Registration of BTx623 dw5 , a New Sorghum Dwarf Mutant

The USDA-ARS has released a new dwarf sorghum [ Sorghum bicolor (L.) Moench) mutant BTx623 dw5 (

D warf genes have played a key role in plant breeding since the Green Revolution (Hedden, 2003;Khush, 2001;Peng et al., 1999).The reduced plant height due to the dwarf gene confers lodging resistance to plant varieties and allows the application of ample irrigation and fertilizers, which significantly boosts grain yield and successfully averted the forecasted famine due to increased population and reduced farmland availability in the 1970s (Khush, 2001).The Green Revolution began with wheat (Triticum aestivum L.) and rice (Oryza sativa L.), for which only one dwarf locus is required to breed semidwarf varieties.The reduced height gene 1 (rht1) in wheat is a dominant mutation in a gibberellic acid (GA) receptor (Ueguchi-Tanaka et al., 2005), whereas the dwarf locus used in rice is GA20 oxidase, a key gene involved in the biosynthesis of bioactive GAs (Sasaki et al., 2002).
Semidwarf is an important breeding trait in grain sorghum [Sorghum bicolor (L.) Moench], which contributes to the increased grain yield and allows the utilization of combine in grain harvesting.Four genetic loci, dw1, dw2, dw3, and dw4, have been used to control the plant height in sorghum (Quinby and Karper, 1954).None of these dwarf loci alone, however, can sufficiently reduce sorghum plant height to meet the requirement for machine harvesting.Therefore, breeders have traditionally incorporated three of the four dwarfing loci into grain sorghum breeding lines to produce semidwarf sorghum cultivars and two dwarf loci for breeding tall biomass sorghum.The Dw1 gene has been mapped to chromosome 9, Dw2 to chromosome 6, Dw3 to chromosome 7, and Dw4 to chromosome 4 (Li et al., 2015).The exact location of dw4 is still debatable, probably, on the lower arm of chromosome 4 (Li et al., 2015).For hybrid production, dwarfing alleles at these loci are required to introduce into both male and female parents.As a result, the semidwarf sorghum hybrids all contain multiple identical chromosome segments flanking the dwarf loci, causing homozygosity in large chromosome regions around these dwarf loci (Thurber et al., 2013).
The genes for three of the four dwarf loci in sorghum have been identified (Hilley et al., 2017;Hirano et al., 2017;Multani et al., 2003;Yamaguchi et al., 2016).However, unlike the ones in wheat and rice, none of these three dwarf genes is involved in the GA signaling or biosynthesis pathway.The Dw1 gene encodes a protein of unknown function, possibly involved in brassinolide signaling (Hirano et al., 2017;Yamaguchi et al., 2016).The Dw2 gene encodes a protein kinase (Hilley et al., 2017).The Dw3 encodes an auxin transporter (Multani et al., 2003).The Dw4 gene has been mapped to a chromosome region on chromosome 4, but the causal gene for dw4 locus has not been identified yet.Nevertheless, the mapped dw4 chromosome region does not contain any gene involved in GA signaling or GA biosynthesis pathway (Higgins et al., 2014;Upadhyaya et al., 2012;Zhang et al., 2015).Thus, there is a great need to identify new dwarf loci that enable breeding for semidwarf sorghum effectively using a single dwarf locus.
We isolated a number of dwarf mutants from a pedigreed sorghum mutant library in a leading inbred line BTx623 treated with ethyl methane sulfonate (EMS) (Jiao et al., 2016;Xin et al., 2008).Here, we report the characterization of the first dwarf mutant, BTx623 dw5 , (Reg.No. GS-787, PI 688506), which is genetically different from the four currently known controlling the dwarf phenotype in sorghum.The dwarf phenotype is caused by a recessive mutation in a single nuclear gene and may have potential use in breeding semidwarf sorghum cultivars.

