Introducing multiple genomic alterations into cyanobacteria is an essential tool in the development of strains for industrial and basic research purposes. We describe a system for generating unmarked mutants in the model cyanobacterial species Synechocystis sp. PCC6803 and marked mutants in Synechococcus sp. PCC7002.
Cyanobacteria are ecologically important organisms and potential platforms for production of biofuels and useful industrial products. Genetic manipulation of cyanobacteria, especially model organisms such as Synechocystis sp. PCC6803 and Synechococcus sp. PCC7002, is a key tool for both basic and applied research. Generation of unmarked mutants, whereby chromosomal alterations are introduced into a strain via insertion of an antibiotic resistance cassette (a manipulatable fragment of DNA containing one or more genes), followed by subsequent removal of this cassette using a negative selectable marker, is a particularly powerful technique. Unmarked mutants can be repeatedly genetically manipulated, allowing as many alterations to be introduced into a strain as desired. In addition, the absence of genes encoding antibiotic resistance proteins in the mutated strain is desirable, as it avoids the possibility of ‘escape’ of antibiotic resistant organisms into the environment. However, detailed methods for repeated rounds of genetic manipulation of cyanobacteria are not well described in the scientific literature. Here we provide a comprehensive description of this technique, which we have successfully used to generate mutants with multiple deletions, single point mutations within a gene of interest and insertion of novel gene cassettes.
Cyanobacteria are an evolutionarily ancient and diverse phylum of bacteria found in nearly every natural environment on Earth. In marine ecosystems they are particularly abundant and play a key role in many nutrient cycles, accounting for approximately half of carbon fixation1, the majority of nitrogen fixation2 and hundreds of millions of tons of hydrocarbon production3 in the oceans annually. Chloroplasts, the organelle responsible for photosynthesis in eukaryotic algae and plants, are likely to have evolved from a cyanobacterium that was engulfed by a host organism4. Cyanobacteria have proved useful model organisms for the study of photosynthesis, electron transport5 and biochemical pathways, many of which are conserved in plants. In addition cyanobacteria are increasingly being used for production of food, biofuels6, electricity7 and industrial compounds8, due to their highly efficient conversion of water and CO2 to biomass using solar energy9. Many species can be cultivated on non-arable land with minimal nutrients and seawater, suggesting that cyanobacteria could potentially be grown at large scale without affecting agricultural production. Certain species are also sources of natural products, including antifungal, antibacterial and anti-cancer compounds10,11.
The ability to generate mutants is key to understanding cyanobacterial photosynthesis, biochemistry and physiology, and essential for development of strains for industrial purposes. The majority of published studies generate genetically modified strains by insertion of an antibiotic resistance cassette into the site of interest. This limits the number of mutations that can be introduced into a strain, as only a few antibiotic resistance cassettes are available for use in cyanobacteria. Strains containing genes conferring antibiotic resistance cannot be used for industrial production in open ponds, which is likely to be the only cost-effective means to produce biofuels and other low value products12. The generation of unmarked mutants overcomes these limitations. Unmarked mutants contain no foreign DNA, unless intentionally included, and can be manipulated multiple times. Therefore it is possible to generate as many alterations in a strain as desired. In addition, polar effects on genes downstream of the modification site can be minimized, allowing more precise modification of the organism13.
To generate mutant strains, suicide plasmids containing two DNA fragments identical to regions in the cyanobacterial chromosome flanking the gene to be deleted (termed the 5' and 3' flanking regions) are first constructed. Two genes are then inserted between these flanking regions. One of these encodes an antibiotic resistance protein; the second encodes SacB, which produces levansucrase, a compound conferring sensitivity to sucrose. In the first stage of the process, marked mutants, i.e. strains containing some foreign DNA, are generated. The plasmid construct is mixed with the cyanobacterial cells and the DNA is taken up naturally by the organism. Transformants are selected by growth on agar plates containing the appropriate antibiotic and the mutant genotype verified by PCR. Suicide plasmids cannot replicate within the strain of interest. Therefore any antibiotic resistant colonies will result from a recombination event whereby the gene of interest in inserted into the chromosome. To generate unmarked mutants, the marked mutant is then mixed with a second suicide plasmid containing just the 5' and 3' flanking regions. However, if insertion of foreign DNA is required, a plasmid consisting of the 5' and 3' flanking regions with a cassette containing the genes of interest inserted between these DNA fragments, can be used. Selection is via growth on agar plates containing sucrose. As sucrose is lethal to cells when the sacB gene product is expressed, the only cells that survive are those in which a second recombination event has occurred, whereby the sucrose sensitivity gene, in addition to the antibiotic resistance gene, has been recombined out of the chromosome and onto the plasmid. As a consequence of the recombinational exchange, the flanking regions and any DNA between them are inserted into the chromosome.
