Strain availability.

N. gaditana strain CCMP1894 was obtained from CCMP, now known as the Bigelow National Center for Marine Algae and Microbiota (NCMA). All strains described in this manuscript were derived from this parental background. The Ng-Cas9+ strain can be requested from Synthetic Genomics, Inc., by contacting the corresponding authors, or through the ATCC. The cell line was authenticated by immunoblotting for Cas9 and confirming GFP expression by flow cytometry; it was not tested for mycoplasma contamination.

Media formulations.

Standard medium (SM-NO 3 −) consisted of 35 g/l Instant Ocean, 10× F/2 trace metals and vitamins, and 0.361 mM NaH 2 PO 4 . The N source was 8.8 mM NaNO 3 for the semicontinuous assay and 15 mM NaNO 3 for the batch growth screen, both of which are described below. When cultures were supplemented with ammonium, 10 mM NH 4 Cl was used as the N source and was buffered with 15 mM HEPES, pH 8.0; this medium is referred to as SM-NH 4 +/NO 3 − in the main text.

Batch growth assessment.

Cultures (175 ml) were grown in 75 cm2 rectangular tissue culture flasks positioned with their narrowest 'width' dimension against an LED light source. Culture flasks were masked with opaque white plastic to provide a 21.1-cm2 rectangular opening allowing irradiance to reach the culture. Incident irradiance was programmed to mimic the light intensity of a 14-h light, 10-h dark diel period: a linear ramp up of irradiance from 0 to 1,200 μE over 4 h (increasing in 15-min intervals) was followed by 6 h at constant 1,200 μE and then a linear ramp down in irradiance from 1,200 to 0 μE over a 4-h period (Supplementary Fig. 7). The temperature of the cultures was regulated by a water bath set at 25 °C. Deionized H 2 O was added to the cultures daily to replace evaporative losses. Cultures were inoculated at an OD 730 nm of 0.5, and 5-ml samples were typically harvested every other day for analytical measurements. Sampling time was kept constant at 30 min before the end of the light phase of the diel cycle.

Semicontinuous productivity assay.

Rectangular tissue culture flasks (225 cm2) were inoculated to an OD 730 nm of 0.9 in a final volume of 550 ml. The inoculum (typically 200 ml) was previously scaled up in batch mode in the same conditions to allow the cells to acclimate to the assay environment. Culture flasks were masked with opaque white plastic to provide a 31.5-cm2 rectangular window allowing irradiance to reach the culture. The flasks were aligned with their width (narrowest dimension) against an LED light bank that was programmed with a light–dark cycle and a light profile that increased until 'solar noon' and then decreased until the end of the light period. The light profile was designed to mimic a spring day in southern California, with 14 h light and 10 h dark, with the light peaking at approximately 2,000 μE (Supplementary Fig. 7). The flasks included stir bars and had stoppers with inserted tubing connected with syringe filters for delivering CO 2 -enriched air (1% CO 2 , flow rate, 300 ml per min). The flasks were set in a water bath programmed to maintain a constant temperature of 25 °C on stir plates set to 575 r.p.m. during the assay period. Cultures were diluted daily at mid-day, when the light intensity was at its peak, by removing 30% of the volume (165 ml) and replacing it with the same volume of the assay medium plus an additional 10 ml of deionized water to compensate for evaporation. A 30% dilution rate was empirically determined as the most productive dilution rate for Nannochloropsis, because 50%, 30% and 15% daily dilutions resulted in average TOC productivities of 6.5, 9 and 8 g/m2/d, respectively. Semicontinuous assays were typically run for 7–14 d. Daily lipid (FAME) and biomass (TOC) productivities were calculated from cultures that had reached steady-state standing-crop TOC and FAME density. Volumetric FAME and TOC productivities (in mg l−1 d−1) were calculated by multiplying the volumetric FAME and TOC amounts by the 30% dilution rate. Aerial productivities (in g m−2 d−1) were calculated by dividing the total productivity of the culture by the size of the aperture through which irradiance was permitted:

RNA extraction.

