Recent Toolboxes for Synthetic Biology in Cyanobacteria

Due to limitations in the molecular biology tools available for cyanobacteria in comparison to the other bacteria (E. coli ), there is a need to design these tools using a synthetic biology approach. Synthetic biology manipulates the already available tools native or foreign and recombining them using different combinations for a better output. In this view, several tools like promoters, riboswitches, ribosome binding sites, CRISPR/Cas system, etc. were developed, which are briefly reviewed here. Strains likeSynechococcus elongatus PCC 7942, Synechocystis sp . PCC 6803, etc. are used as a host to test these genetic tools. One of the major demerits of these strains is the longer doubling time. Recently Yu et al. (2015) discovered Synechococcus elongatus UTEX 2973, which is the fastest-growing strain reported to date. Table 1 summarizes the commonly used cyanobacterial strains as hosts. <Table 1 >

2.1 Promoters

There are number of native and foreign promoters that are used in cyanobacteria and are summarized in Table 2. In bacteria, promoters are recognised by the σ factor of the RNA polymerase (RNA P) enzyme and aids in the transcription of the gene of interest. The promoters can be constitutive or inducible. Constitutive promoters transcribe the genes continually in an unregulated way. Whereas inducible promoters are specific to the signals like light, dark, heavy metals, nitrate/nitrite, etc. and are helpful when the intermediate/end products are toxic to the host cells. Pcpc560 considered as the super-strong promoter, was discovered by Zhou et al. (2014). It has the same expression rate as that of the E. coli producing functional proteins at a level of up to 15 % of total soluble proteins. It has two promoters from the cpcB gene and 14 transcription factor binding sites, which are assumed to be the crucial factor for its strength. In one of the studies done by Liu and Pakrasi (2018), promoter cpcB showed the highest expression (sequence identical to cpc560). They compared 13 different promoters by checking the expression of enhanced yellow fluorescence protein. Out of the thirteen promoters, twelve were native, and one was E. coli originated. Wang et al. (2018) constructed promoters’ library, compared 17 different promoters, and concluded that Ptrc gives two times better expression than the promoter PpsbA (from chloroplast of the flowering plantAmaranthus hybridus ) and seven times expression than PcpcB (Pcpc560) and its variants. The expression level was checked by expressing the ethylene forming enzyme inSynechocystis sp. PCC 6803. Ptrc2O and Ptrc1O (Huang et al., 2010 and Camsund et al., 2014) is the promoter derived from Ptrc (Brosius et al., 1985). Ptrc1O has a strong lac operator than Ptrc, whereas Ptrc2O has two lac operator site showing efficient repression. Markley et al. (2014) constructed a promoter which performs better than trc promoter, giving 48±7 fold expression of YFP. They constructed two orthogonal promoter libraries with the IPTG induction system, which were tried and tested in Synechococcus sp. strain PCC 7002. Promoter MB1, MB2, and MB3 were obtained by a change in sequences in the Biobrick promoter J23119. Similarly, 20 different synthetic promoters were assembled by modifying J23119, Ptrc1O, Ptic10, and Ptac10 promoters, and the intensity of fluorescence was checked in Synechocystis and Synechococcus elongatus UTEX 2973 (Vasudevan et al., 2019). In both the cases, J23119 showed the highest fluorescence levels. Werner et al. (2018) discovered and characterized nineteen native promoters in Synechocystis sp. PCC 6803, which are induced by 12:12 hour light and dark cycles (LD cycles). Out of these nineteen promoters, four of the promoters PhliC, Prbp1, Pslr0006l,and PsigA shows a strong correlation with 12:12 LD cycles when characterized using bacterial luciferase bioluminescent gene. In the same species, metal ions (nickel, cobalt, and zinc) induced nrsB, nrsD, nrsS, coaT, and ziaA promoters were compared with endogenous constitutive promoters (Englund et al., 2016). PnrsB was found to be the most efficient promoter, which can be regulated and tuned with the help of nickel ion. Kelly et al. (2018) inserted a rhamnose-inducible rhaBAD promoter from E. coli toSynechocystis sp. PCC 6803 which showed a controlled expression system. PsynDIF, a short 48 nucleotides long synthetic promoter for heterocyst-specific expression in filamentous cyanobacteria (Wegelius et al., 2018) can be used well for the production of oxygen intolerant enzymes as the promoter gives 10 times more expression after the heterocyst formation. Expression of the promoter changes when it is relocated from the native location to the new one (Albers and Peebles 2016). PpsbAII within the native genomic location ofSynechocystis sp. PCC 6803 showed 15.8 times increase in the transcript in comparison to only 1.6 times when the promoter is moved to neutral region slr0168. Promoter psbA was used to control the expression of mannitol encoding genes (mtlD and mlp ) inSynechococcus sp. PCC 7002, giving a yield of 1.1 g mannitol L−1 with a production rate of 0.15 g mannitol L−1 day−1 (Jacobsen and Frigaard, 2014). Huang and Lindblad (2013) constructed non-inducible promoters and showed that altering a few base pairs can change the strength of the promoters. R40 promoter was used as a template for promoter designing. Specific base pairs changes were done in the R40 promoter at TATAAT site (L12 promoter created whose strength is less than R40) and between -10 element and transcription start site (L12 promoter having strength more than R40). Three line of modifications were done between -10 element and transcription start site creating a total of 19 promoters, L01 to L016, L21 and L22, and L31. The study shows that L21 promoter has 110 (±1) times strength than L22, opening the possibilities to change the region between −27 and +3 at TATAAT sequence. Bioinformatics tools play an important role in the prediction of promoters in cyanobacteria. Btss finder is one of the means for bacterial promoter prediction, which includes E. coli and cyanobacteria. Being novel, it can identify the promoters of different sigma classes of two different phyla. Various native and foreign promoters are shown in table 2.
<Table 2 >

