The starting point of this work was a discovery of QGlc.ipk-5B quantitative trait locus (QTL) on 5B chromosome in ITMI mapping population (Pshenichnikova et al. 2008). We decided to test the reliability of this valuable locus, for breeding by introducing it into another wheat genotype. Line ITMI-43 that exhibited a high value of this trait was selected as a carrier of introgression from the synthetic W7984 in the pericentromeric region of 5B chromosome. In addition, our previous study of the diversity of technological properties of grain and flour among the lines of the Arsenal collection (Shchukina et al. 2020) showed that the 73/00i line is characterized by a particularly high gluten content in grain. By that time, it was known that, among others, it also carried introgression in the proximal region of the long arm of 5B chromosome (Adonina et al. 2012). This line was also chosen to verify the genetic effect of introgression in an unrelated genotype. The recipient cultivar S29 is characterized by a relatively low protein and gluten content in grain and, in addition, the monosomic lines were developed for it. Eight backcrosses were made for each of the lines in order to restore the genetic background of the recipient to the maximum extent possible, with the exception of the donor 5B chromosome. As a result, two substitution lines were obtained, carrying different introgressions in different parts of 5B chromosome. Line S29 (73/00i 5B) carries the introgression from Ae. speltoides, and line S29(ITMI-43 5B) the introgression from T. durum, inherited from the hexaploid synthetic wheat W7984 (Triticum tauschii CIGM86.940 × T. durum, cv. Altar84) (Nelson et al. 1995). The correctness of the substitution was confirmed by the allelic composition of microsatellite loci in the lines, which corresponded to the introgression lines (Fig. 3).
The gluten content was a target trait that was controlled during backcrossing. It was found that in the case of line S29(73/00i 5B) there is a decrease in gluten content compared to the donor line. Initially, donor and recipient differed by an average of 8.6% (Table 1). In the second backcross, the differences between S29 and the line were still high and amounted to more than 12%, in the fourth – about 7%, and by the eighth backcross the average value dropped to 4% (Table 2). This may indicate that introgressions identified in chromosomes 1B and 6B in line 73/00i 5B also carry genes associated with gluten content. We tested this assumption by backcrossing onto 1B monosomic line. After three backcrosses, no increase in gluten content was detected in monosomic hybrids, and the development of this line was discontinued. We are currently carrying out a similar work with 6B chromosome; it is giving a positive result - the line consistently demonstrates a high gluten content at intermediate backcrosses. The results obtained indicate that the target trait has an additive inheritance and can be separated into individual genetic components.
In the case of line S29(ITMI-43 5B), carrying locus QGlc.ipk-5B, a different picture was observed. Initially, the recipient and donor differed in the target trait by 4% (Table 1). This difference was observed both at the intermediate stage of backcrossing and at the final eighth backcross (3% and 4%, respectively). It should be noted that the donor of the mapping population - Synthetic W7984 - was studied twice under the conditions of Novosibirsk in terms of gluten content and demonstrated a difference of 10% from S29. In our previous work (Pshenichnikova et al. 2008) in addition to the QTL on 5B chromosome, we identified another locus QGlc.ipk-7A on 7A chromosome. Börner et al. (2002) identified a locus associated with grain protein content at a comparable position in the ITMI population. It is likely that line ITMI-43 did not inherit this locus. This observation indirectly confirms the additive nature of the inheritance of gluten content in grain.
Long-term field tests of two sister lines S29 (73/00i 5B) under the field conditions of Novosibirsk showed approximately the same average significant excess of protein content by more than 1 percent and gluten by more than 4 percent compared to S29. In line S29(ITMI-43 5B) the average gluten content in the grain was 4% higher than in the S29. The tests carried out in other geographical areas also showed an average significant excess of protein and gluten content in grain (Table 3). They were 0.9% for protein content and 2.5% for gluten content. However, there were large differences in the expression of target traits in different conditions. This was especially true for the protein content. In Barnaul and Omsk, both lines exceeded the recipient in each year of the study. In Ussuriysk, line S29 (73/00i 5B) showed no difference from S29, and line S29 (ITMI-43 5B) differed little from it in Gatersleben. In terms of gluten content in grain, line S29(ITMI-43 5B) in Omsk was also comparable to S29. Figure 4 shows the comparative average gluten contents of the parent cultivar and the lines under different conditions. It can be seen that the greatest differences were observed in the conditions of Novosibirsk, Ussuriysk and Gatersleben. These regions have more favorable summer climatic conditions compared to Omsk and Barnaul. The latter two regions are drier; summer rainfall (on grain producing territories) can be less than 200 mm. Line S29 (73/00i 5B) was grown for the largest number of field seasons in Novosibirsk (six). The most favorable of them were four summer seasons (2016–2018, 2020), when the number of days with precipitation was more than 20 days (https://world-weather.ru/pogoda/russia/novosibirsk/may-2016/). During these years, the
line consistently formed 36–38% gluten in the grain. To varying degrees, it maintained a higher gluten content under all growing conditions. Line S29(ITMI-43 5B) was more or less stable only under normal humidification conditions - in Novosibirsk in 2016–2018, in Gatersleben and Ussuriysk (Fig. 4). Data from a greenhouse experiment also indicate that this line is not drought-tolerant (Table 6).
