Characterization of Chromosome Inheritance of the Intergeneric BC2 and BC3 Progeny between Saccharum spp. and Erianthus arundinaceus

Erianthus arundinaceus (E. arundinaceus) has many desirable agronomic traits for sugarcane improvement, such as high biomass, vigor, rationing ability, tolerance to drought, and water logging, as well as resistance to pests and disease. To investigate the introgression of the E. arundinaceus genome into sugarcane in the higher generations, intergeneric BC2 and BC3 progeny generated between Saccharum spp. and E. arundinaceus were studied using the genomic in situ hybridization (GISH) technique. The results showed that the BC2 and BC3 generations resulted from n + n chromosome transmission. Furthermore, chromosome translocation occurred at terminal fragments from the E. arundinaceus chromosome in some progeny of Saccharum spp. and E. arundinaceus. Notably, the translocated chromosomes could be stably transmitted to their progeny. This study illustrates the characterization of chromosome inheritance of the intergeneric BC2 and BC3 progeny between Saccharum spp. and E. arundinaceus. This work could provide more useful molecular cytogenetic information for the germplasm resources of E. arundinaceus, and may promote further understanding of the germplasm resources of E. arundinaceus for sugarcane breeders to accelerate its progress in sugarcane commercial breeding.


Introduction
Sugarcane (Saccharum spp.) is a large perennial grass that is indigenous to tropical and subtropical regions. As the most important sugar-producing crop worldwide, sugarcane has significant potential to contribute to the global sugar security and produces approximately 75% of the world's raw sugar [1]. In addition, as a C 4 plant, sugarcane is an efficient crop in converting solar energy into chemical energy. Therefore, it has been heralded as an alternative source of fuel and petrochemical feedstock for the production of first-generation bioethanol to alleviate the current energy crisis [2].
The genus Saccharum is an important member of the Poaceae family that consists of six species, including S. officinarum, S. spontaneum, S. robustum, S. barberi, S. sinense, and S. edule. Modern sugarcane cultivars are highly complex aneupolyploids, and most are primarily derived from interspecific hybridization between S. officinarum (2n = 80) and S. spontaneum (2n = 40-128) through nobilization [3]. This term was first coined by Dutch breeders Jesweit in Java during the early 1900s to denote the process of introgression of S. spontaneum into S. officinarum following hybridization and successive backcrossing. During the nobilization process, interspecific F 1 hybrids were obtained from crosses between S. officinarum as the female parent and S. spontaneum as the male parent, and then were repeatedly backcrossed to S. officinarum as the female parent. Using this approach, progeny conserve the entire genome of S. officinarum in the first interspecific cross (F 1 ) and the first backcross (BC 1 ) [4]. Hence, the chromosome inheritance of progeny in F 1 and BC 1 exhibits 2n + n transmission. This not only allows for a quick recovery of the high sugar content from S. officinarum, but also integrates resistance genes to biotic and abiotic stresses from S. spontaneum [5]. Jesweit succeeded in the selective breeding of some new cultivars with high resistance to disease, and significantly contributed to the perseverance through the sugar crisis in Java at that time due to disease outbreaks [2]. POJ2878, hailed as the "wonder cane", is one of the most successful examples of the utilization of nobilization.
However, due to the frequent utilization of a limited number of progenitors in sugarcane breeding programs, modern sugarcane cultivars have given rise to a sharp decline in genetic diversity [6,7]. Genetic erosion renders sugarcane increasingly vulnerable to resistance against biotic and abiotic stresses. As a result, the genetic potential for yield and quality improvement has hardly allowed for any advancement in the past several decades. Therefore, options for remedying the growing concern of a dearth of genetic variation has become an urgent and necessary task for sugarcane breeders. One efficient approach for combating this issue is by tapping into wild relatives to introduce favorable genes for increased productivity and better adaptability to a wide large range of growing conditions as well as providing more robust disease resistance. The genus Saccharum together with the four related genera, namely Erianthus, Miscanthus, Narenga, and Sclerostachya, comprise the "Saccharum complex" [8]. These four related genera serve as a rich gene pool for sugarcane improvement with tolerance to abiotic stresses and resistance to biotic stresses. As one of the most important wild relatives of sugarcane, Erianthus arundinaceus (E. arundinaceus) has many superior traits for sugarcane improvement [8][9][10][11][12]. It has already been considered to be one of the most popular germplasm sources for crossing utilization in sugarcane improvement. However, the taxonomy of Saccharum and Erianthus had been controversial for a long time [13]. Until there is a good deal of evidence that the genetic distance is large between Saccharum and Erianthus, according to morphological characteristics, chromosome number, and phylogenetic relationship, the evidence sheds light on the taxonomic relationship between Saccharum and Erianthus [8,[14][15][16][17]. Despite the large genetic distance between Saccharum and Erianthus, genuine intergeneric F 1 hybrids and their derivatives have been successfully generated in the past [9,11,[18][19][20][21].
