bases if the single oligomer length is 70 bases), it was decided to amplify with the terminal oligomer after its concatenation. Numerous preliminary experiments have shown that the maximum optimal number of oligomer linkages is seven when 70-base oligomers are linked (Figure 1). To obtain a final product of 1 kbp, we have basically concatenated three optimized and extended AESOE products, which are composed of three compartments (part A, part B and part C) (Figure 1A). Boundaries of each part consist of a 70-mer overlapping region which is just the size of a unit oligomer length (Figure 1B). Then, the combined first products were subjected to the double-stranded OE-PCR method in the second step to construct the full-length target product (Figure 1C left), but in many cases, the product included many types of intermediate products (Figure 1C right). The linked AESOE product was amplified with primers at both ends of the synthetic product after its final linkage (Figure 1D).

    Figure 1. Schematic presentation of AESOE2 (dsDNA synthesis of flaviviruses).

    (A) For reliable synthesis, about 1000 bases of target sequences are directed for synthesis. Up to seven oligomers of 70 bases F1-R2-R3-R4-R5-R6-R7, F7-R8-R9-R10-R11-R12-R13, F13-R14-R15-R16-R17-R18-R19 were used, and AESOE reaction was performed. The oligomers are arranged to be continuously linked. The design and arrangement of oligomers can be made automatically by using the program as shown in Supplement B manual. (B) For the ends of the part where AESOE is performed (R7 and F7, R13 and F13), the same sequence is reoriented to allow for linkage. It shows the reaction used to link the parts synthesized by AESOE by multiple overlap extension PCR. (C) After the first step of multiple overlap extension PCR has proceeded to the point where no final product is generated. (D) A final PCR with primers at the ends is performed to obtain a final product of uniform length.

    AESOE: Asymmetric extension supported by a simulator for oligonucleotide extension; AESOE2: the current version or the second AESOE as used in this study.

    Points that need improvement in the synthesis process

    In the first step, as the reaction proceeds, the accumulation of single strands constructed with the reverse oligomers occurs (Figure 2). To eliminate and reuse the accumulation, a short leading forward oligomer was used to allow the reaction to proceed efficiently. When a four-strand length (Figure 2, right end in green) is used as the final product, several types of reverse strands accumulate (Figure 2). A method that prevents the accumulation of reverse strands must be developed. In the first step, a joining product of approximately 400 bp is generated (Figure 1A), and in the third step, PCR amplification using the first and 19th double-ended oligomers yields the final product (Figure 1C & D). The reaction conditions for these reactions were set according to OE-PCR method of the type that links duplexes [2].

    Figure 2. Leftward single-strand accumulation.

    In asymmetric extension supported by a simulator for oligonucleotide extension reaction with seven oligomers, only the 70-mer oligomer accumulates a leftward synthesis product due to linkage reaction. The mechanism of accumulation is illustrated in the figure.

    Oligomer design for the current AESOE method

    The joining of the oligomers was set based on the nearest neighbor method [18] (Supplementary Tables 3 & 4). Each overlap region is divided into 70 bases so that the Gibbs energy at the time of duplex formation at each joining is -13.6 kcal/mol. The target length of approximately 1000 bp is divided into 19 segments, and each of the 19 segments is fixed at 70 bases so that the complementary portions at both ends are approximately 15 bases. Due to the set-up of the region, the 19th sequence at the final 3′ end is shorter than the others (Supplementary Table 3). For the synthesis of the final product, the 19 oligomers are divided into three parts of seven oligomers, which are then linked. If each part is designated as parts A, B and C, the orientation of the first (no. 1, no. 7 and no. 13) of each part was set to be forward sense, while the remaining six parts were set to be reverse sense (Figure 1B). As a result, the objective product of each part was about 400 bp each. Oligomers for 16S rRNAs were designed according to the same strategy (Supplementary Table 4).

    In the first step of the synthesis, leftward single-strand DNA may be accumulated (Figure 2). For the prevention of this accumulation, we used short oligomers in the first step of synthesis (Supplementary Table 5).

    The second step is complementary to the 70 bases at the end of the first step, so the joining region between the parts of the first step is included at its end when the first step is finished (Figure 1C). Therefore, no special design was used. The possibility that the base sequence of the joining region may bind to other regions was not considered anymore, and joining was set up mechanically. This is after assuming the oligomer direction of the first step. This was chosen to minimize doing another redesign.

    The synthesis method of the current AESOE

    PCR reaction

    Oligomers were synthesized by Eurofin Genomics (Tokyo, Japan). The concentration after chemical synthesis was specified as 50 pmol/μl, and if dilution was necessary, a 10 mM Tris-1 mM ethylenediaminetetraacetic acid solution was used. PCR was performed as specified using GoTaq G2 Hot Start Green Master Mix (Promega, MA, USA).

