Two germplasm sets were used for crossing and generating segregating population for combining PsyE1 (Y gene) and GpcB1 gene. First set of germplasm consisted of BC₁F₅bulk plants havingPsyE1 (Y gene) in HD2967 background (HD2967 + PsyE1/Sr25/Lr19). The set consisted of 180 lines, out of which an agronomically better-performing line with high pigment content was chosen and used as parent 1 (Padhy et al., 2022c). A commercial bread wheat variety HD2967 (ALD/COC//URES/3/HD2160 M/HD2278) developed by the Indian Agricultural Research Institute (IARI), New Delhi for cultivation under timely sown irrigated conditions in the northwestern plain zone of India (notification number 2326 (E) dated 10.10.2011, Indian Council of Agricultural Research (ICAR), Government of India). The variety has an average yield of 5.1 t/ha (potential yield 6.6 t/ha) with bold, round, lustrous, and amber-colored grains with a protein content of 10.7%. Similarly, the second set of germplasm consisted of the GpcB1 gene and Yr15 in the PBW550 background (PBW550 + GpcB1+Yr36+ Yr15) in the F₆ generation. Out of 81 available lines having the same genetic composition, the third line (F₆ 3) was used for crossing as parent 2 owing to its better agronomic performance, high protein content, and low polyphenol oxidase activity (Supplementary Table S2, Gill M S, 2017). PBW550 (WH594/RAJ3856//W485) is a double dwarf early maturing bread wheat variety developed and released by Punjab Agricultural University, Ludhiana (notification number 72 (E) dated 10.01.2008, Indian Council of Agricultural Research (ICAR), Government of India) for timely sown irrigated conditions in the northwestern plain zone of India having good grain and chapati quality.

A stepwise hybridization and phenotypic-biochemical-molecular selection strategy in the introgression program were followed to obtain agronomically elite plants with desirable genes under consideration. The germplasm lines of parent 1 and parent 2 were validated using biochemical and molecular analysis for carotenoid content, protein content, and rust resistance genes. Eight individual crosses were made between parent 1 and parent 2 where eight different lines with the same genetic composition (HD2967 + PsyE1/Sr25/Lr19) were used as parent 1 against single parent 2 (‘F₆ 3’ having genetic composition PBW550 + GpcB1+Yr36+ Yr15), during the main season (November to May) at ‘Wheat Experimental Area’, Punjab Agricultural University, Ludhiana (30.9010° N, 75.8573° E, 810 ft above msl) in 2017–18 (Table 1). F_1’s were raised at the off-season (June to October) location of Punjab Agricultural University, Ludhiana situated in the Himalayas at Keylong, Lahaul&Spiti District, Himachal Pradesh (32.210° N, 77.140° E, 10,500 ft above msl). Off-season or alternate season refers to production outside their typical cropping cycle and is used to advance the generations mainly by the breeders.

F₂ was raised at Ludhiana during the main crop season in 2018–19. The selection of phenotypically promising rust-resistant single plants was done in artificially inoculated conditions for stripe rust and leaf rust. Grains from selected plants were screened for carotenoid pigments using spectrophotometric evaluation and high grain protein content (GPC) using the Kjeldahl method as well as for the genes of interest using MAS. Selected single plant progenies along with parental lines HD2967 and PBW550 were planted in 1.5 m paired rows with a plant-to-plant distance of 10 cm and row-to-row distance of 20 cm in a randomized block design. The selected lines were carried forward to F₃/F₄/F₅ in 2019–20, 2020–21, and 2021–22, respectively. In each generation, plant/progenies selection was based on rust resistance, carotenoid content, and GPC. Four promising lines having all the desired genes combined with elite agronomic traits and grain quality (luster, boldness, thousand-grain weight, etc.) were identified for evaluation of yield potential.

