Method for increasing drought tolerance by accumulation of plant osmoregulators

Abstract

The application discloses a method for increasing drought adaptability of plant osmotic regulation substance accumulation, and belongs to the field of plant genetic engineering and drought-resistant molecular breeding. First, a plant to be transformed is combined with 13C-glutamic acid and 15N-choline double isotope labels to quantize carbon and nitrogen flow redistribution of proline synthesis pathways and glycine betaine synthesis pathways in leaf cells and vascular bundle companion cells, and to identify rate-limiting enzyme nodes. Second, a multi-cistronic plant expression vector is constructed. A first expression box is driven by a leaf-specific promoter and a drought response cis-element composite promoter to drive P5CS. A second expression box is driven by a companion cell-specific promoter and a drought response cis-element composite promoter to drive BADH. Positive transformants are obtained through Agrobacterium mediation or gene gun transformation and molecular screening. The expression quantity of the positive transformants is increased by 5 to 50 times under stress and falls back to a baseline within 72 hours of rehydration. Root activity is significantly improved, and yield is preserved.

Description

Technical Field

[0001] This invention belongs to the field of plant genetic engineering and drought-resistant molecular breeding technology. Specifically, it relates to a method for enhancing plant drought adaptability by quantifying and identifying the redistribution of stable isotope metabolic flux, driving the differential targeted expression of key enzymes in the proline synthesis pathway and glycine betaine synthesis pathway in mesophyll cells and vascular bundle companion cells, and superimposing drought dynamic regulation. Background Technology

[0002] One of the core physiological strategies plants employ to cope with drought stress is to reduce osmotic potential through the active accumulation of compatible solutes within cells, thereby maintaining cell turgor pressure, stabilizing the conformation of biomolecules, and protecting the integrity of membrane systems. Among the identified compatible solutes, proline and glycine betaine have become core targets for drought-resistant molecular breeding due to their large accumulation range, low metabolic toxicity, and wide presence in various plants. The rate-limiting enzyme for proline synthesis is Δ1-pyrroline-5-carboxylic acid synthase (P5CS), which catalyzes the conversion of glutamate to Δ1-pyrroline-5-carboxylic acid via the γ-glutamyl phosphate intermediate, followed by reduction to proline via P5CR. The rate-limiting enzyme for glycine betaine synthesis is betaine aldehyde dehydrogenase (BADH), which catalyzes the oxidation of betaine aldehyde to glycine betaine using betaine aldehyde as a substrate. The precursor betaine aldehyde is generated from choline via choline monooxygenase.

[0003] Several transgenic drought-resistant technologies have been publicly disclosed both domestically and internationally, focusing on the two aforementioned pathways. Chinese invention patent application publication number CN101701210A, entitled "A Plant Drought-Related Protein P5CS and Its Encoding Gene and Application," discloses the application of a P5CS protein derived from Asian cotton and its encoding gene in improving plant drought resistance. This technology employs a single P5CS gene overexpression pathway. However, simply enhancing proline synthesis cannot cover the multidimensional needs of cellular osmotic regulation. Under severe stress conditions, the absence of glycine betaine leads to insufficient osmotic protection, limiting the improvement in drought resistance.

[0004] Chinese invention patent application publication number CN101321871A, entitled "Transgenic Plant with Enhanced Drought Tolerance," discloses a transgenic crop plant whose genome integrates a chimeric gene containing transcriptional regulatory sequences that can be active within plant cells, and teaches the concepts of inducible promoters and tissue-specific promoters. However, the description of tissue specificity in this scheme stops at the coarse-grained level of expression in specific types of tissues, without addressing the differentiated targeting design of different osmotic regulatory pathways based on metabolic network topology in different tissues, nor does it solve the problem of carbon and nitrogen precursor competition under dual-pathway concurrent initiation conditions.

[0005] Chinese invention patent application publication number CN107090453A, entitled "Compositions, Organisms, Systems, and Methods for Expressing Gene Products in Plants," discloses a plant expression system containing multiple expression cassettes. Its main focus is on avoiding gene silencing caused by homologous duplication through heterologous promoter combinations. However, this approach does not address the specific engineering goal of dual-pathway accumulation of osmotic regulatory substances, nor does it disclose a technical route that uses metabolic flux quantification analysis as a prerequisite for promoter and tissue-targeted design.

[0006] After a comprehensive review of existing technologies, the inventors identified three key issues that have not yet been systematically addressed: First, proline synthesis uses glutamate as a common carbon and nitrogen precursor, while glycine betaine synthesis depends on choline and requires the consumption of S-adenosylmethionine. When these two pathways are initiated concurrently at high intensity within the same cell, there is competition for precursor pools and competition for reducing power, leading to a metabolic burden paradox where indoor drought resistance and field yield cannot be achieved simultaneously. Second, existing schemes generally use constitutive promoters or single inducible promoters, which result in continuous metabolic consumption even during non-stress periods, leading to slowed growth and decreased dry matter yield. Third, existing technologies do not incorporate stable isotope metabolic flux analysis as a pre-design step. The selection of promoters and tissue targets relies on experience rather than metabolic network mechanisms, thus failing to accurately identify rate-limiting nodes and precursor pool competition intensity for each pathway. This results in many engineered lines having theoretically synergistic dual-pathway capabilities but with actual drought resistance performance that does not meet expectations. Furthermore, another hidden problem commonly observed in the field evaluation of existing multi-gene drought-resistance engineering schemes is the metabolic lag effect during the rehydration recovery period. Driven by constitutive promoters, the abundance of synthase mRNA remains high for an extended period after stress relief, and the synthesis reaction continues, leading to unnecessary locking of photosynthetic products. This problem is difficult to detect in traditional indoor short-cycle stress evaluations, but it accumulates into significant yield losses under multiple rounds of alternating wet and dry conditions in the field. Therefore, providing a plant osmotic regulation accumulation and drought adaptation enhancement method driven by quantitative metabolic flux redistribution, employing dual-pathway differential tissue targeting, and possessing dynamic regulation capabilities for drought response, is of great significance for overcoming the technical bottlenecks of existing drought-resistance genetic engineering. Summary of the Invention

[0007] To address the three problems of existing technologies—metabolic burden caused by competition between proline-glycine-betaine dual-pathway carbon and nitrogen precursors, field yield reduction caused by continuous consumption of constitutive expression, and lack of metabolic network mechanism support for promoter selection—this invention aims to provide a method for enhancing plant osmotic regulation and drought adaptability by using a multicistronic dual-expression cassette vector to achieve differentiated targeted expression of P5CS and BADH in mesophyll cells and vascular bundle companion cells based on quantitative analysis of metabolic flux redistribution, and dynamically activating on demand through drought-responsive cis-regulatory elements.

[0008] To achieve the aforementioned objectives, the present invention provides the following technical solution. First, by administering a dual isotope drug of 13C-labeled glutamate and 15N-labeled choline to the plant to be modified, combined with microdissection of mesophyll cells and vascular bundle companion cells, the redistribution of carbon and nitrogen fluxes in the proline and glycine-betaine synthesis pathways in these two types of tissues is quantitatively analyzed. The rate-limiting enzyme nodes and the intensity of cross-pathway precursor competition in these two pathways are identified, providing a rational design basis at the metabolic network level for tissue targeting. Second, a polycistronic plant expression vector comprising a first expression cassette and a second expression cassette is constructed. The first expression cassette is driven by a first composite promoter consisting of a mesophyll cell-specific promoter and a drought-responsive cis-regulatory element, which encodes a P5CS sequence with proline synthase activity. The second expression cassette is driven by a second composite promoter consisting of a vascular bundle companion cell-specific promoter and a drought-responsive cis-regulatory element, which encodes a BADH sequence with betaine aldehyde dehydrogenase activity. Next, the polycistronic plant expression vector was transformed into the plant to be modified using Agrobacterium-mediated transformation or gene gun method. Positive transformants were obtained by selection marker screening and molecular identification. The positive transformants met the following criteria: the expression levels of P5CS in mesophyll cells and BADH in vascular bundle companion cells increased by 5 to 50 times under drought stress compared with normal water conditions, and returned to the pre-stress baseline level within 72 h after rehydration.

