Overexpression of MdIAA9 confers high tolerance to osmotic stress in transgenic tobacco

Plant materials and treatments

Tissue-specific expression was monitored in the young roots, stems, fully expanded leaves, flowers, and mature fruits from 5-year-old apple plants that had been treated with bud grafting. The apple plant was the combination of a scion (Malus domestica Golden Delicious) and a stock plant (Malus hupehensis). To vary the treatments, 1 year-old plants (Golden Delicious scions and Malus hupehensis rootstocks) were grown in pots (diameter: 300 mm, height: 320 mm) in a greenhouse. One seedling was grown per pot and 60 pots were used per treatment. In order to achieve drought stress, irrigation for the seedlings was discontinued when the plant height reached approximately 0.5 m. The 9th–12th leaves were collected from the base of the plants when irrigation had been withheld for 0, 2, 4, 6, 8, and 10 days. A 200 mM sodium chloride (NaCl) solution was used to irrigate plants using a salt stress treatment and the samples were collected at 0, 1, 5, 10, 15, and 20 days after irrigation. For chilling treatments, plants were put into phytotrons (24 h-constant illumination and photosynthetic photon flux of 270 μmol·m⁻²·s⁻¹), the temperature was adjusted to 4 °C and the samples were collected at 0, 2, 4, 8, 12, and 24 h. For abscisic acid (ABA) and IAA treatments, plants were sprayed with 100 mM ABA or 10 µM IAA and the samples were collected at 0, 2, 4, 8, 12, and 24 h. The samples were frozen immediately in liquid nitrogen, then stored at −80 °C until RNA analysis occurred. The seedlings of the wild type (“NC89” tobacco) were used for genetic transformations. Transgenic plants and wild types were cultured in a growth chamber under a 16-h photoperiod at 23 °C. A total of 15 day old plants were grown on Murashige and Skoog (MS) agar medium supplemented with 0 or 200 mM of mannitol and left to grow for 13 day for the assay of osmotic stress. Seedling growth parameters, such as root lengths, fresh weights, and relative electrolyte leakage (REL), concentrations of chlorophyll, malondialdehyde (MDA), and proline were then measured. All of the experiments were repeated three times.

Identification of apple Aux/IAA genes

Annotated Aux/IAA open reading frame (ORF) translations from Arabidopsis were obtained by searching the TAIR website (TAIR10; http://www.arabidopsis.org/) (Dong et al., 2018a). A total of 34 Arabidopsis Aux/IAA sequences were used as queries in a homology search of annotated ORF translations from the Genome Database for Rosaceae (GDR; http://www.rosaceae.org/) (Tian et al., 2015) using the Basic Local Alignment Search Algorithm (https://www.rosaceae.org/blast). In addition, we searched the annotated apple ORF translations using a Hidden Markov Model-based approach (HMMER v3.0) and the Aux/IAA domain signature PF02309 as maintained by the Pfam database (http://pfam.xfam.org/family/PF02309). The presence of Aux/IAA domains was confirmed in each candidate MdIAA sequence using the Pfam database (http://pfam.sanger.ac.uk/search) and NCBI-Conserved Domain Search (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) Shao et al. (2014) (Dong et al., 2018b).

Multiple sequence alignments, phylogenetic analysis, and exon/intron organization of MdAux/IAA proteins

Multiple sequence alignments were performed for 34 MdIAA protein sequences using DNAMAN 6.0.3.99 with default parameters (Zhou et al., 2017). Full-length protein sequences of Arabidopsis and rice were obtained from the NCBI protein database. The phylogenetic trees were estimated with the MEGA6 program (Dong et al., 2018b) using the neighbor-joining method (Saitou & Nei, 1987) with Poisson corrections and 1,000 replications for the bootstrap analysis. Genomic sequences and location data in apples were observed from the apple genome database (https://www.rosaceae.org/organism/Malus/x-domestica; Genome version 1.0). The structural features of these MdIAA genes which included exons/introns, exon numbers, and locations were obtained and shown using TBTools.

