The ethylene-forming enzyme (EFE) from Pseudomonas syringae catalyzes the synthesis of ethylene which can be easily detected in the headspace of closed cultures. A synthetic codon-optimized gene encoding N-terminal His-tagged EFE (EFEh) was expressed in Synechocystis sp. PCC 6803 (Synechocystis) and Escherichia coli (E. coli) under the control of diverse promoters in a self-replicating broad host-range plasmid. Ethylene synthesis was stably maintained in both organisms in contrast to earlier work in Synechococcus elongatus PCC 7942. The rate of ethylene accumulation was used as a reporter for protein expression in order to assess promoter strength and inducibility with the different expression systems. Several metal-inducible cyanobacterial promoters did not function in E. coli but were well-regulated in cyanobacteria, albeit at a low level of expression. The E. coli promoter Ptrc resulted in constitutive expression in cyanobacteria regardless of whether IPTG was added or not. In contrast, a Lac promoter variant, PA1lacO-1, induced EFE-expression in Synechocystis at a level of expression as high as the Trc promoter and allowed a fine level of IPTG-dependent regulation of protein-expression. The regulation was tight at low cell density and became more relaxed in more dense cultures. A synthetic quorum-sensing promoter system was also constructed and shown to function well in E. coli, however, only a very low level of EFE-activity was observed in Synechocystis, independent of cell density.
Citation: Guerrero F, Carbonell V, Cossu M, Correddu D, Jones PR (2012) Ethylene Synthesis and Regulated Expression of Recombinant Protein in Synechocystis sp. PCC 6803. PLoS ONE 7(11): e50470. doi:10.1371/journal.pone.0050470
Editor: Brett Neilan, University of New South Wales, Australia
Received: July 26, 2012; Accepted: October 23, 2012; Published: November 21, 2012
Copyright: © 2012 Guerrero et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7-ENERGY-2010-1) under grant agreement n. . The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Genetically tractable cyanobacteria are gaining attention as a host for the direct photosynthetic conversion of sunlight and CO2 into chemical energy. Recently, different strains of cyanobacteria have been engineered for the production of hydrogen , ethylene –, ethanol , butanol , isoprene  and fatty acids . The complete genome of >50 cyanobacteria species is available and comprehensive stoichiometric reconstructions have been developed . However, until now there are few reports of comprehensive metabolic engineering of multi-step pathways , . In order to enable economically sustainable biological conversion of solar energy, H2O and CO2 into fuel it is necessary to engineer the catalytic hosts specifically for the intended biotechnological purpose. This requires a molecular toolbox for metabolic engineering including promoters for user-regulated protein expression.
Promoters are the DNA regions that control gene expression and contribute to determine the rate of transcription and therefore the quantity of protein that is synthesized. In bacteria the transcription is initiated when the promoter sequence is recognized by a sigma (σ) factor which allows the formation of the RNA polymerase (RNAP) holoenzyme . Currently there is a vast quantity of standard biological parts and devices described  that has primarily been characterized in E. coli. The performance of diverse promoters in cyanobacteria was recently reviewed by Heidorn and coworkers .
For biotechnological purposes it may be important to regulate the expression of the gene(s) of interest, particularly if the gene products, their catalyzed metabolism or potentially toxic metabolites has a negative effect on growth . Well-repressed promoters may also be important in fundamental studies where the impact from the expression of a particular protein is studied with specific timing. Typically used promoters in cyanobacteria include the strong light inducible psbAI promoter (e.g. ), and the RuBisCO subunit rbcLS promoter (e.g. ) which are both constitutive under standard growth conditions. Several native metal-inducible promoters from cyanobacteria have been demonstrated to allow fine-tuned control of protein expression, including the copper controlled promoters upstream of the petE (plastocyanin gene) and petJ (cytochrome c553) genes , , although the relative strength of expression has not been compared with other promoters. Inducible and well-regulated protein expression has been reported with the E. coli Trc promoter in Synechococcus elongatus PCC 7942 , although expression with the Lac promoter was leaky . Recently, Huang and coworkers  examined a range of E. coli derived systems for protein expression in Synechocystis and reached the conclusion that there was no regulated system that was capable of strong expression in this model cyanobacterium.
Table 1. Functional blocks used in vector construction and the restriction sites used for their cloning, in the order of 5′ to 3′ ends of each functional block.doi:10.1371/journal.pone.0050470.t001
We considered ethylene biosynthesis as metabolic target to (1) establish a model system for photobiological synthesis of volatile hydrocarbon fuel and (2) allow non-invasive monitoring of the performance of engineered expression-systems. There are three known biological pathways for ethylene synthesis , . In most organisms the native precursor for ethylene synthesis is methionine. In higher plants ethylene is synthesized from methionine via the intermediate 1-Aminocyclopropane-1-carboxylic acid (ACC) by the enzymes ACC synthase and ACC oxidase . In most prokaryotes, ethylene is formed from methionine via 2-keto-4-methyl-thiobutyric acid (KMBA) catalyzed by an NADH:Fe(III) EDTA oxidoreductase . In a few plant pathogens ethylene is synthesized by an Ethylene Forming Enzyme (EFE) in a complex multi-step reaction utilizing 2-oxoglutarate (2-OG), arginine and dioxygen as substrates . This 2-OG dependent pathway was first reported in Penicillium digitatum and has been extensively studied in Pseudomonas syringae pv. phaseolicola PK2 by Fukuda and coworkers , . Heterologous expression of EFE in E. coli , , Trichoderma viride , Trichoderma reesei , Saccharomyces cerevisiae  and the cyanobacteria Synechococcus elongatus PCC 7942  and recently in Synechocystis sp. PCC 6803  have all resulted in the accumulation of ethylene in the headspace of closed vessels. Curiously, the first cyanobacterial system was highly unstable resulting in rapid development of mutants that lost the capability to synthesize ethylene . Such instability was not reported in any of the other organisms further prompting continued analysis of ethylene-synthesis in cyanobacteria.