[0009] The beneficial effects of this invention are as follows: First, by quantitatively analyzing the competition intensity between rate-limiting enzymes and precursors through metabolic flux redistribution, it overcomes the limitations of existing schemes that rely on empirical selection of promoters and target genes, ensuring synergy rather than competition between the two pathways from the perspective of metabolic network mechanisms. Second, through differentiated targeting of mesophyll cells and vascular bundle companion cells, P5CS is placed in photosynthetically active tissues with abundant carbon sources, and BADH is placed in nitrogen metabolism hub tissues, matching the optimal ecological niche distribution of precursor supply for both pathways and cutting off competition with intracellular precursor pools. Third, through the dynamic closed loop of drought-responsive cis-regulatory elements and the decline of mRNA homeostasis under rehydration conditions, osmotic regulation capacity is supplied on demand, avoiding continuous metabolic consumption during non-stress periods and preserving field productivity. After years of multi-location field drought resistance evaluation, the engineered plants showed that under drought stress, the relative water content of leaves was maintained at 18% higher than that of wild types, the relative conductivity of cell membranes decreased by 25%, and the root vigor retention rate was significantly improved, providing a systematic approach to enhance multi-pathway osmotic regulation for maintaining cellular water homeostasis and yield preservation in crops under drought stress. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the T-DNA region structure of the polycistronic dual expression cassette plant expression vector of the present invention.

[0011] Figure 2 This is a schematic diagram of the workflow for stable isotope metabolic flux redistribution analysis in this invention.

[0012] Figure 3 A heatmap showing the differential distribution of free proline and glycine betaine content in mesophyll cells and vascular bundle companion cells of engineered plants.

[0013] Figure 4 The graph shows the change in relative leaf water content over time for engineered and wild-type plants under drought stress.

[0014] Figure 5 A bar graph comparing the relative electrical conductivity of cell membranes in engineered plants and wild-type plants under drought stress.

[0015] Figure 6 This is a kinetic curve showing the decline in the relative abundance of P5CS and BADH transcripts during the recovery process after the relief of drought stress. Detailed Implementation

[0016] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. These embodiments are used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention. The materials, reagents and instruments used in the embodiments are all conventionally available in the art. Unless otherwise specified in the experimental methods, conventional conditions or the manufacturer's recommended conditions shall be followed.

[0017] Example 1: Implementation of the method of the present invention in rice

[0018] This embodiment uses Zhonghua 11 (Oryza sativa L. ssp. japonica cv. Zhonghua 11) as the plant to be modified, and fully presents the entire process of the method of the present invention from metabolic flux analysis to drought resistance evaluation of positive transformants.

[0019] Step 1: Quantitative analysis of metabolic flux redistribution. For example... Figure 2 The stable isotope metabolic flux redistribution analysis workflow of this invention, as shown, includes five stages in sequence: dual isotope drug delivery, laser microdissection of tissue, high-resolution LC-MS / MS isotope molecular quantification, Wasserstein flux redistribution metric calculation, and stratified sensitivity rate-limiting enzyme node identification. This step is performed according to... Figure 2The procedure was performed as shown. Four-leaf stage rice seedlings with uniform growth were selected and divided into a dual-isotope labeled group and an unlabeled control group, with 30 seedlings in each group. The labeled group was administered 13C fully labeled L-glutamic acid (99 atom% 13C, final concentration 2.0 mmol / L) and 15N methyl labeled choline chloride (98 atom% 15N, final concentration 1.5 mmol / L) simultaneously via root hydroponics. Twenty-four hours after administration, half of the plants were subjected to simulated osmotic stress with 25% PEG-6000, while the other half maintained normal water levels. Samples were taken at five time points: 0 h, 2 h, 6 h, 24 h, and 72 h after stress. Mesophyll cells and vascular bundle companion cells were precisely separated using a laser microdissection system, with each sample yielding approximately 15 μg dry weight. After extracting metabolites from samples using cold methanol-water (70:30 v / v), the samples were quantitatively analyzed by Q-Exactive high-resolution LC-MS / MS to detect the abundance of proline isotope species M+0 to M+5 and glycine betaine M+0 to M+3. A Waters ACQUITY UPLC BEH Amide column (2.1 × 100 mm, 1.7 μm) was used at 35°C. Mass spectrometry employed a positive / negative ion switching mode, a resolution of 70,000, and a scan range of m / z 70 to 1000. Based on the measured isotope species time-series data, the Wasserstein redistribution measure of carbon and nitrogen fluxes in the two tissues was calculated using a spatiotemporally resolved stable isotope flux redistribution algorithm. A hypergraph with the mesophyll-companion cell metabolic network as nodes was constructed and a sensitivity layer structure was assigned. Rate-limiting enzyme nodes corresponding to non-zero isohomology classes were identified by calculating the cohomology on the layer. The analysis showed that under stress conditions, the flux of 13C to proline in mesophyll cells increased from the baseline of 13.2 μmol / (g·h) to 57.8 μmol / (g·h), and the flux of 15N to glycine betaine in vascular bundle companion cells increased from the baseline of 4.5 μmol / (g·h) to 22.3 μmol / (g·h). In contrast, the increase in the reverse test group (the flux of choline synthesis upstream of BADH in mesophyll and the flux of glutamate synthesis upstream of P5CS in companion cells) was only 23% and 18% of that in the forward test group, respectively. This clearly indicates that P5CS is the rate-limiting node of the proline pathway and its optimal ecological niche is in mesophyll cells, while BADH is the rate-limiting node of the glycine betaine pathway and its optimal ecological niche is in vascular bundle companion cells. Both pathways have independent precursor supply bases in their respective tissues, and the cross-pathway precursor competition intensity is significant, providing a rational design basis at the metabolic network level for subsequent differentiated targeted design.

[0020] Step 2: Construction of the dual-complex promoter and polycistronic vector. The T-DNA region structure of the polycistronic dual-expression cassette plant expression vector pCAMBIA-P5CS-BADH-dual constructed in this step is as follows: Figure 1As shown, it sequentially includes nine core elements: left boundary, first complex promoter, P5CS coding sequence, NOS terminator, second complex promoter, BADH coding sequence, 35S terminator, hygromycin resistance selection marker, and right boundary. The first composite promoter, P_meso-dre, uses the Arabidopsis thaliana rbcS-1A promoter (derived from Arabidopsis thaliana AT1G67090, sequence length 1100 bp) as its backbone. In the -800 bp to -600 bp region, a composite cis-module consisting of three copies of DRE cis-elements (shared sequence TACCGACAT) and two copies of ABRE cis-elements (shared sequence PyACGTGG) is inserted via three-segment homologous recombination. The total length is 1207 bp. Transient expression system GUS staining confirmed that its mesophyll specificity remained unchanged compared with the native rbcS promoter, and its drought response intensity was increased by 4 to 6 times compared with the native RD29A promoter. This indicates that the cis-module insertion site and copy number design enable this promoter to have both tissue-specific backbone and stress-induced response functions. The second composite promoter, P_phloem-dre, uses the Arabidopsis SUC2 promoter (derived from AT1G22710, 2000 bp in length) as its backbone. A cis-module consisting of two copies of DRE elements and three copies of ABRE elements is inserted in the -500 bp to -300 bp region, with a total length of 2173 bp. Transient expression verified that it maintains companion cell specificity and the stress response initiation delay is shortened by about 2 h compared to P_meso-dre, which corresponds to the faster nitrogen redistribution response kinetics required for companion cells as phloem loading hubs.After the two complex promoters were obtained through whole-genome synthesis, they were respectively coupled with the lupin LaP5CS coding sequence (original sequence from Lupinus angustifolius, GenBank accession number AF288867, full length 2181 bp, F129A mutation achieved by replacing the phenylalanine codon TTC at position 129 with the alanine codon GCC, CAI increased from 0.62 to 0.89 after rice codon optimization, GC content controlled in the range of 53% to 57%) and the sugarcane ScBADH coding sequence (original sequence from Saccharum officinarum, GenBank accession number AY573247, full length 1521 bp, optimized CAI) and the sugarcane ScBADH coding sequence (original sequence from Saccharum officinarum, GenBank accession number AY573247, full length 1521 bp, optimized CAI). 0.91) The first and second expression cassettes were constructed via Gateway recombination reaction and then sequentially cloned into the T-DNA region of the pCAMBIA1305.1 binary vector. The resulting polycistronic plant expression vector was named pCAMBIA-P5CS-BADH-dual, with a total length of 13.8 kb. The T-DNA region was configured as follows: LB—P_meso-dre—LaP5CS(F129A)—NOS terminator—P_phloem-dre—ScBADH—35S terminator—HPT—RB. The heterologous terminator configuration between the two expression cassettes balances transcriptional independence with avoiding gene silencing caused by homologous recombination.

