Premethylation of Foreign DNA Improves Integrative Transformation Efficiency in Synechocystis sp. Strain PCC 6803

Abstract

Restriction digestion of foreign DNA is one of the key biological barriers against genetic transformation in microorganisms. To establish a high-efficiency transformation protocol in the model cyanobacterium, Synechocystis sp. strain PCC 6803 (Synechocystis 6803), we investigated the effects of premethylation of foreign DNA on the integrative transformation of this strain. In this study, two type II methyltransferase-encoding genes, i.e., sll0729 (gene M) and slr0214 (gene C), were cloned from the chromosome of Synechocystis 6803 and expressed in Escherichia coli harboring an integration plasmid. After premethylation treatment in E. coli, the integration plasmid was extracted and used for transformation of Synechocystis 6803. The results showed that although expression of methyltransferase M had little impact on the transformation of Synechocystis 6803, expression of methyltransferase C resulted in 11- to 161-fold-higher efficiency in the subsequent integrative transformation of Synechocystis 6803. Effective expression of methyltransferase C, which could be achieved by optimizing the 5′ untranslated region, was critical to efficient premethylation of the donor DNA and thus high transformation efficiency in Synechocystis 6803. Since premethylating foreign DNA prior to transforming Synechocystis avoids changing the host genetic background, the study thus provides an improved method for high-efficiency integrative transformation of Synechocystis 6803.

INTRODUCTION

Driven by energy sustainability and environmental concerns, there have been increasing endeavors in recent years to develop genetic manipulation tools in cyanobacteria (1–3). One of the key elements to ensure success of such efforts is to establish high-efficiency genetic transformation methodologies in cyanobacteria. Since its first demonstration in the cyanobacterium Synechococcus elongatus (formerly Anacystis nidulans) in 1970 (4), genetic transformation protocols for a variety of cyanobacterial species have been developed and optimized (5–17). However, despite the advances made in the past decades, the integrative transformation efficiencies in cyanobacteria are typically lower than for other model systems, such as Escherichia coli and yeast, mainly due to various biological barriers such as the restriction degradation system against foreign DNA (13).

Most bacteria carry specific restriction-modification (R-M) systems that are able to recognize and degrade foreign DNA. Each R-M system typically consists of one methyltransferase (also called methylase) and one restriction endonuclease. The methylase protects the self DNA from restriction digestion by methylating the nucleotides at specific DNA sequences (i.e., restriction sites), while the foreign DNA, which usually bears a different methylation pattern, would be recognized and degraded by the endonucleases (18). It has been reported that R-M systems can dramatically reduce the DNA transformation efficiency in a variety of bacteria (13, 19–24), including cyanobacteria (7, 25–28). For instance, Elhai and coworkers reported that the efficiency of conjugal transfer of shuttle vectors from E. coli into the cyanobacterium Anabaena sp. strain PCC 7120 was reduced by three restriction activities (7).

In order to overcome the restriction barrier and thus enhance the integrative transformation efficiency in cyanobacteria, three possible approaches may be considered: (i) bypassing the R-M system by removing all of the relevant restriction sites from the donor DNA (7, 29), (ii) temporary or permanent inactivation of the R-M systems (30, 31), or (iii) premethylation of the donor DNA using the methylases from the acceptor cyanobacterium (7, 8, 13, 24, 32–34). For instance, Iwai and colleagues reported that after a putative type I restriction endonuclease of the thermophilic cyanobacterium Thermosynechococcus elongatus BP-1 was disrupted, the integrative transformation efficiency was increased by 4- to 6-fold for some constructs and some previously nontransferable constructs were successfully integrated into the genome (31). Elhai and coworkers were able to increase the efficiency of conjugal transfer of shuttle vectors from E. coli into the cyanobacterium Anabaena PCC 7120 via premethylation (7). However, there is still no direct evidence that using premethylated foreign DNA improves the integrative transformation efficiency in cyanobacteria.

