C. reinhardtii CC425 and CC124 were obtained from the College of Life Sciences and Oceanography, Shenzhen University, and grown in tris-acetate-phosphate (TAP) medium at 24°C under continuous illumination (about 120 μmol photons m^–2 s^–1) for 3 days to reach a logarithmic phase as of seeds. For nutrition starvation, cells in the middle log stage were inoculated at the ratio of 10% (v/v) into Sueoka’s High-Salt Medium (HSM) or without nitrogen (-N), sulfur (-S), or iron (-Fe) for 8 days. Cultures were collected by centrifugation and used for total RNA isolation. For cell growth, neutral lipid contents, and fatty acid (FA) assays of CrPrp19 RNAi or overexpression lines, cells were cultured in HSM medium with an initial density of about 5 × 10⁵ cells⋅ml^–1. For synchronized culture, cells were grown with aeration of 5% (v/v) CO₂ enriched air on a 14/10 h light/dark cycle for 4 days (Dillard et al., 2011), then cells were collected and observed using an optical microscope.

CrPrp19 complementary DNA (cDNA) sequence was achieved from JGI Chlamydomonas reinhardtii v5.5 (ID: Cre02.g073650) and isolated by RT-PCR. The homologous sequences of CrPrp19 were obtained using BLASTp in the NCBI database.¹ Multiple sequence alignments were performed using the ClustalW program.² Neighbor-joining (NJ) phylogenetic tree was constructed by MEGA7.0 based on the JTT model, using a bootstrap value of 1,000 as default parameters (Kumar et al., 2016). Compute pI/Mw tool³ was used to calculate molecular weight and pI value of CrPrp19. Motifs were annotated by the conserved domain database using the SMART program.⁴ Subcellular localization was predicted with Cell-PLoc 2.0.⁵ Transmembrane regions and orientation were predicted by TMHMM program.⁶ Comparative 3D protein prediction was obtained using Phyre2.⁷

Total RNA was extracted using Trizol reagent (Invitrogen, Carlsbad, CA, United States) following the manufacturer’s instructions. About 0.1 mg of total RNA was reversed to the first cDNA using the PrimerScript RT Reagent Kit with gDNA Eraser (Takara, Beautilion, Otsu, Japan). Primers used for qRT-PCR are listed in Supplementary Table 1. The quantified expression level of the target gene was normalized against the housekeeping gene, 18S rRNA (GenBank accession no. MF101220.1). ABI-7500 Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, United States) was applied for qRT-PCR using the SYBR Green master mix (Takara). The conditions for quantitative analysis were as follows: 94°C for 2 min; 35 cycles of 94°C for 10 s, 62°C for 15 s, and 72°C for 20 s; and a final extension at 72°C for 10 min. Gene expression was normalized with the 2^–ΔΔCt method setting HSM-cultured groups as controls.

Plasmid pMaa7 IR/ CrPrp19 IR, which was used for knockdown of CrPrp19 in C. reinhardtii, was constructed as follows. The primers that were used in this experiment are listed in Supplementary Table 1, and a fragment 278 bp from 5’-UTR to non-conservative domain-encoding region was amplified from cDNA by PCR and cloned into pMD18-T vector. After being digested by KpnI/BamHI and HindIII/SalI, the fragment was subsequently inserted into a modified vector pMD18T-18S as previously described by Deng et al. (2011). Finally, CrPrp19 IR was double digested with KpnI and HindIII and inserted to pMaa7 IR with EcoRI blunt-end, while overexpression plasmid pCAMBIA1302-CrPrp19 was constructed by restriction digestion and ligation method using other primer pairs, which are also listed in Supplementary Table 1. For RNAi experiment, the plasmid was introduced into C. reinhardtii by the glass beads method (Kindle, 1990). Transformants were screened in TAP medium with 1.5 mM L-tryptophan, 5 mg/L paromomycin, and 5 μM 5-FI (Sigma-Aldrich, San Francisco, CA, United States). To obtain overexpression lines, the recombinant plasmid was transformed into cells via Agrobacterium meditation method and putative transformants were screened using a TAP medium, which was supplemented with 50 mg/L kanamycin. Finally, transgenic algae lines were confirmed by the transcription level of CrPrp19 using qRT-PCR.