The first gene that increases protein content in wheat grain, extracted from a wild relative, is Gpc-B1. This gene has been extensively studied, its function and its belonging to gene families have both been established, along with the variability among cereal species (Tabbita et al. 2016). The primary source of Gpc-B1 is the line LDN(DIC-6B) of Langdon durum wheat with chromosome 6B substitution from the tetraploid wild species T. turgidum L. var. dicoccoides (Joppa, Cantrell, 1990). In the mentioned work, the substitution line for 5B chromosome also demonstrated a high protein content compared to the recipient in 3 of the 4 studied environmental conditions. Two more lines created by the same authors with replacement of 5B chromosome from other accessions of this wild tetraploid species were also high-protein (Xu et al. 2008). Gonzalez-Hernandez et al. (2004), using a Langdon line with 5B chromosome replacement, mapped three loci associated with an increase in protein content. Two of them coincide with the introgression regions in our lines. In line S29(73/00i 5B), the introgression includes the area of marker Xgwm604, which is located 5 cM from marker Xgwm408 (Fig. 2). In Gonzalez-Hernandez et al. (2004) the locus Qgpc.ndsu-5B.1 was mapped here with the greatest phenotypic effect, manifested in all environmental conditions. In the region of the Xgwm408 marker, Golabadi et al. (2012) also mapped a locus associated with protein content in a recombinant population obtained from crossing drought-tolerant and non-drought-tolerant durum wheat cultivars. Line S29(ITMI-43 5B) carries introgression in the pericentromeric region, which includes marker Xgwm 213. It is located 3.4 cM from marker Xwg909, which was the closest to locus Qgpc.ndsu-5B.3 with a weaker phenotypic effect on protein content (Gonzalez-Hernandez et al. 2004). This effect did not manifest itself in all environmental conditions. Thus, chromosome 5B of tetraploid Triticum species can, like 6B, be a source of genes that increase protein content in grain.
The two introgression lines differed in the physical and mixing properties of flour and dough. Introgression lines - donors of 5B chromosomes had very low indicators (Tables 4, 5). Overall, the indicators were substantially improved, and line S29(73/00i 5B) even outperformed S29, the ‘strong’ class wheat. This line had a higher alveograph flour strength and dough extensibility and higher dough resistance to mixing. Line S29(ITMI-43 5B) showed lower mixing performance than the recipient. The line was characterized by lower dough stability and higher liquefaction (Table 5). Previously, in our work with ITMI population (Pshenichnikova et al. 2008), QTLs associated with farinograph mixing indicators were found on chromosome 5B. In particular, locus QDsa.ipk-5B, associated with dough stability during mixing, was localized in the region of the same Xtam72 marker as locus QGlc.ipk-5B, the target locus of this study. Thus, by transferring chromosome 5B from line ITMI-43 to S29 genotype we were able to verify this locus as well.
The study of yield components showed that the lines exhibit variability in individual parameters. Under greenhouse conditions, line S29(73/00i 5B) was less productive under irrigated and drought conditions, while line S29(ITMI-43 5B) reduced yield only under drought conditions (Table 6). Under field conditions in Novosibirsk, the first line did not differ from the recipient in terms of yield, while the second line demonstrated a significant increase in the number of grains from secondary tillers and per plant. In a study conducted by Mancini et al. (2018), quantitative trait loci (QTL) associated with crop components were mapped in a collection of tetraploid wheat using genome-wide association studies (GWAS). Two QTL, QTkw.mgb-5B and Gys.mgb-5B, were found to be associated with TKW and grain yield. These QTL were located on 5B chromosome, approximately 10 cM distal from the microsatellite marker Xgwm499. Based on this finding, it can be assumed that variability in yield components in our substitution line is also linked to these QTL. The fact is that 5B chromosome of line ITMI-43, in addition to centromeric introgression, also contains a distal introgression on the long arm (personal communication from M. Röder). We have previously not localized any QTL related to quality indicators in it (Pshenichnikova et al., 2008). Therefore, we have not considered this introgression region in this study. However, we analyzed the allelic composition of locus Xgwm499 and detected the variability between the lines and S29. According to Mancini et al. ((2018) this locus is 8.7 cM distal to the SNP marker IWB66934, which marks the yield-related loci QTkw. mgb-5B and QGys.mgb-5B. An indirect confirmation of the presence of such loci in lines S29 (73/00i 5B) and S29(ITMI-43 5B) is a decrease in TKW in the line under greenhouse conditions (Table 6), and in the field conditions of Novosibirsk for line S29(ITMI-43 5B) (Table S4).
Previously, we were able to successfully transfer introgression from the species T. timopheevii to S29, which increases the protein and gluten content in grain (Shchukina et al. 2022). The introgression was located on chromosome 2A and characterized by the stable phenotypic expression of the target traits at the same geographic locations as in our study. The increase in protein and gluten content was 2% and 4%, respectively, and was not associated with the formation of yield components. Introgressions into chromosome 5B differ in their phenotypic manifestations, primarily in that their effects are, on average, lower - about 1% of protein and less than 4% of gluten. To establish the statistical significance of these differences, we conducted dozens of experiments in different field settings (Table S2). As it turned out, their manifestation depends on climatic conditions, namely, humidification conditions. In addition, these introgressions significantly affected yield components, in particular, TKW. It is possible that this effect is typical only for S29 genotype, and will not be negative in another genotypic environment. Such a case was described for gene Gpc-B1, which is prone to decrease TKW (Tabbita et al. 2016). With introduction into the Indian wheat gene pool this effect was overcome (Vishwakarma et al. 2014).