ldentifying glyphosate-tolerant maize by soaking seeds in glyphosate solution

YAN Shu-feng , Sher Muhammad , LlU Hai-fang, TlE Shuang-gui SUN Shu-ku

1 Cereal Crops Research Institute, Henan Academy of Agricultural Sciences, Zhengzhou 450002, P.R.China

2 Jamil-ur-Rahman Center for Genome Research, International Center for Chemical and Biological Sciences, University of Karachi,Karachi 75270, Pakistan

3 Institute of Paulownia, Henan Agricultural University, Zhengzhou 450002, P.R.China

4 Qiqihar Fuer Agronomy Incorporation, Qiqihar 161041, P.R.China

Abstract The identification of glyphosate-tolerant maize genotypes by field spraying with glyphosate is time-consuming, costly and requires treatment of a large area. We report a potentially better technique of seed-soaking to identify glyphosate-tolerant maize genotypes. The effects of soaking maize seeds in glyphosate solution under controlled conditions were studied on seed germination rate, seedling morphological indices, seedling growth and leaf chlorophyll content. These responses were compared among a glyphosate-tolerant transgenic maize cultivar CC-2, glyphosate-susceptible inbred line Zheng 58(the recurrent parent of CC-2) and hybrid cultivar Zhengdan 1002. The results showed that the germination rate, seedling morphological indices and leaf chlorophyll content of glyphosate-tolerant CC-2 seeds did not change significantly among five different concentrations of glyphosate treatment (0 to 2%). In contrast, germination rates, seedling morphological indices and leaf chlorophyll contents of Zheng 58 and Zhengdan 1002 seeds were significantly negatively affected by exposure to increasing concentrations of glyphosate. The glyphosate-tolerant inbred line CC-2 displayed a strong tolerance to glyphosate after soaking in 0.1 to 2.0% glyphosate solutions, while both the inbred line Zheng 58 and hybrid Zhengdan 1002 were susceptible to glyphosate. The accuracy of the glyphosate-soaking method for screening glyphosate-tolerant maize was confirmed using a field spraying trial.

Keywords: EPSPS, glyphosate, leaf chlorophyll content, maize, seed-soaking

1. lntroduction

Maize (Zea mays L.) is the most productive and a widely grown crop around the world. Besides, being used in food products and animal feed, maize is also used extensively in industrial products such as ethanol biofuel production (Hill 2007; Shiferaw et al. 2011; Zhu et al. 2013). Competition from weeds results in worldwide crop yield losses of up to 30% in eight major crops, including maize (Armengot et al. 2013). In developing countries, millions of people are leaving rural areas, while in industrialized countries,workers have largely left agriculture. This loss of agricultural workers available to manually remove weeds has led to an increase in herbicide use for weed control (Gianessi 2013).Globally, glyphosate has become the most commonly used herbicide since its commercialization in 1974 because of its non-selective, broad spectrum and highly effective activity against weeds combined with its low toxicity to food crops and the absence of any residues (Spurrier 1973; Duke and Powles 2008a).

The herbicide glyphosate inhibits the activity of 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS),the sixth key enzyme in the shikimate pathway. Inhibition of this key enzyme causes the build-up of shikimic acid in plant tissues which reduces the biosynthesis of aromatic amino acids and chorismate-derived secondary metabolites, which ultimately causes plant death by starvation (Steinrücken and Amrhein 1980; Boocock and Coggins 1983; Wang et al.2016). Because glyphosate is a non-selective herbicide,its use can not only control broad-leaved and grassy weeds but it can also damage non-glyphosate-tolerant crops. The chlorophyll content, uptake and translocation of some mineral nutrients, seed quality and crop yield can be significantly affected by exposure to even very low concentrations of glyphosate or by glyphosate drift (Kitchen et al. 1981; Ellis et al. 2003; Cakmak et al. 2009). The development of transgenic glyphosate-tolerant maize was the solution to overcome the problem of glyphosate’s nonselectivity. Commercialized since 1996, the global cultivated area of genetically modified (GM) crops has increased by 100-fold from 1.7 million hectares in 1996 to 179.7 million hectares in 2015, where GM maize accounted for 53.6 million hectares and represented 29.83% of all GM crops grown in 2015 (James 2016). Taken together, glyphosatetolerant crops represent almost 90% of all GM crops grown worldwide (Duke and Powles 2008b).