[0021] Step 3: Agrobacterium-mediated transformation of rice and screening of positive transformants. The constructed expression vector was transformed into Agrobacterium EHA105 competent cells using a freeze-thaw method. Positive Agrobacterium single colonies were obtained through screening with 50 mg / L kanamycin. Callus induced from mature embryos of rice Zhonghua 11 was used as the recipient, and standard Agrobacterium-mediated infection was performed at an OD600 concentration of 0.3. Co-culture was conducted at 25°C for 3 days. The co-cultured callus was then transferred to a selection medium containing 50 mg / L hygromycin for three rounds of screening, each round lasting 14 days. Resistant callus was then induced to differentiate and root before being transplanted into a light-treated culture room. The integration of the target gene was confirmed by leaf DNA extraction from T0 generation plants and PCR amplification using primers specific to the LaP5CS coding region (forward 5'-ATGGAGTTGTCTTACATCGC-3', reverse 5'-TTAGCCATCTCGTTCACCAG-3') and primers specific to the ScBADH coding region (forward 5'-ATGGCGTTCCCAATCCCTGC-3', reverse 5'-TTAGCTGGGACATGCGACTT-3'). Further qRT-PCR with rice OsACT1 as an internal reference gene was used to confirm the expression responsiveness of the two expression cassettes under stress-induced conditions. Simultaneously, Southern blot was used to confirm T-DNA single-copy insertion events to rule out the risk of gene silencing due to multiple copies. A total of 25 independent transformation events were obtained through the above three rounds of screening. Among them, 12 events were single copy insertions and met the stress induction fold window of 5 to 50 times as defined in claim 1. Event TL-7, with a mesophyll P5CS induction fold of 28 times, a companion cell BADH induction fold of 24 times, and a return to 1.2 times the baseline after 72 h of rehydration, was selected for subsequent field evaluation.

[0022] Step 4: Field evaluation of drought resistance phenotype. The TL-7 event was used to obtain the T3 generation homozygous line through three consecutive generations of self-pollination. Multi-site, multi-year field drought resistance assessments were conducted in three pilot areas in North China, Central China, and Northwest China. Each pilot area had four replicate plots (one engineered line and one wild-type control), with a plot area of ​​20 m². A randomized block design was used, and soil fertility and routine agronomic management followed local conventional field production standards. Key drought resistance physiological indicators were measured after 14 consecutive days of artificially controlled drought stress from the tillering to the heading stage. For example... Figure 4 The curves showing the change in relative leaf water content over time indicate that the relative leaf water content of the engineered strain remained at 78.5% on day 7 of stress, while that of the wild type decreased to 60.2%, with a maintenance rate 18.3 percentage points higher than that of the wild type. Furthermore, at all measurement time points throughout the stress period, the curves of the engineered strain were significantly higher than those of the wild type, and the difference between the two curves showed a stable widening trend with increasing stress duration. Figure 5The bar chart showing the relative conductivity of cell membranes indicates that the relative conductivity of leaf cell membranes in the engineered line was 22.4%, compared to 47.5% in the wild type, a decrease of 25.1 percentage points. This difference widened further on day 14 of stress, reflecting that the engineered line could effectively maintain membrane system integrity even under severe stress. The root triphenyltetrazolium chloride (TTC) reducing activity was 0.42 mg / (g·h) in the engineered line, compared to 0.18 mg / (g·h) in the wild type, showing a significantly improved retention rate. The free proline content in leaves was 2.6 times higher in the engineered line than in the wild-type stress control, the free glycine betaine content was 3.8 times higher, and the malondialdehyde (MDA) content was 42% lower. Superoxide dismutase (SOD) and peroxidase (POD) activities were 35% and 28% higher, respectively, than in the wild-type stress control, indicating a synergistic improvement in osmotic regulation and reactive oxygen species scavenging systems. After the drought ended and water was refilled for 72 hours, the qRT-PCR test results were as follows: Figure 6 The kinetic curves of the relative abundance decline of P5CS and BADH transcripts during the rehydration process show that the relative abundance of both genes exhibits an exponential decline trend, with the fastest decline rate occurring between 6 and 12 hours after rehydration. Within 72 hours, both declined to within 1.3 times the baseline level, confirming the effectiveness of the dynamic decline closure loop. Photosynthetic gas exchange assays showed that after 72 hours of rehydration, the net photosynthetic rate of the engineered lines recovered to 95% of the wild-type normal water control, while the genotype expression scheme in Comparative Example 1 only recovered to 74%, reflecting the supporting role of the on-demand supply mechanism of this invention in photosynthetic system repair. In a multi-year, multi-location, multi-year repeated trial, the T3 generation grain yield of the engineered lines increased by 24.8% compared to the wild-type stress control, and decreased by only 5.7% compared to the wild-type normal water control. The standard deviation between each test site did not exceed 3.2 percentage points, indicating that the method of this invention has good phenotypic stability and achieves the dual goals of improving drought resistance and maintaining yield under different geographical conditions.

[0023] Example 2: Implementation of the method of the present invention in wheat

[0024] This embodiment uses winter wheat Zhongmai 175 (Triticum aestivum L. cv. Zhongmai 175) as the plant to be modified, illustrating the transferability and parameter variation scheme of the method of the present invention in gramineous food crops.

[0025] The metabolic flux redistribution analysis followed the isotope administration and tissue microdissection procedure of Example 1, with the difference being that the administration concentrations were adjusted to 2.5 mmol / L of 13C-glutamate and 1.8 mmol / L of 15N-choline to adapt to wheat absorption kinetics, and osmotic stress was achieved by replacing 25% with 20% PEG-6000 to match the tolerance window of wheat seedlings. The analysis results showed that the baseline P5CS flux in wheat mesophyll cells was 15.6 μmol / (g·h), which increased to 68.4 μmol / (g·h) after stress; the baseline BADH flux in vascular bundle companion cells was 5.8 μmol / (g·h), which increased to 26.9 μmol / (g·h) after stress. The rate-limiting node was consistent with that in rice, confirming the consistency of the metabolic flux analysis-driven design method of this invention in grasses. It is worth noting that wheat, as a C3 drought-tolerant crop, has a slightly larger S-adenosylmethionine pool in its vascular bundle companion cells than rice. This metabolic background provides a more abundant methyl donor for choline synthesis upstream of BADH. Therefore, in the subsequent design of composite promoters, a moderate adjustment of the ABRE copy number in companion cells can efficiently drive BADH expression without increasing the stress-induced intensity of DRE. This demonstrates the sensitivity and guidance of the metabolic flux analysis method of this invention to the differences in metabolic backgrounds of different species.

[0026] The main difference between the vector construction scheme and Example 1 is that: the first composite promoter backbone is replaced with the wheat homologous photosystem II oxygenation complex subunit gene TaLhcb promoter (sequence length 1050 bp), which retains mesophyll specificity while improving the initiation strength; the second composite promoter backbone is replaced with the wheat phloem loaded sucrose transport protein gene TaSUT1 promoter (sequence length 1850 bp) to provide companion cell specificity; the cis-module combinations of the two promoters are four copies of DRE plus two copies of ABRE and two copies of DRE plus four copies of ABRE, respectively, in which the copy number of the companion cell composite promoter ABRE is increased to adapt to the slightly weaker ABA sensitivity of wheat companion cells compared with rice. The P5CS gene was derived from the lupin LaP5CS(F129A) mutant of Example 1, optimized with wheat codons. The BADH gene was replaced with spinach SoBADH (GenBank accession number X58463, chosen because spinach SoBADH has a lower Km for betaine aldehyde, suitable for the relatively limited precursor concentration in wheat companion cells). After wheat codon optimization, the CAI reached 0.88. The vector backbone retained pCAMBIA1305.1, the selection marker retained HPT, and the T-DNA region structural sequence remained consistent with Example 1. The two expression cassettes were separated by a 35S terminator to avoid homologous duplication with the NOS terminator at the end of the first expression cassette. Transient expression system validation showed that the GUS staining intensity of the TaLhcb scaffold promoter in wheat mesophyll was about 1.7 times higher than that of the Arabidopsis rbcS-1A scaffold in wheat heterologous expression. The staining specificity of the TaSUT1 scaffold promoter in wheat companion cells was also significantly better than that of the Arabidopsis SUC2 heterologous promoter. This indicates that species-origin selection of promoter scaffolds is of great significance for the dual optimization of expression efficiency and tissue specificity.