    First-stage reaction solution & reaction procedure

    In the first step, the AESOE joining of seven oligomers of equal proportions was performed part by part (Figure 1B). Only the first oligomer was the forward oligomer and the rest were reverse oligomers. The first part comprised oligomer nos. 1F:2R:3R:4R:5R:6R:7R in the ratio 1:1:1:1:1:1:1, the second part by 7F:8R:9R:10R:11R:12R:13R in the ratio 1:1:1:1:1:1:1, and the final part by 13F:14R:15R:16R:17R:18R:19R in the ratio 1:1:1:1:1:1:1. The concentration of each oligomer was 10 pmol/μl, and 1 μl was mixed and diluted for each part.

    The composition of the three parts (250 μl oligomer mix) was as follows, and oligomer dilution mixtures were prepared for each part: Tris-EDTA (TE) 243 μl, no. 1–7 50 pmol/μl, mixed 1 μl each, a total of 250 μl (A) solution; TE 243 μl, no. 7–13 50 pmol/μl, mixed 1 μl each, a total of 250 μl (B) solution. Mix TE 243 μl and 50 pmol/μl of no. 13–19 in 1 μl each to make a total of 250 μl of (B) solution.

    For the reaction solution, 2 μl from (A) above was reacted with 1 μl of the short first no. 1 forward oligomer 10 pmol/μl, 2 μl of DW and 5 μl of GoTaq Green Mix, for a total of 10 μl. Similarly, (B) and (C) reacted with the same composition.

    Reaction conditions in the thermal cycler were 30 cycles of reaction. The annealing temperature was set at 52°C. The entire reaction started at 95°C for 2 min for a hot start. Then, the conditions were set at (95°C for 30 s → annealing temperature 52°C for 30 s → 72°C for 2 min) × 30-times, final extension 72°C for 3 min, and the temperature was kept at 8°C.

    Second-stage reaction solution & reaction procedure

    In the second step, the seven lengths of products obtained in the first step were mixed in equal proportions. After synthesis was completed, a concatenation reaction was performed for the parts corresponding to a total of 19 lengths to reach 1000 bp. After the first step reaction, 1 μl from each part (A, B and C) sample was reacted with 3 μl total, 1 μl of DW and 5 μl of GoTaq Green Mix, for a total of 10 μl. The reaction conditions in the thermal cycler were the same as in the first step.

    Third-stage reaction solution & reaction procedure

    The third step is the PCR amplification reaction of the final product formed in the second step (Figure 1D). In this last amplification process, the PCR method was carried out by reusing the oligomers at both ends as primers for the product theoretically obtained in small quantities in the second step. The resulting dsDNA of the final product was amplified and obtained.

    Specifically, 1 μl was taken from the sample after the second-step reaction, where 1 μl of each of the two end oligomers (no. 1 70-mer forward oligomer and no. 19 reverse oligomer) was adjusted to 10 pmol/μl, 1 μl of DW and 5 μl of GoTaq Green Mix, for a total of 10 μl of DW and 5 μl of GoTaq Green Mix, for a total of 10 μl. The reaction conditions in the thermal cycler were set as in the first and second steps.

    Confirmation of synthetic products

    Synthetic products were separated by agarose gel electrophoresis at 1.7% in 0.5× Tris/Borate/EDTA buffer and confirmed by luminescence under UV light after staining with 0.5 μg/ml ethidium bromide solution. Nucleotide sequences of the synthetic products were determined using oligomers at the 5′ and 3′ ends (Eurofins Genomics, Tokyo, Japan). Sanger sequencing was performed from the 5′ and 3′ ends of the synthetic products. This was referred to regions where the histogram showed a single base with exception on the oligomer used for AESOE reaction that contained a degenerate base.

    Alternative strategy when the first step fails

    Adding reverse complementary sequences

    Reconstitution in current AESOE method when the first-step synthesis cannot be confirmed is possible by adding a few forward oligomers (Figure 3). Thus, the seven-unit division was changed to a six-unit division. By adding oligomers, a modification was made to allow full-length synthesis. On this reconstruction, four oligomers on the 3′ end of the joining oligomers were subject to the synthesis. Resultantly, the oligomers required were three reverse-complement sequence oligomers numbered 6, 11 and 16, as shown in the middle row of Figure 3. Furthermore, we tried the synthesis with five-partitioning sets. Additional oligomers required were four reverse-complement sequence oligomers numbered 5, 9, 13 and 17 (Figure 3). Additional oligomers on the alternative first step of the synthesis are listed in Supplementary Table 6. Short oligomers were also prepared for the alternative synthesis (Supplementary Table 7).