The EPSPS genes have been classified as either glyphosate-sensitive class-I or glyphosate-tolerant class-II genes. EPSPS class-I genes have been reported in plant and bacteria species while glyphosate-tolerant EPSPS class-II genes have been predominantly cloned from bacteria (Funke et al. 2007; Jin et al. 2007; Tohge et al.2013). Researchers, such as Daniell et al. (1998) and Klee et al. (1987), generated glyphosate-tolerant plants by genetically modifying them with overexpressing EPSPS class-I genes derived from Arabidopsis thaliana and petunia, respectively. To date, however, no commercial GM crops tolerant to glyphosate have been generated to overexpress a wild-type EPSPS class-I gene. The existing glyphosate-tolerant commercial crops were developed through the introduction of various genes that contributed tolerance against glyphosate like EPSPS class-II genes such as CP4-EPSPS (Padgette et al. 1995; Chhapekar et al.2015), EPSPS class-I genes such as aroA and mEPSPS(Comai et al. 1985; Liu et al. 2015; Cui et al. 2016), the glyphosate-N-acetyltransferase (GAT) gene (Castle et al.2004; Green 2009; Lombardo et al. 2016) and the GOXV247 gene (Zhou et al. 1995; Kim et al. 2015). The GOXV247 gene alone, however, provided insufficient tolerance to glyphosate and was, therefore, always combined with a glyphosate-tolerant EPSPS gene in commercial applications(Dill 2005). Notably, a non-transgenic glyphosate-tolerant cotton mutant (R1098) was obtained by mutagenesis of EPSPS gene using Co60radiation that could tolerate 1.48 kg ha–1glyphosate, a glyphosate-tolerant hybrid cotton was then generated using the R1098 mutant as a parent(Tong et al. 2010; Zhu et al. 2011).

Glyphosate field spraying to identify glyphosate-tolerant maize plants is a time-consuming and inefficient method for many reasons. Plants are grown from seed and must reach an appropriate stage (4–6 leaf stage) before they can be sprayed by the post-emergence herbicide. Moreover, it can take more than a week for the symptoms of the glyphosate treatment to appear. Field spraying is inefficient because growing maize requires a large area, more labor and the use of large amounts of herbicide. Altogether, this is a costly assay. However, a seed-based assay conducted under controlled conditions for the detection of glyphosateresistant ryegrass has been developed by some researchers(Perez-Jones et al. 2005; Ghanizadeh et al. 2015).

Here, we present a novel technique of glyphosate application to screen for glyphosate-tolerant individuals:Maize seeds are soaked in a glyphosate solution. To our knowledge, this is the first report describing this method to screen for glyphosate-tolerant plants. The accuracy of identification of tolerant lines by this technique is similar to that of the field spraying method and may counter some of the issues associated with field-based identification.Therefore, this technique would be most useful to breeders and seed companies to control the purity of tolerant parent lines, determine the seed purity of F1glyphosate-tolerant GM hybrid maize and detect the degree of glyphosatetolerance in glyphosate-tolerant maize; it could also be used to positively identify transformed seeds of gene-stacked GM crops that use glyphosate as a selective agent.

2. Materials and methods

2.1. Plant materials

The transgenic glyphosate-tolerant cultivar CC-2, produced by the insertion of the maroACC gene (a modified GAT gene) from Agrobacterium tumefaciens, was supplied by China Agricultural University, Beijing, China and preserved by strict self-pollination. The glyphosate-tolerant cultivar Z58E-1 (containing a mutated EPSPS gene), glyphosatesusceptible hybrid Zhengdan 1002 and the glyphosatesusceptible inbred line Zheng 58 (which has a similar genetic background with CC-2 and Z58E-1), were provided by the Cereal Crops Research Institute, Henan Academy of Agricultural Sciences, China.

2.2. Seed-soaking in glyphosate solution

The seed-soaking experiment was conducted in a completely randomized design with three replications and five treatments. Seeds of CC-2, Zheng 58 and Zhengdan 1002 were soaked in five levels of 41% glyphosate (isopropylamine;Roundup) concentrations: (1) 0 (control); (2) 0.1%; (3) 0.5%;(4) 1.0%; and (5) 2.0% (Yan et al. 2016). After 14 h, the seeds (25 seeds/treatment) were sown in peat moss soil and vermiculite at a 1:1 ratio and placed in an incubator(26°C, 16 h light:8 h dark).

2.3. Survey of seedling morphological indices

After 6 d in the incubator, the germination rates of CC-2,Zheng 58 and Zhengdan 1002 seeds soaked in the different concentrations of glyphosate were calculated for each replicate. After 7 d, the fresh weight (seedlings and seeds)and leaf weight were also measured after washing off any plant debris. The stem and root lengths of each seedling were measured.

