1. Introduction
The introduction of the
Reduced height (
Rht) genes associated with insensitivity to gibberellins, plant growth hormones into new varieties caused a rapid increase in wheat productivity in the second half of the XX century [
1]. Gibberellin-insensitive reduced height genes can enhance the resistance of wheat varieties to lodging, especially under conditions of high doses of nitrogen fertilizers, making them adapted to cultivation under intensive farming [
2]. Currently, several dozen reduced height genes and their alleles have been described [
3]. At the same time, the gibberellin-insensitive short stature trait of most modern commercial wheat varieties is provided by mutations of only two genes—
Rht-B1 and
Rht-D1. Mutations of these genes—
Rht-B1b (previous designation
Rht1) and
Rht-D1b (previous designation
Rht2)—were originally transferred to the CIMMYT-developed varieties (International Center for the Improvement of Maize and Wheat/Centro Internacional de Mejoramiento de Maíz y Trigo), and later to European varieties from the Japanese dwarf cv. Norin 10, and ensured the success of the Green Revolution [
4]. The
Rht-B1e mutation was also widely introduced in Russian wheat cultivars [
5]. Despite the known disadvantages, gibberellin-insensitive dwarfism is still preferred for wheat varieties cultivated under sufficient moisture conditions [
6].
The reduced height of plants can be associated with both the impaired biosynthesis of gibberellins and the accumulation of repressors of the hormonal signal of gibberellins—the DELLA proteins, while tall plant phenotype can be associated with damage to the gibberellin deactivation enzymes or the loss of the repressive function of the DELLA proteins. The
Rht-B1 and
Rht-D1 genes encode DELLA proteins, which function as transcriptional coactivators and corepressors [
7,
8]. These proteins are negative regulators of the gibberellin signaling pathway. A high level of DELLA proteins in plant cells suppresses the growth of their vegetative organs, while the activation of growth by gibberellins is mediated by the degradation of these proteins [
1]. The gibberellin receptor, GID1 (GIBBERELLIN-INSENSITIVE DWARF 1), plays an important role in regulating the cell level of DELLA proteins. GID1 in the presence of these hormones acquires the ability to bind to DELLA proteins, after which the GID1-DELLA complex is recognized by the F-box proteins—SLY1 (SLEEPY1) and GID2 (GIBBERELLIN-INSENSITIVE DWARF 2), which form ubiquitin ligase complex, after which DELLA-protein is ubiquitinated and subjected to proteasome degradation [
9]. The
Gid1 gene (
Gibberellin-insensitive dwarf 1) was first identified in dwarf rice mutants that did not respond to treatment with exogenous gibberellic acid by increasing the growth of experimental plants [
10]. The decrease in expression of the
Gid1 gene and, as a consequence, a dwarf phenotype was revealed in a series of different plant mutants. DELLA proteins contain two main domains—the N-terminal DELLA domain, which is involved in the interaction with GID1, and the GRAS domain, which possesses transactivation, repressive, and regulatory activities [
9]. Reduced height mutations of the wheat
Rht-1 gene are associated with damage or absence of the DELLA domain, which makes these proteins more stable in the cell and, accordingly, significantly reduces the growth processes stimulated by gibberellins in mutant plants [
11].
In bread or common wheat (
Triticum aestivum L.; 2
n = 6
x = 42, BBAADD genome), DELLA proteins are encoded by
Rht-A1,
Rht-B1, and
Rht-D1 genes, referring, respectively to its subgenomes A, B, and D. The
Rht-A1 gene does not have any known reduced-height mutations. The
Rht-B1 gene has the largest number of ones. Its
Rht-B1b,
Rht-B1d,
Rht-B1e, and
Rht-B1p alleles determine the semi-dwarf phenotype caused by the emergence of stop codons within a small region of the
Rht gene, after which translation is likely to be reinitiated [
12,
13]. The
Rht-B1c allele determines the extreme dwarf phenotype caused by the insertion of the retrotransposon, which leads to the occurrence of an intron in the gene and a 30-amino acid insertion in the region of the protein DELLA domain [
12]. In addition, numerous agronomically important reverse mutations of the
Rht-B1c allele with a semi-dwarf phenotype were obtained [
14].