We have successfully used these methods to generate multiple chromosomal mutations in the same strain of Synechocystis sp. PCC6803 (hereafter referred to as Synechocystis)13,14, to introduce single point mutations into a gene of interest13 and for expression of gene cassettes. While generation of unmarked knockouts has been demonstrated prior to our work in Synechocystis15,16, a detailed method, aided by a visual presentation of the critical steps, is not publicly available. We have also applied the same method for generation of marked knockouts in another model cyanobacterium, Synechococcus sp. PCC7002 (hereafter referred to as Synechococcus). This protocol provides a clear, simple method for generating mutants and a rapid protocol for validating and storing these strains.
1. Preparation of Culture Media
2. Growth of Cyanobacterial Strains
3. Generation of Plasmid Constructs
4. Generation of Marked Synechocystis and Synechococcus Mutants
5. Generation of Unmarked Synechocystis Mutants
6. Long-term Storage of Strains
Figure 1: Plasmid construction for generation of marked and unmarked knockouts, e.g. cpcC1 and cpcC2 de Synechocystis. (A) Region of the Synechocystis genome where (B) cpcC1 and cpcC2 and adjacent genes are located. Highlighted in black is the region of the genome to be deleted in the mutant. (C) Sites of the genome which are amplified by PCR. The 5' flanking region (indicated in blue) and 3' flanking region (indicated in red) are amplified with restriction endonuclease sites for cloning into pUC19. The 5' (or 3') flanking region is excised out of pUC19 and inserted into the pUC19 + 3' (or 5') flanking region plasmid to generate plasmid B. (D) The npt1/sacB cassette from pUM24 is excised via BamHI digestion and inserted between the 5' and 3' flanking regions to generate Plasmid A. Please click here to view a larger version of this figure.
Plasmid design is critical for successful generation of both marked and unmarked mutants. Figure 1 gives an example of plasmid A and B used to generate a deletion mutant in the Synechocystis genes cpcC1 and cpcC213. In each case the 5' and 3' flanking regions are approximately 900-1,000 bp. Reduced flanking regions can be used although the smallest we have successfully trialed has been approximately 500 bp. Plasmid B can also contain a gene cassette between the 5' and 3' ~1 kb flanking regions or a modified version of the native gene sequence.
Figure 2: Verification of marked and unmarked mutants, e.g. cpcC1/cpcC2 de Synechocystis and SYNPCC7002_A1173 in Synechococcus. (A) The expected size of the wild-type Synechocystis (top), unmarked (middle) and marked knockout (bottom) amplicons generated using primers cpcC1C2for and cpcC1C2rev, approximately 200 bp on either side of the chromosomal region to be deleted. (B) The expected size of the wild-type Synechococcus (top) and marked knockout (bottom) amplicons generated using primers A1173for and A1173rev, approximately 200 bp on either side of the chromosomal region to be deleted. Agarose gel showing amplicons generated from (C) wild-type Synechocystis (lane 2), unmarked (lane 3) and marked cpcC1/cpcC2 knockouts (lane 4), and (D) wild-type Synechococcus (lane 6) and the marked SYNPCC7002_A1173 knockout (lane 7). Markers are shown in lanes 1 and 5. Please click here to view a larger version of this figure.
Upon transformation of plasmid A into the cells, typically several hundred colonies will appear on a plate after approximately 7-10 days. Colonies are <1 mm in diameter and will not increase in size for the next few weeks. Therefore it is critical to use a blunt end toothpick to remove the colony and streak it on a fresh BG11 + kanamycin agar plate. Approximately half the re-streaked colonies will grow after 4-6 days. If genes are non-essential and mutants demonstrate growth similar to the wild-type strain under continuous light of 20-40 µmol photons m-2 sec-1 (e.g. terminal oxidase mutants in Lea-Smith et al., 201314) (Figure 3), then all the chromosomes should contain a copy of the npt1/sacB cassette inserted sequence, as determined via PCR. If genes are non-essential and mutants demonstrate a slow growth phenotype under continuous light of 20-40 µmol photons m-2 sec-1 (e.g. phycobilisome deficient mutants in Lea-Smith et al., 201413) (Figure 3), then several rounds of re-streaking on BG11 agar plates with gradually increased amounts of kanamycin are essential in order to obtain a segregated marked mutant. Once a segregated mutant is obtained this should be re-streaked on a fresh BG11 plus kanamycin agar plate to ensure that segregation is complete. If repeated rounds of streaking do not result in a segregated marked mutant then the gene is likely essential for survival. Figure 4 gives an outline of the experimental steps involved in unmarked mutant generation.
