Wild and Cultivated Homoeologous Barley Chromosomes Can Associate and Recombine in Wheat in the Absence of the Ph1 Locus

: Bread wheat is an allohexaploid that behaves as a diploid during meiosis, the cell division process to produce the gametes occurring in organisms with sexual reproduction. Knowledge of the mechanisms implicated in meiosis can contribute to facilitating the transfer of desirable traits from related species into a crop like wheat in the framework of breeding. It is particularly interest-ing to shed light on the mechanisms controlling correct pairing between homologous (equivalent) chromosomes and recombination, even more in polyploid species. The Ph1 ( Pairing homoeologous 1 ) locus is implicated in recombination. In this work, we aimed to study whether homoeologous (equivalent chromosomes from different genomes) Hordeum chilense (wild barley) and H. vulgare (cultivated barley) chromosomes can associate and recombine during meiosis in the wheat background in the absence of the Ph1 locus. For this, we have developed H. chilense and H. vulgare double monosomic addition lines for the same and for different homoeology group in wheat in the ph1b mutant background. Using genomic in situ hybridization, we visualized the two (wild and cultivated) barley chromosomes during meiosis and we studied the processes of recognition, association, and recombination between homoeologous chromosomes in the absence of the Ph1 locus. Our results showed that the Ph1 locus does not prevent homoeologous chromosome pairing but it can regulate recombination.


Introduction
Wheat is one of the oldest crops in the world. Its cultivation dates back 10,000 years ago during the Neolithic Revolution and it is considered one of the main reasons for which humans transformed from hunter-gatherer nomad to settled farmer [1]. Presently, wheat is the third most produced cereal in the world (734 million tons), only behind maize (1.1 million tons) and rice (782 million tons) (data taken from 2018; http://faostat3.fao.org/). Nevertheless, this production needs to be increased on account of the growing human population, which means that wheat crops need to be more productive and better adapted to specific agro-climatic conditions [2]. In this framework, plant breeders play an essential role to increase wheat quality and productivity by trying to understand gene functions and interactions, as well as exploiting the existing genetic variability in the development of new varieties better adapted to the environment. In fact, wide crossing is an important tool in plant breeding and sometimes the results are the starting point for new crops. Examples of new plant species developed in the Triticeae tribe are ×Triticosecale, obtained after crossing wheat and rye, or ×Tritordeum, an amphyploid between the wild barley H. chilense Roem. et Schult. and wheat [3,4]. Hence, in these cases, related species have been used by breeders as genetic donors for widening the genetic basis of wheat to get, for example, wheat cultivars better adapted to specific agro-climatic conditions, higher grain quality, or carrying resistance to diseases [5][6][7][8]. The use of relative species to transfer valuable genes into crops to improve grain nutritional quality and increase the total yield potential is based on interspecific hybridization of many cultivars with their wild relatives [9]. In the case of wheat, hybridization between wheat chromosomes and those from the wild related species produces only a low level of chromosome pairing and recombination, increasing the difficulty of introgressing desirable traits from the related species into it. Thus, going deeper into the knowledge of wheat genetics and its genome organization seems to be essential in the framework of plant breeding.
Bread wheat is an allopolyploid with three subgenomes (A, B, and D) which contain seven pairs of homologous (equivalent) chromosomes each one. Equivalent chromosomes from different subgenomes are named homoeologous chromosomes, which complicates the process of recognition and pairing during meiosis. Despite this genome complexity, hexaploid wheat behaves as a diploid during meiosis. This means that chromosomes associate regularly in pairs of homologues (equivalent chromosomes from the same subgenome), being homoeologous pairing prevented through the action of Ph (Pairing homoeologous) genes [10]. Among Ph genes, the Ph1 locus, located on the short arm of chromosome 5B, plays a major effect on chromosome associations during meiosis [11]. Other loci such as the Ph2 and the Ph3, located on the short arm of chromosomes 3D and 3A, respectively, have a weaker effect on meiosis than the Ph1 locus [12][13][14]. Due to its implication in meiosis, the Ph1 locus has been studied for decades and was initially described in charge of homologous chromosome pairing in bread wheat [11,[15][16][17], although the molecular mechanisms behind these observations are still unclear. More recently, other functions of the Ph1 locus during meiosis have appeared. The Ph1 locus also suppresses recombination between homoeologous chromosomes [18][19][20] although it does not prevent chromosome associations between homoeologues [21]. It is clear that the precise control of chromosome pairing is crucial for conferring meiotic regularity, and hence reproductive stability in allopolyploids [22].
