The Yanzhan 4110 (followed as Yanzhan) and natural dwarf mutants of Yanzhan 4110 (followed as Yanzhan mutation) provided by Dong Puhui’s research group at Henan University of Science and Technology (HAUST), were crossed to produce F_2:3 and F_3:4 populations, including 15 dwarf lines and 15 tall lines. Field trials were conducted in 2019-2020 and 2020-2021 at the experimental farm of the Institute of Water-Saving Agriculture for Arid Areas of China, Northwest A&F University, Yangling, Shaanxi, China. The experimental design was randomized complete block design (RCBD), and each line was planted in 2.5m² sized plots (five rows of 2 m long, 25 cm between two rows, and 6.7 cm between two plants). Each plot was repeated three times.

The GA₃ treatment was applied separately on both Yanzhan and Yanzhan mutation, and repeated three times. When wheat plants reached at stem elongation (Z31), early heading (Z51), early flowering (Z61), and grain filling (Z71) stages, GA₃ solution (100 μM) was sprayed on the surfaces of leaves, culms and young panicles (Zadoks et al., 1974), and the control was applied with distilled water of the same volume. GA₃ treatments in field were performed referred to Duan et al. 2020a.

Genomic DNA was extracted by CTAB method (Clark, 1997). The genomic sequence of Rht-B1b was cloned with primers in Supplementary Table 1 , using PCR conditions described. PCR was performed in a Peltier Thermal cycler with a 20μl volume used ApexHF HS DNA Polymerase FL (Accurate Biology, China), and the PCR program was set according to the instructions (temperature was 60°C, extension time was 3min). PCR products were distinguished by 2% agarose gel electrophoresis.

The PCR target band was purified using Universal DNA Purification Kit (Tiangen, China). After adding ‘A’ base at the end, the recovered product was cloned into the T vector (TAKARA pMD18-T Vector Cloning Kit). The vector was transferred into E. coli DH5α competent cells by the 42°C heat shock method, and 950μl of SOC liquid medium was added to the 37°C shaking culture for 45 minutes. The product was plated in AMP-containing LB medium. After overnight incubation at 37°C, individual colonies were detected and sequenced. Primer synthesis and clone sequencing were performed at Tsingke Biology Technique (Beijing, China).

Yanzhan and Yanzhan mutation F₂ population used to create tall and dwarf mixed pools. Genomic DNA was extracted from the second leaf below the spike of 20 tall and 20 dwarf plants at the filling stage. Equal amounts of tall plant genomic DNA were mixed into tall mixed pools, and equal amounts of dwarf plants genomic DNA were mixed into dwarf mixed pools. Combined with 660K chip (Affymetrix), SNP typing was detected (CapitalBio Technology, Beijing, China).

Plant height and the internode length were investigated at the mature stage. Internode diameter and wall thickness were measured with Vernier scale during the grain filling stage (Du et al., 2018), with five replicates.

With reference to the method of Zhao et al. (2021), the internodes strength tester was used to measure the mechanical strength of the 4th and 5th basal internodes. The center of gravity height and fresh weight of the main stem were also measured. The formula for calculating the lodging resistance index is as follows:

LRI: Lodging Resistance Index; MS: Mechanical Strength (N); CGH: Center of Gravity Height (m); FW: Fresh Weight(g).

The standard and method for yield characters were referred to “Physiological Breeding II: A field Guide to Wheat Phenotyping” (Reynolds, 2012). At maturity, tillers plant^-1, spike length, spikelet number spike^-1, grains number spike^-1, 1000-kernel weight, biomass plant^-1, yield plant^-1 and harvest index were measured using a random sample of 20 individual plants.

After threshing and drying, grain length, width and diameter were measured using the SC-G automatic seed analyzer (Wanshen, China). Grain quality traits, including water content, crude protein, wet gluten, water absorption, starch, stable time and formation time were measured using a Diode Array 7250 near-infrared reflectance (NIR) spectrometer (Perten Instrument AB, Sweden).