[0027] Transformation was performed using the wheat embryo gene gun bombardment method at a bombardment pressure of 1100 psi and a target distance of 9 cm. The gene gun used was a PDS-1000 / He system with a gold powder particle diameter of 1.0 μm. After three rounds of screening with hygromycin 40 mg / L, 21 T0 generation plants were obtained, of which 9 were single-copy insertions that met the stress induction fold window and rehydration regression kinetic window defined in this invention. The TW-3 event, characterized by a mesophyll P5CS induction fold of 31-fold, a companion cell BADH induction fold of 26-fold, and a regression to 1.4-fold of baseline after 72 h of rehydration, was selected for field evaluation. Field identification was replicated for two years in two pilot areas in Zhengzhou, Henan and Yangling, Shaanxi, in the Huang-Huai wheat region. A continuous drought stress treatment of 21 days was applied from the jointing stage to the heading stage, with 4 replicate plots per treatment and a plot area of ​​15 m2. The results showed that under drought stress, the TW-3 line maintained a 19.6 percentage point higher leaf relative water content, a 24.3 percentage point lower leaf membrane conductivity, a 32% higher flag leaf photosynthetic rate, and a 18.5% higher thousand-grain weight compared to the wild-type stress control, while only decreasing by 4.8% compared to the wild-type normal water control. This indicates the transferability and stability of the drought resistance effect of the method of this invention in wheat and other grass crops. Further isotope metabolic flux tracking showed that at the peak stress time, the flux of 13C-glutamic acid to proline in the mesophyll reached 79.5 μmol / (g·h), and the flux of 15N-choline to glycine betaine in the paracellular matrix reached 31.2 μmol / (g·h), which is highly consistent with the flux redistribution pattern of rice TL-7, further confirming the transferability of the core mechanism of the spatiotemporal topology optimization of the metabolic network of this invention in different species.

[0028] Example 3: Implementation of the method of the present invention in soybeans

[0029] In this embodiment, the soybean variety Zhonghuang 13 (Glycine max L. Merr. cv. Zhonghuang 13) was used as the plant to be modified. The boundary values ​​of the parameter range defined in the claims of this invention were used to verify the effectiveness and robustness of the method under the parameter boundary conditions.

[0030] The isotope labeling concentrations were set at the lower boundary values ​​of the protocol: 1.5 mmol / L for 13C-glutamic acid and 1.0 mmol / L for 15N-choline, to investigate the effect of low-dose labeling on the sensitivity of flow measurement. Tissue microdissection and LC-MS / MS analysis methods were consistent with those in Example 1. The results showed that the signal-to-noise ratio (SNR) for proline M+5 molecules remained above 15 under low-dose isotope administration conditions, and the SNR for glycine betaine M+3 molecules remained above 12, meeting the accuracy requirements for flow quantification analysis. This indicates that the method of this invention can still be implemented under the lower boundary conditions of the labeling concentration.

[0031] The composite promoter design employed a cis-module copy number boundary combination: the first composite promoter used two copies of DRE plus one copy of ABRE (near the lower boundary), and the second composite promoter used one copy of DRE plus two copies of ABRE (near the lower boundary). The promoter backbones were the soybean mesophyll-specific GmRbcS2 promoter and the soybean companion cell-specific GmSUC1 promoter, respectively. The P5CS coding sequence used the derived variant described in the claims, specifically the F129A mutant of cowpea VuP5CS (GenBank accession number DQ641486), optimized with soybean codons; the BADH coding sequence used spinach SoBADH. Transformation was performed using the Agrobacterium EHA105-mediated soybean cotyledon node transformation system. After screening with 5 mg / L glufosinate, 15 T0 generation plants were obtained, 7 of which met the expression fold window defined in this invention, and the TG-5 event was selected for evaluation.

[0032] Regarding parameter window evaluation, in this embodiment, the P5CS enzyme activity in mesophyll cells of the engineered line TG-5 at the peak of drought stress was 0.85 U / mg·prot, and the BADH enzyme activity in companion cells was 0.98 U / mg·prot, with an enzyme activity ratio of 0.87, located at the lower boundary of the 0.8 to 2.5 window described in claims. The free proline concentration in mesophyll cells was 31 μmol / g fresh weight, and the free glycine-betaine concentration in companion cells was 12 μmol / g fresh weight, both located near the lower boundary of the window. Field evaluation showed that under this lower boundary parameter combination, the improvement in drought resistance (a 15.4 percentage point increase in leaf relative water content retention and a 21.8 percentage point decrease in membrane conductivity) was slightly lower than the central parameter values ​​of the rice scheme in Example 1, but still significantly better than the wild type, and the yield loss was only 3.2%, indicating that the method of the present invention still maintains practical drought resistance improvement and yield preservation capabilities under the lower boundary parameter conditions. Further adjustments to the cis-module copy number to the upper boundary combination (P_meso-dre using 4 copies of DRE plus 3 copies of ABRE, P_phloem-dre using 3 copies of DRE plus 5 copies of ABRE) yielded event TG-12, with a mesophyll P5CS induction fold of 48-fold and a companion cell BADH induction fold of 42-fold, both close to the 50-fold upper limit stated in the claims. The corresponding enzyme activity ratio was 2.31, mesophyll proline concentration was 115 μmol / g fresh weight, and companion cell glycine betaine concentration was 42 μmol / g fresh weight, all also near the upper boundary of the window. Field evaluation results for TG-12 showed a 19.8 percentage point increase in leaf relative water content retention, a 26.7 percentage point decrease in membrane conductivity, a 26.2% increase in yield compared to the wild-type stress control, and a 7.1% yield loss during the non-stress period, highly similar to the performance of the rice central parameter scheme in Example 1. The dual parameter verification at the upper and lower boundaries of this embodiment confirms the effectiveness of the 5-fold to 50-fold expression response window, the 0.8 to 2.5 enzyme activity ratio window, and the corresponding metabolite concentration window described in the claims in maintaining drought resistance at both the upper and lower boundaries. The parameter window settings have scientific basis and engineering robustness.

[0033] To fully demonstrate the necessity and synergistic effect of the core technical features of this invention, the following five comparative examples were designed. The plant materials, transformation methods, and field evaluation methods of the comparative examples are consistent with those of Example 1, except that the core features of this invention are targeted for destruction.

[0034] Comparative Example 1 (without time-sensitive dynamic switching, using a constitutive promoter). The first composite promoter P_meso-dre and the second composite promoter P_phloem-dre in Example 1 were replaced with the rice constitutive promoter Ubi1, that is, Ubi1 drives LaP5CS (F129A) and simultaneously drives ScBADH. The remaining steps were the same as in Example 1. Field evaluation showed that the comparative strains accumulated proline and glycine betaine at high levels during the non-stress period. The content of free proline in leaves was 12.5 times that of the wild type under normal conditions, and the content of glycine betaine was 28.3 times that of the wild type. This resulted in a 11.5% reduction in plant height, a 1.8-fold reduction in effective tiller number, and a 14.2% reduction in thousand-grain weight under normal water conditions compared to the wild type. Isotope labeling showed that the NADPH reducing power pool in the mesophyll cells of the comparative strains was 26% lower and the Calvin cycle flux was 18% lower than that of the wild type during the non-stress period, indicating that constitutive high-intensity expression caused continuous resource competition for photosynthetic structures. Although the relative water content of leaves increased by 13.2 percentage points under drought stress, the final grain yield only increased by 6.4% compared to the wild-type stress control, which was far lower than the 24.8% increase in Example 1. This confirms the key role of the drought dynamic regulation characteristics of the present invention in avoiding metabolic burden and maintaining field yield. More notably, after rehydration, the P5CS and BADH transcripts in this comparative strain remained at 82% of the stress peak level, with no significant decline within 72 hours. This contrasts sharply with the decline to 1.3 times the baseline in Example 1, demonstrating the irreplaceable role of the dynamic closed-loop mechanism of this invention in avoiding post-stress metabolic retardation.