Figure 3: Growth of Synechocystis mutants. Examples of mutants which demonstrate (A) similar growth to wild-type and (B) slower growth than wild-type. The ΔCOX mutant lacks cytochrome oxidase due to deletion of the CtaC1D1E1 genes. The ΔCyd mutant lacks quinol oxidase due to deletion of the CydAB genes. The olive mutant lacks a portion of the phycobilisome due to deletion of the CpcABC1C2D genes. Samples in (B) were bubbled with air to facilitate growth. Reproduced from data published in Lea-Smith et al., 201314 and 201413 (www.plantphysiol.org; Copyright American Society of Plant Biologists). Please click here to view a larger version of this figure.
Generation of unmarked mutants is highly efficient. Upon transformation of plasmid B into the marked mutant, a four day incubation period and subsequent plating on BG11 plus sucrose agar plates, hundreds of colonies are obtained per 1-10 µl of transformed cell suspension. However a series of dilutions should be trialed, ranging from 0.1 to 100 µl, as an excessive amount leads to a reduced concentration of sucrose per cell, resulting in poor selection for unmarked mutants. If a lawn of cells is seen over the entire plate then lower concentrations should be tried. Once individual colonies are obtained on plates, patching on BG11 plus sucrose and BG11 plus kanamycin agar plates is an essential step. Typically for unmarked mutants where a region of the chromosome is being deleted, the majority of colonies will be kanamycin sensitive and sucrose resistant. PCR amplification of the target region in these colonies show that nearly 100% demonstrate the unmarked mutant profile, e.g. Figure 2. If a gene cassette is being inserted into the chromosome then typically a higher proportion of kanamycin resistant and sucrose resistant colonies are observed. These mutants can grow on sucrose due to a mutation in the sacB gene. If no kanamycin sensitive and sucrose resistant colonies are generated then the gene cassette is deleterious to the cell.
Figure 4: Generation of marked and unmarked mutants in Synechocystis. Schematic detailing (A) recombination and (B) experimental steps involved in mutant generation. Plasmid A is first mixed with cells. Following incubation on agar plates containing kanamycin, colonies in which a recombination event occurs between the 5' and 3' flanking regions (indicated in blue and red, respectively) and the homologous sequence in the chromosome, are isolated. In addition, the npt1/sacB cassette between the 5' and 3' flanking regions is inserted into the chromosome. Following segregation a marked mutant is generated. Marked mutant cells are then mixed with plasmid B which can contain either (C) 1: the 5' and 3' flanking regions; 2: the 5' and 3' flanking regions with an expression cassette containing genes of interest inserted between these sequences; 3: the 5' and 3' flanking regions with the wild-type sequence with the desired nucleotide alterations inserted between these sequences. A second homologous recombination event occurs between the 5' and 3' flanking regions and the homologous regions in the chromosome, resulting in removal of the npt1/sacB cassette and either the unmarked knockout or a mutant with an insertion or altered wild-type region introduced into the chromosome. Please click here to view a larger version of this figure.
