Characteristics and Fitness Analysis through Interspecific Hybrid Progenies of Transgenic Brassica napus and B. rapa L. ssp.

Interspecific hybridization between transgenic crops and their wild relatives is a major concern for transgene dispersal in the environment. Under controlled conditions, artificial hand pollination experiments were performed in order to assess the hybridization potential and the fitness of interspecific hybrids between Brassica rapa and genetically modified (GM) Brassica napus. Initially, six subspecies of B. rapa were hybridized with GM B. napus through hand pollination. In the resulting F1 hybrids, the combination of B. rapa ssp. narinosa (♀) × GM B. napus (♂) had the highest crossability index (16.9 ± 2.6). However, the F1 selfing progenies of B. rapa ssp. rapa (♀) × GM B. napus were found to be more effective in producing viable future generations with the highest crossability index (1.6 ± 0.69) compared to other subspecies. Consequently, they were used for the generation of F2 and F3 progenies. The 18 different morphological characteristics among the parental cross-combinations and F1 hybrid progenies were measured and visualized through hierarchical clustering. Different generations were found to be grouped based on their different morphological characteristics. The chromosome numbers among the interspecific hybrids ranged from 2n = 29 to 2n = 40. Furthermore, the SSR markers revealed the presence of genomic portions in the hybrids in comparison with their parental lines. There is a high possibility of transgene flow between GM B. napus and B. rapa. The study concluded that the interspecific hybrids between B. napus and B. rapa can be viable and can actively hybridize up to F3 generations and more. This suggests that the GM B. napus can disperse the transgene into B. rapa, and that it can pass through for several generations by hand pollination in a greenhouse environment.


Introduction
The commercialization of genetically modified (GM) crops started in 1996. The global cultivation area of GM crops has increased dramatically in the last 25 years. The production has also increased dramatically in the last 25 years. The production has experienced over a 100-fold increase [1,2]. However, the main problem is potential transgene flow from GM crops that can affect non-transgenic counterparts, such as closely related or sexually compatible species [3]. Thus, the concerns about the gene flow from GM crops to their wild relatives have been intensified in the countries where their commercial cultivation is authorized. The Brassicaceae family is getting special attention because it has wild relatives throughout the world, and it can hybridize with any close and distant relatives within genera and species [4,5]. In the Brassicaceae, there are six Brassica species with three different genomes (A (n = 10); B (n = 8); C (n = 9)), which include three diploid species, namely Brassica rapa (AA, 2n = 20), B. nigra (BB, 2n = 16), and B. oleracea (CC, 2n = 18), through a natural hybridization process which further formed three allotetraploid species, namely B. juncea (AABB, 2n = 36), B. carinata (BBCC, 2n = 34), and B. napus (AACC, 2n = 38) [6]. Among the crops, B. napus (oilseed rape) is one of the most preferred and suitable for gene flow studies, since it can produce a large amount of pollen and has a huge number of related species, including cultivars and wild relatives [7]. Several studies have shown sexually compatible relatives with this crop: B. rapa [8,9], B. juncea [10], B. oleracea [11], Hirschfeldia incana [12,13], Sinapis arvensis [12,14], and Raphanus raphanistrum [13,15] have been reported. Most of the commercial GM B. napus have potential transgenes that are resistant to herbicides such as glyphosate, glufosinate, and bromoxynil [16]. Selective pressure on herbicides promotes the growth of GM B. napus and increases the risk of the escape of herbicide resistance genes through hybridization with related species [17]. The probability of establishing a transgene with another species depends largely on the suitability of the F 1 hybrid between the crop and wild species and subsequent generations. Despite the classical view that wild crop hybrids should be less suitable than their parents, there are instances when wild crop hybrids may be as suitable or better suited as their parents [18][19][20][21][22].
The majority of gene flow studies on GM Brassica sp. have focused on crosses between transgenic B. napus (2n = 38; AACC) and wild relative B. rapa (2n = 20; AA) [5,23,24]. Spontaneous hybridization occurs in Europe and the United States, and their generations can easily backcross to B. rapa in wild environments [3,25,26]. However, limited information is known about the consequences of invasion between B. napus and B. rapa, and gene establishment is not well documented [27]. Cross-compatibility and callose deposition in pollen tubes are the main reasons for hybridization failure in Brassica [28]. However, reports of artificial hand pollination which has resulted in crop and relative hybridization are important sources of knowledge because they enable the evaluation of species' reproductive compatibility and the identification of hostile species combinations. This makes it easier for us to perform a cautious examination of the species that ought to be taken into account for their potential to serve as transgene escape targets in the local environment [5,29,30]. Most of the studies to evaluate the gene transfer from B. napus to B. rapa were conducted in the F 1 and BC 1 generations. Moreover, few studies have been conducted to investigate the fate of transgenes for more than three generations of interspecific hybridization. To assess whether a transgene can increase persistence across all generations through interspecific hybridization, the frequency of hybridization between the two related species, which increased their fitness, survival rates, and fertility, should be considered in subsequent generations [6,29,31].
Since Korea is one of the prime exporters of diversified B. rapa ssp., the possibility of transgene flow and ecological sustainability from GM rapeseed to B. rapa should be investigated. Therefore, in this study, we tried to analyze the possibility of gene transfer between GM rapeseed and various subspecies of B. rapa, and the diversity of subsequent generations and reciprocal combinations of interspecific hybrids. To this end, the main objectives of the study are: (i) the assessment of crossability indices between GM B. napus and six subspecies of B. rapa through artificial hand pollination; (ii) the morphological characteristics revealing their relative fitness characters for transgene persistence in generational progress; (iii) chromosome counts of individuals of F 1 , F 2 , F 3 progenies; and (iv) the inspection of the genetic similarity using SSR markers for F 1 , F 2 , F 3 , and BC 1 progenies.

