Carbon and nitrogen partitioning of transgenic rice T2A-1 (Cry2A*) with different nitrogen treatments

Nitrogen (N) and carbon(C) metabolisms in plants were investigated to assess different responses of Bt and non-Bt rice to different N treatments. T2A-1 (Bt rice variety) inserted with Cry2A* protein to resist Lepidoptera and its parental line MH63 was adopted in this study. The total N accumulation presented no statistical difference. But nitrogen contents in different parts of rice plant were significantly different between the two lines, especially on leaf and spike part. This study revealed that the nitrogen in leaf of T2A-1 was far more than that of MH63; however, the nitrogen in spike of T2A-1 was less than that of MH63. In addition, MH63 assimilated more carbon than T2A-1. However, the distribution proportion of carbon in leaf, stem and spike of T2A-1 and MH63 were both 1:1:1. What’s more, our study of the difference in metabolism pathway based on proteomics analysis provided more insights on the responses of two lines of Bt and non-Bt rice to different N treatments. And amino acid metabolism, energy metabolism, and carbohydrate metabolism presented significant difference between two lines. In addition, the number of differentially expressed proteins with N deficiency treatment was almost twice as many as that with normal N treatment. It could be inferred that the insertion of Cry2A* in T2A-1 may bring about effects on carbon and nitrogen allocation and related metabolisms, especially under N deficiency environment.