Several studies have explored chromatin arrangements in interphase nuclei, chromosome interactions, and movements during meiotic prophase I, and mechanisms that ensure correct segregation of chromosomes during anaphase. These studies have shed some light on chromosome dynamics in Arabidopsis thaliana as well as in plants with large and complex genomes such as wheat and maize [23]. In wheat, at the onset of meiosis, homologous chromosomes are sorted in pairs and are intimately associated along their lengths while a proteinaceous structure, the synaptonemal complex (SC), is assembled between them. Despite all the existing information regarding these processes, how homologues find each other to associate correctly in pairs is one of the least understood meiotic events [24][25][26]. In wheat, when a chromosome recognizes its homologue and not another chromosome, a localized conformational change in adjacent chromatin is triggered in both partners, facilitating recognition and pairing of homologous versus homoeologous chromosomes, this process also being under the effect of the Ph1 locus [14,27]. Thus, the Ph1 locus has a big influence on fertility but in contrast, has a negative effect for plant breeding purposes since it avoids pairing and recombination between wheat chromosomes and those from related species. Therefore, the study of chromosome associations during meiosis in wheat is necessary and can contribute to developing valuable tools in the framework of breeding.
The development of alien chromosome additions has an important utility not only for breeding but also for plant genetic studies. The availability of chromosome introgressions into a crop like wheat make these lines powerful tools for fundamental research, contributing, for example, to elucidating the processes of homoeologous recombination, distribution of chromosome-specific markers, and repetitive DNA sequences, and regulation of heterologous gene expression [28]. Sets of both cultivated (H. vulgare) and wild (H. chilense) barley addition lines in a hexaploid wheat background were developed several decades ago [29][30][31] and still currently have an enormous potential in plant meiosis studies. These lines allow the targeting of one specific pair of chromosomes or just a chromosome section within the wheat background using genomic in situ hybridization (GISH), and as a result, rearrangements and associations in a pair of homologues can be studied individually [32].
In this study, we developed double monosomic addition lines of cultivated and wild barley, H. vulgare and H. chilense, respectively, in the wheat ph1b mutant background for the same and for different homoeology group with the aim of going deeper into the knowledge of chromosome associations during meiosis. These double monosomic addition lines allow the visualization of chromosomes from two different barley species in wheat in the ph1b mutant background and analyze whether the Ph1 locus might play a role in chromosome recognition/pairing at early meiosis in the absence of homologues. Results showed that the Ph1 locus does not impede homoeologous recognition and pairing between H. chilense and H. vulgare chromosomes during meiosis but recombination, which only occurs in the absence of the Ph1 locus.

Plant Material
In this work we used H. vulgare and H. chilense monosomic addition lines in bread wheat (Triticum aestivum cv. Chinese Spring) in the ph1b mutant background carrying one H. vulgare or one H. chilense chromosome (AABBDD ph1bph1b + H v and AABBDD ph1bph1b + H ch , respectively), which were developed previously in our lab [33]. These lines were allowed to self-pollinate to obtain H. chilense and H. vulgare disomic addition lines in the wheat ph1b mutant background to be used as parental lines with the aim of developing H. chilense and H. vulgare double monosomic addition lines in the wheat ph1b mutant background. Hordeum chilense and H. vulgare double monosomic addition lines in the presence of the Ph1 locus were also used [21].
