Establishment of an Efficient Agrobacterium-Mediated Genetic Transformation System to Enhance the Tolerance of the Paraquat Stress in Engineering Goosegrass (Eleusine Indica L.)

Eleusine indica (goosegrass) is a problematic weed worldwide known for its multi-herbicide tolerance/resistance biotype. However, a genetic transformation method in goosegrass has not been successfully established, making a bottleneck for functional genomics studies in this species. Here, we report a successful Agrobacterium-mediated transformation method for goosegrass. Firstly, we optimized conditions for breaking seed dormancy and increasing seed germination rate. A higher callus induction rate from germinated seeds was obtained in N6 than in MS or B5 medium. Then the optimal transformation efficiency of the gus reporter gene was obtained by infection with Agrobacterium tumefaciens culture of OD600 = 0.5 for 30 min, followed by 3 days of co-cultivation with 300 μmol/L acetosyringone. Concentrations of 20 mg L−1 kanamycin and 100 mg L−1 timentin were used to select the transformed calli. The optimal rate of regeneration of the calli was generated by using 0.50 mg L−1 6-BA and 0.50 mg L−1 KT in the culture medium. Then, using this transformation method, we overexpressed the paraquat-resistant EiKCS gene into a paraquat-susceptible goosegrass biotype MZ04 and confirmed the stable inheritance of paraquat-resistance in the transgenic goosegrass lines. This approach may provide a potential mechanism for the evolution of paraquat-resistant goosegrass and a promising gene for the manipulation of paraquat-resistance plants. This study is novel and valuable in future research using similar methods for herbicide resistance.


Introduction
Grasses were used for food, feed, and beverages by humans 2 million years ago. Some of the grasses have been domesticated by humans as crops for agriculture [1]. Moreover, some of the most devastating agricultural weeds in crop fields also are grasses [2]. Recently, reviews have been made on expounding the grass research and addressing the value of grass diversity in the future human life [3][4][5][6]. However, the global herbicide resistant weeds found in 267 plant species (154 dicots and 113 monocots) have become a serious problem for agricultural production [7].
Goosegrass (Eleusine indica L. Gaertn) is one of the top 10 herbicide-resistant weed species distributed throughout the world [8,9]. Although multiple resistances to paraquat, glufosinate, and glyphosate in goosegrass have been reported, the underlying molecular mechanism of herbicide stress responses remains unclear [10,11]. One of the crucial reasons is that the genetic transformation system in goosegrass has not been established. The genetic transformation system in goosegrass can provide a deeper understanding of how resistant-gene expression responds to herbicide stress through its metabolism to

Effects of the Developmental Regulators on Goosegrass Calli Induction from Seeds
The effects of the developmental regulators showed that the treatment of 30% NaOH with 5-20 min greatly increased the germination rates (up to 26.25%) (Figure 2A). However, the treatment of −20 • C could also improve the germination rates up to 49% ( Figure 2B). The seeds were also treated with 75% ethanol for 1-5 min followed by 0.10% HgCl 2 for 15 min. Although the results did not show that one minute was significantly different than 2 or 5 min, the treatment with 75% ethanol solution for 1 min (and then with 0.10% HgCl 2 for 15 min) had the highest germination rate (32%) without seed contamination ( Figure 2C). To further reduce any negative influence of the induction rate from HgCl 2 sterilization, 0.05%, 0.10%, and 0.20% concentrations of HgCl 2 solution were used to sterilize the seeds for 10, 15, and 20 min. The callus induction rates of the seeds under sterilization with 0.05% HgCl 2 trended higher than those with 0.10% and 0.20% HgCl 2 ( Figure 2D). The developmental regulators may improve callus induction in plants. As one of the critical regulators on callus induction, the use of 0.50 mg L −1 2,4-D (2,4-dichlorophenoxyacetic acid) in the N6 medium showed improved induction of type-I calli over 1.0, 2.0, and 3.0 mg L −1 2,4-D ( Figure 2E). With this 0.50 mg L −1 2,4-D concentration, the total callus induction rate reached as high as 75% on average ( Figure 2F). After initial induction of the calli, they were sub-cultured on the media containing either 7 g agar/L or 3.0 g phytagel/L. The subculture medium with agar was more conducive to keep the growth of type-I calli with a higher induction rate (62.49%), as compared to that containing phytagel, although that use of agar and phytagel respectively in the initial callus induction did not significantly impact induction rates (74.44% and 71.67%) for the type-I calli ( Figure 2G). Furthermore, an air drying (for 3.0 h) of the medium surface before the subculture could promote the conversion of the type-II calli into type-I calli (93.89%) ( Figure 2H), which were suitable for the generation of embryogenic calli.

