Stacked genetically modified soybean harboring herbicide resistance and insecticide rCry1Ac shows strong defense and redox homeostasis disturbance after glyphosate application

Background World agricultural production of genetically modified (GM) products in particular, the combination of different traits/genes in the same plant has been intense over the last decade. The stacking of herbicide and insect-resistant transgenic genes can result in fitness costs that rely on the type and strength of the selection pressure exerted by the environment. Here we report the results of transcriptomic analysis comparing the effect of glyphosate on various biological processes, metabolic pathways, and main shikimic enzymes in stacked versus single soybean resistant varieties. Results Gene expression data were grouped according to treatment, ie the herbicidal factor strongly influenced. Common physiological results between the single and established varieties were mainly in Redox metabolism, energy, and metabolism. Photosynthesis was only found negatively affected in the single variety. The defense components, although present in both varieties, show a more intense presence in staked pathways, that demonstrated pathways related to up-regulated secondary metabolites biosynthesis, a known response when plants are under various stress conditions. RT-PCR results confirmed that native EPSPS expression was up-regulated at the same level for single and stacked events. However, metabolic differences in expression were observed, suggesting a distinct cascade effect between simple and stacked, triggered by glyphosate application.

accumulation of glyphosate in single-event transgenic soybean enhanced cellular oxidation, possibly through mechanisms involving stimulation of the photorespiratory pathway [11]. It also indicated that most of the glyphosate-induced genes are homologous to the known expression sequence tags -ESTs induced by abiotic stress factors [12]. In silico and in vivo studies also showed stress response, i.e. glutathione redox metabolism alteration, in glyphosate-treated resistant soybeans (Ayyadurai et al., 2011). The photosynthesis metabolism has also been affected by herbicide application in glyphosate resistant GM maize varieties [13].
Although the use of these approaches provides a robust understanding of the genetic regulation of the response of resistant genotypes, a comprehensive picture of the metabolic manifestations of resistant genotypes conferring multiple transgenic traits is still lacking. This is partly because previous untargeted omics approaches were limited to the analysis of single transgene resistant plants [12][11] [14][13] [15].
In order to gain new insights into the response of glyphosate application in soybean harboring two or more transgenic events, the current study compared the effect of glyphosate on several biological processes, metabolic pathways and key enzymes of the shikimate and phenylpropanoid pathway and other cascade pathways in stacked INTACTA RR2 PRO soybean varieties. We hypothesized that transgenic plants with a combination of transgenes respond differently to glyphosate accumulation due to (1) the cost of expressing more than one heterologous protein and (2) synergistic and antagonistic interactions of each transgene cascade pathways to glyphosate direct and indirect target pathways. This study was undertaken in order to characterize the interactions between the shikimate pathway and other unsupervised side-affected pathways as determined by changes in the transcriptome of glyphosate-treated stacked transgenic soybean variety, which express the Agrobacterium tumefaciens strain CP4-EPSPS enzyme that is resistant to inhibition by glyphosate. We undertook this work on soybean because it is a major crop species: genetically modified lines currently occupy more than 50% of the acreage planted globally. The data presented here provide new knowledge concerning the influence of recombinant Cry1Ac (rCry1ac) transgene cassette on the defense response and glutathione metabolism, the abundance of beta-glucosidase and 5 oxidoreductase enzymes when glyphosate is applied.

Plant material and herbicide treatment
Two soybean (Glycine max) varieties were used in this transcriptomic profiling study: NA 5909 RG and BRS 1001 iPRO. The first one, NA 5909 RG, is a single-event variety containing one transgenic event The experiment was conducted in a full-factorial no-choice experiment in block design with two factors: soybean variety and herbicide treatment. Seeds of each soybean variety were grown in greenhouse under two treatment conditions: herbicide spray application (treated group); and no herbicide application (control group). Seeds were grown in 14 L plastic pots filled with a substrate (1/3 clay soil; 1/3 cellulose residue and 1/3 poultry organic residue) with pH corrected to 6.0. There were three plants per pot, and three pots per treatment, disposed in three random blocks. The experiment was watered daily. All plants were kept under the same conditions until they reach V2 stage, approximately 34 days after emergence, when the herbicide treatment was applied to a subset of plants (treated group). Glyphosate treatment was conducted through spray application using glyphosate-based herbicide formulation Roundup Transorb® (Monsanto company) under the general maximum dosage informed by the leaflet (4.5 L / ha; 2.2 kg a. i. / ha) for soybean crops [16]. To minimize spray contamination and drift, all plants were placed outside the greenhouse and separate into plastic barriers. Within each block (three) plants were randomized and border protected. Leaf samples were collected eight hours after herbicide application [10]. For each treatment, three leaves of the fourth trifoliate were collected from the three plants present in each pot, immediately frozen in liquid nitrogen and stored at -80 °C. Therefore, each treatment contained three biological pools of 6 three different plants, which were used for the transcriptomic analysis ( Figure 1).