Seeds were germinated on wet filter paper in the dark for 5 days at 4 • C, followed by a period of 24 h at 25 • C. emerging seedling roots were cut, incubated during 4 h in 0.05% w/v colchicine at 25 • C, fixed in Carnoy's solution (three parts 100% ethanol plus, one part glacial acetic acid) and stored at 4 • C until their use. Plants were then grown in the greenhouse at 26 • C during the day and 18 • C during the night (16 h photoperiod). Immature spikes were collected and were also preserved in Carnoy's solution and used to characterise chromosome pairing at meiosis I.

DNA Characterization
Genomic DNA was extracted from frozen seedling leaves following the protocol described by [34] with some modifications [35]. The absence of the Ph1 locus in the lines used in this work was verified using the ABC920 SCAR marker [36]. The presence of each Hordeum sp. chromosome was confirmed by both PCR [37,38] and in italics hybridization.

Fluorescence In Situ Hybridization
Associations between H. vulgare and H. chilense chromosomes in wheat in the ph1b mutant background were studied using fluorescence genomic in situ hybridization (GISH) [39]. Root tips and spikes in meiosis were collected and pre-treated as described previously [21]. Chromosome spreads were obtained from both root tips and anthers in meiosis as described [21]. Genomic DNA extracted from H. vulgare and H. chilense seedling leaves was used as probes for GISH experiments. Hordeum vulgare and H. chilense genomic DNA were labeled by nick translation with biotin-11 (Boehringer Mannheim Biochemicals, Germany) and digoxigenin-11-dUTP (Roche Applied Science, Indianapolis, IN, USA), respectively. The in situ hybridization protocol was performed according to [21].

Fluorescence Microscopy and Image Processing
Hybridization signals were visualized using a Nikon Eclipse 80i epifluorescence microscope. Images were captured with a Nikon CCD camera using the Nikon 3.0 software (Nikon Instruments Europe BV, Amstelveen, The Netherlands) and processed with Photoshop 11.0.2 software (Adobe Systems Inc., San Jose, CA, USA).

Development of Double Monosomic H. vulgare-H. chilense Addition Lines in Wheat in the ph1b Mutant Background
With the aim of obtaining double monosomic H. chilense-H. vulgare addition lines in wheat in the ph1b mutant background, we cultivated 50 monosomic H. chilense and H. vulgare addition lines for chromosomes 5H ch , 7H ch , and 7H v , which were previously developed in our lab in the absence of the Ph1 locus [35]. These monosomic addition lines were self-pollinated in order to identify disomic addition lines for each Hordeum chromosome in the segregation population in the wheat ph1b mutant background. F2 seeds were germinated and plants were screened by PCR using molecular markers to confirm the presence of the H. chilense or H. vulgare chromosomes and the ph1b deletion. Those plants carrying either the H. chilense or H. vulgare chromosomes in the absence of the Ph1 locus were selected and analyzed by GISH to identify disomic addition lines for each 5H ch , 7H ch , and 7H v chromosomes. GISH analysis was required to distinguish among the positive plants for a specific chromosome, those carrying two copies of the desirable H. chilense or H. vulgare chromosome. Only 8, 5, and 5 wheat plants were identified carrying two copies of each 5H ch , 7H ch , and 7H v , respectively, in the ph1b mutant background (Table 1). After in situ hybridization experiments, these 5H ch , 7H ch , and 7H v disomic addition lines in the ph1 mutant background were cultivated until grain was obtained. Then, seeds were germinated, and plants were cultivated and used as parental lines for genetic crosses to generate double monosomic H. chilense-H. vulgare addition lines in wheat in the ph1b mutant background. A summary of the process and the number of plants obtained is shown in detail in Figure 1 and Table 1.   Only 20 grains were obtained from the genetic crosses between the 7H v and the 7H ch disomic addition lines in the absence of the Ph1 locus. Similarly, only 26 grains were obtained from the genetic crosses between the 7H v and the 5H ch disomic addition lines in wheat in the ph1b mutant background (Table 1). All these grains were germinated, and plants were screened by PCR using molecular markers to confirm simultaneously the presence of one H. chilense chromosome, one H. vulgare chromosome, and the ph1b deletion. Unfortunately, it was not possible to detect any H. chilense and H. vulgare double monosomic addition lines carrying the whole chromosome 7H v . Chromosome 7H v was always identified as a telosomic 7H v L chromosome. Thus, we only obtained H. chilense and H. vulgare double monosomic addition lines carrying the whole 5H ch or 7H ch H. chilense chromosome and the telosomic 7H v L chromosome. This means that chromosome associations between H. chilense and H. vulgare chromosomes were only possible to study in the long chromosome arm of both Hordeum chromosomes in the absence of the Ph1 locus. These double monosomic wild and barley addition lines in wheat in the ph1b mutant background were selected to study the effect of the Ph1 locus on chromosome associations. Thus, plant lines were cultivated until meiosis to analyze H. chilense and H. vulgare chromosome associations for both the same homoeology group (7H ch 7H v L) and for different homoeology groups (5H ch 7H v L) in the wheat ph1b mutant background. Only the first spike was taken from each plant for meiosis scoring, leaving the rest of the spikes to develop until maturity to obtain F2 seeds. Spikes entering meiosis were isolated and chromosome spreads from anthers were used for GISH experiments to study chromosome associations between homoeologous chromosomes in the absence of the Ph1 locus.