Cross-Compatibility of Each B. rapa ssp. with Transgenic B. napus
Three crossing experiments were performed based on the flowering times of three different sets of GM B. napus with B. rapa ssp. The artificial hand pollination of six subspecies resulted in an average of 1101 flowers, leading to a 45.3% pod-setting ratio, and an average of seven seeds were obtained from each pod. In parental lines, the maximum crossability index was observed in B. rapa ssp. chinensis (25.1 ± 2.3) (Table 1). Despite this, the maximum crossability index of 16.9 ± 2.6 was observed during initial hybridization in B. rapa ssp.  (Table 1). That produced an average number of seeds in each pod with a 58.8% pod-setting ratio. However, there was no significant difference in producing further generations among the cross-combinations of all the subspecies with GM B. napus. Among them, the F 1 hybrid (selfing) of B. rapa ssp. rapa had the highest crossability index (1.6 ± 0.7) compared to other subspecies (Table 1). Therefore, B. rapa ssp. rapa was taken into F 2 hybrid and F 3 hybrid (selfing) generations, and the resulting crossability indexes were 2 ± 0.7 and 6.4 ± 4.8, respectively (Table 1).
The hybridization of reciprocal combinations resulted in comparatively higher crossability index values, among which, the maximum crossability index was found between GM B. napus (♀) × B. rapa ssp. nipposinica (♂) (27.5 ± 2.9), with a ratio of 17 seeds per pod. Inclusively, the statistical analysis with one-way ANOVA shows that the crossability index was highly significant with respective crossing materials (p < 0.05). A multiple comparison with the Tukey test reveals that differences in the average crossability indexes were largely attributable to parental, cross-combination, and F 1 hybrids ( Table 1). The occurrence of vivipary in parental combinations of B. rapa ssp. (♀) was found to be at higher rates, ranging from 22.7% to 73%. In contrast, reciprocal combinations of GM B. napus (♀) showed less vivipary, and ranged from 0.1% to 4.2% (Table 1 and Supplementary Table S1).