Library construction, sequencing and mapping
Total RNA was isolated using RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The purity of RNA was measured using Nanodrop 2000 (Thermo Scientific), and the integrity of RNA was determined by agarose gel electrophoresis. Samples of 1-2 μg mRNA were used for library construction and sequencing. Libraries were constructed by Novogene Corporation (Sacramento, CA). Briefly, mRNA was enriched from total RNA using oligo (dT) beads.
The mRNA was then randomly fragmented and cDNA was synthesized using random hexamers. After cDNA synthesis and library construction (terminal repair, A-tailing, ligation of sequencing adapters, size selection and PCR enrichment). Library concentration was quantified and then diluted to 1 ng/μl.
The libraries were sequenced on the Illumina HiSeq platform (PE150). Raw reads are filtered to get rid of reads containing adapter or with low quality to get the clean reads for analysis. The Hisat2 v2.0.4 algorithm was used to map the filtered sequenced reads to the reference genome [17].
The mapped transcripts were annotated and counted generating normalized readcount values in terms of fragments per kilobase of transcript per million mapped reads (FPKM). Differentially expressed genes (DEGs) between the different treatments were determined using DESeq [18]. Only genes with FDR adjusted p-value (q-value) < 0.05 after Benjamini-Hochberg correction for multipletesting [19] were considered as significant DEGs.

Functional annotation of DEGs and bioinformatics
The DEGs were annotated for gene ontology (GO) terms using Blast2GO v2.5 [20] and categorized into Molecular Function (MF), Cellular Component (CC), and Biological Process (BP) categories. The DEGs were annotated using KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis aiming to identify significantly enriched metabolic pathways or signal transduction pathways affected by the glyphosate treatment in both single and stacked varieties. Pathways with q-value < 0.05 were significantly enriched. A heatmap clustering analysis of the log10(FPKM+1) values was conducted using pheatmap library [21] in R [22] aiming to find gene expression patterns across the different treatments. Genes were also converted to proteins and used as input for producing flow 7 charts of DEGs and pathways significantly affected by using STRING v11 Protein-Protein Interaction Networks software (http://string-db.org/) [23].

EPSPS Expression by qRT-PCR
The qRT-PCR analysis was performed aiming to quantify the expression of native and transgenic EPSPS in both single and stacked soybean varieties. Primers were designed based on the predicted coding sequence of each of the three target EPSPS genes (MON89788-1; MON4032-6; and EPSPS native) by using the online tool PrimerQuest (Integrated DNA Technologies Inc., Skokie, IL, USA) ( DEGs in response to herbicide treatment for the single and stacked variety respectively, when compared to the control ( Figure 2).
Additionally, we have further explored the data by running a hierarchical clustering analysis of DEGs aiming to find genes with similar expression patterns across the different varieties and treatments.
The heatmap showed that gene expression data clustered according to treatment, which means that the factor 'herbicide' resulted in a major effect of variability as compared to the 'variety' factor For the single variety, the altered components are directly related to defense response mobilization.
Notably, the herbicide treatment up-regulated energy and defense-related pathways while downregulated growth-related pathways ( Figure 5). For instance, carbon metabolism (22 DEGs out of 488 genes, 4.5%); N-Glycan biosynthesis (7/74, 9.45%); carbon fixation (   samples, compared to their respective controls ( Fig. 6a and  On the other hand, relative expression of transgenic EPSPS in the single variety (MON4032-6) showed significant down-regulation after herbicide application (0.79 ± 0.16 Rq; p = 9.67e-7) compared to control samples (Fig. 6c). The same pattern was found for the transgenic EPSPS present in the stacked variety (MON89788-1), in which expression levels in Stacked Treated samples was cut by half (0.47 ± 0.36 Rq; p = 5.33e-11) compared to untreated control samples (Fig. 6d). Interestingly, while cp4-EPSPS has been modulated in the same manner in both single and stacked varieties; differences in the metabolism has been observed and thus suggests it is not related to either native or transgenic EPSPS modulation.