Growth of the C. reinhardtii CrPrp19 RNAi, Maa7, overexpression lines, and wild-type CC124 were determined by a Coulter Multisizer 4 (Beckman Coulter, Fullerton, CA, United States) daily. Cell sizes were measured using a microscope after synchronization culture referred elsewhere (Dillard et al., 2011). Moreover, cell responses to stresses such as -N, -S, and high salinity (8 g/L sodium acetate) were observed using the plated method.

Neutral lipid (NL) was measured based on fluorescence intensities of 7-day-old cells staining with Nile Red (Sigma-Aldrich) according to a modified method reported by Chen et al. (2009). Liquid cell cultures were mixed with Nile Red and DMSO with a final concentration of 5 μg/ml and 20% (v/v), respectively. After being incubated in darkness for 15 min, mixtures were directly measured by a Glomax-Multi Detection System (Promega, Madison, WI, United States), with excitation and emission wavelengths of 530 and 575 nm, respectively. The standard curve between the concentration of neutral lipid (μg/ml) and the Nile Red fluorescence level was made based on using Triolein (Sigma-Aldrich) as a reference standard. NL contents (μg/10⁶ cells) were calculated using the following formula: NL(g/106cells) = cell0004×FI(530/575)−0.0038]×50/celldensity. For lipid drop observation, cells were cultured for 7 days and stained with Nile Red dye, then observed with a Nikon Eclipse 80i fluorescence microscope (Nikon, Tokyo, Japan) using an excitation and emission wavelength of 480 and 560 nm, respectively.

For fatty acid methyl ester (FAME) transformation, 5–10 mg weighed freeze-dried cell powder was used. Cells were suspended in 1 ml NaOH–CH₃OH solution (2 mol/L) and incubated at 75°C for 30 min with shaking. Methyl non-adecanoate (C15:0, 200 μg/ml; Sigma-Aldrich) was applied as the internal standard. After cool-down, 1 ml of HCl–CH₃OH (4 mol/L) was added, and pH was adjusted to below 2.0 using HCl, and the mixture was then incubated at 75°C for 30 min with shaking. The methyl esters of FAs were extracted with 2 ml hexane, then dried by nitrogen blowing, and dissolved in 10 μl of CH₂Cl₂ for GC/MASS analysis. FAMEs were examined by Thermo Trace GC Ultra gas chromatograph coupled to Thermo Polaris Q mass spectrometry (Agilent, Santa Clara, CA, United States) equipped with an HP-5MS column (30 mm × 0.25 mm, film thickness 0.25 μm) as previously described (Wang et al., 2018).

A yeast two-hybrid assay was performed to find the binding proteins of CrPrp19 by using Matchmaker GAL4 Two-Hybrid Systems (Takara), following the manufacturer’s instructions. To construct bait vector pGBKT7-CrPrp19, CDS of CrPrp19 was amplified using primers, which are listed in Supplementary Table 1 with the BamHI and PstI flanking restriction sites, then cloned through double digests and ligation. A total of 5 mg of C. reinhardtii total RNA was used to create the cDNA library. Both bait and cDNA library vectors were transformed together into yeast strain Y2H Gold and screened on quadruple dropout QDO/X/A medium (SD/-Ade/-His/-Leu/-Trp supplemented with X-a-Gal and Aureobasidin A). The plasmids of potential yeast positives were extracted and amplified in E. coli. All plasmids from E. coli were sequenced and blasted.

For CrPrp19 localization assay, a GFP- CrPrp19 fusion protein expression vector was constructed and transformed into C. reinhardtii CC124 using Agrobacterium-mediated genetic transformation method. The coding region of CrPrp19 cDNA without stop codon was amplified by using primer pairs, which are listed in Supplementary Table 1, then it was inserted into the NcoI and SpeI sites at the initiator codon downstream of the cauliflower mosaic virus (CaMV) 35S promoter in the plant cell expression vector pCAMBIA1302. The control plasmid, pCAMBIA1302, and plasmids containing the fused gene construct were transformed to Agrobacterium LBA4404, which were then co-incubated with cultured CC124 cells. After dark-repairing culture for 24 h at 24°C, cells were plated on TAP mediums with 50 mg/L kanamycin and 15 mg/L rifampin. The clones were imaged with the Nikon Eclipse 80i fluorescence microscope with an excitation wavelength of 488 nm and an emission wavelength of 530 nm.