Stock solution recipes | |
Chemical | Amount (g) |
100x BG11 (per L) | |
NaNO3 | 149.6 |
MgSO4.7H2O | 7.49 |
CaCl2.2H2O | 3.6 |
Citric acid | 0.6 |
Add 1.12 ml 0.25 M Na2EDTA, pH 8.0 | |
0.25 M Na2EDTA, pH 8.0 (per 100 ml) | |
Na2EDTA | 9.3 |
Trace elements (per 100 ml) | |
H3BO3 | 0.286 |
MnCl2.4H2O | 0.181 |
ZnSO4.7H2O | 0.022 |
Na2MoO4.2H2O | 0.039 |
CuSO4.5H2O | 0.008 |
Co(NO3)2.6H2O | 0.005 |
Iron stock (per 100 ml) | |
Ferric ammonium citrate | 1.11 |
Phosphate stock (per 100 ml) | |
K2HPO4 | 3.05 |
Na2CO3 stock (per 100 ml) | |
Na2CO3 | 2 |
TES buffer, pH 8.2 (per 100 ml) | |
TES | 22.9 |
NaHCO3 stock (per 100 ml) | |
NaHCO3 | 8.4 |
HEPES, pH 8.2 (per 500 ml) | |
HEPES | 119.15 |
Vitamin B12 (Per 50 ml) | |
Cyanocobalamin | 0.02 |
Luria Bertani media (Per 500 ml) | |
Luria Bertani broth | 12.5 |
1 M MgCl2 (Per 100 ml) | |
MgCl2.6H2O | 20.33 |
Solution A (Per 200 ml) | |
MnCl2.4H2O | 0.395 |
CaCl2.2H2O | 1.47 |
2-(N-Morpholino)ethanesulfonic acid hydrate, 4-Morpholineethanesulfonic acid (MES) | 0.4265 |
Solution A + glycerol | |
10 ml solution A | |
1.5 ml glycerol |
Table 1: Solutions used in this study.
Primer | Sequence |
cpcC1C2leftfor | GTACTCTAGAGCGGCTAAATGCTACGAC |
cpcC1C2leftrev | GATCGGATCCGCGGTAATTGTTCCCTTTGA |
cpcC1C2rightfor | GATCGAGCTCTGCACTGGTCAGTCGTTC |
cpcC1C2rightrev | GACTGAATTCATCGTTGCTTGAACGGTCTC |
M13 forward | TGTAAAACGACGGCCAGT |
M13 reverse | CAGGAAACAGCTATGAC |
cpcC1C2for | GTTTTCATTGGCATCGGTCT |
cpcC1C2rev | ATGTCCCAGGAACGACTGAC |
A1173for | AGCAAACCGTTTTTGTGACC |
A1173rev | TGCAAGGTGGCGAACTGTAT |
Table 2: Primers used in this study. Restriction endonuclease sites are underlined.
The most critical steps in generation of unmarked mutants are: 1) careful plasmid design to ensure only the targeted region is altered; 2) ensuring that samples remain axenic, especially when cultured on sucrose; 3) plating transformed cells for marked mutant generation initially on BG11 agar plates lacking antibiotics, followed by addition of agar plus antibiotics 24 hr later; 4) culturing marked mutants for 4 full days prior to plating on BG11 plus sucrose agar plates: 5) ensuring that marked mutants are fully segregated and 6) thoroughly confirming the genotype of mutant strains. For this last step, additional primers designed to amplify part of the deleted region, can be used to ensure that it has been removed. Southern blotting, while laborious, can also be used. However, our experience is that the procedure outlined in this paper is sufficient for proper verification of mutants. This procedure has also been used to generate marked mutants in Synechococcus elongatus PCC7942. However, repeated transformation of this cyanobacterium has proved challenging.
If marked mutants cannot be segregated then different environmental conditions high CO2, low light (<20 µmol photons m-2 sec-1) or additional nutrients (i.e. glucose) can be tested. For example, the addition of glucose is essential in order to generate photosystem II mutants21. If marked mutants never fully segregate then the gene is probably essential for viability. However, there are examples from the literature where some research groups have been unable to knockout a gene (For example, Vipp de Synechocystis)22, only for other groups to later show that the gene is not essential23. This could be due to differences in the wild-type strains or incorrect plasmid design, resulting in polar effects on adjacent, essential genes. If a mutant does not fully segregate we would recommend that the plasmid containing the npt1 cassette from pUC18K20 between the left and right fragments be used for transformation. It is easier to verify the presence of bands corresponding to the wild-type and mutant by PCR, since this fragment is approximately 1.2 kb, compared to the 3.8 kb npt1/sacB cassette. This result is an important piece of evidence demonstrating that the gene is essential.
Generation of unmarked mutants with inserted expression cassettes is generally more challenging than development of knockout strains. We generally express genes under control of the strong cpcBAC1C2D promoter13. In some cases this may decrease the chances of successful insertion of the gene cassette, if over-expression of a protein is deleterious to the cell. Weaker promoters should then be tested. In general we have observed that the larger the gene cassette is, the more difficult it is to insert it into the genome. We have not been able to insert gene cassettes larger than 5 kb. Care must also be taken in choosing sites to insert expression cassettes into the genome. Neutral sites that do not affect cell viability or growth should be used. Examples in Synechocystis include phaAB and phaCE, which encode the proteins encoding the polyhydroxybutyrate biosynthetic pathway24,25. More recently an extensive list of neutral sites in Synechocystis has been identified26.