Cloning of MdIAAs and gene expression analysis

Total RNA was extracted from frozen samples using the cetyl trimethyl ammonium bromide method (Wang et al., 2017a). Two µg of total RNA was used for synthesizing the first-strand cDNA. MdIAA ORF sequences were obtained using RT-PCR, with fully expanded leaves from the Golden Delicious apple as an RNA source, in addition to 17 gene-specific primer pairs as listed in Table 1. For quantitative real time PCR (qRT-PCR) assays, gene-specific primers were designed and synthesized by Sangon Biotech (Shanghai, China). One μg of total RNA from each sample was used to perform reverse transcription and one μL of the product was used for PCR-amplification. All reactions contained 10 µL of SYBR^® Premix Ex Taq™ (TaKaRa, Kusatsu, Japan), 0.4 µL of each specific primer, 8.2 µL of ddH₂O₂, and 1.0 µL of cDNA template, and the reactions were run by an iQ5 instrument (Bio-Rad, Hercules, CA, USA) (Dong et al., 2018a). The PCR conditions included an initial 95 °C for 3 min, then 40 cycles of 95 °C for 10 s, 58 °C for 30 s, and 72 °C for 15 s; this was followed by 72 °C for 3 min and then 81 cycles of 7 s each, increasing by an increment of 0.5 °C from 55 to 95 °C. Three biological replicates were conducted for each treatment and the values of ΔCt were calculated using the MdMDH gene or NtActin as the endogenous control (Wang et al., 2017b). The relative expression levels of MdIAA genes were obtained according to the 2^−ΔΔCt method (Livak & Schmittgen, 2001) and the specificity of the amplifications was examined by dissociation curve analysis.

Prediction of cis-acting elements in promoters

The upstream regions (1,500 bp upstream of the translational start codon) of the selected MdIAAs were used to identify putative cis-acting elements (Plant CARE database; http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Wang et al., 2017a).

Vector construction and plant transformation

In order to isolate the full-length cDNA of MdIAA9 used to construct the OE vectors, we conducted RT-PCR using fully expanded Golden Delicious apple leaves. The cDNA of MdIAA9 was cloned into pBI 121 vectors and was driven by a 35S promoter known as the cauliflower mosaic virus. For tobacco transformation, wild type (“NC89” tobacco) plants were transformed using the Agrobacterium tumefaciens EHA105-mediated leaf dip method (Xia et al., 2012). The PCR-positive plantlets were transplanted into soil for growing in the greenhouse. Seeds were screened with 50 mg L⁻¹ kanamycin after the transgenic plants were harvested individually. Homozygous transgenic T3 plants were used in osmotic stress investigations.

Measurement of physiological indices

Relative electrolyte leakage was determined according to the method described by Tan et al. (2017). Chlorophyll concentrations and free proline were measured using the method of Dong et al. (2018a), and MDA levels were measured as described by Wei et al. (2018).

Statistical analysis

The experimental data obtained was analyzed using SPSS 20 software (SPSS, Inc., Chicago, IL, USA) and indicated by means ± standard deviation. Data was analyzed using One-way ANOVA and Tukey’s tests at a significance level of P < 0.05.

Aux/IAA gene family members in apple

In order to identify AUX/IAA proteins in apples, 34 Arabidopsis Aux/IAA proteins were used to search the apple genome database by BLASTP. In addition, we carried out a Hidden Markov Model search based on the AUX/IAA domain signature. In all, 35 candidate genes were found. In order to further confirm and identify the conserved Aux/IAA domains, all candidate proteins were analyzed using Pfam and SMART databases. One gene, MD15G1305900, was not analyzed further because it contained an apparently incomplete reading frame. We obtained basic information for the remaining 34 Aux/IAA genes, which included gene names, locus IDs, intron numbers, ORF lengths, chromosome locations, and numbers of amino acids (Table 2). The ORF lengths of MdIAAs ranged from 498 base pairs (MdIAA19) to 1,158 base pairs (MdIAA34), with an average length of 789 base pairs. The size of MdIAA proteins were found to have a range of 165–385 amino acids. The corresponding molecular mass varied from 18.21 kD (MdIAA19) to 40.61 kD (MdIAA34).

