DNA-free two-gene knockout in Chlamydomonas reinhardtii via CRISPR-Cas9 ribonucleoproteins

Microalgae are versatile organisms capable of converting CO2, H2O, and sunlight into fuel and chemicals for domestic and industrial consumption. Thus, genetic modifications of microalgae for enhancing photosynthetic productivity, and biomass and bio-products generation are crucial for both academic and industrial applications. However, targeted mutagenesis in microalgae with CRISPR-Cas9 is limited. Here we report, a one-step transformation of Chlamydomonas reinhardtii by the DNA-free CRISPR-Cas9 method rather than plasmids that encode Cas9 and guide RNAs. Outcome was the sequential CpFTSY and ZEP two-gene knockout and the generation of a strain constitutively producing zeaxanthin and showing improved photosynthetic productivity.

As a result, small insertions and deletions (indels), detected by targeted deep sequencing, were induced at frequencies of up to 0.56% ( Supplementary Fig. 2). From RGEN-transfected colonies in each Petri dish (hundreds of colonies per dish), we picked several putative CpFTSY knockout cells by visual coloration examination ( Supplementary Fig. 3). We confirmed six Δ CpFTSY mutants by confirming pale green color (Fig. 1b), by measuring a higher Chl a to Chl b ratio compared to that in the wild type (Fig. 1c), and by performing Sanger sequencing ( Fig. 1d and Supplementary Fig. 4), in each isolated line. For all Δ CpFTSY mutants we obtained, total amounts of Chl were three times lower but Chl a to Chl b ratios were two or three times higher than those of wild type (Fig. 1c). In addition, indel patterns were measured at the expected positions, 3 nt upstream of protospaceradjacent motif (PAM) sequence (Fig. 1d), indicating the targeted gene mutations by CRISPR RGENs. Notably, we obtained targeted Δ CpFTSY mutants more efficiently and directly by this one-step method, compared to the time consuming conventional generation of knock out mutants 19 . Targeted ZEP gene knockout using RGEN RNPs. We next applied RGEN-RNPs to block zeaxanthin (Zea) epoxidation and, thereby, to accumulate this xanthophyll in C. reinhardtii. Zea is a macular pigment of retina, which can prevent the development of chronic diseases such as age-related macular degeneration by filtering hazardous blue light and UV 20 . Blocking the epoxidation step from Zea to violaxanthin (Vio), catalyzed by the zeaxanthin epoxidase (ZEP), can lead to constitutive accumulation of Zea 21,22 (Fig. 2a). Accordingly, we designed five sgRNAs in the ZEP locus ( Supplementary Fig. 5) as described above and transfected each sgRNA with the purified Cas9 protein into C. reinhardtii cells. Targeted deep sequencing showed that indels were induced at a frequency of 0.46% (Supplementary Fig. 6). We picked several putative ZEP knockout cells by measuring the Chl  Fig. 9). As expected, indel patterns were shown at the expected positions ( Fig. 2b) and Zea were significantly increased more than ten times in Δ ZEP mutants compared to the wild type, even under low light growth conditions (Fig. 2c).
Two-gene of ZEP and CpFTSY knockout using RGEN RNPs. We then sought to combine the two genotypes to enhance productivity and to accumulate Zea in mass culture under bright sunlight conditions. Thus, we transfected RGEN-RNPs targeted to CpFTSY into one of Δ ZEP mutants, Δ Z1 in Fig. 2b, to obtain two-gene knockout mutants. As a result, indels were measured by targeted deep sequencing at the target site with a frequency of 1.1% ( Supplementary Fig. 10). We were ultimately able to obtain fourteen Δ ZEP/Δ CpFTSY mutant lines by confirming pale green color ( Supplementary Fig. 11), by performing Sanger sequencing ( Supplementary  Fig. 12), and by measuring a higher Chl a to Chl b ratio compared to that in the wild type (Supplementary Table 1). One of the Δ ZEP/Δ CpFTSY mutants which have indel patterns at target loci ( Fig. 3a and Δ ZF6 in Supplementary Fig. 12) showed that Zea was accumulated similarly as in the Δ ZEP single mutant, while the total amount of Chl was significantly lower that the Δ ZEP single mutant and comparable to that in the Δ CpFTSY mutant (Fig. 3b). Despite the total absence of ZEP and CpFTSY proteins, confirmed by Western blot analysis, Δ ZEP/Δ CpFTSY lines could grow in the light, attaining golden rather than green coloration (Fig. 3c), caused by the lower Chl and greater Zea content.
To confirm that the phenotypes in the double mutant cells were not caused by off-target effects, we investigated whether the CpFTSY and ZEP-targeted RGENs induced off-target mutations. We identified potential off-target sites that differed from on-target sites by 4 nucleotides with no bulge and by 2 nucleotides with one DNA or RNA bulge using Cas-OFFinder 23 (www.rgenome.net/cas-offinder). No indels were found at these sites above sequencing error rates, in line with previous studies (Supplementary Fig. 13) 24 .
Photosynthetic productivity of two-gene knockout mutant. We finally compared the photosynthetic productivity of the Δ ZEP/Δ CpFTSY double mutant with that of the Δ ZEP single mutant. The Chl antenna size of the photosystems is well known to impact the photosynthetic rate of mutants and wild type 16 . The quantum yields of photosynthesis of Δ ZEP/Δ CpFTSY, Δ ZEP and wild type were essentially same, indicating that this parameter was not affected by single or double mutation (Fig. 3d). However, the maximum photosynthetic rate (P max ) on a Chl basis of the Δ ZEP/Δ CpFTSY mutant was about 54% greater than that of the Δ ZEP mutant at saturating irradiance. This difference is attributed to the truncated antenna size of the Δ ZEP/Δ CpFTSY mutant, indicating a higher productivity per Chl in high light (HL) conditions. Furthermore, growth of the Δ ZEP/Δ CpFTSY mutant line under HL conditions was dramatically better than that of the wild type and the Δ ZEP mutant (Fig. 3e). Therefore, the Δ ZEP/Δ CpFTSY double mutant showed greater photosynthetic activity and greater light use efficiency than the Δ ZEP mutant, suggesting that the double mutant could lead to greater biomass accumulation under HL growth conditions. Photosynthetic activities of the wild type, Δ ZEP, Δ CpFTSY and Δ ZEP/Δ CpFTSY mutants were summarized in Table 1.
In conclusion, we achieved DNA-free targeted gene editing in C. reinhardtii. This simple RGEN RNP method can be applied to other microalgae without a requirement of a time-consuming cloning step. Moreover, the resulting transformants would be except from GMO regulation, enabling applications of microalgae for pharmaceutical, nutraceutical, food and animal feed and in the medical treatment of specific diseases.

