Generation of aroma in three-line hybrid rice through CRISPR/Cas9 editing of BETAINE ALDEHYDE DEHYDROGENASE2 (OsBADH2)
Abstract
Aroma or fragrance in rice is a genetically controlled trait; Its high appreciation by consumers increases the rice market price. Previous studies have revealed that the rice aroma is controlled by a specific gene called BETAINE ALDEHYDE DEHYDROGENASE (OsBADH2), and mutation of this gene leads to the accumulation of an aromatic substance 2-acetyl-1-pyrroline (2-AP). The use of genetic engineering to produce aroma in commercial and cultivated hybrids is a contemporary need for molecular breeding. The current study reports the generation of aroma in the three-line hybrid restorer line Shu-Hui-313 (SH313). We created knock-out (KO) lines of OsBADH2 through the CRISPR/Cas9. The analysis of KO lines revealed a significantly increased content of 2AP in the grains compared with the control. However, other phenotypic traits (plant height, seed setting rate, and 1000-grain weight) were significantly decreased. These KO lines were crossed with a non-aromatic three-line hybrid rice male sterile line (Rong-7-A) to produce Rong-7-You-626 (R7Y626), R7Y627 and R7Y628. The measurement of 2-AP revealed significantly increased contents in these cross combinations. We compared the content of 2-AP in tissues at the booting stage. Data revealed that young spike stalk base contained the highest content of 2-AP and can be used for identification (by simple chewing) of aromatic lines under field conditions. In conclusion, our dataset offers a genetic source and illustrates the generation of aroma in non-aromatic hybrids, and outlines a straightforward identification under field conditions.
1 INTRODUCTION
Rice (Oryza sativa L.) is one of the important crops in the world and feeds more than half of the world's population. There are five major subgroups according to the level of genetic diversity (Garris et al., 2005; Caicedo et al., 2007). Among those, aromatic cultivars have a pleasant odor/aroma and high market price compared to others (Jana et al., 2011). The aroma in rice grains is regarded as a superior trait and indicative of cooking and eating quality (Sakthivel et al., 2009). Previous studies indicated the presence of more than 300 aromatic compounds in aromatic rice (Petrov et al., 1996; Widjaja et al., 1996; Wakte et al., 2017). Among them, 2-acetyl-1-pyrroline (2-AP) is a key aromatic compound present in aromatic cultivars, especially basmati and jasmine (Lorieux et al., 1996; Jezussek et al., 2002; Sakthivel et al., 2009; Peng et al., 2018). It produces a popcorn-like and nutty pleasant aroma while cooking rice, and breeders are trying to engineer high 2-AP-containing cultivars. According to the level of aroma, rice cultivars can be divided into non-aromatic, moderate-aromatic, light-aromatic and strong-aromatic types (Jana et al., 2011).
The polyamine metabolism is considered the main pathway of 2-AP biosynthesis, which depends on BETAINE ALDEHYDE DEHYDROGENASE (BADH2) (Prodhan & Shu 2020). Polyamines are small organic compounds that contain two or more amino groups. In this pathway, polyamine is converted to γ-amino butyraldehyde (GABald), which is a precursor of γ-aminobutyric acid (GABA). GABald cyclizes to Δ1-pyrroline, which is a precursor of 2-AP (Chen et al., 2008). In aromatic rice, loss of function of OsBADH2 enzyme represses the conversion of GABald to GABA, resulting in the accumulation of 2-AP (Bradbury et al., 2008).
Previous studies reported that aroma is a quantitative trait and is affected by many aromatic compounds. Other studies reported that it is controlled by a single recessive gene and located on chromosome 8 (Ahn et al., 1992; Petrov et al., 1996; Dong et al., 2001). Map-based cloning and fine mapping helped to narrow down the interval to a region containing OsBADH2 (Bradbury et al., 2005; Wanchana et al., 2005; Chen et al., 2006). Later on, Chen et al. confirmed through genetic complementation that OsBADH2 is a candidate gene for aroma and contains 15 exons and 14 introns (Chen et al., 2008). Genome-wide analysis indicated the presence of a 8-base deletion in the seventh exon and three SNPs in the OsBADH2 in aromatic cultivars (Bradbury et al., 2005); mutations resulting in a decreased enzyme activity that leads to an increase in 2AP content and popcorn-like aroma (Niu et al., 2008; Shan et al., 2013). Previous studies reported other variation types in OsBADH2 indicating the negative correlation with 2-AP expression and differences in aroma level are determined mainly by the variation types (Amarawathi et al., 2008; Shi et al., 2008; Kovach et al., 2009).