Rht-D1 gene has a smaller number of known reduced-height mutations—the
Rht-D1b allele (originally described as the
Rht2 gene), due to a stop codon, and the
Rht-D1c allele that determines the extreme dwarf phenotype and represents a duplication of the
Rht-D1b allele [
15]. In addition to the alleles listed, a significant number of functionally neutral mutations have been described in the
Rht-1 genes of polyploid wheat [
16]. There are also many mutations associated with an increase rather than a decrease in plant height [
17]. The allelic diversity of the wheat
Gid1 and
Gid2 genes known to date is scarce [
18,
19]. The lack of involvement of the
Gid1 and
Gid2 genes in breeding programs for the development of commercial semi-dwarf wheat varieties could be explained by the recessiveness of their dwarfing mutations, which could be assumed based on the function of the encoded proteins, and the requirement of the simultaneous presence of recessive mutations in all homoeologous genes of allopolyploid wheat for expression of the dwarfism.
The alleles of the wheat
Rht-1 genes are known to differ in the degree of influence on plant height and other agronomically important traits, while their pleiotropic effects do not necessarily strictly correlate with each other. Thus, the
Rht-D1b allele reduces the resistance of wheat plants to
Fusarium head blight to a greater extent than the
Rht-B1b allele [
20], and the
Rht-B1c.23 and
Rht-B1c.26 alleles shorten the stem length only slightly more than the
Rht-B1b allele, but at the same time significantly increase the duration of the seed dormancy period, rising the resistance of wheat to preharvest sprouting [
14]. This means that the study of the polymorphism of the gibberellin signaling pathway genes can enrich the toolkit of breeders with new alleles that have unique valuable combinations of multiple phenotypic expressions.
Many studies have shown the existence of a significant allelic diversity of agronomically important genes in diploid ancestral wheat species and its wild relatives in comparison with widely cultivated tetraploid and hexaploid wheat species [
21,
22].
Aegilops tauschii Coss. (= syn.
Ae. squarrosa L.; 2
n = 2
x = 14, DD genome) is one of the ancestral species of bread wheat and a donor of its D subgenome [
23,
24]. It also participated in the formation of many polyploid species of the genus
Aegilops L. as a donor of their cytoplasmic genome [
25]. However, during bread wheat evolution, only a handful of
Ae. tauschii accessions from a small region hybridized with wheat leading to a narrow genetic base of the wheat D subgenome. Therefore,
Ae. tauschii should be used more widely in bread wheat breeding.
At present, there is an intensive search for molecular polymorphisms [
26,
27], which could be used in wheat breeding in the future. The genetic diversity of
Ae. tauschii has been proved to exceed significantly that of the D genome of polyploid wheat [
28]. For this reason,
Ae. tauschii is among the most promising donors of economically valuable traits for bread wheat, and, therefore, the study of diploid ancestral species of polyploid wheat is promising for the search for the new gene variants. These variants may prove to be agronomically important if transferred to widely cultivated wheat species through specially designed bridge species [
29] or by producing synthetic allopolyploids [
30,
31].
The objective of this research is to study the allele diversity of the gibberellin signaling pathway genes—Rht-D1, Gid1-D, and Gid2-D—in Ae. tauschii.
3. Discussion
Currently, wild-related species of crop plants are increasingly used to search for new alleles of genes that provide resistance to biotic and abiotic stresses [
22], as well as genes that control other agronomically important traits. This is due to their high adaptability, great genetic diversity, and the possibility of transferring valuable alleles to crop species by distant hybridization [
22,
30,
33].
Ae. tauschii is a wild self-pollinating grass, a donor of the D subgenome of bread wheat. The species has a vast natural range in central Eurasia, spreading from Turkey to western China, and is mainly found in the Caucasus and Iran along the coast of the Caspian Sea [
34]. Within the entire range,
Ae. tauschii is represented in the form of small, isolated from each other populations [
35]. Populations of
Ae. tauschii have adapted well to a variety of growing conditions, including sandy shores, rocky hills, roadside, and wet forests.
Ae. tauschii is often found as a weed in wheat and barley fields [
36]. Nowadays, the efforts of geneticists and breeders have aimed at the recruitment of the
Ae. tauschii gene pool for the improvement of modern cultivars of bread wheat.
Morphologically,
Ae. tauschii is traditionally divided into two subspecies—
Ae. tauschii Coss. ssp.
tauschii and
Ae. tauschii Coss. ssp.
strangulata (Eig) Tzvel. Using molecular methods, it was divided into two evolutionary lines [
37]. The
tauschii subspecies has the widest distribution and a wide variety of forms, which are traditionally grouped into four botanical varieties [
38]. The
strangulata subspecies is widespread in the southern part of the Caspian Sea and, according to some early studies, in the Transcaucasia and is distinguished by wider spikelets [
34,
39]. The
strangulata subspecies is considered the most likely donor of the D subgenome of bread wheat [
39]. The
tauschii subspecies is much less studied for this reason. Therefore, we mainly studied the
tauschii subspecies, the gene pool of which was not previously involved in the formation of the biodiversity of the D genome of hexaploid wheat. In addition, the
strangulata subspecies is believed to be more polymorphic than the
tauschii subspecies [
40].