Cell pellets were resuspended in 1.8 ml extraction solution (5 ml grinding buffer, 5 ml phenol, 1 ml 1-bromo-3-chloropropane and 20 μl mercaptoethanol, with grinding buffer comprising 9 ml of 1 M Tris, pH 8, 5 ml of 10% SDS, 0.6 ml of 7.5 M LiCl, and 450 μl 0.5 M EDTA in a final volume of 50 ml) and vortexed vigorously for 5 min at 4 °C in the presence of 200-μm zirconium beads. After centrifugation, 1 ml of 25:24:1 phenol extraction solution (25 ml phenol, pH 8.1, 24 ml 1-bromo-3-chloropropane and 1 ml isoamyl alcohol) was added to the aqueous phase in a separate tube. Tubes were shaken vigorously and centrifuged for 2 min at 21,000g. The extraction was repeated with 1 ml 1-bromo-3-chloropropane, and the resulting aqueous layer was treated with 0.356 volumes of 7.5 M LiCl to precipitate the RNA overnight at −20 °C. After LiCl precipitation, RNA pellets were resuspended in 50 μl H 2 O, and RNA quality was assessed by on-chip gel electrophoresis with an Agilent 2100 Bioanalyzer, according to the manufacturer's instructions.

RNA-seq.

Next-generation sequencing libraries were prepared from total RNA with a TruSeq Stranded mRNA Sample Prep Kit (Illumina), according to the manufacturer's instructions. For the nitrogen-replete and nitrogen-depleted comparison, TruSeq libraries were sequenced on an Illumina HiSeq platform (Ambry Genetics) to generate 100-bp paired-end reads (described in ref. 35). For the comparison of the ZnCys-RNAi-7 to WT, libraries were sequenced in house on an Illumina NextSeq platform to generate 75-bp paired-end reads. Trimmomatic software (http://www.usadellab.org/cms/?page=trimmomatic) was used to filter and trim reads on the basis of base-quality scores. Mapping of the trimmed-read libraries and expression estimation relative to a reference gene set were performed with RSEM36. The reference gene set comprised predicted coding sequences based on the N. gaditana B-31 genome annotation (http://www.nannochloropsis.org/page/ftp/). Differential expression analysis was performed with the R package edgeR37. Gene ontology annotation for the N. gaditana coding sequences were obtained from the UniProt-GOA database (ftp://ftp.ebi.ac.uk/pub/databases/GO/goa/proteomes/357410.N_gaditana.goa). Gene set enrichment testing was performed with goseq in R to identify enriched GO categories among differentially expressed genes (FDR-adjusted P value <0.05) for a given comparison and fold-change direction38.

Quantitative real-time PCR.

Total RNA was reverse transcribed with an iScript Reverse Transcription Supermix kit (Bio-Rad). The enzyme mix Ssofast EvaGreen Supermix (Bio-Rad) was used for amplification of genes of interest. Primer and cDNA concentrations were as recommended by the manufacturer, and reactions were performed on a C1000 Thermal Cycler coupled with a CFX Real-time System (Bio-Rad). The primer sequences for ZnCys (Naga_100104g18) were F, 5′-atacaggaagcgtggttacag-3′ and R, 5′-gaagtattaagggactggccg-3′. qRT–PCR primers were evaluated for efficiency, and the 2−ΔΔCt method was used to estimate gene expression normalized to that of a control gene (Naga_100004g25, with primer sequences F, 5′-ctctcctattgctttccctcg-3′ and R, 5′-ctaccaacacctctacacttcc-3′) that was empirically determined to have a low coefficient of variation across different conditions.

Generation of the Cas9 editor line.

The high-efficiency N. gaditana Cas9 editor line (Ng-Cas9+) was generated by random integration of a ZraI restriction enzyme–linearized vector containing expression cassettes for the selectable marker blasticidin deaminase (BSD), Cas9 and TurboGFP (Evrogen) (Supplementary Fig. 8). The BSD and Cas9 genes were optimized for N. gaditana codon usage, whereas TurboGFP was directly amplified from pTurboGFP-C purchased from Evrogen. The Cas9 coding sequence contained an N-terminal nuclear localization signal (NLS) and a FLAG epitope followed by a linker region, as described in ref. 39. The expression of all three genes was driven by endogenous promoters and terminators; the source genes are described in Supplementary Table 8. The Cas9 expression construct was assembled with a Gibson Assembly HiFi 1 Step Kit (Synthetic Genomics) into a minimal pUC-vector backbone; the confirmed DNA sequence of this plasmid is shown in GenBank format in Supplementary Note 1.