Table 2. Genetic elements used in this work.doi:10.1371/journal.pone.0050470.t002
Table 3. Primers used in this work. Sequences recognized by restriction enzymes are designated in uppercase font. Bases that are not complementary are underlined and overhangs are shown in italic.doi:10.1371/journal.pone.0050470.t003
Materials and Methods
1.1 Organisms and growth conditions
Escherichia coli strain DH5α was grown routinely in LB medium at 37°C. Liquid cultures were incubated in a rotary shaker at 150–200 rpm. LB was supplemented with 1.5% (w/v) bactoagar for solid cultures and FeCl3 (where indicated). The growth medium was supplemented with the appropriate antibiotics at the following concentrations: 50 µg/ml spectinomycin (Sp), 20 µg/ml streptomycin (Str), 100 µg/ml ampicillin (Amp). Synechocystis sp. PCC 6803 (glucose-tolerant strain, hereafter called Synechocystis) was grown in BG-11 medium supplemented with streptomycin and spectimomycin when harboring the self-replicating vectors. The final antibiotic concentrations were Sp = 50 µg ml−1 and Str = 20 µg ml−1. To induce the synthesis of the EFE protein in strains bearing the Lac-inducible promoters, 1 mM of IPTG was added to both, E. coli and Synechocystis. Cells were cultured at 30°C in all cases. The performance of the metal inducible promoters was measured in modified BG-11 media lacking the appropriate metal to ensure maximum repression. The induction of heterologous protein synthesis was carried out by the addition of 0.5 µM of CuSO4 for cells harboring pDF-pet-EFEh, 6 µM of CoCl2 for cells harboring pDF-coa-EFEh, or 2 µM of ZnCl2 for cells harboring pDF-smt-EFEh. All the cyanobacteria cultures were routinely grown in 250 ml Erlenmeyer flasks at 30°C in a 1% CO2 enriched atmosphere with continuous shaking and a light intensity of ≈100 µE m−2 s−1.
Table 4. Strains obtained in this work. Plasmids and strains obtained in this work and their relative ethylene production. n.d. indicates that no ethylene was detected.doi:10.1371/journal.pone.0050470.t004
Figure 1. Work-flow for construction of the primary self-replicating wide-host-range vector pDF-trc.doi:10.1371/journal.pone.0050470.g001
1.2 Codon optimization and gene synthesis
The amino acid sequence of the ethylene forming enzyme from Pseudomonas syryngae pv. phaseolicola, GenBank accession D13182.1, was synthesized by ATG:biosynthetics (Germany) with the following modifications: DNA sequence was optimized for the preferred codon usage in Synechocystis; Six histidine residues (His-tag) were introduced in the protein N-terminal right after the first methionine codon; the repeated sequence cgatg was avoided; and selected unique restriction sites used for plasmid construction were avoided (table 1). For the plant pathway, the Arabidopsis thaliana polypeptide sequences ACS7 (UniProt ID Q9STR4) and ACO3 (UniProt ID O65378) were combined as a synthetic operon where each open reading frame was preceded by a ribosomal binding site. The construct was synthesized by GenScript (USA) with the following modifications: codon optimized for Synechocystis, and avoiding the selected unique restriction sites used for plasmid construction (Table 1).
Figure 2. The relationship between the amount of EFE and accumulation of ethylene in the headspace of closed cultures of E. coli.
E. coli DH5α harboring the pDF-trc-EFEh vector with Ptrc was cultivated in LB medium and induced with various concentrations of added IPTG. (A) EFEh was purified by affinity chromatography and the amount of target protein was quantified using Image Lab software (BioRad) relative to the total protein content and the final OD600 of the cultures. (B) The ethylene accumulation in closed E. coli DH5α cultivation vessels is plotted relative to the amount of recombinant EFE (in percentage relative to the amount of EFEh with maximum dose of IPTG) that was present in each vessel. Protein synthesis was induced by the addition of 25 µM or 1 mM IPTG that was added at an optical density of 0.1 (600 nm). Cultures to which no IPTG was added were used as controls. The concentration of IPTG that was added to each sample is shown in both panels.doi:10.1371/journal.pone.0050470.g002
The LuxRI synthetic quorum-sensing system was synthesized (GenScript, USA) using the protein sequences of Vibrio fischeri LuxR (UniProt ID P12746) and LuxI (UniProt ID P12747) as template, optimizing the DNA sequence for Synechocystis, and removing the restriction enzymes recognition sites shown in Table 1. For the RhlRI system the protein sequences of Pseudomonas aeruginosa PAO1 RhlR (UniProt ID P54292) and RhlI (UniProt ID P54291) were used for synthesis (GenScript, USA) with the same modifications as LuxRI. In both constructs the intergenic regions were not modified from the native sequences.
Figure 3. The rate of ethylene synthesis in E. coli and Synechocystis in response to promoter choice and IPTG.
Two different plasmids were evaluated, pDF-trc-EFEh (Ptrc), pDF-lac-EFEh (PA1lacO-1). The + symbol indicates that expression was induced by the addition of 1 mM IPTG. Cultures to which no IPTG was added are indicated by n.a.doi:10.1371/journal.pone.0050470.g003
1.3 Construction of the self-replicating wide-host-range pDF-series vectors
The general strategy to construct this set of vectors was to work in “functional blocks” (Table 1) which could be exchanged easily due to the presence of unique restriction sites flanking them. To achieve this, different genetic elements from previously available vectors (Table 2) were either digested and ligated using standard molecular biology procedures, or cloned by PCR using primers (Table 3) designed specifically to generate the appropriate restriction site. When DNA was synthesized (GenScript, USA; ATG:biosynthetics, Germany) the constructs were designed to carry the appropriate restriction sites in the borders.