Statistical analysis was performed using SPSS 11.5 program, and differences were analyzed with one-way ANOVA followed by Duncan’s multiple comparison test. Data are shown as mean ± SD from three independent experiments unless specified otherwise.

The cDNA of CrPrp19 comprises an ORF of 1,512 bp encoding a Prp19-like protein of 503 aa with a calculated molecular weight of 53.95 kDa and an isoelectric point of 6.61. CrPrp19 protein contains a conserved U-box at the N terminal, a coiled coil motif, and a WDR domain with seven WD repeats at C terminal (Figure 1A). None of the transmembrane helices was found in CrPrp19 according to TMHMM program. A total of 37 CrPrp19 homologous protein sequences were retrieved by BLASTp from the NCBI database. After multiple alignments, a Neighbor-Joining phylogenetic tree of CrPrp19 was constructed with bootstrap values of 1,000 replicates (Figure 1B). The phylogenetic analysis showed that the overall amino acid residues of Prp19 proteins had a relatively low conservation from yeast to human. Prp19 proteins from eukaryotes were divided into four major clades, microalgae (clade I), higher plants (clade II), animals (clade III), and fungi (clade IV). CrPrp19 was mostly closely related to GpPrp19 from Gonium pectoral, which was grouped with microalgae Prp19 proteins. Rather, Prp19 proteins from yeast formed a separate branch, showing a distinct generation relationship (Figure 1B). The three-dimensional protein prediction by Phyre 2 (Kelley et al., 2015) showed that CrPrp19 was structurally homologous with pre-mRNA-processing factor 19 of the yeast spliceosome at 3.6 Å resolution (PDB ID:c3jb9U_1), with a confidence of 100% (Figure 1C).

Nutrition deprivation has been generally considered as an effective method to induce lipid biosynthesis in microalgae. To evaluate the expression profile of CrPrp19 gene under nutrition deprivation conditions, C. reinhardtii cells cultured under nitrogen (N), sulfur (S), and iron (Fe) deficiencies stress for 8 days were collected for RNA isolation, and the untreated cells (HSM) were used as control. Gene expression of CrPrp19 fold changes at the starvations were normalized with controls. As shown in Figure 2, mRNA expressions of CrPrp19 were detected in all cells under treatments. However, compared with the controls, a significant decrease in CrPrp19 gene expression was found under N and S deficiency stresses. More specifically, the transcript level of CrPrp19 was remarkably decreased under N deficiency since 2 days and reached its lowest point at 4 days, which was approximately 30.9% lower than that of control. Meanwhile, the expression level of CrPrp19 under S deprivation showed a similar expression pattern, which was dramatically reduced by S deficiency after 2 days, and the inhibition effect was on until 8 days, with a minimum transcript, 9.7% of control. Unlike N and S deprivation, the expression of CrPrp19 was induced by deprivation of Fe at 2 days, but was markedly repressed after 4 and 6 days, which decreased by 47.1 and 31.3%, respectively (Figure 2). Taken together, the transcript of CrPrp19 is deduced by nutrition deprivation, implying that it may be a regulator in response and adaptation to these nutrition stresses.

To investigate the roles of CrPrp19 gene in C. reinhardtii cells, RNA interference (RNAi) silencing and gene overexpression were carried out, respectively. A 278-bp fragment of CrPrp19 cDNA was subcloned and inserted to pMaa7 IR vector, generating the CrPrp19 gene RNAi construct pMaa7 IR/CrPrp19 IR (Figure 3A). The coding region of CrPrp19 was also cloned by RT-PCR and used to construct the gene overexpression binary vector pCAMBIA1302/CrPrp19 (Figure 3B). After transforming pMaa7 IR/CrPrp19 IR and pCAMBIA1302/CrPrp19 constructs into C. reinhardtii CC425 and CC124, respectively, antibiotic resistance transformants were generated. pMaa7 IR vector was also transformed to C. reinhardtii CC425 as controls for RNAi experiment, and named as Maa7 lines. After twice plating on selected plates, the survival transformants were randomly chosen for CrPrp19 gene expression analysis by qRT-PCR. Two RNAi and two overexpression lines with the biggest differences in gene expression level compared with controls were selected, and named as CrPrp19i-2, CrPrp19i-42 and CrPrp19o-18, CrPrp19o-31, respectively. As shown in Figures 3C,D, CrPrp19 mRNA abundance was decreased by 98.64 and 98.45% in RNAi lines, CrPrp19i-2 and CrPrp19i-42, respectively, while it increased by 44.67 and 74.42 times in overexpression lines, CrPrp19o-18 and CrPrp19o-31, respectively, compared with controls. The results suggested the high effectiveness of CrPrp19 silencing and overexpression was achieved.