Non-Homologous Chromosome Associations between H. vulgare and H. chilense Are Allowed during Early Meiosis in Wheat in the ph1b Mutant Background
Chromosome associations between H. vulgare and H. chilense chromosomes for the same (7H v L and 7H ch ) and for different (7H v L and 5H ch ) homoeology groups were studied in pachytene in double H. chilense-H. vulgare monosomic addition lines in wheat in the absence of the Ph1 locus. In addition, equivalent H. chilense and H. vulgare double monosomic addition lines, for the same and for different homoeology groups in the presence of the Ph1 locus, were also grown. Hordeum chilense and H. vulgare chromosomes were visualized during meiosis by GISH to accurately assess the effect of the Ph1 locus on chromosome associations. It is worth saying that H. vulgare chromosome 7H v was complete in these wheat lines carrying the Ph1 locus.
In situ hybridization experiments in meiosis were developed in more than 250 pollen mother cells (PMCs) of each genomic combination (7H v L7H ch and 7H v L5H ch ) in the ph1b mutant and results were visualized in prophase I of meiosis ( Figure 2). GISH experiments showed that H. vulgare and H. chilense homoeologous chromosomes (7H v L7H ch ) were always fully-associated in pairs by the long chromosome arms during pachytene in both wheat backgrounds, in the presence and in the absence of the Ph1 locus (Figure 2a,c). In contrast, non-homoeologous Hordeum chromosomes 7H v L and 5H ch were not observed associated at any time in the presence of the Ph1 locus (Figure 2b), confirming previous results from our group [21]. However, we visualized some minor chromosome associations between 7H v L and 5H ch non-homoeologous chromosomes in 10.89% of the PMCs during pachytene in the ph1b mutant background (Figure 2d), involving a small chromosome portion of both non-homoeologues. Our observations showed that homoeologous barley chromosomes could associate both in the presence and in the absence of the Ph1 locus. Chromosome pairing between non-homoeologous chromosomes from two different Hordeum species can also be allowed in the absence of the Ph1 locus. Thus, homoeologous barley chromosomes can recognize each other by equivalent chromosome regions and trigger a chromatin conformational change independently of the Ph1 locus (Figure 2a,c).