Morphological Characteristics and Relative Fitness of Parental Genotypes and Interspecific Hybrids
Based on 18 morphological characteristics, the parental and all the crossing materials were grouped into two data sets. The B. rapa ssp. and their respective cross-combinations with GM B. napus (parental cross-combinations/PCC) are included in one group, whereas another one with F 1 , F 2 , and F 3 selfing progenies of B. rapa ssp. rapa (♀) × GM B. napus (♂) (interspecific hybrids) is included in another group. The hierarchical clustering of parental cross-combinations ( Figure 1A Table S2).
Regarding the selfing progenies of B. rapa ssp. rapa (♀) × GM B. napus (♂) (KSF 1 to KSF 3 ) interspecific hybrid classification, a total of three clusters have been inferred ( Figure 1B). Cluster 1 (40 progenies) indicated the individuals presenting long and wide flowers, whereas cluster 2 (22 progenies) informed about the plant architecture regarding pod-setting ratio, branches, and plant height. Cluster 3 (11 progenies) grouped individuals showing better reproductive fitness, with higher values of the number of seeds, number of pods, and pod-setting ratio (Supplementary Figure S1 and Table S3). Overall, good reproductive fitness was observed for the cluster 3 of parental cross-combinations (PCC) and the interspecific hybrids ( Figure 1A,B), suggesting a good fitness of generative agricultural characteristics that can help us assess the further generations.    Regarding the selfing progenies of B. rapa ssp. rapa (♀) x GM B. napus (♂) (KSF1 to KSF3) interspecific hybrid classification, a total of three clusters have been inferred ( Figure  1B). Cluster 1 (40 progenies) indicated the individuals presenting long and wide flowers, whereas cluster 2 (22 progenies) informed about the plant architecture regarding podsetting ratio, branches, and plant height. Cluster 3 (11 progenies) grouped individuals showing better reproductive fitness, with higher values of the number of seeds, number of pods, and pod-setting ratio (Supplementary Figure S1 and Table S3). Overall, good reproductive fitness was observed for the cluster 3 of parental cross-combinations (PCC) and the interspecific hybrids ( Figure 1A,B), suggesting a good fitness of generative agricultural characteristics that can help us assess the further generations.

Chromosome Numbers of Interspecific Hybrids and Progenies
The parental genotypes, B. rapa ssp. rapa (KS) KSF1 hybrid and the selfing progenies of KSF2 and KSF3, were found to have variable chromosome numbers in microscopic observation ( Figure 2). The chromosome numbers of parental genotypes, such as B. rapa ssp.

Assessment of Intergenomic Recombination and Their Progenies Validation by Using SSR Markers
To validate the interspecific hybrids and their selfing progenies with the above-mentioned morphological characteristics and chromosome number variations, the 17 SSR markers were used for the genetic analysis of 23 KSF2 and 28 KSF3 selfing progenies and 33 KSBC1 plants. KSF2 and KSF3 extensively revealed a heterozygous nature. As shown in Figure 3A,B, 93.09% and 90.23% of KSF2 and KSF3 were found to be of a heterozygous nature (presence of the marker in both A and C genomes). In the A and C genomes of the KSF2 and KSF3 plants, fewer SSR loci were missed ( Figure 3A,B). Contrarily, in KSBC1 plants, only 53.65% were found to be heterozygous. Due to homeologous recombination,

Assessment of Intergenomic Recombination and Their Progenies Validation by Using SSR Markers
To validate the interspecific hybrids and their selfing progenies with the abovementioned morphological characteristics and chromosome number variations, the 17 SSR markers were used for the genetic analysis of 23 KSF 2 and 28 KSF 3 selfing progenies and  Figure 3A,B, 93.09% and 90.23% of KSF 2 and KSF 3 were found to be of a heterozygous nature (presence of the marker in both A and C genomes). In the A and C genomes of the KSF 2 and KSF 3 plants, fewer SSR loci were missed ( Figure 3A,B). Contrarily, in KSBC 1 plants, only 53.65% were found to be heterozygous. Due to homeologous recombination, the A genome (46%) was found to be in higher frequencies in KSBC 1 than in KSF 2 and KSF 3 hybrid progenies (Supplementary Figure S2).  Cluster analysis using the Jaccard distance matrix was used to evaluate the SSR marker data. The maximum distance was found in backcross generations (0.971), and a Cluster analysis using the Jaccard distance matrix was used to evaluate the SSR marker data. The maximum distance was found in backcross generations (0.971), and a minimum distance (0.029) was recorded on KSBC 1 , KSF 2 , and KSF 3 generations. Using the distance matrix, the UPGMA dendrogram was constructed, which revealed a good degree of fit by the values of the cophenetic correlation coefficient (r = 0.940, p < 0.001) (Supplementary Figure S3). The KSF 2 and KSF 3 hybrids and KSBC 1 progenies were clustered into eight major clusters. This is in accordance with the tree constructed with 18 morphological traits (Supplementary Figure S1). The parental plants of B. rapa ssp. rapa were clustered with KSBC 1 progenies, whereas B. napus and KSF 1 were grouped with KSF 2 progenies. Similarly, the control plants of B. oleracea were clustered separately and out-branched far from all other clusters.

