Skip to main content
NCBI home page
Microbes and Environments logoLink to Microbes and Environments
. 2020 Nov 7;35(4):ME20056. doi: 10.1264/jsme2.ME20056

Impact of a G2-EPSPS & GAT Dual Transgenic Glyphosate-Resistant Soybean Line on the Soil Microbial Community under Field Conditions Affected by Glyphosate Application

Minkai Yang 1,†, Zhongling Wen 1,†, Aliya Fazal 1,†, Xiaomei Hua 2, Xinhong Xu 1, Tongming Yin 3, Jinliang Qi 1, Rongwu Yang 1, Guihua Lu 1,3,4,*, Zhi Hong 1,*, Yonghua Yang 1,3,*
PMCID: PMC7734404  PMID: 33162465

Abstract

In the past thirty years, the biosafety of the aboveground part of crops, including horizontal gene transferal through pollen dispersal and hybridization, has been the focus of research; however, microbial communities in the underground part are attracting increasing attention. In the present study, the soybean root-associated bacterial communities of the G2-EPSPS plus GAT transgenic soybean line Z106, its recipient variety ZH10, and Z106 treated with glyphosate (Z106J) were compared at the seedling, flowering, and seed filling stages by high-throughput sequencing of the V4 hypervariable regions of 16S rRNA gene amplicons using Illumina MiSeq. The results obtained showed no significant differences in the alpha/beta diversities of root-associated bacterial communities at the three stages among ZH10, Z106, and Z106J under field growth conditions; however, the relative abundance of four main nitrogen-fixing bacterial genera significantly differed among ZH10, Z106, and Z106J. Ternary plot results indicated that in the root compartment, the proportional contributions of rhizobial nitrogen-fixing Ensifer fredii and Bradyrhizobium elkanii, which exhibit an extremely broad nodulation host range, markedly differed among the three treatments at the three stages. Thus, the present results indicate that transgenic G2-EPSPS and GAT soybean may induce different changes in functional bacterial species in soil, such as E. fredii and B. elkanii, from ZH10, which were compensated for/enriched at the flowering and seed filling stages, respectively, to some extent through as of yet unknown mechanisms by transgenic soybean treated with glyphosate.

Keywords: G2-EPSPS & GAT genes, transgenic soybean, glyphosate, root-associated bacterial community, Ensifer fredii

The world’s first genetically modified (GM) plant was introduced in 1983, and the United States of America took the lead in promoting the industrialization of GM crops in 1996 (Brookes and Barfoot, 2013). After entering the 21st century, the development momentum of this industrialization has become increasingly fierce worldwide. Developed and developing countries are both increasingly focusing on the application of GM technology in agriculture, including China. After 20 years of development, China has achieved advances in the field of biotechnology research and development (Zhang et al., 2018; Feng and Yang, 2019). While GM crops bring significant benefits to humans, biosafety issues, including environmental and biological factors, need to be considered (Benbrook, 2016; ISAAA, 2017; 2018). The impact of GM herbicide-resistant crops on the soil microbial ecosystem has also become the focus of GM safety assessments (Dale et al., 2002; Babujia et al., 2016), particularly the importance of and issues associated with the development of GM soybean in China, which is partly due to its origin.

Two approaches are currently used to obtain glyphosate-resistant soybean: the introduction of the glyphosate-tolerant 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) gene into GM soybeans to improving their tolerance as well as the introduction of a glyphosate-degradable gene, which degrades glyphosate before it exerts its effects, thereby reducing damage to soybean (Duke and Powles, 2008; Guo et al., 2015). Many genes have been found to degrade glyphosate, and the glyphosate N-acetyltransferase (GAT) gene exhibits good resistance to glyphosate. Under the effects of GAT, glyphosate may be converted into N-acetylglyphosate, which is non-toxic to plants (Castle et al., 2004).

Three soybean treatments were examined in the present study: the soybean cultivar Zhonghuang10 (ZH10) as the recipient, its G2-EPSPS & GAT dual transgenic soybean cultivar ZH10-6 (Z106) (Guo et al., 2016), and ZH10-6 treated with glyphosate (Z106J), with the following aims: 1) to compare and analyze the composition and abundance of root-associated bacterial communities among three soybean treatments with three different sampling compartments at the seedling, flowering and seed filling stages, and 2) to elucidate the impact on the enrichment/derichment of specific bacterial species under field growth conditions using the high-throughput sequencing of 16S rRNA gene (V4 region) amplicons via the Illumina MiSeq sequencing platform.

Materials and Methods

Plant materials and sampling methods

The recipient soybean cultivar ZH10, the G2-EPSPS and GAT transgenic herbicide-tolerant soybean line Z106, and Z106 treated with glyphosate (Z106J) were used as three treatments in this experiment (Guo et al., 2016). All soybean lines were treated with diammonium hydrogen phosphate ([NH4]2HPO4) 300 kg hectare–1, and potassium chloride (KCl) 150 kg hectare–1 before planting, while no fertilizer was applied during the experiment. Regarding Z106J, glyphosate was applied at 900‍ ‍g hectare–1 at the seedling stage. The experimental field (the Shunyi Experimental Field Station of the Crop Research Institute, CAAS) is located in Shunyi district, Beijing, China (N 40.237°, E 116.570°). This field was divided into 36 plots (5×4‍ ‍m per plot) with a randomized complete block design for national joint experiments in May 2015, and the layout and design of the experimental field is shown in Fig. S1 (Lu et al., 2017), including the above three treatments with three replications. The climate conditions of the experimental location were a sub-humid warm temperate continental monsoon climate. Experimental samples were collected at the seedling stage on July 5, flowering stage on July 29, and seed filling stage on August 24 in 2015.

