Mutation of Inositol 1,3,4-trisphosphate 5/6-kinase6 Impairs Plant Growth and Phytic Acid Synthesis in Rice

Inositol 1,3,4-trisphosphate 5/6-kinase (ITPK) is encoded by six genes in rice (OsITPK1-6). A previous study had shown that nucleotide substitutions of OsITPK6 could significantly lower the phytic acid content in rice grains. In the present study, the possibility of establishing a genome editing-based method for breeding low-phytic acid cultivars in rice was explored, in conjunction with the functional determination of OsITPK6. Four OsITPK6 mutant lines were generated by targeted mutagenesis of the gene’s first exon using the CRISPR/Cas9 method, one (ositpk6_1) with a 6-bp in-frame deletion, and other three with frameshift mutations (ositpk6_2, _3, and _4). The frameshift mutations severely impaired plant growth and reproduction, while the effect of ositpk6_1 was relatively limited. The mutant lines ositpk6_1 and _2 had significantly lower levels (−10.1% and −32.1%) of phytic acid and higher levels (4.12- and 5.18-fold) of inorganic phosphorus compared with the wild-type (WT) line. The line ositpk6_1 also showed less tolerance to osmotic stresses. Our research demonstrates that mutations of OsITPK6, while effectively reducing phytic acid biosynthesis in rice grain, could significantly impair plant growth and reproduction.


Introduction
Myo-inositol-1,2,3,4,5,6-hexakisphosphate (IP 6 ), also known as phytic acid (PA), is the main storage form of phosphorous (P) (65-80%) in cereal and legume seeds, accounting for~1.5% of the dry weigh [1]. In most cereal grains, PA exists as mixed salts (phytates) in protein storage bodies and can chelate several mineral cations, including Zn 2+ , Fe 2+ , Ca 2+ , and Mg 2+ [2]. During seed germination, endogenous grain phytase is activated to degrade phytate, releasing myo-inositol, phosphorus, and bound mineral cations [3], which are utilized by the developing seedlings. The PA biosynthetic pathway is still not well defined, but a number of genes involved in its biosynthesis or transport have already been cloned in several plants. Mutations of these genes could result in low-phytic-acid (lpa) grains in rice [4][5][6][7][8][9][10][11][12][13][14] and other plants, e.g., wheat [15] and maize [3,16,17]. In rice, 12 genes have been identified that catalyze the production of intermediate inositol polyphosphates in seeds [18]. Inositol 1,3,4-trisphosphate 5/6-kinase (ITPK) plays a pivotal role in phytic acid biosynthesis, whereby the inositol triphosphate (IP 3 ) molecule is further phosphorylated at the 5th or 6th position [19,20]. ITPK belongs to the ATP-grasp fold proteins group [21] and is conserved from plants to humans with diverse functions. ITPK has even been found in the anaerobic protozoan Entamoeba histolytica [22], where The amino acid sequences of mutant proteins were aligned to that of the wild-type protein using the Clustal Omega Multiple Sequence Alignment (https://www.ebi.ac.uk/Tools/msa/clustalo/). The numbers represent the total number of amino acids, and the amino acid V521 and P522 are highlighted in red box. (C) The three-dimensional structures of OsITPK6 and its mutants were analyzed on SWISS-MODEL (https://www.swissmodel.expasy.org/). The amino acid sequences of mutant proteins were aligned to that of the wild-type protein using the Clustal Omega Multiple Sequence Alignment (https://www.ebi.ac.uk/Tools/msa/clustalo/). The numbers represent the total number of amino acids, and the amino acid V521 and P522 are highlighted in red box. (C) The three-dimensional structures of OsITPK6 and its mutants were analyzed on SWISS-MODEL (https://www.swissmodel.expasy.org/). The ositpk6_1 (a 6-nt in-frame deletion) mutation would result in the loss of two amino acids at positions 91 to 92 ( Figure 1B). In contrast, all the other three mutations, i.e., ositpk6_2 (a 1-nt insertion), ositpk6_3 (a 5-nt deletion), and ositpk6_4 (a 2-nt deletion), would generate a premature stop codon almost right after the mutation site and, hence, significantly shorten the encoded proteins ( Figure 1B). Consequently, the ositpk6_2, _3 and _4 mutant alleles were predicted to produce proteins of only 105, 103, and 104 amino acids, respectively ( Figure 1C). Analysis of ITPK6 proteins of six organisms indicated that the two amino acids missing in the ositpk6_1 mutant were located in a highly conserved segment ( Figure S1), suggesting the mutation of ositpk6_1 could have a potential functional consequence.

