Reduced Carbon Dioxide by Overexpressing EPSPS Transgene in Arabidopsis and Rice: Implications in Carbon Neutrality through Genetically Engineered Plants

Simple Summary Overexpressing the EPSPS gene may increase the consumption of ambient CO2 by plants. To explore the relationship between CO2 consumption by plants and expression of the EPSPS gene, we measured the expression level of the EPSPS gene, intercellular CO2 concentration, expression of global genes, and photosynthetic ratios in genetically engineered (GE) and non-GE Arabidopsis and rice plants. Our results showed substantial increases in EPSPS gene expression, CO2 consumption, expression of genes in carbon-fixation related pathways, and photosynthetic ratios. The results indicate that overexpressing the EPSPS gene can enhance CO2 consumption by GE plants with increased expression of genes in carbon-fixation related pathways. The findings not only expand our knowledge on the relationship between overexpression of the EPSPS gene(s) and carbon fixation, but also provide a novel strategy to reduce global CO2 for carbon neutrality, in addition to increases in crop production by genetic engineering of the EPSPS gene(s) in plants. Abstract With the increasing challenges of climate change caused by global warming, the effective reduction of carbon dioxide (CO2) becomes an urgent environmental issue for the sustainable development of human society. Previous reports indicated increased biomass in genetically engineered (GE) Arabidopsis and rice overexpressing the 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) gene, suggesting the possibility of consuming more carbon by GE plants. However, whether overexpressing the EPSPS gene in GE plants consumes more CO2 remains a question. To address this question, we measured expression of the EPSPS gene, intercellular CO2 concentration, photosynthetic ratios, and gene expression (RNA-seq and RT-qPCR) in GE (overexpression) and non-GE (normal expression) Arabidopsis and rice plants. Results showed substantially increased EPSPS expression accompanied with CO2 consumption in the GE Arabidopsis and rice plants. Furthermore, overexpressing the EPSPS gene affected carbon-fixation related biological pathways. We also confirmed significant upregulation of four key carbon-fixation associated genes, in addition to increased photosynthetic ratios, in all GE plants. Our finding of significantly enhanced carbon fixation in GE plants overexpressing the EPSPS transgene provides a novel strategy to reduce global CO2 for carbon neutrality by genetic engineering of plant species, in addition to increased plant production by enhanced photosynthesis.


Introduction
The great challenge of global climate change caused by the gradual increases in greenhouse gases (GHG) has posed tremendous threats to the global economic security and sustainable development of human society [1].Carbon dioxide (CO 2 ), particularly Biology 2024, 13, 25 2 of 18 that produced from various anthropogenic activities, constitutes an important GHG in the world [1].In response to the challenge posed by excessive GHG emissions, the global community has reached the Paris Climate Agreement [2,3] to take decisive actions and make significant strides toward the development of a low-carbon economy.In addition, scientific investigations are carried out to explore effective resolutions for achieving global carbon reduction or carbon neutrality to maintain the continued world development.These resolutions include applying clean energy [4][5][6], conserving energy and reducing CO 2 emission [7,8], regulating carbon sink in the ecosystems [9][10][11][12], and increasing the terrestrial vegetation coverage [13].
As an important carbon pool, the terrestrial ecosystems fix ~30% global atmospheric CO 2 emitted by anthropogenic activities [14,15].Consequently, terrestrial ecosystems have played an important role in reducing the atmospheric CO 2 , which is critical for maintaining the stability of the global climate [14].In terrestrial ecosystems, forests cover ~30% of the total terrestrial area [15,16], contributing a larger fraction to the carbon sink, meaning CO 2 fixation by organisms [15,17,18].Grasslands, croplands (crop fields), and other terrestrial floral ecosystems are also important for global terrestrial CO 2 sink [15,18].Usually, different types of stored carbon (sink) in plants are in the form of biomass that is transformed from solar energy and CO 2 through photosynthesis [15].
Photosynthesis is a pivotal source for carbon sink in terrestrial ecosystems [19] due to its capacity of converting atmospheric CO 2 into the fixed carbon by plants [20].Given that the further increase in terrestrial vegetation is relatively difficult, we can significantly increase the efficiency of photosynthesis and carbon fixation by green plants.It is reported that the efficiency of photosynthesis can determine the possibility for plants to absorb or consume CO 2 , thereby impacting global climate change [21].As a result, terrestrial plants have great potential to extensively increase the efficiency of photosynthesis and carbon fixation at the global scale.Thus, enhancing photosynthetic efficiency of terrestrial plants provides an ideal strategy for carbon neutrality to achieve GHG CO 2 reduction.In addition, enhancing photosynthesis can also increase crop production, thereby contributing to food security.
Previous studies indicated that genetically engineered (GE) plants overexpressing the exogenous or endogenous EPSPS genes substantially increased biomass in rice [22][23][24] and Arabidopsis thaliana [25,26].In addition, overexpressing the EPSPS gene also significantly increased photosynthetic ratios [22].These results together suggest that GE plants overexpressing the EPSPS gene may consume more atmospheric CO 2 to meet the demand for the increased photosynthesis and biomass.The EPSPS gene, encoding the essential enzyme in the shikimate pathway, is present in all plant species, as well as in many microorganisms [27].Furthermore, the EPSPS gene is also important for the biosynthesis of metabolites such as aromatic amino acids, lignin, and auxin [28].Therefore, it is possible to use green plants to increase the ratios of photosynthesis and CO 2 fixation through genetical engineering.
Whether overexpression of the EPSPS gene would generally increase the consumption of atmospheric CO 2 by GE plants is still unclear, but presents interesting questions in terms of its practical implications for carbon neutrality.To address this question, we investigated the intercellular CO 2 concentration of Arabidopsis and rice plants overexpressing the EPSPS gene produced through genetic engineering [22,25], in addition to their photosynthetic ratios.We also applied the RNA-seq and RT-qPCR molecular techniques to determine the possible impact of the EPSPS transgene on the global genes, and particularly carbonfixation related genes, in Arabidopsis or rice plants.The objectives of this study were to answer the following questions: (1) Does overexpression of the EPSPS gene increase the consumption or fixation ratios of CO 2 in the GE experimental plants?(2) What are the changes of the gene expression pattern globally in the experimental plants caused only by overexpressing the EPSPS gene?(3) What is the impact of overexpressing the EPSPS gene on the carbon-fixation related genes, as well as photosynthetic ratios in the GE experimental plants?Answers to these questions will facilitate our understanding for designing novel strategies to reduce atmospheric CO 2 by overexpressing the EPSPS genes in GE plants as the carbon sink, in addition to increases in crop production through enhanced plant photosynthesis.