2.4. Determination of chlorophyll content in maize leaves

The chlorophyll content of leaves from each treatment was determined using a SPAD-502 Plus Portable Chlorophyll Meter (Minolta Camera Co., Ltd., Osaka, Japan); six measurements were taken from different parts of the maize leaf for each plant and averaged in each replicate.

2.5. Comparison of the effectiveness of the field identification and glyphosate seed-soaking methods for identifying glyphosate-tolerant individuals

The comparative experiment on the two methods of glyphosate application was conducted in the summer of 2016. There were three replicates (25 seeds per replicate)for each of four groups of genotypes of maize. The first was a test sample comprised of a known ratio (3:1) of CC-2 to Zheng 58 seeds. The other three types of seeds were from the glyphosate-tolerant genotype Z58E-1 (positive control)and two glyphosate-susceptible genotypes, Zheng 58 and Zhengdan 1002 (negative controls). All plants of each genotype in the field-spray treatment were sprayed with a 0.5% glyphosate solution at the 4–6 leaf stage in the field (T1). For the next three treatments, the seeds (25 seeds/replicate, three replicates for each of four groups of genotypes of maize) were soaked in 0.5% glyphosate solution before being sown in peat moss (T2), in sand culture(T3) and on wet filter paper (T4) under controlled conditions in an incubator (26°C, 16 h light:8 h dark). The germination rates of all four treatments were calculated for each replicate.

2.6. Statistical analysis

For data analyses, all traits were subjected to an analysis of variance using Data Processing System Statistical Software(DPS version 7.05, Zhejiang University, China) and the means were separated using an LSD test at the 5% level of probability.

3. Results

3.1. The effects of glyphosate seed-soaking on maize seed germination rate

After soaking the seeds in 0.1, 0.5, 1.0 or 2.0% glyphosate solutions or distilled water (0% glyphosate) for 14 h, followed by 6 d in a light incubator, the germination rates of CC-2,Zheng 58 and Zhengdan 1002 seeds were determined(Fig. 1). As the glyphosate concentration increased,the germination rate of glyphosate-tolerant CC-2 seeds did not change significantly. After seed-soaking in 0.1%glyphosate, however, the germination rates of Zheng 58 and Zhengdan 1002 seeds fell significantly. None of the Zheng 58 or Zhengdan 1002 seeds sprouted when treated with glyphosate concentrations of 0.5% or higher (Figs. 1 and 2). At the glyphosate concentration of 0.1%, the seed germination rates of CC-2, Zheng 58 and Zhengdan 1002 were 93.33, 29.33 and 57.33%, respectively; the glyphosatetolerant CC-2 germination rate was significantly higher than the glyphosate-susceptible hybrid, Zhengdan 1002. The germination rate of the hybrid, Zhengdan 1002, seeds was also significantly higher than that of the Zheng 58 inbred line at a glyphosate concentration of 0.1%.

3.2. The effects of glyphosate seed-soaking on maize seedling morphological indices and seedling growth

At 7 d after treatment, the fresh weight of CC-2, Zheng 58 and Zhengdan 1002 seedlings grown from glyphosatesoaked seeds was analyzed (Table 1). As the glyphosate concentration increased, the seedling fresh weight of CC-2 did not change significantly. Conversely, the fresh weights of Zheng 58 and Zhengdan 1002 seedlings were significantly reduced; glyphosate concentrations of 0.5% or greater completely inhibited seed germination so that no seedlings were available for testing. There were no significant differences among the fresh weights of CC-2, Zheng 58 and Zhengdan 1002 seedlings grown from seeds soaked in distilled water. When the glyphosate concentration was 0.1%, CC-2 seedlings had the highest fresh weight (1.91 g),followed by Zhengdan 1002 (1.21 g) and Zheng 58(0.77 g) seedlings; the differences in weight among these lines were significant. The seedling fresh weight of CC-2 was significantly higher than the seed weights of Zheng 58 and Zhengdan 1002 when treated with a concentration of glyphosate equal to or greater than 0.5%.

Fig. 1 The germination rates of CC-2, Zheng 58 and Zhengdan 1002 seeds after glyphosate-soaking. Data are mean±SD.

The maize seedlings were divided into three parts (root,stem and leaf) and the data for each part were analyzed separately (Table 1). Leaf weight, stem length and root length measurements showed patterns similar to the seedling fresh weight pattern. The leaf weight, stem length and root length of CC-2 did not change significantly at different glyphosate concentrations while the leaf weight,stem length and root length of glyphosate-susceptible Zhengdan 1002 and Zheng 58 fell significantly when soaked in 0.1% glyphosate concentration. At this concentration, the respective measurements for CC-2 were all significantly higher than those for Zhengdan 1002 and Zheng 58. The leaf weight, stem length and root length of Zhengdan 1002 and Zheng 58 in glyphosate concentrations equal to or greater than 0.5% were not measured as the seeds soaked in these concentrations did not germinate.