Figure 4. Topography of the lac derived promoters.
The -10 and -35 hexamers are highlighted in grey, and the lac operators are underlined. The predicted center of each operator is highlighted in red font. The transcriptional start site  is highlighted with bold font and blue color.doi:10.1371/journal.pone.0050470.g004
Figure 5. The effect of IPTG concentration, promoter and cell density on the relative rate of ethylene synthesis in Synechocystis.
(A) The rate of ethylene synthesis with pDF-lac-EFEh (PA1LacO-1) in response to the optical density of the culture and the presence or absence of IPTG. Open squares: No IPTG addition. Closed circles: 1 mM IPTG. (B) is identical to (A) with the exception that the rate of ethylene synthesis is shown relative to the optical density. (C) The rate of ethylene synthesis in response to IPTG concentration (logarithmic scale) with pDF-trc-EFEh (Ptrc, black closed symbols), pDF-lac-EFEh (PA1lacO-1, white open symbols). All cultures were measured at approximately the same optical density (OD750 = 0.5). The inset graph differs only from the larger graph by having a linear scale on the X-axis.doi:10.1371/journal.pone.0050470.g005
The primary plasmid, named pDF-trc, was constructed by fusion of a DNA region containing the LacIq repressor, Trc promoter, poly-linker and transcription terminators derived from pTrc99a with the self-replicating region from the broad-host range plasmid pVZ321 (a RSF1010 derivative). The antibiotic resistance marker (Sp/Str) of pZE13-MCS (Expressys, Germany) was used to replace the existing selection marker. Finally, the codon-optimized efe-gene from Pseudomonas syringae (encoding an N-terminal 6-His tag) was inserted into the poly-linker. Additional vectors were constructed from this base plasmid by exchange of the promoter region and/or gene(s) of interest (Table 4).
Figure 6. The genomic structure of the native metal-inducible promoter elements.
The red line indicates the approximate fragment cloned for the construction of each of the pDF plasmids. The names of the generated vectors are indicated below each red line.doi:10.1371/journal.pone.0050470.g006
Figure 7. The rate of ethylene synthesis in E. coli and Synechocystis using three different metal-inducible promoters from cyanobacteria.
The rate of ethylene synthesis was analyzed in cultures of E. coli (light grey bars) and Synechocystis (dark grey bars) carrying the metal-inducible plasmids pDF-pet-EFEh (PpetE), pDF-coa-EFEh (Pcoa) and pDF-smt-EFEh (Psmt). Cyanobaterial cultures were grown in BG-11 medium, lacking copper or cobalt for the evaluation of PpetE and Pcoa, respectively. E. coli cells were grown in LB medium. The ethylene concentration was measured in closed vials that were either not induced (no addition) or induced using 0.5 µM of CuSO4 for cells harbouring pDF-pet-EFEh, 6 µM of CoCl2 for cells harbouring pDF-coa-EFEh, or 2 µM of ZnCl2 for cells harbouring pDF-smt-EFEh.doi:10.1371/journal.pone.0050470.g007
1.4 Transformation of cyanobacteria
Figure 8. The structure of the synthetic quorum-sensing induction systems used for the constructs (A) pDF-luxRI-EFEh and (B) pDF-rhlRI-EFEh.doi:10.1371/journal.pone.0050470.g008
1.5 Ethylene evolution measurement
In vivo measurements in both E. coli and cyanobacteria were carried out in cultures in the exponential growth phase. A 1 ml culture was incubated in a 10 ml serum bottle, sealed with a butyl rubber stopper, and incubated at 30°C in a rotary shaker (with growth light in the case of cyanobacteria). Gas samples were extracted from the head-space (25 to 250 µl) with a gas-tight syringe and injected in to a gas chromatograph with flame ionization detector (Perkin Elmer). Samples were separated on a CP-CarboBOND fused silica capillary column from Varian (L = 50 m x ID = 0.53 mm x OD = 0.75 mm). The injection and oven temperature were 80°C, and detector temperature 200°C. Ethylene eluted at approximately 3.5 minutes using helium as the carrier gas at a flow rate of 7 ml min−1. A mixture of 99% (v/v) N2 with 1% (v/v) C2H4 was used as a reference standard. Ethylene evolution values are expressed as the mean and standard deviation of at least three replicate cultures. To allow direct comparison between the different strains the ethylene production value was normalized against the optical density of the culture (600 nm for E. coli, 750 nm for Synechocystis).
Figure 9. The accumulation of ethylene (black line) and optical density (blue line) with either the (A) constitutive Ptrc or (B) quorum-sensing PluxRI promoters in E. coli.
Note that there is effectively no ethylene synthesis until OD600 has reached approximately 0.5 with the quorum-sensing promoter. 10 mM FeCl3 was added to the LB media in all cultures.doi:10.1371/journal.pone.0050470.g009
1.6 Protein purification
To purify the recombinant His-tagged EFE protein from E. coli, 0.3 ml of overnight LB-grown precultures were used to inoculate 15 ml of LB supplemented with different concentrations of IPTG. Cultures were incubated during 5 hours at 30°C. After incubation, 3 ml of each culture was used to measure ethylene evolution, while the remaining culture was pelleted and stored at −20°C. Frozen pellets were thawed and resuspended in a lysis buffer (20 mM Na2HPO4, 500 mM NaCl, 20 mM imidazole, 1 mg ml−1 lysozyme). After the addition of an EDTA-free protease inhibitor cocktail (Roche) cells were sonicated three times for 30 s. Lysate was clarified centrifuging 5 mins at 17000 g to pellet insoluble debris. Supernatant was used to purify the recombinant His-tagged EFE protein using the His SpinTrap™ kit (GE Healthcare) following the manufacturer recommendations. Protein content was quantified using Image Lab software (BioRad).