Cultivation of aromatic rice is popular in India, Thailand, Cambodia, Pakistan and different regions of China (e.g., Pang-Xie-Gu in Yunnan, Xiang-He in Guizhou and Jing-Xi-Xiang-Nuo in Guangxi, etc.). However, the most common defects (especially in China) associated with aromatic cultivars are prolonged growth periods, poor lodging resistance and low grain yield. These defects limit the commercial application and promotion of these cultivars on a large scale. Therefore, attention is being diverted to the breeding of high-yielding aromatic hybrids to replace the current aromatic cultivars (Han, 2003; Zhou & Liao, 2002). The conventional breeding method for aromatic rice is laborious and time-consuming. The advancements in biotechnology and genome editing methods ease their application and can be exploited to decrease the cost and save time. RNA silencing (RNAi) has been successfully utilized to produce aroma in non-aromatic cultivars (Niu et al., 2008; Chen et al., 2012). Similarly, transcription activator-like effector nucleus (TALEN) technology has also been deployed to edit OsBADH2 and can induce aroma in non-aromatic rice cultivars (Shan et al., 2013; Shan et al., 2015). Clustered regularly interspaced short palindromic repeats-associated protein9 (CRISPR/Cas9) is a new genomic editing technique that could be used to replace DNA sequence (Guo et al., 2023; Li et al., 2021). A recent study reported the use of CRISPR/Cas9 to knock out OsBADH2 to produce aromatic cultivars (Hui et al., 2022; Zhang et al., 2023). Previous studies have shown that editing of OsBADH2 could confer rice aroma, but it also has significant negative effects on yield-related traits, making it difficult to apply in production.
Whether knock-out of OsBADH2 can be used in hybrids to rapidly create aroma and what would be possible effects on other heterotic yield-related traits, was not studied before. Shu-Hui-313 (SH313) is a hybrid restorer line and is a genetic source of five non-aromatic cultivars. To create the aromatic SH313, we targeted the second exon of OsBADH2 using CRISPR/Cas9 to obtain KO lines (KO626, KO627, and KO628). Those lines had aroma but their yield-related traits were significantly decreased. We further explored their breeding utilization by crossing KO lines with a male sterile line Rong-7-A and produced R7YKO626, R7YKO627 and R7YKO628. Compared with control (R7Y313), the grain aroma of hybrids of KO lines was significantly increased and yield-related traits had no significant change. Together, our data provides a practical example of rapid improvement of aroma in rice hybrids using CRISPR/Cas9.
2 MATERIALS AND METHODS
2.1 Plant materials and growth conditions
SH313 is a non-aromatic restorer line of the three-line hybrid system and was used as the male parent to cross with Rong-7-A (corresponding male sterile line of SH313). Rong-7-A was used as a control (CT) parent throughout this study. Rong-7-A was crossed with SH313 and knockout lines (KO626, KO627 and KO628) of SH313 and four hybrid combinations were created [Rong-7-You-313 (R7Y313), R7YKO626, R7YKO627, R7YKO628]. Hybrid R7Y313 is a non-aromatic three-line hybrid cultivar that has passed the Trials of National Examination and officially approved to be released (approval number: 20190058) and registered at Rice Data Centre https://www.ricedata.cn/variety/index.htm. All experimental materials were grown alternatively in the paddy fields at Chengdu (N30.67°, E104.06°), Sichuan and Lingshui (N18.47°, E110.04°), Hainan in the People's Republic of China.
2.2 Agronomic traits measure
Each hybrid/knock-out lines were planted in a plot consisting of 10 rows, with 10 plants in each row. The rows were separated by a distance of 25 cm, and the plants within each row were spaced 15 cm apart. 5-point sampling method was carried out, and two plants were taken at each point for a total of 10 plants. To evaluate plant height and tiller number, 10 plants were measured at heading stage. Seeds of 10 individual plants were collected to calculate seed setting rate and 1000-grain weight.