The formula for the difference between dwarf and tall plants as follow:

Data was analyzed using analysis of variance (ANOVA) for a t-test of the difference between the dwarf and tall lines using IBM SPSS Statistics 23.0. * means significant at p< 0.05. ** means significant at p< 0.01, none means not significant at p > 0.05. Oligo 7 was used for primer design. Origin 2018 and Microsoft Office Excel 2016 were used to complete the figures and tables.

Rht-yz significantly reduced plant height. The height of Yanzhan was approximately 75.4 cm, and the height of the Yanzhan mutant carrying Rht-yz was reduced by 22.5 cm (29.8%). In addition, Rht-yz reduced plant height by 36.4% and 33.4% in the F_2:3 and F_3:4 populations, respectively ( Figure 2 ). Rht-yz also had significant effects on internode length, diameter and thickness in wheat. Both Yanzhan and Yanzhan mutations had 5 internodes, and the Yanzhan mutations had significantly shorter internode lengths compared with Yanzhan ( Figures 3A, B ), in which the peduncle was significantly shortened by 8.0 cm (31.3%), I2, I3, I4 and I5 were significantly shortened by 3.9 cm (25.8%), 4.5 cm (38.8%), 2.7 cm (36.0%) and 4.0 cm (56.3%), respectively. In addition, compared with Yanzhan the diameters of I2, I3, I4, and I5 with Yanzhan mutations were significantly increased by 0.4 mm (9.8%), 0.4 mm (9.1%), 0.3 mm (7.0%), and 0.2 mm (4.4%), respectively, the thickness of I3, I4 and I5 was significantly increased by 0.2 mm (33.3%), 0.2 mm (25.0%) and 0.3 mm (37.5%) respectively (Figure 3C and Table 3 ).

Similar results were observed in the F_2:3 and F_3:4 populations, where Rht-yz significantly reduced plant height and internode length. Rht-yz also significantly increased the diameter of I2 and I5 and the wall thickness of I3, I4 and I5 ( Table 3 ).

This study focused on the lodging resistance index of the basal stem (I4 and I5), which are most relevant for lodging resistance traits ( Figure 4 ). The results showed that Rht-yz significantly increased the mechanical strength of Yanzhan mutants I4 and I5 by 7.4% and 15.5% respectively. Meanwhile, Rht-yz significantly increased the lodging resistance index of I4 (72.7%) and I5 (90.6%) ( Table 4 ).

In the parental materials, F_2:3 and F_3:4 populations, Rht-yz significantly reduced the tillers plant^-1, biomass plant^-1 and yield plant^-1. In addition, 1000-kernel weight and harvest index of the Yanzhan mutant and Rht-yz dwarf lines were also significantly reduced in the F_2:3 population. For spike traits, Rht-yz increased spike length, spikelet number spike^-1 and grain number spike^-1, but the difference was not significant ( Table 5 ).

Grain length, width and diameter of the Yanzhan mutation were not significantly different from those of Yanzhan, but Rht-yz reduced grain width by 6.5% and also had a negative effect on grain plumpness ( Figure 5 ).

Rht-yz significantly increased grain crude protein (12.8%), wet gluten (10.0%), starch (6.9%), dough stability time (70.9%) and dough developing time (7.4%), but had no significant effect on water content and water absorption ( Table 6 ).