1.1 Expression of EFE in Synechocystis sp. PCC 6803 results in stable synthesis of ethylene
Synechocystis harbors a gene (slr1212) annotated as a (putative) ethylene-receptor that has been shown to directly bind ethylene  and regulate phototaxis in response to UV-light . Given the reported genetic instability of Synechococcus elongatus PCC 7942 transformed with EFE , we first confirmed that the addition of ethylene, at a level sufficient to influence a signaling event (1% (v/v) ethylene), to the headspace of closed cultures of Synechocystis did not influence the growth and absorption spectra (400–750 nm) over a 3 day period (data not shown). A series of constructs harboring two different ethylene pathways under the control of a diverse set of promoters were thereafter prepared in a broad-host range RSF1010 plasmid (hereafter named pDF) from a primary vector bearing the Trc promoter (Figure 1). All final plasmid constructs were confirmed to be correct by DNA sequencing. Following transformation of both E. coli DH5α and Synechocystis, ethylene accumulation was observed in the headspace of closed cultures of most strains. No ethylene synthesis could be observed in some strains (noted in Table 4) despite repeated attempts to transform the host and confirmation of successful transformation by PCR and/or plasmid isolation (and verification by sequencing or diagnostic restriction digestion). Plasmids were maintained in both hosts throughout the project period. For example Synechocystis strain transformed with plasmid pDF-trc-EFEh (harboring Ptrc and codon-optimized EFE with N-terminal His-tag) was in all cases stable and produced ethylene with a similar rate of evolution when maintained for more than 6 months in liquid culture. The addition of 10 µM iron (as FeCl3) tripled the rate of ethylene synthesis in E. coli, although no such effect was observed in cyanobacteria. Omission of the EFE-encoding gene resulted in no measurable accumulation of ethylene in either of the two hosts. The addition of six histidine residues to the N-teminal end of EFE did not influence ethylene-synthesis (data not shown).
1.2 Using EFE as a non-invasive promoter-reporter to identify systems for well-regulated expression in Synechocystis
In order to utilize EFE as a promoter-reporter, we first confirmed that the rate of ethylene synthesis correlated with the amount of recombinant EFEh protein in E. coli using Ptrc (pDF-trc-EFEh) which is an established tunable promoter . The accumulation of ethylene was measured and recombinant His-EFE protein was thereafter purified by nickel-affinity chromatography and quantified following SDS-PAGE (Figure 2). The protein content correlated with the accumulation of ethylene in the headspace suggesting that total EFE activity is the main rate-limiting factor for ethylene synthesis in E. coli. This confirms that EFE-dependent ethylene synthesis can be used as a facile and non-invasive promoter-reporter. The relative degree of control and promoter strength of a series of promoters was thereafter evaluated. The studied promoters could be divided into three groups: (1) Lac-derived (IPTG induced) promoters, (2) native cyanobacterial metal inducible promoters, and (3) synthetic quorum-sensing promoters.
1.1 1. Lac-derived promoters.
A commonly used E. coli promoter (Ptrc) was initially tested in both E. coli and Synechocystis. In E. coli the repression in the absence of IPTG was effectively complete (Figure 3). The transformation of Synechocystis with the same plasmid resulted in approximately the same rate of ethylene accumulation compared to E. coli and the presence or absence of IPTG only had a minor impact on ethylene synthesis. The rate of ethylene synthesis observed in Synechocystis was similar to that previously described when the native EFE gene was expressed in Synehococcus elongatus PCC 7942  and Synechocystis .
Since repression of the Ptrc in Synechocystis was poor, in accordance with earlier studies , further promoters were tested including a variant of the Lac promoter, PA1lacO-1  (Figure 4). Initial experiments with PA1lacO-1 showed surprisingly strong repression and similar rates of ethylene synthesis as with Ptrc (Figure 3). Upon closer inspection it was found that the degree of repression in the absence of IPTG related to the optical density of the culture at the time of ethylene-measurement (Figure 5a, 5b), with tight repression in low-density cultures and progressively weaker repression the more dense the culture became. This prompted a re-evaluation of the repression afforded by Ptrc in low-density cultures; however, the superior control with the PA1lacO-1 promoter was still evident (Figure 5c).
1.1.2. Native metal-inducible promoters from cyanobacteria.
The few systems described up to date in the literature that are able to carry out finely tuned heterologous expression in cyanobacteria are native metal-inducible promoters, including promoters preceding genes encoding plastocyanin (petE ), cytochrome c6 (petJ ), the nrsBACD operon , CoaT (coaT ) and metallothionein (smtA ). In order to compare the relative utility of these promoters with the E. coli promoters, we selected three of these systems (PpetE, Pcoa, Psmt – Figure 6), sub-cloned them into the self-replicating plasmid in place of the lacIq and Ptrc elements and evaluated all constructs in both E. coli and Synechocystis (Figure 7). Not surprisingly, the cyanobacterial promoters performed poorly in E. coli. In cyanobacteria the repression afforded by the metal-inducible promoters was much better than with Ptrc in the absence of inducing conditions, although the rate of ethylene synthesis was at best 25% compared with the strong lac-derived promoters. The relatively poor regulation observed with PpetE is probably due to the difficulty in removing all residual copper in standard water and laboratory glassware. Interestingly Psmt from Synechococcus sp. PCC 7002 allowed stronger induction of protein synthesis in E. coli compared to Synechocystis, with only residual ethylene evolution activity detected in the latter host (Figure 7, Table 4).