At all three stages, each soybean treatment was sampled from three different plots with two sampling points per plot, which were randomly distributed across the field, and composite samples from two sampling points per plot were made and treated as one replicate. Regarding the naming of samples, A represents bulk soil, i.e., soil before planting, while B, C, and D represent the seedling (V4–V5), flowering (R1–R2), and seed filling (R5–R6) stages, respectively. The SO, RH, and RT compartments represent surrounding soil, rhizospheric soil, and root samples, respectively. Soil loosely adhering to the roots was shaken off the soybean plant as the surrounding soil sample (SO). The rhizospheric soil sample (RH) was collected by brushing off soil that tightly adhered to the root surface. The root sample (RT) (a mixed sample of the rhizoplane/endosphere and the layer of endophytes in roots) was collected after washing with phosphate-buffered saline (PBS) (Lu et al., 2017). All samples were then stored at 4°C for a basic physicochemical analysis and at –80°C for DNA extraction (Lu et al., 2018a, 2018b).

Basic physicochemical properties of soils and plants

To analyze the carbon and nitrogen contents of surrounding soil and dried plant tissues, all samples were sent to the Modern Analysis Center of Nanjing University. The pH and water content of samples were also tested for surrounding soil. In the acetylene reduction assay (ARA) on surrounding soil, 10‍ ‍g of each sample was mixed with 1.4‍ ‍mL of 0.1 mol L–1 glucose in a 65‍ ‍mL conical flask, followed by an injection of 7‍ ‍mL of acetylene instead of air with a syringe and an incubation at 28°C for 15 hours before measurements of ethylene production by gas chromatography.

DNA extraction from soil and root samples

In the present study, every biological replicate was a mixture of total metagenomic DNA extracted using approximately 0.6‍ ‍g soil from each sampling point by the Power Soil DNA Isolation Kit (MoBio Laboratories), following the instructions described by Lu et al. (2017) with minor modifications. In root sample preparation, 2×0.7‍ ‍g of root segments from the root tip of every biological replicate was carefully homogenized with liquid nitrogen using a mortar and pestle and then subjected to total metagenomic DNA extraction using the same method described above.

16S rRNA gene amplicon sequencing via the Illumina MiSeq platform

We used an improved dual-index high-throughput sequencing approach with paired-end 250 nt. A ‘heterogeneity spacer’ (0 to 7‍ ‍bp) was introduced into the index sequence, which allowed an equal proportion of samples to be sequenced (Fadrosh et al., 2014). In brief, fusion primers included appropriate P5 or P7 Illumina adapter sequences, an 8-nt index sequence, and gene-specific primers for amplifying the V4 region of 16S rRNA gene (Peiffer et al., 2013), namely, 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) (Caporaso et al., 2010). PCR amplification, product purification, and library quality determinations and quantification were performed as previously described by Lu et al. (2017). High-throughput sequencing of the qualified libraries was performed on the Illumina MiSeq platform (Illumina) with the MiSeq Reagent Kit by BGI Tech Solutions. The sequencing clean data of 90 samples have been submitted to the Sequence Read Archive (SRA) with the SRA accession number PRJNA613555.

Analysis of 16S rRNA gene amplicon sequencing data

A total of 10,532,365 qualified paired-end clean reads with an average count per sample of 102,256 (range: 44,754–161,581) were obtained from all samples at the seedling, flowering, and seed filling stages. Total qualified paired clean reads at the seedling stage were 2,933,264 and the average count per sample was 97,775‍ ‍(range: 44,754–140,183). At the flowering stage, total qualified paired-end clean reads were 3,361,258 and the average count per sample was 112,042 (range: 50,930–156,604). At the seed filling stage, total qualified paired-end clean reads were 2,989,410 and the average count per sample was 87,924 (range: 49,695–145,012) (Table S1). Operational taxonomic units (OTUs) were selected as previously described (Lu et al., 2018a, 2018b) with minor modifications. Chimeras were filtered out using UCHIME (v4.2.40) and connected tags were filtered to eliminate low quality and short sequences using QIIME (v1.7.0) before clean tags were obtained (Edgar et al., 2011; Edgar, 2013). As described previously by Lu et al. (Lu et al., 2018a, 2018b), OTU counts in each sample’s library were normalized after species annotation and phylogenetic tree construction, mitochondrion/chloroplast-related sequences (phylum Cyanobacteria, family Mitochondria) from the sequence data set were simultaneously removed, and all of the operations described above were conducted in the I-Sanger platform (http://www.i-sanger.com). Alpha and beta diversity analyses were then conducted based on OTUs and species annotation results (Wen et al., 2019). In addition, the OTU rank abundance curve was drawn to reflect species richness and evenness (Fig. S2), and the Pan/Core OTU analysis was used to observe increases in total species and decreases in common species with higher sample numbers (Fig. S3).

Alpha diversity, beta diversity, and the taxonomic analysis

Alpha diversity is applied to analyses of the complexity of species diversity for a sample, which may be expressed by different indices, including observed species, Chao1, ACE, Shannon, and Simpson (Fig. 1 and S4) (Schloss et al., 2009), and a boxplot of alpha diversity was made by Graphpad Prism. The beta diversity analysis is used to assess differences in species complexity in samples through several values, such as Bray-Curtis, weighted UniFrac (WUF), and unweighted UniFrac (UUF). Beta diversity was calculated by QIIME (Caporaso et al., 2010) to evaluate differences in species complexity in samples among ZH10, Z106, and Z106J. A principal coordinate analysis (PCoA) based on WUF was also performed with QIIME (Peiffer et al., 2013; Tian et al., 2019). A Partial Least Squares Discriminant Analysis (PLS-DA) was performed with the mix Omics package of software R (Fig. S5) (Rohart et al., 2017). Taxa clustering was conducted based on the relative abundance of each taxon; longitudinal clustering indicates the similarity of all taxa among different samples, while horizontal clustering indicates the similarity of certain taxa among different samples (Fig. S6) (Jie et al., 2019). All data were processed in I-Sanger (Wen et al., 2019). In addition, according to previous studies (Lu et al., 2017; Wen et al., 2019), the relative abundance of the main nitrogen-fixing bacterial genera was compared using a one-way ANOVA, and four genera that showed significant differences among the three treatments, namely, Bradyrhizobium (Kaneko et al., 2002; de Souza et al., 2012), Bacillus (Seldin and Dubnau, 1985; Talebi et al., 2013), Cupriavidus (Andam et al., 2007; Liu et al., 2012), and Ochrobactrum (Ngom et al., 2004; Wackerow-Kouzova, 2007), were examined in more detail.