Discussion
Many studies have explored the interspecific hybridization and gene flow between transgenic B. napus and various subspecies and varieties of B. rapa [32][33][34]. In our previous report, the gene flow of an early flowering gene (BrAGL20) was characterized in F 1 hybrids between B. rapa ssp. pekinensis and GM B. napus [29]. Apart from F 1 hybrids, there are no reports on selfing progenies' transgene persistence in subsequent generations (F 2 and F 3 ). It is critical to investigate transgene persistence over multiple generations. Hence, in this study, to reveal the gene flow of the transgene to more generations, interspecific hybridization of six B. rapa ssp. and GM B. napus (as a paternal) was performed through artificial hand pollination. Several subspecies of B. rapa are known for their higher levels of phenotypic and genetic diversity. They can have varying degrees of cross-compatibility and self-incompatibility by nature [5,35,36]. However, we preliminarily investigated the fertilization barriers or self-incompatibilities that occurred during self-pollination in six subspecies. However, through artificial hand pollination, they showed no self-incompatibility with the flower buds. Different levels of crossability have been recorded for each subspecies (Table 1). In cross-combination, the average crossability of B. rapa ssp. with GM B. napus is four seeds per pod, with a range from 2 to 12. However, in reciprocal crosses, 12 seeds per pod were detected, with a range of 4 to 17. Our findings were consistent with earlier research, indicating that seed-setting is more successful when the maternal parent has a greater ploidy level than the paternal parent [21,[37][38][39].
Selfing progenies of the F 1 hybrid (B. rapa ssp. (♀) × GM B. napus (♂)) and successive progenies of B. rapa ssp. rapa (KSF 2 and KSF 3 ) exhibited extremely low crossability index values and, thus, less compatibility. Although there is no experimental evidence to support this, we hypothesized that it was caused by pollen viability, pollen rejection, or pre-zygotic barriers during self-pollination. Thus, it may have an inhibition of pollen hydration and germination or pollen tube growth on the stigma [40][41][42]. Crossability is also influenced by reproductive barriers, which are dependent on parental fertility and pollen-pistil interactions [43,44]. Even if pollen germination and fertilization are successful, precocious or viviparous germination will occur, as previously reported [29,34,45]. Seed development is influenced by aberrant endosperm growth, embryo abortion, cross-species hybridization, parent ploidy levels, and hybridization directions [34,41]. The morphological characteristics and the number of progenies or individuals produced were strongly correlated with fitness [20,[46][47][48]. In cluster analysis, 18 morphological characters were positively correlated with all the B. rapa ssp. and F 1 hybrids, except the subspecies, 'pekinensis' and 'rapa' (Figure 1A). F 1 selfing progenies of B. rapa ssp. rapa (♀) × GM B. napus (♂) had morphological characteristics similar to F 1 hybrids ( Figure 1B). However, they decreased their fitness values in all aspects compared to F 1 hybrids. The transgene may have a direct contribution to their fitness increase/vigor or decrease/depression in the progenies [49,50]. In F 3 , the progenies belong to cluster 3, which is highly correlated with the number of seeds and number of pods ( Figure 1B). Our results indicate that the generative characters in cluster 3 are similarly expressed in F 1 hybrids, and F 1 and F 3 progenies (see Supplementary Tables S2 and S3). This information could be useful in the effective characterization of interspecific hybrid progenies of B. rapa (♀) × GM B. napus (♂) from self-pollination.
As expected, due to genetic imbalance, chromosome numbers may have varied significantly across all of these F 2 and F 3 hybrid selfing progenies. Interspecific hybridization causes chromosomal changes that can lead to transcriptional modifications that might affect the morphological characteristics of the plants [51]. Similarly, homologous recombination and increasing the chromosome numbers lead to reduced fitness [52], affect the seed yield [53], and induce genomic instability, thus reducing the probability of gene flow [54]. An assessment of relative fitness can prove the rational chromosome number variations in the interspecific hybrids. The interspecific hybridization and the chromosomal segregations were confirmed with Brassica A and C genome-specific SSR markers [26,55]. Through homologous recombination in F 1 hybrids, they had a closer genetic similarity, a higher percentage of C genome, and transgene presence in all progenies than the backcross generation. These results concur with previously reported studies [6,[56][57][58]. However, F 2 and F 3 progenies were found to have missed loci in both the A and C genomes. The homologous recombination between the A and C genomes leads to the deletion, rearrangements, and duplications of the chromosome (Zhang et al., 2016). Whereas nearly 46% of C genome loci were lost in KSBC 1 progenies, only three A genome loci were lost. The transgene presence on one of the chromosomes of the C genome is transmitted at a low frequency. This suggests that the transgenes can more safely integrate into the C-chromosome than into the A chromosome [24]. That may be due to the higher level of homologous recombination with the AA-genome-containing maternal parent (B. rapa ssp. rapa). Based on the UPGMA cluster analysis results, KSF 2 and KSF 3 progenies were shown to be genetically distant from the KSBC 1 generation. However, a few KSBC 1 generations were more closely placed with KSF 2 and KSF 3 progenies. Interspecific hybridization experiments were performed by using B. rapa ssp. as a maternal parent (♀) and GM B. napus as a paternal parent (♂). In addition, we also perform hybridization with reciprocal combinations (Supplementary Figure S4). An average of 1328 young flower buds was used for artificial hand pollination in different plants for each crossing experiment. The emasculated B. rapa flower buds were pollinated with pollen from GM B. napus flowers the next day and then immediately covered with sealed, prelabeled bags after pollination. Then, the plants were allowed to grow, and the fructification events of siliques were observed. We measured medium-sized pods (10 no.) for each plant to determine the crossability indexes for all of the cross-combination plants. B. napus and GM B. napus were used as standard controls, and the number of seeds per pod was calculated as a hybridization crossability index between GM B. napus and different B. rapa ssp. The resultant F 1 hybrids were self-pollinated (5 plants), and produced F 2 and F 3 selfing progenies. Furthermore, the F 1 hybrids (pollen donor) were crossed with B. rapa (seed parent), and produced BC 1 progenies. For all of the progeny, the crossability index was calculated as the number of seeds obtained per pod. The survival rate (%) of seedlings after herbicide treatment was used to calculate the herbicide resistance rate. Briefly, seedlings were sprayed with 0.3% Basta (Bayer Crop Science GmbH, Manheim am Rhein, Germany) at the 4-5 leaf stage and again 4 days later, and seedling survival was measured at 4-7 days after the second application (details are in Supplementary Table S4). For the backcross generation detection of bar proteins in transgenic plants, a qualitative detection of bar proteins in the leaves of transgenic plants was conducted using a commercial immunostrip specific to bar proteins (Agrastrip ® seed & leaf TraitCheck LL, Company: Romer Labs) according to the manufacturer's instructions (Supplementary Figure S5). PCR reactions for bar genes were performed according to Sohn et al. [29] (Supplementary Figures S6-S8).