1.1.3. Synthetic quorum-sensing promoters.
It has been suggested that quorum-sensing promoters have evolved to enable “communication” within microbial communities . This is achieved with a two-component system where the first component encodes the protein that catalyzes the synthesis of the chemical autoinducer (N-acyl homoserine lactone), and the second component acts as sensor and regulator (Figure 8). In theory a self-regulated system may be useful in order to avoid the addition of costly chemicals to induce gene expression in cultures that cannot utilize constitutive promoters. There are no studies, as far as we are aware, that have attempted to utilize orthogonal synthetic quorum-sensing based promoters in cyanobacteria.
We designed two synthetic quorum-sensing systems (LuxRI and RhlRI, Table 2 and 4) and placed them in front of the His-tagged efe gene in the pDF vector (Figure 8). LuxRI from Vibrio fischerii is induced by 3-oxo-hexanoyl homoserine lactone , and RhlRI from Pseudomonas aeruginosa is induced by butanoyl homoserine lactone . Both constructs were used to transform both E. coli and cyanobacteria and evaluated for ethylene synthesis. In E. coli, the LuxRI construct functioned as expected, with ethylene accumulation only being observed after the culture reached a threshold cell density (OD600 ≈ 0.5, Figure 9). The rate of ethylene synthesis was similar between strains harboring the Ptrc and PluxRI constructs, suggesting that the pDF-LuxRI-EFEh construct was functional. In Synechocystis, however, only a low rate of ethylene synthesis (≈10 nl C2H4 ml−1 h−1) was observed independent of cell density (data not shown). The presence of the plasmid in the cyanobacterial cells was confirmed by PCR, together suggesting that the LuxRI promoter as designed in the present study did not function as expected. Constructs carrying the Pseudomonas aeruginosa RhlRI promoter did not function in E. coli DH5α (rate of ethylene synthesis <5 nl C2H4 ml−1 h−1, data not shown) at different cell densities, even after external addition of the inducer butanoyl homoserine lactone. The construct was therefore not tested in cyanobacteria.
1.3 Alternative ethylene pathway
Li and coworkers  demonstrated that over-expression of the two key-genes in the plant ethylene biosynthesis pathway, ACC oxidase (sourced from tomato) and ACC synthase (sourced from soybean), as a fusion product in E. coli enabled the synthesis of ethylene in vitro using crude extracts fed with S-adenosyl-L-methionine. We attempted to reconstruct the plant pathway in vivo by co-expression of the two individual proteins in E. coli designed with Arabidopsis thaliana genes as template. No ethylene accumulation was observed under standard conditions used with EFEh. Supplementation of the medium with 5 mM bicarbonate, 5 mM sodium ascorbate and 10 µM FeCl3 resulted in a production of 30 nl ethylene ml−1 h−1. In the cyanobacterium Synechocystis no activity was detected with or without supplementation.
In previous studies, when the EFE-pathway was introduced to Synechococcus elongatus PCC 7942, a rapid loss of catalytic function was repeatedly observed through targeted mutation of the encoding gene . In contrast, no loss of activity was observed in the present study in Synechocystis despite repeated sub-culturing, allowing EFE-dependent ethylene synthesis to be used as a reporter for the evaluation of promoter constructs. During the preparation of this manuscript another study appeared that also demonstrated stable ethylene evolution in Synechocystis using a codon-optimized efe gene . What is the reason for this contrast in genetic stability between the two organisms? The two previous studies ( and ) both utilized light-induced promoters and chromosomal integration of the efe gene, while in the present study we employed orthogonal promoters to express an optimized efe gene in a self-replicating plasmid system. Interestingly, under similar conditions, the ethylene evolution in all three studies was in the range of 200 nl C2H4/ml culture/hour. As it appears that neither the promoter nor the system used to express the gene (chromosomal integration versus plasmid-based) have an influence on the ethylene evolution stability, we may conclude that the instability observed in the first study () is caused by: (1) host strain, and/or (2) efe codon optimization. Although it is not possible to exclude any one of these factors, the mutations in the efe gene introduced to Synechococcus elongatus PCC 7942 were repeatedly located in the same region of the gene. This region contained repeated sequences that were removed by codon optimization in our study. Further investigation will be necessary to resolve this outstanding question.
The Lac derived promoters enabled the greatest rates of ethylene synthesis in cyanobacteria in the present study. Previously, GFP-expression with Trc and Lac promoters were compared in Synechocystis . It was concluded that the Trc promoter was strong but not regulated, whilst the Lac promoter was very weak. In contrast, we found that with only moderate variation in the promoter structure (Figure 4), a variant of the Lac promoter (PA1lacO-1) enable strong protein expression with fine-tuned regulation, although the degree of control became progressively more relaxed the higher the density of the cultures became. Since strong repression was observed with PA1lacO-1, the lack of repression with some of the Lac promoters is unlikely to be due to the functionality of the Lac repressor protein in Synechocystis.
The difference in repression between Ptrc and PA1lacO-1 is most likely due to the presence of a second lac operator sequence in the PA1lacO-1 promoter between the -35 and the -10 region (Figure 4, operator regions are underlined). The apparent enhanced competitiveness of LacIq with PA1lacO-1 may then be explained by either (1) an increased chance of binding, assuming that LacIq binding at either of the two sites will negatively influence the binding of the sigma factor, and/or (2) the additional operator is more favorably positioned to negatively influence binding by the sigma factor.