Screening for stripe rust and leaf rust was performed in the field along with MAS to validate the effectiveness of rust resistance genes linked to gene GpcB1(Yr36) and PsyE1 (Lr19) as well as for Yr15. Progenies from F₂/F₃/F₄ and F₅ generations were screened against stripe rust and leaf rust at the adult plant stage in the field during seasons 2017–18 to 2021–22. For screening, artificial epiphytotic conditions for rust were created by spraying the urediniospores of Pt pathotypes (100S119, 78S84), and Pt pathotypes (77–1, 77–2, 77–5, 104–2) diluted in water containing Tween-20. For the uniform spread of disease, highly susceptible cultivars HD2967, PBW343, and PBW550 were planted as spreader rows all around the field and after every 20 rows (Kaur et al., 2020). Data was recorded using the Cobb’s scale, as illustrated by Peterson et al., (1948), when ‘check’ cultivars showed complete susceptibility. The screening for stem rust could not be done in field conditions in Punjab since it is not prevalent in the region and hence, it was monitored using linked molecular markers only. As Sr25 is linked to PsyE1, therefore, the positive selection for PsyE1 implies the presence of Sr25.

The selection for the traits i.e., high carotenoid content, high protein content, and resistance to foliar rusts was achieved in a combined phenotypic—genotypic strategy. For the marker-assisted selection, the DNA of F₂ plants was extracted from young leaves using the CTAB method (Saghai-Maroof et al., 1984) with some modifications (Kaur et al., 2020). PsyE1/Lr19/Sr25 was screened using a combination of gene-specific markers i.e., Psy1-E1 specific and Psy1-D1 specific (Zhang and Dubcovsky 2008). F₂s were screened for GpC-B1/Yr36 using the dominant marker Xucw108 (Uauy et al., 2006). Similarly, the rust-resistant gene, Yr15, was screened with marker Xbarc8 (https://wheat.pw.usda.gov/) using specific primers (Table 2). The primers were used to amplify the desired DNA segment using specific PCR (polymerase chain reaction) profiles (Supplementary Table S1) and separated by gel electrophoresis using 2% agarose gel in 0.5X TBE buffer. The amplicons (Table 2) were visualized in a Biorad-gel-documentation system with an ethidium bromide stain.

Based on the screening of individual populations for high protein content (PBW 550 + Yr15 + GpcB1/Yr36) and high carotenoid content (HD 2967 + PsyE1/Sr25/Lr19), the phenotypic expression of both PsyE1 and GpcB1was observed to be variable among the plants, as guided by their genotype constitution. Therefore, it is necessary to choose the line with the best agronomic characters as the parental type for developing the segregating population. Amongst the eight crosses made among selected lines, four crosses (i.e. BC₁F₅ 20/F₆ 3, BC₁F₅ 18/F₆ 3, BC₁F₅ 154/F₆ 3, BC₁F₅ 23/F₆ 3) survived, while the F₁generation of the remaining crosses suffered hybrid necrosis due to low temperature and the genetic makeup of the cultivar HD2967 (Supplementary Figure S1) (Mishra et al., 2017; Takumi and Mizuno, 2011). Hybrid necrosis (gradual premature death of leaves or plants in certain wheat F₁ hybrids) is caused by the interaction of two dominant complementary genes, Ne1 and Ne2, located on chromosome arms 5BL and 2BS, respectively (Chu et al., 2006) and the HD2967 genotype carrying both the genes is often reported to cause hybrid necrosis in F₁s. The seeds harvested from the remaining four crosses were sown as a single seed to raise a total of 2748 F₂ individuals which were phenotyped for rust resistance (stripe rust and leaf rust) along with yellow grain pigment content. A subset of 649 F₂ plants was selected for further screening of biochemical constituents (carotenoid content, protein content, and polyphenol oxidase activity) and MAS (using gene-specific markers for the genes, Yr15, GpcB1/Yr36 and PsyE1/Sr25/Lr19) (Figure 1).