2.3 CRISPR/Cas9
To knock out OsBADH2, a CRISPR/cas9 vector was constructed. Briefly two guide RNA (gRNA) with the sequences “badh2-Y (CAAGTACCTCCGCGCAATCG) “and “badh2-B (TATGGCTTCAGCTGCTCCTA)” with protospacer adjacent motif (PAM) sequence “CCA” were identified from the coding sequence of OsBADH2. To increase knock-out efficiency, two targets were simultaneously aimed at. Oligos were designed and possible off-targets were prevented using BLAST search and knock-out constructs were amplified using PCR, according to a previous study (Ali et al., 2022). Badh2-Y was inserted into Pbwa(V)-cas9i2 and badh2-B into pBWD (LB) DNA at the EcoRI cloning site. Reaction system (10 μL): 10 × buffer 1.0 μL, vector 1.5 μL, fragment 2.0 μL, EcoRI 0.5 μL, T4-ligase 0.5 μL and ddH2O 4.5 μL, 37°C 2 h. Finally, pBWA(V)-cas9i2-badh2-Y and pBWD(LB)DNAi-badh2-B were assembled into the final vector pBWA(V)-cas9i2-badh2 using endonuclease SapI and T4 ligase. The constructs were verified by sequencing and transferred into Agrobacterium to transform rice callus (Ma et al., 2016; Toki et al., 2006).
In brief, mature seeds of wild-type SH313 were collected, peeled and disinfected. These seeds were placed on the induction medium and cultivated in 30°C under light. After 6–8 days of cultivation, the seeds were induced to develop callus. Then the calluses were inoculated into the subculture medium and were cultured in 30°C at light for 3 days. Then, calluses were placed in the Agrobacterium culture solution (OD600 = 0.08–0.1) and incubated for 30 minutes. Then, the calluses were transferred to the culture medium (N6D + 100 μM AS+30 g/L sucrose, pH 5.8) cultured at 22–25°C in darkness for 2–3 days. After that, the calluses were transferred to the screening medium (containing 50 mg/L Hyg) and were cultured in 30°C at light for 15 days. After that, the calluses were transferred to a new screening medium (subculture medium as previously described) and were cultured under the same conditions for 15 days. After two rounds of screening cultivation, most of the negative calluses died. The positive calluses were transferred to the differentiation medium and were cultured at light under 30°C for 20 days. The calluses turned green and began to differentiate. When the green calluses grew into 3–5 cm seedlings, they were transferred to the rooting medium for further cultivation, until transgenic seedlings were obtained. The sequence of positive knock-out lines was verified from genomic DNA from T1 transgenic plants using primers (for target one, F: 5′-TCTCCACCCTCTGCTTCTGCCTCT-3′, R: 5’-TTTGGAATAAGTTGGAAGCATGGCTG-3′), (for target two, F:5’-CAGGACTTGTTTGGAGCT-3′, R:5’-TAATGCCATGCCAACTG-3′). Mutations of knock-out lines were detected by the software SnapGene 6.0.2.
2.4 Evaluation of aroma using leaching-method
Aroma evaluation was carried out according to the protocol published by Scod et al. (1980). Briefly, 1 g of fresh leaves (tillering stage) were cut into pieces and put into a 10-mL centrifuge tube containing 4 mL KOH solution (1.7%). After 10 minutes, selecting a non-aromatic sample(Shu-Hui-313)as the first control and an aromatic sample(Dao-Hua-Xiang)as the second control, and 10 testers (ten students from the lab) were used to evaluate the aroma of leaves. In the test, based on the level of the aroma, we classified three types: non-aroma, light-aroma and strong-aroma (Table S1).
2.5 Determination of 2-AP content
A gas chromatography-mass spectrometer (GC–MS) was used to measure 2-AP contents. Briefly, 10 mg of 2,4,6-3-methyl-pridine was placed in a 10-mL volumetric flask and was diluted with ethanol for a standard stock solution. 1 mL standard stock solution was placed in a 100-mL volumetric flask and diluted to 10 μg/mL with ethanol to have an extractant. Crushed grain sample (500 mg of rice flour) was passed through a 0.5-mm sieve and then poured into a 10-mL tube. 0.8 mL extractant was added, the vessel was sealed and placed in an oven at 80°C for 3 hours. The vessel was moved out and let it cool down to room temperature. Samples were sterilized by 0.22 μm pore size filtration. At the end, 150 μL filtrate was placed in a 2-ml sample bottle and GC–MS analysis was performed. Dao-Hua-Xiang (DHX) is a widely cultivated aromatic rice cultivar in China and was used as an aromatic rice standard.