It is also possible that the difference in regulation between these promoters may be caused by the variation in the structure of the -35 and -10 regions. Ptrc possesses the “standard” bacterial structure TTGACA-17n-TATAAT, while PA1lacO-1 has a TTGACT-17n-GATACT structure. It is therefore possible that the relevant Synechocystis sigma factors display different selectivity for these two promoter regions, resulting in differing expression of the EFE protein and consequently different rates of ethylene biosynthesis.
The gradual relaxation in EFE-expression at increasing culture density that is observed with PA1lacO-1 in the absence of IPTG is unfortunate but interesting. A possible explanation is that endogenous sugars, such as allolactose, may accumulate in high-density culture cells and bind to LacIq. Alternatively, the distribution of sigma factors may change in response to cell culture density, in turn influencing binding to the promoter-region and/or competition with the LacIq repressor.
The metal-inducible promoters PpetE and Pcoa functioned well in Synechocystis as demonstrated previously . They provide a useful complement to the IPTG-inducible promoters, although the utility of PpetE may be compromised by the difficulty in completely removing copper from laboratory glassware and water. In addition, as the repressor for the PetE promoter is unknown it was not included in the plasmid constructs. Therefore, the weak repression observed with the PetE promoter may also have been caused by an imbalance between PetE and its corresponding repressor. In contrast, the repression of the transcription using Pcoa in a cobalt-depleted BG-11 medium was excellent, even without extra precautions in preparation of the media. In the case of Psmt it is not a surprise that this system from Synechococcus sp. PCC 7002 does not function in Synechocystis, taking into account that Synechocystis lacks the gene for metallothionein. Nevertheless, Synechocystis contains a zinc exporter – ziaA – that is controlled by an SmtB-like repressor (ziaB) . In addition, the upstream region of the zinc response gene ziaA shows some homology with the sequence upstream of the Synechococcus sp. PCC 7002 metallothionein gene. Still, these similarities are not sufficient to make the heterologous Psmt system function in Synechocystis.
The lack of induction in cyanobacteria with the synthetic quorum-sensing systems LuxRI and RhlRI may be explained by a lack of readily available substrate (acyl-ACP and S-adenosyl-L-methionine) or incompatibility between the native sigma factors and the heterologous promoter regions from which the LuxRI and RhlRI regions were sourced. In addition, RhlRI is in Pseudomonas aeruginosa associated with the LasRI system in a manner not fully understood yet . It is therefore possible that also factors other than substrate are missing. The addition of commercial homo-serine lactones from multiple suppliers was also tried without effect, most likely reflecting the instability of the compounds since GC-MS analysis repeatedly failed to yield any distinct peak with the expected mass spectra (data not shown).
The ACS-ACO pathway strains did not produce ethylene in neither of the two tested prokaryotes even though ethylene synthesis earlier was reported with a recombinant synthetic ACS-ACO fusion . Closer inspection of the work of Li and coworkers  indicate that only minute quantities of ethylene were obtained with crude extracts of the synthetic ACS-ACO fusion and no information regarding in vivo synthesis in E. coli cultures were provided. There are several possible reasons for the lack of activity in the present study: (1) unfortunate choice of gene-source, in turn influencing protein synthesis and/or activity with available substrate, (2) lack of stabilizing ACS factors present in native environment , (3) lack of pyridoxal phosphate (PLP) as a cofactor, (4) inactivation of ACO by Cobalt present in standard BG-11 media. Given that ethylene synthesis was observed upon addition of substrate to E. coli whole cells, and no ethylene was observed with negative control cells that harbored the empty pDF-trc vector, the lack of functional recombinant ACS and ACO could at least be discarded. Given that the EFE-pathway functioned well and the ACS-ACO pathway is expected to also generate HCN as an undesirable by-product, we did not further examine the cause of the lack of function with the ACS-ACO pathway.
We recommend the PA1lacO-1 promoter for use in Synechocystis in cases when reasonably well-regulated and potentially strong protein expression is desired, although it will be important to monitor and control culture density for reproducible outcome. If lower protein expression levels are acceptable, the metal-inducible promoters such as Pcoa will be able to offer effectively complete repression in the absence of inducing conditions. Since stable production of ethylene in cyanobacteria is now possible with Synechocystis, this opens the road for the establishment of photo-biotechnological systems for direct conversion of sunlight, CO2 and water into this highly versatile industrial chemical. However, further optimization of the host and cultivation system will most likely be needed to enable economically sustainable production systems.
Conceived and designed the experiments: FG PRJ. Performed the experiments: FG VC MC DC. Analyzed the data: FG VC MC DC PRJ. Contributed reagents/materials/analysis tools: FG VC MC DC. Wrote the paper: FG PRJ.
- 1. Ducat DC, Sachdeva G, Silver PA (2011) Rewiring hydrogenase-dependent redox circuits in cyanobacteria. Proc Natl Acad Sci U S A 108: 3941–3946. doi: 10.1073/pnas.1016026108
- 2. Sakai M, Ogawa T, Matsuoka M, Fukuda H (1997) Photosynthetic Conversion of Carbon Dioxide to Ethylene by the Recombinant Cyanobacterium, Synechococcus sp. PCC 7942, Which Harbors a Gene for the Ethylene-Forming Enzyme of Pseudomonas syringae. Journal of Fermentation and Bioengineering 84: 434–443. doi: 10.1016/s0922-338x(97)82004-1
- 3. Takahama K, Matsuoka M, Nagahama K, Ogawa T (2003) Construction and analysis of a recombinant cyanobacterium expressing a chromosomally inserted gene for an ethylene-forming enzyme at the psbAI locus. J Biosci Bioeng 95: 302–305. doi: 10.1016/s0922-338x(97)82004-1
- 4. Ungerer J, Tao L, Davis M, Ghirardi M, Maness P-C, et al. (2012) Sustained photosynthetic conversion of CO2 to ethylene in recombinant cyanobacterium Synechocystis 6803. Energy & Environmental Science.