Plant Materials and Growth Conditions
Two GE Arabidopsis (Arabidopsis thaliana) lineages (coded as A1 and A2) overexpressing the EPSPS transgene (cloned from cultivated rice) and their corresponding non-GE counterparts were used in this study to represent the dicotyledonous plants.The Arabidopsis plants were the 4th generation (T 4 ) derived from the GE and non-GE homozygous T 3 Arabidopsis, respectively [25].The GE Arabidopsis lineages, containing a single copy of the EPSPS gene driven by the CaMV 35S promoter, were generated through Agrobacterium-mediated transformation.The GE Arabidopsis plants were proven to tolerate the glyphosate herbicide [25].A total of 204 Arabidopsis plants were included in the experiment with the following greenhouse growth conditions: 22 • C all day, 16-h light, 8-h dark.
One GE rice (Oryza sativa) lineage (coded as W) overexpressing the same rice EPSPS transgene and its corresponding non-GE counterpart were used to represent the monocotyledonous plants.The rice plants were derived from the hybrids between an EPSPS GE rice line and a weedy rice population at the 4th generation (F 4 ) from Wang's experiment [22].The GE rice lineage, containing a single copy of the EPSPS gene driven by the ubiquitin (Ubi) promoter (from maize), was generated through Agrobacterium-mediated transformation.The GE rice plants also tolerated the glyphosate herbicide [22].A total of 42 rice plants were included in the experiment with the following greenhouse growth conditions: 30 • C during the day, 27 • C during the night, 8-h light, 16-h dark.Sampling and the measurement were made on 30-day-old Arabidopsis plants and 35-day-old rice plants with fully expanded leaves.

Gas-Exchange Measurements
Carbon fixation ratios were estimated by the intercellular CO 2 concentration (Ci) and photosynthetic ratios using a portable photosynthetic system equipped with the multiphase flash fluorometer and chamber (LI-6800, LI-COR, Inc., Logan, NE, USA).To ensure the accurate measurement, all gas-exchange measurements were conducted on three plants for each Arabidopsis or rice lineage at a constant atmospheric CO 2 concentration (Ca) with 400 µmol mol −1 , gas flow rate of 500 µmol s −1 , fan speed of 10,000 rpm, a leaf temperature of 25 • C, and 60% relative humidity (all controlled by the LI-6800).All the experimental leaves were sufficiently large enough to cover the entire window area of the equipment.The leaves were clamped on the window area for ~10 min before the measured parameters being stable, then the formal measurement started.To determine the light-saturated photosynthetic ratio of Arabidopsis plants, the photosynthetic measurement was made using a series of light intensities at the 1500, 1200, 800, and 500 µmol m −2 s −1 .The lightsaturated photosynthetic ratio and the intercellular CO 2 concentration (Ci) was determined within ten minutes at photosynthetic photon flux density (PPFD) of 1200 µmol m −2 s −1 for Arabidopsis.In addition, a PPFD of 1000 µmol m −2 s −1 was used for measuring Ci and photosynthetic ratio of rice plants within ten minutes following the Wang's description [22].Further, we used the Ca-Ci value to indicate the amount of atmospheric CO 2 consumption and fixation during the photosynthetic assimilation.The calculation of Ca-Ci value is based on the differences between measured air CO 2 concentrations (Ca) and intercellular CO 2 concentrations (Ci).

RNA-Sequencing and Analysis
Two GE Arabidopsis lineages overexpressing the EPSPS gene and two non-GE counterparts were selected for RNA-sequencing (RNA-seq).Each lineage included five biological replicates, each consisting of six Arabidopsis plants, and were randomly sampled from fully expanded leaves.Extraction of total RNA, reverse transcription, and adaptor ligation for hybridization were performed using the NEBNext UltraTM RNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA) for cDNA library construction.A total of 20 cDNA libraries were subjected to RNA-seq analysis using an Illumina novaseq6000 with a 150 bp read length and the sequencing was conducted using pair-end cDNA libraries.Our study had a sequence depth of ~20 million uniquely mapped reads per sample using the following procedure.First, adaptor sequences and low-quality sequences were trimmed using the Trimmomatic.Second, the filtered sequences were mapped to the reference A. thaliana genome using HISAT2.Subsequently, the aligned read files were processed by String Tie.After the reads were assembled into transcripts, their abundance was estimated and normalized using the numbers of reads Fragments Per Kilobase of transcript per Million fragments mapped reads (FPKM).
The differential expression analysis was performed using DEseq2.The threshold for differentially expressed genes (DEGs) was set as false discovery rates (FDR) ≤ 0.05 and fold change (FC) ≥ 1.5.Volcano plots were created based on these DEGs.To elucidate the particular DEGs' functional characteristics, all DEGs were compared using the Kyoto Encyclopedia of Genes and Genomes (KEGG) reference database of pathway networks or integration by BLASTx.For the KEGG enrichment analysis, pathways with p ≤ 0.05 were considered significantly enriched.For heatmap analysis, the key genes enriched into pathways and with the similar expression pattern were targeted and selected to perform expression profiles analysis.

Confirmation of the RNA-Seq Data by Real Time Quantitative PCR (RT-qPCR)
The key genes involved in important steps of carbon fixation, including rubisco activase (RCA), glyceraldehyde 3-phosphate dehydrogenase subunit A (GAPA), fructose 1,6-bisphosphate aldolase (FBA), and sedoheptulose-1, 7-bisphosphatase (SBPASE), were selected for expression confirmation in the Arabidopsis and rice experimental plants, based on the obtained RNA-Seq data.The primer pairs for each selected gene were designed automatically using Primer6 (Table S1).The RT-qPCR reactions were performed on Biorad (CFX 96) with Takara TB Green Kit (RR420Q) (TaKaRa, Dalian, China).The reaction volume was 25.0 µL, containing 12.5 µL TB Green Premix Ex Taq (Tli RNaseH Plus), 2.0 µL cDNA, 1.0 µL forward primer and 1.0 µL reverse primer (10 µM), and 8.5 µL ddH 2 O, following the manufactory protocols.All reactions were repeated three times.The RT-qPCR procedure was determined as 95 • C for 30 s and 40 cycles of 95 • C for 5 s, 60 • C for 30 s, and 72 • C for 30 s.The relative gene expression level was calculated with the 2 −△△Ct method [29] using ubiquitin 5 (UBQ5) [25] and ACTIN [30] as the reference gene in Arabidopsis and rice, respectively.