In sugarcane, there are various types of chromosome transmission such as n + n, 2n + n, n + 2n, and 2n + 2n [19,[21][22][23][24][25][26]. Compared to the common type of chromosome transmission (n + n), the other three specific types of chromosome transmission (2n + n, n + 2n, and 2n + 2n) are derived from unilateral and bilateral sexual polyploidization. In the plant kingdom, sexual polyploidization, leading to unreduced gametes (2n gametes) with the somatic chromosome number rather than the gametophytic number (n gamete), is generally believed to be the predominant mechanism of polyploidization [27,28]. Based on the different gametogenesis, the unreduced gametes are divided into 2n egg gametes and 2n male gametes. 2n egg gametes typically originate from the unreduced ovule during megasporogenesis, whereas 2n male gametes are the result of the unreduced pollen during microsporogenesis [29]. The consequence of unilateral sexual polyploidization is 2n + n (result from the fertilization of unreduced ovule by normal haploid pollen) or n + 2n (result from the fertilization of normal ovule by unreduced pollen), while the result of bilateral sexual polyploidization is 2n + 2n (result from the fertilization of unreduced ovule by unreduced pollen). Several mechanisms have been described that the types of meiotic abnormalities responsible for the production of 2n gametes. Due to the different parental heterozygosity rate that each mechanism transmits to the progeny, the genetic consequences of different types of 2n gametes formation are highly divergent [28]. Hence, the use of 2n gametes, resulting in the different types of chromosome transmission during the establishment of sexual polyploids, is of prime importance to develop and conduct breeding strategies for crop improvement [30]. Indeed, it has already been proven effective for improvement of crops such as lily, potato, banana, and citrus [31][32][33][34][35][36][37][38]. In nobilization of S. spontaneum, the utilization of 2n gametes transmission from S. officinarum in interspecific crosses with S. spontaneum is such a typical example of the cytological peculiarity of 2n gametes.
Genomic in situ hybridization (GISH) is a powerful molecular cytogenetic tool to unravel the chromosome composition for the detection of different chromosomes sets derived from two or more distinct species and even recombinant chromosome segments in allopolyploids. During allopolyploid speciation and evolutionary process, the occurrence of chromosomal rearrangement is common, such as translocation and inversion. So far, this molecular cytogenetic technology has been widely used in investigating the chromosome composition and chromosomal translocation in a wide range of natural allopolyploids or artificial polyploid progeny [39][40][41][42][43][44][45][46]. Thus, knowledge of inferring the chromosome transmission from the chromosome composition in allopolyploid will make it possible to implement a strategy for developing useful varieties through breeding. Using GISH, much insight has been gained into sugarcane chromosomal inheritance and genomic recombination over the past several decades [11,[19][20][21]24]. A previous study indicated that modern sugarcane cultivars possess approximately 120 chromosomes, with 70-80% derived from S. officinarum, 10-20% from S. spontaneum, and a few chromosomes derived from interspecific recombination [24].
In this study, two generations, including nine BC 2 progeny and eight BC 3 progeny, were characterized by GISH. The objectives were as follows: (1) to determine the chromosome transmission in these two generations, which can provide a reference for breeding strategies for further deployment of genes and traits from E. arundinaceus; and (2) to determine the presence of various types of intergeneric chromosomal translocation and obtain information on whether they can be inherited, which can provide a basic understanding for efficient utilization in sugarcane breeding.