- 5. Deng M, Coleman J (1999) Ethanol synthesis by genetic engineering in cyanobacteria. Appl Environ Microbiol 65: 523–528. doi: 10.1016/s0922-338x(97)82004-1
- 6. Lan EI, Liao JC (2011) Metabolic engineering of cyanobacteria for 1-butanol production from carbon dioxide. Metab Eng 13: 353–363. doi: 10.1016/j.ymben.2011.04.004
- 7. Lindberg P, Park S, Melis A (2009) Engineering a platform for photosynthetic isoprene production in cyanobacteria, using Synechocystis as the model organism. Metab Eng.
- 8. Liu X, Sheng J, Curtiss R (2011) Fatty acid production in genetically modified cyanobacteria. Proc Natl Acad Sci U S A 108: 6899–6904. doi: 10.1073/pnas.1103014108
- 9. Knoop H, Zilliges Y, Lockau W, Steuer R (2010) The metabolic network of Synechocystis sp. PCC 6803: systemic properties of autotrophic growth. Plant Physiol 154: 410–422. doi: 10.1104/pp.110.157198
- 10. Lan EI, Liao JC (2012) ATP drives direct photosynthetic production of 1-butanol in cyanobacteria. Proc Natl Acad Sci U S A 109: 6018–6023. doi: 10.1073/pnas.1200074109
- 11. Imamura S, Asayama M (2009) Sigma factors for cyanobacterial transcription. Gene Regul Syst Bio 3: 65–87.
- 12. Canton B, Labno A, Endy D (2008) Refinement and standardization of synthetic biological parts and devices. Nat Biotechnol 26: 787–793. doi: 10.1038/nbt1413
- 13. Heidorn T, Camsund D, Huang HH, Lindberg P, Oliveira P, et al. (2011) Synthetic biology in cyanobacteria engineering and analyzing novel functions. Methods Enzymol 497: 539–579. doi: 10.1016/B978-0-12-385075-1.00024-X
- 14. Kämäräinen J, Knoop H, Stanford NJ, Guerrero F, Akhtar MK, et al. (2012) Physiological tolerance and stoichiometric potential of cyanobacteria for hydrocarbon fuel production. J Biotechnol. 162: 67–74.
- 15. Ghassemian M, Wong B, Ferreira F, Markley JL, Straus NA (1994) Cloning, sequencing and transcriptional studies of the genes for cytochrome c-553 and plastocyanin from Anabaena sp. PCC 7120. Microbiology 140 (Pt 5): 1151–1159.
- 16. Tan X, Yao L, Gao Q, Wang W, Qi F, et al. (2011) Photosynthesis driven conversion of carbon dioxide to fatty alcohols and hydrocarbons in cyanobacteria. Metab Eng 13: 169–176. doi: 10.1016/j.ymben.2011.01.001
- 17. Geerts D, Bovy A, de Vrieze G, Borrias M, Weisbeek P (1995) Inducible expression of heterologous genes targeted to a chromosomal platform in the cyanobacterium Synechococcus sp. PCC 7942. Microbiology 141 (Pt 4): 831–841. doi: 10.1099/13500872-141-4-831
- 18. Huang H, Camsund D, Lindblad P, Heidorn T (2010) Design and characterization of molecular tools for a Synthetic Biology approach towards developing cyanobacterial biotechnology. Nucleic Acids Res 38: 2577–2593. doi: 10.1093/nar/gkq164
- 19. Wang KL, Li H, Ecker JR (2002) Ethylene biosynthesis and signaling networks. Plant Cell 14 Suppl: S131–151 doi: 10.1099/13500872-141-4-831
- 20. Fukuda H, Ogawa T, Tanase S (1993) Ethylene production by micro-organisms. Adv Microb Physiol 35: 275–306. doi: 10.1099/13500872-141-4-831
- 21. Lin Z, Zhong S, Grierson D (2009) Recent advances in ethylene research. J Exp Bot 60: 3311–3336. doi: 10.1093/jxb/erp204
- 22. Ogawa T, Takahashi M, Fujii T, Tazaki M, Fukuda H (1990) The Role of NADH:Fe(III)EDTA Oxidoreductase in Ethylene Formation from 2-Keto-4-Methylthiobutyrate. Journal of Fermentation and Bioengineering 69: 287–291. doi: 10.1016/0922-338x(90)90107-8
- 23. Nagahama K, Ogawa T, Fujii T, Tazaki M, Tanase S, et al. (1991) Purification and properties of an ethylene-forming enzyme from Pseudomonas syringae pv. phaseolicola PK2. Journal of General Microbiology 137: 2281–2286. doi: 10.1099/00221287-137-10-2281
- 24. Fukuda H, Ogawa T, Ishihara K, Fujii T, Nagahama K, et al. (1992) Molecular cloning in Escherichia coli, expression, and nucleotide sequence of the gene for the ethylene-forming enzyme of Pseudomonas syringae pv. phaseolicola PK2. Biochem Biophys Res Commun 188: 826–832. doi: 10.1016/0006-291X(92)91131-9
- 25. Fukuda H, Sakai M, Nagahama K, Fujii T, Matsuoka M, et al. (1994) Heterologous expression of the gene for the ethylene-forming enzyme from Pseudomonas syringae in the cyanobacterium Synechococcus. Biotechnology Letters 16: 1–6. doi: 10.1016/0922-338x(90)90107-8
- 26. Ishihara K, Matsuoka M, Inoue Y, Tanase S, Ogawa T, et al. (1995) Overexpression and in vitro reconstitution of the ethylene-forming enzyme from Pseudomonas syringae. Journal of Fermentation and Bioengineering 79: 205–211. doi: 10.1016/0922-338x(95)90604-x
- 27. Tao L, Dong H-J, Chen X, Chen S-F, Wang T-H (2008) Expression of ethylene-forming enzyme (EFE) of Pseudomonas syringae pv. glycinea in Trichoderma viride. Applied Microbiology and Biotechnology 80: 573–578. doi: 10.1007/s00253-008-1562-7
- 28. Chen X, Liang Y, Hua J, Tao L, Qin W, et al. (2010) Overexpression of bacterial ethylene-forming enzyme gene in Trichoderma reesei enhanced the production of ethylene. Int J Biol Sci 6: 96–106. doi: 10.7150/ijbs.6.96
- 29. Pirkov I, Albers E, Norbeck J, Larsson C (2008) Ethylene production by metabolic engineering of the yeast Saccharomyces cerevisiae. Metabolic Engineering 10: 276–280. doi: 10.1016/j.ymben.2008.06.006
- 30. Eaton-Rye JJ (2011) Construction of gene interruptions and gene deletions in the cyanobacterium Synechocystis sp. strain PCC 6803. Methods Mol Biol 684: 295–312. doi: 10.1007/978-1-60761-925-3_22
- 31. Matsuoka M, Takahama K, Ogawa T (2001) Gene replacement in cyanobacteria mediated by a dominant streptomycin-sensitive rps12 gene that allows selection of mutants free from drug resistance markers. Microbiology 147: 2077–2087.