2.6 Measurement of chalkiness degree
Chalkiness degree was measured using a grain appearance analyzer instrument (Model SC-E, Wanshen Ltd.) that automatically calculated the chalkiness degree by scanning the rice grains.
2.7 Measurement of gel consistency
Gel consistency was measured according to the previous study by Fu et al. (2023) and Vandeputte et al. (2003). Sample was ground into powder and passed through a 100-mesh sieve for subsequent experiments. 0.1 g of rice flour was placed in a test tube and 0.2 mL 0.025% thymol and 2 mL 0.2 mol/L KOH were added. The reaction tube was boiled for 8 min and was allowed to cool for 20 min. At room temperature (25°C), the reaction tube was horizontally placed with a graph paper behind to measure the colloid length obtained after 1 h incubation.
2.8 Determination of total protein content
Protein content was determined according to the previous study by Lynch et al. (1999) and Yang et al. (2019). Sample was ground into powder and 2 g of rice flour was placed in a 250-mL test tube. Two pieces of Celt catalyst plate(4.5 g K2SO4 and 0.5 g CuSO4·5H2O) and 12 mL concentrated sulfuric acid were added and the tube was gently shaken and digested for 90 min. After distillation, the Kjeldahl instrument (Hanon Advanced Technology Group Co., Ltd) was used for determination.
2.9 Determination of amylose content
Y: amylose content; a: intercept of standard curve; b: slope of standard curve; X: absorbance value of sample.
2.10 Statistical analysis
All the data with three replications were processed and analyzed by Student's t-test, where the difference was significant with P < 0.05 (*) and very significant with P < 0.01 (**).
3 RESULTS
3.1 CRISPR/Cas9-induced mutagenesis of OsBADH2
To create OsBADH2-mutated transgenic lines of SH313, a non-aromatic restorer line, a CRISPR/Cas9 vector targeting the second exon of OsBADH2 was introduced into SH313. Eleven independent knock-out lines (KO) were obtained (Figure 1A) and three (KO626, KO627 and KO628) were used for target site sequencing (Figure 1B). Chromatograms revealed two bases deletion were present in KO626, one base insertion in KO627and one base deletion in KO628 (Figure 1C).
3.2 KO lines of OsBADH2 exhibited the presence of 2-AP contents
The positive transgenic lines were tested for 2-AP contents. The KOH leaching method was used to confirm the aroma in KO lines of OsBADH2. We conducted aroma testing on all 11 lines, and the results indicated that 10 of the lines exhibited a significant aroma, while one line did not display aromatic characteristics (Table S1). 2-AP is a key volatile compound conferring aroma in rice cultivars. We determined the contents of 2-AP using GC–MS and results indicated that the grains of the KO lines had higher 2-AP content than the wild-type SH313. The 2-AP content of KO626, KO627 and KO628 were 28.4, 22.3 and 32.3 times higher, respectively, than that of SH313 (Figure 2). Dao-Hua-Xiang (DHX) is an aromatic japonica conventional rice cultivar planted on a large scale in China. The 2-AP contents of KO626 and KO628 were comparable to that of DHX.
We then evaluated whether the knock-out of OsBADH2 affects agronomic and quality traits. The KO lines showed a significant reduction in plant height, seed setting rate and 1000-grain weight compared to SH313 (Figure 3 A-G). However, the KO lines displayed no significant changes in the degree of chalkiness, gel consistency and protein content. In contrast, the amylose content was significantly increased (Figure 3 H-K). These results revealed that the editing of OsBADH2 can confer aroma to rice but simultaneously causes a significant negative effect on agronomic traits.