In this work, we studied the genes of the gibberellin hormone signaling pathway—
Rht-D1,
Gid1-D, and
Gid2-D, and described new alleles for each of them (
Table 1,
Table 2,
Table 3 and
Table 4). We found seven
Rht-D1 haplotypes in
Ae. tauschii, which give four protein isoforms differing in amino acid sequence. Moreover, all these haplotypes are not natural for bread wheat. The
Rht-D1a_5 and
Rht-D1a_7 haplotypes were previously described only in synthetic allopolyploids [
16]. Most of the
Rht-D1 alleles of bread wheat, which control the tall plant phenotype, encode isoform C of the RHT-D1 protein, which was found in only one accession in our study—
Ae. tauschii ssp.
tauschii KT120-10 from China (
Figure 1). Thus, only a small part of the diversity of the
Ae. tauschii Rht-D1 gene was transferred to bread wheat, and there is a prospect of expanding the wheat gene pool through the development and involvement of synthetic allopolyploids in modern breeding programs.
According to PROVEAN, isoform A of RHT-D1 protein should suppress stem growth to a lesser extent than isoform B. This is partially confirmed by the data of previous studies, according to which,
Ae. tauschii plants are shorter in the eastern regions of the species range than in the western parts [
34,
41]. According to our data, protein isoform B is more common in the eastern regions—Pakistan, Uzbekistan, Afghanistan (
Table S1). Perhaps the development of isoform B of the RHT-D1 protein is the result of plant adaptation to the sharply continental climate of this part of the range. Statistical analysis of our data on
Ae. tauschii plant height showed no significant association of this trait with the isoforms of the RHT-D1 protein. It is possible that an adaptation of
Ae. tauschii through the
Rht-D1 mutation is not directly related to plant height, but to other traits regulated by gibberellins, for example, the timing of anthesis (heading) or the duration of the seed dormancy. It is also possible that the expression of various
Rht-D1 alleles was compensated for by other genes—
Gid1-D and
Gid2-D, the alleles of which, as we have shown, are not randomly combined with the
Rht-D1 alleles (
Table 5).
The
Ae. tauschii Gid1-D gene has 13 different alleles found to encode two isoforms of the protein, designated by us with letters A and B. We found that isoform B of the RHT-D1 protein occurs only together with GID1-D isoform A, and GID1-D isoform B is found only with RHT-D1 isoform A. This joint occurrence of two proteins’ isoforms can be explained by the natural selection of their combinations. The GID1-D isoform B is encoded by a relatively young monophyletic group of alleles. According to PROVEAN, this isoform should be less functional, which means, most likely, associated with the short stature of plants. However, for the GID1-D isoforms, we were unable to establish a statistically significant relationship with plant height. Most likely, the study of the phenotypic expression of the detected alleles will become possible only through hybridological analysis. Nevertheless, the dendrogram (see
Figure 2) clearly shows three clusters. The upper one is formed by the accessions from Central Asia, the middle one—from the Transcaucasia. An accession from China is located separately. This concordance of genetic and geographical divergence makes it possible to select specific accessions from specific regions to transfer the vast genetic diversity of
Ae. tauschii into bread wheat by a limited number of interspecific crossings.
The Gid2-D gene showed to have eight alleles encoding two isoforms of the GID-2 protein. Isoform A corresponds to most of the identified alleles, and isoform B corresponds to the Gid2-D1b allele, which is closest to the ancestral variant of the gene.
The closest to the ancestral variants of the
Rht-D1 and
Gid2-D genes are found in Armenia, Azerbaijan, and Iran. Combinations of rare alleles are observed in the same area. All this confirms the hypothesis about the center of origin of
Ae. tauschii in Transcaucasia [
38,
42]. As for the identification of a
Gid1-D allele that is close to the ancestral, in plants collected in China, this one could have been introduced there by humans as part of the
Ae. tauschii genome, which existed as a weed in bread wheat fields. This allele persisted there due to the isolation of the Chinese populations of
Ae. tauschii from the rest of the range. Probably, this form of
Ae. tauschii introduced into China was a descendant of the one that once donated the D subgenome for bread wheat. This is also evidenced by the discovery of the
Rht-D1 allele encoding the protein isoform C in KT 120–10 accession from China, which is characteristic of bread wheat, but rarely occurs in the studied plants of
Ae. tauschii.
We do not know any examples of height-reducing genes transferred from
Aegilops species with D genome to bread wheat [
43]. Therefore, our findings provide useful information for further unlocking of the genetic mechanism of agronomic trait control in wheat.