Generation of giant embryo mutants
ZH11 is a WT with a small embryo; three sgRNA targets in the first exon of GE were selected for CRISPR/Cas9 knockout of ZH11 (Fig. 1a). Hygromycin gene detection and sequencing were used to analyze the mutations in 34 T0 transgenic lines. At T0, there were 10 homozygous mutants, 18 heterozygous mutants, and 6 negative plants without mutations. Among them, ge-1 inserted a base T at + 355 (A in the start codon ATG of GE is + 1) in the second target, resulting in a frameshift mutation and premature termination of protein translation at amino acid position 328. In ge-2, the deletion of base A at + 355 in the second target resulted in a frameshift mutation and premature termination of protein translation at amino acid position 126. ge-3 mutated the base from A to C in the third target at + 661, and the corresponding codon changed from ATC (Ile) to CTC (Leu) (Fig. 1b).
Gaba Content, Embryo Weight, And Embryo Volume Of Giant Embryo Mutants Were Significantly Increased
Transgenic plants were planted and T2 homozygous giant embryo mutants with stable genotypes were obtained. There were three giant embryo mutants (ge-1, ge-2, and ge-3) with different genotypes and stable phenotypes. The grain phenotypes of GABA content and embryo size were also investigated (Fig. 2). GABA content was measured in the three giant embryo mutants. Compared with the WT, giant embryo mutant ge-1, ge-2, and ge-3 showed large embryos of different sizes in grain appearance (Fig. 2a). With an increase in embryo volume, the GABA content also gradually increased. The GABA content of WT was 0.0056 mg/g, and that of brown rice ge-3, ge-1, and ge-2 was 0.0155, 0.0416, and 0.0434 mg/g, respectively, which increased by 176.79%, 642.86%, and 675.00%, respectively, compared with that of the WT (Fig. 2b). Compared with WT, ge-1, ge-2, and ge-3 showed highly significant differences in the weight of the embryo, endosperm, and brown rice and the weight ratio of embryo to brown rice, respectively. The ge-2 mutant had the highest seed embryo weight and weight ratio of embryos to brown rice. Compared with WT, 1000-grain embryo weights of ge-3, ge-1, and ge-2 increased by 95.38%, 130.77%, and 169.23%, respectively. The 1000-grain weight of brown rice of ge-3, ge-1, and ge-2 decreased by 11.18%, 35.18%, and 34.81%, respectively, compared with that of the WT. Compared with WT, the 1000-seed weight of the endosperm of ge-3, ge-1, and ge-2 decreased by 14.89%, 40.95%, and 41.91%, respectively. Compared with the WT, the weight ratios of embryos to brown rice of ge-3, ge-1, and ge-2 increased by 119.98%, 256.00%, and 313.00%, respectively (Fig. 2c). Similarly, it was found that the volume of a 100-grain embryo and the volume ratio of embryo to brown rice of ge-3, ge-1, and ge-2 were significantly different between the mutants and WT, and showed an increasing trend. The volumes of the 100-grain brown rice, embryo, and endosperm of the three mutants were significantly different from those of the WT. The volume of 100-grain embryos of ge-3, ge-1, and ge-2 increased by 160.87%, 284.78%, and 332.39%, respectively, compared with the WT. Compared with the WT, the volume of 100-grain brown rice in ge-3, ge-1, and ge-2 decreased by 1.38%, 1.84%, and 2.30%, respectively, and the volume of the 100-grain endosperms of ge-3, ge-1, and ge-2 decreased by 6.57%, 11.01%, and 13.01%, respectively. At the same time, compared with the WT, the volume ratio of embryo to brown rice of ge-3, ge-1, and ge-2 increased by 164.51%, 292.00%, and 342.58%, respectively (Fig. 2d). Using CRISPR/Cas9 multi-target technology to knockout GE, multiple ge mutants can be obtained, and a variety of rice germplasms with embryos larger than WT can be created.