Transformation of the ZraI-linearized Cas9 expression construct by electroporation was conducted according to ref. 11, except that 1 × 109 cells were transformed in a 0.2-cm cuvette with a field strength of 7,000 V/cm, delivered with a Gene Pulser II (Bio-Rad). After overnight recovery in liquid SM-NO 3 − medium, cells were plated on SM-NO 3 − agar containing 100 mg l−1 blasticidin (Invivogen). Putative transformant colonies were replated on the same selective-medium plates, and after sufficient growth, an inoculation loop full of cells was resuspended in liquid medium and analyzed on an Accuri C6 cytometer (BD Biosciences) for GFP fluorescence (Supplementary Fig. 9a). Stable transgenic Cas9 expression was confirmed in the Ng-Cas9+ line by western blotting with an anti-FLAG antibody from Sigma-Aldrich (cat. no. F1804; validation on manufacturer's website; Supplementary Fig. 9b). The Ng-Cas9+ line was assessed for potential differences in FAME or TOC productivity compared with those of WT under semicontinuous dilution growth mode and was found to be equivalent to WT in these parameters (Fig. 3a and Supplementary Fig. 4).

Generation of targeted insertional mutants in the Ng-Cas9+ editor line.

Synthetic gRNAs were designed according Cho et al.40 and Mali et al.41. Target regions were selected by searching for the presence of NGG (the protospacer-adjacent motif) in the gene of interest. The spacer sequence comprised 17–20 nt directly upstream of the protospacer-adjacent motif, as shown in Supplementary Table 9. gRNAs were synthesized in vitro according to ref. 40 and coelectroporated with a PCR-amplified expression cassette (pHygR) containing a codon-optimized hygromycin-resistance gene (HygR) driven by the endogenous EIF3_P promoter and FRD_T bidirectional terminator in inverted orientation (descriptions of DNA-element sources in Supplementary Fig. 9b and Supplementary Table 8). The confirmed DNA sequence of pHygR is shown in GenBank format in Supplementary Note 2. Approximately 1 μg each of gRNA and pHygR was added to the cuvette, and electroporation was conducted as described above. Selection of HygR transformants was as described above except that 500 mg l−1 hygromycin was added to agar plates instead of blasticidin. Targeted insertion of the pHygR fragment via NHEJ-mediated repair of the double-stranded DNA break catalyzed by Cas9 in loci targeted by gRNAs was assessed by colony PCR with primers flanking the gRNA target site by ∼200 bp on either side. In this manner, PCR amplification of WT loci would result in ∼400-bp products, whereas pHygR-targeted insertional mutants would result in ∼2,800-bp products—thus accounting for the insertion of the 2,400-bp pHygR fragment. All insertional-mutant lines presented in this study were confirmed to have a single pHygR insert at the intended locus and to lack the WT PCR product. PCR products were sequence confirmed for all mutants, thus allowing the determination of insert orientation and potential loss of chromosomal and/or insert DNA, or the gain of small insertions generated during the NHEJ-mediated dsDNA-break-repair process (example colony PCR results in Fig. 1d). In general, loss-of-function KO mutants were generated by selecting a gRNA-target locus in the first half of an exonic coding sequence, whereas Cas9-derived insertional-attenuation mutants were created by targeting insertions to the promoter or untranslated regions of a given gene, as described in the main text (gRNA sequences in Supplementary Table 9).

Generation of ZnCys-RNAi lines.