Methods
Cell cultivation. Chlamydomonas reinhardtii strains CC-4349 cw15 mt-and the mutant strains were cultivated and maintained as described previously 25 . Briefly, Cells were maintained at 25 °C under continuous white light at low light (50 μ mol photons m −2 s −1 ). Cells were cultivated photoheterotrophically in Tris-acetate phosphate (TAP) medium, or photoautotrophically in high-salt (HS) medium under continuous low and high irradiance (50 and 700 μ mol photons m −2 s −1 , respectively). Data in all experiments indicate mean the average and SE from at least three biological replicates.

RNA preparation.
Purified recombinant Cas9 protein was purchased from ToolGen, Inc. Guide RNA was transcribed in vitro using the MEGAshortscript T7 kit (Ambion) as previously described 9 . Transcribed RNA was purified by phenol:chloroform extraction, chloroform extraction, and ethanol precipitation. Purified RNA was quantified by spectrometry. Chlamydomonas transformation. To generate target-specific knockout mutants using RNP complex in Chlamydomonas, 50 × 10 4 cells were transformed with Cas9 protein (200 μ g) premixed with in vitro transcribed sgRNA (140 μ g). Cas9 protein in storage buffer (20 mM HEPES pH 7.5, 150 mM KCl, 1 mM DTT, and 10% glycerol) was mixed with sgRNA dissolved in nuclease-free water and incubated for 10 minutes at room temperature.
Cells were transformed with a Biorad Gene Pulser Xcell ™ Electroporation Systems, according to the recommended protocol from the GeneArt Chlamydomonas Engineering kit. After transformation, cells were incubated 12 hours and harvested for genomic DNA extraction, or immediately diluted the number of cells 2000 and plated on TAP medium containing 1.5% agar to obtain single colonies for further investigation (Supplementary Fig. 14).
Targeted deep sequencing. Genomic DNA segments that encompass the nuclease target sites were amplified using Phusion polymerase (New England Biolabs). Equal amounts of the PCR amplicons were subjected to paired-end read sequencing using Illumina MiSeq. Rare sequence reads that occur only once were excluded to remove errors associated with sequencing reaction and amplification. Indels located around the RGEN cleavage site (3 bp upstream of the PAM) were considered to be mutations induced by RGENs. The deep sequencing data are available at the NCBI Sequence Read Archive (www.ncbi.nlm.nih.gov/sra) under accession number PRJNA327012.
Examination of potential off-target sites. To examine whether there were nuclease-induced indels at hundreds of thousands of potential off-target sites in each genome sequence, we used Cas-OFFinder 23 to list potential off-target sites that differed from on-target sites by up to 4 nucleotides or that differed by up to 2 nucleotides with a DNA or a RNA bulge in length. As a result, 17 potential off-target sites were obtained and we carried out targeted deep sequencing in independent clones. Genotypic characterization. The genomic DNA was extracted from the cells harvested after transformation, or the individual colonized mutants selected from TAP agar plate, for targeted deep sequencing and Sanger sequencing, respectively. To isolate genomic DNA, cells were harvested, resuspended in a microprep buffer containing 2.5× extract buffer (0.35 M Sorbitol, 0.1 M Tris/HCl pH 7.5, and 5 mM EDTA), 2.5× nuclei lysis buffer (0.2 M Tris/HCl pH 7.5, 0.05 M EDTA, 2 M NaCl, and 2% (w/v) CTAB), and 1 × 5% N-Lauroylsarcosine, and incubated for 2 hr at 65 °C. The genomic DNA was extracted with Chloroform:Isoamylalcohol 24:1 and precipitated with Isopropanol. For Sanger sequencing, the target regions were PCR amplified with specific primers (Supplementary Table 2). The PCR products were verified by agarose gel electrophoresis, eluted from the gel and sequenced using the Sanger method (Macrogen, South Korea).