Since rice yield suffers great loss due to lepidopteran pests 1 , scientists worked on possible control measures including developing transgenic Bt rice in the last few decades [2][3][4][5][6][7] . Cultivation area of transgenic Bt crop was up to 98.5 million hectares worldwide in 2016 8 . In recent years, planting Bt crops has been widely recognized as a pest prevention measure. Although many Bt genes have been found, only a few of them were selected for developing transgenic crops. Cry2A protein was reported to have different binding regions in brush border membrane vesicles and to present no cross-resistance [9][10][11][12] . Cry2A gene could be applied to develop marker-free Bt transgenic rice as a resistance source of gene pyramid. As a result, the evolution of insects' resistance to toxicity of Bt rice would be postponed. Thus, Cry2A was considered as an important protein for bio-breeding scientists. However, several previous studies demonstrated the differences in the agronomy traits and physiological metabolism in some rice lines inserted by exogenous insect-resistant genes [13][14][15][16][17][18][19] . These differences included reduced plant heights and root lengths 14 , fewer grains per panicle 17 , decreased setting rates 5,14,17,18,20 , and so on, which in turn led to reductions in grain yield. The obvious advantage of Bt rice in pest resistance does not balance out its unintended effects resulting in lower yield 21 . These unintended effects brought by transgenic crop were supposed to have three sources: the disruption of endogenous genetic background 22 , the somaclonal variation happened during tissue culture processes 23 , and "bio-burden" brought by additional transgenic protein synthesis process 24 . "Bio-burden" was described as burden on material and energy, which could influence N and C metabolisms in plant.
The N amount demanded by plants is highest among all mineral nutrients. And it was one of the most important elements influencing crop yield directly. Generally, with the increase of N application, more Bt protein in plant tissues could be detected 20,25,26 . According to Bruns and Abel 25 , there was no significant difference in the total N uptake between Bt maize and their non-Bt counterparts. However, some other Bt hybrid maize strains were demonstrated to have lower N accumulation in grains but higher N accumulation in straws compared with their non-Bt control lines 27 . Moreover, Pioneer 38W36Bt, a Bt hybrid maize, accumulated more N in its kernels and entire maize plant than its conventional control Pioneer 38W36. But the two maize lines showed similar N and leaf chlorophyll contents at the late growth stages 28 . Besides, Bt cotton was reported to have higher N content Nitrogen accumulation and distribution in Bt (T2A-1) and non-Bt rice. There was no significant difference in nitrogen assimilation between Bt (T2A-1) and non-Bt rice. However, the distribution of nitrogen in leaf, stem and spike presented several differences at the two sampling times respectively on FS and day 15 after FS, especially on leaf and spike part (Table 3). To be more specific, mean value of nitrogen concentration in leaf of T2A-1 were 6.41% and 6.35% higher than that of MH63 respectively (Table 4). While the mean value in spike of T2A-1 were 4.74% and 8.38% lower than that of MH63. And the nitrogen content in each part of both lines T2A-1 and MH63 exhibited the same ranking, namely, leaf > stem > spike at FS. Whereas 15 days after FS, the ranking of the two lines was changed into stem > leaf > spike. www.nature.com/scientificreports www.nature.com/scientificreports/ C/N ratio in different part of Bt (T2A-1) and non-Bt rice. Under the four nitrogen treatments, C/N ratio in different parts of T2A-1 and MH63 presented significant difference (Table 5). At both sampling time, i.e., FS and day 15 after FS, C/N ratio was the highest in stem, and the lowest in leaf. Furthermore, C/N ratio of T2A-1 was significantly lower under N4 treatment than under other N treatments. And the C/N ratio of MH63 didn't show obvious difference under its four treatments. At FS, C/N ratio in stem of T2A-1 is higher under N1 treatment than under N2 and N3 treatments. While, on day 15 after FS, there was no difference on C/N ratio under all treatments N1-4 of T2A-1. And C/N ratio in leaf and spike was the lowest under N4 treatment, which was significantly lower than that under N1 treatment. The C/N ratio of whole aboveground part of T2A-1 was significantly lower than that of MH63, and so was the C/N ratio in leaf.
Differentially expressed proteins related to C and N metabolism in leaves of MH63 and T2A-1 by iTRAQ. Flag leaves of MH63 and T2A-1 under RN(1 g N·plant −1 ) and N0 (0 g N·plant −1 ) treatments were collected for proteomics analysis in iTRAQ (isobaric tags for relative and absolute quantification) method. The analysis results indicated that of a total of 6040 proteins identified in the experiment, 206 proteins were up-regulated and 315 proteins were down-regulated under RN treatment (Fig. 1). While 320 up-regulated and  Table 3. Accumulation and distribution of N of T2A-1 and MH63 under different nitrogen treatments at several growth periods (mg). Data are presented as the means ± standard deviation (SD, n = 3). Lowercase letters indicate SNK variance between groups under same nitrogen treatments at same sampling time. The * indicate a significant source of variance at P = 0.05, while ** at P = 0.01. NS means no significance. N1: 0.2 g N pot −1 ; N2: 0.35 g N pot −1 ; N3: 0.5 g N pot −1 ; N4: 1 g N pot −1 . FS: flowering stage.
www.nature.com/scientificreports www.nature.com/scientificreports/ 217 down-regulated proteins were identified in the leaves of two rice lines at N0 level. In addition, the number of those differentially expressed proteins related to C and N metabolism under the two N treatments were 55 and 61 (Tables 6 and 7), occupying 17.2% and 28.1% of the total difference at RN and N0 levels respectively, and the KEGG analysis of those differentially expressed proteins were shown in Fig. 2.
The differentially depressed proteins were devoted to energy metabolism, carbohydrate and lipid metabolism, nucleotide metabolism, amino acid metabolism, glycan biosynthesis and metabolism, terpenoids and polyketides metabolism, and other secondary metabolites biosynthesis, all of which belonged to C and N metabolism. The concerning differentially expressed proteins were listed in Tables 6 and 7.