Statistical Analysis
The independent t test [30] was conducted to compare differences in expression of the EPSPS (trans)gene, intercellular CO 2 concentration (Ci), expression of DEGs, expression of the four carbon-fixation related genes, and photosynthetic ratios between GE plants and their corresponding non-GE counterparts.

Increases in EPSPS Gene Expression and Carbon Consumption in EPSPS Tansgenic Arabidopsis and Rice Lineages
Our results, based on the RT-qPCR method, showed significant increases (p < 0.001) in the average expression level of the EPSPS transgene in Arabidopsis and rice lineages.The EPSPS expression level was ~100-150 folds' higher in the GE Arabidopsis lineages overexpressing the EPSPS gene (Figure 1, A1 and A2, black columns) than their corresponding non-GE lineages with normal EPSPS expression (Figure 1, A1 and A2, white columns).Similarly, the expression level of EPSPS gene was significantly higher (p < 0.001) in the GE Biology 2024, 13, 25 5 of 18 rice lineage with ~50 fold higher gene expression than its corresponding normal non-GE lineage (Figure 1, W).

Arabidopsis and Rice Lineages
Our results, based on the RT-qPCR method, showed significant increases (p < 0.001) in the average expression level of the EPSPS transgene in Arabidopsis and rice lineages.The EPSPS expression level was ~100-150 folds' higher in the GE Arabidopsis lineages overexpressing the EPSPS gene (Figure 1, A1 and A2, black columns) than their corresponding non-GE lineages with normal EPSPS expression (Figure 1, A1 and A2, white columns).Similarly, the expression level of EPSPS gene was significantly higher (p < 0.001) in the GE rice lineage with ~50 fold higher gene expression than its corresponding normal non-GE lineage (Figure 1, W).The concentration of CO2 in intercellular (Ci), as an estimate of the carbon consumption, was directly measured in the experimental plants.Results from this study indicated a significantly lower level (p < 0.05) of remaining CO2 in the GE Arabidopsis and rice lineages with overexpression of the EPSPS gene (Figure 2, black columns) than that in their corresponding non-GE counterparts with normal EPSPS expression (Figure 2, white columns).The Ci decreased 3.2% to 3.7% in the GE Arabidopsis and rice lineages, compared with their corresponding non-GE counterparts with normal EPSPS expression.These results suggested that GE Arabidopsis and rice lineages consumed more CO2 than their corresponding non-GE counterparts in terms of the directly measured Ci values.The concentration of CO 2 in intercellular (Ci), as an estimate of the carbon consumption, was directly measured in the experimental plants.Results from this study indicated a significantly lower level (p < 0.05) of remaining CO 2 in the GE Arabidopsis and rice lineages with overexpression of the EPSPS gene (Figure 2, black columns) than that in their corresponding non-GE counterparts with normal EPSPS expression (Figure 2, white columns).The Ci decreased 3.2% to 3.7% in the GE Arabidopsis and rice lineages, compared with their corresponding non-GE counterparts with normal EPSPS expression.These results suggested that GE Arabidopsis and rice lineages consumed more CO 2 than their corresponding non-GE counterparts in terms of the directly measured Ci values.
Further, we included another value (the Ca-Ci value) to indicate the amount of CO 2 consumption during the photosynthetic assimilation for accurate estimate of the relationships between the level of increased exogenous CO 2 consumption and expression of the EPSPS gene in the GE and non-GE Arabidopsis (Figure 3a) and rice plants (Figure 3b).The obtained results showed that overexpression of the EPSPS gene consumed 14-20% greater CO 2 in the GE Arabidopsis and rice lineages than their corresponding non-GE counterparts with normal EPSPS expression.There was a significantly positive correlation (R 2 = 0.70-0.72,p < 0.05) between the levels of CO 2 consumption (as represented by the Ca-Ci values) and EPSPS gene expression in the GE and non-GE Arabidopsis and rice lineages (Figure 3a,b).These results indicated that a greater amount of CO 2 was consumed within a given time to meet the demands for carboxylation in the photosynthesis of the EPSPS GE plants having greater EPSPS gene expression.Further, we included another value (the Ca-Ci value) to indicate the amount of CO2 consumption during the photosynthetic assimilation for accurate estimate of the relationships between the level of increased exogenous CO2 consumption and expression of the EPSPS gene in the GE and non-GE Arabidopsis (Figure 3a) and rice plants (Figure 3b).The obtained results showed that overexpression of the EPSPS gene consumed 14-20% greater CO2 in the GE Arabidopsis and rice lineages than their corresponding non-GE counterparts with normal EPSPS expression.There was a significantly positive correlation (R 2 = 0.70-0.72,p < 0.05) between the levels of CO2 consumption (as represented by the Ca-Ci values) and EPSPS gene expression in the GE and non-GE Arabidopsis and rice lineages (Figure 3a,b).These results indicated that a greater amount of CO2 was consumed within a given time to meet the demands for carboxylation in the photosynthesis of the EPSPS GE plants having greater EPSPS gene expression.Further, we included another value (the Ca-Ci value) to indicate the amount of CO2 consumption during the photosynthetic assimilation for accurate estimate of the relationships between the level of increased exogenous CO2 consumption and expression of the EPSPS gene in the GE and non-GE Arabidopsis (Figure 3a) and rice plants (Figure 3b).The obtained results showed that overexpression of the EPSPS gene consumed 14-20% greater CO2 in the GE Arabidopsis and rice lineages than their corresponding non-GE counterparts with normal EPSPS expression.There was a significantly positive correlation (R 2 = 0.70-0.72,p < 0.05) between the levels of CO2 consumption (as represented by the Ca-Ci values) and EPSPS gene expression in the GE and non-GE Arabidopsis and rice lineages (Figure 3a,b).These results indicated that a greater amount of CO2 was consumed within a given time to meet the demands for carboxylation in the photosynthesis of the EPSPS GE plants having greater EPSPS gene expression.