Plant Material and Field Management
The dw5 mutant was isolated in 2015 from the M 4 mutant library of EMS-treated BTx623 seeds as described previously (Xin et al., 2008).Seed from a single dwarf plant was harvested and used for subsequent characterization.The dwarf phenotype of the mutant plants was confirmed in M 5 progenies and F 2 segregation population under field condition at the USDA-ARS Cropping Systems Research Laboratory in Lubbock, TX (33.58°N, 101.85°W, elevation 976 m).In 2017, the seed of original wild-type BTx623, the male sterile near-isogenic line (NIL) of BTx623 (BTx623 ms8 ) (Xin et al., 2018) used as female for backcrossing, the dw5, and backcrossed F 1s were planted in the winter nursery field in Guayanilla, Puerto Rico (18.0373°N, 66.7963° W, elevation 49 m), on 11 Dec. 2017.The plot size for all plantings at both Lubbock and Puerto Rico locations was 4.6 m long with 1.02 m row spacing.About 60 sorghum seeds per row were planted per plot, and the planting depth was about 3 cm.The major agronomic traits and plant height data were collected from six randomly selected plants per plot.Plant height was measured from the soil surface to the tip of the main panicle.Panicle length was measured from the first node to the tip of the panicle.Exsertion was measured from the flag leaf base to the first node of the panicle.The panicles from six plants were harvested at maturation and dried for 4 d in a forced air oven at 42°C before threshing.The 1000 seed weight was determined after drying the seeds for 2 d in 65°C oven.In 2018, the same lines along with backcrossed F 2 s were planted in Lubbock field on 11 May as three replicates per line, with two plots as a replicate for F 2 s (segregation population) and a single plot as a replicate for the rest of the lines (homozygous).Numbers of plants segregating for wild-type plant height and dwarf phenotypes were recorded.Data for agronomic traits were collected from six randomly selected plants per replicate the same way as those described above.For each replicate of F 2 s, data for major agronomic traits were collected from six randomly selected dwarf and six randomly selected wild-type height plants, respectively.Days to flower were recorded as the number of days from planting to mid-anthesis.Plant height was measured from the ground to the tip of the main panicle.Number of leaves was counted from the first green leaf (round) to the flag leaf.For accurate leaf number counting, the leaf number was marked on every oddnumbered leaf after it was fully expanded.
The experiment fields at both locations were managed by routine farm practices.One week before planting, all plots were furrow irrigated.The surface soil was gently loosed by a rolling cultivator prior to planting.Field irrigation started 2 wk after germination and ended at grain maturation.At the Lubbock location, plots were irrigated via an automated subsurface drip system at 3 mm per day, while the winter nursery plots at Puerto Rico were also watered daily via surface drip lines.At both locations, preplanting N-P-K fertilizer at a rate about 34 kg ha −1 (30 lb acre −1 ) was applied to the experiment field.In addition, nitrogen fertilizer at a rate about 90 kg ha −1 (80 lb acre −1 ) was applied at midseason through irrigation lines.Other activities such as weed control and pest control were managed by routine farm practices.

Genetic Characteristics
The dw5 mutant was backcrossed to the male sterile NIL of BTx623, BTx623 ms8 (Xin et al., 2018) and the backcrossed F 1 and F 2 populations were used to examine the genetic control of dwarf phenotype in dw5 mutant.The results of plant height segregation are provided in Table 1, and the plant height measurement data are provided in Table 2.The average height of wild-type parent (BTx623 ms8 ) minus two times of standard derivation and the average height of dw5 plus two times of standard derivation were used as a general guide, respectively, to classify wild-type and dwarf plants in the F 1 and F 2 populations.The plant height segregation ratio of F 1 plants at both locations was 1 to 0 of wild-type to dwarf (Table 1), indicating that the dwarf phenotype in dw5 is a recessive trait.The plant height data of F 2 population segregated for 218 wild-type plants to 69 dwarf plants, a ratio close to 3:1 with a Chi-square of 0.71, indicating the dwarf phenotype in dw5 is controlled by a single nuclear gene.Hence, we conclude that the dw5 contains a recessive mutation in a single nuclear gene.
The wild-type BTx623 used to generate the sorghum EMS mutant library is a "three-dwarf gene" inbred line that contains the recessive dwarf alleles at dw1, dw3, and dw4 loci but the tall allele at DW2 locus (Miller, 1977).The M 4 mutant line (25M2-0345) from which the dwarf mutant was isolated has already been sequenced and annotated for gene mutations by bioinformatic analysis (Jiao et al., 2016).We searched the mutant database and found that this newly isolated dwarf mutant line contains no mutation in the Dw2 or in the Dw3 and Dw1 genes and thus, represents a new dwarf locus.Therefore, this dwarf mutant was named dwarf 5 (dw5).
The dw5 mutant was characterized in two environments: the hot summer and long-day condition in Lubbock and the mild temperature and short-day conditions in the winter nursery in Puerto Rico, in 2017 and 2018.The dwarf phenotype was observed in both locations (Fig. 1).In addition, while the plant heights of the wildtype parent (BTx623 ms8 ) differed under shortday (range 138-161 cm, average 151.7, in Puerto Rico, 2017) and long-day (range 107-119, average 112.6 cm, in Lubbock, 2018) conditions, the heights of dw5 plants remained essentially the same (74.1 vs. 72.0,Table 2).The plant heights of dw5 parental line ranged from 66 to 81.5 cm among data collected at both locations.The results indicate that the dwarf phenotype of dw5 is stable under both long-and short-day environments as well as under variable temperature conditions.
The growth and development of dw5 mutant plants are almost identical to BTx623.The seed size and 1000 kernel weight of dw5 mutant are also similar to those of BTx623 (Table 2).However, the panicle length of dw5 mutant is slightly shorter than that of BTx623.This is likely due to the background mutations present in the dw5 since the mutant was isolated from a heavily mutagenized sorghum EMS mutant library (Jiao et al., 2016).Deep sequencing of the 256 M 4 sorghum mutant lines selected from this EMS mutant library has shown that on average, each M 4 mutant line contains about 7660 mutations (Jiao et al., 2016).Consequently, the dw5 mutant may contain many other mutations unrelated to dwarf phenotype.It is necessary to remove most of the background mutations that may affect its agronomic performance.Indeed, after backcrossing to BTx623 only once, the dw5 mutant displayed remarkable improvement in agronomic performance (Table 2).

Utility of the dw5 Mutants
The dwarf phenotype of the dw5 mutant provides a useful source for breeding dwarf sorghum hybrid.It contains a mutation in a new genetic locus from the three dwarf genes resided in BTx623 and differs from the genetic control of the dwarf phenotype of the known dwarf genes (Miller, 1977).Therefore, the dw5 mutation provides a new locus to breed semidwarf sorghum lines.