Morphological Characteristics
The morphological characteristics (vegetative and generative) of all the parental lines and F 1 hybrids, followed by the generation of 29 F 1 , 20 F 2 , and 23 F 3 selfing progenies, were investigated (Supplementary Figures S9 and S10). The morphological characters of all the plant components were classified using the multigrade International Union for the Protection of New Varieties of Plants descriptors for Brassica [60]. The vegetative characters are as follows: PH, plant height; BS, branch segment; NOB, no. of branches (1,2,3). The generative characters are: NPF, no. of pollinated flowers; NOP, no. of pods; PSR, podsetting ratio; NOS, no. of seeds; SPP, seeds per pod; VV, vivipary; NFS, non-filled seeds; FL, flower length; FW, flower width; FLD, flower diagonal; FIS, filament short; FIL, filament long; STL, style length (Supplementary Table S1).

Chromosome Numbers
The root tips were collected at 8 a.m. because of the high mitotic activity. Immediately after harvesting, the roots were pre-treated with 8-hydroxyquinoline at room temperature (RT) for 4 h. Following the pre-treatment, the root tips were rinsed with distilled water and treated with a 3:1 (v/v) mixture of ethanol and acetic acid. This was used to fix the pre-treated roots for 24 h at RT. The roots were rinsed again using distilled water and kept in 70% ethanol and stored at −20 • C until the roots were ready to be used. The fixed roots were washed with distilled water and the meristematic portions were cut off. The cells were then immersed in a hydrolyzed enzyme buffer (Cytohelicase 250 mg, Cellulose 250 mg, Pectolyase 250 mg in 25 mL of 0.01 M citrate) for 1 h at 37 • C. After washing the enzyme, the roots were gently tapped or crushed with a pin. Then, a drop of acetic acid (60%) was added to clean and evenly distribute the roots, and they were placed in an oven at 46 • C for 2 min. Finally, the slides were counterstained with Vectashield (H-1000) with DAPI (4,6-diamidino-2-phenylindole, Sigma), and covered with filter paper by applying firm thumb pressure. To avoid autofluorescence, the prepared slides were treated with a drop of immersion oil before being examined under a Nikon Eclipse 50i fluorescence microscope at a magnification of 100×. The method described here is a slight modification to the protocols of Tagashira et al. [61] and Hoshi [62].