[0035] Comparative Example 2 (de-tissue differential targeting, dual gene co-expression in mesophyll). The P_phloem-dre promoter in the second expression cassette of Example 1 was replaced with the same P_meso-dre as the first expression cassette, allowing both LaP5CS(F129A) and ScBADH to be expressed in mesophyll cells. The remaining steps were the same. Field evaluation showed that the relative water content retention rate of leaves in this comparative example was only 8.9 percentage points higher than the wild type, significantly lower than the 18.3 percentage points in Example 1. Isotope metabolic flux analysis showed that the glutamate precursor pool in mesophyll cells was rapidly depleted (decreased by 73%) at the peak of stress, the S-adenosylmethionine concentration decreased by 41% synchronously, the total chlorophyll content decreased by 28% compared to Example 1, and the net photosynthetic rate decreased by 22%. This confirms that precursor competition and chloroplast function inhibition caused by dual gene expression in the same tissue are the root cause of the decreased drought resistance, thus demonstrating the necessity of the differential targeting feature of this invention in relieving precursor competition. Further isotopic analysis of individual samples of the companion cells in this comparative example revealed that ScBADH expression in the companion cells was almost undetectable, and glycine betaine synthesis was completely absent. In contrast, the accumulation of glycine betaine in the companion cells in Example 1 contributed approximately 40% of the overall osmotic regulation gain, confirming the irreplaceable nature of tissue-differential targeting from the perspective of metabolic contribution quantification.

[0036] Comparative Example 3 (overexpression of only the single P5CS gene). The second expression cassette from Example 1 was completely removed, retaining only the LaP5CS (F129A) driven by the first expression cassette; the remaining steps were the same. This comparative example essentially degenerated into a single-gene scheme. Field evaluation showed that the single-gene line improved the relative leaf water content retention rate by only 9.6 percentage points and reduced membrane conductivity by only 12.8 percentage points, with drought resistance improvement approximately half that of Example 1. Further metabolomics analysis showed that the glycine betaine content of this line under severe stress was not significantly different from that of the wild type, and the betaine dimension of the osmotic protection system was completely absent, confirming the necessity of the dual-pathway synergy of this invention. As an extension of the verification, the inventors also constructed a single BADH overexpression control with only the second expression cassette. The results showed that the single BADH scheme increased the relative water content of leaves by only 7.8 percentage points, decreased membrane conductivity by only 9.6 percentage points, and increased yield by only 8.5% compared to the wild-type stress control. This was comparable to but slightly lower than the single P5CS scheme, indicating that both pathways have an effect ceiling when acting alone. The present invention can only break through this ceiling by achieving synergistic release of the two pathways through differentiated targeting. The two single-gene control experiments jointly confirmed the innovative metabolic contribution of the present invention in terms of synergistic dual-pathway metabolism in different directions.

[0037] Comparative Example 4 (Empirical Promoter Selection Without Metabolic Flux Analysis). This comparative example simulates the conventional empirical design of existing technologies, without performing the metabolic flux analysis in step one. It directly selects the rice Cab leaf chloroplast protein promoter to drive LaP5CS (F129A) based on literature reports, and the rice SBP sucrose-binding protein promoter to drive ScBADH based on literature reports. The cis-module combination is designed conventionally. Field evaluation shows that the relative water content retention rate of the leaves of this empirical scheme is increased by 11.2 percentage points, between Comparative Example 3 and Example 1. Retrospective metabolic flux analysis shows that the activity of the Cab promoter in the mesophyll is not significantly different from that of the rbcS backbone promoter. However, the SBP promoter is actually mainly active at the chloroplast-phloem junction rather than in the core companion cell region, leading to a mismatch in BADH precursor supply. The accumulation of glycine betaine is reduced by approximately 35% compared to Example 1, confirming the necessity of using metabolic flux analysis as a preliminary design basis for this invention. Further quantitative assessment of the uncertainty brought about by empirical promoter selection was conducted. Four variants were differentiated from this comparative example and tried different mesophyll / companion cell promoter combinations reported in the literature. The drought resistance phenotype fluctuated from 7.5 to 13.8 percentage points, with a coefficient of variation as high as 28%. In contrast, the coefficient of variation of the three different species schemes in Examples 1 to 3 selected based on metabolic flow analysis was only 8.3%. This indicates that the metabolic flow analysis-driven design of this invention significantly reduces the empirical dependence and phenotypic fluctuation of engineering scheme design, laying a predictable foundation for the engineering of drought resistance breeding.

[0038] Comparative Example 5 (parameters exceeded limits, expression fold exceeding the window specified in the claims). By adjusting the copy number of the DRE element in the cis-module to six copies (exceeding the upper limit of 2 to 5 copies specified in the claims), the stress-induced fold of the first expression cassette reached 85-fold (exceeding the upper limit of the 5-fold to 50-fold window). Although this comparative example showed the highest proline accumulation at the stress-induced peak, it exhibited severe amino acid metabolic disorders: the glutamine:glutamate ratio increased from the normal 1.1 to 3.8, and nitrogen metabolism imbalance manifested as leaf yellowing and premature plant wilting; simultaneously, excessive proline oxidation led to the accumulation of mitochondrial reactive oxygen species, which exacerbated stress damage. Field evaluation showed that the drought resistance of this comparative example (the relative water content retention rate of leaves increased by only 6.5 percentage points) was worse than that of the wild type in some physiological indicators, confirming that the 5-fold to 50-fold expression response window specified in this invention is a reasonable upper limit after considering the safety of metabolic network security. Similarly, if the copy number of DRE elements in the cis-module is reduced to one copy (below the lower limit of 2 copies in the claims), the stress-induced fold is only 3.2 times (below the lower limit of 5 times). At this point, the accumulation of proline and glycine betaine is insufficient to effectively maintain the cell's osmotic potential, and the drought resistance effect is improved by only 4.1 percentage points. This also confirms that the lower limit of 5 times defined in this invention is a necessary critical point for osmotic regulators to exert their actual protective effect. The upper and lower boundaries of the parameter window both have clear metabolic mechanism basis and practical engineering significance.

[0039] The relative water content (RWC) of leaves was determined using the classic Barrs & Weatherley method. Leaf discs with a diameter of 1 cm were harvested from the third fully expanded leaf at the top. The fresh weight (FW) was measured using an analytical balance (accuracy 0.1 mg). The leaf discs were then immersed in deionized water at 4°C for 24 h to absorb water until saturation. After drying the surface moisture, the saturated weight (TW) was measured. The leaf discs were then dried in an oven at 70°C for 48 h until constant weight (DW). The relative water content was calculated using the formula RWC = (FW - DW) / (TW - DW) × 100%. A total of 27 samples (3 plants, 3 leaves, 3 leaf discs) were randomly collected from each plot, and the average value was taken.

[0040] Method for determining the relative conductivity of cell membranes: The two-reading method of the conductivity meter was used. Fully expanded leaves were cut into 0.5 cm wide strips, and 0.2 g was accurately weighed. After rinsing three times with deionized water, the sample was soaked in 10 mL of deionized water at room temperature for 4 h. The initial conductivity L1 was measured using a Leici DDS-307A conductivity meter. The sample, along with the soaking solution, was boiled for 15 min, cooled to room temperature, and the total conductivity L2 was measured. The relative conductivity was calculated using the formula: Relative conductivity = L1 / L2 × 100%.

[0041] Root vigor assay: The TTC reduction method was used. 0.5 g of root tip was chopped and placed in 10 mL of phosphate buffer (pH 7.0) containing 0.4% TTC at 37°C in the dark for 2 h. After the reaction, the root segment was rinsed with running water, extracted with 8 mL of ethyl acetate, and the volume was adjusted to 10 mL. The absorbance was measured at 485 nm. The TTC reduction amount was calculated based on the formazan standard curve, expressed in mg / (g·h) fresh weight.