To determine the function of CrPrp19 in cell growth and neutral lipid (NL) metabolism, CrPrp19 silencing and overexpression lines were cultured in HSM medium for 7 days with the same initial cell density, and then the cell densities and NL contents were examined. The time-course profiles of cell growth and NL accumulation of different transgenic lines are shown in Figure 4. Compared with control Maa7-24 line, the cell densities of CrPrp19 RNA interference lines were decreased significantly in Maa7-24 line as compared with control after cultured for 4 days (p < 0.01). However, cell growth had no discernible difference between CrPrp19 overexpression line and wild-type strain, p > 0.05 (Figure 4A). The growth curves also showed that CrPrp19 RNAi lines had a maximum cell concentration at day 5 with the highest values of 6.8 × 10⁷ and 7.0 × 10⁷ cells/ml. They were 21.5 and 19.2% less than that of the control group, respectively. Figure 4B shows that the contents of NL (μg/10⁶ cells) of two CrPrp19 RNAi lines were more than that of controls since cultured for 4 days. After being cultured for 6 days, NL contents of CrPrp19 RNAi lines became stable, which had a similar value of about 0.8 μg/10⁶ cells, which were 1.46 and 1.33 times more than that of the control line at day 6 and day 7. In contrast, overexpressed CrPrp19 gene seems unregulated on NL accumulation in C. reinhardtii. CrPrp19o-18 line had less NL at day 2 and day 3 than CC124 strain, while lipid contents of CrPrp19o-31 lines were increased by 119% than that of control at 6 days. These results suggested that the silencing of CrPrp19 gene could increase lipid accumulation, but this effect was at the price of less cell densities.

CrPrp19 RNAi silencing and overexpressed lines and their respective controls were aerated cultured for 4 days, and the result showed that CrPrp19 RNAi silencing seemed more yellow than others (Figure 4C). Because of the growth difference between CrPrp19 RNAi and Maa7 line, a further comparison and analysis between the growth of CrPrp19 RNAi and Maa7 line under stresses conditions were performed. Here, a high concentration of sodium acetate (NaAc, 8 g/L) was also applied. Acetate is the carbon source usually used in heterotrophic cultivation to promote biomass of C. reinhardtii, but high concentration of NaAc inhibits C. reinhardtii metabolism. Appropriate amount of NaAc could also increase the yield of intracellular FA in C. reinhardtii mutant (Ramanan et al., 2013). The results showed that CrPrp19 RNAi and Maa7 line showed no obvious difference under nitrogen or sulfur deficiency, but much better tolerability to high concentration of NaAc than control (Figure 4D). The growth abilities of CrPrp19 RNAi lines under high concentration of NaAc could be applied to produce lipids with NaAc addition.

Moreover, to reveal the changes of lipid synthesis–related genes in CrPrp19 RNAi and overexpressed lines, qRT-PCR was applied. CrDGAT2-3 was induced to be expressed in CrPrp19 RNAi line, while the expression of CrPDAT increased in CrPrp19 overexpressed line (Figure 4E). The transcript of CrPEPCK gene was stable in both CrPrp19 RNAi and overexpressed lines, but expressions of other chosen lipid biosynthesis–related genes decreased. In summary, when CrPrp19 was silenced or overexpressed, the majority of transcriptions of genes participating in lipid synthesis were changed, suggesting that CrPrp19 could regulate lipid synthesis through control gene transcription.