Fig. 1.

Fig. 1.

Open in a new tab

A boxplot of alpha diversity among three treatments. A) Observed OTUs index of the seedling stage; B) Observed OTUs index of the flowering stage; C) Observed OTUs index of the seed filling stage; D) Shannon index of the seedling stage; E) Shannon index of the flowering stage; F) Shannon index of the seed filling stage. Z106 and ZH10 represent the transgenic soybean (ZH10-6) and its recipient cultivar (Zhonghuang 10), respectively. Z106J represents Z106 treated with glyphosate. A represents soil before planting, B, C, and D represent the seedling, flowering, and seed filling stages, respectively. SO, RH, and RT represent surrounding soil, rhizospheric soil, and root samples, respectively. The asterisk (*, P<0.05) indicates a significant difference according to the Student’s t-test.

Ternary plot analysis

A ternary plot is an equilateral triangle describing the ratio relationship of the different attributes of three variables. In the analysis, the species composition of three groups of samples may be compared and analyzed according to species classification information. The proportion and relationship of different species in samples may be visually displayed by a triangle diagram. A ternary plot was drawn by GGTERN (http://www.ggtern.com/) (Bulgarelli et al., 2012; Schlaeppi et al., 2014).

Statistical analysis

Our samples belonged to different groups, and the sample number per group was 3. The Student’s t-test for alpha diversity was performed in I-Sanger. An analysis of similarities (ANOSIM) and permutational MANOVA (PERMANOVA/Adonis) were complementary, non-parametric analyses that were performed using the database Silva128/16S, with/without subsampling, as well as the vegan package of software R (v3.1.3) in I-Sanger based on the Bray–Curtis, WUF, and UUF distance metrics (Schloss et al., 2009; Wen et al., 2019).

Ethics statement

The Ministry of Agriculture of the People’s Republic of China issued permissions for the locations. Field studies did not involve endangered species. The experimental field was not protected or privately owned in any way.

Results

Analysis of basic physicochemical properties of soils and plants

According to the information obtained from the China Soil Database (http://vdb3.soil.csdb.cn/), the soil type in Shunyi district, Beijing city, is Aquic cinnamon soil. As shown in Table 1, the results of soil pH and water content showed significant differences among the three soybean treatments from the seedling to seed filling stages. The soil nitrogen content did not significantly differ among the three treatments at the three stages, while a significant difference was observed in the soil carbon content at the seed filling stage only. The soil nitrogenase activities of Z106 and Z106J were significantly higher than that of ZH10 at the seedling stage only. Significant differences were observed in the plant carbon content at the seedling and flowering stages, while the plant nitrogen content significantly differed at all three stages among the three treatments.

Table 1.

Surrounding soil and plant analyses of ZH10, Z106, and Z106J showed significant differences in some traits among three treatments.

Analysis Trait Seedling Stage (Mean±SD) Flowering Stage (Mean±SD) Seed Filling Stage (Mean±SD)
ZH10 Z106 Z106J ZH10 Z106 Z106J ZH10 Z106 Z106J
Soil analysis pH value 8.12±0.04a 7.79±0.14b 7.82±0.05b 7.95±0.04b 8.01±0.03b 8.11±0.03a 7.92±0.04b 8.00±0.02a 7.96±0.01b
Water content (%) 10.85±0.01a 12.67±0.01b 11.53±0.01ab 14.65±0.00b 15.54±0.00a 14.70±0.00b 10.58±0.01b 10.98±0.01b 11.48±0.00a
C content (%) 0.95±0.05 1.00±0.03 1.00±0.05 1.03±0.10 0.96±0.06 0.97±0.06 0.99±0.03b 1.12±0.02a 0.97±0.06b
N content (%) 0.10±0.00 0.11±0.01 0.10±0.01 0.10±0.01 0.10±0.01 0.10±0.01 0.10±0.01 0.10±0.01 0.10±0.01
Ethylene production (nmol [g*h]–1) 14.91±4.90a 22.53±2.28b 23.64±9.69b 56.64±23.71 63.53±14.98 52.33±14.02 21.06±5.11a 25.30±4.89ab 32.27±5.25b
Plant analysis C content (%) 38.38±0.20a 39.91±0.41b 39.76±0.63b 40.17±0.81b 37.75±1.15a 39.38±0.55b 38.79±1.36 38.88±0.56 39.51±0.57
N content (%) 4.03±0.11a 4.41±0.16b 4.62±0.11c 2.96±0.17a 3.37±0.13b 3.40±0.15b 2.34±0.18a 2.56±0.26ab 2.85±0.10b
Open in a new tab

SD represents the standard deviation (n=3). Z106 and ZH10 represent transgenic soybean ZH10-6 and its recipient cultivar Zhonghuang 10. Z106J represents Z106 treated with glyphosate. C and N contents represent carbon and nitrogen contents, respectively. Statistical analyses were performed using a one-way ANOVA. The values in bold indicate a significant difference (P<0.05) among the ZH10, Z106, and Z106J groups.

