2.1 Pearl millet genome retrieval and primer designing

The rice G-box binding protein containing the B12D domain from Oryza sativa #OsJ_20800 in the STRING database (https://string-db.org/) was used as a query in the NCBI tBLASTn (BLAST: Basic Local Alignment Search Tool (nih.gov) against the Whole Genome Shotgun Contigs database, limited by the Pennisetum glaucum genome. The PgB12Dg primers were designed based on the similarity between the genomic region of P. glaucum from chromosome 2 (#LKME02052028.1) from nucleotides 114,581,452 to 114,582,550 and mRNA sequence #CP050798.1 from Setaria viridis. Of the tested primer pairs, the primer pair F: 5′-CAGCAGCACAAGATCGAAAA-3′ and R: 5′- CTCGGAGAAGAAGCGGTTGAGG-3′ were found to amplify the cDNA of PgB12Dg successfully.

2.2 cDNA amplification and sequencing

The local pearl millet (P. glaucum subsp. monodii (Maire) Brunken) variety used in this study is under registration #1316 in the Centre of Genetic Resource in Saudi Arabia. This local variety is grown during the summer season in Saudi Arabia and is known to be tolerant of extreme heat. Total RNA was extracted using TRI reagent (Sigma) on seedlings. The cDNA was synthesized with SensiFAST cDNA Synthesis (Bioline). PCR was performed using High-fidelity PCR Master (Sigma), 0.3 µmol of the designed primers, and 4 µL of cDNA in a total reaction of 20 µL. PCR program started with denaturation at 94 °C for 2 min; then 30 cycles of 94 °C for 15 s, 57 °C for 25 s, and 72 °C for 30 s; and final elongation at 72 °C for 5 min. The PCR product was sequenced by the Sanger method (Sanger et al., 1977) using a BigDye Terminator v3.1. Resulted sequences were assembled using MUSCULE (EMBL-EBI), and the consensus sequence was submitted to in silico investigation.

2.3 In silico analysis of the deduced protein

Translation of the open reading frame in the sequence of cDNA and prediction of the isoelectric point (PI) and molecular weight were performed by ExPASy (https://www.expasy.org/). Multiple sequence alignment was conducted for the deduced protein with its orthologues from monocots imported from the NCBI database using MUSCULE (EMBL-EBI). Accession numbers for the aligned proteins are as follows: Saccharum (AFO59572.1), Panicum hallii (XP_025813718.1), Zea mays (NP_001148351.1), Eragrostis curvula (TVU11840.1), Panicum virgatum (XP_039842942.1), Setaria italica (XP_004965125.1), Digitaria exilis (KAF8641711.1), Sorghum bicolor (XP_002438146.1), Setaria viridis (XP_034589112.1), O. sativa (XP_015643421.1), Oryza brachyantha (XP_006655963.1), Oryza meyeriana (KAF0903233.1), Brachypodium distachyon (XP_003564009.1), Aegilops tauschii (XP_020187542.1), and Hordeum vulgare (KAE8785662.1).

Tools from the PSIPRED website (Nugent and Jones, 2013) were used to predict the secondary and tertiary structures of the deduced protein. The molecular function and cellular components of the deduced proteins were predicted by FFPred 3, and the MEMEMBED algorithm predicted the topology of the transmembrane helices. Domains in PgB12Dg were predicted by InterPro (https://www.ebi.ac.uk/interpro/). Subcellular localization of the deduced protein was conducted using the LocTree2 server https://rostlab.org/services/loctree2/ (Goldberg et al., 2012). Functional protein association for the deduced PgB12Dg protein was predicted by the STRING database using the ortholog protein from O. sativa.