Pathways with Differential Gene Expression Caused by Overexpressing the EPSPS Gene between GE and Non-GE Arabidopsis Lineages
To further investigate the underlying mechanism concerning why overexpression of the exogenous EPSPS gene increases the CO 2 consumption, we analyzed the GE and non-GE Arabidopsis lineages-the most commonly used model plants through the analysis of RNA-seq.Our results, based on the assay of differentially expressed genes (DEGs) in the Arabidopsis plants, indicated the considerable impact of overexpressing the EPSPS gene on the expression pattern of the global genes, although considerable variation in gene expression was observed between the two Arabidopsis lineages (Figure 4a,b; Tables S2 and S3).As shown in Figure 4a, 1113 DEGs were detected in the A1 Arabidopsis GE and non-GE lineages (Figure 4a).Of these DEGs, 656 genes displayed down-regulation and 457 showed upregulation.However, different from the A1 Arabidopsis lineage, 3766 DEGs were detected in the A2 Arabidopsis GE and non-GE lineages (Figure 4b).Of these DEGs, 1857 genes displayed down-regulation and 1909 showed upregulation.
pression was observed between the two Arabidopsis lineages (Figure 4a,b; Tables S2 and  S3).As shown in Figure 4a, 1113 DEGs were detected in the A1 Arabidopsis GE and non-GE lineages (Figure 4a).Of these DEGs, 656 genes displayed down-regulation and 457 showed upregulation.However, different from the A1 Arabidopsis lineage, 3766 DEGs were detected in the A2 Arabidopsis GE and non-GE lineages (Figure 4b).Of these DEGs, 1857 genes displayed down-regulation and 1909 showed upregulation.Results based on further analyses of the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment indicated that DEGs associated with overexpression of the EPSPS (trans)gene were categorized into different biological pathways of two Arabidopsis lineages (Figure 5).Unexpectedly, 271 (~31%) and 1281 (~39%) of the total DEGs were annotated by KEGG and were enriched in a total of 97 and 121 important pathways of the A1 and A2 Arabidopsis lineages, respectively (Figure 5a,b; Tables S4 and S5).Among these top 25 selected and displayed pathways, the porphyrin and chlorophyll metabolism, carbon fixation in photosynthetic organisms, phenylalanine, tyrosine and tryptophan biosynthesis, tryptophan metabolism, flavone and flavonol biosynthesis, and phenylpropanoid biosynthesis were included, which all showed a significant enrichment.These pathways were determined to be closely associated with overexpression of the EPSPS gene and carbon fixation.The above results might provide some insights into the mechanisms to explore the relationships between increased CO2 consumption and overexpression of the EPSPS (trans)gene.Results based on further analyses of the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment indicated that DEGs associated with overexpression of the EPSPS (trans)gene were categorized into different biological pathways of two Arabidopsis lineages (Figure 5).Unexpectedly, 271 (~31%) and 1281 (~39%) of the total DEGs were annotated by KEGG and were enriched in a total of 97 and 121 important pathways of the A1 and A2 Arabidopsis lineages, respectively (Figure 5a,b; Tables S4 and S5).Among these top 25 selected and displayed pathways, the porphyrin and chlorophyll metabolism, carbon fixation in photosynthetic organisms, phenylalanine, tyrosine and tryptophan biosynthesis, tryptophan metabolism, flavone and flavonol biosynthesis, and phenylpropanoid biosynthesis were included, which all showed a significant enrichment.These pathways were determined to be closely associated with overexpression of the EPSPS gene and carbon fixation.The above results might provide some insights into the mechanisms to explore the relationships between increased CO 2 consumption and overexpression of the EPSPS (trans)gene.
Results from the KEGG enhancement further indicated similar expression profiles of the critical genes (Figure 6) that were supposedly involved in the pathways closely associated with EPSPS gene overexpression or carbon fixation.Interestingly, some pivotal carbon fixation genes (i.e., GAPA, GAPB, EMB3119, and FBA1), particularly those involved in the photosynthetic ratio limiting steps (i.e., RCA and SBPASE), showed upregulated expression although with considerable variation in their expression levels between the A1 and A2 Arabidopsis lineages (Figure 6).The expression level of most chlorophyll biosynthesis genes (e.g., HEMA1, GUN5, PCB2, and ATG4) was significantly increased with overexpression of the EPSPS gene in this study (Figure 6).In addition, most genes (e.g., ASA1, TSB2, CM1, PAT, and PD1) that hypothetically participated in the biosynthesis of aromatic amino acids (i.e., phenylalanine, tyrosine, and tryptophan) and genes (e.g., CAD4, PAL2, CYP83B1, and UGT78D1) that were associated with metabolisms also showed significantly higher expression (Figure 6).Interestingly, genes that were supposed to be involved in the translocation (PPT2) and the last-three-steps in the biosynthesis of phosphoenolpyruvate (PEP)-one of the natural substrates of EPSPS enzyme-were also detected to be upregulated in terms of their expression (Figure 6).amino acids (i.e., phenylalanine, tyrosine, and tryptophan) and genes (e.g., CAD4, PAL2, CYP83B1, and UGT78D1) that were associated with metabolisms also showed significantly higher expression (Figure 6).Interestingly, genes that were supposed to be involved in the translocation (PPT2) and the last-three-steps in the biosynthesis of phosphoenolpyruvate (PEP)-one of the natural substrates of EPSPS enzyme-were also detected to be upregulated in terms of their expression (Figure 6).