Discussion
This study found no significant difference in total biomass between T2A-1 and MH63 under different N treatments, indicating that the two lines had similar response to biomass assimilation. However, Bt rice line T2A-1 presented higher dry matter content in leaves and lower content in spike compared with its counterpart MH63. Previous studies reported other differences in agronomy traits. For example, lower yield caused by lower setting rates was also reported in several Bt rice lines introduced with a Cry2A* 31 , Cry1C* 20 , Cry1Ab 17 , Cry1Ac 14 , Xa21 4 or Bt/CpTI 18 gene. And shorter plant heights and root lengths 14 , fewer grains per panicle 17 and lower setting rates 5,14,17,18,20 have also been reported in recent years. These differences in agronomy traits could not be ignored for optimizing cultivation of transgenic crops.
Besides, the results of accumulation and distribution of N examination were similar to those of biomass in the two lines, T2A-1 and MH63. N content in spike of T2A-1 was lower than that in spike of MH63, potentially resulting in the weight loss of spike of T2A-1. The experiment with maize indicated that lower one in grains but higher N accumulation in stovers were found in several Bt hybrid maize strains, compared with their non-Bt counterparts 27 . And more N in kernels was detected in Pioneer 38W36Bt, a Bt hybrid maize, compared with the control Pioneer 38W36 (non-Bt), while N concentration and leaf chlorophyll content at the silking and maturity stages in Bt hybrid maize were similar to that in control group 28 . Bt cotton was reported to have more active N metabolism in the vegetative stage than its non-Bt counterpart, resulting in a reduction in boll size 29 . It could be concluded that there existed differences in N metabolism between Bt crops and their non-Bt crops including rice, maize and cotton.
Additionally, the significant differences in the accumulation of C between the two lines under several N treatments were also observed. But C distribution ratios in plant parts of the two lines were the same, i.e., 1:1:1. Proteomics analysis results revealed that the greatest difference in the number of differentially expressed proteins related to carbohydrate metabolism was observed between RN and N0 treatment. As we know, C metabolism provides material and energy for all metabolism processes in plant. Thus, changes on C metabolism would have enormous influence on plant growth and development. Our proteomics analysis results obtained by iTRAQ method exhibited that differentially expressed proteins mainly functioned in energy metabolism, carbohydrate metabolism, and so on. Previous reports on Bt cotton revealed that differentially expressed proteins identified between transgenic Bt + CpTI cotton SGK321 and its counterpart SY321 accounted for approximate 10% of all proteins identified by using proteomics analysis. These differentially expressed proteins mainly involved in carbon fixation and photosynthesis, glyoxylate and dicarboxylate metabolism pathway, oxidative pentose phosphate pathway 32   www.nature.com/scientificreports www.nature.com/scientificreports/ that most of these differentially expressed proteins were involved in energy transfer, oxidative respiration, and photosynthesis. In addition, Gong et al. used comparative proteomics approaches to determine proteome differences in seeds between 2 transgenic rice lines and their corresponding control groups, namely, D68 and MH86 34 . Mass spectrometry analysis exhibited that differentially expressed proteins participated in several cellular and metabolic processes, including protein synthesis and transportation, and defense response. And differentially expressed proteins were also detected in seeds of transgenic and non-transgenic soybean by comparative proteomics approach 35 . Therefore, it can be concluded that changes in C metabolism were also widely found in transgenic plants.

Analysis of variance
It is well known that C metabolism and N metabolism are always intertwined in plant 36 . Up to 55% of net plant carbon was devoted to nitrogen assimilation and metabolism in some tissues 37 . Carbon and nitrogen metabolism are intimately linked. This study revealed different C/N ratios in leaf between T2A-1 and MH63. The C/N ratio in whole aboveground part of T2A-1 was significantly lower than that of MH63. C/N ratio has influence on glucose metabolism, assimilate transportation, senescence process of leaf, etc. Thus, as plant material and energy suppliers, C and N metabolism ought to be taken into consideration simultaneously in their function assessment.     Continued www.nature.com/scientificreports www.nature.com/scientificreports/ Until now, there were three hypotheses of unintended effects brought by transgenic technology. But insertion position of transgenic rice line T2A-1 used in our study was noncoding region, and it was chosen to exhibit the best field resistance and excellent agronomy traits by several field experiment among 102 transgenic rice lines 6 . Thus, we are inclined to attribute these effects to bio-burden caused by additional Bt protein synthesis. As a soluble protein, Bt protein would consume extra N and energy in Bt crops. And C and N metabolism were assumed as basic processes regulating N allocation and energy distribution in plant. Thus, it has drawn a wide concern whether the Bt protein synthesis process will influence the fundamental C and N metabolism in Bt crops resulting in a possible "bio-burden" affecting plant growth and development 18,24,38,39 . Poor adaptabilities to nutrient deficiency and several abiotic stresses were reported in some Bt cotton lines in previous studies. For example, experiment conducted by Wei et al. revealed that root activity (less root organic acid exudation) in several Bt cotton lines under low N supply condition was lower than that in their non-Bt wild types 40 . And lower biomass was also observed in some Bt cotton lines under low K application condition than their non-Bt counterparts 41 . Besides, some Bt cotton lines were reported to have suffered more from pathogens stress 42 and CeO 2 nanoparticles stress 43 than their non-Bt wild types. Another experiment by Li et al. (2015) indicated that Bt cottons adapted poorly to drought stress or elevated O 3 stress 42 . And inhibited nitrogen metabolism by salinity, waterlogging and the combined stress was observed to result in the decline of Bt protein expression, causing bollworm control reduction 44 .