Effects of Regulators on Agrobacterium-Mediated Infection and Regeneration of Goosegrass Calli
To evaluate the infection efficiency of goosegrass calli, the empty plasmid with the GUS gene was transformed into the Agrobacterium tumefaciens strain (EHA105). First, the effect of Agrobacterium-mediated infection was determined by the A. tumefaciens cell culture with OD 600 values from 0.40 to 0.80. The results showed that the optimal cell density for infection was 0.4-0.5 (OD 600 ), which obtained the maximum transformation efficiency (39.17%) of the GUS gene on the goosegrass calli ( Figure 3A). Secondly, we investigated the infection time of the goosegrass calli dipped with the A. tumefaciens culture for various time intervals from 10 to 40 min. The infection duration of 30 min was found to be optimal with 37.50% of the calli showing the GUS expression ( Figure 3B). Thirdly, goosegrass calli were co-cultivated with A. tumefaciens cells for different periods from 0 to 4 days. The transformation efficiency of GUS was significantly improved with 2-3 days of cocultivation, reaching as high as 36% ( Figure 3C). On the other hand, different concentrations of acetosyringone (AS) with 100-500 µmol/L in the co-cultivation were tested. The results showed that use of 300 µmol/L AS produced the highest efficiency (58.67%) of goosegrass transformation ( Figure 3D). Next, the optimal concentration of hygromycin (0, 20, 25, 30, 40, 50, 100 mg L −1 ) was determined for screening goosegrass transformants with the selectable marker gene HPT (encoding hygromycin phosphotransferase). We observed that the use of 25 mg L −1 of hygromycin dramatically decreased the fresh weight of the goosegrass calli ( Figure 3E). Thus, the concentration of 20 mg L −1 of hygromycin was used to select the transformant cells during the callus culture. In addition, the use of a suitable antibiotic, such as timentin, for A. tumefaciens counter selection is crucial during the callus culture, which can avoid contamination of the A. tumefaciens cells over the surface of the calli. A test showed that the use of 100 mg L −1 timentin could drastically decrease the contamination rate of A. tumefaciens on the surface of the goosegrass calli ( Figure 3F). Thus, this concentration was used for later goosegrass transformation. The conversion rates of the type-II and type-III calli into type-I calli in subculture with an air drying of the medium surface for 0 h and 3.0 h, respectively, before the subculture.

Effects of Regulators on Agrobacterium-Mediated Infection and Regeneration of Goosegrass Calli
To evaluate the infection efficiency of goosegrass calli, the empty plasmid with the GUS gene was transformed into the Agrobacterium tumefaciens strain (EHA105). First, the observed that the use of 25 mg L of hygromycin dramatically decreased the fresh weight of the goosegrass calli ( Figure 3E). Thus, the concentration of 20 mg L −1 of hygromycin was used to select the transformant cells during the callus culture. In addition, the use of a suitable antibiotic, such as timentin, for A. tumefaciens counter selection is crucial during the callus culture, which can avoid contamination of the A. tumefaciens cells over the surface of the calli. A test showed that the use of 100 mg L −1 timentin could drastically decrease the contamination rate of A. tumefaciens on the surface of the goosegrass calli ( Figure 3F). Thus, this concentration was used for later goosegrass transformation.  For callus regeneration in plant transformation, MS or 1/2 MS medium is often used. Thus, the selected hygromycin-resistant goosegrass calli were transferred to MS medium supplemented with different combinations of 6-BA (0.50-2.00 mg L −1 ), 2,4-D (0.50 mg L −1 ), and kinetin KT (0.50-2.00 mg L −1 ) for regeneration ( Table 1). The results showed the highest regeneration rate (71.05%) on MS medium supplemented with 0.50 mg L−1 6-BA and 0.50 mg L−1 KT.