Various substances were significantly altered in the ge-1 mutant
The phytohormones auxin and cytokinin (CK) play important roles in regulating organ size. In this study, IAA, trans-zeatin riboside (tZR), trans-zeatin (tZ), and cis-zeatin (cZ) of the ge-1 embryo were significantly lower than those of the WT when the embryo grew larger (Fig. 3a). The contents of total starch, crude protein, crude fat, and various minerals in WT and ge-1 brown rice were detected (Fig. 3b, c). Compared with the WT, the crude fat and protein contents of ge-1 brown rice were significantly increased, while the total starch content was significantly decreased (Fig. 3b). Most of mineral contents of ge-1 brown rice were significantly increased compared with WT, including boron (B), iron (Fe), magnesium (Mg), zinc (Zn), calcium (Ca), copper (Cu), potassium (K), manganese (Mn), strontium (Sr), and barium (Ba). The phosphorus (P) content increased significantly. However, sodium (Na) was the only mineral that showed a significant decrease between WT and ge-1 brown rice, and there were no significant differences in vanadium (V), nickel (Ni), molybdenum (Mo) and lead (Pb) (Fig. 3c).
Analysis Of Gene Expression Related To Gaba Metabolism
To explore the molecular regulation mechanism of the significantly increased GABA content in the ge mutant, grain samples of WT and ge-1 at 5 and 9 DAF were used for RNA-seq. The expression levels of a lot of genes were significantly different between WT and ge-1 (Fig. S1, S2). Compared with the WT, there were 231 DEGs in the ge-1 mutant at 5 DAF, among which 135 and 96 DEGs were upregulated and downregulated, respectively. A total of 1128 DEGs were detected between WT and ge-1 at 9 DAF, among which 868 and 260 DEGs were upregulated and downregulated, respectively (Fig. S1, S2). The alanine, aspartate, and glutamate metabolism pathways were related to the GABA metabolism pathway. KEGG analysis of significant DEGs in this metabolic pathway revealed significant increases in the expression of genes associated with glutamate, glutamine, and GABA synthetase in the ge-1 mutant 9 DAF. Among them, the expression levels of genes LOC_Os08g36320 (GAD1), LOC_Os05g48200 (OsNADH-GOGAT2), LOC_Os09g26380, and LOC_Os03g50490 (OsGS1;3) corresponding to five enzymes (EC4.1.1.15, EC1.4.1.3, EC1.4.1.14, EC2.6.1.2, and EC6.3.1.2) were significantly upregulated, whereas the expression level of LOC_Os06g10420 (EC3.5.1.3) was significantly downregulated (Fig. S3). At the same time, the expression levels of these genes were also verified by qRT-PCR in the ge-1 mutant at 9 DAF (Fig. 4a-e). Therefore, quantitative analysis of the expression levels of genes encoding metabolic enzymes in the GABA shunt and polyamine degradation pathways showed that the expression levels of many genes were significantly different between the WT and ge-1 (Fig. 4f-t). In the GABA shunt, GAD is the rate-limiting enzyme of GABA synthesis is GAD and GABA-T is a GABA-degrading enzyme. The expression levels of the genes LOC_Os08g36320 (GAD1), LOC_Os04g37500 (GAD2), and LOC_Os04g37460 (GAD5) encoding GAD were significantly upregulated in the ge-1 mutant at 9 DAF, but there was no significant difference at 5 DAF (Fig. 4a, f, g). The genes encoding GABA-T, LOC_Os08g10510 (GABA-T2), and LOC_Os04g52440 (GABA-T3) were significantly downregulated in the ge-1 mutant at 5 and 9 DAF, and LOC_Os04g52450 (GABA-T1) was only significantly downregulated at 9 DAF (Fig. 4h, j, k). The rate-limiting enzymes in the polyamine degradation pathway are DAO and PAO. LOC_Os04g20164 (DAO2), LOC_Os04g53190 (PAO3), and LOC_Os04g57550 (PAO4) was significantly upregulated in the ge-1 mutant at 5 and 9 DAF, and LOC_Os04g40040 (DAO4), LOC_Os01g51320 (PAO1), and LOC_Os04g57560 (PAO5) were significantly upregulated at 5 DAF but not at 9 DAF (Fig. 4i-q). LOC_Os06g23114 (DAO1) was significantly downregulated and LOC_Os09g20260 (PAO6) was significantly upregulated in the ge-1 mutant at 9 DAF (Fig. 4n, r). The expression of LOC_Os09g20284 (PAO7) was significantly upregulated at 5 DAF and downregulated at 9 DAF in the ge-1 mutant (Fig. 4s). The key enzyme linking amino acid metabolism and the TCA cycle is mainly glutamate dehydrogenase (GDH) (Kim and Baik 2019). Glu, the substrate for GABA synthesis, is converted into α-ketoglutarate by GDH to enter the TCA cycle, which can reduce the accumulation of GABA. The gene LOC_Os04g45970 (GDH2), encoding GDH, was significantly downregulated at 5 and 9 DAF in the ge-1 mutant (Fig. 4t). In conclusion, the above results suggest that the expression levels of most of genes conducive to GABA synthesis are significantly upregulated in the ge-1 mutant, while the expression levels of most of genes conducive to GABA degradation are significantly down-regulated in the ge-1 mutant.