Discussion
The impact of directly related metabolic pathways to shikimate pathway has been observed in both 14 single and stacked variety. However, in the stacked variety the impact on the isoflavonoid, flavonoid, glutathione, cysteine and methionine and phenylalanine metabolism were most prominent (Figure 7).
In addition, alteration in the jasmonic acid metabolism was also observed. Increased levels of jasmonic acid have been also observed after glyphosate and drought stress application in NK603 herbicide resistant GM maize [13]. Glyphosate was also shown to interfere with other hormones such as ethylene [30] and abscisic acid (ABA) [10]. Although extremely important, the effects of glyphosate in plant hormone metabolism are still unclear.

Strong Defense Response triggered by Glyphosate
Defense imposes a substantial demand for resources that can negatively impact growth and diminish the overall set of energy reserves and/or promote resource diversion for growth, defense, and reduction of photosynthesis [31]. Previous transcriptome studies using microarray technique to investigate the metabolic impact of glyphosate treatment in susceptible and resistant soybean, Arabidopsis and brassica showed that most affected pathways are involved with defense metabolism [12][32] [10]. In this study, genes related to defense metabolism were up-regulated, indicating that glyphosate may be related to alterations in gene cascades and unexpected pathways. Rapid recognition of injuries by cellular signal transduction pathways occurs through various signaling molecules, including calcium, protein phosphorylation and ROS, which are well-known triggers of stress resistance in plants [39]. Herbicides are considered abiotic stressors that can disrupt the balance between the production and elimination of reactive oxygen species (ROS) [40]. There is a close relationship between calcium-dependent ROS production and a specific group of genes. as key factors in activating innate and mobilized immunity during oxidative stress damage [43].
Another example of defense regulatory circuit was the identification of WRKY transcription factors that is phosphorylated by MAPK and a W-box in the promoter region of Nicotiana tabacum Rboh, interconnecting the phosphorylation events of MAPK in response to pathogen recognition with the accumulation of Rboh protein [42]. Strict regulation and fine-tuning of WRKY proteins are directly linked to plant stress signaling responses [44][45], such as saline stress [46], drought [47][48] and heat stress [49]. We observed up-regulation of WRKY genes in both varieties, with higher expression and number of genes in the staked variety (five genes with an average of 2.5 fold change). WRKY genes have not yet been found affected by glyphosate.
The pathogenesis-related proteins (PR), known as an indispensable component of innate immune responses in plants under biotic or abiotic stress conditions were also observed in this study. In the single-variety, we find one gene PR1, up-regulated with a 2.9 log2FC. These proteins are also involved in hypersensitive response or systemic acquired resistance against a variety of plant infections [50] and an important response mechanism to multiple stresses [51]. PR proteins are considered the signature genes of salicylic acid and jasmonic acid pathways in many crop plants [52] [53][51] [54].
Immune sensors in plants are well-known substrates for heat shock proteins, such as heat shock protein 90 (Hsp90). To recognize potential pathogens, higher eukaryotic organisms use extra-or intracellular sensors as the initial switch in the induction of disease defense responses [55] [56].
Beyond defense, the environmentally responsive HSP90 chaperone complex is suggested to be involved in multiple signaling cascades, with the potential to be of great importance for sensing the environment and mediating appropriate phenotypic plasticity [57]. In our study, Hsp90 gene families were found up-regulated in the both single and stacked varieties. In soybean, Hsp90 gene was induced by heat, salt, and osmotic stresses but the response times and expression abundances were diverse [58].
Many plant defense compounds are used in a non-active glycosylated form suitable for storage in the vacuole and further protection from toxic side-effects as a consequence of its defense system [59].
These are recognized as class of secondary metabolites called phytoanticipins. When plant tissue in which they are present is disrupted, the phytoanticipins are bio-activated by the enzymatic removal of a protecting glucose group by a β-glucosidase. These are binary systems in which two sets of components that, when separated, are relatively inert provide plants with an immediate chemical defense against protruding herbivores and pathogens [60]. Strikingly, the stacked variety upregulated seven β-glucosidase-related genes with an average of 5 log2FC. We also found a regulated isoflavone 7-O-methyltransferase gene found 8.5 log2FC. This specific change in the metabolism of secondary metabolites has not been reported before in plants treated with glyphosate. Thus, suggesting that the presence of rCry1Ac transgenic cassette has an impact on the defense metabolism when glyphosate is sprayed.