2.4 Plant treatments and gene expression analysis

Germinated seeds were treated by plant hormones and three types of abiotic stress to examine PgB12Dg expression using real-time PCR. The expression of PgB12Dg was measured in the three-day-old seedlings' root, shoot, and seeds under three types of stress; drought, salt, and cold. Pearl millet seeds were surface-sterilized by 5 % sodium hypochlorite for 30 min and washed thrice before germinating. Seven groups of seeds were germinated in Petri dishes at room temperature.; four groups for stress treatment and their control and three other groups for hormones treatment with their control. Their replicates were germinated in each group for real-time PCR biological replicates. The conditions of the stress treatments have been determined based on an initial gradient experiment for each stress type. The tested temperatures for cold stress were 2, 5, 8, and 12 °C. The tested sodium chloride concentrations for salt stress were 100, 150, 200, and 250 mM. These gradient conditions were applied to 3-day-old seedlings for 24 h. The stress conditions affecting growth below optimum levels were selected for plant treatments. Stress treatment was performed by exposing seedlings separately to each type of stress as follows; drought stress was induced by withholding water and leaving the plants in open dishes for 24 h in a relative humidity of 13 %. Seedlings were treated with 200 mM sodium chloride for 6 h to induce salt stress. Another group of seedlings were incubated in the dark at 10 °C for 6 h to induce cold stress. The expression of PgB12Dg in untreated seedlings was used as a control for stressed seedlings. The expression of PgB12Dg was analyzed in separated tissues of the seedlings: root, shoot, and seed. For the hormones' treatment, the seeds were soaked in H2O for 24 h and then divided into three groups. Two groups were soaked for 6 h in 1 mg/ml of ABA and GA, respectively, while the third group of seeds was left in H2O as a control group.

Total RNA was extracted from the treated samples and controls, and then cDNAs were amplified using SensiFAST cDNA Synthesis (Bioline). PgB12Dg-specific primers were designed to span the second and the third exons (F: 5′- CAAGCGACCGGACGCCGA-3′, R: 5′-TCAACAGCCAAGCAAAATGC-3′). Expression of the control gene Elongation factor-1 alpha (primers F: 5′-GTTACAACCCAGACAAGATTGC-3′ and R: 5′-TGGACCTCTCAATCGTGTTG-3′) was used to normalize the PgB12Dg expression in all samples. A real-time PCR reaction was performed with 12.5 µL of Power SYBR Green Master Mix from Thermo Fisher Scientific, 2 µL of a cDNA template, and 0.1 µM of each primer in a reaction of 25 µL. The initial cycle was 50 °C for 2 min, then 95 °C for 10 min. The 55 synthesis cycles were 15 s at 95 °C and 1 min at 60 °C. For expression assessment, cycles to threshold (Ct) values were calculated according to Livak and Schmittgen (2001). The Ct values of PgB12Dg in untreated tissues were used as calibrators to obtain ddCt and 2^-ddCt values.

3.1 PgB12Dg cDNA sequencing and characterization

The obtained cDNA sequence of PgB12Dg consisted of 314 base pairs and had an open reading frame of 261 base pairs encoded for 87 amino acids. The cDNA sequence of PgB12Dg is available in NCBI GenBank under accession numbers MW092093 and QQO98583 for the cDNA sequence and the deduced protein, respectively. The molecular weight for the deduced protein was 9.8 K_D, and the PI was 9.74. An NCBI BLASTn search against Whole Genome Shotgun Contig revealed a single hit in chromosome 2 (#LKME02052028.1) from nucleotides 114,581,452 to 114,582,550, which shared 92 to 100 % identity with the PgB12Dg cDNA. Three exons were detected in the PgB12Dg cDNA by aligning it with the sequence of the genomic region from 114,581,417 to 114,582,817 of chromosome 2. The length of the three exons were 52, 87, and 175 base pairs, respectively, whereas the introns were 232 and 865 base pairs, respectively (Supplementary Fig. S1).

3.2 The deduced PgB12Dg in silico analysis

The secondary and tertiary structures of the deduced PgB12Dg protein showed three alpha-helices, as illustrated in Fig. 1. InterPro domain prediction indicated the B12D domain #IPR010530 from 7 to 74 amino acids. Based on Panther prediction, the deduced protein appeared as B12D and G-box binding protein #PTHR33417:SF21. The transmembrane region embedded in the membrane started from 10 to 32 amino acids, based on TMHMM prediction. Additional non-cytoplasmic region (outside of the membrane) from 1 to 11 amino acids and cytoplasmic region (inside the membrane) from 30 to 87 were shown, based on Phobius prediction (Fig. 2A). High conservation was shown in the entire amino acids sequence of PgB12Dg protein. Its orthologues from monocots ranged from 86 to 89 amino acids in length (Fig. 2B). Only the homologous proteins that shared identity above 80 % were selected for multiple sequence alignment to avoid aligning other members from the B12D family.