    Figure 3. Configuration of most split oligomers and additionally synthesized oligomers at the time of failure.

    The figure shows how to change the number of oligomers in the AESOE set in case of reaction failure. By resetting the position and AESOE set of the forward-directed oligomer (shown in green), which is the linkage portion of AESOE product when conducting overlap extensionPCR, and its complementary reverse-directed sequence oligomer (right end of each AESOE set), the final product can be synthesized without replacing the entire oligomer.

    AESOE: Asymmetric extension supported by a simulator for oligonucleotide extension.

    Results & discussion

    First-step reaction optimization

    Examination of the maximum length of the first-step consolidation reaction

    If the 1 kbp level is targeted, the overlap bases between oligomers and the number of oligomers on the first step are important factors for whole synthesis reactions. Since joining in the second step is performed between duplexes, a joining region longer than 50 bases, that is, 70 bases for both ends, is necessary (overlap region of Figure 1C). For the artificial gene synthesis oligomer, the maximum size was assumed for low-cost synthesis (70 bases in Japan and Europe). Therefore, we performed a preliminary experiment to determine the maximum size of the first step with an oligomer length of 70 bases. Of the size of the final product of the first step, the size exceeding 70 bases becomes the substantial extension unit of the second step. In each first-step synthesis set for flaviviral sequences, the number of oligomers was set from six to nine. For evaluation of the oligomer numbers per set, the AESOE product was reamplified by PCR using the end oligomers. The preliminary experiment showed that more than six joining was possible by employing an equal concentration of oligomers, while original AESOE reactions require graded concentrations. Full-length products were synthesized on an oligomer set including six or seven oligomers, while an oligomer set including eight oligomers often failed (Figure 4). Success rates were 100, 90, 40 and 10% on six, seven, eight and nine oligomer sets (Table 1). From these results, we used seven oligomers as the basic set for first step AESOE reactions. For all second-step products, the entire sequence was confirmed and 100% matched the sequence assumed in the design.

    Figure 4. Electrophoresis images of each set of oligomers joining at the first stage.

    Preliminary asymmetric extension supported by a simulator for oligonucleotide extension reactions for first step of flaviviral sequences (Supplementary Table 3). Six to nine oligomers of each flaviviral oligomer were selected and synthesized as described in ‘Materials & methods’. The resultant joined molecules were separated at 1.2% agarose gel electrophoresis. Set numbers are indicated on the lines and the number of oligomers is indicated below the lines.

    M: Molecular weight marker.

    Table 1. Success rates for the first stage of oligomer joining.
    Number of first-step linked oligomers6789
    Number of sets for which the total length is synthesized10941
    Number of sets where synthesis stopped at partial length0169

    Reimprovement of the first-step synthesis

    Once the basic configuration of seven oligomers was determined, the joining reaction was reviewed again. The simulation of reactions suggested that the reaction could be improved by adding a short-forward directed oligomer at the beginning (Figure 5). It was confirmed that the concentration of intermediate double-stranded products of different sizes increased compared with the results on no short oligomers (Figure 6A), facilitating the increased efficiency of synthetic reactions (Figure 6B & Supplementary Figure 1).

    Figure 5. Role of short forward oligomer.

    Illustrates the effectiveness of introducing short right-handed oligomers to eliminate the accumulation of left-handed oligomers in AESOE. Left-handed oligomers of various lengths are synthesized as byproducts in AESOE, and the addition of short right-handed primers in AESOE can reduce synthesis problems.

    AESOE: Asymmetric extension supported by a simulator for oligonucleotide extension.

    Figure 6. Effect of short forward oligomer on first-step synthesis.

    Preliminary asymmetric extension supported by a simulator for oligonucleotide extension reactions for first step for flaviviral sequences (Supplementary Table 1) with joining seven 70-mer oligomers described in ‘Materials & methods’ (Supplementary Table 3). The joining reactions were performed in (A) with the short forward oligomer, or in (B) without the short forward oligomer (Supplementary Table 5). The resultant joined molecules were separated on 1.2% agarose gel electrophoresis. Each set number is indicated by the origin of the agarose gel.

    M: Molecular weight marker.