The cyanobacterium Synechocystis sp. strain PCC 6803 (Synechocystis 6803) is naturally transformable, and the transformation procedure was optimized over the past several decades (12, 35, 36). It was reported that the transformation efficiency in Synechocystis 6803 was increased by two orders of magnitude when the sll1354 gene, encoding the exonuclease RecJ, was knocked out (35). However, the physiological effect of inactivation of the exonuclease RecJ was unclear. In contrast, using premethylation of foreign DNA to increase the integrative transformation efficiency does not cause any change to the host genetic background and therefore is more desirable. In this study, two cytosine-specific methylase genes, sll0729 (gene M) and slr0214 (gene C), were cloned from the chromosome of Synechocystis 6803 (37). Specifically, gene C from Synechocystis 6803 encodes a cytosine-specific methyltransferase that putatively targets the first cytosine of the PvuI site (5′-CGATCG-3′) (38), and gene M has been predicted to encode a cytosine-specific methyltransferase that recognizes and functions on the first cytosine of the sequence 5′-GGCC-3′ (37, 39, 40). These two genes were cloned and coexpressed in E. coli strains harboring the integration plasmids, and the effects of premethylation of foreign DNA on the integrative transformation efficiency in Synechocystis 6803 was investigated.

MATERIALS AND METHODS

Strains and culture conditions.

All strains used in this paper are listed in Table 1. E. coli strain XL1-Blue MRF′ (Stratagene, La Jolla, CA, USA) was used as the host for all plasmids. All recombinant E. coli strains were cultivated in LB medium at 37°C, with shaking at 175 rpm when appropriate. Solid LB plates were prepared by adding agar to a final concentration of 1.5% (wt/vol). Antibiotics were added to the LB medium to final concentrations of 100 μg/ml for ampicillin, 50 μg ml⁻¹ for kanamycin, 100 μg ml⁻¹ for spectinomycin, and 100 μg ml⁻¹ for chloramphenicol when necessary to maintain the plasmids. Synechocystis 6803 was cultivated in BG11 medium under light with an intensity of 35 μE m⁻² s⁻¹. Ten millimolar N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid (TES) (pH 8.2), 3 g liter⁻¹ thiosulfate, and 1.5% (wt/vol) agar were added to BG11 before autoclaving to prepare solid agar plates. A final concentration of 10 μg ml⁻¹ kanamycin or 50 μg ml⁻¹ spectinomycin was added to the BG11 plates for selection of successful transformants.

TABLE 1.

Strains and plasmids used in this research

Strain or plasmid	Description	Reference or source
Strains
E. coli XL1-Blue MRF′	Δ(mcrA)183 Δ(mcrCB-hsdSMR-mrr)173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F′ proAB lacIqZ ΔM15 Tn10 (Tetr)]	Stratagene
Synechocystis sp. strain PCC 6803	Wild type	ATCC
Plasmids
pBS-SPtK	Ampr, pUC ori, f1(+) ori, Ptac-Kanr	41
pJU102	Tetr Smr, pUC ori, PpsbA-efe-FLAG	43
pACYC184	Cmr Tetr, p15A ori	New England BioLabs
pAC-M	sll0729 inserted between BamHI and SalI sites of pACYC184	This study
pAC-C	slr0214 inserted between BamHI and SalI sites of pACYC184	This study
pAC-MC	sll0729 and slr0214 inserted between BamHI and SalI sites of pACYC184	This study
pAC-Mv	pAC-M but with 5′-UTR optimized for sll0729	This study
pAC-Cv	pAC-C but with 5′-UTR optimized for slr0214	This study
pAC-MvC	pAC-MC but with 5′-UTR optimized for sll0729	This study

Construction of plasmids.

The integration plasmid pBS-SPtK, targeting a neutral site of the Synechocystis chromosome, was constructed as described previously (41). Briefly, the DNA fragments Arm 1 (slr1362) and Arm 2 (sll1274) were cloned from Synechocystis 6803 genomic DNA and serve as homologous regions for DNA exchange during integrative transformation of Synechocystis 6803. The P_tac promoter upstream of the Kan^r gene was directly synthesized, and Kan^r was amplified from pET-30a(+) (Novagen, Madison, WI). The schematic structure of plasmid pBS-SPtK is illustrated in Fig. 1.