Goosegrass Transformed with the Paraquat-Resistant Gene EiKCS
After establishing the the Agrobacterium-mediated genetic transformation system in goosegrass, we applied this system to study the gene function of EiKCS that confers paraquat-resistance in goosegrass [29]. Type-I calli were induced from seeds of the paraquatsusceptible goosegrass biotype (MZ04) on N6 medium with 0.50 mg L −1 2,4-D ( Figure 4A). After callus subculture ( Figure 4B), the calli were infected by A. tumefaciens cells containing the binary vector pCUbi1309::EiKCS for overexpressing EiKCS driven by the promoter of the maize Ubiquitin gene ( Figure 4C). Then, the calli were successively transferred to co-cultivation media with 0.50 mg L −1 AS and screening media with 20 mg L −1 hygromycin ( Figure 4D). Next, the hygromycin-resistant calli were differentiated on MS with 0.50 mg L −1 6-BA, 0.50 mg L −1 KT, 0.05 mg L −1 NAA, 0.05 mg L −1 IAA, and 100 mg L −1 timentin, until buds appeared and then they were transferred to light until green leaves grew to 1-2 cm in length ( Figure 4E). The transformants were separately transferred to root-inducing media until green leaves grew to 4-6 cm in length ( Figure 4F). Finally, acclimatization and soil transfer of the transgenic plants were carried out ( Figure 4G). Each independent transformant was propagated into more than 14 plants in pots ( Figure 4H).

Characterization of Paraquat-Resistant Transgenic Goosegrass Plants
To determine the paraquat-resistance and genetic characters of the transgenic goosegrass plants, they (T2 to T5 generations) were treated with the recommended paraquat solution (3750 mg L −1 ). After spraying with a full dose of the paraquat solution, all leaves of the transgenic (T2, T3) and wild-type plants withered within two days. However, the transgenic plants survived after 30 days and almost recovered after 90 days, whereas the wild-type plants completely died ( Figure 5A). Moreover, the abiotic stress of paraquat stimulated the growth of transgenic goosegrass. Although the leaves of the transgenic goosegrass (T4) were visibly damaged after two days of full-dose or half-dose paraquat treatments, and their vegetative growths were still flourishing to the reproductive stage after 30 days of the treatments ( Figure 5B). In addition, the paraquat resistance of the transgenic goosegrasses (T5) was equivalent to that of naturally selected resistant-biotype goosegrass QY05. However, the susceptible-biotype goosegrass (JM) as a control died under a low concentration of 270 mg L −1 paraquat treatment ( Figure 5C). Hence, these results determined that a stable genetic resistance to paraquat could be successfully obtained by using this Agrobacterium-mediated genetic transformation system in goosegrass.