    Second-step reaction optimization

    The incomplete outcome of second-step synthesis & full-length synthesis on third-step reactions

    In the second step, after joining the first-step seven oligomers, we aimed for reaction conditions in which the final synthesized product is mixed with several intermediate products (Figure 7A). PCR amplification was performed using the no. 1 and no. 19 oligomers at both ends used in the first step. Although the final full-length synthetic product was present in a mixed form with many intermediate products in the second step, PCR using the two oligomers at the full-length ends (no. 1 forward oligomer and no. 19 reverse oligomer) confirmed the amplification of the full-length synthetic artificial gene product (Figure 7B). The efficient synthesis of 16s rRNAs (Figure 8) was also shown after the same steps. The sequence independence of the linking regions of the oligomers and the success of a total of 41 sets (ten Flavivirus and 31 16s rRNAs) in preliminary experiments suggest that the linking of seven oligomers is effective for the synthesis of the final product. In particular, no problem was demonstrated in the second step for all 41 sets. However, during the condition-setting process, the linkage with seven oligomers failed 10% of the time in the first step. Therefore, for the sake of stable synthesis for many researchers, we considered an efficient response to the failure of the synthesis in the first step.

    Figure 7. Reaction results of the second-stage completion type.

    Second stage with (A) second- and (B) third-step asymmetric extension supported by a simulator for oligonucleotide extension reactions for flaviviral sequences (Supplementary Table 1). First-step reactions, which are described in ‘Materials & methods’, were performed with seven oligomers of each flaviviral oligomer (Supplementary Table 3). The resultant joined molecules were separated on 1.2% agarose gel electrophoresis. Each set number is indicated by the origin of the agarose gel.

    M: Molecular weight marker.

    Figure 8. Synthesized DNA fragments for 3′ end of 16s rRNAs.

    Final products of AESOE reactions for bacterial 16s rRNA (Supplementary Table 2). Reactions were performed as described in ‘Materials & methods’ with oligomers of each bacterial 16s rRNA (Supplementary Table 4). The resultant joined molecules were separated on 1.2% agarose gel electrophoresis. Each set number is indicated by the origin of the agarose gel.

    M: Molecular weight marker.

    Alternative configurations in case synthesis are not completed in the first step

    In this method, the nucleotide sequence of the linked region of the oligomer does not need to be considered during design. In addition, the success rate of synthesis in the first step increases as the number of oligomers is reduced. As a final improvement of the method, we considered a method in which even if synthesis fails, the entire process succeeds with the replacement of the minimum number of oligomers. Since the redesign of the first step succeeded with a reduction in the number of oligomers, we reduced the number of oligomers in a particular first-step set and reversed the direction of the starting oligomer in the subsequent first-step sets. The composition of the part was changed from the basic seven oligomers to six or five (Figure 3) oligomers per part, showing that both first oligomer sets worked (Figure 9 & Supplementary Figures 2 & 3). These results suggest that full-length synthesis is possible by adjusting the number of oligomers per part even when the first step of synthesis does not work.

    Figure 9. Artificial gene synthesis after reconstruction of first step oligomers.

    Final products of AESOE with different numbers of oligomer sets on (A) first step, and (B) second step reactions as described in ‘Materials & methods’. First-step reactions were performed with six or five oligomer sets. The second and third steps of the reactions were performed as described in ‘Materials & methods’. The resultant joined molecules were separated on 1.2% agarose gel electrophoresis. Each viral abbreviation is indicated on the origin of the agarose gel.

    M: Molecular weight marker.

    Comparison with previous studies of current AESOE

    Comparison with one-way bilateral synthesis method

    There are reports of OE-PCR-like methods that can synthesize more than 1 kbp, but they require advanced technology for designing the sequence of the linkage portion, so stability needs to be practically improved. For example, a 1.1 kbp gene was formed by randomly mixing 56 oligomers [19,20]. This method is called OE-PCR because DNA synthesis is carried out continuously by linking oligomers. Although it has been improved using computer programs [21], it was known that when the target synthetic product is larger than 0.5 kbp, the synthesis becomes quite difficult due to the increase of unscheduled and accidental joining reactions [22–24]. We experienced similar low synthetic efficiency when synthesizing fragments of flaviviruses [25].

    An effective attempt to stabilize synthesis by OE-PCR beyond 1 kbp has been proposed by using a constant orientation of the linking oligomers. As reported, the thermodynamically balanced inside-out (TBIO) procedure has successfully synthesized 2.4 kbp by organizing oligomer joining and linking six oligomers in the reverse-reverse-reverse direction to oligomers in the forward direction [26]. However, TBIO does not work with some genes, indicating the need to optimize oligomer settings for each gene [27]. Because of the importance of sequence selection of the linkage in TBIO, it is expected that all oligomers need a redesign if the synthesis is not successful.