FIG 1.

Integration plasmid pBS-SPtK and schematic representation of the homologous recombination. Arm 1, left homologous DNA fragment, part of which is slr1362. Arm 2, right homologous DNA fragment, part of which is sll1274. Asterisks indicate the sites with the DNA sequence 5′-GGCC-3′; arrowheads represent the sites with the DNA sequence 5′-CGATCG-3′ (PvuI site).

Helper plasmids used to express Synechocystis methylases were constructed as follows. Gene sll0729 (gene M), which encodes a modification methylase, was cloned from the genome of Synechocystis 6803 via PCR using primers MMS1 and MMS2, digested with BglII and SalI, and inserted between the BamHI and SalI sites on plasmid pACYC184 to construct plasmid pAC-M. Similarly, gene slr0214 (gene C), which encodes a cytosine-specific methyltransferase, was cloned from the genome of Synechocystis 6803 via PCR using primers CSM1 and CSM2, digested with BamHI and SalI, and inserted into pACYC184 and pAC-M digested with the same restriction enzymes, resulting in plasmids pAC-C and pAC-MC, respectively. When gene M or C was cloned by PCR, the native ribosome binding site (RBS) was included. Particularly, a point mutation was introduced by the primer in PCR amplification of gene M in order to improve the RBS (Table 2). To optimize the 5′ untranslated region (5′-UTR) for gene M, primers Ptet2 and MMS5 were used in PCR with plasmid pAC-M or pAC-MC as the template. The PCR products were digested with EcoRV before being ligated to form plasmids pAC-Mv and pAC-MvC, respectively. To optimize the 5′ untranslated region for gene C, primers CSM3 and CSM4 were used in PCR with plasmid pAC-C as the template. The PCR products were digested with HindIII and BamHI before being inserted between these two sites on plasmid pACYC184 to construct plasmid pAC-Cv. High-fidelity Phusion DNA polymerase (Thermo Fisher Scientific, Grand Island, NY, USA) was utilized in all PCR amplifications. All the plasmid constructs were confirmed by DNA sequencing. All plasmids are listed in Table 1, and the PCR primers are listed in Table 2.

TABLE 2.

Primers used for DNA recombination

Primer	Sequence (5′ to 3′)a	Usage
MMS1	GAAGATCTGAGGAATAGAACTATGGAGGAAAC	pAC-M
MMS2	ATGGTCGACTAGGATCCGTTATAACCTTCAGGATTACTCATG	pAC-M
MMS5	GACGATATCAGGAGGAATAGAACTATGGAGGAAAC	pAC-Mv, pAC-MvC
Ptet2	GACGATATCAGCAATTTAACTGTGATAAACTAC	pAC-Mv, pAC-MvC
CSM1	TAGGATCCAGGAAAAACCATGGCCAGAC	pAC-C
CSM2	ATGGTCGACTTGGAGTGGTAATTCTAACTGC	pAC-C
CSM3	GATAAGCTTTAATGCGGTAGTTTATCACAGTTAAATTGCTAGGAGGAAAAACCATGGCCAGAC	pAC-Cv
CSM4	CATGGATCCTAATTCTAACTGCTTTAGGAATG	pAC-Cv

^aRibosome binding sites are underlined, and start codons are in italic. The bold “G” in primer MMS1 indicates a deliberate C-to-G point mutation.

Preparation of DNA for transformation.

Each helper plasmid was cotransferred with the integration plasmid into E. coli strain XL1-Blue MRF′. As a control, plasmid pACYC184 was cotransferred with the integration plasmid into E. coli XL1-Blue MRF′.