Increases in Expression of the Key Carbon-Fixation Genes and Photosynthetic Ratios of GE Arabidopsis and Rice Lineages
Four key carbon-fixation genes were selected to confirm their actual expression level in the Arabidopsis and rice experimental plants.Consistently, the expression level of the four key genes that showed generally high expression in the RNA-seq analysis also displayed significant upregulation (p < 0.05), both in the Arabidopsis and rice plants using the RT-qPCR method (Figure 7).Expression of the RCA gene, which activated Rubisco responsible for catalyzing the primary step of CO 2 fixation, was significantly increased in the GE Arabidopsis (A1 by 1.9 folds and A2 by 1.8 folds) and in the GE rice lineage (W by 1.7 folds), compared with that in their corresponding non-GE counterparts (Figure 7a).Expression of the GAPA gene that is associated with the reduction action step, the main product exit point of Calvin-Benson Cycle, also showed significant upregulation in the GE Arabidopsis lineages (A1 by 2.4 folds and A2 by 2.2 folds) and in the GE rice lineage (W by 4.6 folds), compared with that in their corresponding non-GE counterparts (Figure 7b).Similarly, the two genes that are involved in RuBP regeneration (e.g., the FBA and SBPASE genes), affecting the maximum CO 2 assimilation rates, showed upregulation in the GE Arabidopsis and rice lineages for different folds (Figure 7c,d).These results all indicated that overexpressing the EPSPS gene could elevate expression of the carbon-fixation related genes, and consequently increased the CO 2 consumption of the GE plants.
(W by 4.6 folds), compared with that in their corresponding non-GE counterparts (Figure 7b).Similarly, the two genes that are involved in RuBP regeneration (e.g., the FBA and SBPASE genes), affecting the maximum CO2 assimilation rates, showed upregulation in the GE Arabidopsis and rice lineages for different folds (Figure 7c,d).These results all indicated that overexpressing the EPSPS gene could elevate expression of the carbon-fixation related genes, and consequently increased the CO2 consumption of the GE plants.To confirm whether the overexpression of the EPSPS gene could enhance photosynthesis, we measured photosynthetic ratios of the experimental plants using the LI-6800 portable photosynthetic system.Our results indicated significantly greater photosynthetic ratios (p < 0.01) in the GE Arabidopsis and rice lineages than those in their corresponding non-GE counterparts (Figure 8).The photosynthetic ratios were ~30% (A1) and ~40% (A2) greater in the two GE Arabidopsis lineages than those in their corresponding non-GE counterparts (Figure 8, A1 and A2).Similarly, our results also showed significantly greater photosynthetic ratio (p < 0.001) in the GE rice lineage, by ~34%, than that in its corresponding non-GE counterpart (Figure 8, W).These results generated from both monocotyledonous Arabidopsis and dicotyledonous rice suggested that overexpressing the EPSPS gene might generally increase the photosynthesis of all plants.
greater in the two GE Arabidopsis lineages than those in their corresponding non-GE counterparts (Figure 8, A1 and A2).Similarly, our results also showed significantly greater photosynthetic ratio (p < 0.001) in the GE rice lineage, by ~34%, than that in its corresponding non-GE counterpart (Figure 8, W).These results generated from both monocotyledonous Arabidopsis and dicotyledonous rice suggested that overexpressing the EPSPS gene might generally increase the photosynthesis of all plants.

Overexpressing the EPSPS Transgene Substantially Increased CO2 Consumption in Plants
In this study, we found that the expression level of the EPSPS gene increased substantially in the fourth generation (T4) of genetically engineered (GE) Arabidopsis plants and GE rice/weedy rice hybrid lineages (F4).The EPSPS gene was cloned from cultivated rice [31] and driven by the constitutive promotors CaMV35S (for GE Arabidopsis) and Ubiquitin (for GE rice) [22,25].This finding was generated based on the comparative analyses of average mRNA expression between the GE Arabidopsis/rice plants and their corresponding non-GE counterparts.The GE plants overexpressed the EPSPS gene, whereas the non-GE plants had normal EPSPS expression.Importantly, our experimental materials included in this study represented both dicotyledonous (Arabidopsis) and monocotyledonous (rice) plants.Therefore, we expect that the proper modification of the EPSPS gene(s) through GE technologies can substantially increase the expression level of this gene both in dicots and monocots.This prediction is supported by previous results published by different authors [22,23,25,26].In addition, the detected expression level of the EPSPS gene(s) in the GE Arabidopsis and rice plants was largely variable.Therefore, it is possible to select transgenic events with substantially enhanced EPSPS overexpression in particular plants using more effective GE technologies.
Interestingly, we also found in this study that the intercellular CO2 concentration (Ci) was significantly reduced both in the GE Arabidopsis and rice plants with significantly greater EPSPS gene expression, although the measurement for Ci only lasted for about ten minutes.The intercellular CO2 concentration, as a critical parameter related to carbon

Overexpressing the EPSPS Transgene Substantially Increased CO 2 Consumption in Plants
In this study, we found that the expression level of the EPSPS gene increased substantially in the fourth generation (T 4 ) of genetically engineered (GE) Arabidopsis plants and GE rice/weedy rice hybrid lineages (F 4 ).The EPSPS gene was cloned from cultivated rice [31] and driven by the constitutive promotors CaMV35S (for GE Arabidopsis) and Ubiquitin (for GE rice) [22,25].This finding was generated based on the comparative analyses of average mRNA expression between the GE Arabidopsis/rice plants and their corresponding non-GE counterparts.The GE plants overexpressed the EPSPS gene, whereas the non-GE plants had normal EPSPS expression.Importantly, our experimental materials included in this study represented both dicotyledonous (Arabidopsis) and monocotyledonous (rice) plants.Therefore, we expect that the proper modification of the EPSPS gene(s) through GE technologies can substantially increase the expression level of this gene both in dicots and monocots.This prediction is supported by previous results published by different authors [22,23,25,26].In addition, the detected expression level of the EPSPS gene(s) in the GE Arabidopsis and rice plants was largely variable.Therefore, it is possible to select transgenic events with substantially enhanced EPSPS overexpression in particular plants using more effective GE technologies.
Interestingly, we also found in this study that the intercellular CO 2 concentration (Ci) was significantly reduced both in the GE Arabidopsis and rice plants with significantly greater EPSPS gene expression, although the measurement for Ci only lasted for about ten minutes.The intercellular CO 2 concentration, as a critical parameter related to carbon fixation, was reported in a previous study where carbon fixation was determined by the carboxylation activity or activated amount of ribulose-1,5-bisphosphate carboxylase/oxygenase [32].The published results supported our findings that reduced intercellular CO 2 concentration is a critical determinant to measure carbon fixation.In addition, we also detected a significant positive correlation (R 2 > 0.70, p < 0.05) between the levels of atmospheric CO 2 consumption (Ca-Ci) and EPSPS gene expression when the parameter Ci was converted into Ca-Ci.Altogether, these findings suggest that overexpression of the EPSPS gene in our experimental plants can considerably increase the consumption or fixation of atmospheric CO 2 .This conclusion addressed our first question concerning overexpression of the EPSPS transgene and fixation of CO 2 in plants.In our study, ~20% greater amount of CO 2 was consumed by the GE experimental materials than their corresponding non-GE counterparts.Based on these results, we can make a prediction that the appropriate application of overexpression of the EPSPS gene(s) in plants through GE technologies can considerably increase fixation of CO 2 .Statistical data showed that one third of CO 2 is produced by anthropogenic activities [14,15].Therefore, our findings provide a new opportunity for global carbon neutrality through genetically engineered plants.