N treatment
Based on the discussion above, it could be concluded that the phenotypic and yield changes of Bt crops under different conditions were frequently occurring phenomena. However, the mechanism of such changes has not been known well until now. Our study revealed that differences in C and N partitioning could be one of possible reasons accounting for changes in yield and phenotypic traits in T2A-1. Future study could make a further exploration of energy utilization related to C and N metabolism under the framework of the "bio-burden" hypothesis comparing more transgenic plants and their wild types.  www.nature.com/scientificreports www.nature.com/scientificreports/

Materials and Methods
Materials. Bt rice line T2A-1, expressing Cry2A* protein, was used for the assessment of C and N metabolism processes in the plant. And its conventional (non-Bt) parent line MH63, an excellent three-line rice restorer line, served as a control in this study. Both of the two rice lines were provided by National Key Laboratory of Crop Genetic Improvement, Huazhong Agricultural University. During the transformation process, a series of Cry2A* rice lines were obtained from the modification of the Cry2A* gene with no amino acid sequence increased. Besides, the insertion position of Cry2A* in T2A-1 was a noncoding region, which theoretically had few effects on exogenous gene expression. And T2A-1 was found to exhibit the best field resistance and excellent agronomy traits, therefore it was selected for the experiment among 102 independent transformants obtained 6 .

Soils.
Considering its low nitrogen content, loess soil was chosen for this nitrogen concentration gradient experiment at the Agricultural Experiment Station of Huazhong Agriculture University, Wuhan, China. The soil was first air-dried in open air environment at normal temperature, and then ground, finally mixed with sand before use. The soil contained 18.48 g kg −1 of organic matter, 0.11 g kg −1 of total N, and 1.19 g kg −1 of C with a pH being 6.56. Experimental Design. The nitrogen gradient experiment using T2A-1 and MH63 was carried out from May to September 2016 in net house. Four nitrogen fertilizer treatments were performed in this study, namely, supplying 0.2 g (N1 treatment), 0.35 g (N2 treatment), 0.5 g (N3 treatment), and 1 g (N4 treatment) nitrogen, respectively, in the whole growing period of rice. Nitrogen fertilizer in form of urea was applied three times during the whole developing period with applying ratio being 50% as base fertilizer, 30% at mid-tillering stage, and 20% at young panicle differentiation stage, respectively. In addition, potassium and phosphorus fertilizers were applied in a common way. Seedlings were transplanted to PVC pipes (16 cm × 55 cm). One seedling was planted into each pipe, then three pipes were placed into one bucket as a group. And there were three buckets as three replicates for each nitrogen fertilizer treatment using T2A-1 and MH63. These treatments were arranged in a completely random arrangement. One whole plant would be sampled and separated into leaf, stem and spike part for C and N content measurement at each sampling time. And two sampling times were set during reproductive periods including the flowering stage (FS), and 15 days after the flowering, named as FS, and 15 d after FS, respectively.
In addition, to explore the different expression of proteins related to C and N metabolism between MH63 and T2A-1, the other experiment concluding two nitrogen treatments was conducted for proteomics detection at the same time in the same screen house. And 0 g (N0 treatment) and 1 g (RN treatment) nitrogen fertilizer application treatments during the whole rice growth period were chosen to enlarge the difference between the two lines. Flag leaves were collected on day 15 after the flowering for proteomics testing.
Total Carbon and Nitrogen Measurement. Leaf, stem and spike parts were separated from one sampling plant. All samples were oven-dried at 70 °C to constant weight before testing, and the final content of C and N was obtained through the formula that content percentage measured by elemental analyzer (Vario ISOTOPE, Elementar) multiplied dry weight of each sample. Nitrogen accumulation equated to the sum of N content of leaf, stem with sheath, and spike.
Procedure of iTRAQ (isobaric tags for relative and absolute quantifica)Protein Extraction. The samples were frozen in liquid nitrogen and ground with pestle and mortar. Five times volume of TCA/acetone

Protein Quantification
The protein ratios are calculated as the median of only unique peptides of the protein Experimental Bias Normalizes all peptide ratios by the median protein ratio. The median protein ratio should be 1 after the normalization. Table 9. Parameters setting of MASCOT engine.