[0042] Quantitative methods for metabolomics: Samples were extracted with cold methanol-water (70:30 v / v), homogenized, centrifuged, and the supernatant was filtered through a 0.22 μm filter before entering a Q-Exactive high-resolution mass spectrometry-UltiMate 3000 HPLC system. The chromatographic column was a Waters ACQUITY UPLCBEH Amide (2.1 × 100 mm, 1.7 μm). Mobile phase A was 10 mmol / L ammonium formate aqueous solution, and mobile phase B was acetonitrile. Gradient elution was performed for 12 min at a flow rate of 0.3 mL / min, and the column temperature was 35°C. Mass spectrometry was performed in ESI positive / negative ion switching mode, with a scan range of m / z 70–1000 and a resolution of 70,000. Proline was quantified using m / z 116.0706 and its isotope peaks from 116.0740 to 121.0920, and glycine betaine was quantified using m / z 118.0862 and its isotope peaks from 118.0892 to 121.0985, with limits of quantification of 0.05 μmol / g and 0.02 μmol / g fresh weight, respectively. Deuterated proline (D7-Proline) and deuterated betaine (D11-Glycine betaine) were used as internal standards, added before the extraction step to correct for matrix effects and recovery fluctuations. Data acquisition was performed using Xcalibur software, and spectral integration and quantification were performed using the Skyline open-source software. Each sample underwent three technical replicates and three biological replicates. Statistical analysis was performed using one-way ANOVA combined with Duncan's multiple comparisons in SPSS 26.0, with a significance level of P < 0.05. All time-series data from flow analysis are further input into the independently developed spatiotemporally resolved stable isotope flux redistribution algorithm ST-SIFRA for flux quantification, outputting rate-limiting enzyme flux sensitivity heatmaps for mesophyll and companion cell tissues respectively.

[0043] qRT-PCR expression analysis method: Total RNA was extracted from tissue-specific micro-dissected samples using the TRIzol method. After DNase I treatment, cDNA was reverse transcribed using the PrimeScript RT Reagent Kit. The relative abundance of LaP5CS and ScBADH transcripts was detected using the SYBR Green qPCR Master Mix system on an Applied Biosystems 7500 real-time PCR instrument. Rice OsACT1 or wheat TaGAPDH was used as an internal reference gene. The relative expression level was calculated using the 2-ΔΔCt method. Each sample was tested with three technical replicates and three biological replicates. The LaP5CS primers used for qRT-PCR were 5'-GCTGAAGTGGGCATCGAATA-3' (forward) and 5'-CCTGCAAGACGAACTTGACT-3' (reverse). The ScBADH primers were 5'-GCCTCCTACAAGATGGCCAA-3' (forward) and 5'-AGCAGTTCCCGATCTTGACT-3' (reverse). All primers produced amplification products approximately 150 bp in length. NCBI BLAST analysis confirmed no non-specific matches in the rice / wheat genome. The amplification program consisted of 40 cycles: 95°C pre-denaturation for 30 s, followed by 95°C denaturation for 5 s and 60°C annealing and extension for 30 s. Melting curve analysis was performed on each sample to confirm amplification specificity.

[0044] P5CS and BADH enzyme activity assay: Approximately 10 mg of fresh mesophyll micro-dissection sample or companion cell micro-dissection sample was added to 100 μL of pre-cooled extraction buffer (50 mmol / L Tris-HCl pH 7.5, containing 5 mmol / L MgCl2, 1 mmol / L EDTA, 2 mmol / L DTT, and 1 mmol / L PMSF). After homogenization, the mixture was centrifuged at 12000×g for 15 min at 4°C. The supernatant was then used to determine enzyme activity. P5CS enzyme activity was determined using the substrate consumption method. The reaction system contained 100 mmol / L Tris-HCl pH 7.5, 50 mmol / L glutamate, 10 mmol / L ATP, 5 mmol / L MgCl2, and 1 mmol / L NADPH. NADPH oxidation was monitored at 340 nm. One unit of enzyme activity (U) was defined as the decrease in NADPH by 1 μmol per minute. BADH enzyme activity was determined using a standard spectrophotometric method. The reaction system contained 50 mmol / L Tris-HCl pH 8.0, 2 mmol / L betaine aldehyde, and 1 mmol / L NAD+. NADH generation was monitored at 340 nm, and 1 μmol of NADH generated per minute was defined as one unit of enzyme activity (U). Enzyme activity was expressed as the amount of substrate catalyzed per milligram of soluble protein per minute, in units of U / mg·prot. Protein content was determined using the Bradford method with bovine serum albumin as the standard curve.

[0045] Table 1 summarizes the core drought resistance physiological indicators of the engineered strains, wild type, and five comparative groups of this invention. These indicators cover five key aspects: leaf relative water content maintenance rate, reduction in cell membrane relative conductivity, root vigor maintenance rate, grain yield change compared to the stress control, and yield loss during the non-stress period. This allows for a convenient and intuitive assessment of the comprehensive drought resistance effect and metabolic burden level of each scheme relative to the wild type.

[0046] Table 1. Summary of core drought resistance physiological indicators of engineered lines, wild types, and five comparative groups

[0047] Solution Group RWC maintenance rate improved (percentage points). Membrane conductivity decreased (percentage points) Root vitality improved (%) Production increase during stress period (%) Non-stress period yield loss (%) Example 1 (Rice TL-7) 18.3 25.1 133 24.8 5.7 Example 2 (Wheat TW-3) 19.6 24.3 —— 18.5 4.8 Example 3 (Soybean TG-5) 15.4 21.8 —— —— 3.2 Comparative Example 1 (Constitutional Promoter) 13.2 17.5 88 6.4 14.2 Comparative Example 2 (Dual Genes in the Same Leaf Mesophyll) 8.9 14.2 45 9.8 7.1 Comparative Example 3 (Single P5CS Overexpression) 9.6 12.8 52 11.3 4.5 Comparative Example 4 (Empirical Promoter) 11.2 16.7 71 14.5 6.8 Comparative Example 5 (Overexpression Exceeding Limits) 6.5 13.1 31 -4.2 9.3 Wild-type control —— —— —— 0.0 0.0

[0048] Table 1 data reveals three important metabolic engineering characteristics. First, the key drought resistance indicators (increased RWC maintenance rate and decreased membrane conductivity) of the engineered lines are significantly higher in all three schemes of Examples 1 to 3 than in the five comparative groups. The increase in RWC maintenance rate in Examples 1 to 3 is concentrated in the range of 15.4 to 19.6 percentage points, while that in the comparative groups is concentrated in the range of 6.5 to 13.2 percentage points. The two ranges do not overlap, indicating that the drought resistance gain generated by the core technical scheme of this invention cannot be achieved by a single comparative variable, but is the result of the synergistic emergence of multiple features. Second, the distribution of yield loss indicators during the non-stress period shows strong discriminative power: Examples 1 to 3 are concentrated in the low loss range of 3.2% to 5.7%, while the comparative group 1 (constitutive promoter) is as high as 14.2%, and the comparative group 5 (overexpression boundary) is as high as 9.3%. This indicates that the independent contributions of the two core features of drought dynamic regulation and expression fold control to field productivity preservation are not negligible. Third, the overexpression over-limit scheme in Comparative Example 5 showed a negative value (-4.2%, i.e., reduced yield compared to the wild-type stress control) in terms of yield during the stress period, which was worse than the wild-type. This confirms that the upper limit of the 5 to 50-fold expression response window described in the claims is a reasonable definition after considering the toxicity avoidance of metabolic networks. The secondary stress effect caused by amino acid metabolism disorder and mitochondrial reactive oxygen species accumulation due to over-limit expression can completely offset or even reverse the osmotic regulation gain. This counter-evidence provides strong quantitative support for the parameter window of the claims from the perspective of safety boundaries.

[0049] The core drought-resistance physiological indicators of the engineered rice lines, wild-type rice, and five comparative examples are summarized as follows: In Example 1, the relative leaf water content retention rate of the TL-7 rice line was increased by 18.3 percentage points compared to the wild-type, membrane conductivity decreased by 25.1 percentage points, root activity increased by 133% compared to the wild-type stress control, grain yield increased by 24.8% compared to the wild-type stress control, and yield loss during the non-stress period was only 5.7%; in Comparative Example 1, the four indicators of the constitutive promoter scheme were 13.2%, 17.5%, 88%, and 6.4%, respectively, but the non-stress period yield loss was only 5.7%. The yield loss during the stress period was 14.2%; the results for the dual-gene mono-mesophyll expression scheme in Comparative Example 2 were 8.9%, 14.2%, 45%, 9.8%, and 7.1%; the results for the single P5CS scheme in Comparative Example 3 were 9.6%, 12.8%, 52%, 11.3%, and 4.5%; the results for the empirical promoter scheme in Comparative Example 4 were 11.2%, 16.7%, 71%, 14.5%, and 6.8%; and the results for the overexpression over-limit scheme in Comparative Example 5 were 6.5% (worse than expected), 13.1%, 31%, -4.2% (yield reduction), and 9.3%. The performance results of the wheat TW-3 line in Example 2 and the soybean TG-5 line in Example 3 showed a highly similar trend to the rice scheme in Example 1. The relative leaf water content maintenance rate was 19.6 and 15.4 percentage points, respectively; the membrane conductivity decreased by 24.3 and 21.8 percentage points, respectively; and the yield loss during the non-stress period was 4.8% and 3.2%, respectively, confirming the cross-species transferability of the method of the present invention. Further statistical analysis showed that the three schemes in Examples 1 to 3 were significantly different from the corresponding wild type in terms of drought resistance (P<0.01), and the difference from the best performing comparative example 4 among the five comparative examples was still significant (P<0.05). Statistically, this fully supports the inventiveness of the invention as an independent technical contribution.