MAS was carried out to select the genotypes combining PsyE1/Sr25/Lr19, GpcB1/Yr36, and two stripe-rust-resistant genes along with Yr15 in the F₂ plants sown as single plants. The four crosses which survived made up a total of 2748 F₂ individual plants. The conventional MAS was modified to achieve the best results out of the selected F₂ plants. For the gene pyramiding experiments, expression of the genes in terms of the phenotype is equally important as their presence in a genotype. Therefore, sequential screening was performed for rust resistance (stripe rust and leaf rust) followed by estimation of carotenoid content and protein content before molecular validation to minimize the time, effort, and cost for the selection of pyramided plants (Figure 1).

PsyE1 gene introgression resulted in a significant increase in the carotenoid content, which varied up to 9.49 ppm in the selected pyramided lines (Supplementary Table S6). Further, the GpcB1 gene was pyramided with PsyE1in the current experiment, and the former was observed to increase the protein content up to 14% and 13.5% with and without yield penalty, respectively. The gene also resulted in increased remobilization of nutrients, thereby increasing the Fe and Zn concentration by 10% and 5%, respectively (Uauy et al., 2006). Uauy et al., 2006 first cloned and characterized the grain protein content associated with gene GpcB1 as well as developed the functional markers which have opened numerous avenues to further increase the nutritional significance of bread wheat. The study also showed that GpcB1 is also responsible for increased iron and zinc content in grain. Although the expression of this gene is influenced by the environment with a coupled yield penalty (Brevis and Dubcovsky, 2010), examples of introgression of the GpcB1 gene through MAS without compromising crop yield are also available (Uauy et al., 2006; Brevis and Dubcovsky, 2010). However, GpcB1-induced accelerated senescence resulting in shriveled grain seems to be the significant reason that even after widespread use of this gene in almost all wheat breeding programs for a decade, no commercially released variety has this gene introgressed in it. In the present study, the GPC was increased by 2% of existing content, making a total of 13%–14% with minimal yield compromise along with the carotenoid content raised up to 8ppm.

The yellow color in wheat is controlled by the Phytoene Synthase (Psy) gene and its alleles which produces the enzyme. Phytoene Synthase dimerizes two geranylgeranyl pyrophosphate molecules and is also the rate-limiting factor in the carotenoid biosynthesis pathway (Lindgren, 2003). The wheat Psy1 gene shows homology with the maize Psy1 gene and is associated with carotenoid accumulation (Palaisa et al., 2003) in the endosperm. In maize, it was mapped on chromosome 6 which shows synteny with chromosome 7 of bread wheat (Gale and Devos, 1998; Palaisa et al., 2003). In wheat, the gene is present in all three homologs of chromosome 7 (PsyA1, PsyD1, and PsyB1) (Zhang and Dubcovsky, 2008). The original 7E chromosomal segment from Th. ponticum was incorporated into the wheat cultivar “Agatha” containing the Psy gene which was translocated into chromosomes 7A and 7D of wheat and was named the Y gene due to the undesired yellow coloration of the flour (Sharma and Knott, 1966). Plants heterozygous for the 7 EL segment showed 56% higher lutein content than the lines without the 7 EL segment (Zhang et al., 2005). Carotenoid (lutein) is the main component mediating the yellow coloration of wheat flour (Adom et al., 2003; Zhang et al., 2005). Further, the yellow color of the flour is desirable for yellow alkaline noodles consumed mostly in Japan and Southeast Asia (Kruger et al., 1992). Similarly, in North India, missi roti is consumed, by adding chickpea flour to wheat flour for added taste and colour is quite popular (Padhy et al., 2022b; Padhy, 2019).