Recombination Can Occur Indistinctly between Wild Barley, Cultivated Barley, and W Chromosomes in the Absence of the Ph1 Locus
Taking into account that homoeologous and non-homoeologous chromosome ciations between H chilense and H. vulgare chromosomes did occur in wheat, we fo on studying whether or not these chromosome associations could allow overcrossin recombination events in the absence of the Ph1 locus. Thus, we analyzed chromo associations in pollen mother cells (PMCs) in metaphase I from H. vulgare and H. ch double monosomic addition lines for the same and different homoeologous gro wheat in the ph1b mutant background. 7H v 7H ch and 7H v 5H ch double monosomic add in wheat in the presence of the Ph1 locus were used as control. Cultivated and wild b chromosomes remained always as univalent in GISH experiments in metaphase I presence of the Ph1 locus, in both situations (same and different homoeologous g Figure 3a,b), as was previously described [21]. In contrast, several types of associ between H. vulgare and H. chilense chromosomes were detected in the absence of th locus for both the same and different homoeologous groups (Figure 3c (Figure 3k,l). The most frequent association in both wheat lines ing homoeologous (7H v L7H ch ) and non-homoeologous chromosomes (7H v L5H c curred between H. vulgare and H. chilense chromosomes (20.85% and 7.26%, respect Table 2). Our results showed that recombination was, as expected, more frequent bet

Recombination Can Occur Indistinctly between Wild Barley, Cultivated Barley, and Wheat Chromosomes in the Absence of the Ph1 Locus
Taking into account that homoeologous and non-homoeologous chromosome associations between H chilense and H. vulgare chromosomes did occur in wheat, we focused on studying whether or not these chromosome associations could allow overcrossing and recombination events in the absence of the Ph1 locus. Thus, we analyzed chromosome associations in pollen mother cells (PMCs) in metaphase I from H. vulgare and H. chilense double monosomic addition lines for the same and different homoeologous group in wheat in the ph1b mutant background. 7H v 7H ch and 7H v 5H ch double monosomic additions in wheat in the presence of the Ph1 locus were used as control. Cultivated and wild barley chromosomes remained always as univalent in GISH experiments in metaphase I in the presence of the Ph1 locus, in both situations (same and different homoeologous group, Figure 3a,b), as was previously described [21]. In contrast, several types of associations between H. vulgare and H. chilense chromosomes were detected in the absence of the Ph1 locus for both the same and different homoeologous groups (Figure 3c-l). Associations in metaphase I between H. vulgare and H. chilense, H. chilense and wheat, and H. vulgare and wheat chromosomes were observed (Figure 3c-h) in both 7H v L7H ch and 7H v L5H ch double monosomic addition lines in the ph1b mutant background, as well as double H. chilense-wheat and H. vulgare-wheat associations (Figure 3i,j) and tripartite H. vulgare-H. chilense-wheat associations (Figure 3k,l). The most frequent association in both wheat lines carrying homoeologous (7H v L7H ch ) and non-homoeologous chromosomes (7H v L5H ch ) occurred between H. vulgare and H. chilense chromosomes (20.85% and 7.26%, respectively, Table 2). Our results showed that recombination was, as expected, more frequent between homoeologous chromosomes than non-homoeologous chromosomes, supporting chromosome associations observed during pachytene between 7H v L and 7H ch chromosomes, which were also more frequent than associations occurring between 7H v L and 5H ch chromosomes. The H. vulgare-H. chilense-wheat tripartite chromosomes association was the least frequent in both lines carrying 7H v 7LH ch and 7H v L5H ch chromosomes in the ph1b background (0.47% and 0.33%, respectively) ( Table 2). These observations suggested that, although homoeologous chromosomes 7H v L and 7H ch can associate in pairs during pachytene both in the presence and in the absence of the Ph1 locus, recombination can only occur in the absence of the Ph1 locus. In addition, although chromosome associations between non-homoeologues can occur in equivalent chromosome regions, recombination events were rare between non-homoeologues. Moreover, chromosome associations and recombination events between wild or cultivated barley chromosomes and wheat were less frequent, suggesting that chromosome associations might depend on genome homology, as H. chilense and H. vulgare species are phylogenetically closer between them than to wheat. The absence of homologues in the case of Hordeum chromosomes might also contribute to increase chromosome associations between both wild and cultivated barley chromosomes, which did not occur in the case of wheat chromosomes where the two homologues for all wheat chromosomes were always present. In summary, our observations suggested that the Ph1 locus cannot suppress homoeologous chromosome associations in wheat but recombination, because no recombination events were detected between homoeologous H. chilense and H. vulgare chromosomes in any case in the presence of the Ph1 locus.