    In addition, the biochemical synthesis of dsDNA is associated with the problems of incorrect formation of double-strands on the joining region; and the elongation step of oligomers results in a certain frequency of sequence mutations. The OE-PCR method requires all primers to be mixed in one tube and therefore shorter overlaps do not allow unambiguous annealing of complementary primers; this most definitely results in nonspecific sequences that inhibit full-length product formation. From this problem associated with OE-PCR, it has been noted that the manual design of oligonucleotides does not always guarantee the successful synthesis of the desired gene. For the method to work, it has been suggested that the melting temperatures of the overlaps must be similar for all oligonucleotides [20,26]. Consequently, specialized oligonucleotide design programs must be used, which is not only time-consuming but could also lead to research discontinuation due to a lack of prospects for success [28].

    Oligomer joining optimization is not required in this study because the joining reactions are sequentially assembled by AESOE arrangement, which suppresses formation of ambiguous byproducts [18]. Unlike our previous AESOE, PCR is performed in the final step of AESOE synthesis so that the product is obtained even at a 1:1 concentration ratio of oligomers. Since simulations of AESOE reaction predicted an overproduction of intermediate products in one direction, the reaction efficiency was increased by introducing a forward oligomer with a short 5′ end. Limitations were placed on the number of oligomers linked, but it was possible to prepare an alternative method in case the first-step synthesis did not reach the target number of oligomers, thus setting a standard number of linked oligomers (Figure 6). The practicality of this method is realized by the alternative number of combinations in case of failure of the first step (Figure 8 & Supplementary Figures 2 & 3). It was also important to note that the ability to obtain a final length product during these alternative combinations was made possible by the fact that the synthesis goal in the second step of the reaction was not to obtain the final product by electrophoresis but to succeed in the third step of the synthesis. In the second step, the reaction was kept to the extent that a small amount of the final product was attained, and by moving to the third step, a clear final product was obtained (Figure 7). A remarkable 94.4% of the products were confirmed by Sanger sequencing, and all the confirmed bases matched with the templates (Supplementary Figures 4 & 5 & Supplementary Tables 8 & 9). The error rate of the Taq polymerase used in this study is 1 × 10-5 errors/base according to the manufacturer's information. For the individual bases, it is about 1 × 10-3 errors/base after 90 cycles. Synthetic product reliability was found to be consistent [29]. This result also substantiates the fact that sets that fail to synthesize or have inadequate yields can also be resolved by dilution of the product after the first and second stages. Taken together, AESOE2 is unique in that it achieves reliable synthesis by stabilizing the reaction up to the third step, leaving the number of oligomers in the first step at seven oligomers.

    To overcome reaction randomness during joining reactions, dual-asymmetric PCR avoids these problems and enables stable artificial gene synthesis to take place [30,31]. Furthermore, some authors have further optimized these methods to enable error-free DNA synthesis [32]. However, these methods still require gel purification to remove some byproducts [26]. In contrast to what has been done by those researchers, it can be gleaned that the method described in this study does not require any gel purification in the process, nor does it require optimization of the nucleotide sequence to optimize the joining sequence of oligomers for each gene.

    Differences from using OE-PCR for joining in duplexes

    In the first step, the reactions are sequential and unidirectional, a kind of cascade reaction (Figure 2). This is illustrated by the increased efficiency of synthesis due to the introduction of short forward primers (Figure 6). The unique feature of this method is the introduction of such a design that assumes a reaction. As a result, stable synthesis in the first step has been achieved, making it possible to set a joining length of 70 bases, which is longer than the 35 bases introduced in multiplex overlap extension PCR, for the joining in the second and subsequent steps [2]. Thus, the joining length in the second step is determined at the time of designing the first step, which is a unique feature of this method.

    Oligomer length in the first stage

    In addition to novelty, the economics of synthesis were considered in the design. The synthesis cost of OE-PCR, which was previously done in our laboratory, had a much higher appraisal than the price of ordering it from a company [18]. The reaction process that determined its economic efficiency was the first step, and the most influential factor was the oligomer length to the extent that it was offered at a low price. As in the case of Japan, a private genomic processing company can provide a reasonably priced 70-mer. Such a service can be said to be the most efficient in terms of just the right price and length. Although this length and price may change in the region and future years, this method is expected to be able to accommodate changes in oligomer size that can be sold at the low-cost maximum length because there is an alternative method in case the first step fails. Conversely, technological innovation and cost reduction in oligomer synthesis in companies may increase the maximum length setting of low-cost oligomers [17]. The current method could increase synthesis efficiency without apprehensive major changes.