Changes in Carbon Allocation
The insertion of transgenes controlled by strong promoters has been always a concern as to the potential physiological effects on carbon allocation metabolism. In this paper, we applied glyphosate, an inhibitor of the enzyme EPSPS, present in the shikimate pathway as an abiotic stressor at concentrations present in real HR crop fields.
While under normal growth conditions, more than 20% of plant-fixed carbon flows through the shikimate pathway [61] [62]. Under stress, plants mobilize their carbon stocks to transform energy and resist harmful effects on cells. In our study, in the single variety, we observed the up-regulation of enzymes related to energy transformation processes and structural functions, such as lactate dehydrogenase, which participates in the process of transforming glucose into energy formed from pyruvate [63] and N-acetylglucosamine, a cellulose analogous structural polysaccharide, involved in structural roles on the cell surface [64].
The total biomass production of soybean depends on energy supplied by photosynthesis for synthesizing carbon compounds [66]. Alteration in carbon metabolism has been already observed after glyphosate application. This occurs because the inhibition of EPSPS deregulates the pathway, which results in an uncontrolled flow of carbon and subsequent massive accumulation of shikimate and other acids in metabolic sinks such as leaves and nodules of legumes [67]. Two of the major metabolic checkpoints co-ordinating primary nitrogen and carbon assimilation in leaves are nitrate reductase (NR) and PEPC [68].
The enzyme PEPC has been found up-regulate and is biologically related to maintaining load balance during the upward flow of xylem sap in vegetables [69] [70] and in the supply of substrates for symbiotic organism, developing a central role for biological nitrogen fixation [71]. Our data demonstrate a relationship of up-regulated genes that involve balancing the energy supply to the nodules and retaining sufficient carbon for growth, at a level of change beyond the PECP enzyme.
Several PEPC isoforms are controlled by an interaction between allosteric regulation and reversible phosphorylation. In legume leaves and root nodules, these regulatory functions of PEPC are governed primarily by phosphoenolpyruvate carboxykinase (PEPc Kinase) (Also found as altered in this study) [72] that promotes reversible protein phosphorylation of major importance in controlling legume nodule carbon metabolism and related metabolite transport [73]. PEPc Kinase is a member of the Ca2+/calmodulin-regulated group of protein kinases. However, it lacks the auto-inhibitory region and EF-hands of plant Ca2+-dependent protein kinases [74]. Regulatory mechanisms related to the formation and maintenance of root nodules for biological nitrogen fixation are fundamental for the proper adjustment of metabolic flow between host plants and symbolic organisms [75]. In the cytoplasmic compartment of plants, glucose and fructose-free hexoses (two genes for fructose bisphosphate aldolase were found with an average of 1.7 log2FC) are phosphorylated by glucose or pentose pathways [76] [77]. PEPC and malate dehydrogenase (a gene also found in this study with a 1.7 log2FC) convert the carbon flux of glycolysis to malate [78] which is used as carbon skeletons for N2 amino acid synthesis [79] [80]. Although malate is the main source of energy for symbiotic organisms, high volume malate may inhibit N2 fixation and nitrogen uptake [81].
This leads to the belief that in the single variety there was a change in carbon metabolism relative to storage strategies. Both in the structural form and in the carbon flow demand required by nodular organisms. This may be involved in the change in carbon flow required for growth and development.
In the stacked variety, on the other hand, carbon appears to be stored as starch, the carbohydrate used as the energy source for the defense response. We find altered starch and sucrose metabolism, especially gene related to trehalose-phosphate with 3.6 log2FC. Sucrose and starch management and balance promote optimization of growth rates [82] [83]. Trehalose (α-d-glucopyranosyl-1,1-α-dglucopyranoside) is a nonreducing disaccharide that is found in many organisms and has various functions: osmolyte, storage reserve, transport sugar, and stress protectant [84] [85].
It is also involved in growth and development metabolism [86] with clear links to abscisic acid and auxin signaling [87] as well as to the activation of starch synthesis [88]. The levels of trehalose increase in response to osmotic stress [89] as well as to dehydration stress tolerance [90] [89].
Trehalose is one of the most effective osmoregulatory sugars in terms of the minimal concentration required to establish a normal balance [91]. Many plants accumulate substantial starch reserves in their leaves to provide carbon and energy for maintenance and growth [92] [93]. Therefore, the accumulation of soluble sugars, such as trehalose, is suggested to be a protective mechanism under oxidative stress conditions [94][95].
Wingler et al. [96] showed a strong accumulation of starch in response to trehalose [96]. In this study, in the stacked variety we suggest that the trehalose transcripts may also be involved in the accumulation of starch, carbohydrate required to cope in the energy balance due to the need for response observed by up-regulation in various secondary metabolites (flavonoids, isoflavonoids, monoterpenoids).