The transformed E. coli strains were cultivated in antibiotic-supplemented LB medium at 30°C for 12 h before plasmids were extracted using the QIAprep Spin Miniprep kit (Qiagen, Valencia, CA, USA). The plasmid mixture purified from each strain was digested with restriction enzymes XhoI, SacI, and XbaI to confirm the coexistence of the integration plasmid and each helper plasmid or the control plasmid pACYC184. Each digested sample was further diluted and analyzed using the Agilent DNA 12000 kit on the Agilent 2100 Bioanalyzer system (Agilent Technologies, Inc., Wilmington, DE, USA) to quantify the concentrations of DNA fragments and thus to eventually calculate the concentration of the integration plasmid. Each plasmid mixture sample was diluted so that the concentration of the integration plasmid was 100 μg ml⁻¹.

Transformation of Synechocystis.

Synechocystis 6803 was grown to an optical density at 730 nm (OD₇₃₀) of ∼0.4. To transfer the integration plasmid pBS-SPtK into Synechocystis, 50 μl of Synechocystis 6803 was directly taken into each 1.5-ml Eppendorf tube and mixed with 5.5 μl of prepared plasmid DNA. To transfer integration plasmid pJU102 into Synechocystis, 2.5 ml of Synechocystis culture was concentrated to a final volume of 50 μl. Cells were then transferred into a 1.5-ml Eppendorf tube and mixed with 5.5 μl of prepared plasmid DNA. The final concentration of each integration plasmid was 10 μg ml⁻¹ in each transformation mixture. The Eppendorf tubes were then incubated at 30°C under light with an intensity of ∼15 μE m⁻² s⁻¹ for 5 h, with one shake at 2.5 h. The transformation mixture was subsequently transferred onto BG11 plates amended with 10 μg ml⁻¹ kanamycin. Colonies were counted after 1 to 2 weeks.

RESULTS AND DISCUSSION

Preparation of DNA for Synechocystis transformation.

A total of three helper plasmids were constructed to express Synechocystis DNA methylases Sll0729 (M), Slr0214 (C), or both in E. coli (see Fig. S1 in the supplemental material). Each helper plasmid has an origin of plasmid pACYC184 which is compatible with the pUC origin of the integration plasmid pBS-SPtK. Each helper plasmid could therefore coexist with the integrative plasmid in the E. coli host. After cotransformation of E. coli, the coexistence of the integration plasmid and the helper plasmid was confirmed by restriction digestion followed by agarose gel electrophoresis (see Fig. S2 in the supplemental material). The integrative plasmid pBS-SPtK was digested to 2.5 kb and 2.8 kb; plasmid pACYC184 and the helper plasmids pAC-M, pAC-C, and pAC-MC, were linearized to DNA fragments larger than 4.0 kb (see Fig. S2 in the supplemental material).

As the DNA concentration is critical to the transformation efficiency in Synechocystis (12, 35, 36), each plasmid mixture sample was further analyzed with the Bioanalyzer to quantify the DNA concentration for the integrative plasmid pBS-SPtK. Figures S2 and S3 in the supplemental material show examples of how DNA concentrations were measured using both agarose gel electrophoresis and the Bioanalyzer after restriction digestion of the plasmid mixture extracted from the cotransformed E. coli. After three replicates of DNA analysis via the Bioanalyzer, each DNA mixture sample was then diluted with deionized water to a final concentration of 100 μg ml⁻¹ before being used for transformation of Synechocystis 6803.

Effects of premethylation of DNA on the integrative transformation efficiency.