Possible Mechanisms for Increased CO 2 Consumption in Genetically Engineered Plants Overexpressing EPSPS Gene
Our results, based on the transcriptomic analyses, evidently demonstrated that overexpressing the EPSPS gene had a substantial impact on the expression level of many genes in different biological pathways.This finding is based on the comparison of differentially expressed genes (DEGs) between GE and non-GE Arabidopsis plants.Particularly, our results from KEGG (Kyoto Encyclopedia of Genes and Genomes) analyses indicated significant gene enrichment in the carbon-fixation related pathways.All the carbon-fixation related pathways enriched in this study are found in the down-streams of the shikimate pathway, including the phenylalanine, tyrosine, and tryptophan biosynthesis pathways, and the phenylpropanoid biosynthesis pathway.The above results indicate that more carbon from the atmosphere can be fixed due to overexpression of the EPSPS gene(s).Obviously, overexpression of the EPSPS gene can trigger upregulation of many other genes in the carbon-fixation related pathways.
This explanation is supported by many previously published results in which overexpression of the EPSPS gene significantly increased the production of aromatic amino acids [22,23], lignin [33], and biomasses [22,23,25] of the GE Arabidopsis and rice plants.All the produced metabolic matters mentioned above are closely associated with the carbonfixation related pathways [27,28].As a matter of fact, metabolites of phenylpropanoid and aromatic amino acids, such as phenylalanine, tyrosine, and tryptophan, are synthesized in the carbon-fixation related pathways, accounting for ~20% of total carbon fixed in plants [27,28].In addition, our results also indicated upregulated expression of the PPT2 gene that is reported to be responsible for translocating EPSPS's substrate PEP-the carbon skeleton of the shikimate-pathway derived metabolites [34,35].These results again suggest more carbon being fixed in plants by overexpressing the EPSPS gene.Published results also indicated that increased expression of the PPT gene can stimulate carbon fixation by increasing the content of aromatic amino acids, the number of fruits, and biomasses of plants [36].Therefore, our conclusion in this study, that overexpression of the EPSPS gene(s) can fix more carbon by enhancing expression of many genes, gains supports from previous published results.
Accordingly, we examined the expression of four important carbon-fixation related genes, RCA, GAPA, FBA, and SBPASE, with a consistent upregulating pattern in the two examined GE Arabidopsis lineages based on our RNA-seq analysis.In addition to our results, published reports also indicated that these genes played critical roles in the three phases of the carbon-fixation pathway in most organisms with photosynthesis [37].The four genes were found to be significantly upregulated in Arabidopsis and rice plants overexpressing the EPSPS gene.This finding confirms the consistency of the upregulation patterns concerning carbon-fixation related genes between the RT-qPCR and RNA-seq analyses from our experiment.We consider that the increases in CO 2 fixation in the GE Arabidopsis and rice plants is likely due to upregulation of the key genes in the carbonfixation related pathways because of the overexpression of the EPSPS gene.This conclusion is based on our findings about the upregulation of carbon-fixation related genes.
In fact, the expression changes of the four genes (RCA, GAPA, FBA, and SBPASE) have downstream effects on plants' physiology or metabolism, including photosynthesis, biomass, and plant growth.The previous studies also indicated that the four genes were the key to the three phases of carbon fixation, namely, carbon fixation (RCA), reduction reaction (GAPA), and ribulose 1,5-bisphosphate regeneration (FBA and SBPASE) [37].First, the RCA gene is essential for the activation and maintenance of catalytic activity of rubisco that catalyzes the first step of carbon fixation [38][39][40].Overproduction of RCA can enhance the carbon fixation in Arabidopsis and rice [41,42].Second, the GAPA gene encodes the subunits GapA of GAPDH involving in the reduction action of carbon fixation.Overexpression of GAPDH can increase the ratios of carbon fixation in rice [43].Third, the FBA gene is associated with photosynthetic carbon flux [44,45], and increased FBA in plastids can enhance carbon fixation of tobacco plants [46].The SBPASE gene is essential for maintaining the balance between the carbon required for the regeneration of ribulose 1,5-bisphosphate and required for the biosynthesis of different polysaccharides [47].In addition, it was found in wheat and Arabidopsis that increased SBPASE activities could also improve carbon fixation [48,49].All these results and findings support our conclusion that upregulation of the key genes in the carbon-fixation related pathways through overexpressing the EPSPS gene enhances carbon fixation.
We also found increased photosynthetic ratios and critical genes that are involved in chlorophyll biosynthesis [50] in the GE Arabidopsis and rice plants overexpressing the EPSPS gene.Previous studies reported that enhanced chlorophyll biosynthesis is closely related to photosynthesis [51,52].All these findings provide evidence for the explanation of increased photosynthesis in the GE rice [22] and Arabidopsis in our study.Given that increased photosynthesis is also associated with increased CO 2 assimilation [53], our findings regarding overexpression of the EPSPS gene that enhances carbon fixation provide indirect evidence of increased photosynthesis from another perspective.In summary, we propose that increased CO 2 fixation caused by overexpression of the EPSPS gene is most likely owing to the upregulation of the key genes in the carbon-fixation related pathways.Increased photosynthetic ratios provide additional evidence for increased carbon fixation in plants.These findings facilitate our understanding about the relationships between overexpression of the EPSPS gene(s) and carbon fixation in plants.These findings can also provide theoretical foundation for regulating carbon fixation and crop production through genetic engineering of plant species with enhanced expression of the endogenous or exogenous EPSPS gene(s).