Impact of ositpk6 Mutations on Plant Growth and Seed Germination
Plant growth of ositpk6_2, _3, and _4 was significantly impaired. First, their panicles were significantly shorter (−30.1%, −28.8%, and −29.1%, respectively) than that of the WT parental cultivar Xidao 1 (Figure 2A). Second, the mutant panicles had a high percentage of empty grains with darkened glumes ( Figure 2B). Third, the height of the mutant plants was significantly reduced (−37.5%, −36.9%, and −39.1%, respectively) compared with that of their parental cultivar Xidao 1 ( Figure 2C). The impact of ositpk6_1 on plant growth and seed set was limited and not obvious (Figure 2A-C). No significant differences of tiller number per plant were observed between Xidao 1 and all four ositpk6 mutant lines ( Figure 2D). Compared with Xidao 1, the seed set and 1000-grain weight of ositpk6_1 were also significantly decreased (−11.7% and −10.8%, respectively) ( Figure 2E,F). Due to the extremely low seed set, we were not able to harvest enough seeds from ositpk6_3 and ositpk6_4 for the evaluation of 1000-grain weight and other characteristics. The germination of ositpk6_1 was slower in the first three days ( Figure 3A) but gradually caught up with that of the WT after five days and reached~80% on the 7th day ( Figure 3B). The germination rate of ositpk6_2 was far lower than that of the WT ( Figure 3A), being only~20% on the 7th day ( Figure 3B).

Impact of ositpk6 Mutations on Plant Growth and Seed Germination
Plant growth of ositpk6_2, _3, and _4 was significantly impaired. First, their panicles were significantly shorter (−30.1%, −28.8%, and −29.1%, respectively) than that of the WT parental cultivar Xidao 1 (Figure 2A). Second, the mutant panicles had a high percentage of empty grains with darkened glumes ( Figure 2B). Third, the height of the mutant plants was significantly reduced (−37.5%, −36.9%, and −39.1%, respectively) compared with that of their parental cultivar Xidao 1 ( Figure 2C). The impact of ositpk6_1 on plant growth and seed set was limited and not obvious ( Figure  2A-C). No significant differences of tiller number per plant were observed between Xidao 1 and all four ositpk6 mutant lines ( Figure 2D). Compared with Xidao 1, the seed set and 1000-grain weight of ositpk6_1 were also significantly decreased (−11.7% and −10.8%, respectively) ( Figure 2E, F). Due to the extremely low seed set, we were not able to harvest enough seeds from ositpk6_3 and ositpk6_4 for the evaluation of 1000-grain weight and other characteristics. The germination of ositpk6_1 was slower in the first three days ( Figure 3A) but gradually caught up with that of the WT after five days and reached ~80% on the 7th day ( Figure 3B). The germination rate of ositpk6_2 was far lower than that of the WT ( Figure 3A), being only ~20% on the 7th day ( Figure 3B).   (B) The germination rate was recorded from 1 to 7 days after soaking, and three replicates were examined in each group.

Effect of ositpk6 Mutations on Inorganic Phosphorus (Pi), Phytic Acid Phosphorus (PA-P), and Total Phosphorus (TP) in Brown Rice
A colorimetric assay showed that ositpk6_1 and ositpk6_2 had significantly higher Pi levels than the control ( Figure 4A). To quantify the mutational effect of ositpk6_1 and ositpk6_2, the Pi, PA-P, and TP contents were assessed in seeds of these two mutant lines, together with Xidao 1 as the WT control. All mutant lines had significantly lower levels of PA-P and higher levels of Pi compared with the control, while TP was not significantly different from that of the control ( Figure 4B-D). ositpk6_1 and ositpk6_2 had Pi levels of 1.13 mg g −1 and 1.43 mg g −1 , respectively, which were 4.12-and 5.18-fold higher than those of the control (0.28 mg g −1 ), respectively ( Figure 4B). Xidao 1 seeds had a PA-P content of 2.30 mg g −1 , which was significantly greater than those of the two mutant lines; the reduction of PA-P levels was 10.1% and 32.1% in ositpk6_1 and ositpk6_2, respectively ( Figure 4C). Xidao 1 seeds had a TP content of 3.9 mg g −1 , which was not significantly different from those of the two mutant lines ( Figure 4D). (B) The germination rate was recorded from 1 to 7 days after soaking, and three replicates were examined in each group.