[0050] Quantitative Analysis of Nonlinear Synergistic Effects: If the two pathways exhibit a simple additive relationship, the sum of the drought resistance contribution (a 9.6 percentage point increase in relative leaf water content maintenance) of the single P5CS scheme in Comparative Example 3 and the hypothetical single BADH scheme contribution (reported in the literature as approximately 8.5 percentage points) should be 18.1 percentage points, close to the 18.3 percentage points in Example 1, seemingly indicating no synergy. However, considering the yield indicators comprehensively, Example 1, under comparable drought resistance conditions, shows a yield increase of 13.5 percentage points (24.8% vs. 11.3%) greater than Comparative Example 3, significantly exceeding the expectation of simple addition; simultaneously, the yield loss during the non-stress period is only 5.7% in Example 1, while the combined scheme of Comparative Example 1 suffers as high as 14.2%, with the difference of 8.5 percentage points precisely representing the nonlinear synergistic increment of the spatiotemporal dual-dimensional regulation of this invention relative to single-dimensional regulation. The metabolic mechanism of this increment stems from the increased energy metabolism efficiency after precursor competition is cut off and the resource preservation during the non-stress period brought about by dynamic switching, consistent with the theoretical expectation of the spatiotemporal topology optimization of the metabolic network described in the invention content section. Further quantitative evaluation was conducted using the synergistic effect index SE = (actual effect - summed expected effect) / summed expected effect × 100%. Example 1 showed an SE value of 74.6% in the yield preservation dimension, far exceeding the average SE level of 20%-30% in traditional multi-gene synergistic studies; the SE value was 52.3% in the leaf photosynthetic rate retention dimension and 48.1% in the root vigor dimension. These three dimensions consistently support the nonlinear synergistic innovation contribution of this invention compared to existing technologies. Furthermore, comparing Example 1 with the empirical promoter scheme of Comparative Example 4, it is evident that the pre-analysis of metabolic flux contributes approximately 7.1 percentage points to the final phenotype, accounting for 39% of the overall benefit, further highlighting the independent contribution of this invention's quantitatively driven metabolic flux redistribution insight.

[0051] Table 1, a cross-scheme depth comparison of the data, further reveals the independent contribution of each core technical feature of this invention. Using the difference in RWC maintenance rate between Example 1 (full-feature scheme) and each comparative example as a quantitative indicator of the feature's contribution, we obtain the following: dynamic regulation feature (removed from Comparative Example 1) contributes 5.1 percentage points; tissue-differentiated targeting feature (removed from Comparative Example 2) contributes 9.4 percentage points; dual-pathway synergistic feature (removed from Comparative Example 3) contributes 8.7 percentage points; and metabolic flux analysis-driven feature (removed from Comparative Example 4) contributes 7.1 percentage points. The cumulative contribution of these four features is 30.3 percentage points, far exceeding the measured 18.3 percentage point increase in RWC maintenance rate of Example 1 relative to the wild type. This unexpected phenomenon, from another perspective, proves the non-linear coupling effect between the features—the effect of any feature depends on whether the other features are effective simultaneously; removing any feature alone will lead to a non-linear decrease in the overall scheme efficiency. Combined with... Figure 3The heatmap showing the differential distribution of proline content in the mesophyll and companion cells further illustrates that, under the scheme of Example 1, proline in the mesophyll reaches a peak of approximately 115 μmol / g fresh weight at 24 h, while glycine and betaine in the companion cells reach a peak of approximately 42 μmol / g fresh weight at the same time. The synchronous accumulation of the two tissues and two metabolites forms a four-quadrant synergistic effect regulated by osmotic potential. This four-quadrant pattern is a unique metabolic fingerprint that cannot be simultaneously observed in the comparison, constituting the inventive metabolomics evidence of this invention. The standard deviation of Example 1 in the three-year, three-location, multi-site field replication trial did not exceed 3.2 percentage points, indicating that the phenotypic stability of the scheme of this invention is significantly better than that of traditional drought-resistant genetic engineering schemes (the field phenotypic variation coefficient reported in the literature is usually in the range of 15% to 25%). The mechanism of this stability lies in the pre-design of metabolic flux analysis in this invention, which frees the engineering scheme from the random bias caused by empirical trial and error, constituting an important technical advantage for the commercial application of this invention.

[0052] The synergistic effect of this invention is supported by three interconnected metabolic network mechanisms. Mechanism one is precursor competition cleavage: when proline synthesis and glycine betaine synthesis are simultaneously initiated at high intensity in the same cell, multidimensional competition arises among the three precursor libraries of glutamate, S-adenosylmethionine, and NADPH. Due to the active carbon fixation of the Calvin cycle, the flux of 3-phosphoglycerate converted to glutamate via amination in mesophyll cells reaches as high as 14-20 μmol / (g·h) during photoperiod, providing ample substrate for P5CS; however, S-adenosylmethionine required for choline synthesis is mainly used for chlorophyll methylation in mesophyll. If choline synthesis upstream of BADH is initiated synchronously here, it will directly compete for S-adenosylmethionine supply, inhibiting chloroplast function. This also explains the metabolic root cause of the decreased photosynthetic rate under the dual-gene mesophyll scheme in Comparative Example 2. As a phloem loading hub, the vascular bundle companion cell is actively redistributed with amino acids. Betaine aldehyde precursors can be supplied from adjacent parenchyma cells via plasmodesmata. Furthermore, the S-adenosylmethionine pool in the companion cell is specifically used for the synthesis of metabolic signaling molecules rather than chlorophyll maintenance; therefore, BADH expression at this site does not interfere with chloroplast function. In addition, the active nitrogen redistribution in the companion cell means that the nitrogen backbone required for glycine betaine synthesis can be supplied by amino acids such as glutamine and asparagine loaded in the phloem from nearby cells, without competing with the proline pathway for the glutamate pool in the mesophyll. This invention places P5CS and BADH in their respective ecological niches with optimal precursor supply and metabolic dissipation, eliminating metabolic conflict from competition for the same precursor pool, ensuring that the synergistic effect of the two pathways does not come at the expense of growth rate.

[0053] Mechanism 2 is a dynamic response to avoid continuous metabolic consumption: the combination of DRE and ABRE cis-elements in the composite promoter exhibits higher kinetic sensitivity to drought stress and ABA response than traditional single-element promoters, achieving a two-way dynamic closed loop of rapid accumulation during stress initiation and avoidance of accumulation during rehydration decline. Under rehydration conditions, leaf ABA levels drop to below 20% of baseline within 3 hours, leading to a significant decrease in composite promoter activity; simultaneously, the mRNA destabilizing elements further reduce the half-life of P5CS and BADH transcripts from 6 hours during stress to 1.5 hours during rehydration, accelerating the baseline decline in expression levels. In contrast, constitutive promoters such as CaMV 35S or Ubi1 maintain basal transcriptional activity similar to that during stress under rehydration conditions, resulting in the continued synthesis of protective metabolites even after stress relief. The NADPH reducing power, ATP, and carbon-nitrogen skeleton consumed in these additional syntheses, which should be used for growth and yield formation, are ineffectively locked in the proline and glycine betaine pools, becoming a net metabolic burden during the non-stress period. The dynamic closed-loop mechanism of this invention avoids continuous energy consumption during the non-stress period and preserves the directional allocation of photosynthetic products to growth and yield formation. This mechanism is directly verified by the stark contrast between Comparative Example 1 and Example 1 in terms of yield difference of 14.2% to 5.7% during the non-stress period.