    Causes & remedies for failure to achieve full length with this method

    No products of the first stage are produced

    This case was our most common observation and once it occurred, there was no way to correct it. Therefore, in such cases, we resynthesized the full length by changing the number of oligomers in the first step and, in turn, the forward oligomer of the next part. The first-step configuration was changed in various combinations with the flavivirus sequence, and the synthesis problem was solved by changing the oligomer set, which had a base of seven, to six or five oligomers only in the part that was difficult to synthesize (Supplementary Figures 2 & 3). The overlapping parts are effects of AESOE's characteristic of sequential reactions starting with the forward oligomer. This change was made without any consideration of interactions between oligomers, but in OE-PCR, such a change required an entire redesign, and if one design was unsuccessful, numerous oligomer changes along with trials and errors were required [25]. In this method, the number of combinations in the first step is changed rather than doing complicated calculations, thereby reducing tedious redesigning as well as time-consuming trials and errors that frequently occur during artificial gene synthesis.

    The full length of the third step is not generated

    The end point of the second stage synthesis cycle was critical for obtaining the full product length in the third stage (Figure 7). Although it is difficult to provide clear guidelines for this process because the degree of synthesis is determined empirically, the optimum conditions can be obtained through a limited number of trial-and-error experiments because, at this stage, there is already a large amount of synthetic product from the first stage. Improvement of this stage should be considered in the future.

    The openness of the technology of this method

    This method was designed with utmost emphasis on practicality. This is because we believe that the publication of a practical method is necessary to improve methods of artificial gene synthesis of genetic materials that are company-centered and dictated by cost. Therefore, in the development of this method, the method must be published in a research paper and be reproducible in the laboratory of the researcher who implements it. The following two points may satisfy these conditions: first, the final form of this method is simple and reproducible. It does not utilize any special software. Second, since the procedure has been stabilized through numerous preliminary experiments, it can be used by many researchers by including an easy-to-understand manual in Supplement B. Furthermore, the release of such technology may improve gene synthesis business contracts in companies.

    Scope of application of this method

    The total number of reaction cycles in this method is high because economy and reproducibility were of utmost importance. Therefore, this method may not be suitable for expression analysis of genes in which base substitutions are not allowed. The most promising use of this method is as a standard DNA fragment to be used when examining conditions of PCR and LAMP. This approach can be reliably practical even when the bases of a few molecules are different from the original sequence. Categorically, due to the increasing number of test subjects with required specificity, our design is assumed to cover a greater scope of application as a standard template when examining conditions of PCR and other molecular-based methods.

    Conclusion

    The improvements in AESOE design through ligation of dsDNA produced by AESOE reactions have successfully synthesized DNAs of 1 kbp size or more. With this, it was concluded that DNAs can be synthesized even beyond the synthetic length limit of AESOE. Our findings can now respond to the efficiency demands of producing synthetic genes.

    Future perspective

    Synthetic gene formulations require constant validation considering the dynamic nature of synthetic gene technology. Our current approach to synthesizing larger-than-usual-sized (>1 kB) DNA products from gene fragments of pathogenic bacteria and viruses can serve to respond to growing demands of providing inductive stimuli for the development of therapeutics, and diagnostic tools that are beneficial for human and veterinary medicine.

    Executive summary

    Background

    • To develop a novel method to synthesize 1 kbp or greater-sized DNA fragments using ‘oligomer unidirectional joining method’ via asymmetric extension supported by a simulator for oligonucleotide extension (AESOE).

    Experimental

    • Forty-one sets of different genomic pieces of ten flaviviral genomes, and 31 bacterial 16s rRNA fragments with sizes ranging from 500 bases to 1.0 kbp were utilized for artificial gene synthesis.

    • Short-length oligomers were joined and accomplished in three steps: the first step is a seven-linked AESOE, the second step is the linking of the 400-base fragments from the first step, and the third step is the final amplification.

    • Sanger sequencing was performed from the 5′ and 3′ ends of the synthetic products.

    Results & discussion

    • Careful designing of equal concentration of oligomers by overlapping their bases in stepwise phases (steps 1, 2 and 3) has resulted in synthesis of a 1 kbp dsDNA size or more.

    • Limited product of 230 bp or less with four-linked oligomers on initial AESOE has been improved using the seven-linked oligomers designed in a noncomplex method.

    • Consistent synthetic product reliability was that all the confirmed bases matched with the templates on 94.4% of the products that were confirmed with Sanger sequencing.

    • Synthesis failures and coping with failure through adjustments of oligomer settings along with trials and errors have partly resolved some failure issues.

    Conclusion

    • DNAs can be synthesized even beyond the synthetic length limit of AESOE which can now respond to the efficiency demands of producing synthetic genes.