Characterization of Paraquat-Resistant Transgenic Goosegrass Plants
To determine the paraquat-resistance and genetic characters of the transgenic goosegrass plants, they (T 2 to T 5 generations) were treated with the recommended paraquat solution (3750 mg L −1 ). After spraying with a full dose of the paraquat solution, all leaves of the transgenic (T 2 , T 3 ) and wild-type plants withered within two days. However, the transgenic plants survived after 30 days and almost recovered after 90 days, whereas the wild-type plants completely died ( Figure 5A). Moreover, the abiotic stress of paraquat stimulated the growth of transgenic goosegrass. Although the leaves of the transgenic goosegrass (T 4 ) were visibly damaged after two days of full-dose or half-dose paraquat treatments, and their vegetative growths were still flourishing to the reproductive stage after 30 days of the treatments ( Figure 5B). In addition, the paraquat resistance of the transgenic goosegrasses (T 5 ) was equivalent to that of naturally selected resistant-biotype goosegrass QY05. However, the susceptible-biotype goosegrass (JM) as a control died under a low concentration of 270 mg L −1 paraquat treatment ( Figure 5C). Hence, these results determined that a stable genetic resistance to paraquat could be successfully obtained by using this Agrobacterium-mediated genetic transformation system in goosegrass. To confirm the positive transgenic goosegrass, the T2 to T6 generations of transgenic goosegrass were detected by molecular screening. PCR certified positive transgenic lines in T2 to T6 generation goosegrasses with specific primers. The amplifying fragments contained the EiKCS gene with the Ubiquitin-promoter sequences in the over-expression To confirm the positive transgenic goosegrass, the T 2 to T 6 generations of transgenic goosegrass were detected by molecular screening. PCR certified positive transgenic lines in T 2 to T 6 generation goosegrasses with specific primers. The amplifying fragments contained the EiKCS gene with the Ubiquitin-promoter sequences in the over-expression vector. The presence of the marker hygromycin (HPT) gene and Ubi were the positive controls. And the original paraquat-susceptible goosegrass biotype (MZ04) was used as the negative control ( Figure 6A). The result showed the successful identification of positive transgenic goosegrass lines. Meanwhile, the effect of the EiKCS function was evaluated in the paraquat-resistant response of the transgenic goosegrass. The qRT-PCR showed different expression levels of EiKCS in leaves of the wild-type (MZ04) and three OE-EiKCS lines (T 2 ) under treatments of spraying H 2 O, 270 mg L −1 paraquat, and 270 mg L −1 paraquat plus 1.5 mmol/L spermidine ( Figure 6B). The results of qRT-PCR confirmed that the relative expressions of the EiKCS gene in transgenic goosegrass were all higher than those in wildtype goosegrass. Moreover, paraquat stress promoted significant overexpressing of EiKCS in these transgenic lines with an average of 7.38-fold compared to the wild-type goosegrass. However, the increased expression of EiKCS by paraquat in the transgenic goosegrass was down-regulated by 4.25-fold when applying exogenous spermidine (Para + spd). Moreover, the gene-targeted protein EiKCS showed higher expression levels in the transgenic goosegrass than in all wild-type goosegrass (MZ04) by PRM under the same treatments ( Figure 6C). The results of PRM confirmed the relative abundance of protein EiKCS in these transgenic lines with a 1.16-fold (H 2 O), 1.10-fold (Para), and 1.32-fold (Para + spd) increase compared to the wild-type goosegrass on treatments. The quantification methodology showed that the total expressed mRNA and protein of the EiKCS in transgenic goosegrass were higher than the endogenous expression of those in wild-type goosegrass. Therefore, this transgenic goosegrass was confirmed as an engineering plant for overexpressing EiKCS.
With these treatments, the resistance phenotypes of the transgenic goosegrass were consistent with the relative expression levels of EiKCS ( Figure 6D). The paraquat-susceptible goosegrasses (MZ04) thoroughly died, while transgenic goosegrass survived with green leaves under the same paraquat stress. Moreover, the pre-treatment with exogenous spermidine significantly alleviated the toxicity of paraquat, making the stems and leaves of the plant green again. The fresh weights of goosegrass were statistically analyzed at 48 h after these treatments ( Figure 6E). The results showed that paraquat-susceptible goosegrasses were more significantly inhibited by paraquat, while the transgenic grass improved its tolerance to paraquat. The high expressions of the targeted EiKCS and its translated proteins positively affect the paraquat resistance of goosegrass. Meanwhile, the exogenous polyamine inhibited the level of the EiKCS expression, indicating that the EiKCS might involve in polyamine metabolism. The results are generally consistent with those of rice, but whether the same biosynthesis pathway of polyamines between goosegrass and rice still requires further study on this transgenic material. In summary, this effective Agrobacterium-mediated genetic transformation system in goosegrass is a beneficial biotechnological approach for exploring the gene functions of engineering weed species in response to abiotic stresses.