Mutant Selection.
The targeted knockout mutants were selected upon the basis of coloration of the cells and measurement of chlorophyll (Chl) fluorescence by Walz image-PAM system M-series Chl Fluorescence System equipped with a CCD camera as described 21 . Initially screened transformants were measured pigments to verify their changed pigment compositions.
Pigments and photosynthetic activity determination. Cells for pigments determination were grown under the low light conditions, 50 μ mol photons m −2 s −1 . The Chl content of the cells was spectrophotometrically determined in 80% (v/v) acetone extracts as described 26,27 . HPLC analysis was conducted with a Shimadzu Prominence HPLC model LC-20AD equipped with a Waters Spherisorb 5.0 μ m ODS1 4.6 × 250 mm cartridge column. The pigment was extracted in 90% (v/v) acetone and the supernatant of sample was subjected to HPLC analysis. The pigments were separated using a solvent mixture of 0.1 M Tris-HCl pH 8.0, acetonitrile, methanol, and ethylacetate. During the run, the solvent concentrations were 14% 0.1 M Tris-HCl, 84% acetonitrile, and 2% methanol from 0 to 15 minutes. From 15 to 19 minutes, the solvent mixture was consisted of 68% methanol and 32% acetonitrile. A post-run was performed for 6 minutes with the initial solvent mixture. The flow rate was constant at 1.2 mL per minute. Pigments were detected at 445 nm and 670 nm. Concentration of the individual pigment was determined from the HPLC profiles calibrated with standard pigments of Chl and carotenoids (14C Centralen; DHI, Hørsholm, Denmark). The photosynthetic activity was measured at 25 °C with a Hansatech Clark-type oxygen electrode illuminated with a halogen lamp. An aliquot of 1 mL cell suspension containing 2 μM Chl was transferred to the oxygen electrode chamber. To ensure that oxygen evolution was not limited by the carbon supply available to the cells, 50 μ L of 0.5 M NaHCO 3 , pH 7.4 was added to the suspension before the measurements. The oxygen evolution was measured at increasing light intensities, 0 to 1,200 μ mol photons m −2 s −1 , and each step was recorded for 2 minutes.

SDS-PAGE and Western-Blot Analysis.
For protein analysis, cells grown in TAP under 70 μ mol photons m −2 s −1 were harvested, resuspended in lysis buffer (20 mM HEPES-KOH pH 7.5, 5 mM MgCl 2 , 5 mM β -mercaptoethanol and 1 mM PMSF) and disrupted by sonication. Total protein samples loaded and separated on SDS-PAGE gel as described26. The SDS-PAGE gels were stained with 0.1% (w/v) Coomassie Brilliant Blue R for visualization, or blotted onto an ATTO P PVDF membrane via a semi dry transfer system. Membranes were probed with specific polyclonal antibodies raised against the Zeaxanthin epoxidase (ZEP), cpFTSY and β subunit of ATP synthase (ATPβ ) from C. reinhardtii. Signals were visualized using the Abfrontier west save up ECL Reagent and exposed to an X-ray film for signal detection.
Growth analysis. Chlamydomonas cells were photoautotrophically grown in 200 mL HS media in the bubble column photobioreactor (40 mm in diameter and 500 mm in height) illuminated with continuous high light (700 μ mol photons m −2 s −1 ) at 25 °C. Each culture was aerated with 5% CO 2 at a feed velocity of 40 mL/min. The initial cell concentration for each culture was 100 × 10 4 cells/mL.