Segregation of the Inter-Specific Chromosome Associations Occurred Randomly in the Absence of the Ph1 Locus
Chromosome reorganizations in PMCs as the result of inter-specific chromosome associations were studied in later stages of meiosis in wheat only in the ph1b mutant background, since no recombination events were detected in the presence of the Ph1 locus. Almost 300 PMCs were screened in total in meiosis anaphase I using GISH. Several possible situations of inter-specific chromosome associations between H. chilense, H. vulgare, and wheat were identified in both lines carrying 7H v L7H ch and 7H v L5H ch chromosomes in wheat in the ph1b mutant background (Figure 4). Hordeum chilense and H. vulgare chromosomes were usually observed to be delayed in the metaphase plate during anaphase I/telophase I, independently of being associated or unpaired (Figure 4). Hordeum chilense and H. vulgare chromosomes remained associated during anaphase I in the 7H v L7H ch double monosomic addition line in the ph1b mutant background, but other chromosome associations between wheat and H. chilense or H. vulgare chromosomes were also visualized in anaphase I/telophase I, corresponding to previous chromosome associations observed in metaphase I (Figure 4a-h). When chromosomes 7H v L and 7H ch did not associate in metaphase I, both univalents segregated randomly, both in one cell, each one in each daughter cell or suffered chromosome misdivision (Figure 4i-k).  Similarly, we visualized in anaphase I/telophase I interspecific chromosome associations between H. chilense and H. vulgare in the non-homoeologous 7H v L5H ch double monosomic addition line in the ph1b mutant background. We scored 150 PMCs by GISH and we observed chromosome associations between H. chilense and H. vulgare chromosomes in few cells corresponding with the low frequency of pairing previously visualized in metaphase I (Figure 4h), but we did not detect other chromosome associations such as H. vulgare-wheat, double H. chilense-wheat and H. vulgare-wheat, or the tripartite H. chilense-H. vulgare-wheat. As expected, non-homoeologous chromosome associations between H. chilense and H. vulgare were less frequent than H. chilense and H. vulgare associations for the same homoeologous group, corresponding to previous observations in pachytene and metaphase I (Table 2). Hordeum chilense and H. vulgare chromosomes segregated randomly when they did not associate previously in 7H v L5H ch double monosomic addition lines in the ph1b mutant background (Figure 4l). Inter-specific chromosome associations were also visualized in tetrads, where several possibilities of H. chilense and H. vulgare segregation were detected, including the non-integration of the Hordeum chromosome, especially in the case of the H. vulgare 7H v L chromosome ( Figure 5).   Finally, it is worth mentioning that other chromosome rearrangements can occur during meiosis in the absence of the Ph1 locus. We detected one mosaic plant carrying the 7H v L5H ch chromosomes in the ph1b mutant background. In situ hybridization in meiosis chromosome spreads of this plant revealed that both H. chilense and H. vulgare chromosomes were present only in some of the cell surrounding the meoicytes (Figure 6a,b). In fact, most of these cells and the meiocytes only carried the H. chilense chromosome, and just a few cells accompanying the meiocytes retained the H. vulgare chromosome. This observation suggested that a misdivision event occurred in an early mitosis event before meiosis, resulting in the loss of the H. vulgare chromosome in one cell, which later generated the meiocytes. This case of aneusomaty was only detected in anthers, as in somatic root cells, both H. chilense and H. vulgare chromosome were present in all cells (Figure 6c). Whether or not the Ph1 locus might be involved in this event remains to be elucidated.