Alpha diversity of root-associated bacterial communities among ZH10, Z106, and Z106J

The mean and standard deviation (SD) were calculated based on six alpha diversity indices of all replicates of the surrounding soil, rhizospheric soil, and root samples of ZH10, Z106 and Z106J, and the Student’s t-test was used for multigroup comparisons. The results obtained revealed no significant differences in alpha diversity among the three treatments at the three stages, except for some comparison groups in one alpha diversity index (Table S2), such as the Observed OTUs in the rhizospheric soil samples of ZH10 and Z106 at the seed filling stage, and the Shannon index of the root samples of ZH10 and Z106 at the flowering stage (Fig. 1 and S4). In addition, the Observed OTUs and Shannon index of root samples at the seedling, flowering, and seed filling stages were markedly lower than those of the surrounding soil and rhizospheric soil samples (Fig. 1).

Beta diversity of root-associated bacterial communities among ZH10, Z106, and Z106J

The normalized abundance of each OTU in each sample was calculated based on the OTU table for the biom format, and a beta diversity analysis was performed by PCoA based on the WUF distance (Fig. 2) of the root-associated bacterial communities of ZH10, Z106, and Z106J. The distance between samples was assessed by composition similarities among all soil samples. As shown in Fig. 2, bacterial communities in the surrounding soil, rhizospheric soil, and root samples of ZH10, Z106, and Z106J were not distinct from each other at the seedling, flowering, or seed filling stages, indicating no marked difference among the three treatments. In addition, the bacterial communities in the root samples of ZH10, Z106 and Z106J were distinct from those in the surrounding soil and rhizospheric soil samples at all three stages. The results of PLS-DA were also consistent with the above results (Fig. S5).

Fig. 2.

Fig. 2.

Open in a new tab

PCoA based on the weighted UF distance matrix at three stages among three treatments. A) Seedling; B) flowering; C) seed filling. Z106 and ZH10 represent the transgenic soybean (ZH10-6) and its recipient cultivar (Zhonghuang 10), respectively. Z106J represents Z106 treated with glyphosate. A represents soil before planting, B, C, and D represent the seedling, flowering, and seed filling stages, respectively. SO, RH, and RT represent the surrounding soil, rhizospheric soil, and root samples, respectively.

We then performed statistical ANOSIM and Adonis on bacterial communities. The results of ANOSIM and Adonis based on the Bray–Curtis distance both indicated that the taxonomic beta diversities of bacterial communities among ZH10, Z106, and Z106J in the same compartment at one stage were not significantly different (Table S3).

In order to verify the results obtained, we performed ANOSIM and Adonis based on WUF and UUF (Table S3), and these results also indicated no significant differences among the phylogenetic beta diversity of bacterial communities among ZH10, Z106, and Z106J in the same compartment at different stages.

Comparative analysis of taxonomic levels of root-associated bacterial communities among ZH10, Z106, and Z106J

We used a ternary plot analysis to identify the bacteria responsible for the observed community differentiation at the species level and the results obtained showed no marked differences among the three treatments in the surrounding soil or rhizospheric soil samples; however, several species had higher proportions in root samples (Fig. S7C), including Ensifer (previously described as Sinorhizobium) fredii belonging to the family Rhizobiaceae, order Rhizobiales, and class Alphaproteobacteria within the phylum Proteobacteria. In comparisons with Z106, E. fredii accounted for a lower proportional contribution in Z106J at the seedling stage (Fig. S7D) and a higher proportional contribution in ZH10 and Z106J at the flowering stage (Fig. S7E). In the seed filling stage, the proportional contribution of E. fredii among the three treatments appeared to be similar (Fig. S7F). Based on these results, we conducted the same analysis of root samples at the three stages among the three treatments, and the results obtained were consistent with the ternary plots shown in Fig. S7D, E, and F (Fig. 3). Regarding the species Bradyrhizobium elkanii, which belongs to the family Bradyrhizobiaceae, order Rhizobiales, and class Alphaproteobacteria within the phylum Proteobacteria, its proportional contribution in Z106 was higher than that in ZH10 and Z106J at the seedling and flowering stages, but was lower at the seed filling stage.

Fig. 3.

Fig. 3.

Open in a new tab

Ternary plot of species based on different treatments in root samples at three stages. Different proportional contributions were observed: A) Seedling; B) flowering; C) seed filling. Z106 and ZH10 represent the transgenic soybean (ZH10-6) and its recipient cultivar (Zhonghuang 10), respectively. Z106J represents Z106 treated with glyphosate. B, C, and D represent the seedling, flowering, and seed filling stages, respectively. RT represents the root sample.

We also compared the relative abundance of major five bacterial taxa at the phylum, class, order, family, genus, and species taxonomic levels using a one-way ANOVA (Fig. S8). The results revealed that the relative abundance of Betaproteobacteria (Fig. S8D), Burkholderiales (Fig. S8G) and Comamonadaceae (Fig. S8J) in root samples at the seedling stage was significantly higher in Z106J than in Z106, while the relative abundance of Burkholderiales (Fig. S8G) and Comamonadaceae (Fig. S8J) was significantly higher in ZH10 than in Z106. Regarding root samples at the flowering stage, the relative abundance of Alphaproteobacteria (Fig. S8E), Rhizobiales (Fig. S8H), and Rhizobiaceae (Fig. S8K) was significantly higher in ZH10 than in Z106.

Comparison of the composition of main nitrogen-fixing bacterial genera at different growth stages

We examined the relative abundance of the main nitrogen-fixing bacterial genera and found that the relative abundance of Cupriavidus was significantly higher in the Z106J root sample than in the Z106 root sample at the seedling and seed filling stages (Fig. 4A and C). At the flowering stage, the relative abundance of Cupriavidus was significantly higher in the Z106 root sample than in the ZH10 and Z106J root samples (Fig. 4B). Regarding root samples at the seed filling stage, the relative abundance of Bradyrhizobium was significantly higher in ZH10 than in Z106, while the relative abundance of Bacillus and Ochrobactrum was significantly higher in Z106 than in Z106J, and Ochrobactrum was present at a significantly higher abundance in Z106 than in ZH10 (Fig. 4C).

Fig. 4.

Fig. 4.