In our experiment, the genes PsyE1+Lr19+Sr25 in HD2967 background with GpcB1+Yr36+ Yr15 in PBW550 background were converged and stacked together. After the selection of the desirable lines utilizing gene-specific markers, the resultant lines showed total rust resistance to both stripe and leaf rust, thus demonstrating the expression and functionality of the Lr19, Yr36, and Yr15 genes. Yr36 is the temperature-sensitive stripe rust resistance gene which confers enhanced resistance to stripe rust as the temperature increases. Combining one major stripe rust resistance gene (Yr15) which ensures completely clean plant foliage with temperature sensitive Yr36 gene takes care of stripe rust resistance. Further, Lr19, linked to the Y gene, partially covers leaf rust. The other minor/defeated leaf rust genes in several backgrounds may add to the complete resistance when combined with other genes. This kind of additive effect of minor defeated genes has been also shown to provide resistance in segregating generations of two susceptible cultivars who reported that in the two crosses (PBW621×PBW343 and HD2967×PBW343) the resistant segregants possessed two genes, one contributed by PBW621 or HD2967 (depending on the cross) and the other, unexpectedly but obviously, came from the most susceptible cultivar, PBW 343 (Sharma et al., 2022). Also, the gene Sr25, linked to the Y gene, makes a case for pre-emptive breeding for black rust in wheat.

An increased carotenoid content (up to 12ppm) in the segregating population was limited by a yield penalty, such as reducing the number of tillers in the lines than the checks. However, it was observed that the lines pyramiding all the desired genes had carotenoid content up to 9 ppm without any substantial yield penalty. An increased carotenoid content of 50%–120% was observed in these pyramided lines as compared to those previously reported by Zhang et al. (2005), which showed 56% higher lutein content than the lines without the 7 EL segment. The agronomic characterization of the translocated isogenic lines demonstrated that the presence of 7E was associated with high grain yield under irrigated conditions (Monneveux et al., 2003). After the characterization of the gene by Zhang and Dubcovsky, (2008), the gene was used positively for the first time in a durum background by Singh et al. (2014). Sun et al. (2014) found a significant negative correlation between carotenoid content and thousand-grain weight and a positive correlation between carotenoid content and grain length, grain length/grain width, and grain length/grain thickness. They also reported a negative correlation between carotenoid content and grain width and grain thickness. Similarly, Padhy A. K. et al. (2022) reported a significant negative correlation between carotenoid content and thousand-grain weight and spike length. Spikelets per spike and grains per spike were negatively correlated with carotenoid content but were insignificant. However, they reported a slight positive correlation between carotenoid content and tillers per meter row as well as yield. It implies that the Y gene can be effectively used for inbreeding wheat with high carotenoid content without any major yield penalty provided the population size is kept large so as not to lose out the rare recombinants combining high carotenoid content, high grain weight, and disease resistance.

Micronutrient deficiencies affect a large section of the world’s population. Biofortification is an evidence-based nutrition strategy to address some of the most common and preventable global micronutrient gaps and can help improve consumer health (Bouis and Saltzman, 2017; Foley et al., 2021). Enhancing the micronutrient concentration, specifically protein, iron, and zinc content in wheat via genetic fortification using modern plant breeding approaches has been exploited as the most effective and safe approach to alleviate micronutrient malnutrition by plant breeders. Further, a number of issues concerning the nutritional quality of wheat revolve around the seed phosphorus (P) storage compound called phytic acid (myo-inositol-1,2,3,4,5,6-hexakisphosphate) for which studies report genetic variability in available wheat germplasm (Sharma et al., 2022). Enhancing carotenoid content in wheat is a novel addition to biofortification components to be considered when breeding biofortified wheat cultivars (Padhy et al., 2022b). Lutein, the major component of carotenoids, is an antioxidant that reduces oxidative damage to biological membranes by scavenging peroxy-radicals, which are involved in certain human diseases and aging, as well as responsible for the degeneration of food quality. Therefore, the carotenoid pigments contribute to an increased nutritional value of wheat and wheat products (Bast et al., 1996). The PsyE1 gene is present in the distal region of the chromosome arm 7 EL from Th. ponticum increases carotenoids in the endosperm and is linked with the rust-resistant genes Lr19 and Sr25 (Zhang et al., 2005; Padhy et al., 2022a). The incorporation of the PsyE1 gene increases all the carotenoids that are produced in the carotenoid biosynthesis pathway. There is an increase in lutein, zeaxanthin, and β-carotein content due to the introgression of PsyE1 (Zhang et al., 2005; Ceoloni et al., 2017; Padhy A. K. et al., 2022). In a bread wheat background, despite the numerous health benefits of carotenoids, the gene was never used commercially as the yellow coloration of wheat flour lacked consumer acceptability and health-associated awareness. But recent trends and demand for biofortified quality wheat and wheat products have encouraged us to use the gene positively by analyzing its role in increased carotenoid content, agronomic effects, and potential health benefits (Padhy et al., 2022b).