Phylogenetic relationships of Aux/IAA proteins

A phylogenetic tree was constructed which included 34 members in apple, 34 members in Arabidopsis, and 31 members in rice. According to their phylogenetic relationships, Aux/IAA members of the three species could be split into nine subgroups, which were numbered I–IX (Fig. 1). The number of MdIAAs in the nine subgroups ranged from four to 17. The 34 MdIAAs were distributed unevenly among the subgroups. The numbers of MdIAAs in Subgroups I, II, III, IV, V, VI, VII, VIII, and IX were 4, 2, 4, 4, 1, 2, 6, 3, and 8, respectively.

Sequence alignment, gene structure, and conserved motifs of Aux/IAA proteins in apple

There were four conservative domains from a sequence alignment of the ORF translation of the Aux/IAA genes (Fig. 2). Nuclear localization signals were also found in most MdIAAs. Amino acids at five positions were absolutely conserved among all MdIAAs, but amino acids in 30 positions were strongly conserved (75–99%). The similarity between the other amino acids was <50%.

To explore the MdIAA gene’s structural diversity in apples, we evaluated the conservation of the exon-intron organization. The MdIAA proteins were divided into groups I and II according to their phylogenetic relationships (Fig. 3). The gene structure analysis suggested that in group I, although MdIAA18 had one exon and the other two genes (MdIAA9 and MdIAA33) had six exons, most genes had four exons. In group II, only MdIAA19 contained two exons and the remaining genes contained four exons.

We identified common motifs using the MEME web server (http://meme-suite.org/tools/meme) (Fig. 4). The following parameter settings were used: size distribution, zero or one occurrence per sequence (zoops); motifs count, 4; and motif width, between six and 50 wide. In our research, the four motifs identified by MEME were consistent with the four classical domains. This identified four motifs, which corresponded to the four conserved domains found through sequence alignment above. All four motifs were apparent in 22 of the proteins, although only three motifs were found in seven additional proteins. Motif I was not contained in MdIAA3, MdIAA4, MdIAA22, and MdIAA34; Motif IV was not in MdIAA12 and MdIAA4; and Motif II was not in MdIAA27. Three MdIAAs, MdIAA1, MdIAA25, and MdIAA32, contained only Motif II and Motif III, and two other MdIAAs, MdIAA18 and MdIAA19, contained only Motif II. It is obvious that most of the domains and motif regions overlapped with each other. Additionally, several highly conserved key sequences, such as LxLxLx in Domain I and Motif I, and VGWPP in Domain II and Motif II, strongly suggests that these regions may be the key sites conferring the functions of MdIAAs.

Cloning of apple IAA genes

To confirm the coding sequence of MdIAAs as annotated in the reference genome and to further uncover the biological function of these MdIAAs, 17 MdIAA genes were cloned: MdIAA4, MdIAA5, MdIAA8, MdIAA9, MdIAA10, MdIAA14, MdIAA15, MdIAA16, MdIAA20, MdIAA21, MdIAA24, MdIAA26, MdIAA27, MdIAA28, MdIAA30, MdIAA31, and MdIAA33, respectively. As a result, all cloned sequences showed a high similarity to the predicted apple genome (Data S1). The sequences were submitted to NCBI GenBank.

