Artificial complementary chromatic acclimation gene expression system in Escherichia coli

Background The development of multiple gene expression systems, especially those based on the physical signals, such as multiple color light irradiations, is challenging. Complementary chromatic acclimation (CCA), a photoreversible process that facilitates the control of cellular expression using light of different wavelengths in cyanobacteria, is one example. In this study, an artificial CCA systems, inspired by type III CCA light-regulated gene expression, was designed by employing a single photosensor system, the CcaS/CcaR green light gene expression system derived from Synechocystis sp. PCC6803, combined with G-box (the regulator recognized by activated CcaR), the cognate cpcG2 promoter, and the constitutively transcribed promoter, the PtrcΔLacO promoter. Results One G-box was inserted upstream of the cpcG2 promoter and a reporter gene, the rfp gene (green light-induced gene expression), and the other G-box was inserted between the PtrcΔLacO promoter and a reporter gene, the bfp gene (red light-induced gene expression). The Escherichia coli transformants with plasmid-encoded genes were evaluated at the transcriptional and translational levels under red or green light illumination. Under green light illumination, the transcription and translation of the rfp gene were observed, whereas the expression of the bfp gene was repressed. Under red light illumination, the transcription and translation of the bfp gene were observed, whereas the expression of the rfp gene was repressed. During the red and green light exposure cycles at every 6 h, BFP expression increased under red light exposure while RFP expression was repressed, and RFP expression increased under green light exposure while BFP expression was repressed. Conclusion An artificial CCA system was developed to realize a multiple gene expression system, which was regulated by two colors, red and green lights, using a single photosensor system, the CcaS/CcaR system derived from Synechocystis sp. PCC6803, in E. coli. The artificial CCA system functioned repeatedly during red and green light exposure cycles. These results demonstrate the potential application of this CCA gene expression system for the production of multiple metabolites in a variety of microorganisms, such as cyanobacteria. Supplementary Information The online version contains supplementary material available at 10.1186/s12934-021-01621-3.


Introduction
For the development of bioprocesses using genetically modified bacteria, the physically controllable gene expression systems are recognized as an alternative to the conventional chemical induction system, considering the cost, waste, and recycling of cultivated media. Among several physical signals, light-based regulation is ideal for the regulation of cyanobacterial genes, considering the
One of the remarkable abilities of the cyanobacterial light-regulated gene expression system is the complementary chromatic acclimation (CCA), which enables cyanobacteria to change the protein or pigment composition of phycobilisomes and accessory light-harvesting complexes linked to photosystems during exposure to red and green light [14][15][16]. In CCA, cyanobacteria recognize and adapt their photosynthetic machinery to a wavelength of light by altering the phycobilisomes composition at the transcriptional level. Extensive studies have been carried out on the most representative cyanobacterial strain that displays CCA, Fremyella diplosiphon [14] (Additional file 1: Figure S1). The same system was also reported in the marine cyanobacterial strain Synechococcus sp. PCC7335 (PCC7335), which also shows type III CCA [17] (Scheme 1).
These studies have elucidated the molecular mechanism of type III CCA, which involves reversible photo-control under both red and green light. RcaE, RcaF, and RcaC protein are reportedly responsible for the regulation of type III CCA. Each protein plays a significant role in the regulation of both phycocyanin and phycoerythrin expression in a two-component regulatory system. RcaE functions as a sensor histidine kinase, and RcaF and RcaC function as cognate response regulators [4,[18][19][20][21][22][23]. Under red light, RcaE phosphorylates RcaC via RcaF, which binds to L-boxes in CCA-regulated promoters and activates the expression of red light-upregulated genes, including cpc2. Simultaneously, RcaC binds to L-box locates at the opposite orientation upstream of pebAB (green light-upregulated operon in PCC7335), at a greater distance from the transcription start site, and decreases the transcriptional activity and represses the expression of the key green lightupregulated operon, pebAB. Under the green light, RcaC is inactivated and cannot bind to L-boxes in CCA-regulated promoters nor activate red light-upregulated genes; simultaneously, RcaC cannot repress the expression of the key green light-upregulated operon, pebAB. Type III CCA recognizes and discriminates between green and red light environments using only one sensor histidine kinase, RcaE, which phosphorylates (activates) its cognate response regulator RcaC. Moreover, depending on the location of L-boxes in the promoters where activated RcaC binds, the promoter is activated or repressed [17]. The molecular mechanism of the type III CCA light-regulated gene expression regulation mechanism prompted us to design a dual light-regulated gene expression system by employing a single photosensor system to control gene expression in a recombinant microorganism.
In this study, we designed and developed an artificial CCA green and red light-regulated gene expression system, inspired by the type III CCA system in the cyanobacteria, using a single two-component regulated photosensor system, CcaS/CcaR, combined with a G-box inserted downstream of the promoters to mimic of type III CCA light-regulated gene expression (Scheme 2). We employed the CcaS/CcaR green light system derived from PCC6803 as the only photosensor. The rfp gene was inserted downstream, whereas the inherent CcaR binding site, G-box, was located upstream, of the cpcG2 promoter. The bfp gene was inserted downstream of the P trcΔLacO and another G-box was inserted between P trcΔLacO and bfp. Under the green light illumination (Scheme 2a), CcaS activates CcaR to bind to the G-box upstream of the cpcG2 promoter, thereby upregulating the transcription of rfp; activated CcaR is also able to bind to the G-box located between P trcΔLacO and bfp, thereby downregulating the transcription of bfp. Under the red light illumination (Scheme 2b), CcaS inactivates and dephosphorylates CcaR, thereby releasing CcaR from G-boxes. Consequently, the transcription of rfp is downregulated, whereas the transcription of the bfp gene is upregulated.
To realize the designed artificial CCA gene expression system, a plasmid vector was constructed encoding the gene components of artificial CCA. E. coli DH5α was used as the model microorganism, after transformation with the plasmid encoding the phycocyanobilin (PCB) synthesis gene cassette, to demonstrate the expression of red fluorescent protein and repression of blue fluorescent protein under green light illumination, and the repression of red fluorescent protein and expression of blue fluorescent protein under red light illumination, which were regulated at the transcriptional level.