Discussion
Given the successful Agrobacterium-mediated genetic transformation in important crops, such as rice [30,31] and maize [32], this approach has been widely applied in many plant species. The breeding of transgenic herbicide-resistant crops has been drastically increased to address the limitations posed by weed damage and improve crop productivity [33][34][35]. However, further research is needed to identify genes with multiple resistances for creating herbicide-resistant crops and to explore the molecular mechanism of nontarget genes in herbicide detoxification [36,37]. In our previous studies, the mechanism of paraquat resistance was considered to involve non-target genes [27]. The EiKCS was found in goosegrass with not mutated in the resistant biotype (R-NX) compared to the susceptible biotype (S-HN). Further, the EiKCS has been shown as an ideal candidate gene of the paraquat-resistant genes for transgenic rice by promoting the polyamine synthesis involved with the arginine decarboxylase (ADC) pathway [29]. In this study, we transformed goosegrass for overexpression of EiKCS. We found that the transgenic goosegrass showed a more distinct paraquat-resistance based on visual injury assessment, as compared to the transgenic rice. However, the new molecular mechanism responses (ADC pathway or not) of the polyamine pathway regulated by the EiKCS gene to paraquat resistance could be further elaborated in further metabolic research of this transgenic goosegrass. Thus, analyses of paraquat-resistance phenotype, stable inheritance, and expressions of the transgenes showed that this Agrobacterium-mediated transformation method should be more useful in studying gene function and the related metabolism pathways in this weed species.
In plant Agrobacterium-mediated transformation processes, the problems such as bacterial contamination of seeds for callus induction, hydration and browning of calli are not uncommon. In this study, we established the first efficient Agrobacterium-mediated transformation system in goosegrass by use of a combination of optimal factors including explants and regulators on the callus induction and culture, Agrobacterium-mediated infection, and callus regeneration. It is worth mentioning that the contamination of QY04 and PY07 calli came from their seeds at rates of 11.83% and 21.89%, compared to almost no contamination in MZ04 under the same treatment of seed-sterilizing. This difference in the contamination rate of seeds may be related to their originally collected geographical locations. Second, primary calli were easy to hydrate in the process of the subculture. We found that in the subculture medium, the use of 7 g agar per L instead of 3 g phyto gel per L could effectively generate type-I calli to promote the growth of embryogenic callus, but did not affect the formation of type-I calli in the callus induction medium, possibly due to that use of 7 g agar in the subculture decreased the hydration degree of calli. Next, the browning of the calli also was caused by the growth of Agrobacterium cells on the surfaces of the calli after the infection. It is best to adjust the cell OD 600 to 0.5 during the infection, but it also is recommended to add an antibiotic (such as 100 mg L −1 timentin as used in this study) when the surfaces of the calli are contaminated severely by Agrobacterium cells. Thus, it is clear that we performed a thorough assessment of the methods and attempted to explain the reasons for testing certain components, how they were evaluated, and what the best recommendation should be.
In conclusion, this is the first establishment of an Agrobacterium-mediated genetic transformation in goosegrass, which is novel and has value in future research using similar methods for herbicide resistance research.

Plant Materials
The experiments were carried out at Weed Research Laboratory, South China Agricultural University, Guangzhou City, Guangdong Province, China. Natural seeds of four paraquat-susceptible biotypes (MZ04, PY07, JM and QY04) and one paraquat-resistant biotype (QY05) were selected from 16 biotypes of mature goosegrass in Guangdong province, which were MZ04, PY07, JM, QY04, and QY05 with their paraquat GR 50 values of 24.29, 54.15, 66.11, 67.29 and 314.43 g a.i.ha −1 respectively [38]. Among 16 biotypes of goosegrass, the MZ04 biotype was the most susceptible to paraquat (GR 50 24.29 g a.i.ha −1 ), while the QY05 biotype (GR 50 314.43 g a.i.ha −1 ) was the most resistant. Mature seeds of the MZ04 biotype goosegrass were used for the Agrobacterium-mediated transformation with EiKCS in this study. Paraquat-susceptible seeds of goosegrass (S-HN) and paraquat-resistant seeds of goosegrass (R-NX) in Guangdong Province of China were also used in determining developmental regulators in the genetic transformation system. The level of paraquatresistance in the R-NX goosegrass was 59.48-fold higher than that in the S-HN goosegrass as previously described [27].