Plasmid pBS-SPtK cannot replicate and thus acts as an integrative plasmid when transferred into the cyanobacterium Synechocystis 6803 (1, 3). Synechocystis cells would not grow on the kanamycin-amended BG11 plates unless the kanamycin resistance marker was integrated into the genome via homologous recombination (Fig. 2A). To validate the integration of the antibiotic resistance marker gene into the targeted Synechocystis chromosome locus, 16 colonies were randomly picked up from the transformation plates and restreaked onto BG11 plates amended with the same antibiotics. All of the restreaked colonies were able to grow up after 1 week. Eight out of the 16 colonies were then picked up for colony PCR using primers flanking the integration site, and it was found that all these colonies had the resistance marker gene correctly inserted into the targeted Synechocystis chromosome locus (see Fig. S4 in the supplemental material). The number of colonies shown on the kanamycin-amended BG11 plates thus can be used as an indicator of the integrative transformation efficiency. As shown in Fig. 2A, passing the integration plasmid pBS-SPtK through E. coli cells expressing Synechocystis methylase M or C individually had little impact on the integrative transformation efficiency in Synechocystis 6803 (Fig. 2A). However, after the integrative plasmid pBS-SPtK was passed through E. coli harboring the helper plasmid pAC-MC, the efficiency of integrative transformation of Synechocystis 6803 dramatically increased, being about 7.5-fold higher than that of the control (n = 4; P < 0.001 using a two-tailed t test) (Fig. 2A). The results suggest that either cooperation of methylases M and C occurred upon premethylation of the integration plasmid pBS-SPtK or the expression of methylase C via plasmid pAC-C was very poor but was significantly improved when methylase C was expressed via plasmid pAC-MC.

FIG 2.

Transformation efficiency of the integrative plasmid pBS-SPtK into Synechocystis 6803. The transformation efficiency is calculated as the colony number per microgram of pBS-SPtK plasmid DNA. The asterisks indicate statistically significant results. The error bars indicate standard deviations for independent biological replicates. (A) Integrative transformation efficiency of Synechocystis by deionized water (dH₂O) (negative control), pBS-SPtK, and pBS-SPtK passed through an E. coli host that harbors either pACYC184, pAC-M, pAC-C, or pAC-MC. (B) Integrative transformation efficiency of Synechocystis by integration plasmid pBS-SPtK passed through an E. coli host that harbors either pACYC184, pAC-Mv, pAC-Cv, or pAC-MvC.

Improved expression of methylase C facilitates integrative transformation.

Transcription of methylase genes M and C in E. coli cells was verified by analyzing the mRNA using reverse transcription-quantitative PCR (RT-qPCR). On all the pACYC184-derived helper plasmids, methylase gene M or C or both were expected to be expressed via the P_tet promoter (see Fig. S1 in the supplemental material). Both M and C mRNA molecules were successfully detected from the total RNA extract of E. coli strain XL1-Blue MRF′/pAC-MC (see Fig. S5 in the supplemental material), indicating that transcription of methylase genes through the P_tet promoter was effective. Next, translation efficiencies of the methylase genes M and C were considered. The Gibbs free energy related to the translation initiation rate (TIR) of each methylase gene was calculated using a previously established method (42). It was found that the change of the Gibbs free energy, ΔG_total, was 5.13 kcal mol⁻¹ for translation initiation of gene M when expressed on pAC-M or pAC-MC, indicating that expression of gene M was very poor. The ΔG_total for gene C translation initiation on plasmids pAC-MC and pAC-C was 3.65 kcal mol⁻¹ and 6.35 kcal mol⁻¹, respectively, leading to a 3.4-fold-higher TIR for gene C via pAC-MC than via pAC-C (Table 3). This is consistent with the aforementioned improved integrative transformation efficiency when using pAC-MC but not pAC-C as the helper plasmid (Fig. 2A).

TABLE 3.

Gibbs free energies and translation initiation rates

Plasmid	ΔGtotala (relative TIR) for methylase:
	M	C
pAC-M	5.13 (249)
pAC-C		6.35 (143)
pAC-MC	5.13 (249)	3.65 (484)
pAC-Mv	−2.02 (6,214)
pAC-Cv		−4.79 (21,556)
pAC-MvC	−2.02 (6,214)	3.65 (484)

^aValues are in kilocalories mole⁻¹.