Construction of red light-regulated gene expression system using a green light sensor
First, we constructed a plasmid for the red light-regulated gene expression system, by mimicking the Rca system of the CCA system of PCC7335, but utilizing the CcaS/ Scheme 1 CCA system with red and green light-regulated gene expression system, derived from Synechococcus sp. PCC7335 [17] with some modifications. a Under red light, RcaE, RcaF, and RcaC are phosphorylated. Phosphorylated RcaC binding to L-box within the promoter region activates pcyA and cpc2 transcription and represses pebAB transcription. b Under green light, RcaE, RcaF, and RcaC are dephosphorylated. Unphosphorylated RcaC does not bind to L-box; consequently, pcyA and cpc2 transcriptions are inactivated, and pebAB is transcribed CcaR green light-sensing system of PCC6803 as the photosensor. The CcaS photoreceptor and cognate response regulator CcaR together with their native promoters were assembled with a strong P trcΔLacO promoter, which was modified by inserting a G-box sequence, between P trcΔLacO and the structural gene of a bfp as a reporter Scheme 2 An artificial CCA system with green and red light-regulated dual gene expression system by employing a single photosensor, CcaS/CcaR, combined with G-box inserted upstream of promoters. a Under green light illumination, CcaS, a green light sensor, activates the response regulator CcaR. Activated CcaR binds to the G-boxes, and upregulates the cpcG2 promoter, consequently, the rfp gene is transcribed. Simultaneously, CcaR binds to the G-box located between P trcΔLacO and reporter gene, bfp, and downregulates the transcription of bfp. b Under red light illumination, CcaS inactivates and dephosphorylates CcaR. Inactivated CcaR is no longer able to bind to the G-boxes, and downregulates the cpcG2 promoter; consequently, the rfp gene is not transcribed. Simultaneously, CcaR does not bind to the G-box located between P trcΔLacO and reporter gene, bfp, transcription under the P trcΔLacO promoter is upregulated, and the bfp gene is transcribed gene (BBa_K592024) [24]. The final construct, named pBR-RSS-bfp, functioned as a red light-regulated plasmid.
After 15 h of red light illumination, the transcription of the bfp reporter gene derived from the pBR-RSS-bfp plasmid reached its highest level of approximately 69.9 AU; this was 36.2-fold higher than that achieved under green light illumination (1.93 AU; Fig. 1a). To represent BFP at the translational, a fluorescence intensity assay was used. The fluorescence intensity of BFP was 7.35 AU/OD 595 nm after red light illumination; this was 8.75-fold higher than that after to green light illumination (0.84 AU/OD 595 nm ). These results indicated that the constructed red light-regulated gene expression system functioned, and that gene expression was induced under red light but repressed under green light.