    Supplementary data

    To view the supplementary data that accompany this paper please visit the journal website at: www.future-science.com/doi/suppl/10.2144/btn-2022-0127

    Author contributions

    D Endoh and K Kayama conceptualized the study; K Kayama, D Endoh and Y Nishida designed the experiments; K Kayama, D Endoh, T Endoh and GA Camer wrote and critically enhanced the manuscript; K Kayama, Y Nishida, T Endoh and K Hanazono performed the experiments and prepared tables; K Kayama, Y Nishida, T Endoh, K Hanzono and GA Camer prepared the figures; and D Endoh designed and analyzed the data.

    Acknowledgments

    The authors thank Y Sanekata for his help in 16s rRNA genome artificial synthesis. Thanks is also due to RGU Laboratory of Radiation Biology students including Y Sasamura for assisting us in PCR-related experiments.

    Financial & competing interests disclosure

    This work was supported by research funding of Rakuno Gakuen University with Lagrange Co. Ltd (Tokyo, Japan). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

    No writing assistance was utilized in the production of this manuscript.

    Open access

    This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/

    Papers of special note have been highlighted as: •• of considerable interest

    References

    • 1. Benner SA, Sismour AM. Synthetic biology. Nat. Rev. Genet. 6(7), 533–543 (2005). •• Presented chemically synthesized DNA oligomers as a major technology of synthetic biology.
    • 2. Kadkhodaei S, Memari HR, Abbasiliasi S et al. Multiple overlap extension PCR (MOE-PCR): an effective technical shortcut to high throughput synthetic biology. RSC Adv. 6(71), 66682–66694 (2016). •• Supported the notion of developing tools for growing synthetic gene demands.
    • 3. Torii S, Ono C, Suzuki R et al. Establishment of a reverse genetics system for SARS-CoV-2 using circular polymerase extension reaction. Cell Rep. 35(3), 109014 (2021). •• Expressed the need for a DNA fragment as a positive standard for PCR.
    • 4. Villalobos A, Ness JE, Gustafsson C, Minshull J, Govindarajan S. Gene Designer: a synthetic biology tool for constructing artificial DNA segments. BMC Bioinform. 7(1), 1–8 (2006). •• Dealt with the development of synthetic genes useful for PCR.
    • 5. Jayaraman K, Fingar SA, Shah J, Fyles J. Polymerase chain reaction-mediated gene synthesis: synthesis of a gene coding for isozyme c of horseradish peroxidase. Proc. Natl Acad. Sci. USA 88(10), 4084–4088 (1991). •• Supported the view about studies being done on synthesis of genes based on chemically synthesized DNA oligomers.
    • 6. Dillon PJ, Rosen CA. A rapid method for the construction of synthetic genes using the polymerase chain reaction. BioTechniques 9(3), 298–300 (1990). •• Demonstrated a method of gene synthesis using PCR.
    • 7. Horton RM, Hunt HD, Ho SN, Pullen JK, Pease LR. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77(1), 61–68 (1989). •• Reported another mode of synthetic gene engineering.
    • 8. Hilgarth RS, Lanigan TM. Optimization of overlap extension PCR for efficient transgene construction. MethodsX 7, 100759 (2020). •• Reported improvement in the use of overlap extension PCR (OE-PCR).
    • 9. Zhang M, Liu Y, Chen L et al. One simple DNA extraction device and its combination with modified visual loop-mediated isothermal amplification for rapid on-field detection of genetically modified organisms. Anal. Chem. 85(1), 75–82 (2013). •• Presented methods for DNA manipulation and its practical use.
    • 10. Kosuri S, Church GM. Large-scale de novo DNA synthesis: technologies and applications. Nat. Methods 11(5), 499–507 (2014). •• Dealt with substantial simultaneous gene synthesis.
    • 11. Hao Y, Li Q, Fan C, Wang F. Data storage based on DNA. Small Structures 2(2), 2000046 (2021). •• Dealt with technology that could be used for computational and data storage.
    • 12. Eimerman P, Jeddeloh J, Champion E et al. Development of a simple and versatile enzymatic DNA synthesis system that enables accurate, fast, and long oligos on demand. J. Biomol. Tech. 31(Suppl.), S10 (2020). •• Dealt with developing accuracy and speed in synthetic gene technology.
    • 13. Meiser LC, Antkowiak PL, Koch J et al. Reading and writing digital data in DNA. Nat. Protoc. 15(1), 86–101 (2020). •• Presented the mode of DNA data storage and safekeeping.
    • 14. Czar MJ, Anderson JC, Bader JS, Peccoud J. Gene synthesis demystified. Trends Biotechnol. 27(2), 63–72 (2009). •• Described the value for new applications of synthetic technology.