The RNAi plasmid designed to silence the ZnCys locus was constructed by using a minimal pUC vector as the backbone. The confirmed DNA sequence of the final RNAi plasmid is provided in GenBank format in Supplementary Note 3. Briefly, the sense and loop structures of the ZnCys hairpin were amplified from N. gaditana cDNA with primers F, 5′-gacagagacacacagggatcgtttaaacgatcagccacgacggctctcg-3′ and R, 5′-cactatgcaccttcccccttgactcaccggttcgtggcactgctgttg-3′. These primers contained ∼20-bp overlaps with the minimal pUC vector backbone to allow for cloning with a Gibson AssemblyHiFi 1 Step Kit (SGI-DNA). The antisense fragment was amplified with 5′-cacaacagcagtgccacgaaccgccccgactgcttttactgc-3′ and 5′-ctatgcaccttcccccttgactcgtttaaacgatcagccacgacggctc-3′ primers and cloned into the vector containing the sense and loop fragments by digesting the vector with AgeI and by using the Gibson Assembly method (Synthetic Genomics). The final vector was digested with ScaI, and 1 μg of the pertinent linear fragment was transformed into WT N. gaditana as described above. 32 hygromycin-resistant transformant colonies were assessed for GFP expression as described above for the Cas9 editor line NgCas9+. Eight GFP-expressing constructs were screened for increased FAME/TOC, and the line ZnCys-RNAi-7 was selected because it displayed maximal FAME/TOC with little apparent decrease in TOC productivity.

Analytical methods (FAME, TOC and LC/GC–MS).

FAME analysis was performed on 2-ml samples that were dried with a GeneVac HT-4X. To each dried pellet, the following were added: 500 μl of 500 mM KOH in methanol, 200 μl of tetrahydrofuran containing 0.05% butylated hydroxyltoluene, 40 μl of a 2 mg/ml C11:0 free fatty acid/C13:0 triglyceride/C23:0 fatty acid methyl ester internal standard mix and 500 μl of glass beads (diameter 425–600 μm). The vials were capped with open-top PTFE-septa-lined caps and placed in an SPEX GenoGrinder at 1,650 r.p.m. for 7.5 min. The samples were then heated at 80 °C for 5 min and allowed to cool. For derivatization, 500 μl of 10% boron trifluoride in methanol was added to the samples, which were then heated at 80 °C for 30 min. The tubes were allowed to cool, and 2 ml of heptane and 500 μl of 5 M NaCl were then added. The samples were vortexed for 5 min at 2,000 r.p.m. and finally centrifuged for 3 min at 1,000 r.p.m. FAME were quantified with gas chromatography with a flame ionization detector (GC–FID) on an Agilent 7980A instrument equipped with a DB-FFAP column (Agilent P/N 127-3212) with the C23:0 externally spiked FAME as a standard, by using methods similar to the American Oil Chemists' Society Methods Ce 1b-89 and Ce 1-62 (http://www.aocs.org/Methods/).

Total organic carbon (TOC) was determined by dilution of 2 ml of cell culture to a total volume of 20 ml with DI water. For each measurement, three injections into a Shimadzu TOC-Vcsj Analyzer were performed for determination of total carbon (TC) and total inorganic carbon (TIC). The combustion furnace was set to 720 °C, and TOC was determined by subtraction of TIC from TC. The four-point calibration range was from 2 p.p.m. to 200 p.p.m., corresponding to 20–2,000 p.p.m. for nondiluted cultures with a correlation coefficient of r2 >0.999.

For LC–MS analysis of lipids, 2-ml samples of each culture were spun down at maximum speed for 5 min, the supernatants were removed, and total lipids were extracted from the pellets via a modified Bligh–Dyer extraction42. The bottom, lipid-containing layer was concentrated to dryness with an EZ-2 Genevac. The residue was dissolved in 1 ml of 1:1 chloroform/methanol containing 10 μg/ml BHT and 10 μg/ml TAG (17:1/17:1/17:1) as an internal standard. LC–MS was performed on an Agilent 1200 LC interfaced with an Agilent 6538 Q-ToF with a dual ESI source. 10 μl of the extract was injected onto a Halo C8 column, 2.7 μm, 4.6 × 150 mm (MacMod Analytics) and eluted in a gradient of mobile phase A (375:375:250:1 methanol/isopropanol/water/acetic acid + 10 mM ammonium acetate) and mobile phase B (500:375:125 acetonitrile/isopropanol/tetrahydrofuran + 10 mM ammonium acetate). The flow rate was 0.6 ml/min, and the following gradient elution was used: 0 min, 15% B; 1 min, 15% B; 15 min, 56% B; 19 min, 85% B; and 26 min, 85% B. The source parameters were as follows: gas temperature, 300 °C; drying gas, 10 l/min, nebulizer, 35 psig; capillary voltage, 4,000 V. The ToF scan rate was 1 spectrum/s. Individual TAGs were identified by accurate mass and quantified by summing the ion intensities of the individual TAGs and using a relative response factor of 1 compared with the ion intensity of the 10 μg/ml TAG(17:1/17:1/17:1) internal standard.

