A tuber mustard AP2/ERF transcription factor gene, BjABR1, functioning in abscisic acid and abiotic stress responses, and evolutionary trajectory of the ABR1 homologous genes in Brassica species

Plant growth and stress treatments to tuber mustard

Seeds of tuber mustard cultivar Yong’an were provided by Chongqing Fuling Institute of Agricultural Sciences, Chongqing, China. The seeds were sown in October and plants grew in their natural environments in Chongqing, China. The roots, stems, and leaf of 14-week-old (before the start of stem swelling) and 16-week-old (stem welling stage) tuber mustard were respectively harvested and frozen immediately in liquid nitrogen for total RNA preparation and semi-quantitative analysis. For ABA treatment, abscisic acid (ABA, 10 µM) were sprayed onto 1-month-old tuber mustard seedlings to ensure total coverage of the foliage area. The seedlings were collected respectively at 0, 2, 4, 6, and 12 h after ABA treatments with three biological replicates, and frozen immediately in liquid nitrogen for total RNA preparation and qRT-PCR analysis.

Cloning of BjABR1 promoter and sequence analysis

The tuber mustard genomic DNA were used for genome walking according to the described method previously (Li et al., 2002). The TAIL-PCR method and the Genome Walking kit (TaKaRa, Dalian, China) were used to perform genome walking to isolate the promoter region of BjABR1 gene following the manufacturer’s protocol. The random primers were improved following a previous study (Liu & Chen, 2007), and the three specific primers, SP1 (5′-AAGGAGGCGTGAGGGAGTT-3′), SP2 (5′-TCACGGTGGAGGTAGTCATC-3′), and SP3 (5′-TTCCTGATTTGCCACTTTT-3′), were designed as reverse primers according to the genomic DNA sequence of BjABR1. The PLANTCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) was used to identify potential cis-regulatory elements within the promoter.

Construction of the plant expression vector and generation of transgenic plants

To generate the plant transformation vector pCAMBIA1302::BjABR1, the cDNA fragment with the complete ORF of BjABR1 was synthesized from the total RNA using a pair of primers (sense primer: 5′- AGATCT (BalII) TGCGTGCCTTAAAAGTG -3′, reverse primer: 5′-GGTTACC (Bst EII) TTAAGAGGATGGGCTAT -3′). After validating its sequence, the PCR fragments were digested by BalII and Bst EII and ligated into the plasmid pCAMBIA1302 expression vectors which were also digested by BalII and Bst EII but removed the GFP gene. The pCAMBIA1302::BjABR1 constructs were then transferred into Agrobacterium tumefaciens GV3101 cells.

Arabidopsis thaliana (ecotype Col-0) seeds were sown on Murashige and Skoog (MS) medium containing 1% sucrose and 1% agar in a growth incubator. The growth conditions were as below: 23 °C, 75% relative humidity, and the photoperiod of 16 h light and 8 h darkness. The pCAMBIA1302::BjABR1 constructs were transferred into Arabidopsis plants using the floral dip method to obtain first-generation (T0) seeds. To obtain the homozygous, T0 seeds were planted on MS medium with 25 mg L⁻¹ hygromycin (Dingguo, Beijing, China) to obtain second-generation (T1) seeds. The T1 seeds were then planted on MS medium with 25 mg L⁻¹ hygromycin, and the transgenic lines with a segregation ratio of 3:1 (resistant:sensitive) were selected to be third-generation (T2) seeds. The T2 seeds were planted on MS medium with 25 mg L⁻¹ hygromycin. When the T3 lines displayed 100% hygromycin resistance, they were considered as homozygous and their seeds were harvested to use for following experiments.

Stress treatments to Arabidopsis

Approximately 100 seeds each from the wild-type Arabidopsis and transgenic line were sown on MS medium containing 100 mM NaCl, 200 mM NaCl, 200 mM mannitol, 400 mM mannitol, one µM ABA, or two µM ABA to test the germination (radicle emergence). Seeds were firstly incubated for 3 days at 4 °C in darkness to break seed dormancy, and then transferred into a growth incubator keeping 23 °C. The germination was scored daily for 9 days after transferring to 23 °C. For plant growth assays, seeds of wild-type Arabidopsis and transgenic lines were sown on MS medium to germinate (radicle emergence). The germinated seeds were then transferred to MS medium containing 100 mM NaCl, 200 mM NaCl, 200 mM mannitol, 400 mM mannitol, one μM ABA, or two μM ABA, and the seedlings were photographed on day 10 after transferring. Three biological replicates were performed.