Discussion
Homologous chromosomes need to recognize each other to correctly associate, recombine and segregate. Plant breeders have always wished to transfer genetic variability from related species into a crop like wheat using genetic crosses, expecting recombination between chromosomes from the donor species and those from wheat. However, wheat genome has an enormous stability and recombination between non-homologous chromosomes is hampered. The genetic control of pairing and the physical divergence of the homoeologous genomes are considered the genetic systems responsible for the stable cytological diploid behavior of polyploid wheat [40,41]. The Ph1 locus has been described controlling correct pairing and suppressing homoeologous recombination in wheat [19,24,42]. In this work, to go deeper into the knowledge of the Ph1 locus function, we developed H. chilense and H. vulgare double monosomic addition lines in the wheat ph1b mutant background for the same and for different homoeology groups. For this, we used H. chilense and H. vulgare monosomic addition lines for chromosomes 7H v , 5H ch , and 7H ch in the wheat ph1b mutant background developed previously [35] to be used as parental lines in genetic crosses. The wheat lines developed here in the absence of the Ph1 locus are equivalent to double monosomic addition lines for the same Hordeum chromosomes previously developed in our laboratory in the presence of the Ph1 locus [21]. However, we found difficulties in obtaining wheat lines carrying both Hordeum chromosomes intact at the same time in the ph1b mutant background. In fact, Hordeum vulgare chromosome 7 was always introgressed as a telochromosome 7H v L and never the entire chromosome in the absence of the Ph1 locus. Chromosome misdivision by the centromere was frequently observed when univalents were present in the wheat background [43][44][45]. Particularly, 7H ch and 7H v telosomic chromosomes were found in wheat in the ph1b mutant background in a higher frequency than the rest of the H. chilense and H. vulgare chromosomes [35]. This confirms the difficulties found in this work to retain the whole 7H v chromosome in the H. chilense and H. vulgare double monosomic addition lines in the absence of the Ph1 locus. Thus, chromosome association and recombination in the ph1b mutant background was

Discussion
Homologous chromosomes need to recognize each other to correctly associate, recombine and segregate. Plant breeders have always wished to transfer genetic variability from related species into a crop like wheat using genetic crosses, expecting recombination between chromosomes from the donor species and those from wheat. However, wheat genome has an enormous stability and recombination between non-homologous chromosomes is hampered. The genetic control of pairing and the physical divergence of the homoeologous genomes are considered the genetic systems responsible for the stable cytological diploid behavior of polyploid wheat [40,41]. The Ph1 locus has been described controlling correct pairing and suppressing homoeologous recombination in wheat [19,24,42]. In this work, to go deeper into the knowledge of the Ph1 locus function, we developed H. chilense and H. vulgare double monosomic addition lines in the wheat ph1b mutant background for the same and for different homoeology groups. For this, we used H. chilense and H. vulgare monosomic addition lines for chromosomes 7H v , 5H ch , and 7H ch in the wheat ph1b mutant background developed previously [35] to be used as parental lines in genetic crosses. The wheat lines developed here in the absence of the Ph1 locus are equivalent to double monosomic addition lines for the same Hordeum chromosomes previously developed in our laboratory in the presence of the Ph1 locus [21]. However, we found difficulties in obtaining wheat lines carrying both Hordeum chromosomes intact at the same time in the ph1b mutant background. In fact, Hordeum vulgare chromosome 7 was always introgressed as a telochromosome 7H v L and never the entire chromosome in the absence of the Ph1 locus. Chromosome misdivision by the centromere was frequently observed when univalents were present in the wheat background [43][44][45]. Particularly, 7H ch and 7H v telosomic chromosomes were found in wheat in the ph1b mutant background in a higher frequency than the rest of the H. chilense and H. vulgare chromosomes [35]. This confirms the difficulties found in this work to retain the whole 7H v chromosome in the H. chilense and H. vulgare double monosomic addition lines in the absence of the Ph1 locus.
Thus, chromosome association and recombination in the ph1b mutant background was only possible to study between the long arm of chromosome 7H v and the long arm of both 7H ch (homoeologous) and 5H ch (non-homoeologous) chromosomes.