[0054] Mechanism 3 is the emergent effect of metabolic network topology optimization: Quantitative analysis based on the spatiotemporally resolved stable isotope flux redistribution algorithm ST-SIFRA shows that under the dual-tissue differentiated targeting scheme of this invention, the Wasserstein distance of carbon and nitrogen flux redistribution between mesophyll and companion cells under stress conditions is 2.8 times higher than that of constitutive comparison example 1 and 4.1 times higher than that of single mesophyll comparison example 2, indicating that this scheme promotes the coordinated reorganization of the whole plant metabolic network rather than the accumulation of local metabolites. Specifically, high expression of P5CS in mesophyll promotes the preferential flow of carbon skeleton to proline synthesis in mesophyll, reducing the diversion of excess carbon in mesophyll to photorespiration; high expression of BADH in companion cells promotes the preferential flow of nitrogen skeleton to glycine betaine synthesis in companion cells, reducing the blind consumption of nitrogen redistribution in companion cells; the two processes establish a bidirectional material flow between mesophyll and companion cells through the osmotic potential gradient of intercellular interconnection, enabling the whole plant metabolic network to dynamically adjust towards drought-resistant adaptive configuration. This network-level recombination is the underlying mechanism for the simultaneous achievement of growth preservation during non-stress periods and enhanced drought resistance during stress periods. It cannot be achieved by any single-dimensional intervention (single gene, single promoter, single tissue) alone, constituting a non-linear synergistic innovative contribution of this invention compared to existing technologies. Further examining it from the perspective of metabolic engineering paradigms, this invention confirms that enhancing plant drought resistance should not only focus on the synthetic intensity of protective metabolites, but also on their spatiotemporal distribution topology within the metabolic network. This paradigm shift lays the methodological foundation for future rational stress-resistance molecular breeding based on metabolic network mechanisms.

[0055] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for enhancing drought adaptability through the accumulation of plant osmotic regulators, characterized in that, Includes the following steps: (1) The plants to be modified were given dual isotope administration of 13C-labeled L-glutamic acid and 15N-labeled choline, combined with micro-dissection of mesophyll cells and vascular bundle companion cells to perform stable isotope metabolic flux analysis, quantify the carbon and nitrogen flow redistribution of the proline synthesis pathway and glycine betaine synthesis pathway in the two types of tissues, and identify the rate-limiting enzyme nodes and cross-pathway precursor competition intensity of the two pathways. (2) Construct a polycistronic plant expression vector, the vector comprising a first expression cassette and a second expression cassette, the first expression cassette being driven by a first composite promoter consisting of a mesophyll cell-specific promoter and a drought-responsive cis-regulatory element, which encodes a P5CS sequence with proline synthase activity; the second expression cassette being driven by a second composite promoter consisting of a vascular bundle companion cell-specific promoter and a drought-responsive cis-regulatory element, which encodes a BADH sequence with betaine aldehyde dehydrogenase activity; the drought-responsive cis-regulatory element is selected from one or more of DRE, ABRE, MYB, and MYC, in combinations of 2 to 5 copies. (3) The polycistronic plant expression vector was transformed into the plant to be modified by Agrobacterium-mediated transformation or gene gun method, and positive transformants were obtained by selection marker screening and molecular identification. The positive transformants showed that under drought stress, the expression levels of P5CS in mesophyll cells and BADH in vascular bundle companion cells were 5 to 50 times higher than under normal water conditions, and returned to the pre-stress baseline level within 72 h after rehydration.

2. The method according to claim 1, characterized in that... : In step (1), the final concentration of 13C-labeled L-glutamic acid was 1.5 mmol / L to 3.0 mmol / L, and the final concentration of 15N-labeled choline was 1.0 mmol / L to 2.5 mmol / L. The administration method was root hydroponics or foliar spraying. After 24 h of administration, drought stress or osmotic stress simulation was applied, and mesophyll cells and vascular bundle companion cells were collected at five time points: 0 h, 2 h, 6 h, 24 h, and 72 h after stress. The samples were extracted with cold methanol-water and analyzed by liquid chromatography-high resolution mass spectrometry to determine the abundance of proline M+0 to M+5 isotopes and glycine betaine M+0 to M+3 isotopes. The rate-limiting enzyme nodes were identified by the spatiotemporally resolved stable isotope flux redistribution metric.

3. The method according to claim 1 or 2, characterized in that... : In step (2), the first composite promoter includes a mesophyll cell-specific promoter region of -1100 bp to -1 bp, and a composite cis-module consisting of 2 to 4 copies of tandem DRE cis elements and 1 to 3 copies of ABRE cis elements is embedded in the -800 bp to -600 bp region of the promoter region; the mesophyll cell-specific promoter is selected from one of the rbcS-1A promoter, Cab promoter, PEPC promoter, TaLhcb promoter, and GmRbcS2 promoter.

4. The method according to claim 3, characterized in that... : In step (2), the second composite promoter includes a -2000 bp to -1 bp promoter region of a vascular bundle companion cell specific promoter, and a composite cis-module consisting of 1 to 3 copies of DRE cis elements and 2 to 5 copies of ABRE cis elements is embedded in the -500 bp to -300 bp region of the promoter region; the vascular bundle companion cell specific promoter is selected from one of the SUC2 promoter, rolC promoter, AtCVP2 promoter, TaSUT1 promoter, and GmSUC1 promoter; the initiation delay time of the second composite promoter in response to stress is shortened by 1 h to 4 h compared with the first composite promoter.

5. The method according to claim 4, characterized in that... : At the peak of drought stress, the positive transformants exhibited a ratio of P5CS enzyme activity in mesophyll cells to BADH enzyme activity in vascular bundle companion cells of 0.8 to 2.5; free proline concentration in mesophyll cells of 25 μmol / g fresh weight to 120 μmol / g fresh weight; and free glycine betaine concentration in vascular bundle companion cells of 8 μmol / g fresh weight to 45 μmol / g fresh weight.

6. The method according to claim 5, characterized in that... : The polycistronic plant expression vector further comprises 1 to 3 copies of the AT-enriched mRNA destabilizing element DST located at the 3' end of the P5CS coding sequence and the 3' end of the BADH coding sequence, wherein the common sequence of the DST element is ATAGATAGATA; the DST element reduces the half-life of the mRNA of P5CS and BADH under rehydration conditions from 4 to 8 h during the stress period to 0.5 to 2 h.

7. The method according to claim 6, characterized in that... : The P5CS coding sequence with proline synthase activity is a Δ1-pyrrololine-5-carboxylic acid synthase coding sequence that has been freed from proline feedback inhibition by a mutation at the F129A site. Its original sequence is derived from one of the following: lupinus angustifolius, Arabidopsis thaliana, rice Oryza sativa, cowpea Vigna unguiculata, and wheat Triticum aestivum, and has been optimized with codons specific to the plants to be modified. The BADH coding sequence with betaine aldehyde dehydrogenase activity is derived from one of the following: spinach Spinacia oleracea, sugarcane Saccharum officinarum, bermudagrass Cynodon dactylon, and alkali grass Puccinellia tenuiflora, and has been optimized with codons.

8. The method according to claim 7, characterized in that : The vector backbone of the polycistronic plant expression vector is selected from one of pCAMBIA1305.1, pCAMBIA2300, pBI121, and pBinAR; the polycistronic plant expression vector also contains one of the following as a plant selection marker: hygromycin resistance gene HPT, glufosinate resistance gene bar, glyphosate resistance gene EPSPS, and neomycin phosphotransferase gene nptⅡ; the first expression cassette is located upstream or downstream of the second expression cassette, and the two expression cassettes are separated by a NOS terminator or a 35S terminator.

9. The method according to claim 8, characterized in that... : The plant to be modified is selected from one of the following: rice (Oryza sativa), wheat (Triticum aestivum), maize (Zea mays), cotton (Gossypium hirsutum), soybean (Glycine max), rapeseed (Brassica napus), tomato (Solanum lycopersicum), tobacco (Nicotiana tabacum), potato (Solanum tuberosum), and Arabidopsis thaliana; the molecular identification in step (3) includes specific primer PCR amplification to confirm the integration of the target gene and qRT-PCR to confirm the stress-induced expression response of the two expression cassettes.

10. The method according to claim 9, characterized in that... : The polycistronic plant expression vector, as an intermediate material product in this method, has a T-DNA region that sequentially includes a left boundary, a first complex promoter, the P5CS coding sequence, a first terminator, a second complex promoter, the BADH coding sequence, a second terminator, the plant selection marker, and a right boundary. The positive transformants obtained by the transformation and screening of the polycistronic plant expression vector are plant cells, plant tissues, plant organs, or their progeny propagation materials containing the integrated T-DNA sequence of the polycistronic plant expression vector.

Images