For expression analysis of previously identified stress-inducible genes, RD29B and RD22, 3-week-old wild type and transgenic seedlings were sprayed with water (CK), 200 mM NaCl, and 400 mM mannitol to ensure total coverage of the foliage area, respectively. For expression analysis of previously identified ABA-inducible genes, ABI3, 3-week-old wild type and transgenic seedlings were also sprayed with water (CK) and 10 µM ABA, respectively. A total of 3 h after stress or ABA treatments, total RNAs of seedlings were then extracted, and qRT-PCR analyses were performed. Three biological replicates were performed.

Quantitative real-time PCR (qRT-PCR) validation

AMV RNA PCR Kit 3.0 (Takara, Kusatsu, Japan) was used to reverse-transcribe the total RNA according to the manufacturer’s protocol. SYBR Premix Ex Taq™ Kit (Takara, Kusatsu, Japan) was used to perform qRT-PCR reactions with primers for BjABR1, RD29B, RD22, ABI3, and 18S rRNA genes. The 18S rRNA gene was used as the internal control. The primer pairs are listed in Table S1. A Bio-Rad iQ5 realtime PCR detection was used to perform the qRT-PCR reactions. Three replicates per sample were examined.

Identification and phylogenetic analysis of ABR1 homologous genes in Brassica species

Five whole-genome sequenced Brassica species were used in the present study, including three diploid species (B. oleracea, B. rapa, and B. nigra) and two allotetraploid species (B. juncea and B. napus). The genomic information of B. rapa (version 2.0), B. nigra (version 1.1), B. juncea (version 1.1), and B. napus (version 5) was provided in the Brassica Database (BRDB, http://brassicadb.org/brad/index.php), while the genomic information of B. oleracea (version 2.1) was provided in the Ensembl Plants Database (http://plants.ensembl.org/Brassica_oleracea/Info/Index). The genomic information of A. thaliana was also provided in BRDB. The cloned protein sequence, intron sequence, and promoter sequence of BjABR1 gene was analyzed by BLAST searches against B. juncea database of BRDB. The amino acid sequence of AtABR1 (AT5g64750) gene was analyzed by BLAST searches against A. thaliana database of BRDB to confirm no other ABR1 homologous genes in A. thaliana. The amino acid sequence of AtABR1 gene was submitted as a query to identify ABR1 homologous genes in Brassica species, including B. nigra, B. rapa, B. oleracea, B. juncea, and B. napus, and the genes whose sequence similarity scores were greater than 300 were ABR1 homologous genes, while the other genes, whose sequence similarity scores were all less than 100 were not homologous to ABR1 gene.

The nucleotide or amino acid sequences of the ABR1 homologous genes were aligned using the software DANMAN 7.0 to view their nucleotide or amino acid sequences changes. Multiple alignment of amino acid sequences of the ABR1 homologous genes were performed together using ClustalX (Thompson, Higgins & Gibson, 1994) with default options and a nj format file containing genetic distance was generated. One minus the genetic distance was the similarity of pairwise sequences. A subsequent manual alignment correction was accomplished by using MEGA 5.1 (Tamura et al., 2011). Phylogenetic trees were constructed by means of the bootstrap neighbor-joining (NJ) method and a Kimura 2-parameter model that were provided by MEGA 5.05. The stability of internal nodes was assessed by bootstrap analysis with 1,000 replicates. The physical location data of exons and introns of ABR1 homologous genes were retrieved from the genomic information and the exon-intron structures were displayed by using the Gene Structure Display Server (GSDS, http://gsds.cbi.pku.edu.cn).

Semi-quantitative analysis of ABR1 homologous genes in tuber mustard

The spatial and temporal expression of six ABR1 homologous genes of tuber mustard, BjuA008532, BjuA008665, BjuA023050, BjuA032667, BjuB001787, and BjuB007684_correction, were examined using semi-quantitative method. AMV RNA PCR Kit 3.0 (Takara, Kusatsu, Japan) was used to reverse-transcribe the total RNAs of roots, stems, and leaf of 14-week-old and 16-week-old tuber mustard, respectively. To control for genomic DNA contamination, a reaction lacking reverse transcriptase was performed in parallel for each sample. PCR amplification of a 235 bp fragment of tuber mustard Actin3 gene (BjuA014894), using intron-flanking and gene-specific primers (Table S1), was controlled for the presence of equal amounts of cDNA template in each reaction. Gene-specific primers of BjuA008532, BjuA008665, BjuA023050, BjuA032667, BjuB001787, and BjuB007684_correction genes were designed according to specific region of nucleotide sequences based on the alignment result of nucleotide sequences among the ABR1 homologous genes. These gene-specific primers were listed in Table S1 and used to amplify the coding regions of BjuA008532 (286 bp), BjuA008665 (303 bp), BjuA023050 (324 bp), BjuA032667 (371 bp), BjuB001787 (228 bp), and BjuB007684_correction (417 bp) by PCR. The amplified products were also validated by sequencing. In the PCR reactions of six ABR1 homologous genes, three parameters, the annealing temperature, the extension time, and the quantities of templates were set according to Tm values of primers, the length of amplified segments, and equal amounts of cDNA template in each reaction adjusted by Actin3 gene, respectively, and the cycle numbers all are 30 times.