Generation of unmarked mutants in cyanobacteria is a slow process, taking approximately 5-7 weeks if all steps are conducting properly. This is slower than the standard method of generating marked knockouts utilized by the majority of research groups investigating cyanobacteria. However, the flexibility of being able to introduce further mutations into unmarked mutants partially compensates for this, since additional plasmids containing a range of cassettes conferring resistance to different antibiotics, do not have to be constructed. For research purposes the ability to mutate multiple genes is sometimes necessary in order to fully characterize the functional role of proteins. For example, we identified a deleterious phenotype only upon deletion of the two terminal oxidase electron sinks localized to the thylakoid membrane, since loss of only one of these complexes could be compensated for by activity of the other14. Development of a strain for industrial applications will also require multiple modifications to a strain, not just for introduction of foreign genes but also to increase photosynthetic efficiency, light harvesting optimization and deletion of competing pathways for the desired substrate.
The major factor limiting the speed of unmarked mutant generation is the slow division time of model cyanobacterial species, between 8-20 hr depending on light conditions. Under higher light intensities and CO2 concentrations, growth is faster. However, there is a risk that mutant strains which cannot tolerate either high light or CO2 will be selected against, or that mutant strains will undergo undesirable alterations prior to phenotypic characterization. Therefore this is not recommended. However, it would be highly advantageous if a more rapid protocol to generate unmarked mutants was developed. Overall, this would facilitate the development of strains for both basic research and applied applications. Such strains could be used for biofuel, biomass or chemical production or in understanding many aspects of cyanobacterial biochemistry, genetics and physiology.
The authors have nothing to disclose.
We are grateful to the Environmental Services Association Education Trust, the Synthetic Biology in Cambridge SynBio fund and the Ministry of Social Justice and Empowerment, Government of India, for financial support.
NaNO3 | Sigma | S5506 | |
MgSO4.7H2O | Sigma | 230391 | |
CaCl2 | Sigma | C1016 | |
citric acid | Sigma | C0759 | |
Na2EDTA | Fisher | EDT002 | |
H3BO3 | Sigma | 339067 | |
MnCl2.4H2O | Sigma | M3634 | |
ZnSO4.7H2O | Sigma | Z4750 | |
Na2MoO4.2H2O | Sigma | 331058 | |
CuSO4.5H2O | Sigma | 209198 | |
Co(NO3)2.6H2O | Sigma | 239267 | |
Ferric ammonium citrate | Sigma | F5879 | |
K2HPO4 | Sigma | P3786 | |
Na2CO3 | Fisher | SODC001 | |
TES | Sigma | T1375 | |
NaHCO3 | Fisher | SODH001 | |
HEPES | Sigma | H3375 | |
cyanocobalamin | Sigma | 47869 | |
Na2S2O3 | Sigma | 72049 | |
Bacto agar | BD | 214010 | |
Sucrose | Fisher | SUC001 | |
Petri dish 90 mm triple vented | Greiner | 633185 | |
0.2 µm filters | Sartorius | 16534 | |
100 mL conical flasks | Pyrex | CON004 | |
Parafilm M 100 mm x38 m | Bemis | FIL003 | |
Phusion high fidelity DNA polymerase | Phusion | F-530 | |
Agarose | Melford | MB1200 | |
DNA purification kit | MoBio | 12100-300 | |
Restriction endonucleases | NEB | ||
T4 ligase | Thermo Scientific | EL0011 | |
Luria Bertani broth | Invitrogen | 12795-027 | |
MES | Sigma | M8250 | |
Kanamycin sulfate | Sigma | 60615 | |
Ampicillin | Sigma | A9518 | |
GeneJET plasmid miniprep kit | Thermo Scientific | K0503 | |
14 mL round-bottom tube | BD falcon | 352059 | |
GoTaq G2 Flexi DNA polymerase | Promega | M7805 | |
425-600 µm glass beads | Sigma | G8772 | |
Glycerol | Sigma | G5516 | |
DMSO | Sigma | D8418 | |
Fluorescent bulbs | Gro-Lux | 69 | |
HT multitron photobioreactor | Infors |