Effect of ositpk6 Mutations on Inorganic Phosphorus (Pi), Phytic Acid Phosphorus (PA-P), and Total Phosphorus (TP) in Brown Rice
A colorimetric assay showed that ositpk6_1 and ositpk6_2 had significantly higher Pi levels than the control ( Figure 4A). To quantify the mutational effect of ositpk6_1 and ositpk6_2, the Pi, PA-P, and TP contents were assessed in seeds of these two mutant lines, together with Xidao 1 as the WT control. All mutant lines had significantly lower levels of PA-P and higher levels of Pi compared with the control, while TP was not significantly different from that of the control ( Figure 4B-D). ositpk6_1 and ositpk6_2 had Pi levels of 1.13 mg g −1 and 1.43 mg g −1 , respectively, which were 4.12-and 5.18-fold higher than those of the control (0.28 mg g −1 ), respectively ( Figure 4B). Xidao 1 seeds had a PA-P content of 2.30 mg g −1 , which was significantly greater than those of the two mutant lines; the reduction of PA-P levels was 10.1% and 32.1% in ositpk6_1 and ositpk6_2, respectively ( Figure 4C). Xidao 1 seeds had a TP content of 3.9 mg g −1 , which was not significantly different from those of the two mutant lines ( Figure 4D).

Effect of ositpk6 Mutation on Stress Tolerance
To further test whether the mutation also had any impact on stress tolerance, we subjected the ositpk6_1 and Xidao 1 plants to osmotic stress treatment (because of the limited number of seeds harvested and the low germination rate, ositpk6_2, was not further analyzed). The growth of ositpk6_1 seedlings appeared to be inferior to that of the WT control grown either under normal or stressed conditions ( Figure 5A). The shoot length of ositpk6_1 was shorter than that of the WT with or without stress treatment (100 mM NaCl or 20 mM mannitol), while the root length of ositpk6_1 was shorter than that of the WT only under stress ( Figure 5A,B). There was no significant difference in the number of leaves and roots with or without stress treatment between ositpk6_1 and WT ( Figure 5C).

Effect of ositpk6 Mutation on Stress Tolerance
To further test whether the mutation also had any impact on stress tolerance, we subjected the ositpk6_1 and Xidao 1 plants to osmotic stress treatment (because of the limited number of seeds harvested and the low germination rate, ositpk6_2, was not further analyzed). The growth of ositpk6_1 seedlings appeared to be inferior to that of the WT control grown either under normal or stressed conditions ( Figure 5A). The shoot length of ositpk6_1 was shorter than that of the WT with or without stress treatment (100 mM NaCl or 20 mM mannitol), while the root length of ositpk6_1 was shorter than that of the WT only under stress ( Figure 5A,B). There was no significant difference in the number of leaves and roots with or without stress treatment between ositpk6_1 and WT ( Figure 5C).

Discussion
ITPK6 is a unique gene in the ITPK gene family, and knowledge of its function has so far been very limited. The identification and characterization in rice of two itpk6 mutant lines were reported