Plant materials
The plant materials used in this study consisted of 17 progeny derived from two generations (BC 2 and BC 3 ) of intergeneric hybrids between Saccharum spp. and E. arundinaceus (Table 1). In the F 1 generation, F 1 hybrids between S. officinarum and E. arundinaceus were derived from crosses between Badila (S. officinarum, 2n = 80) as the female parent and HN 92-77 (E. arundinaceus, 2n = 60) or HN 92-105 (E. arundinaceus, 2n = 60) as the male parent. In the BC 1 generation, F 1 hybrids between S. officinarum and E. arundinaceus were used as the female parent. CP 84-1198 (2n = 120), a commercial cultivar containing germplasms from S. officinarum, S. spontaneum, S. barberi and S. robustum without contribution from E. arundinaceus, was used as the male parent [21]. In the BC 2 and BC 3 generation, all the female parents and the male parents are listed in detail in Table 1. Among them, ROC10, ROC20, ROC23, YT73-204, YT91-976, YT93-159, NJ57-416, and YC95-46 are the commercial cultivars containing germplasm from Saccharum spp. without contribution from E. arundinaceus, while "YCE" series are the progeny of E. arundinaceus. The progeny analyzed in this study were generated at the Hainan Sugarcane Breeding Station of Guangzhou Sugarcane Industry Research Institute. All plant materials were grown in the greenhouse at Fujian Agriculture and Forestry University.
Genomic in situ hybridization (GISH) procedure Chromosome preparation and the GISH experiment were carried out according to the method described by D'hont et al. [9]. Genomic DNA from Badila (S. officinarum) and YN82-114 (S. spontaneum) was labelled with Biotin, the Biotin-labeled probe was detected with Avidin D, Rhodamine 600 (XRITC) and a Biotinylated anti-avidin antibody (Vector Laboratories, Burlingame, CA), respectively. Genomic DNA from HN92-77 or HN92-105 (E. arundinaceus) was labelled with Digoxigenin, and the Digoxigenin-labeled probe was detected with sheep-anti-Digoxin-FITC (Roche, Lewes, UK) and rabbit-anti-sheep-FITC secondary antibody (Roche, Lewes, UK). Chromosomes were then counterstained with 4 0 , 6-diamidino-2-phenylindole (DAPI) in a Vectashield anti-fade solution (Vector Laboratories, Burlingame, CA). FISH signals were captured using an AxioScope A1 Imager fluorescent microscope (Carl Zeiss, Gottingen, Germany). In this study, results are presented as the modal number and the range of chromosomes counting four to 22 metaphases for each progeny ( Table 2). The images were processed using an AxioCam MRc5 and AxioVision v.4.7 imaging software (Carl Zeiss, Gottingen, Germany).    [19,20] reported that the similar transmission was in some different intergeneric BC 2 and BC 3 progeny between Saccharum spp. and E. arundinaceus.
In this study, YCE05-150 is a sibling line of YCE06-140 with five E. arundinaceus chromosomes. However, the number of E. arundinaceus chromosomes in YCE05-150 is eight. That is, more than half of the E. arundinaceus chromosomes in YCE03-218 was transmitted to YCE05-150. Given the results of chromosome count, the number of total chromosomes in YCE05-150, YCE03-218 (as the female parent) and ROC10 (as the male parent) were 116, 107 and 112, respectively. We can exclude the possibility that YCE05-150 was the product of 2n + n or n + 2n transmission. In fact, YCE05-150 was also the product of n + n transmission.
In nobilization of S. spontaneum, the interspecific F 1 hybrids and BC 1 progeny result from 2n + n transmission. The speedy process of nobilization is conducted to recover high biomass yield and sugar content from S. officinarum while retaining the stress tolerance characteristics from S. spontaneum. As a general rule, the improved varieties are obtained in the BC 2 or BC 3 . However, the increasing times of nobilization may negatively impact the recovery of vigor and resistance to biotic or abiotic stresses. Molecular cytogenetic studies of E. arundinaceus indicated that chromosome transmission was n + n in F 1 , BC 2 , and BC 3 generations, but was 2n + n in the BC 1 generation [19][20][21]. Compared to the progress of nobilization for utilization of S. spontaneum, this slows down the progress of nobilization in utilization of E. arundinaceus and may require the improved varieties from the BC 3 , BC 4 , or even higher generations. Interestingly, Wu et al. [21] recently reported that an unexpected inheritance pattern of E. arundinaceus chromosomes resulted from more than a 2n + n transmission in the BC 1 generation. This may lead to the presence of a larger number of new, multilocus allelic combinations and potentially creates a massive opportunity for selection of desirable traits in newly synthesized germplasm.