SSR Analysis
The SSR markers used in this study were derived from a previous report by Zhang et al., 2016, and the markers which can produce two bands were selected based on a comparison between the A and C genomes in B. napus (Supplementary Table S5). Among them, 17 SSR primers were generated, with clearly distinguishable bands, which were used for further analysis. The genomic DNA was extracted from leaf tissue using the cetyl trimethyl ammonium bromide (CTAB) method [63]. The polymerase chain reaction (PCR) mixture in a 20 µL volume contains forward and reverse primer (1 µL) (10 picomol each), gDNA (1 µL), Taq PCR mix (http://cells-safe.com/, accessed on 14 February 2022), and RNAase-free water (18 µL). The PCR amplification was performed in a thermal cycler (Biometra Thermal cycler) with the following conditions: an initial denaturing step at 95 • C for 3 min; followed by 35 cycles of 95 • C for 30 s, 56 • C for 30 s, 72 • C for 30 min, and 72 • C for 10 min [55]. The amplified PCR products were visualized using a QSEP400 highthroughput gel electrophoresis system (Qsep400 multi-channel Bio-fragment analyzer). The amplifications were scored on the basis of the presence or absence of bands (H: 1,1; A: 1,0; C: 0,1;) and were depicted as binary characters. To find the genetic relationships among the progenies, Jaccard's distance matrix was plotted using DARwin software for Windows version (6.0.021) [64], and clustering was carried out using the unweighted pair group method and arithmetic average (UPGMA). The resulting phylogenetic tree was exported using Evolview [65] for graphical annotation.

Statistical Analysis
The data on morphological characteristics were analyzed using the R program v 4.1.2 (https://cran.r-project.org/bin/windows/base/old/4.1.2/R-4.1.2-win.exe, accessed on 6 September 2022). Using the package, 'agricolae' [66], a one-way analysis of variance (ANOVA), followed by Tukey's mean separation, was carried out with a significance difference at p = 0.05. Principal component analysis, followed by a hierarchical clustering analysis, was performed to assess the relationship among the genotypes based on morphological characteristics using FactoMinerR [67] and Factoextra [68]. Prior to the multivariate analyses, missing data were imputed with the missMDA [69] package.

Conclusions
The GM B. napus can effectively hybridize with different subspecies of B. rapa through artificial hand pollination in a controlled environment. In particular, it can produce several viable and fertile generations (F 1 , F 2 , F 3 , and BC 1 ) with B. rapa ssp. rapa, and can transfer the herbicide-resistant transgene to their progenies. In greenhouse conditions, artificial hand pollination with transgenic B. napus resulted in a 100% outcrossing rate. However, in field conditions, spontaneous hybridization has an outcrossing rate, ranging from 0.02 to 2.78% in field conditions [34,70]. Due to several environmental factors, the outcrossing rate is much lower compared to greenhouse conditions. According to our data, greenhouse containment is the most successful approach for preventing natural gene flow. So far, no examples of greenhouse containment failure have been observed. The few conditions that have a significant impact on the outcrossing rate are unlikely to occur naturally: (i) in nature, there will be fewer flowering possibilities for transgenic B. napus and B. rapa at the same period; (ii) the B. rapa flowering period was controlled using the vernalization process; (iii) the young flower buds are the determining factor for successful cross-pollination/hybridization in other subspecies; and (iv) the pollen of transgenic B. napus was manually transferred by artificial hand pollination to B. rapa ssp., and the plants were maintained at controlled conditions throughout their life cycle. It is necessary to understand the transgene expression characteristics of hybrid progenies to assess the transgene persistence. Further gene flow studies are needed for the enhanced understanding of the process, and to assess its impacts on the environment and ecology.