Construction and characterization of artificial CCA gene expression system
The artificial CCA gene expression system was constructed by inserting a red light sensor region (P trcΔLacO -G box-bfp-Terminator), from the pBR-RSS-bfp plasmid, downstream of the rfp gene, from the previously constructed pBR-GSS-rfp plasmid [11], under the regulation, a single photosensor derived from two-component regulated photosensory system CcaS/CcaR green Fig. 1 Evaluation of E. coli transformants harboring pBR-RSS-bfp after 15 h under red or green light. a Transcriptional analyses of the bfp gene normalized by 16S rRNA housekeeping gene. b Translational analysis using fluorescence assay; BFP is normalized by cell density at OD 595 nm. Data represent mean ± SD from three independent experiments light-sensing system of PCC6803. The response of pBR-GSS-RSS plasmid, to red and green light illumination was evaluated, at the transcriptional and translational levels under pseudo-continuous cultivation [25].
After 18 h of cultivation at 30 °C, in an incubator shaker at 140 rpm, under dark light, the cells transformants was reached the exponential phase. Decreasing the temperature to 25 °C, and with media replacement every 12 h, the stationary phase was maintained until 54 h of the cultivation, at which point the transformants were exposed to the designated light illuminations.
The transcription of the rfp gene reached the highest level of 115.96 AU after 18 h of continuous green light illumination, which was 8.40-fold higher than that of the bfp gene (13.80 AU), (Fig. 2a). Additionally, the trend of RFP and BFP fluorescence intensity levels were consistent with these results, as RFP fluorescence intensity reached its highest at value 47.55 AU/OD 595 nm after 18 h of green light illumination, which was 8.05-fold higher than that of BFP fluorescence intensity (5.90 AU/OD 595 nm ; Fig. 2b).
In contrast, the transcription of the bfp reporter gene reached its highest level of 126 AU after 18 h of continuous red light illumination, which was 9.09-fold higher than that of the rfp gene (13.80 AU; Fig. 2c). Additionally, the BFP fluorescence intensity reached its highest value of 35.8 AU/OD 595 nm after 18 h of red illumination light induction, which was 4.06-fold higher than that of RFP fluorescence intensity (8.80 AU/OD 595 nm ; Fig. 2d). These results demonstrate that the designed and constructed artificial CCA gene expression system can regulate the expression of two genes under the red and green light using a single light sensor. Here, bfp gene expression was induced under red light while rfp gene expression was repressed, whereas rfp gene expression was induced under green light while bfp gene expression was repressed, as we expected.

Investigation of CCA under repeated red and green light illumination
Using pseudo-continuous cultivation conditions, the effect of the red and green light exposure cycles on the artificial CCA gene expression system were investigated. After 18 h of cultivation at 30 °C, in an incubator shaker at 140 rpm, under dark light and transformants reached the exponential phase, red light was applied and its effects at the transcriptional and translational levels were evaluated under pseudo-continuous cultivation conditions. The transcription of bfp reached level of 75.1 AU, after 6 h of red light illumination, which was 4.93-fold higher than that of the rfp gene (15.2 AU). When the light was changed to green light 6 h later, the level of bfp transcription decreased to 15.8 AU; in contrast, the level of the rfp transcription increased to 58.6 AU, or 3.69-fold higher than that of bfp. When the light was changed back to red 6 h later, the transcript level of bfp increased to 142 AU, and that of rfp decreased to 27.5 AU. This pattern was repeated until 54 h of illumination time (Fig. 3a).
The trend was consistent at the translational level. After 6 h of red light illumination, the BFP fluorescence intensity reached 38.4 AU/OD 595 nm , which was 4.46-fold higher than the RFP fluorescence intensity (8.61 AU/OD 595 nm ). When changed to green light 6 h later, the fluorescence intensity of BFP decreased to 11.9 AU/OD 595 nm and that of RFP increased to 41.3 AU/OD 595 nm . Moreover, when changed back to red light 6 h later, the BFP fluorescence intensity increased to 61.2 AU/OD 595 nm and RFP fluorescence intensity decreased to 13.7 AU/ OD 595 nm. Overall, during the red and green light exposure cycles, the expression of BFP increased under red light exposure while that of RFP was repressed, and that of RFP increased under green light exposure while that of BFP was repressed, respectively (Fig. 3b).