The accumulation of lipid bodies in CrPrp19 RNAi and overexpressed transformants was examined by fluorescence microscopy (Figure 5). After Nile Red staining, numerous fluorescent lipid droplets (LDs) were observed in CrPrp19 RNAi cells, but none or only a few were present in control cells and overexpressed lines (Figure 5). CrPrp19 RNAi cells were almost twice bigger than the control Maa7-24 line, while overexpression of CrPrp19 did not change cell morphology. To verify this phenomenon, all microalgae strains were synchronized with 5% CO₂ aerated cultured under long-day station. After synchronization culture, the CrPrp19 RNAi cell lines also seemed larger than others (Figure 5), indicating that CrPrp19 plays a critical role in cell size control.

In general, the composition of FAs determined the quality of natural edible oils and biofuels. Therefore, we analyzed the FA composition of CrPrp19 RNAi and overexpressed transformants. After transesterification, FAMEs were examined using GC-MS with methyl non-adecanoate (C15:0) as the internal standard. As shown in Figure 6, FAMEs of CrPrp19 overexpressed line had no difference with the wild-type CC124, with p > 0.05, which was similar to the contents of total lipids (Figure 4B). Total FAMEs in CrPrp19 RNAi showed a significant increase compared with control Maa7-24 line and wild-type CC124. Among FAs in microalgae lines, the highest contents were C16:0 and C18:3n3. The C16:0 in the CrPrp19 RNAi line compared with Maa7 lines and wild type was increased 1.98 and 1.59 times, respectively, whereas there was no obvious distinction among CrPrp19 RNAi lines and controls in C18:3n3 contents. Another high-content FA in CrPrp19 RNAi transformants was C18:1n9c, which was 2.65 and 1.37 times more than that of Maa7 lines and wild type. Others, such as C14:0 and C18:2 in CrPrp19 RNAi lines, also showed significant accumulation than controls, with 2.02- and 2.20-fold increase. Besides, C16:1, C18:0, C18:3 (5, 9, 12), and C20:0 of CrPrp19 RNAi lines were much more than that of Maa7 lines. C18:1n9t in transformant CrPrp19i-2 compared with wild-type CC124 was increased by 36.6%, but the same as the Maa7 line. The other FAs showed no significant difference. CrPrp19 may negatively regulate FA synthesis.

Protein subcellular localization predicted by Cell-PLoc 2.0 showed that CrPrp19 was a nuclear protein. To verify this result, a fusion construct containing the CrPrp19 gene fused in-frame with the GFP gene and driven by the CaMV 35S promoter was used to transform C. reinhardtii CC124 cells by Agrobacterium-mediated genetic transformation method. As shown in Figure 7, the cells of C reinhardtii CC124 successfully transformed with the chimeric CrPrp19-GFP gene exhibited GFP fluorescence only in the nucleus. Transformation of C. reinhardtii CC124 cells with the empty construct pCAMBIA1302, which contains CaMV35S/GFP gene construct but lacks the CrPrp19 coding sequence, showed no evidence for nuclear localization of GFP activity. That is, in pCAMBIA1302, transformed cells of C. reinhardtii showing GFP were uniformly distributed throughout the cytoplasm of each cell. So CrPrp19 is a nucleoprotein.

Generally, a well understanding of protein interactions could help reveal the cellular functions of target proteins. Here, the yeast two-hybrid (Y2H) assay was conducted to screen interactions of CrPrp19 protein. The full-length coding CrPrp19 gene was cloned in frame with GAL4 DNA-binding domain, and C. reinhardtii cDNA library was cloned with GAL4 activation domain. After detection of non-autoactivation of CrPrp19, the Y2H assays were performed according to the manufacturer of the yeast two-hybrid assay kit. Finally, 52 positive clones grown in SD/QDO/X-α-Gal/AbA (selects for activation of auxotrophic, chromogenic, and antibiotic resistance markers through positive interaction) medium were selected. The gene sequences were identified by PCR amplification, sequencing, and BLASTp analysis. After removing the repeat and invalid sequences, 12 proteins were chosen for gene cloning. Since some of these proteins had long coding genes, only gene fragments were cloned and sequenced from first time Y2H assays. Target genes were cloned into pGADT7 vector and interacted with CrPrp19 bait protein. Finally, six binding proteins were identified, and their information is listed in Table 1, i.e., structural maintenance of chromosomes (SMC) protein, growth factor receptor domain containing protein, peroxisomal coenzyme A diphosphatase, peroxisomal membrane protein 2, phosphoserine transaminase, and pyrophosphate phosphohydrolase.