Cloning and sequence analysis of BjABR1 promoter

With reference to the genomic DNA sequence of BjABR1, a 1,212 bp sequence upstream of the coding region was isolated using the genome walking method. A promoter motif search was performed using the PLANTCARE database showed that a number of putative plant cis-elements were present. As shown in Fig. 1, some hormone-related elements were recognized, including the MeJA-responsive element (TGACG-motif), salicylic acid responsive element (TCA-element), and gibberellin-responsive element (GARE-motif); some stress-related elements were recognized, such as heat stress responsiveness element (HSE), low-temperature responsiveness element (LTR), defense and stress responsiveness element (TC-rich repeats), fungal elicitor responsive element (Box-W1), anaerobic induction element (ARE); light-responsive elements, such as the Box 4, ACE, Box I, G-box, Sp1, chs-CMA1a were observed; some development-related elements were also explored, including meristem specific activation element (CCGTCC-box), and cell cycle regulation element (MSA-like). The presence of these putative cis-acting elements indicates that the BjABR1 gene could function in hormone- and stress-related responses, and plant development.

Expression patterns of BjABR1 in response to ABA and abiotic stress in tuber mustard

The expression levels of BjABR1 under abiotic stress conditions were investigated in tuber mustard in a previous study. The expression of BjABR1 were strongly induced at 2 and 6 h after NaCl and Mannitol treatment, respectively, then decreased, and the expression of BjABR1 increased gradually with time under cold stress (Xiang et al., 2014). The expression of BjABR1 was also inducible by ABA. As shown in Fig. 2, the transcripts of BjABR1 increased rapidly and accumulated to the peak at 4 h after ABA treatment, then declined gradually. It is indicated that BjABR1 gene may function in responses to ABA and abiotic stress in tuber mustard.

Alleviative plant sensibility to ABA and abiotic stress in transgenic Arabidopsis compared to wild type

To further investigate the biological functions of BjABR1 gene, the transgenic Arabidopsis plants expressing BjABR1 gene were generated. The expression levels of BjABR1 in transgenic lines were examined using qRT-PCR. All transgenic lines constitutively expressed higher levels of BjABR1 transcript compared to the wild-type plants (Fig. 3), and three transgenic lines expressing more BjABR1 transcripts, TL2, TL4, and TL7, were used to the following study.

The transgenic plants under normal MS medium were indistinguishable from the wild-type Arabidopsis (Fig. 4), but they exhibited alleviative plant sensibility to ABA, osmotic stress, and salt stress compared to the wild type (Figs. 4 and 5). Firstly, the seed germination rates of transgenic lines were improved in different levels of ABA and stress conditions (Fig. 5). For example, at one µM ABA, there are about 60% of transgenic seeds germinating and about 40% of wild-type seeds germinating in the second day, and there are more than 90% of transgenic seeds germinating and less than 80% of wild-type seeds germinating in the fifth day. At two µM ABA, there are more than 50% of transgenic seeds germinating and less than 40% of wild-type seeds germinating in the second day, and there are more than 80% of transgenic seeds germinating and less than 70% of wild-type seeds germinating in the fifth day. At 100 mM NaCl, there are more than 60% of transgenic seeds germinating and less than 40% of wild-type seeds germinating in the second day, and there are about 90% of transgenic seeds germinating and about 70% of wild-type seeds germinating in the fifth day. At 200 mM NaCl, there are about 50% of transgenic seeds germinating and about 30% of wild-type seeds germinating in the second day, and there are about 80% of transgenic seeds germinating and less than 70% of wild-type seeds germinating in the fifth day. Similarly, the germination rate of transgenic seeds was greater than that of wild-type seeds on the MS medium with 200 mM or 400 mM mannitol. Secondly, the growth of transgenic seedlings on ABA and stress conditions was comparable to the wild type, showing that the wild-type plants were significantly more inhibited by ABA, osmotic and salt stress compared to the transgenic plants, and the transgenic plants distinctly had more and larger leaves and longer roots compared to WT under the same condition (Fig. 4). It is indicated that BjABR1 gene involved in responses to ABA and abiotic stresses.