2.2 Ribosome Binding Sites

As the promoters regulate the initiation of transcription, in the same way, ribosome binding sites (RBS) regulate translation initiation rate of downstream target genes (Kierzek et al., 2001). Upon translation initiation, with the help of complementary base pairing of the nucleotides, the 3-terminal sequence of the 16S rRNA interacts with the core Shine-Dalgarno (SD) sequence of RBS. Ma et al. (2002) showed this in Synechocystis 6803 that the 3 terminal sequence of the 16S rRNA is AUCACCUCCUUU and its complementary SD sequence is AAAGGAGGUGAU (core SD sequence underlined). To enhance the production of 2,3-butanediol in Synechococcus 7942, expression levels of the three genes (alsS, alsD and adh) are coordinated by utilizing four different RBS from E. coli (Oliver et al., 2014). Wang et al. 2016 increased limonene synthesis by RBS engineering in Synechococcus elongatus PCC 7942. Strain L1113 showed limonene production of 32.8 μg/L/OD/d by changing the original RBS of the trc promoter. Similarly, a synthetic RBS introduced in psbA promoter increased the limonene production to 885.1 μg/L/OD/d. RBS library for cyanobacteria was created by Englund et al. (2016) by utilizing 8 RBS sequences from BioBrick Registry of standard biological parts and two from Synechocystis sp . PCC 6803 and predicted by “RBS library calculator”. The mentioned library was used to express an enhanced yellow fluorescent protein (eYFP) with the help of PnrsB, PnrsD, PnrsS, PcoaT and PziaA promoters which are induced by nickel, cobalt and zinc metal efflux pumps. More recently, Liu et al. (2018) evaluated 20 native RBS, which were 22 base pair long. Ptrc1O was selected as the promoter to check the strength of the RBS sequences by the EYFP gene. In the same manner, Thiel et al. (2018) also assessed 13 RBS out of which 7 were native ofSynechocystis 6803 , and 6 were from E. coli . Codon-optimized GFPmut3, sYFP2, and ethylene forming enzyme were used as the reporter proteins for checking translation efficiency. These studies help in the selection of suitable RBSs for overexpression of the end product. According to Reeve et al. (2014), the same RBS can have inconstant translational efficiencies in different microorganisms or different genes in the same organisms. RBS calculating tools plays an important role which is based on the thermodynamic model to predict the changes in the start codon and 5 untranslated regions in an mRNA transcript. RBS calculator, UTS design and RBS designer are the majorly used tools to determine translation rates. Each calculator is used efficiently for reverse and forward engineering. RBS calculator prognosticates TIR by enumerating the strength of 30S complex and mRNA transcript interaction (Salis et al., 2009 and Salis et al., 2011). RBS designer (Na and Lee, 2010) works by designing RBS sites synthetically on the RNA transcript, while the UTR designer (Seo et al., 2013) focuses on changing 5-UTR to alter protein expression and predicts translation efficiency. These tools serve an importance purpose of generating RBS libraries, but efficiency of them can vary. Wang et al. (2017) reported the low efficiency of RBS library created by RBS calculator and established a rational RBS design strategy. Likewise, Thiel et al. (2018) also stated that the data predicted by UTR designer and RBS calculator shows different translation efficiency than the experimental one. In another study numerous RBS calculated for bisabolene synthase gene gives 7.8 mg/L titer (Sebesta and Peebles, 2020).