Total carbohydrate analysis was conducted on an ∼0.7 mg TOC equivalent of cell culture purified by three rounds of centrifugation and washing with PBS, and concentrated to 0.5 ml in PBS. Acid hydrolysis was used to convert carbohydrates into their constituent monomers by the addition of 0.5 ml deionized H 2 O and 1 ml 6 N HCl and [U-13C]glucose and [U-13C]galactose as internal standards at a final concentration of 50 μg/ml each. Samples were heated at 105 °C for 1 h in glass vials with PTFE-lined caps. 100-μl aliquots of the room-temperature-cooled samples that had been centrifuged at 3,000g for 1 min were dried in an EZ-2 Genevac, derivatized with MSTFA/TMCS and analyzed by GC–MS according to ref. 43. Internal 13C-labeled standards were used to quantify the concentration of the major carbohydrate monomers, glucose and galactose, and to estimate the concentrations of less abundant sugars (arabinose, rhamnose, xylose and mannose). These concentrations were summed to yield a total saccharide concentration (in micrograms per milliliter) which was converted to the carbon content of total carbohydrates by a multiplication factor of 0.45 (i.e., ∼45% of the carbohydrate mass represented by carbon). This value was divided by the amount of TOC detected in an identical aliquot of concentrated cell culture to estimate the percentage of carbon allocated to carbohydrate, as shown in Figure 3b.

Analysis of total amino acids was conducted by derivatization of whole amino acid hydrolysate to propoxycarbonyl propyl esters through a modified method, according to the manufacturer's instructions for the EZ:faast kit from Phenomenex. Briefly, to 0.5 ml concentrated cells (as described for carbohydrate analysis above), 800 μl of 6 N HCl containing 200 μl/ml thioglycolic acid, 10 μl of β-mecraptoethanol and 200 μl of 2 mM norvaline (internal standard) was added, and the vortexed sample was incubated at 110 °C for 24 h. Samples cooled to room temperature were centrifuged at 1,500g for 1 min, and a 50-μl aliquot was transferred to a fresh 2 ml GC vial. Aliquots were derivatized and analyzed by GC–MS according to the EZ:faast manual and ref. 44. This method allowed for the quantification of alanine, glycine, valine, leucine, isoleucine, proline, aspartate and asparagine, methionine, glutamate and glutamine, phenylalanine, lysine, tyrosine and cysteine. Tryptophan, threonine, serine, arginine and histidine were excluded. Volumetric concentrations of each detected hydrolyzed amino acid were converted to the carbon content present in that amount. These values were summed and normalized to TOC, as described for total carbohydrates above, thus yielding an estimate of carbon allocated to protein. However, when referring to FAME/TOC (as in Fig. 1e), values in grams FAME per gram TOC are presented.

Light-saturating photosynthesis capacity (P max ) and chlorophyll determination.