Wheat breeding in combination with advances in biotechnology tools has made remarkable progress in increasing crop yields in the recent past. Wheat breeders must constantly respond to ever-emerging challenges, primarily, nutritional security, which defines the major mandate of a wheat breeder in the present scenario. Wheat breeders in India have always faced the endless task of continually developing new wheat varieties combining enhanced yield and disease resistance and have emerged successful, leading to self-sufficiency in wheat. However, in light of the malnourishment statistics of the country, the task of combining two or more nutritional components into agronomically elite backgrounds, and, if not enhancing, at least maintaining substantial yields represents an unprecedented challenge for wheat breeders. Varieties with improved nutritional quality, protein content, high grain yield, high carotenoid content, and desirable processing quality in adapted elite genetic backgrounds with tolerance to stresses and diseases can help alleviate nutrient deficiencies. Breeding wheat with enhanced levels of nutritional components is a cost-effective, sustainable solution to malnutrition problems. It is therefore paramount that suitable biofortified wheat varieties are developed, released, and disseminated for widespread adoption. Since grain nutrition is a non-visible trait, it is essential that new cultivars not only have enhanced nutritional components but also higher yield levels.

Shuttle breeding, started at CIMMYT, Mexico, was originally used to accelerate the breeding cycle by growing segregating generations in contrasting environments. This could ensure rapid generation advance by taking more than a single generation per year. Additional benefits of shuttle breeding were observed in terms of wider as well stringent selection due to exposure of the breeding material to variable disease spectrums, soil types, photoperiod lengths, and diverse abiotic stresses. Wheat breeders in North India, including Punjab Agricultural University located in India, Punjab, exploit the shuttle breeding concept and grow alternate wheat crops at Keylong, situated in the Himalayas. This has been highly effective in achieving the breeding objectives as the wheat after harvesting (end of April to first fortnight of May) from the main crop season is planted at an alternate location or commonly known as the off-season location at Keylong in the second fortnight of May. The crop is harvested from Keylong at the end of September and the seed is ready to be sown back in the main crop season in November. This cycle fits well for major breeding objectives focusing on grain yield. However, when the breeding is aimed at stacking together the various subcomponents of quality, this system poses a major challenge in terms of the time available for quality component analysis in the screening of the generations before the next sowings. Wheat grain quality primarily refers to two main components, namely the nutritional content (which may be for domestic or commercial use) and industrial processing or end product specificity. Enhancing wheat quality improves processing specificity, makes more desirable and more diverse consumer products and ensures the competitiveness of farmers, grain merchandisers, millers, and end processors. Wheat quality criteria may vary drastically depending on the end use. Usually, the conventional small-scale quality tests and/or marker-assisted selection (MAS) are done in the advanced breeding stages (F₆–F₈) or the final product. But combining together different components of quality definitely requires stringent screening for the traits under consideration from the F₂ generation onwards. Moreover, this testing must be based on more specific food processing (dough visco-elasticity and mixing properties, starch pasting properties, baking performance, etc.). Since the breeding for complex multigenic quality characters in wheat by biochemical analyses and functional pilot tests is traditionally a slow process, shuttle breeding has limited utility. Instead, rapid generation advance (RGA) under controlled conditions or doubled haploid production can be of much use. This explains the breeding scheme presented here which uses an off-season location for advancing F₁ to F₂ only and the other generations were evaluated for selection in the main season only.