2.3 Riboswitches

In comparison with the inducible promoters, riboswitches do not require additional protein factors (RNA P) and are cis-acting regulatory element which changes the conformation on binding with its ligand controlling TIR (Henkin 2008; Domin et al., 2017). This makes riboswitches an ideal tool for gene regulation. Nakahira et al. (2013) illustrated that modified theophylline-responsive riboswitches regulate gene (luciferase) expression more efficiently than inducible promoters. Further, this riboswitch was used in many studies in Synechococcus elongatusPCC 7942, Leptolyngbya sp . strain BL0902, Anabaena sp . strain PCC 7120, and Synechocystis sp . Strain WHSyn andSynechocystis 6803 to check the expression regulation of yellow and green fluorescent protein (Ma et al., 2014; Ohbayashi et al., 2016). The theophylline-responsive riboswitches used were earlier screened and characterized in the past in Gram-negative alpha- and gamma-proteobacteria and Gram-positive bacteria (Lynch and Gallivan, 2009; Topp et al., 2010). It is also used to regulate intracellular glycogen content (40 to 300% of wild type) in Synechococcus elongatus PCC7942 by controlling ADP-glucose pyrophosphorylase (GlgC). Optimised level of glycogen increases cellular robustness (Chi et al., 2019). Apart from theophylline-responsive riboswitches, cobalamin-dependent riboswitch work well in Synechococcus 7002 as the strain cannot synthesize the cobalamin itself (Perez et al., 2016). But this riboswitch cannot work in the strain which synthesize cobalamin such as Synechococcus 7942, Synechocystis 6803,Crocosphaerawatsonii WH8501 and Synechococcus sp. WH7803 (Helliwell et al., 2016). Other riboswitches used in cyanobacteria includes S-box (SAM), SAM-II (a-proteobacteria) and SAMI/IV-variant riboswitch, thiamine pyrophosphate (TPP)-riboswitch, Glycine riboswitch, SMK box translational riboswitch, Purine riboswitch, FMN riboswitch (RFN element), Lysine riboswitch, SAH (S -adenosyl-l-homocysteine) riboswitch, THF (Tetrahydrofolate) riboswitch Moco (molybdenum cofactor) riboswitch, (Sun et al., 2013; Zhang and Gladyshev, 2008; Singh et al., 2018). Some of the riboswitches inducer are toxic to the host organism and are key metabolic intermediates, therefore, only theophylline dependent riboswitch is widely used in cyanobacterial systems.