Rht-B1a is wild-type and encodes a normal DELLA protein. The Rht-B1b and Rht-D1b have single base mutations in the coding region of the DELLA protein (Rht-B1b has a C-T mutation and Rht-D1b has a G-T mutation in the coding region), which ultimately leads to inactivity of the DELLA protein and reduced plant height (Peng et al., 1999; Pearce et al., 2011). The Rht-B1c transcript has a 90bp insertion, while Rht-B1d and Rht-B1e are prematurely terminated by single base (AT) mutations (Pearce et al., 2011), while Rht-B1k has displacement mutations in the VHIID domain to produce abnormal DELLA proteins. In this study, it was found that the number of differential SNPs enriched in the chromosome 4B was higher (36) and that there were 28 differential SNPs in the 30-33 MB interval, forming a differential SNPs enrichment peak. However, the number of differential SNPs was less and the distribution was scattered on other chromosomes, which did not form an obvious differential SNPs enrichment peak. Therefore, we can only obtain one interval (4B: 30-33MB) that regulates plant height, which contains a dwarf gene, Rht-B1b. The mutation site of the Rht-B1b did not differ between the Yanzhan and Yanzhan mutations. Therefore, Rht-yz is a new dwarfing site that regulates plant height and its mechanism of action is different from that of Rht-B1b (C-T mutation), but it is not clear whether it affects the Rht-B1a (TraesCS4B02G043100) or other genes.

Improving lodging resistance in wheat is one of the most important factors in increasing wheat yields. Many pre-Green Revolution wheat varieties were replaced due to lodging resistance defects. The wheat dwarf genes Rht-B1b and Rht-D1b reduced plant height by 23%, Rht-B1b+Rht-D1b reduced plant height by 50% (Butler et al., 2005). The dwarfing genes Rht8 and Rht12 reduced plant height by 8.0% and 40%, respectively (Tang et al., 2009; Chen et al., 2013). In this study, Rht-yz reduced plant height by approximately 29.8%, which was a moderate strength effect.

Studies have shown that the length of I4 and I5 is significantly correlated with lodging resistance in wheat (Kashiwagi et al., 2005), and lodging resistance is positively correlated with internode diameter, thickness and fullness (Zhao et al., 2021). Studies have shown that Rht18 increased basal internode strength (Yang et al., 2015), while Rht13 decreased basal internode diameter (Wang et al., 2014). In addition, the height of center of gravity is also an important factor in the lodging resistance index. In this study, Rht-yz mainly improved the lodging resistance by reducing the plant height and increasing the diameter, wall thickness and mechanical strength of the basal internode, which has important application value for dwarf breeding in wheat.

The study reported that Rht3 increased tiller number (Zhang et al., 1995), Rht5 reduced spikelet number and grain number spike^-1 in two different genetic backgrounds (Daoura et al., 2014), and Rht8 had little effect on yield traits (Rebetzke et al., 2012). In this study, Rht-yz had a significant negative effect on tiller number plant^-1, biomass plant^-1 and yield plant^-1, but no significant effect on harvest index.

An unexpected disease (wheat stripe rust) during the 2019-2020 growing season led to a yield reduction in the F_2:3 population. During the 2020-2021 growing season, no signs of disease were observed. Therefore, the effect of F_3:4 on yield traits is closer to the actual effect of Rht-yz. The growth period of the Yanzhan mutation was 5-7 days later than that of Yanzhan, suggesting that Rht-yz may be involved in regulating the growth and development of wheat. Studies have also shown that Rht4+Rht-B1b has higher yield and harvest index under different moisture conditions (Liu et al., 2017). Rht5 reduced wheat yield, but Ppd-D1 effectively eliminated the negative effects of Rht5 on yield and growth by advancing the growth period (Chen et al., 2018). These studies remind us that the negative effects of Rht-yz on yield can also be eliminated by introducing other new genes.

Protein content, starch content and gluten content in wheat have an important impact on the promotion and application of wheat varieties. The quality of flour is mainly determined by protein and starch content (Ahmed, 2015). The dwarf gene Rht-B1c significantly increased the protein content, Rht14 had a negative effect on quality traits (Duan et al., 2020b), and Rht15 increased the protein and wet gluten content (Zhao et al., 2021). In this study, Rht-yz significantly increased crude protein, wet gluten and starch content. Therefore, Rht-yz is of great significance for improving the grain quality in wheat.