Altered expression of stress/ABA-induced genes in transgenic Arabidopsis

To elucidate the molecular mechanism of BjABR1 action during the stress/ABA response, the expression of known stress/ABA-induced genes was analyzed in wild type and transgenic lines following NaCl, mannitol or exogenous ABA application. qRT-PCR analysis was performed to monitor the expression levels of two previously identified stress-inducible genes, RD29B and RD22, and one ABA-inducible gene, ABI3. Under normal growth conditions, transcript levels of RD29B, RD22, and ABI3 were comparable in both wild type and transgenic Arabidopsis (Fig. 6). The transcript levels of RD29B and RD22, however, were significantly higher in the transgenic lines compared to the wild type under salt and mannitol conditions (Figs. 6A and 6B), while activity of ABI3 gene was down-regulated in the transgenic lines (Fig. 6C). The result indicated that alleviative plant sensibility to abiotic stress in transgenic Arabidopsis compared to wild type may be due to the increased expression of the stress-induced genes correlating with stress tolerance. The ABI3 gene product is an important component of ABA signaling pathway (Giraudat et al., 1992) and overexpression of ABI3 in Arabidopsis conferred hypersensitivity to ABA (Lopez-Molina et al., 2002). Therefore, the alleviative plant sensibility of transgenic Arabidopsis to ABA may result from the reduced expression of ABI3 gene in the transgenic Arabidopsis.

Identification and phylogenetic analysis of ABR1 homologous genes

By BLASTP searches against B. juncea database of BRDB using the protein sequence of BjABR1 gene, the gene, named BjuA032667 in B. juncea database, shared the highest amino acid sequence similarity (about 99%) with BjABR1 (Fig. S1). The intron, and promoter sequences of BjABR1 gene cloned by us were also aligned with the corresponding sequences of BjuA032667 gene, respectively, showing that the intron and promoter sequences between the two genes are also highly similar, and there were only four and five nucleotide changes in their intron and promoter sequences, respectively (Fig. S2). It is believed that BjABR1 and BjuA032667 are the same gene and their sequence difference may result from the error of sequencing or PCR amplification, or nucleotide mutation or polymorphism. The BjuA032667 gene instead of BjABR1 was used to the following analysis.

By BLASTP searches against A. thaliana database of BRDB using the protein sequence of AtABR1 (AT5g64750) gene, the alignment score of AT5g64750 gene was 571, while the alignment scores of other genes were all less than 100, so there were no genes that were homologous to AtABR1 except itself in A. thaliana.

By BLASTP searches against B. juncea, B. rapa, B. nigra, and B. napus database of BRDB, and B. oleracea database in Ensembl Plants using the protein sequence of AtABR1 gene, three (named BniB045887-PA, BniB004262-PA, and BniB047045-PA), three (named Scaffold000071.53_Bra037794, Scaffold000004.823_Bra024325, and Scaffold000021.116_Bra031903), three (named Bo9g018330.1, Bo3g100930.1, and Bo2g166380.1), six (named BjuB001787, BjuB007684, BjuA023050, BjuA008665, BjuA008532, and BjuA032667), and five (named GSBRNA2T00079211001, GSBRNA2T00000822001, GSBRNA2T00047471001, GSBRNA2T00086262001, and GSBRNA2T00134741001) ABR1 homologous genes in B. nigra, B. rapa, B. oleracea, B. juncea, and B. napus were identified, respectively.

According to the location data of exons and introns of ABR1 homologous genes from genomic information, all ABR1 homologous genes had one intron (Fig. 7B) except BjuB007684 and GSBRNA2T00134741001 which had two introns (Fig. 8). Comparing the nucleotide sequence of BjuB007684 gene with that of its the closest genes, BniB045887-PA, the results showed that the first intron sequence, the first and second exon sequence of BjuB007684 were highly similar with the intron sequence, the first and second exon sequence of BniB045887-PA, respectively (Fig. S3), while the second intron and third exon sequence of BjuB007684 were highly similar with the partial sequence of the second exon and the downstream sequence of the coding region of BniB045887-PA gene (Fig. S4). Further sequence alignment analysis showed that the termination codon (TGA) was present in BjuB007684 in the same location where the termination codon of BniB045887-PA in (Fig. 8A), indicating that the second intron and the third exon of BjuB007684 may be wrongly annotated. Then, the gene structure and the termination codon of BjuB007684 were corrected as that of BjuB007684_correction according to the nucleotide sequence of BniB045887-PA (Fig. 8A).

