**1. Introduction**

Starch is the main component of wheat grain, constituting up to 75% of its dry weight. This polysaccharide contains two different glucose polymers: amylose (22%–35% of the total) and amylopectin (68%–75% of the total) [1]. Changes in the ratio between these polymers have a clear influence on starch gelatinization, pasting and gelation properties [2], affecting the end-use quality levels of different wheat products, such as bread, pasta, and noodles [3–5], as well their shelf-lives [6] and nutritional values [7].

Starch synthesis involves several starch synthases, starch branching enzymes, and starch debranching enzymes [8]. The most studied of these proteins has been the granule-bound starch synthase I (GBSSI) or waxy proteins (ADP glucose starch glycosyl transferase, EC 2.4.1.21), which are solely responsible for amylose synthesis [9]. In wheat, these proteins are synthesized by genes located in the short arm of the seven-group homeologous chromosome, with the exception of the *Wx-B*1 gene that, owing to a translocation event, is located in the 4AL chromosome [10]. In wheat relatives, this gene is located in similar positions, and its molecular configuration of 12 exons and 11 introns is highly conserved in all of these species [11]. In other Poaceae species, such as barley (*Hordeum vulgare* L.), this gene has shown the identical structure and location [12].

The variability of waxy proteins has been studied in common and durum wheat, as well as in some wild and cultivated relatives [13]. However, the variability in modern wheat cultivars is not very wide, according to data in the Wheat Gene Catalogue [14]. In the search for new waxy variants, species from the primary and secondary wheat pools could contain good candidates. These species have been successfully used to transfer useful traits to wheat. In some cases, these transfer events have generated amphiploids that have also been used as bridge species [15]. In other cases, these amphiploids have been derived to produce human-made crops, such as tritordeum (×*Tritordeum* Ascherson et Graebner), that have shown promising characteristics [16].

Tritordeum was synthesized using mainly durum wheat and *Hordeum chilense* Roem. et Schult. (2*n* = 2× = 14, *Hc<sup>h</sup> Hc<sup>h</sup>*), a wild barley species native to Chile and Argentina, included in the section *Anisolepis* Nevski [17]. In this species, variation in genes related to quality has been widely evaluated over the last decade [18–21] and has been used to expand the genetic base of tritordeum. Furthermore, this species exhibits advantageous agronomic and quality characteristics [22–24], which, together with its ability to be crossed with other members of the *Triticeae* tribe [25], make it useful in cereal breeding.

In its natural distribution area, *H. chilense* shows some ecotypes, based on morphological and ecophysiological traits [26,27]. The two main groups are related to the first two *H. chilense* lines used to develop tritordeum, H1 and H7. Tritordeum developed using these lines showed di fferences in fertility, grain size, and life cycle, depending on the female parent (H1 or H7) used [25]. An analysis of the genes related to the flour quality from both lines also showed that these ecotypes could have di fferent e ffects on the tritordeum quality. These di fferences have been evaluated for the seed storage proteins [18,28], hordoindolines [29], and pigment enzymes [30,31].

The main goals of this study were to analyze allelic variation and molecularly characterize of the *Wx* genes in the H1 and H7 lines of *H. chilense*, and to determine the gene's chromosomal location.