Open in a new tab

Relative abundance of main nitrogen-fixing bacterial genera among three treatments at three stages. A) Seedling; B) flowering; C) seed filling. Z106 and ZH10 represent the transgenic soybean (ZH10-6) and its recipient cultivar (Zhonghuang 10), respectively. Z106J represents Z106 treated with glyphosate. B, C, and D represent the seedling, flowering, and seed filling stages, respectively. SO, RH, and RT represent the surrounding soil, rhizospheric soil, and root samples, respectively. The asterisk (*, P<0.05) indicates a significant difference according to a one-way ANOVA.

Discussion

The general concept of the present study was partly derived from previous studies (Mendes et al., 2014; Edwards et al., 2015) that examined the microbial communities associated with three rhizo-compartments. In the present study, we used similar compartments, including the surrounding soil, rhizospheric soil and associated roots containing the rhizoplane and endosphere (Bulgarelli et al., 2012; Edwards et al., 2015; Lu et al., 2018b; Wen et al., 2019), to effectively compare differences among the three soybean treatments at the three growth stages and assess the impact of GM crops on underground microbial communities.

In the present study, we selected the glyphosate-resistant soybean line Z106 with the dual transgenes of G2-EPSPS and GAT and its recipient cultivar ZH10 as research materials. In contrast to conventional comparisons, which generally focus on the effects of transgenic or only herbicide application separately (Babujia et al., 2016; Fan et al., 2017), we evaluated the influence of glyphosate-resistant transgenes together with the application of glyphosate. Thus, the impact of the dual transgenic soybean between Z106 and ZH10 and that of glyphosate between Z106 treated with glyphosate (Z106J) and Z106 with clean water on soil root-associated microbial communities were effectively analyzed at the seedling, flowering, and seed filling stages, which appeared to clearly provide an important reference for the evaluation of soil ecological and environmental safety (Bulgarelli et al., 2012).

In the alpha diversity analysis, no significant differences were observed in the surrounding soil, rhizospheric soil, or root samples of ZH10, Z106, and Z106J at the seedling, flowering, or seed filling stages (Table S2). Similarly, the beta diversity analysis failed to detect significant differences among the three treatments in the same compartment at any of the three stages (Table S3). These results indicate that neither the G2-EPSPS & GAT transgenes nor the application of glyphosate significantly changed the community structure of soil microbial communities.

However, in comparisons of the composition of the major bacterial taxa at the three growth stages, the abundance of some soil bacterial communities showed significant differences among the three treatments (Fig. S8). Ternary plots also revealed differences in the proportional contributions of several root-associated microbial communities based on different compartments, treatments, or stages. As shown in Fig. 3, the proportional contribution of E. fredii was higher in root samples at the flowering stage of ZH10 and Z106J than that of Z106, which is reported to be involved in nitrogen-fixing nodule formation in plants and is a very common symbiont of soybean (Temprano-Vera et al., 2018) and other legumes (Rathi et al., 2018) in alkaline soils, in which it forms determinate nodules in soybean and cowpea (Krishnan, 2002; Lopez-Baena et al., 2016). Furthermore, E. fredii is a fast-growing root nodule bacterium with a broad spectrum of hosts that was initially isolated from Chinese soil and nodulates the roots of 79 diverse genera of legumes (Krishnan, 2002; Albareda et al., 2009).

The lower proportional contribution of E. fredii in Z106 than in ZH10 indicated that the glyphosate-resistant transgenes G2-EPSPS & GAT affected root-associated microbial diversity in the root system of soybean, possibly through changes in gene expression and root exudation, as reported previously in GM crops (Tesfaye et al., 2003; Steinauer et al., 2016). Compared to Z106, the exogenous application of glyphosate recovered the proportional contribution of E. fredii in Z106J, which was similar to that of ZH10. A previous study reported that a novel class II EPSPS from E. fredii showed a high level of glyphosate tolerance (Wang et al., 2014), which reflected the potential ability of E. fredii to resist glyphosate, suggesting positive feedback through which glyphosate resistance occurs when Z106 is treated with glyphosate. GAT may also degrade glyphosate, which then acts as a nutrient source for E. fredii, thereby neutralizing its herbicidal effects (Saxton et al., 2011).

The relative abundance of nitrogen-fixing bacterial genera in soil was also investigated in the present study, and the results obtained showed that the relative abundance of Bradyrhizobium was significantly influenced by the G2-EPSPS & GAT transgenes at the seed filling stage (Fig. 4). The genus Bradyrhizobium is generally reported to be related to soybean nitrogen fixation, such as B. elkanii, which is a nitrogen-fixing bacterium that nodulates soybean (Barcellos et al., 2007; de Souza et al., 2012; Mason et al., 2018). In addition, B. elkanii showed different proportional contributions among the three treatments at the three stages (Fig. 3), which also demonstrated that the G2-EPSPS & GAT transgenes and the application of glyphosate may affect the abundance of nitrogen-fixing bacteria. Notably, the higher proportional contribution of B. elkanii in ZH10 and Z106J at the seed filling stage was similar to the proportional contribution of the nitrogen-fixing species E. fredii among the three treatments at the flowering stage.

In the present study, differences in relative abundance among the three treatments may also be partly due to variations in soil pH (Table 1) caused by transgenic soybean planting or the application of glyphosate. Soil pH is generally regarded as one of the major factors influencing microbial activity or the community structure (Aciego Pietri and Brookes, 2009; Cao et al., 2016); however, this needs to be examined in more detail in order to elucidate the mechanisms responsible for the different responses.

To clarify the mechanisms by which the crop affected the rhizosphere diazotrophic community, the relative abundance of nitrogen-fixing bacteria in root nodule samples and biological nitrogen fixation (BNF) by soybean need to be assessed (Bohm et al., 2009; Hungria et al., 2014; Fan et al., 2017); however, in the present study, no significant differences were observed in the number of nodules among the three treatments (data not shown). This may be due to the sampling error of nodules, which ultimately resulted in the cessation of attempts to extract bacterial genomic DNA from nodules. However, this issue needs to be examined in future studies to fully elucidate the relative abundance and responses of nitrogen-fixing bacteria residing not only in root-associated compartments, but also in root nodules.