General Improvements of the Transformation System in Goosegrass
The genetic transformation system of goosegrass was improved by determining the developmental regulators. A semi-manufactured product of N6, MS, and 1/2 MS medium (Qingdao Hope Bio-Technology Co., Ltd., Qingdao, China) respectively used as the basal N6 as a calli-induction, differentiation, and root-induction medium for calli induction of MZ04 biotype goosegrass. The product of hygromycin and timentin (RealTimes, Bio-Tech Co., Ltd., Beijing, China) solved a contamination problem by inhibiting Agrobacterium overgrowth on the surface of calli. The X-Gluc kit (Coolaber, Beijing, China) was used to detect the transient transformation of gus in goosegrass calli by monitoring the presence of blue color. Other reagents in the study were purchased from Sigma-Aldrich ® Brand (Merck-Sigma, Darmstadt, Germany and/or its affiliates). The explant, variety, and developmental regulators of goosegrass were analyzed using more than 50 biological replicates. Different tissues of goosegrass (R-NX) such as seeds, stems, leaves, and roots were used as explants for callus induction. The effects of the developmental regulators were evaluated on the induction of calli from goosegrass seeds. To break the seed dormancy, the goosegrass seeds were respectively treated by ddH 2

Vector Construction for Overexpressing EiKCS
Vector construction with the genetic map for overexpressing EiKCS was already performed in transgenic rice according to our previous studies, including the sequence of its translated protein EiKCS [29]. In brief, a paraquat-resistant gene (EiKCS/PqE) was cloned by PCR from R-NX and subsequently was constructed into a pCUbi1390 vector with the strong promoter of the maize Ubiquitin gene. The pCUbi1309::EiKCS construct was transferred into Escherichia coli DH5α and then introduced into Agrobacterium strain EHA105 by electroporation (1800 V/2 mm). Positive Agrobacterium colonies of pCUbi1309::EiKCS were screened using yeast extract broth (YEP) media with 50 mg mL −1 hygromycin and 20 mg mL −1 rifampicin [29]. Figure 7 shows a graphical overview of vector construction and transformation system for goosegrass.
with the strong promoter of the maize Ubiquitin gene. The pCUbi1309::EiKCS construct was transferred into Escherichia coli DH5α and then introduced into Agrobacterium strain EHA105 by electroporation (1800 V/2 mm). Positive Agrobacterium colonies of pCUbi1309::EiKCS were screened using yeast extract broth (YEP) media with 50 mg mL −1 hygromycin and 20 mg mL −1 rifampicin [29]. Figure 7 shows a graphical overview of vector construction and transformation system for goosegrass. Figure 7. The graphical overview of the vector construction and transformation system for goosegrass.

Transformation System for Goosegrass
The EHA105 cells containing pCUbi1309::EiKCS were used to infect goosegrass calli (Type Ⅰ) of the paraquat-susceptible biotype (MZ04). The Agrobacterium-mediated transformation process for goosegrass (MZ04) was divided into six steps. (1) Breaking seed dormancy. Mature goosegrass seeds of MZ04 were refrigerated at −20 °C for more than 30 days. (2) Seed sterilization. Seeds (MZ04) were soaked in 75% ethanol for 1 min, followed by surface sterilization using 0.05% HgCl2 solution for 10 min with stirring per 3 min. (3) Induction of calli. Each liter of callus induction medium was composed of 24.1 g basal N6, 0.5 g proline, 0.6 g hydrolytic casein, 10 g sucrose, 3 g phytagel, 1 mL 2,4-D (0.5 mg L −1 ), and with pH 5.8. Seeds (MZ04) were placed on the callus induction medium at 24 °C/24 h in darkness for 30 days. (4) Callus subcultures. Each liter subculture medium was composed of 24.1 g basal N6, 0.5 g proline, 0.6 g hydrolytic casein, 10 g sucrose, 7 g agar, 0.5 mL 2,4-D (0.5 mg L −1 ), and pH 5.8. The sponge tissue, buds, and seed coats attached to the calli needed to be removed. The newly formed pale yellow, compact, and nuclear embryogenic calli were transferred to the subculture dish containing the subculture medium at 24 °C/24 h in darkness for 60 days. After being transferred to new subculture dishes per 30 days, these calli grew to a diameter of about 1 cm. (5) Preparation of A. tumefaciens competent cells. The strain EHA105 containing the overexpression vector (pCUbi1309::EiKCS) was cultured in the 100 mL YEP media by 250 mL conical flask at 28 °C for 12 h at 230 rpm on an incubator shaker to obtain the OD600 at 0.5000-0.5999. (6) Callus infection, co-cultivation and screening. ① The dry granular calli with pale yellow (Type I callus) were pretreated at 4 °C for 1-2 days before in the 100 mL Agrobacterium solution completely dipped for 30 min. ② The calli spread over filter paper and covered with filter paper for 3 h air-dry, and then papers were replaced 4-5 times to absorb the excess bacterial solution. ③ The dry calli were transferred to a co-culture medium in 24 °C/24 h darkness for three days. The co-culture medium was the induction medium supplemented with 300 μmol/L acetosyringone (AS) and pH 5.2. ④ Then, the calli were