By comparing KEGG pathway enrichment analysis of DEGs at 5 and 9 DAF, it was found that phenylalanine metabolism and galactose metabolism had significant gene enrichment in both periods (Fig. 5).
DEGs in RNA-seq in the two pathways were analyzed, and DAO2 and DAO4, which are related to DAO in the GABA polyamine degradation pathway, were detected in the phenylalanine metabolism pathway at 9 DAF. The expression levels of DEGs in the two pathways were detected, and it was found that the expression levels of most of genes were significantly different between WT and the ge-1 mutant at 5 and 9 DAF (Table 1).
Table 1
Gene expression levels involved in GABA metabolism in WT and the ge-1 mutant grains at 5 and 9 DAF
Metabolic pathway | Locus ID | WT-5 DAF | ge-1-5 DAF | WT-9 DAF | ge-1-9 DAF |
Galactose metabolism | LOC_Os02g12730 | 8502.16 ± 80.37 | 48331.57 ± 160.86** | 6891.62 ± 51.48 | 708.06 ± 31.48* |
| LOC_Os03g59190 | 275.01 ± 6.70 | 205.30 ± 9.91 | 202.47 ± 11.26 | 194.44 ± 8.44 |
| LOC_Os04g38530 | 13105.95 ± 72.78 | 10050.38 ± 84.95 | 16404.90 ± 37.23 | 2086.05 ± 33.11** |
| LOC_Os05g35360 | 8322.77 ± 38.28 | 5412.54 ± 15.41 | 1369.19 ± 57.37 | 4030.38 ± 104.90 |
| LOC_Os05g44922 | 673965.36 ± 308.21 | 1156569.73 ± 371.10* | 300250.26 ± 240.79 | 184498.80 ± 210.83* |
| LOC_Os06g46284 | 16597.38 ± 46.76 | 29423.48 ± 99.84 | 119344.53 ± 213.44 | 596688.68 ± 318.61** |
| LOC_Os07g26900 | 32128.30 ± 106.90 | 18496.32 ± 30.03* | 2161.39 ± 49.05 | 134.95 ± 9.37** |
| LOC_Os07g48160 | 45102.80 ± 75.07 | 46600.29 ± 76.98 | 48988.75 ± 172.24 | 3015.06 ± 330.15* |
| LOC_Os09g30240 | 9663.93 ± 80.99 | 5461.94 ± 57.34 | 7241.17 ± 51.48 | 15589.76 ± 20.16* |
| LOC_Os09g38340 | 1174.35 ± 22.02 | 60.28 ± 5.53** | 754.03 ± 31.45 | 53.89 ± 17.26** |
| LOC_Os10g35330 | 470.19 ± 1.36 | 111.17 ± 4.99** | 3.26 ± 1.13 | 6.16 ± 0.73 |
Phenylalanine metabolism | LOC_Os01g56380 | 167.52 ± 7.51 | 886.36 ± 71.03* | 24.84 ± 4.03 | 6.87 ± 5.43 |
| LOC_Os02g19924 | 144644.24 ± 280.49 | 68835.99 ± 189.83* | 35482.10 ± 108.70 | 42628.89 ± 138.18 |
| LOC_Os02g19970 | 7962.91 ± 21.33 | 6489.79 ± 39.19 | 23022.93 ± 73.45 | 16992.16 ± 189.80 |
| LOC_Os03g15340 | 6214.79 ± 115.26 | 1589.80 ± 35.76 | 15631.42 ± 156.96 | 550.94 ± 12.39** |
| LOC_Os04g02754 | 87862.71 ± 161.87 | 33412.36 ± 57.32* | 35828.20 ± 194.28 | 26489.90 ± 103.49 |
| LOC_Os11g33090 | 23267.23 ± 105.63 | 6824.88 ± 41.25* | 27458.77 ± 87.94 | 28020.06 ± 49.14 |
| LOC_Os11g42510 | 6659.72 ± 104.64 | 48750.81 ± 93.01** | 15929.08 ± 57.34 | 3323.77 ± 48.68* |
Phenotypic statistics are presented as mean ± s.e.m; * and ** perform significant differences at P < 0.05 and P < 0.01 between WT and the ge-1 mutant, respectively |