The 5′ untranslated regions (5′-UTRs) of genes M and C were subsequently engineered in order to increase the TIRs, resulting in plasmids pAC-Mv, pAC-Cv, and pAC-MvC. In these new constructs, the ΔG_total values for genes M and C were much smaller, and the TIRs for genes M and C on plasmids pAC-Mv and pAC-Cv were increased by 25- and 151-fold, respectively (Table 3). When pAC-Mv was utilized in premethylation of the integration plasmid pBS-SPtK, no improvement of the Synechocystis transformation efficiency was found. Using pAC-MvC in premethylation resulted in a 3-fold increase of the Synechocystis transformation efficiency (n = 3; P < 0.05 using a two-tailed t test) (Fig. 2B). However, when pAC-Cv was utilized in premethylation of the integration plasmid pBS-SPtK, it led to 161-fold-higher efficiency in the subsequent integrative transformation of Synechocystis 6803 (n = 3; P < 0.001 using a two-tailed t test) (Fig. 2B).

It was reported that methylase C specifically methylates the first cytosine of the sequence 5′-CGATCG-3′, which blocks restriction digestion from the PvuI and SgfI endonucleases (which recognizes 5′-GCGATCGC-3′) (38), while modification methylase M (Sll0729) methylates the first cytosine of the sequence 5′-GGCC-3′ (37, 39, 40), which blocks restriction digestion from the restriction endonuclease HaeIII (38). We screened the sequence of the integrative plasmid pSPtK and found in total two 5′-CGATCG-3′ sites and six of 5′-GGCC-3′ sites along the DNA sequence of the integration fragment (Fig. 1). We speculate that the dramatic increase of the transformation efficiency using helper plasmid pAC-Cv was probably due to the protection of the integration fragment from restriction digestion in Synechocystis 6803.

Restriction digestion of plasmids with and without premethylation treatment indicated that while pBS-SPtK and pBS-SPtK/M DNA were fully digestible by restriction endonuclease PvuI, the digestibility of pBS-SPtK/C, pBS-SPtK/MC, and pBS-SPtK/Cv gradually decreased, with pBS-SPtK/Cv being completely nondigestible by PvuI (Fig. 3). This implied that the expression level of methylase gene C increased continuously in E. coli harboring plasmids pAC-C, pAC-MC, and pAC-Cv, respectively, which was consistent with the predicted increasing TIRs (Table 3) and the increasing integrative transformation efficiency (Fig. 2).

FIG 3.

Restriction digestion of plasmids to probe DNA methylation levels. (A) Plasmids digested with restriction endonucleases SacI, XbaI, and XhoI. Solid triangles indicate digested integration plasmid pBS-SPtK; the open triangle indicates sizes of expected DNA fragments of digested helper plasmids. (B) Plasmids digested with PvuI. Solid triangles indicate expected sizes of DNA fragments of digested integration plasmid pBS-SPtK.

Our results indicate that optimization of the expression of cytosine-specific methylase C via optimizing the 5′-UTR played an essential role in improving the transformation efficiency, and the improvement of the transformation efficiency was roughly consistent with the predicted TIR of methylase gene C (Table 3) and the degree of DNA premethylation (Fig. 2 and 3). Expression of cytosine-specific methylase M via helper plasmid pAC-M seems to be already efficient enough to completely methylate the DNA of interest, and the improved 5′-UTR for the methylase gene M also led to complete methylation of the DNA of interest (see Fig. S6 in the supplemental material).

Interestingly, although there are multiple 5′-GGCC-3′ sites (Fig. 1) and premethylation of these sites successfully prevented the DNA from restriction digestion by restriction endonuclease HaeIII in the cases of pBS-SPtK/M and pBS-SPtK/Mv (see Fig. S6 in the supplemental material), no increase in integrative transformation efficiency in Synechocystis 6803 was observed at all (Fig. 2). Methylase M of Synechocystis 6803 was predicted to be very similar to the N4mC methylase (with 55% amino acid identity), which methylates the first cytosine of 5′-GGCC-3′ along the genomic DNA of cyanobacterium Anabaena PCC 7120 (40). It was previously found that the isolated Synechocystis genomic DNA was indeed methylated at the sequence 5′-GGCC-3′ and became nondigestible by HaeIII (38). We postulate here that methylation of the 5′-GGCC-3′ DNA sequence by methylase M in Synechocystis 6803 probably plays a regulatory role in other cellular processes instead of in the R-M system.

