O-linked N-acetylglucosamine transferase is involved in fine regulation of flowering time in winter wheat

Vernalization genes underlying dramatic differences in flowering time between spring wheat and winter wheat have been studied extensively, but little is known about genes that regulate subtler differences in flowering time among winter wheat cultivars, which account for approximately 75% of wheat grown worldwide. Here, we identify a gene encoding an O-linked N-acetylglucosamine (O-GlcNAc) transferase (OGT) that differentiates heading date between winter wheat cultivars Duster and Billings. We clone this TaOGT1 gene from a quantitative trait locus (QTL) for heading date in a mapping population derived from these two bread wheat cultivars and analyzed in various environments. Transgenic complementation analysis shows that constitutive overexpression of TaOGT1b from Billings accelerates the heading of transgenic Duster plants. TaOGT1 is able to transfer an O-GlcNAc group to wheat protein TaGRP2. Our findings establish important roles for TaOGT1 in winter wheat in adaptation to global warming in the future climate scenarios.


Supplementary Method 1. Genetic backgrounds of Duster and Billings
Duster (PI 644016) and Billings (PI 656843) are two hard red winter wheat cultivars released by the Oklahoma Agricultural Experiment Station due to their wide adaptation across the southern Great Plains of the USA but with diverse genetic backgrounds [1,2]. A doubled-haploid (DH) population of 260 lines was generated from a cross of the two cultivars, and the DH population was used to map a gene on the short arm of chromosome 1A for resistance against Hessian fly [3], Lr34 on the short arm of chromosome 7D for resistance against leaf rust [4], as well as a yield gene on the short arm of chromosome 1B [5]. These previous studies validated the linkage groups of the DH population.

Supplementary Method 2. Positional cloning of QHd.osu-6A
The same set of Duster x Billings DH population was tested in a greenhouse with constant temperatures at 25/20ºC and with long days (16 h light/8 h darkness) in two experiments, one was the population without vernalization throughout the whole plant life, and the other was the population with vernalization. Vernalization was performed in a cold room with temperature at 2-6°C and the same long day as the greenhouse. The population at the 4 th -5 th leaf stage was moved into the cold room and returned to the same greenhouse after six weeks. Heading date was scored for each line of two populations tested with and without vernalization. The Duster x Billings DH lines were previously genotyped using genotyping-by-sequencing (GBS) markers, and a total of 2,358 GBS markers were identified and archived in the NCBI SRA (accession number SRP051982, https://www.ncbi.nlm.nih.gov/bioproject/PRJNA271346). A total of 148 GBS markers of linkage group 10 was integrated with heading date from the populations tested in the field and the greenhouse to construct QHd.osu-6A using WinQTLCart 2.5 (North Carolina State University, Raleigh). The sequences of the GBS markers allowed a physical distance covering QHd.osu-6A according to the recently released IWGSC RefSeq v1.0 databases [6]. Both Duster and Billings were identified to carry the same 2174 allele for each of vrn-A1b, PPD-D1b and vrn-D3b [7]. Two mutually exclusive hypotheses were designed to test if there was any new gene for heading date in the Duster x Billings population. If a QTL was associated with any of these three known genes, the QTL should reveal a new mechanism of the mapped known gene. If a QTL was not associated with any of these three known genes, the mapped gene/QTL should be new. In this study, QHd.osu-6A located on the short arm of chromosome 6A, in a region where was never reported to affect this trait; therefore, QHd.osu-6A was cloned using the positional cloning approach.

Supplementary Method 3. Quantitative RT-PCR
RNAs were extracted from leaf samples and performed using reverse transcription kit.
Quantitative real time polymerase chain reaction (qRT-PCR) was conducted using the SYBR Green PCR Master Mix, and actin was used as an endogenous control. qRT-PCR was carried out using a 7500 Real-time PCR System (Applied Biosystems. Foster City, CA) and iQ TM SYBR ® Green Supermix (Bio-Rad Laboratories. Hercules, CA), with actin used as endogenous control.
There is a three-step cycling program consisting of an initial denaturation step at 95ºC for 3 min, followed by 39 cycles at 95ºC for 15 s, 57ºC for 30 s, and 72ºC for 30 s. Primers used in qRT-PCR to amplify TaOGT1, vrn1, vrn3, PPD, TaVRT2, TaGRP2, and actin are listed in Supplementary Table 5.

Supplementary Method 4. MBP-TaVRN1 proteins
MBP-VRN1a protein from Jagger (aa1-180) and MBP-VRN1b from 2174 (aa 1-180) used in the EMSA were from a previous study [8]. The two cDNAs were respectively cloned into pMAL-c2 vector with an MBP-tag (New England Biolabs), and were expressed in the E.
coli (BL21). The cDNA of TaVRT2 (DQ022679) was cloned into pMAL-c2 vector with an MBP-tag (New England Biolabs) by using the primers TaVRT2-EcoRI-F1 and TaVRT2-BamHI-R1 (Supplementary Table 5). An amylose column (New England Biolabs) was used to purify the proteins fused with MBP-tag. Purified proteins were used in the EMSA.

TaOGT1
The cDNA of TaOGT1b was cloned in the BD vector using primers TaOGT1-EcoRI-F1 and TaOGT1-BamHI-R1 (Supplementary Table 5), and expressed TaOGT1b was used as a bait to screen a yeast-two hybrid (Y2H) library established using whole seedlings of winter wheat cultivar 2174. Positive clone sequencing results revealed two proteins TaK1 and TaK4 that might have interactions with TaOGT1b. The full length cDNAs of TaK1 (using primers TaK1-Y2H-NdeI-F1 and TaK1-Y2H-BamHI-R1), TaK4 (using primers TaK4-Y2H-NdeI-F1 and TaK4-Y2H-BamHI-R1), and TaGRP2 (using primers TaGRP2-Y2H-NdeI-F1 and TaGRP2-Y2H-BamHI-R1) were respectively cloned into the AD vector to confirm the interactions with in co-transformation of yeast cells. The sequences of the primers are provided in Supplementary   Table 5.