    • 15. Kersey DC, Dehnhard M. The use of noninvasive and minimally invasive methods in endocrinology for threatened mammalian species conservation. Gen. Comp. Endocrinol. 203, 296–306 (2014). •• Supported the need for synthetic genes that could be useful for wild animal conservation research.
    • 16. Khalil AS, Collins JJ. Synthetic biology: applications come of age. Nat. Rev. Genet. 11(5), 367–379 (2010). •• Supported the nondisclosure of gene synthesis technology on a gene-by-gene basis.
    • 17. Kosuri S, Church GM. Large-scale de novo DNA synthesis: technologies and applications. Nat. Methods 11(5), 499–507 (2014). •• Confirmed the use of method with efficient and economically produced synthetic DNAs.
    • 18. Kayama K, Hashizume H, Camer GA, Endoh D. An improved gene synthesis method with asymmetric directions of oligonucleotides designed using a simulation program. BioTechniques 69(3), 211–219 (2020). •• Described initial work on artificial gene synthesis via asymmetric extension supported by a simulator for oligonucleotide extension (AESOE) along its framework on current troubleshooting of synthesis failures.
    • 19. SantaLucia J Jr. A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proc. Natl Acad. Sci. USA 95(4), 1460–1465 (1998). •• Presented similar reports on OE-PCR-like methods.
    • 20. Stemmer WP, Crameri A, Ha KD, Brennan TM, Heyneker HL. Single-step assembly of a gene and entire plasmid from large numbers of oligodeoxyribonucleotides. Gene 164(1), 49–53 (1995). •• Described other reports of OE-PCR-like methods that can synthesize larger DNA products.
    • 21. Keiper F, Atanassova A. Regulation of synthetic biology: developments under the convention on biological diversity and its protocols. Front. Bioeng. Biotechnol. 8, 310 (2020). •• Backed up the development of methods that regulate gene synthesis.
    • 22. Hoover DM, Lubkowski J. DNAWorks: an automated method for designing oligonucleotides for PCR-based gene synthesis. Nucleic Acids Res. 30(10), e43–e43 (2002). •• Supported the idea that complex synthesis methods may be resolved through automated designs.
    • 23. Baedeker M, Schulz GE. Overexpression of a designed 2.2 kb gene of eukaryotic phenylalanine ammonia-lyase in Escherichia coli. FEBS Lett. 457(1), 57–60 (1999). •• Showed and supported some of the difficulties in gene synthesis.
    • 24. Withers-Martinez C, Carpenter EP, Hackett F et al. PCR-based gene synthesis as an efficient approach for expression of the A+ T-rich malaria genome. Protein Eng. 12(12), 1113–1120 (1999). •• Substantiated the value of PCR-based gene synthetic method in overcoming unexpected joining reactions.
    • 25. Camer GA, Oikawa Y, Omaki H, Endoh D. Novel synthetic nucleotides of notifiable dengue (1–4), Japanese encephalitis, yellow fever and Zika flaviviruses. Future Sci. OA 5(1), FSO353 (2018). •• Described the pros and cons of OE-PCR gene synthesis method.
    • 26. Gao X, Yo P, Keith A, Ragan TJ, Harris TK. Thermodynamically balanced inside-out (TBIO) PCR-based gene synthesis: a novel method of primer design for high-fidelity assembly of longer gene sequences. Nucleic Acids Res. 31(22), e143–e143 (2003). •• Corroborated a newer method of primer design which can synthesize larger DNA products.
    • 27. Young L, Dong Q. Two-step total gene synthesis method. Nucleic Acids Res. 32(7), e59–e59 (2004). •• Supported the need to optimize oligomer settings for each gene.
    • 28. Lin Y, Cheng G, Wang X, Clark TG. The use of synthetic genes for the expression of ciliate proteins in heterologous systems. Gene 288(1–2), 85–94 (2002). •• Delved into the use of specialized oligonucleotide design programs.
    • 29. Mehta RK, Singh J. Bridge-overlap-extension PCR method for constructing chimeric genes. BioTechniques 26(6), 1082–1086 (1999). •• Supported the current work of attempting to match the expected product sequences.
    • 30. Caruthers MH. Gene synthesis machines: DNA chemistry and its uses. Science 230(4723), 281–285 (1985). •• Described the linking of oligomers and gene optimization.
    • 31. Sandhu GS, Aleff RA, Kline BC. Dual asymmetric PCR: one-step construction of synthetic genes. BioTechniques 12(1), 14–16 (1992). •• Described how to avoid reaction randomness during joining reactions in gene synthesis.
    • 32. Xiong AS, Yao QH, Peng RH et al. A simple, rapid, high-fidelity and cost-effective PCR-based two-step DNA synthesis method for long gene sequences. Nucleic Acids Res. 32(12), e98–e98 (2004). •• Founded and presented the values of various PCR methods as well as the enabling tool to avoid DNA synthesis errors