Comparing the nucleotide sequence of GSBRNA2T00134741001 gene with that of its the closest genes, Bo9g018330.1, the results showed that the second intron sequence, the second and third exon sequence of GSBRNA2T00134741001 were highly similar with the intron sequence, the first and second exon sequence of Bo9g018330.1, respectively (Fig. S5), while the first exon sequence and first intron sequence of GSBRNA2T00134741001 were highly similar with the upstream sequence of the coding region of Bo9g018330.1 gene (Fig. S6). Further sequence alignment showed that the initiation codon of Bo9g018330.1 gene is ATG, while the corresponding position of GSBRNA2T00134741001 gene was GTG (Figs. S5B and S6B), indicating that nucleotide error in GSBRNA2T00134741001 sequence may be present. Then, the gene structure and the initiation codon of GSBRNA2T00134741001 were corrected as that of GSBRNA2T00134741001_correction according to the nucleotide sequence of Bo9g018330.1 (Fig. 8B).

The amino acid sequences of ABR1 homologous genes of Brassica species and A. thaliana were aligned and the results showed that they shared high sequence similarity, especially the AP2/ERF domain (Fig. 9).

The phylogenetic tree was constructed on the basis of NJ method with a bootstrap value of 1,000 using MEGA5.1. The phylogenetic tree revealed that the ABR1 homologous genes of Brassica species were divided into three groups, group1, group 2, and group3, showed, respectively by red, blue, and green branches in Fig. 7A. Each diploid Brassica species had three ABR1 homologous genes (nine genes in total), and the nine genes were divided into three group and each group contained three genes which belonged, respectively to three diploid Brassica (Fig. 7A), indicating that the ancestor species of three diploid Brassica also had three ABR1 homologous genes which were all then inherited to each diploid Brassica. The genes of B. rapa and B. oleracea, Bo3g100930.1 and Scaffold000004.823_Bra024325, Bo9g018330.1 and Scaffold000071.53_Bra037794, and Bo2g166380.1 and Scaffold000021.116_Bra031903 shared closer relationship compared to BniB004262-PA, BniB045887-PA, and BniB047045-PA of B. nigra, respectively (Fig. 7A), and the amino acid sequence similarity percentage of ABR1 homologous genes between B. rapa and B. oleracea were much greater than that between B. rapa and B. nigra, and between B. oleracea and B. nigra (Fig. 7D), indicating that the ABR1 homologous genes may have experienced more rapid evolution in B. nigra species.

The two allotetraploid Brassica species, B. juncea and B. napus, have six and five ABR1 homologous genes, respectively. B. juncea was originated by hybridization between diploid B. rapa and diploid B. nigra, while B. napus was originated by hybridization between diploid B. rapa and diploid B. oleracea. The phylogenetic tree revealed that B. juncea inherited all the six ABR1 homologous genes from B. rapa and B. nigra with no gene lost, while B. napus inherited all the three ABR1 homologous genes from B. oleracea and two ABR1 homologous genes from B. rapa with one gene (Scaffold000071.53_Bra037794) lost (Fig. 7A). Furthermore, the intron sequences, and upstream and downstream sequence of the coding sequence of the ABR1 homologous genes in the same group were analyzed by pairwise sequence alignment (data not shown) and the results of sequence similarity among these genes were consistent with the gene relationships shown in the phylogenetic tree. Therefore, the evolutionary trajectory of ABR1 homologous genes in Brassica species can be summarized that the ancestor of B. rapa, B. nigra, and B. oleracea had three ABR1 homologous genes, and B. rapa, B. nigra, and B. oleracea inherited all the three ABR1 homologous genes from the ancestor species, respectively, and then, B. juncea inherited all the six ABR1 homologous genes from B. rapa and B. nigra with no gene lost, while B. napus inherited all three ABR1 homologous genes from B. oleracea and two ABR1 homologous genes from B. rapa with one gene lost (Fig. 7C).

Expression analysis of ABR1 homologous genes in tuber mustard

To understand the roles of six ABR1 homologous genes, BjuA008532, BjuA008665, BjuA023050, BjuA032667, BjuB001787, and BjuB007684_correction, in growth and development of tuber mustard, their spatial and temporal expression in roots, stems, leaves before and after the start of stem swelling were examined using semi-quantitative method, respectively. The analysis revealed that the six ABR1 homologous genes possessed different expression models (Fig. 10). BjuA008665, BjuA008532, BjuA032667 were more clearly expressed in samples compared to BjuB001787, BjuA023050, and BjuB007684_correction, while the expression levels of each gene in six samples were different (Fig. 10). For example, BjuA032667 was expressed in all samples, especially in the roots in stem swelling stage (Fig. 10). It is indicated that the ABR1 homologous genes may play a role in different tissues in different growing stages.