3.3 The knock-out of OsBADH2 enhanced hybrid grain aroma without compromising growth
As KO line of OsBADH2 showed a negative impact on agronomic traits, we explored whether their utilization in hybrid breeding has potential and whether heterotic loci can offset these negative impacts. To study this, KO lines (KO626、KO627 and KO628) were crossed with a non-aromatic three-line hybrid rice male sterile line Rong-7-A, producing R7YKO626, R7YKO627 and R7YKO626 (Figure 4A, B). The hybrid Rong-7-You-313 (R7Y313) was used as a control. The grain 2-AP content of R7YKO626, R7YKO627 and R7YKO628 was 3.1 times, 2.8 times and 2.7 times, respectively increased compared to R7Y313 (Figure 4C). However, compared to R7Y313, plant height, number of tillers, seed setting rate and 1000-grain weight of R7YKO626, R7YKO627 and R7YKO628 were not significantly changed (Figure 4 D-J). Consistent with the results of the KO lines, R7Y313, R7YKO626, R7YKO627 and R7YKO628 exhibited no significant changes in the degree of chalkiness and gel consistency, but the protein content of R7YKO626 was decreased significantly (Figure 4K-M). However, the amylose content of R7YKO626, R7YKO627 and R7YKO628 were significantly increased (Figure 4N). These results indicated that OsBADH2 mutations in hybrids could confer aroma and simultaneously avoid the negative growth impacts in a three-line hybrid system.
3.4 The young spike stalk base has the highest 2-AP content and can be used as a marker for aroma identification under field conditions
The marker-assisted selection or the determination of 2-AP content of aromatic phenotype is usually identified in the laboratory, which is cumbersome and costly. Therefore, an easy and practical identification method applicable under field conditions for aromatic phenotype would be desirable. To clarify the distribution of 2-AP content in rice among different tissues, we quantified the 2-AP in different tissues of the aromatic DHX cultivar at the booting stage. The result showed that the 2-AP content in the young spike stalk base was the highest, while it was lowest in 10–15 cm and 20–25 cm panicle (Figure 5). The 2-AP content of the young spike stalk base was 2.3 and 10.5 times higher than in the leaf blade and 20–25 cm panicle, respectively. This data indicated that chewing a young stalk base at the booting stage can be an ideal tissue for aroma identification under field conditions.
4 DISCUSSION
Aromatic jasmine and Indian basmati cultivars are famous in the world for their pleasant aroma (Hori et al., 1994; Kohji et al., 1992). Molecular studies have revealed that aroma is a complex trait and is mainly controlled by OsBADH2 (Chen et al., 2008; Hui et al., 2022; Zhang et al., 2023). OsBADH2 is involved in the conversion of betaine aldehyde into a 2-AP in rice grains, which produces a popcorn or nutty aroma. Studies have revealed that aromatic cultivars contain a higher quantity of volatile compounds, including 2-AP, and a mutation in OsBADH2 decreases its corresponding betaine aldehyde dehydrogenase enzyme activity and leads to an increase of the aroma due to the accumulation of 2-AP content (Chen et al., 2008; Niu et al., 2008). These studies paved a way to breed new aromatic cultivars by editing OsBADH2. SH313 is a three-line hybrid restorer line with a good combining ability with multiple CMS A-lines; however, these hybrids are non-aromatic. The addition of aroma in hybrid cultivars will increase their market price. To create the aromatic SH313, we edited OsBADH2 using the CRISPR/Cas9 system. Three independent KO lines were produced with significant aroma but their grain yield-related traits were significantly decreased (Figure 3). These results were inconsistent with Hui et al. (2022), who reported that the knock-out lines of OsBADH2 in a japonica cv. Ning-Jing-1 displayed an increase of 1000-grain weight. However, Hui et al. (2022) reported that the KO line of OsBADH2 in an indica cv. Huang-Hua-Zhan showed a decrease of 1000-grain weight. This paradox can be attributed to the difference in knock-out sites and genetic backgrounds of cultivars used for transformations. However, the real reason is still unknown.