Implications of Overexpressing EPSPS in Genetically Engineered Plants for Carbon Neutrality
Our results clearly indicated significant variation in CO 2 fixation among different GE Arabidopsis and rice plants coped with diverse levels of EPSPS gene expression.This finding provides great opportunities for us to select a certain transgenic event that has an extremely high level of EPSPS gene expression as a candidate for ideal CO 2 fixation in the plant breeding programs.There are many methods available that can regulate the expression level of GE plants, such as the use of different promotors with variable driving efficiencies for the same genes [54] and the selection of exogenous and endogenous genes from different sources for genetic engineering [22,25,55,56].We also found a significantly positive correlation between the levels of carbon fixation, as represented by Ca-Ci that showed considerable variation, and expression of the EPSPS gene in this study.These results suggest that the expression level of the EPSPS gene can determine the efficiency of carbon fixation in transgenic events of different GE plants of the same species or hybrid lineages.If the hypothesis is true, we can deliberately apply different GE technologies to increase the expression level of the EPSPS gene(s) in different plant species to achieve the most ideal carbon fixation efficiencies.Furthermore, we found in our transcriptomic analyses that overexpressing the EPSPS gene enhanced the expression level of the key genes in carbon-fixation related pathways, resulting in increased carbon fixation.These findings also suggest the possibility of an alternative strategy by selecting other proper genes, such as the RCA and SBPASE genes, in the carbon-fixation related pathways to enhance their (over)expressing level through GE technologies for achieving the direct increases in carbon fixation.
Accordingly, we can develop better strategies to effectively reduce the concentration of atmospheric CO 2 to achieve successful carbon neutrality through carbon fixation, applying the biotechnology of genetic engineering, particularly in the cropland, even in artificial woodland and artificial grassland.Noticeably, the EPSPS genes are found in all plant species, responsible for the biosynthesis of important secondary metabolites, such as lignin and auxin [28].Theoretically, it is possible to increase expression of the EPSPS genes in all plant species for enhanced CO 2 fixation by genetic engineering.Taking the above facts into consideration, we believe that more atmospheric CO 2 can be assimilated or consumed by GE plants that overexpress the EPSPS genes.Therefore, different technologies can be selected and applied to modify plant species that are suitable to grow in different terrestrial ecosystems, such as in croplands and artificial woodlands.Ideal strategies that combine the proper genetic engineering technologies to modify the EPSPS genes with the properly selected plant species to fix carbon (CO 2 ) efficiently in different areas will offer a great potential for human to meet the challenges of the global climate change and to ensure carbon neutrality, which may offset a large portion of anthropogenic CO 2 emissions.
Cropland that has a great potential to develop GE plants covers ~10% of the Earth's land surface [57], with the estimated coverage of 1.24 billion hectares in 2019 [58].Given that GE crops are extensively grown in the world, with more than 190 million hectares in 2019 [59], it is most likely to develop GE crops overexpressing the EPSPS genes.For example, crop rice (Oryza sativa) is cultivated worldwide, accounting for ~11% of the global cropland with an estimated cultivation area of 165.5 million hectares [60].As an important crop that provides staple food for nearly one half of the global population [61], this plant species alone already provides significant opportunities for reducing atmospheric CO 2 by genetic engineering of its endogenous, as well as exogenous, EPSPS genes.The alternative example is to engineer tree species, because the global coverage of forests is ~4.06 billion hectares-approximately 30% of the total terrestrial area [15].Usually, tree species occurring in forests have a long growth period, which can promote sustainable carbon fixation during their lifespans.It is reported that the global coverage of artificial woodlands increased by 4.1 million hectares per year between 2000~2020 and the current coverage is ~293 million hectares, as estimated in 2023 [62].Therefore, tree species, particularly the artificial woodland species, provide ideal plants for enhanced carbon fixation by genetic engineering.In addition, plant species occurring in the grasslands can also be used to meet the objectives, as grasslands cover ~3.5 billion hectares, accounting for ~26% of the ice-free land on Earth [63].
Therefore, the possibilities of applying plant species in different areas in combination with advanced biotechnologies for genetic engineering allow human being to manage CO 2 fixation from multiple dimensions to achieve global carbon neutrality.However, it is necessary to point out the potential challenges and limitations regarding the application of GE plants.Currently, GE plants cannot be freely grown in the world.According to the biosafety regulations, all GE plants that are intended to be released to the environment for large-scale application should be subjected to risk assessment [30,[64][65][66].Only those GE plants that receive a biosafety certificate are permitted to grow in a given area.Therefore, the potential challenges and limitations of applying GE plants overexpressing the EPSPS genes associated with biosafety issues need to be considered and addressed properly.

Conclusions
In this study, we found substantially reduced intercellular carbon dioxide (CO 2 ) concentration (Ci) in genetically engineered (GE) Arabidopsis (dicot) and rice (monocot) plants that overexpressed the rice EPSPS gene.Further analyses indicated a positive correlation between the levels of atmospheric CO 2 consumption (Ca-Ci) and EPSPS gene expression in these Arabidopsis and rice plants.The possible reason for substantially increased atmospheric CO 2 consumption by overexpressing the EPSPS gene is upregulation of many key genes in biological pathways, particularly those in the carbon-fixation related pathways.For example, expression of the carbon-fixation associated genes such as RCA, GAPA, FBA, and SBPASE showed significant increases.Given that our obtained results are based on the dicot (Arabidopsis) and monocot (rice) plants, overexpression of EPSPS gene(s) promoting increased CO 2 consumption may represent more plant species in nature.
Altogether, findings from this study have a great implication to design a novel strategy not only for the global carbon neutrality as the carbon sink, but also for increases in food production due to increases in photosynthetic ratios.In other words, overexpression of endogenous or exogenous EPSPS genes in plants through GE biotechnologies can promote a higher level of consumption of atmospheric CO 2 in different ecosystems, which is important for global carbon neutrality.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/biology13010025/s1,Table S1: Primers of the genes from the transcriptome analysis; Table S2: Differentially expressed genes (DEGs) between the A1 GE Arabidopsis lineage overexpressing the EPSPS gene and its corresponding non-GE counterpart; Table S3: Differentially expressed genes (DEGs) between the GE A2 Arabidopsis lineage overexpressing the EPSPS gene and its corresponding non-GE counterpart; Table S4: Enriched KEGG pathways between the A1 GE Arabidopsis lineage overexpressing the EPSPS gene and its corresponding non-GE counterpart; Table S5: Enriched KEGG pathways between the A1 GE Arabidopsis lineage overexpressing the EPSPS gene and its corresponding non-GE counterpart.

Figure 1 .
Figure 1.Comparison of the average expression (relative mRNA) level of the rice EPSPS (trans)gene between the GE lineages (black columns) overexpressing the EPSPS gene and their corresponding non-GE lineages (white columns), based on the independent t test [30].A1 and A2 represent the Arabidopsis lineages; W represents rice lineage.Bars indicate standard errors; *** p < 0.001.