Discussion
In this study, an artificial CCA gene expression system was developed using the CcaS/CcaR green light twocomponent regulation system derived from PCC6803. This was inspired by the single photosensor-based regulation of type III CCA reported in PCC7335, which is similar to that reported in F. diplosiphon [17], to achieve dual control of the light-regulated gene expression system in E. coli.
The constructed plasmid pBR-GSS-RSS consisted of rfp and bfp genes under the regulation of the green light-inducible promoter P cpcG2 [7] and constitutive strong P trcΔLacO promoter [26], which were modified by adding a G-box sequence between the P trcΔLacO and the reporter gene, bfp. The expression of rfp was induced by P cpcG2 under green light illumination and that of bfp was induced by the P trcΔLacO promoter, modified by the addition of a G-box sequence, under red light illumination. Moreover, these effects were successfully observed using only a single photosensor.
The transcriptional level induced by cpcG2 promoter was seemed to be slower than the one in trcΔLacO promoter (Fig. 2a, c). By defining the normalized arbitrary unit (nAU), which is the proportion of AU at certain time compared with the maximum AU observed in the experiment (after 18 h of light induction), the speed of the transcription (dnAU/dt), is compared. The speed of the transcription at t = 6 for cpcG2 promoter was 4.95 h −1 , whereas the one for the trc∆LacO promoter was 5.29 h −1 . These results confirm that the transcriptional level is slower in the cpcG2 promoter, compared with the one in the trc∆LacO promoter. This difference would be attributed to the differences of the (− 10) and (− 35) elements between cpcG2 promoter and trc∆LacO promoter. The (− 10) and (− 35) elements of trc∆LacO promoter harbors the E.coli derived (− 10) and (− 35) elements (5′-TAT AAT -3′ and 5′-TTG ACA -3′), which is the Type I promoter, typically recognized by sigma factor, σ 70 , of E. coli. In contrary, cpcG2 promoter (5′-ATT CAA -3′ and 5′-AAC CGA -3′) is derived from cyanobacteria PCC6803, and lack σ 70 promoter consensus sequence, therefore, the transcription of this exogenous promoter was slower than the one from trc∆LacO promoter [26,27].
During cultivation under pseudo-continuous cultivation, the transcription level after 30 h decreased. The decrease in transcription might be attributed to the change of the status of the cells due to the low dilution rate we employed. The dilution rate achieved in this pseudo-continuous cultivation was 0.021 h −1 , which was much lower than the maximum specific growth rate of E.coli observed in this study, which is 1.9 h −1 where the highest transcription would be kept. Under continuous cultivation, the specific growth rate will be synchronized with the dilution rate, therefore, under this pseudo-continuous cultivation, the status of the cells would be later log phase or stational phase, where the cellular activity including transcriptional activity gradually decreases.
The multichromatic control of a gene expression system in E. coli has been reported previously. The expression of the lacZ gene was regulated by utilizing two photosensors, an engineered Cph8 chimera and a cyanobacterial CcaS photoreceptor, and their respective response regulators, OmpR and CcaR [28,29]. Although our current achievements were not directly compared with these-multi-sensor-based dual light-regulated gene expression systems, the utilization of a single photosensor, the CcaS/CcaR green light-sensing system derived from PCC6803 for the dual control of a light-regulated gene expression system, offers the benefit of more simpler plasmid circuit design, which will affect in the transcription factor involved as well as the regulation system, and be a significant contribution of this system.
The technology used in this study was designed for the application in the regulation of biosynthetic pathways that require multiplex control at different steps using different light colors. The results of this study hold great promise for application in cyanobacterial bioprocesses. The utilization of phytochrome-based family photoreceptors, will provide full control, and various light sensors connected to different transcription factors are available, which allows the multichromatic control of several transcription units using light [30][31][32]. Moreover, a light-based regulation system is beneficial, as neither additional expensive chemicals nor physical inducers are needed; thus, this system can be easily applied. A regulatory system that can be switched on and off by shifting two different light colors will facilitate a gene expression system that can be tightly regulated at specific times and locations.