Altered Cellular Redox Homeostasis
Exposure to glyphosate-based herbicides is directly linked to accumulation of antioxidant enzymes, indicating that glyphosate treatment might result in oxidative stress [85]. Glutathione (GSH) is a key of the complex antioxidant network in plants, acting to control ROS accumulation and facilitating cellular redox homeostasis especially under stress conditions [86]. For instance, GSH plays an important role in herbicide detoxification via the glutathione S-transferase (GST) system [87]. We found evidence for cellular detoxification response through significant up-regulation of GST in both soybean varieties under herbicide stress (Single: average log2FC = 3.1; Stacked: average log2FC =

3.5).
On the other hand, other genes encoding important enzymes related to glutathione metabolism showed to be differently affected in the single and stacked varieties, revealing that both genotypes may respond in a different manner in response to oxidative stress. For instance, we found glucose 6phosphate dehydrogenase (G6PDH) -an enzyme participating in the first two reactions of oxidative pentose phosphate pathway -being significantly down-regulated in the single variety. Reduced levels of G6PDH is related to glutathione depletion and consequent high oxidative stress in the cell [88]. It is known that reduced glutathione (GSH) is required to combat oxidative stress and maintain the normal reduced state in the cell, a phenomenon known as the redox homeostasis [85][86] [11]. Oxidized glutathione (GSSG) is reduced to GSH by NADPH generated by G6PDH in the pentose phosphate pathway [89]. Complete depletion of glutathione in its reduced form (GSH), or the production of GSSG from GSH, with concomitant accumulation of formaldehyde have already been reported as signs of undergoing oxidative stress in single-event GM soybean varieties as compared to its non-GM isogenic line [14][90]. On the other hand, for the stacked variety, although G6PDH gene expression has not been significantly affected, herbicide treatment up-regulated the expression of 6-phosphogluconate dehydrogenase (6PGDH) gene (log2FC = 1.25). 6PGDH, a second enzyme participating in the OPPP, catalyses the NADP-dependent oxidative decarboxylation of 6-phosphogluconate generating NADPH and ribulose-5-phosphate, a precursor for the synthesis of nucleotides and nucleic acids [91]. We hypothesize that the production of such reducing equivalents is being used in further reductive reactions in stacked plants, such as keeping GSH in its reduced form, aiming at maintaining the cell redox homeostasis.

20
Our results also showed protein processing in endoplasmatic reticulum (ER) as one of the most upregulated pathways in both, single and stacked varieties when glyphosate is applied. Glutathione homeostasis in response to oxidative stress has been also described as active in the ER In agreement, ER was annotated as the most up-regulated cellular component term under the GO enrichment analysis for the single variety. The only ER genes substantially affected in the stacked variety were those encoding for ER-associated degradation (ERAD) enzymes, such as derlin, Hsp40, Hsp70, Hsp90, and small heat shock factors (sHSF), which were also affected in the single variety.
Abiotic stress causes significant increase in protein unfolding metabolism, leading to the accumulation of misfolded proteins in the ER [93]. Such accumulation triggers the increase in degradation capacity of ERAD system aiming to maintain ER homeostasis. Over time, this process can lead to a variety of cellular signaling pathways which determine the state and fate of cell, which can include autophagy, apoptosis, inflammation, and even activation of cell death under severe conditions [94] [95]. In our study, the metabolic responses to the oxidative stress caused by glyphosate seems to be highly correlated to ER-related genes; most probably due to GSH depletion or elevated production of GSSG as already suggested by previous studies [14][90]. Since glutathione is oxidized, transport proteins must export GSSG from the ER to the cytosol aiming to reach an ideal glutathione homeostasis [92]. Conversely, the stacked variety showed evidence of oxidative stress responses due to the up-regulation of cytosolic glutathione genes (GST log2FC = 3.5; 6PGDH log2FC = 1.25), while 21 only genes encoding ERAD enzymes were significantly up-regulated in ER. Vivancos et al. [11] have also found effects of herbicide on cellular redox homeostasis of single event glyphosate-resistant soybean variety. More specifically, the authors reported that accumulation of high levels of glyphosate in GM tolerant plants have enhanced cellular oxidation, possibly through mechanisms involving increasing of photorespiratory pathway [11]. Moreover, a recent integrative in silico model of C1 metabolism in single event glyphosate-resistant GM soybean, predicted complete depletion of glutathione and accumulation of formaldehyde as a result of oxidative stress compared to its non-GM counterpart. The authors alert on how a single event modification can potentially create a large perturbation to molecular system equilibria [14]. According to our findings, single and stacked GM soybean showed oxidative stress at different levels and cellular components.