The researchers have manipulated the expression of OsBADH2 to regulate the contents of 2-AP in conventional cultivars. However, its negative impact on yield-related traits was still a bottleneck. These cultivars have some common defects, such as low yield, long growth period, and are more prone to lodging. This has limited their large-scale application and promotion of aroma rice in breeding. Hybrids have been reported to mask the deleterious effects due to their heterotic loci and genetically diverse parents exhibit superior traits compared to their parents. The creation of aromatic hybrid cultivars without compromising growth is one of the objectives of breeders. In the current study, we utilized aromatic KO lines of OsBADH2 and crossed with a non-aromatic three-line male sterile line Rong-7-A. Compared with control R7Y313, the grain 2-AP content of R7YKO626, R7YKO627 and R7YKO628 were significantly increased and their yield-related traits were not significantly affected (Figure 4). In contrast to the KO lines of OsBADH2 (Figure 3), hybrid combinations R7YKO626, R7YKO627 and R7YKO628 did not exhibit a reduction in yield-related traits (Figure 4). Aroma is a complex genetically controlled trait regulated by a recessive trait controlled by OsBADH2 in rice. Hybrid grains are derived from F1 plants from two different parents, which masks the deleterious effects arising from one parental allele. That is why the grain 2-AP content of R7YKO626, R7YKO627 and R7YKO628 was higher than that of the control (Figure 4C) but relatively lower than the KO lines of OsBADH2 (Figure 2). Our results suggested the use of KO lines to cross with locally cultivated non-aromatic cultivars to introduce aroma through a breeding strategy.
On top of breeding for aroma improvement, it is also necessary to have a simple and cost-effective way of assessing aroma on the field. At present, various methods are employed to evaluate aroma e.g., sensory evaluation, Gas Chromatography (GC), and electronic nose. Previously chewing method (Lander et al., 1987), leaching method (Scod et al., 1980), and determination of 2-AP content (Niu et al., 2008; Hui et al., 2022; Zhang et al., 2023) have been employed in different studies to evaluate aroma. The sensory evaluation is easy and less accurate, as only a trained panel of individuals with a good sense of smell assess the aroma of cooked rice. Their opinion may vary depending on their taste buds, eating habits and smelling characteristics. The second GC method is relatively accurate but it is not suitable for operations under field conditions. The third method is the most accurate and capable of detecting volatile compounds with greater efficiency but its cost is relatively higher. In the current study, we measured the 2-AP content of various tissues of aroma rice cultivar at the booting stage and found that the young stalk base has the highest 2-AP and can be an ideal tissue for aromatic identification by a simple chewing method. Booting stage was used as it is an easily identifiable stage.
5 CONCLUSIONS
In the current study, we created KO lines of OsBADH2 using the CRISPR/Cas9 in hybrid rice restorer line SH313. The phenotypic analysis revealed that KO has a significantly increased content of 2-AP, but other yield-related traits were significantly decreased. The negative association of aroma with agronomic traits is a bottleneck in the practical utilization of conventional cultivars. In the current study, KO lines were crossed with a non-aromatic three-hybrid male sterile line Rong-7-A. Results showed that the F1 hybrid of the KO lines exhibited the presence of aroma without compromising yield-related traits. These results show that the introduction of aroma in hybrid cultivars rather than conventional cultivars will mask the deleterious effects on yield-related due to the presence of heterotic loci. In addition, our analyses indicate that the young spike stalk base can be used for chewing for aromatic identification at the booting stage under field conditions. Taken together, our study provided a practical example of a rapid improvement of aroma in hybrids and its application in heterosis, as well as a potential on-field test of the aromatic power of the new cultivars.
AUTHOR CONTRIBUTIONS
X.W and Y.L (Yongxiang Liao) and designed and approved the project; M.L., Z.X and J.X performed experiments and H.W., A.A., H.Z (Hao Zhou)., K.D and D.X helped in data analysis. J.X., X.Y., Y.L (Yong Li)., Y.L (Yingxiu Liao)., A.W., D.G., G.Z., and Z.W assisted in the experiments. P.X., H.Z (Hongyu Zhang)., H.W., and X.C assisted in funding acquisition; Y.L (Yongxiang Liao) and A.A. wrote the original manuscript; AA., Y.L (Yongxiang Liao)., and M.L performed data curation. All authors have read and agreed to the published version of the manuscript.
FUNDING INFORMATION
We acknowledge grants-in-aid from the program of the Science and Technology Department of Sichuan Province (2022ZDZX0016, 2022JDRC0111), the Chengdu Science and Technology Bureau (2022-YF09-00036-SN, 2023-YF08-00005-SN), and the Science and Technology Bureau of Changde, Hunan Province (Chang Ke Han 2021–59). The funding bodies played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Open Research
DATA AVAILABILITY STATEMENT
The data that supports the findings of this study are available in the supplementary material of this article.