2.4 CRISPR Based Technique

The most recent synthetic biology tool is the CRISPR/Cas system, which is marker less. CRISPR/Cas stands for Clustered regularly interspaced short palindromic repeats/ CRISPR associated proteins. Its targetability is provided by the single-guide RNA (sgRNA), which is specific to the target genomic site. SgRNA directs the Cas protein to the target site which cleaves both the strand of the genome. Cyanobacteria being oligoploid and polyploid in nature shows difficulty in producing homozygous mutants (Watanabe et al., 2015 and Zerulla et al., 2016). CRISPR based technique improve this editing efficiency. Wendt et al (2016) used CRISPR/Cas 9 system to produce nonbleaching protein A (nblA) mutants of Synechococcus elongatus UTEX 2973. Once all the copies of the genes are deleted the mutants show visible results (within 1 week) as the nblA serves as visual reporter gene. These results were verified by Li et al (2016) by increasing the succinate concentration inSynechococcus elongatus PCC 7942 through glgc knock-out glta/ppc (citrate synthase/phosphoenol pyruvate carboxylase) knock in by CRISPR-Cas 9 editing. Cas 9 at higher concentrations showed toxicity inS. elongatus UTEX 2973 and PCC 7942 cells (Wendt et al., 2016 and Li et al., 2016). The quick fix to the problem was the transient expression of Cas 9 through temperature-controlled plasmid. This gave inkling to Ungerer & Pakrasi (2016) to prospect CRISPR/Cas 12a (also known as Cpf1). Markerless point mutation, a knock-out mutation or a knock-in mutation were generated S. elongatus UTEX 2973,Synechocystis sp. PCC 6803 and Anabaena sp. PCC 7120. Cas 12 have several merits over Cas 9 which includes, no requirement of tracrRNA to activate crRNA, cost efficient as it needs only 42 nucleotides RNA which is cheaper to produce, and PAM (Protospacer adjacent motif) sequence is more specific to the cleavage site. A recent application of Cas 12a was seen in developing high-throughput prototyping tool for promoter characteristics (Choi et al., 2021). This gives opportunity to rapidly characterize promoters by cell free transcription. A newer version of CRISPR/Cas system is CRISPR-interference (CRISPRi) which utilizes the dead Cas9 (dCas9) which binds to the target DNA but has lost the ability to cleave it. This type of system is essential for the genes which are crucial for the cell viability and can only be downregulated rather deleted. Yao et al. (2016) first reposted the application of CRISPRi in Synechocystis sp . PCC 6803. The sgRNAs were placed in the neutral sites (slr2030-slr2031) due to this 94 % of repression was seen in the GFP protein. CRISPRi can be used for producing carbon-based products as shown by Huang et al. (2016). They effectively repressed extrinsic (EYFP) and intrinsic genes (glgc, sdhA and sdhB) to redirect the carbon flow. This laid the foundation for the metabolic pathways in cyanobacteria. Similarly, CRISPRi is being also used to redirect fatty acid flux (Kaczmarzyk et al., 2018). Regulating plsx (phosphate acyltransferase) gene by CRISPRi system fatty alcohol production enhanced to a great extent at 10.3 mg/g dry weight. Modulation of glutamine sythetase (glnA helps in nitrogen assimilation) inAnabaena sp . PCC 7120 was shown by Higo et al. (2018). The process is regulated in such a way that ammonium is produced only when dCas inducer is present in the system. Lately, dCas12a-mediated CRISPR interference system (CRISPRi-dCas12a) was developed in cyanobacteria for repressing genes which are not needed to produce value added chemicals (repression upto 53-94%). The technique was implemented in PCC 7942 to increase squalene production by repressing aconitase (Choi and Woo, 2020).

2.5 Vectors

After successfully finalizing the genetic elements like promoters, RBS, riboswitches, etc. a platform is required for taking the gene of interest into the host cell (here, cyanobacteria). For this purpose, vectors come into role, which is a plasmid with properties having antibiotic resistance gene for selection, mobilization elements for transfer and unique restriction sites for cloning. The heterologous gene can be inserted into the genome of the cyanobacteria or replicate autonomously. The former is done with the help of integrative and latter with replicative vectors. Replicative vectors are easy to use to insert gene of interest in cyanobacteria for the bioproduction and other purposes (Heidorn et al., 2011). Autonomous expression of gene without getting inserted in the genome gives higher expression (Xia et al., 2019). Shuttle vectors, commonly used are replicative plasmids as they can express in two hosts. Jin et al. (2018) constructed shuttle vector for PCC 6803 utilizing its own plasmid. PCC 6803’s plasmid PCC 5.2 consists a replicon and combining it with pMB1 (origin of replication ofE. coli ) leads to the formation of shuttle vector pSCB-YFP. Replicative vectors require antibiotic selection/stress to stably maintain them. Table 3 shows the list of replicative vectors available for research purposes. There are number of neutral sites detected in the genome of cyanobacteria like NS I, NS II, etc. (Ng et al., 2015). Replacing these neutral sites with the gene of interest with the help of homologous recombination is the widely used method as shown in figure 1 (Lee et al., 2017). Another strategy includes integrating the heterologous gene in place of the genes which do have function in cyanobacteria or does not affect the robustness of the strain. In this case the heterologous gene utilizes the promoter, RBS, and terminator sequences of the source gene. High ploidy level in cyanobacteria prove to be the major drawback in integration of gene of interest as it is to be ensured that each and every copy has the gene (Heidorn et al., 2011). This is epitomized by Griese et al. (2011) stating PCC 7942 having 3 to 4 genomic copies per cell, and PCC 6803 having 218 and 58 genomic copies in exponential and stationary phase respectively. Earlier there was no modular cloning (MoClo) (Engler et al., 2014) system for cyanobacteria, however Vasudevan et al. (2019) combined plant MoClo with cyanobacteria making CyanoGate system kit. The kit consists of 96 parts and these can be combined with each other from level 0 to level T to form replicative or integrative vectors. All vectors are submitted at addgene for research purposes (Addgene Kit #1000000146). Commercially available replicative plasmids are listed in table 3 and integrative plasmids have been listed in table 4.
<Figure 1 > <Table 3 > <Table 4 >