The effects of glyphosate seed-soaking on maize seedling growth are shown in Fig. 2. The growth and development of glyphosate-tolerant CC-2 were normal and no significant differences were observed among the different glyphosate concentrations. Seedlings of the glyphosate-susceptible inbred line Zheng 58 were dwarfed with ultra-short roots and smaller, colorless leaves after 0.1% glyphosate treatment. The seedling phenotypes of the glyphosate-susceptible hybrid Zhengdan 1002 treated with 0.1% glyphosate were variable; most seedlings exhibited symptoms of herbicide damage, although a small number of them appeared relatively normal.

3.3. The effects of glyphosate seed-soaking on maize leaf chlorophyll content

When soaked in distilled water, the leaf chlorophyll contents of glyphosate-tolerant CC-2, glyphosate-susceptible inbred line Zheng 58 and hybrid Zhengdan 1002 were 28.34, 27.36 and 27.26 SPAD unit, respectively; these values did notdiffer significantly. When treated with 0.1% glyphosate, the chlorophyll content of CC-2 was significantly higher than that of Zheng 58 and Zhengdan 1002. Chlorophyll contents of CC-2 did not differ significantly at different glyphosate concentrations. When comparing the chlorophyll contents of Zheng 58 and Zhengdan 1002 seedlings, content of those seedlings in the distilled water treatment was significantly higher than that of seedlings in the 0.1% glyphosate treatment. Glyphosate concentrations higher than 0.1%completely inhibited germination in these two lines, resulting in the lack of seedlings available for measurement at these concentrations (Table 2).

Table 1 Seedling fresh weight, fresh leaf weight, stem length and root length of CC-2, Zheng 58 and Zhengdan 1002, 7 d after soaking of seeds in different glyphosate concentrations

3.4. Comparison of the field identification and glyphosate-soaking methods

The effectiveness of the field identification and glyphosatesoaking techniques on identifying the seed purity of glyphosate-tolerant maize was assessed. The test sample(comprised of seeds of CC-2 and Zheng 58 mixed at a 3:1 ratio), along with glyphosate tolerant Z58E-1 as a positive control and glyphosate-susceptible Zheng 58 and Zhengdan 1002 as negative controls, were sprayed with a 0.5% glyphosate solution at the 4–6 leaf stage in the field (T1) and seeds of each line were soaked in 0.5%glyphosate solution before being sown in a peat moss (T2),in sand culture (T3) or on wet filter paper (T4) (Table 3).The germination rates of glyphosate-susceptible Zheng 58 and Zhengdan 1002 were 0 in each treatment while the germination rate of glyphosate-tolerant Z58E-1 was 100% in each treatment. For the test sample, the germination rates were 76.00, 73.33, 78.67 and 74.67% in treatments T1to T4,respectively. The germination rates of the test sample did not differ significantly across the four treatments, indicating that the glyphosate-soaking method was as accurate as the field spraying method for screening glyphosate-tolerant maize, regardless of whether the seeds were sown in peat moss, in sand culture or on wet filter paper. Therefore,seed-soaking is a feasible method to determine the seed purity of glyphosate-tolerant maize.

4. Discussion

Fig. 2 The effects of glyphosate-soaking on CC-2, Zheng 58 and Zhengdan 1002 seedling growth. A, glyphosate-tolerant maize CC-2. B, glyphosate-susceptible inbred Zheng 58. C, glyphosate-susceptible hybrid Zhengdan 1002. Glyphosate-soaking concentrations of 0.1, 0.5, 1.0 or 2.0% are indicated, with 0 representing the negative control of distilled water.

Table 2 The leaf chlorophyll contents (SPAD unit) of CC-2, Zheng 58 and Zhengdan 1002 seedlings 7 d after glyphosate-soaking

Table 3 The germination rates of the test sample and relevant controls with different treatment methods