Photosynthesis imbalance
Glyphosate has been shown to have detrimental effects on many plant physiological and biochemical processes, which reduce photosynthesis efficiency and inhibits chlorophyll function [96][97] [98].
Chlorophyll is essential in photosynthesis and provides matter and energy for plant growth [99]. The concentration of total chlorophyll is the sum of chlorophyll A and chlorophyll B [100] [101]. In our study, the single variety showed a decrease in the light-harvesting chlorophyll A and B contend (complex I of class LhcA 2,3 and 4 with four genes involved, and the complex II of class LhcB 1,2,3 and 6 with nine genes involved). These findings are supported by Li et al. [102], who also observed a decline in the content of chlorophyll A and B in GM and conventional soybean varieties under glyphosate treatment [102].
After light excitation of chlorophyll molecules in the light-harvesting complexes, the energy is transferred to the reaction centers of photosystems I (PSI) and II (PSII) [103]. The electron transfer chain, which mediates the transmembrane charge separation, is the functionally most important part of photosystem [104]. Ferredoxins (FDXs) in chloroplasts are electron transfer proteins that deliver reducing equivalents from PSI in photosynthetic organisms [105]. Electrons from reduced FDXs are accepted by FDX-NADPH-oxidoreductase to generate NADPH, which is required for carbon assimilation in the Calvin cycle [106]. Limited capacity of electron transport after glyphosate exposure was shown by [96] [11]. In this study, we find two genes down-regulated related to putative FDX. The amount of FDX is also decreased in tobacco under various stresses, including those from herbicide treatment [107].
In general, we have observed, down-regulation of genes involved in photosynthesis and this is in agreement with previous studies when glyphosate application is performed. Iquebal et al. [108] observed that genes involved in the photosynthetic pathway were deregulated after exposure to herbicides in resistant chickpea variety [108]. In Lolium perene sensitive plants, chlorophyll fluorescence was also affected by glyphosate [109].

Relevance to risk assessment of stacked GM crops
Worldwide, a growing number of GM crops with stacked transgenic traits are being developed to confer resistance to herbicide active ingredients and some insect species. For most varieties, the single events might never reach market and pre-market risk assessment. Therefore, an assessment of the risks of a stacked GM plant to cause combinatorial and cumulative effects should be considered in the context of the closely related non-modified recipient organism in the receiving environment.
Omics profiling analysis can contribute to the identification of combinatorial effects that may occur due to interactions among the proteins and metabolites produced by the transgenes or endogenous genes of a stacked GM plant. In addition, interactions between the stacked transgenes or their products, or interactions among the physiological pathways in which the transgenes are involved, taking into account the possibility that these interactions could result in potentially harmful substances, such as anti-nutritional factors, some of which may persist or accumulate in the environment should be also considered. However, distinct stress responses were found among the varieties. The stacked variety showed a more pronounced stress response (activation of specific stress defense proteins (Rboh, WRKY) and secondary compounds (β-glucosidase, isoflavone 7-O-methyltransferase).
Both varieties were energetically impacted, the single showed changes in carbon metabolism about storage and accumulation strategies, while the stacked variety showed starch accumulation we have found a regulated response to trehalose and suggest that this carbohydrate is being stored to help regulate energy balance due to the regulation of some secondary metabolites (flavonoids, isoflavonoids, monoterpenoids). There were responses to cellular detoxification through significant up-regulation of GST in both varieties, but in the stacked, we hypothesized that the production of some reducing equivalents is being used in additional reductive reactions as a way to keep GST low to maintain cellular redox homeostasis.
Similar results to ours for the single variety were also found using single GM with glyphosate application and omic approaches.

Competing interests
The authors declare that they have no competing interests.

Funding
This project has received funding from the The German Federal Agency for Nature Conservation  Volcano plots of significantly differentially expressed genes, DEGs (p-adjusted value ≤ 0.05 and |FC| ≥ 1.5; red, up-regulated; green, down-regulated) when applied herbicide. The Xaxis represents fold change; Y-axis represents significance.  Changes caused due to the spray of glyphosate in the stacked variety. The picture shows that the herbicide triggered a series of events. There were mainly changes in the pathways and products related to the biosynthesis of secondary compounds.