Figure 1 .
Figure 1.Comparison of the average expression (relative mRNA) level of the rice EPSPS (trans)gene between the GE lineages (black columns) overexpressing the EPSPS gene and their corresponding non-GE lineages (white columns), based on the independent t test [30].A1 and A2 represent the Arabidopsis lineages; W represents rice lineage.Bars indicate standard errors; *** p < 0.001.

Figure 2 .
Figure 2. Comparison of average intercellular carbon dioxide (CO2) concentration between the GE lineages (black columns) overexpressing the EPSPS gene and their corresponding non-GE lineages (white columns), based on the independent t test [30].A1 and A2 represent Arabidopsis lineages; W represents rice lineage.Bars represent standard errors; * p < 0.05, *** p < 0.001.

Figure 3 .
Figure 3. Correlation between the calculated consumption level (Ca-Ci) of carbon dioxide (CO2) and EPSPS (trans)gene expression in the GE (black dots) and non-GE (black triangles) Arabidopsis (a) and rice (b) plants.The Ca-Ci value is calculated as the differences between measured air CO2 concentrations (Ca) and intercellular CO2 concentrations (Ci).Dashed lines indicate linear regression.

Figure 2 .
Figure 2. Comparison of average intercellular carbon dioxide (CO 2 ) concentration between the GE lineages (black columns) overexpressing the EPSPS gene and their corresponding non-GE lineages (white columns), based on the independent t test [30].A1 and A2 represent Arabidopsis lineages; W represents rice lineage.Bars represent standard errors; * p < 0.05, *** p < 0.001.

Figure 2 .
Figure 2. Comparison of average intercellular carbon dioxide (CO2) concentration between the GE lineages (black columns) overexpressing the EPSPS gene and their corresponding non-GE lineages (white columns), based on the independent t test [30].A1 and A2 represent Arabidopsis lineages; W represents rice lineage.Bars represent standard errors; * p < 0.05, *** p < 0.001.

Figure 3 .
Figure 3. Correlation between the calculated consumption level (Ca-Ci) of carbon dioxide (CO2) and EPSPS (trans)gene expression in the GE (black dots) and non-GE (black triangles) Arabidopsis (a) and rice (b) plants.The Ca-Ci value is calculated as the differences between measured air CO2 concentrations (Ca) and intercellular CO2 concentrations (Ci).Dashed lines indicate linear regression.

Figure 3 .
Figure 3. Correlation between the calculated consumption level (Ca-Ci) of carbon dioxide (CO 2 ) and EPSPS (trans)gene expression in the GE (black dots) and non-GE (black triangles) Arabidopsis (a) and rice (b) plants.The Ca-Ci value is calculated as the differences between measured air CO 2 concentrations (Ca) and intercellular CO 2 concentrations (Ci).Dashed lines indicate linear regression.

Figure 4 .
Figure 4. Significantly up-regulated (red dots) and down-regulated genes (light-blue dots) indicated by the volcano plots based on the differentially expressed genes (DEGs) between the GE plants overexpressing the EPSPS gene and their corresponding non-GE Arabidopsis plants (A1, (a) and A2, (b)).FDR represents the false discovery rates with significances; FC represents the fold changes, indicating the expression fold changes of genes between the GE plants overexpressing the EPSPS gene and their corresponding non-GE plants; Grey dots indicate genes with no significant expression changes.Dashed lines indicate the significance threshold of FDR ≤ 0.05 (horizontal) and the fold change threshold of FC ≥ 1.5 (longitudinal), respectively.

Figure 4 .
Figure 4. Significantly up-regulated (red dots) and down-regulated genes (light-blue dots) indicated by the volcano plots based on the differentially expressed genes (DEGs) between the GE plants overexpressing the EPSPS gene and their corresponding non-GE Arabidopsis plants (A1, (a) and A2, (b)).FDR represents the false discovery rates with significances; FC represents the fold changes, indicating the expression fold changes of genes between the GE plants overexpressing the EPSPS gene and their corresponding non-GE plants; Grey dots indicate genes with no significant expression changes.Dashed lines indicate the significance threshold of FDR ≤ 0.05 (horizontal) and the fold change threshold of FC ≥ 1.5 (longitudinal), respectively.

Figure 5 .
Figure 5. Biological pathways with gene expression changes caused by overexpression of the EPSPS gene based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment in the GE plants overexpressing the EPSPS gene and their corresponding non-GE Arabidopsis plants (A1, (a) and A2, (b)).Red stars represent pathways closely related to carbon fixation.The rich factor indicates the significance of the enrichment level; P represents the reliability of the significance of the enrichment level.

Figure 5 .
Figure 5. Biological pathways with gene expression changes caused by overexpression of the EPSPS gene based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment in the GE plants overexpressing the EPSPS gene and their corresponding non-GE Arabidopsis plants (A1, (a) and A2, (b)).Red stars represent pathways closely related to carbon fixation.The rich factor indicates the significance of the enrichment level; P represents the reliability of the significance of the enrichment level.

Figure 6 .
Figure 6.Variation in the expression profiles of the target genes in the selected pathways closely associated with EPSPS function and carbon fixation between the A1 and A2 Arabidopsis lineages based on the Heatmap.FC indicates the expression fold change of genes between the GE lineages overexpressing the EPSPS gene and their corresponding non-GE lineages.

Figure 6 .
Figure 6.Variation in the expression profiles of the target genes in the selected pathways closely associated with EPSPS function and carbon fixation between the A1 and A2 Arabidopsis lineages based on the Heatmap.FC indicates the expression fold change of genes between the GE lineages overexpressing the EPSPS gene and their corresponding non-GE lineages.

Figure 7 .
Figure 7.Comparison of expression levels of the four carbon-fixation related genes using the RT-qPCR method between GE (overexpressing the EPSPS gene, black columns) and non-GE (white columns) Arabidopsis (A1 and A2) and rice (W) lineages, based on the independent t test [30].Bars represent standard errors; (a-d) indicate the RCA, GAPA, FBA, and SBPASE genes, respectively.* p < 0.05, ** p < 0.01.

Figure 7 .
Figure 7.Comparison of expression levels of the four carbon-fixation related genes using the RT-qPCR method between GE (overexpressing the EPSPS gene, black columns) and non-GE (white columns) Arabidopsis (A1 and A2) and rice (W) lineages, based on the independent t test [30].Bars represent standard errors; (a-d) indicate the RCA, GAPA, FBA, and SBPASE genes, respectively.* p < 0.05, ** p < 0.01.