Conclusions
In this study, an artificial CCA system was successfully developed to realize a multiple gene expression system, which was regulated by two colors, red and green lights, using a single photosensor, the CcaS/CcaR system derived from Synechocystis sp. PCC6803. The designed and constructed artificial CCA gene expression system regulated the expression of two genes, rfp, and bfp, under the red and green light; bfp expression was induced under red light while rfp expression was repressed, and rfp expression was induced under green light while bfp expression was repressed, as we expected. Moreover, the artificial CCA system functioned repeatedly during the red and green light exposure cycles; the expression of BFP increased under red light exposure while RFP was repressed, and that of RFP increased under green light exposure while that of BFP was repressed. These results demonstrate the potential application of this gene expression system for production of multiple metabolites in a varieties of microorganisms, including cyanobacteria.

Construction of plasmid encoding red light-regulated gene expression system
The gene encoding the blue fluorescent protein, bfp (BBa_ K592024) followed by a terminator (BBa_B1006) derived from the BioBrick collection [24], was inserted downstream of the constitutive P trcΔLacO promoter, by overlap PCR. The P trcΔLacO promoter was modified by inserting a G-box sequence (CTT TCC GAT TTC TTT ACG ATTT), upstream of the reporter gene, bfp. The PCR products were assembled corresponding to the BamHI and XbaI sites of pBR-GSS-rfp (Fig. 4a). pBR-GSS-rfp contains the CcaS/CcaR two-component green light-regulated system [11], and was digested with a similar restriction enzyme to omit the rfp gene under the regulation of the P cpcG2 promoter from the plasmid construction. The resulting plasmid was designated pBR-RSS-bfp (Fig. 4c), which was used as a control plasmid for red light illumination. Cloning was performed using E. coli DH5α as the host, and the pSTV-PCB plasmid [11] which encodes protein PCB production (Fig. 4b).

Construction of plasmid encoding dual control of light-regulated gene expression system
A designated plasmid harboring dual control of the light-regulated system was constructed by inserting the P trcΔLacO promoter, which has been modified as previously described in the Methods section, followed by the bfp gene encoding blue fluorescent protein (BBa_ K592024) and a terminator (BBa_B1006), derived from BioBrick collection [24], downstream of the rfp gene corresponding to the BamHI and XbaI sites of pBR-GSS-rfp [12], carrying the CcaS/CcaR two-component regulation system from Synechocystis sp. PCC6803 (PCC6803) by overlap PCR. The final designated plasmid, pBR-GSS-RSS (Fig. 4d) was cloned using E. coli DH5α as the host and the pSTV-PCB plasmid [11], encoding protein PCB production. All constructed plasmids are shown in Fig. 4, and the components are described in Table 1.

Evaluation of plasmid encoding for red light-regulated gene expression system
Transformants of E. coli harboring the plasmid encoding pBR-RSS-bfp were pre-cultured in tubes containing 2 mL of fresh LB broth medium with appropriate antibiotics (100 μg/mL ampicillin and 30 μg/mL chloramphenicol), at 37 °C, in an incubator shaker at 140 rpm, overnight in the dark. The cell culture was transferred to 20 mL of fresh LB broth medium with appropriate antibiotics, as described above, in 100-mL Erlenmeyer flasks with the initial optical density at 595 nm (OD 595 nm = 0.02) and incubated at 30 °C, in an incubator shaker at 110 rpm, under dark conditions until they reached the exponential phase, which was confirmed at OD 595 nm. After cell growth reached the exponential phase, half of the culture was exposed to red light at an intensity of 40 μmol photon m −2 s −1 , measured using a light meter (LI-250A, LI-COR, Bioscience) and the other half maintained under green light (40 μmol photon m −2 s −1 ). During cultivation, samples were withdrawn from the culture to evaluate the effect of the different colors at the transcriptional and translational levels.