Smut caused by Sporisorium scitamineum is a destructive and worldwide disease of sugarcane, resulting in severe yield reductions and considerable loss in sugar content. Wild relatives represent potentially important sources of desirable genes for sugarcane improvement. As one of the most important wild relatives of sugarcane, E. arundinaceus has the potential to improve  Fig 1D and Fig 1G show the translocated disease resistance in sugarcane. Introgressing from wild relatives into sugarcane is an effective approach to broadening the genetic basis of sugarcane germplasm. More strikingly, recent reports have proven that YCE05-179, a BC 2 progeny, is resistant to sugarcane smut [47,48]. The rest of the other yet-to-be-identified progeny between Saccharum spp. and E. arundinaceus might provide superior lines resistant to abiotic and biotic stresses. It is necessary to screen the progeny between Saccharum spp. and E. arundinaceus for those most likely to improve commercial and agricultural traits.
Chromosomal translocation in the intergeneric BC 2 and BC 3 progeny between Saccharum spp. and E. arundinaceus The current results obtained in this study together with those present in a previous study indicate that eight progeny harbored an intergeneric chromosomal translocation between Saccharum spp. and E. arundinaceus (Table 2, Fig 1D, 1G; Figs D, G in S1 File; and Fig 2A, 2E, 2F, 2G; Figs A, E, F, G in S2 File). In our previous study, out of 13 BC 1 progeny analyzed, two BC 1 progeny (YCE01-36 and YCE01-92) both carried an intergeneric translocated chromosome, and the chromosomal translocation occurred at a terminal fragment from the E. arundinaceus chromosome [21]. In this present study, out of nine BC 2 progeny analyzed, two BC 2 progeny (YCE03-168 and YCE03-378) both contained one translocated chromosome, and the chromosomal translocation occurred at a terminal fragment from the E. arundinaceus chromosome ( Table 2, Fig 1D and 1G; Figs D and G in S1 File). Moreover, out of eight BC 3 progeny analyzed, there were four BC 3 progeny (YCE05-64, YCE06-92, YCE06-111, and YCE06-140) with chromosome translocations. Piperidis et al. [20] also reported the presence of intergeneric chromosomal translocations between Saccharum spp. and E. arundinaceus in BC 3 . In our study, YCE06-111 and YCE06-140 possessed an intergeneric translocated chromosome, and YCE06-92 and YCE05-64 carried two and five translocated chromosomes, respectively. The chromosomal translocation occurred at the terminal fragment from E. arundinaceus chromosomes in all these cases (Table 2, Fig 2A, Fig 2E, Fig 2F and Fig 2G show  Translocated chromosomes are stably transmitted to the progeny Based on the pedigree, YCE01-92 (BC 1 ) is the female parent for YCE03-378 (BC 2 ), and YCE03-168 (BC 2 ) is the male parent for YCE06-111 (BC 3 ), respectively. In YCE01-92 there is one recombinant chromosome which is transmitted to YCE03-378. Similarly, in YCE03-168 there is also one recombinant chromosome which is transmitted to YCE06-111. That is, according to the transgenerational inheritance of the translocated chromosome, we can conclude that these translocated chromosomes could be stably transmitted to the progeny in subsequent generations. We recently reported that chromosome translocations only occur in the terminal regions and not the centromeric regions. This finding demonstrated that the terminal regions of the E. arundinaceus and/or Saccharum spp. chromosomes are more actively involved in translocations than the centromeric regions. It is possible that this is because intercalary translocations require more chromosome breakage events than terminal translocations, and therefore rarely occur. In addition, recombination events occur in different generations and in different progeny, suggesting that translocation events are not restricted to an individual progeny.

Significance of intergeneric chromosome translocation for sugarcane improvement
Previous molecular cytogenetic studies have suggested that a few chromosomes were derived from interspecific recombination between S. officinarum and S. spontaneum in modern sugarcane cultivars [24]. In our study, despite a large genetic distance between Saccharum spp. and E. arundinaceus [15,16,49], the occurrence of intergeneric chromosomal translocations occurred within the BC 1 , BC 2 , and BC 3 generations. These results suggest that intergeneric chromosome translocations can occur during an early generation. A considerable number of the recombinant chromosomes confirmed that homologous recombination occurs in the resulting progeny from Saccharum spp. and E. arundinaceus. Undoubtedly, the intergeneric chromosome translocations between Saccharum spp. chromosomes and E. arundinaceus chromosomes in this study represent a new genetic variation between these two genomes. From the point-of-view of sugarcane, intergeneric chromosome translocations can import E. arundinaceus chromosome segments or useful genes of E. arundinaceus into sugarcane. A translocation event can lead to a break in the genetic linkage, which increases the opportunity to segregate genetic variation and opens up the possibility for generating genetic and phenotypic novelty. Importantly, this kind of genetic variation may have a positive impact on sugarcane improvement.