G2-EPSPS & GAT transgenic herbicide-tolerant soybean treated with or without glyphosate had no significant impact on rhizosphere microbial communities in soil during a single growth season. However, G2-EPSPS & GAT transgenes and the application of glyphosate specifically affected the soil microbial community structure, such as the relative abundance and proportional contributions of some nitrogen-fixing bacteria in soybean root-associated bacterial communities, which may have an impact on rhizosphere nitrogen fixation efficiency in soybean, particularly those used for soil remediation under abiotic stresses. For example, in acidic soils lacking efficient phosphorus with aluminum toxicity (Barcelo and Poschenrieder, 2002; Zhao et al., 2004; Liao et al., 2006), soybean plants with nitrogen fixation ability were planted with the expectation of soil improvement (Cheng et al., 2009; Yang et al., 2012); therefore, there are concerns about the efficiency of soil remediation, in addition to assessment of biosafety issues, if application of these GM soybeans is expanded. In addition, the present results need to be confirmed in studies using more biological replicates and field experiments under different geographical and physiological conditions.

Citation

Yang, M., Wen, Z., Fazal, A., Hua, X., Xu, X., Yin, T., et al. (2020) Impact of a G2-EPSPS & GAT Dual Transgenic Glyphosate-Resistant Soybean Line on the Soil Microbial Community under Field Conditions Affected by Glyphosate Application. Microbes Environ 35: ME20056.

https://doi.org/10.1264/jsme2.ME20056

Supplementary Material

Supplementary Material (3.3MB, pdf)

Acknowledgements

We are grateful to Dr. Longguo Jin and Dr. Lijuan Qiu at the Institute of Crop Sciences, the Chinese Academy of Agriculture Sciences (CAAS), for providing soybean seeds of the GM line Z106 and its control cultivar ZH10. This work was supported by grants from the National Important Science & Technology Specific Project (2016ZX08011-003), the National Natural Science Foundation of China (31870495, 31372140), and the Program for Changjiang Scholars and Innovative Research Team in University from the Ministry of Education of China (IRT_14R27).