Most of the studies performed during meiosis in wheat plants carrying alien chromosomes are focused on meiosis metaphase I or later stages [35,46,47]. Previously, we studied interspecific chromosome associations in early meiosis stages (pachytene) and we managed to track simultaneously extra wild and cultivated barley chromosomes (for the same and for different homoeology groups) in the wheat background using in situ hybridization [21]. Both wild and cultivated barley chromosomes were shown to be associated correctly in pairs in the presence of the Ph1 locus, suggesting that the Ph1 locus does not affect to specific chromosome associations [21]. In this work, our results confirmed these previous observations related to the effect of the Ph1 locus on homoeologous chromosome associations. Wild and cultivated barley homoeologous chromosomes can associate intimately either in the presence or in the absence of the Ph1 locus. In addition, we have clearly shown here that the Ph1 locus hampers crossing-overs between wild and cultivated barley homoeologous chromosomes in the absence of homologous chromosomes. Thus, crossing-overs between H. chilense and H. vulgare homoeologous chromosomes are only observed in the absence of the Ph1 locus. Crossovers between H. chilense and H. vulgare were previously detected in the ph1b mutant background [33]. Previously, an increment on the crossover frequency between wheat homoeologous chromosomes was described in the ph1b mutant background by targeting the MHL1 (a crucial component of the meiotic recombination machinery); sites that progressed to crossovers when environmental growing conditions were altered [48]. To the best of our knowledge, we have described here for the first time an increment of the crossovers and therefore recombination between both wild and cultivated barley species in the wheat background.
Other homoeologous chromosome associations have been previously described in the presence of Ph genes. For example, recombination between H. vulgare and H. bulbosum homoeologous chromosomes or between 5M g chromosome from Ae. geniculata and 5D chromosome were also observed in the wheat background [49,50]. Robertsonian translocations between H. chilense and H. vulgare were described before in the presence of the Ph1 locus [51]. However, different levels of allosyndesis between wheat chromosomes and H. vulgare and H. chilense or wheat and rye (Secale cereale) were detected in the ph1b mutant background [33,52]. Our results show that, although H. vulgare and H. chilense are phylogenetically quite distant [53], both species have a lot of similarities at the chromosomal level [54] allowing homoeologous chromosome associations and recombination between them in the absence of the Ph1 locus. Thus, our results also showed that the Ph1 locus suppresses recombination between homoeologous chromosomes as bivalents between homoeologous chromosomes were detected in metaphase I in the ph1b mutant background.
Alterations in chromosome content can also be frequently generated by mistakes during cell division, producing aneuploidy or polyploidy progeny cells. It is not strange that high polyploids show a variable chromosome number even for the same ploidy level. This phenomenon was labelled as aneusomaty, referring to an intra-individual aneuploidy [55]. This term can be equivalent to mosaicism and in wheat can be assumed to arise as the result of the meiotic instability characteristic of freshly created allopolyploids [56]. In our work, we detected this aneusomaty or mosaicism phenomenon in one plant developed in the ph1b mutant background carrying the H. chilense and H. vulgare non-homoeologous chromosomes. In our case, the H. vulgare chromosome was detected only in some somatic cells in the anther but in all somatic cells in the root tissue. Chromosome number variations in a population and at individual level can be frequent in high polyploids, especially in plants with an active vegetative reproduction [55,[57][58][59]. This phenomenon has not been reported previously in the ph1b mutant background and more observations would be required to clarify whether or not the Ph1 locus might have an effect on it.
Subtelomeres are high polymorphic DNA structures, located near telomeres at the end of the chromosome arms [60], and include recombination hot spots [61,62]. This extraordi-nary polymorphism contributes to complicating the picture about the roles that subtelomeres play on chromosome associations and recombination during meiosis, although some studies in Arabidopsis and Hordeum species have shed some light on it [21,[63][64][65]. In our work, crossovers were always observed at the distal region of the associated homoeologous chromosomes. Our results here also suggest that subtelomeres might play a function in the interspecific recombination between wild and barley chromosomes, as chiasmas are always located at the end of the chromosomes near the telomeres. The mechanism by which recombination at the subtelomeres between homoeologues is suppressed by the Ph1 locus remains to be elucidated.
Author Contributions: Both authors contributed to this manuscript. P.P. designed the research; P.P. and M.C.C. performed the experiments; P.P. and M.C.C. analyzed and discussed the results. All authors have read and agreed to the published version of the manuscript.
Funding: This research was supported by grants AGL2015-64833R and PDI2019-103339RB-100 from Spanish Ministerio de Economía y Competitividad and Ministerio de Ciencia e Innovación, respectively.