Transformation System for Goosegrass
The EHA105 cells containing pCUbi1309::EiKCS were used to infect goosegrass calli (Type I) of the paraquat-susceptible biotype (MZ04). The Agrobacterium-mediated transformation process for goosegrass (MZ04) was divided into six steps. (1) Breaking seed dormancy. Mature goosegrass seeds of MZ04 were refrigerated at −20 • C for more than 30 days. (2) Seed sterilization. Seeds (MZ04) were soaked in 75% ethanol for 1 min, followed by surface sterilization using 0.05% HgCl 2 solution for 10 min with stirring per 3 min. (3) Induction of calli. Each liter of callus induction medium was composed of 24.1 g basal N6, 0.5 g proline, 0.6 g hydrolytic casein, 10 g sucrose, 3 g phytagel, 1 mL 2,4-D (0.5 mg L −1 ), and with pH 5.8. Seeds (MZ04) were placed on the callus induction medium at 24 • C/24 h in darkness for 30 days. (4) Callus subcultures. Each liter subculture medium was composed of 24.1 g basal N6, 0.5 g proline, 0.6 g hydrolytic casein, 10 g sucrose, 7 g agar, 0.5 mL 2,4-D (0.5 mg L −1 ), and pH 5.8. The sponge tissue, buds, and seed coats attached to the calli needed to be removed. The newly formed pale yellow, compact, and nuclear embryogenic calli were transferred to the subculture dish containing the subculture medium at 24 • C/24 h in darkness for 60 days. After being transferred to new subculture dishes per 30 days, these calli grew to a diameter of about 1 cm. (5) Preparation of A. tumefaciens competent cells. The strain EHA105 containing the overexpression vector (pCUbi1309::EiKCS) was cultured in the 100 mL YEP media by 250 mL conical flask at 28 • C for 12 h at 230 rpm on an incubator shaker to obtain the OD 600 at 0.5000-0.5999. (6) Callus infection, co-cultivation and screening. 1 The dry granular calli with pale yellow (Type I callus) were pretreated at 4 • C for 1-2 days before in the 100 mL Agrobacterium solution completely dipped for 30 min. 2 The calli spread over filter paper and covered with filter paper for 3 h air-dry, and then papers were replaced 4-5 times to absorb the excess bacterial solution. 3 The dry calli were transferred to a co-culture medium in 24 • C/24 h darkness for three days. The co-culture medium was the induction medium supplemented with 300 µmol/L acetosyringone (AS) and pH 5.2. 4 Then, the calli were vigorously shaken for 30 s followed by standing for 5 min each time until the ddH 2 O became clear. This rinse was needed to change the ddH 2 O around 4-5 times. The calli were dipped for 30 min in a ddH 2 O supplemented timentin (100 mg L −1 ), and they were repeatedly filtered as in 2 after discarding this ddH 2 O. 5 The clean and dry calli were transferred to an A-screening medium for 25 days and then to a B-screening medium for 20 days. The Ascreening medium was the induced medium supplemented with hygromycin (15 mg L −1 ), carbenicillin (200 mg L −1 ) and timentin (100 mg L −1 ). The B-screening medium was the induced medium supplemented with hygromycin (20 mg L −1 ), carbenicillin (300 mg L −1 ) and timentin (100 mg L −1 ). All screening media altered 3 g phytagel to 7 g agar and pH to 6.0. During the screening period, the calli needed to be rinsed as in 4 , dried as in 2 , and transferred to a corresponding screening medium when A. tumefaciens appeared on the surface of the calli.