- 32. Rodríguez FI, Esch JJ, Hall AE, Binder BM, Schaller GE, et al. (1999) A copper cofactor for the ethylene receptor ETR1 from Arabidopsis. Science 283: 996–998. doi: 10.1126/science.283.5404.996
- 33. Song JY, Cho HS, Cho JI, Jeon JS, Lagarias JC, et al. (2011) Near-UV cyanobacteriochrome signaling system elicits negative phototaxis in the cyanobacterium Synechocystis sp. PCC 6803. Proc Natl Acad Sci U S A 108: 10780–10785. doi: 10.1073/pnas.1104242108
- 34. Amann E, Brosius J, Ptashne M (1983) Vectors bearing a hybrid trp-lac promoter useful for regulated expression of cloned genes in Escherichia coli. Gene 25: 167–178. doi: 10.1016/0378-1119(83)90222-6
- 35. Lutz R, Bujard H (1997) Independent and tight regulation of transcriptional units in Escherichia coli via the LacR/O, the TetR/O and AraC/I1-I2 regulatory elements. Nucleic Acids Res 25: 1203–1210. doi: 10.1093/nar/25.6.1203
- 36. Briggs LM, Pecoraro VL, McIntosh L (1990) Copper-induced expression, cloning, and regulatory studies of the plastocyanin gene from the cyanobacterium Synechocystis sp. PCC 6803. Plant Mol Biol 15: 633–642. doi: 10.1007/BF00017837
- 37. Kuchmina E, Wallner T, Kryazhov S, Zinchenko VV, Wilde A (2012) An expression system for regulated protein production in Synechocystis sp. PCC 6803 and its application for construction of a conditional knockout of the ferrochelatase enzyme. J Biotechnol.
- 38. Peca L, Kós PB, Máté Z, Farsang A, Vass I (2008) Construction of bioluminescent cyanobacterial reporter strains for detection of nickel, cobalt and zinc. FEMS Microbiol Lett 289: 258–264. doi: 10.1111/j.1574-6968.2008.01393.x
- 39. Turner JS, Glands PD, Samson AC, Robinson NJ (1996) Zn2+-sensing by the cyanobacterial metallothionein repressor SmtB: different motifs mediate metal-induced protein-DNA dissociation. Nucleic Acids Res 24: 3714–3721. doi: 10.1093/nar/24.19.3714
- 40. Waters CM, Bassler BL (2005) Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol 21: 319–346. doi: 10.1146/annurev.cellbio.21.012704.131001
- 41. Hawkins AC, Arnold FH, Stuermer R, Hauer B, Leadbetter JR (2007) Directed evolution of Vibrio fischeri LuxR for improved response to butanoyl-homoserine lactone. Appl Environ Microbiol 73: 5775–5781. doi: 10.1128/AEM.00060-07
- 42. Lamb JR, Patel H, Montminy T, Wagner VE, Iglewski BH (2003) Functional domains of the RhlR transcriptional regulator of Pseudomonas aeruginosa. J Bacteriol 185: 7129–7139. doi: 10.1128/JB.185.24.7129-7139.2003
- 43. Li N, Jiang XN, Cai GP, Yang SF (1996) A novel bifunctional fusion enzyme catalyzing ethylene synthesis via 1-aminocyclopropane1-carboxylic acid. J Biol Chem 271: 25738–25741. doi: 10.1074/jbc.271.42.25738
- 44. Thelwell C, Robinson NJ, Turner-Cavet JS (1998) An SmtB-like repressor from Synechocystis PCC 6803 regulates a zinc exporter. Proc Natl Acad Sci U S A 95: 10728–10733. doi: 10.1073/pnas.95.18.10728
- 45. Heurlier K, Dénervaud V, Pessi G, Reimmann C, Haas D (2003) Negative control of quorum sensing by RpoN (sigma54) in Pseudomonas aeruginosa PAO1. J Bacteriol 185: 2227–2235. doi: 10.1128/JB.185.7.2227-2235.2003
- 46. Chae HS, Faure F, Kieber JJ (2003) The eto1, eto2, and eto3 mutations and cytokinin treatment increase ethylene biosynthesis in Arabidopsis by increasing the stability of ACS protein. Plant Cell 15: 545–559. doi: 10.1105/tpc.006882