P max (based on 14C incorporation) was measured in samples collected from cultures in semicontinuous growth mode 30 min before daily dilutions. Chlorophyll extractions were performed as previously described45. 0.5-ml aliquots of culture were placed into 2.0-ml screw-top tubes and centrifuged at 21,000g for 10 min. The supernatant was discarded, and 1.0 ml of 1:1 (vol/vol) dimethyl sulfoxide and acetone was added to each tube. Samples were agitated for 3 min in a prechilled block in a Mini-Beadbeater-96 (BioSpec Products) and then centrifuged for 10 min at 21,000g. Extracts were measured in disposable UV–vis cuvettes (BrandTech, cat. no. 759165) on a SpectraMax M2e (Molecular Devices) at 663 nm and 720 nm and quantified with the following equation: chlorophyll a (in μg ml−1) = (A 663 − A 720 ) × 20.15 (ref. 45). Samples were normalized to 5 μg ml−1 chlorophyll a by centrifugation of cultures at 2,100g for 5 min, discarding of the supernatant and resuspension of the cells in the appropriate volume of standard medium (SM-NO 3 −) supplemented with 0.5 g l−1 sodium bicarbonate. Cultures were adapted to low light (∼10 μmol photons m−2 s−1) for approximately 1 h before P max measurements, and 2-ml aliquots were placed into 7-ml glass scintillation vials (VWR, cat. no. 66022). A 20.4 μCi ml−1 working stock of 14C-labeled sodium bicarbonate was prepared from a 1 mCi ml−1 stock (PerkinElmer, cat. no. NEC086H005MC) diluted in freshly prepared 3.57 mM sodium bicarbonate (Sigma-Aldrich, cat. no. S5761). 70 μl of the 14C working stock was added to the 2.0 ml of diluted culture and mixed well with a multichannel pipette. 70 μl of each sample was added to a new set of 7-ml scintillation vials containing 100 μl of phenethylamine (Sigma-Aldrich, cat. no. 128945), filled with 4 ml of Ultima Gold AB (PerkinElmer, cat. no. 6013309) and capped immediately. An additional 1-ml aliquot of each sample was placed into another set of vials and acidified immediately with 2 N HCl. The remaining 1 ml of each samples was exposed to 2,500 μmol photons m−2 s−1 of white light for 10 min at 25 C. Samples were then immediately acidified with 2 N HCl and allowed to off-gas overnight. The following day, 4 ml of Ultima Gold AB was added to each acidified sample, and all samples were measured with an LS6500 scintillation counter (Beckman Coulter). Quantification was performed with equations from Littler and Arnold46.

Western blotting.

Cell pellets corresponding to 1.5 × 109 cells were resuspended in 500 μl buffer consisting of 125 mM Tris, pH 8.8, 10% glycerol and 2% SDS. Zirconium beads were added to the cell slurry, and samples were shaken twice for 3 min on a Mini Beadbeater (BioSpec Products). Lysates were boiled, incubated at 90 °C for 10 min and centrifuged, and the supernatant was diluted with 4× Laemmli sample buffer. Total protein extracted was determined with an RC DC protein assay kit (Bio-Rad). A total of 40 μg of protein was loaded per well onto a Tris-glycine (4–20%) denaturing gel. FAS proteins and ACCase were detected with antibodies raised by ProSci (peptide sequences in Supplementary Table 7). The primary antibody to NR was obtained from Agrisera (cat. no. AS08310; validation on manufacturer's website). iBind western blotting devices (Thermo Fisher Scientific) were used to incubate the blots with primary and secondary antibodies (anti-mouse AP, Novex, cat. no. A16081, 1:500 dilution). Immunosignals were detected with a colorimetric BCIP/NBT kit from Invitrogen (cat. no. 002209).

Transmission electron microscopy.

Samples were pelleted and resuspended in fixing solution consisting of 2.5% glutaraldehyde, 2.5% paraformaldehyde and 0.1 M sodium cacodylate, pH 7.4. Transmission electron micrographs were captured by the Michigan State University Center For Advanced Microscopy with a high-resolution JEOL100 CXII microscope at an accelerating voltage of 100 kV.

Statistical analysis.

No statistical methods were applied to predetermine sample size. Sample sizes were chosen on the basis of industry standards and experience. Differences between the performances of engineered strains and controls were analyzed with box-and-whisker plots showing median productivity, maximum and minimum data points, and first and third quartiles, as indicated in the corresponding figure caption. Differential expression analysis for RNA-seq experiments was performed with the R package edgeR37. For TF identification, FDR <0.01 was considered significant. Gene set enrichment testing was performed with goseq in R, and an FDR-adjusted P value <0.05 was considered significant.

Data availability.

Sequence reads for all RNA-seq experiments described in this work have been deposited in NCBI's sequence read archive under accession number SRP096211. All other data supporting the findings of this study are available from the corresponding authors upon reasonable request.