Evaluation of CCA under repeated red and green light illumination
To investigate whether the CCA system functioned after repeated red and green light illumination, the effects of light on the constructed plasmid pBR-GSS-RSS was evaluated under pseudo-continuous conditions according to [25] with some modifications for E. coli. In a 500-mL Erlenmeyer flask, containing 150 mL of fresh LB medium with appropriate antibiotics (100 μg/mL ampicillin and 30 μg/mL chloramphenicol), transformants were cultivated at 30 °C, in an incubator shaker at 140 rpm, under the dark conditions until cell growth reached the exponential phase. After the transformants reached the exponential phase, cells were exposed to continuous red, continuous green, or repeated red and green (6 h of red light followed by 6 h of green light) light illumination. Every 12 h, 37.5 mL of culture was replaced with fresh medium containing appropriate antibiotics (100 μg/ mL ampicillin and 30 μg/mL chloramphenicol). During Fig. 4 Plasmid used in this research. a pBR-GSS-rfp, encodes for rfp gene driven by P cpcG2 [12]. b pSTV-PCB, encodes protein PCB production [11]. c pBR-RSS-bfp, encodes bfp gene driven by P trcΔLacO with an additional G-box upstream of the reporter gene, bfp. d pBR-GSS-RSS, encodes for rfp and bfp gene driven by P cpcG2 and P trcΔLacO , respectively, with an additional G-box upstream of the reporter gene, bfp Table 1 Plasmid construction use in this study

Plasmid name
Feature Sources pBR-GSS-rfp Plasmid encodes ccaS, ccaR, and rfp genes transcriptionally driven by P cpcG2 promoter under green light [12] pSTV-PCB Plasmid encodes protein phycocyanobilin (PCB) production [11] pBR-RSS-bfp Plasmid encodes ccaS, ccaR, and bfp genes transcriptionally driven by P trcΔLacO promoter under red light In this study pBR-GSS-RSS Plasmid encodes assembly of pBR-GSS-rfp and pBR-RSS-bfp, transcriptionally driven by P cpcG2 and P trcΔLacO under green and red light, respectively In this study cultivation, samples were withdrawn from the culture every 3 h to evaluate the effects of each condition on each designated reporter gene at the transcriptional and translational levels (Additional file 1: Figure S2).

Transcriptional analysis of rfp and bfp reporter gene by quantitative PCR (qPCR)
During cultivation, 1 mL of E. coli cells harboring pBR-RSS-bfp or pBR-GSS-RSS was periodically collected for RNA extraction. RNA was extracted by centrifugation at 4000×g for 10 min at 25 °C, using a TRI reagent base of chloroform phenol (Molecular Research Inc.). The isolated RNA was treated with DNase to remove genomic DNA and reverse transcribed into cDNA using a PrimeScript ® RT reagent kit with gDNA Eraser (Takara Bio Inc.). qPCR was performed to measure the expression of rfp and bfp reporter genes and the 16S ribosomal RNA (rRNA) housekeeping genes of E.coli using SYBR ® Premix Ex TaqTM II (Tli RNase H Plus, Takara Bio Inc.). The expression of these genes at the transcriptional level was analyzed using the ∆∆Ct method and normalized by calculating the expression of the 16S rRNA housekeeping gene.

Evaluation of red and green light-regulated protein expression
To monitor the RFP and BFP protein expression levels, 700 μL of each culture was periodically harvested and centrifuged at 12,000×g for 2 min at 25 °C, and the supernatant was discarded. Each cell pellet was resuspended in 200 μL of phosphate-buffered saline (PBS); this procedure was repeated twice. The resuspended cell pellet was transferred to a black 96-well plate. The fluorescence of each protein was measured using a plate reader (Variouskan flash spectral scanning microplate reader, Thermo Scientific) with excitation and emission wavelengths (RFP at 584 nm and 607 nm; BFP 399 nm and 456 nm, respectively). The cell density at OD 595 nm was measured after transferring 20 μL from each cell suspension to a clear 96-well plate and diluting with 180 μL of PBS.