The phenomenon of heterosis has been exploited in maize breeding since 1908, with hybrid maize now accounting for most of the maize acreage worldwide (Crow 1998; Feng et al.2014). Incomplete detasseling or incomplete inactivation of pollen has resulted in self-pollination of the female inbred lines and a resultant decrease in hybrid seed purity that has led to reductions in maize yield (Astini et al. 2009). Existing methods to determine the level of self-inbreeding occurring in F1hybrids include the grow-out test, isozyme analysis and isoelectric focusing and the use of AFLP (amplified fragment length polymorphism), RAPD (random-amplified polymorphic DNA), ISSR (inter-simple sequence repeat),SRAP (sequence-related amplified polymorphism) and SSR (simple sequence repeat, microsatellite) markers(Crockett et al. 2002; Liu et al. 2007; Zhao et al. 2012).Compared with these methods, however, a simpler method of determining the seed purity of F1glyphosate-tolerant GM hybrid maize involves the use of glyphosate tolerance genes as selectable markers. Following hybridization of the female inbred line (without glyphosate-tolerant genes)and male inbred line (carrying glyphosate-tolerant genes),F1hybrid seeds containing the glyphosate-tolerant genes could be identified by successful germination after soaking the seeds in a glyphosate solution. Self-pollinated seeds of the female inbred line soaked in a 0.1% glyphosate solution would produce seedlings displaying herbicide damage;when soaked in a solution of 0.5% (or higher) glyphosate,these seeds would not germinate. Furthermore, the glyphosate-soaking method could also be used to control the glyphosate-tolerant seed purity of female or male parents in backcrossing or self-crossing experiments in place of the more time-consuming and expensive field-spraying method.

The seed germination rate, growth and vigor of seedlings and tolerance to biotic and abiotic stresses have previously been shown to increase following seed-soaking in plant growth regulators, micronutrients, fungicides and insecticides (Thomas 1973; Imran et al. 2013; Li et al. 2013;Sharma et al. 2016). However, the use of the glyphosate herbicide as a seed-soaking agent to identify glyphosate tolerance characteristics of GM maize, to evaluate the seed purity of F1hybrids and to control the purity of glyphosatetolerant genes in the parents in backcrossing or selfcrossing experiments has not been previously reported.The results of this study show that the glyphosate-soaking method was as accurate as the field-spraying method for screening glyphosate-tolerant maize. Moreover, the soaking method was less expensive and time-consuming.As seed-soaking agents can often be used as seed coating agents, it is possible that glyphosate can be applied as a seed coating agent in future seed production (Gressel and Joel 2000; Tekrony 2006; Zeng and Shi 2009). The glyphosate seed-soaking method may also be suitable for other glyphosate-tolerant crops or gene-stacked GM crops that use glyphosate as a selective agent, although the correct soaking time and glyphosate concentration will be determined again in each instance.

Because many studies have shown that the application of glyphosate severely inhibits leaf chlorophyll content, seed germination and seedling growth of glyphosate-susceptible crops, the germination rate, seedling morphological indices and leaf chlorophyll content after glyphosate treatment are useful parameters to measure glyphosate damage(Hoagland 1980; Ketel 1996; Reddy 2004; Gomes et al.2016; Wu et al. 2016). Increasing glyphosate concentrations in the soaking assay did not significantly affect the germination rate, seedling morphology or leaf chlorophyll content of glyphosate-tolerant CC-2. These parameters were, however, negatively affected in the glyphosatesusceptible Zheng 58 and Zhengdan 1002 lines, suggesting that the CC-2 glyphosate-tolerance gene was activated in the same way during the glyphosate seed-soaking process as it was following glyphosate field-spraying.

5. Conclusion

The effects of soaking maize seeds in glyphosate solution under controlled conditions were studied on different parameters of maize by using the glyphosatetolerant transgenic maize cultivar CC-2, the glyphosatesusceptible inbred line Zheng 58 and the hybrid cultivar Zhengdan 1002. The glyphosate-tolerant inbred line CC-2 displayed a strong tolerance to glyphosate after soaking in 0.1 to 2.0% glyphosate solutions, while both the inbred Zheng 58 and hybrid Zhengdan 1002 lines were susceptible to glyphosate. To our knowledge, this is a novel technique in which maize seeds are soaked in a glyphosate solution to screen for glyphosate tolerance. This method was as accurate as field spraying for the identification of tolerant lines and due to its ease of application and more cost-efficient advantages, its implementation could counter some of the issues associated with field-based identification. This technique could be used by breeders and seed companies to control the purity of tolerant parent lines, determine the seed purity of F1glyphosate-tolerant GM hybrid maize and detect the degree of glyphosate-tolerance in glyphosatetolerant maize; it could also be used to positively identify transformed seeds of gene-stacked GM crops that use glyphosate as a selective agent.

Acknowledgements

This study was funded by grants from the Genetically Modified Organisms Breeding Major Project of China(2016ZX08011-003) and the Science and Technology Cooperation Project of Henan Province and the Chinese Academy of Agricultural Sciences ( 162106000012).