Discussion
ITPK6 is a unique gene in the ITPK gene family, and knowledge of its function has so far been very limited. The identification and characterization in rice of two itpk6 mutant lines were reported in a study, which is the only study on the function of ITPK6 in all organisms [30]. Our present study demonstrated that the knockout of OsITPK6 could severely impair plant growth and reproduction, implying that ITPK6 may play important roles in plant growth and development, in addition to the biosynthesis of inositol polyphosphates.
First, we observed that the OsIPTK6 knockout mutants (ositpk6_2, _3, and _4) generated in the present study grew poorly, e.g., their plant height was reduced to almost half of that of the WT, and their fertility was almost abolished (Figure 2). These results suggested that ITPK6 plays an important role not only in the vegetative growth but also in the reproduction of rice. Because the reduction of phytic acid content in ositpk6_1 grains (−32.1%) was less than in the P522L mutant line (−46%) [30], the reduction of phytic acid alone could not explain the inferior performance of ositpk6_1. Further studies are needed to uncover the biological basis leading to the discrepancy between our present study and that reported in reference [30] regarding the mutational effect on rice growth and reproduction.
In the present study, we identified an ositpk6 mutant, i.e., ositpk6_1, with a 6-bp deletion. Though only two amino acids are expected to be removed from the derived protein, ositpk6_1 did reduce IP 6 content. This may be because these two amino acids are located in a conserved region ( Figure S1). We previously also observed a similar case of lpa rice, where a 6-bp deletion (and, hence, a deletion of two amino acids) in OsSultr3;3 significantly reduced grain phytic acid content [14]. Although this mutation only reduced phytic acid content by less than 20%, it did exert a negative effect on seed set, grain weight, seed germination, and tolerance to abiotic stresses. This was somehow unexpected, because the P522L mutant line with a 46% reduction of phytic acid content still had normal plant growth as its WT parent [30]. Further studies are needed to fully evaluate the mutational effect by examining more ositpk6 mutants.
The usefulness of CRISPR/Cas9-based mutagenesis for improving a particular trait is strictly dependent on the performance of the generated mutants. There are often trade-offs for mutating a gene for a specific purpose, and the overall performance of the mutated plant could be affected as a consequence of pleiotropic effects. Enlightened by the findings of Kim and Tai [30], we hoped to establish a fast and effective method for breeding yield-competitive, lpa rice cultivars by using genome-editing techniques. However, our results suggest that OsITPK6 or its product plays an important role in multiple cellular processes, and simply knocking out OsIPTK6 would impair rice growth and reproduction and, hence, would not work for our purpose.
Because [30] of the success in the production of lpa mutants without a significant negative impact on plant growth and seed development, it is still possible to generate ositpk6 mutants without a significant effect on plant growth, if more appropriate vectors can be designed and more mutants are identified and assessed.
In summary, the present study demonstrates that the OsITPK6 gene is essential for rice growth and reproduction.

CRISPR/Cas9 Vector Construction and Rice Transformation
To generate OsITPK6 mutants, the 1st exon of OsITPK6 (Os09g0518700) was chosen as a target ( Figure 1A). The sgRNAs were designed by searching UniProt for precise positions (http://www.uniprot. org/), and CRISPR-P program (http://cbi.hzau.edu.cn/cgi-bin/CRISPR/) was used to minimize off-target effects [37]. Because of the homology of the target sequence among OsITPK genes, it is unlikely to cause mutations in the other five OsITPK homolog genes ( Figure S2). DNA oligonucleotides were synthesized (Tsingke, Hangzhou, China) for the construction of a CRISPR/Cas9 vector, pH_itpk6, using the pHun4c12s as the starting plasmid, which harbors a CYP81A6-hpRNAi element [38] and was modified from pHun4c12 [39]. Correspondingly, the pH_itpk6 plasmid was transformed into Agrobacterium tumefaciens and used for rice transformation.
Rice calli were induced from mature seeds of the cultivar 'Xidao 1' (O. sativa L. japonica) and were transformed with the pH-itpk6 vector by Agrobacterium-mediated transformation according to reference [40]. Transgenic plantlets were regenerated from hygromycin-resistant calli and acclimatized inside a moist growth chamber (28 • C with a 12 h photoperiod) for one week before being transplanted to experimental facilities.
Ethics Approval and Consent to Participate: The experiments did not involve endangered or protected species. No specific permits were required for these locations/activities.

Mutation Detection in T 0 Plants
For detection of transgenes and mutations in regenerated T 0 plants, total genomic DNA was extracted from leaf tissues following a modified cetyltrimethylammonium bromide (CTAB) method [41]. The presence of the HPT gene was assessed by PCR using the primers HygR-F (5 -AGAAGAAGATGTTGGCGACCT-3 ) and HygR-R (5 -GTCCTGCGGGTAAATAGCT-3 ) [42]. Site-specific mutations were detected by PCR amplification using primer pairs flanking the designated target sites in OsITPK6, i.e., P6-F (5 -CTCGACCCATCCGGTGTTAC-3 ) and P6-R (5 -AAATCGCAGGGGAGAGATCG-3 ) ( Figure 1A). The following generalized PCR program was used: 5 min at 94 • C, followed by 35 cycles of 30 s at 94 • C, 1 min at 60 • C, and 1 min at 72 • C, with a final extension of 10 min at 72 • C.
The PCR products were first subjected to HRM analysis for mutations according to reference [43], putative mutants were sequenced (TSINGKE, Hangzhou, China), and mutated sequences were decoded using the DSDecode program (http://skl.scau.edu.cn/dsdecode/) [44]. Mutations of a few selected plants were further confirmed by clone sequencing.