To reveal whether premethylation of foreign DNA could serve as a general method to improve the integrative transformation efficiency in Synechocystis 6803, helper plasmid pAC-Cv was transferred into E. coli XL1-Blue MRF′ harboring the integration plasmid pJU102 (see Fig. S7 in the supplemental material). Integration plasmid pJU102 was previously constructed to insert the ethylene-forming enzyme expression cassette into the slr0168 neutral site on the Synechocystis chromosome to catalyze ethylene production in the engineered Synechocystis (43). In this study, after premethylation treatment by helper plasmid pAC-Cv, the pJU102 plasmid DNA was completely nondigestible by the restriction endonuclease PvuI (Fig. 4A), and the integrative transformation efficiency of Synechocystis increased by 11-fold compared to that of the control (n = 6; P < 0.001 using a two-tailed t test) (Fig. 4B). The results show that premethylating foreign DNA prior to transformation may serve as a general method for high-efficiency transformation in Synechocystis.

FIG 4.

Premethylation of plasmid pJU102 and increased integrative transformation efficiency in Synechocystis 6803. (A) Plasmid samples were digested with restriction endonuclease PvuI for 2 h before being loaded onto the agarose gel. Solid triangles indicate expected DNA fragments of digested plasmid pJU102 (1,164 bp, 1,198 bp, and 5,603 bp). Sample pJU102 indicates plasmid extracted directly from E. coli without expression of methylase gene C; sample pJU102/Cv indicates plasmid DNA extracted from E. coli with methylase gene C being expressed via pAC-Cv. (B) The transformation efficiency is calculated as the colony number per microgram of pJU102 plasmid DNA. Synechocystis transformation efficiency using pJU102 with or without premethylation treatment is shown. pJU102 indicates sample without premethylation. pJU102/Cv indicates DNA sample with premethylation via helper plasmid pAC-Cv. The asterisk indicates that the result is statistically significant. The error bars indicate standard deviations for independent biological replicates.

Conclusions.

As research interest in genetic manipulation of cyanobacteria has dramatically increased in recent years, strategies to enhance the genetic transformation efficiency in cyanobacteria are urgently needed. Although premethylation of exogenous DNA was proven to be effective in increasing the genetic transformation efficiency in several bacterial species (24, 33, 34), experimental evidence on improving integrative transformation efficiency in cyanobacteria via DNA premethylation treatment was still missing. In this study, two cytosine-specific methylase genes, M (sll0729) and C (slr0214), were cloned from the chromosomal DNA of Synechocystis 6803 and were functionally expressed via helper plasmids in recombinant E. coli strains harboring integration plasmids. Methylase M methylates the first cytosine at the sequence 5′-GGCC-3′, and methylase C methylates the first cytosine at the sequence 5′-CGATCG-3′ as indicated by the sensitivity to restriction digestion from endonucleases HaeIII and PvuI, respectively (Fig. 3B; see Fig. S6 in the supplemental material). The transformation results indicated that while expression of the methylase M and thus premethylation of 5′-GGCC-3′ had little effect on the integrative transformation efficiency in Synechocystis 6803, expression of the methylase C was able to increase the transformation efficiency dramatically in Synechocystis 6803. Complete premethylation of the sequence 5′-CGATCG-3′ along the foreign DNA was achieved by optimization of the expression of methylase gene C via redesign of the 5′-UTR (Fig. 3 and 4A), leading to 11- to 161-fold-higher integrative transformation efficiency in Synechocystis 6803 (Fig. 2B and 4B). The method presented in this study does not affect the genetic background of Synechocystis 6803 and may be combined with other optimization strategies to further increase the efficiency. With minor modification, the same strategy could also be applied to other cyanobacteria.