References

  1. Aciego Pietri J.C., and Brookes P.C. (2009) Substrate inputs and pH as factors controlling microbial biomass, activity and community structure in an arable soil. Soil Biol Biochem 41: 1396–1405. [Google Scholar]
  2. Albareda M., Rodríguez-Navarro D.N., and Temprano F.J. (2009) Use of Sinorhizobium (Ensifer) fredii for soybean inoculants in South Spain. Eur J Agron 30: 205–211. [Google Scholar]
  3. Andam C.P., Mondo S.J., and Parker M.A. (2007) Monophyly of nodA and nifH genes across texan and costa rican populations of Cupriavidus nodule symbionts. Appl Environ Microbiol 73: 4686–4690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babujia L.C., Silva A.P., Nakatani A.S., Cantão M.E., Vasconcelos A.T.R., Visentainer J.V., and Hungria M. (2016) Impact of long-term cropping of glyphosate-resistant transgenic soybean [Glycine max (L.) Merr.] on soil microbiome. Transgenic Res 25: 425–440. [DOI] [PubMed] [Google Scholar]
  5. Barcellos F.G., Menna P., Batista J.S.D., and Hungria M. (2007) Evidence of horizontal transfer of symbiotic genes from a Bradyrhizobium japonicum inoculant strain to indigenous diazotrophs Sinorhizobium (Ensifer) fredii and Bradyrhizobium elkanii in a Brazilian Savannah soil. Appl Environ Microbiol 73: 2635–2643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Barcelo J., and Poschenrieder C. (2002) Fast root growth responses, root exudates, and internal detoxification as clues to the mechanisms of aluminium toxicity and resistance: a review. Environ Exp Bot 48: 75–92. [Google Scholar]
  7. Benbrook C.M. (2016) Trends in glyphosate herbicide use in the United States and globally. Environ Sci Eur 28: 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bohm G.M.B., Alves B.J.R., Urquiaga S., Boddey R.M., Xavier G.R., Hax F., and Rombaldi C.V. (2009) Glyphosate- and imazethapyr-induced effects on yield, nodule mass and biological nitrogen fixation in field-grown glyphosate-resistant soybean. Soil Biol Biochem 41: 420–422. [Google Scholar]
  9. Brookes G., and Barfoot P. (2013) The global income and production effects of genetically modified (GM) crops 1996–2011. GM Crops Food 4: 74–83. [DOI] [PubMed] [Google Scholar]
  10. Bulgarelli D., Rott M., Schlaeppi K., van Themaat E.V.L., Ahmadinejad N., Assenza F., et al. (2012) Revealing structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488: 91–95. [DOI] [PubMed] [Google Scholar]
  11. Cao H.C., Chen R.R., Wang L.B., Jiang L.L., Yang F., Zheng S.X., et al. (2016) Soil pH, total phosphorus, climate and distance are the major factors influencing microbial activity at a regional spatial scale. Sci Rep 6: 25815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Caporaso J.G., Kuczynski J., Stombaugh J., Bittinger K., Bushman F.D., Costello E.K., et al. (2010) QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7: 335–336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Castle L.A., Siehl D.L., Gorton R., Patten P.A., Chen Y.H., Bertain S., et al. (2004) Discovery and directed evolution of a glyphosate tolerance gene. Science 304: 1151–1154. [DOI] [PubMed] [Google Scholar]
  14. Cheng F.X., Cao G.Q., Wang X.R., Zhao J., Yan X.L., and Liao H. (2009) Isolation and application of effective nitrogen fixation rhizobial strains on low-phosphorus acid soils in South China. Chin Sci Bull 54: 412–420. [Google Scholar]
  15. Dale P.J., Clarke B., and Fontes E.M.G. (2002) Potential for the environmental impact of transgenic crops. Nat Biotechnol 20: 567–574. [DOI] [PubMed] [Google Scholar]
  16. de Souza J.A.M., Tieppo E., Magnani G.D.S., Alves L.M.C., Cardoso R.L., Cruz L.M., et al. (2012) Draft Genome sequence of the nitrogen-fixing symbiotic bacterium Bradyrhizobium elkanii 587. J Bacteriol 194: 3547–3548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Duke S.O., and Powles S.B. (2008) Glyphosate: a once-in-a-century herbicide. Pest Manage Sci 64: 319–325. [DOI] [PubMed] [Google Scholar]
  18. Edgar R.C., Haas B.J., Clemente J.C., Quince C., and Knight R. (2011) UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27: 2194–2200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Edgar R.C. (2013) UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat Methods 10: 996–998. [DOI] [PubMed] [Google Scholar]
  20. Edwards J., Johnson C., Santos-Medellín C., Lurie E., Podishetty N.K., Bhatnagar S., et al. (2015) Structure, variation, and assembly of the root-associated microbiomes of rice. Proc Natl Acad Sci U S A 112: E911–E920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Fadrosh D.W., Ma B., Gajer P., Sengamalay N., Ott S., Brotman R.M., and Ravel J. (2014) An improved dual-indexing approach for multiplexed 16S rRNA gene sequencing on the Illumina MiSeq platform. Microbiome 2: 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Fan L., Feng Y.C., Weaver D.B., Delaney D.P., Wehtje G.R., and Wang G.Y. (2017) Glyphosate effects on symbiotic nitrogen fixation in glyphosate-resistant soybean. Appl Soil Ecol 121: 11–19. [Google Scholar]
  23. Feng J., and Yang F. (2019) The regulation of genetically modified food in China. Biotechnol Law Rep 38: 289–293. [Google Scholar]
  24. Guo B.F., Guo Y., Hong H.L., Jin L.G., Zhang L.J., Chang R.Z., et al. (2015) Co-expression of G2-EPSPS and glyphosate acetyltransferase GAT genes conferring high tolerance to glyphosate in soybean. Front Plant Sci 6: 847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Guo B.F., Guo Y., Hong H.L., and Qiu L.J. (2016) Identification of genomic insertion and flanking sequence of G2-EPSPS and GAT transgenes in soybean using whole genome sequencing method. Front Plant Sci 7: 1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hungria M., Mendes I.C., Nakatani A.S., dos Reis F.B., Morais J.Z., de Oliveira M.C.N., and Fernandes M.F. (2014) Effects of the glyphosate-resistance gene and herbicides on soybean: Field trials monitoring biological nitrogen fixation and yield. Field Crops Res 158: 43–54. [Google Scholar]
  27. ISAAA (2017) Global Status of Commercialized Biotech/GM Crops: 2016. New York, NY: The International Service for the Acquisition of Agri-biotech Applications (ISAAA). [Google Scholar]
  28. ISAAA (2018) Global Status of Commercialized Biotech/GM Crops in 2017: Biotech Crop Adoption Surges as Economic Benefits Accumulate in 22 Years. New York, NY: The International Service for the Acquisition of Agri-biotech Applications (ISAAA). [Google Scholar]
  29. Jie W.G., Lin J.X., Guo N., Cai B.Y., and Yan X.F. (2019) Community composition of rhizosphere fungi as affected by Funneliformis mosseae in soybean continuous cropping soil during seedling period. Chil J Agric Res 79: 356–365. [Google Scholar]
  30. Kaneko T., Nakamura Y., Sato S., Minamisawa K., Uchiumi T., Sasamoto S., et al. (2002) Complete genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum USDA110. DNA Res 9: 189–197. [DOI] [PubMed] [Google Scholar]