Development of Transgene-Free Mutant Lines
To obtain homozygous transgene-free mutants, T 1 seedlings were foliar-sprayed with 1000 mg/L bentazon (approximately 100 mL/m 2 ) at about the four-leaf stage according to a previous study [38]. At least five surviving T 1 plants from each independent T 0 plants were selected for further analysis of the presence of T-DNA and site-specific mutations. Eventually, four OsITPK6 mutant lines (ositpk6_1, _2, _3, and _4) ( Figure 1A) were identified; the T 2 and advanced-generation seeds were used for the experiments.

Agronomic Traits Assay
The transgene-free mutant lines and their wild-type parent control (Xidao 1) were grown at the Experimental Farm of Zhejiang Zhijiang Seed Tec. Ltd., Hangzhou, China, during the summer season, and their agronomic traits were evaluated either in fields or post-harvest. Plants were grown in randomized plots, each with 60 plants. Twenty inner plants of each plot were evaluated for each parameter.

Seed Germination Assay
A controlled germination test was performed to assess seed germination capability, each with 100 seeds and replicated for three times [45]. Seeds were soaked in water for 48 h at 30 • C and then germinated on filter paper soaked with distilled water at 30 • C in the dark for one week, and the germination percentage was recorded daily.

Seed Phosphorus Assay
Seed inorganic P (Pi) levels were assayed qualitatively according to the micro-determination method developed in reference [46] with modifications. The qualitative assay was used for the identification of the high-inorganic-P (HIP) phenotype. The seeds were transferred to 96-well plates and extracted in 0.4 M HCl solution (10 µL per mg sample) overnight at 4 • C. Aliquots of 10 µL of supernatant per each sample were used for Pi level determination, according to reference [4] with slight modifications [6]. The development of a blue color implied increased level of Pi (HIP), while colorless samples typified the WT levels of parent varieties ( Figure S1).
Seed Pi levels were also quantitatively determined according to reference [47] in triplicates, as follows. Brown rice grains were ground into rice flour, and~400 mg rice flour per sample were extracted in 12.5% (w/v) TCA (trichloroacetic acid containing 25 mM MgCl 2 ) by gentle shaking overnight at 4 • C. After centrifugation at 15,000× g for 10 min, the supernatants were used for Pi assay according to reference [2].
The PA content was determined for brown rice flour according to reference [48] by using a commercial assay kit (Megazyme, Irland) in triplicate. Briefly,~1 g of brown rice grains was mixed with 20 mL of hydrochloric acid (0.66 M), stirred vigorously for 3 h at room temperature, and then centrifuged at 13,000 rpm for 10 min. Then, 0.5 mL of the supernatant was immediately neutralized by the addition of 0.5 mL of sodium hydroxide solution (0.75 M). The neutralized sample extract was subjected to an enzymatic dephosphorylation reaction procedure. The absorbances at 655 nm was determined using an ultraviolet spectrophotometer.
Total seed phosphorus and mineral elements were determined according to reference [49] in triplicate. Briefly, brown rice samples were digested in a microwave digestion system (Mars6, USA) at 160 • C for 40 min, using~200 mg of sample in 7 mL of HNO 3 . The digested solution was concentrated at 140 • C for 2 h until less than 1 mL of solution was left and then brought to 30 mL with ultrapure water [50]. Total seed phosphorus and mineral elements were analyzed by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) (PerkinElmer, MA, USA).

Stress Treatment
For stress treatment, 14-day-old seedlings were planted in 1× Murashige and Skoog (MS) liquid medium [51], supplemented with 100 mM NaCl or 20 mM mannitol and grown for 7 days before sampling at midday.

Statistical Analysis
All statistical analyses were performed using the student's t-test. The experimental data are presented with the mean standard errors (SE) based on three to six replications. The means were compared by ANOVA, and the significance of the differences between group means were calculated by the Bonferroni Post-tests.