Regeneration System of Goosegrass Calli
The regeneration system of the goosegrass (MZ04) transformant calli after hygromycin screening was divided into four steps.  4) Single-plant propagation of transgenic goosegrass in the greenhouse. Transgenic goosegrass (T 0 plants) was self-pollinated to produce the T 1 generation, and T 2 generation was molecularly screened from the T 1 generation segregation under paraquat stress, and so on.

Parallel Reaction Monitoring
The PRM (Parallel Reaction Monitoring) was based on targeted quantitative proteome technology that used mass spectrometry (MS) to identify the target protein. The protein EiKCS was detected by PRM using its unique peptide segments (SGLGEETYLPAAVLR and CFGCVTQEEDGEGR of EiKCS) in the same samples of RT-PCR, which were three biological replicates. The method of PRM detection was previously reported [41].

Paraquat-Resistance Analysis of Transgenic EiKCS Goosegrass
The different lines of transgenic goosegrass (T 2 to T 6 generation) were sprayed at the vegetative and reproductive stages with 3750 mg L −1 paraquat (Gramoxone Max 2.5 pts/a, Syngenta, Basel, Switzerland) as the recommended dose in the field. The MZ04 biotype goosegrass was the control. The nine T 2 transgenic goosegrasses were sprayed with full-dose paraquat. The 15 T 3 transgenic goosegrasses from seeds of the same T 2 line (12-1-1) were sprayed with full-dose paraquat. Moreover, the plants of T 4 trangenic goosegrass in 28 pots from seeds of the same T 3 line (12-1-1-2) were sprayed with full-dose and half-dose paraquat. The 9 pots of them were sprayed with full-dose paraquat, and 3 pots were without spraying. And the 12 pots of them were sprayed with half-dose paraquat, and 4 pots were without spraying. Furthermore, the 24 seedings at the 4-6 leaves stage of T 5 transgenic goosegrass from the seeds of T 4 lines (12-1-1-2-8) were sprayed with full-dose paraquat. T 5 transgenic goosegrass planted in 4 pots was selected from the seeds of two T 4 lines (12-1-1-2-8 and 12-1-1-2-28) with each pot of four plants. The 3 pots of them were sprayed with 270 mg L −1 paraquat (Syngenta, Shanghai, China) and 1 pot without spraying, keeping the same treatment as the transgenic rice [20]. The paraquat-susceptible goosegrass (JM) and paraquat-resistance goosegrass (QY05) were treated similarly for comparison. In addition, the T 6 transgenic goosegrass lines from the seeds of T 5 lines (12-1-1-2-28-3) were sprayed with 270 mg L −1 paraquat (Syngenta, Shanghai, China) or sprayed with 270 mg L −1 paraquat 12 h after spraying with 1.5 mM spermidine, keeping the same treatments as the transgenic rice [20]. The paraquat-susceptible goosegrass (MZ04) and paraquat-resistance goosegrass (QY05) were treated similarly for comparison. Photographs were taken before and after paraquat treatments to record the resistant phenotypes of complete death or survival with green leaves in goosegrass. The seeds of T 6 transgenic goosegrass, MZ04, and QY05 were simultaneously sown and transplanted. Although MZ04 and QY05 grew better than T 7 transgenic goosegrass, their fresh weights were calculated 48 h after the same treatments.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to protecting parts of the data by a Chinese Patent (ZL202110962351.0).