  31. Krishnan H.B. (2002) NolX of Sinorhizobium fredii USDA257, a type III-secreted protein involved in host range determination, Iis localized in the infection threads of cowpea (Vigna unguiculata [L.] Walp) and soybean (Glycine max [L.] Merr.) nodules. J Bacteriol 184: 831–839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liao H., Wan H., Shaff J., Wang X., Yan X., and Kochian L.V. (2006) Phosphorus and aluminum interactions in soybean in relation to aluminum tolerance. Exudation of specific organic acids from different regions of the intact root system. Plant Physiol 141: 674–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Liu X., Wei S., Wang F., James E.K., Guo X., Zagar C., et al. (2012) Burkholderia and Cupriavidus spp. are the preferred symbionts of Mimosa spp. in southern China. FEMS Microbiol Ecol 80: 417–426. [DOI] [PubMed] [Google Scholar]
  34. Lopez-Baena F.J., Ruiz-Sainz J.E., Rodriguez-Carvajal M.A., and Vinardell J.M. (2016) Bacterial molecular signals in the Sinorhizobium fredii-soybean symbiosis. Int J Mol Sci 17: 755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lu G.H., Zhu Y.L., Kong L.R., Cheng J., Tang C.Y., Hua X.M., et al. (2017) Impact of a glyphosate-tolerant soybean line on the rhizobacteria, revealed by Illumina MiSeq. J Microbiol Biotechnol 27: 561–572. [DOI] [PubMed] [Google Scholar]
  36. Lu G.H., Hua X.M., Cheng J., Zhu Y.L., Wang G.H., Pang Y.J., et al. (2018a) Impact of glyphosate on the rhizosphere microbial communities of an EPSPS-transgenic soybean line ZUTS31 by metagenome sequencing. Curr Genomics 19: 36–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lu G.H., Tang C.Y., Hua X.M., Cheng J., Wang G.H., Zhu Y.L., et al. (2018b) Effects of an EPSPS-transgenic soybean line ZUTS31 on root-associated bacterial communities during field growth. PLoS One 13: e0192008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mason M.L.T., Tabing B.L.C., Yamamoto A., and Saeki Y. (2018) Influence of flooding and soil properties on the genetic diversity and distribution of indigenous soybean-nodulating bradyrhizobia in the Philippines. Heliyon 4: e00921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mendes L.W., Kuramae E.E., Navarrete A.A., van Veen J.A., and Tsai S.M. (2014) Taxonomical and functional microbial community selection in soybean rhizosphere. ISME J 8: 1577–1587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ngom A., Nakagawa Y., Sawada H., Tsukahara J., Wakabayashi S., Uchiumi T., et al. (2004) A novel symbiotic nitrogen-fixing member of the Ochrobactrum clade isolated from root nodules of Acacia mangium. J Gen Appl Microbiol 50: 17–27. [DOI] [PubMed] [Google Scholar]
  41. Peiffer J.A., Spor A., Koren O., Jin Z., Tringe S.G., Dangl J.L., et al. (2013) Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc Natl Acad Sci U S A 110: 6548–6553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Rathi S., Tak N., Bissa G., Chouhan B., Ojha A., Adhikari D., et al. (2018) Selection of Bradyrhizobium or Ensifer symbionts by the native Indian caesalpinioid legume Chamaecrista pumila depends on soil pH and other edaphic and climatic factors. FEMS Microbiol Ecol 94: fiy180. [DOI] [PubMed] [Google Scholar]
  43. Rohart F., Gautier B., Singh A., and Lê Cao K.A. (2017) mixOmics: An R package for ‘omics feature selection and multiple data integration. PLoS Comput Biol 13: e1005752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Saxton M.A., Morrow E.A., Bourbonniere R.A., and Wilhelm S.W. (2011) Glyphosate influence on phytoplankton community structure in Lake Erie. J Great Lakes Res 37: 683–690. [Google Scholar]
  45. Schlaeppi K., Dombrowski N., Oter R.G., Ver Loren van Themaat E., and Schulze-Lefert P. (2014) Quantitative divergence of the bacterial root microbiota in Arabidopsis thaliana relatives. Proc Natl Acad Sci U S A 111: 585–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schloss P.D., Westcott S.L., Ryabin T., Hall J.R., Hartmann M., Hollister E.B., et al. (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 75: 7537–7541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Seldin L., and Dubnau D. (1985) Deoxyribonucleic-acid homology among Bacillus-polymyxa, Bacillus-macerans, Bacillusazotofixans, and other nitrogen-fixing Bacillus strains. Int J Syst Bacteriol 35: 151–154. [Google Scholar]
  48. Steinauer K., Chatzinotas A., and Eisenhauer N. (2016) Root exudate cocktails: the link between plant diversity and soil microorganisms? Ecol Evol 6: 7387–7396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Talebi M., Emtiazi G., Sepahy A.A., and Zaghian S. (2013) Zymogram analysis of alkaline keratinase produced by nitrogen fixing Bacillus pumilus ZED17 exhibiting multiprotease enzyme activities. Jundishapur J Microbiol 6: e7974. [Google Scholar]
  50. Temprano-Vera F., Rodríguez-Navarro D.N., Acosta-Jurado S., Perret X., Fossou R.K., Navarro-Gomez P., et al. (2018) Sinorhizobium fredii strains HH103 and NGR234 form nitrogen fixing nodules with diverse wild soybeans (Glycine soja) from central China but are ineffective on Northern China accessions. Front Microbiol 9: 2843. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Tesfaye M., Dufault N.S., Dornbusch M.R., Allan D.L., Vance C.P., and Samac D.A. (2003) Influence of enhanced malate dehydrogenase expression by alfalfa on diversity of rhizobacteria and soil nutrient availability. Soil Biol Biochem 35: 1103–1113. [Google Scholar]
  52. Tian X.L., Wang C.B., Bao X.G., Wang P., Li X.F., Yang S.C., et al. (2019) Crop diversity facilitates soil aggregation in relation to soil microbial community composition driven by intercropping. Plant Soil 436: 173–192. [Google Scholar]
  53. Wackerow-Kouzova N.D. (2007) Ochrobactrum intermedium ANKI, a nitrogen-fixing bacterium able to decolorize azobenzene. Appl Biochem Microbiol 43: 403–406. [PubMed] [Google Scholar]
  54. Wang L., Peng R., Tian Y., Han J., Zhao W., Wang B., et al. (2014) Characterization of a class II 5-enopyruvylshikimate-3-phosphate synthase with high tolerance to glyphosate from Sinorhizobium fredii. World J Microbiol Biotechnol 30: 2967–2973. [DOI] [PubMed] [Google Scholar]
  55. Wen Z.L., Yang M.K., Du M.H., Zhong Z.Z., Lu Y.T., Wang G.H., et al. (2019) Enrichments/derichments of root-associated bacteria related to plant growth and nutrition caused by the growth of an EPSPS-transgenic maize line in the field. Front Microbiol 10: 1335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Yang T.Y., Liu G.L., Li Y.C., Zhu S.M., Zou A.L., Qi J.L., and Yang Y.H. (2012) Rhizosphere microbial communities and organic acids secreted by aluminum-tolerant and aluminum-sensitive soybean in acid soil. Biol Fertil Soils 48: 97–108. [Google Scholar]
  57. Zhang J., Zhang B., Liu Y., Guo Y., Shi P., and Wei G. (2018) Distinct large-scale biogeographic patterns of fungal communities in bulk soil and soybean rhizosphere in China. Sci Total Environ 644: 791–800. [DOI] [PubMed] [Google Scholar]
  58. Zhao J., Fu J.B., Liao H., He Y., Nian H., Hu Y. M., et al. (2004) Characterization of root architecture in an applied core collection for phosphorus efficiency of soybean germplasm. Chin Sci Bull 49: 1611–1620. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material (3.3MB, pdf)

Articles from Microbes and Environments are provided here courtesy of Nakanishi Printing

RESOURCES