Close

Fig. 1

(A) Secondary structure of the deduced PgB12Dg protein predicted by the PSIPRED website. Helices are shown in pink, and the confidence of prediction is indicated by the degree of color of the ‘Conf’ line. The dark color indicates high confidence, while the light color indicates the low confidence (B) The predicted tertiary structure of the PgB12Dg protein were generated by the DMPfold tool on the PSIPRED website. The N-terminal is on the blue side and the C-terminal is on the red side.

Export to PPT

Close

Fig. 2

(A) Domains in the deduced PgB12Dg protein as predicted by the InterPro. (B) Multiple sequence alignment for the PgB12Dg protein with 15 homologous proteins forms monocots from the NCBI. Only hits that shared more than 80% identity were selected for alignment. Accessions were designated with their scientific names, followed by their NCBI accession numbers. Alignment was conducted by MUSCULE and visualized by Jalview software.

Export to PPT

In silico subcellular localization indicated that the deduced PgB12Dg protein is a mitochondrial protein with high probability (99 %, Fig. 3A). The MEMEMBED topology indicated that the transmembrane helix in PgB12Dg protein spanned from 14 to 30 amino acids. As predicted by the three in silico tools together, LocTree2, InterPro, and MEMEMBED, the sequence from 10 to 32 amino acids of the deduced PgB12Dg protein appeared to be a transmembrane helix embedded in the inner mitochondrial membrane. Transmembrane helix topology showed that the N-terminus of the PgB12Dg protein was in the interspace between the double mitochondrial membranes and that the C-terminus was inside the mitochondrion (Fig. 3B).

Close

Fig. 3

(A) Subcellular localization of the deduced PgB12Dg protein as predicted by the LocTree2 server. (B) Topology of the PgB12Dg transmembrane protein in the membrane as predicted by the MEMEMBED algorithm.

Export to PPT

Regarding the cellular components prediction of the deduced proteins, PgGB-B12D by FFPred 3 indicated that PgB12Dg was an integral component of the mitochondrial inner membrane, indicated by gene ontology numbers GO:0005743 and GO:0016021 with probability 0.963 and 0.995, respectively. The biological process prediction showed that PgB12Dg was involved in ion transmembrane transport and in the regulation of the metabolic process (GO:0006810, GO: 0034220, and GO:0019222) with high reliability (0.899–0.851 probability value). Molecular function prediction indicated that PgB12Dg has transferase activity transferring glycosyl groups (GO: 0016757) with a probability of 0.922 and transmembrane transporter activity (GO:0022857) probability of value 0.879. Functional protein association for the deduced PgB12Dg protein was predicted by STRING, using the ortholog protein from O. sativa, the B12D-G-box binding protein #OsJ_20800 which was found to share 91.8 % similarity with the PgB12Dg protein. Seven sugar transmembrane transporters proteins: SWEET1A, SWEET1B, SWEET2A, SWEET2B, SWEET3A, SWEET3B, and SWEET4, were associated with PgB12Dg homologs in rice (Supplementary Fig. S2).

3.3 PgB12Dg expression profiling

Gene expression profiling revealed that PgB12Dg was downregulated (<0.5 folds to untreated seedlings) in shoots and seeds of seedlings stressed with drought, salt, and cold, except seeds under drought, which is considered non-significant downregulation (0.63 folds, Fig. 4). Under the same stresses, roots showed upregulation (greater than2 folds) except for roots under salt stress, which exhibited non-significant upregulation (1.81 folds). The highest expression for PgB12Dg was observed in roots under drought (3.33 folds), followed by roots under cold (2.08). Treatment of soaked seeds with plant hormones showed downregulation with ABA (0 folds) and upregulation with GA (2.35 folds).

Close

Fig. 4

The expression of the PgB12Dg in P. glaucum germinated seeds examined by real-time PCR. The expression levels of PgB12Dg in root, shoot, and seeds of the three-day-old germinated seedlings treated with drought, salt, and cold. RNA was extracted after 24 h of water withholding (drought) and after 6 h of treatment by cold and salt. The expressions of PgB12Dg in one-day-old seedlings after being treated for 6 h with ABA and GA are shown in the right of the chart. Elongation factor-1 alpha was used as reference gene in real-time PCR analysis. The expression level is indicated by the 2^-ddCt value along the y-axis.

Export to PPT