Analysis of promoter cis-elements of apple IAA genes

Promoter cis-regulatory elements are of great significance in regulating gene expression. In this study, the putative cis-elements within the MdIAA promoter (<1.5 kb upstream from the start site of putative translation) were identified through the PlantCARE database (Fig. 5) and different kinds of cis-elements related to stress and hormone responses were identified. Most MdIAA promoters contained ABRE elements, which usually participate in ABA related responses. Nine of the MdIAA genes contained a TGA-box, which is an auxin-responsive element. Six of the MdIAA genes contained gibberellins-responsive elements such as P-box, GARE-motif or TATC-box. A total of 11 of the MdIAA genes contained a G-box, CGTCA-motif, and/or TGACG-motif, which participate in JA/MeJA-responses. Five of the MdIAA genes contained a TCA-element, which is related to salicylic acid responsiveness. A large number of putative hormone-responsive elements suggested that these genes were involved in hormone signal transduction. Moreover, the promotors contained drought-inducibility elements (MBS), low-temperature response motifs (LTR), and defense and stress responsiveness elements (TC-rich repeats). The results indicated that MdIAA genes may be involved in defense signaling and biotic and abiotic stress in apple plants.

The spatio-specific profiles of MdIAA genes expression in various tissues

In order to determine the additional roles that MdIAAs play in growth and development, 17 selected MdIAA genes were examined in the roots, stems, leaves, flowers, and fruits using qRT-PCR. Some MdIAAs showed higher expressions in multiple tissues (Fig. 6). For instance, there were high levels of expression of MdIAA20 and MdIAA33 in all of the examined tissues. The expression of MdIAA20 was the highest in flowers and fruit while the expression of MdIAA33 was the highest in the roots, stems, and leaves. In addition, a higher transcriptional accumulation in the flowering and fruiting stages of MdIAA4, MdIAA9, and MdIAA24 were found, suggesting that they may play specific roles in the development of flowers and fruit. The mRNA abundances of the MdIAA family genes were higher in fruit than in other tissues. Furthermore, the expression of MdIAA genes in the stems was the lowest among all the tissues. These results indicated that MdIAA genes might play roles in fruit. Overall, the tissue-specific spatial differential expression of MdIAA suggests that these genes were specifically related to apple growth and development.

Expression profiles of MdIAA genes following IAA treatment

It is well known that auxin induces the strong expression of the Aux/IAA genes (Liu et al., 2017). qRT-PCR was performed to confirm the response of the MdIAA genes following IAA treatment. In this work, five testing time points (0, 2, 4, 8, 12, and 24 h) were selected. Under IAA treatment, the MdIAA family genes displayed various expression styles. The expressions of the majority of the MdIAA genes were up-regulated at 4 h as compared with controls (Fig. 7), but the expressions of most MdIAA genes were down-regulated at 8 and 12 h as compared with controls. No significant changes in gene expression were observed in other genes, such as MdIAA27, after IAA treatment. Taken together, this data indicated that most genes from the MdIAA family were auxin early-response genes.

The profiles of MdIAA gene expression following abiotic stresses and ABA spray

To investigate the effects of abiotic stresses and ABA treatment on MdIAA gene expression, the patterns of transcription were observed after drought, cold, NaCl, and ABA treatment (Fig. 8). Under drought stress, the expressions of MdIAA genes showed diverse responses at different time points (Fig. 8A). Among the 17 MdIAA selected genes, MdIAA4, MdIAA26, and MdIAA33 increased significantly. The expression of MdIAA4 was 5.15 times that of the control group at 8 days of drought treatment and the expression of MdIAA26 was 8.95 times that of the control group at 2 days and 5.50 times that at 4 days of drought treatment, respectively. The expression levels of MdIAA33 showed a trend similar to MdIAA26, being 7.50 times and 6.24 times that of the control group at 2 and 4 days, respectively. However, in this study, the expression levels of several MdIAA genes were significantly lower than those of the control group after drought treatment. For instance, the expression of MdIAA20 was only 12.7% of the control group at 8 d under drought treatment, while the expression of MdIAA24 was only 23.8% of the control group at 10 days under drought treatment.

Two genes showed an obvious response to the NaCl treatment (Fig. 8B). The expression of MdIAA24 was 6.82 and 4.49 times that of the control group at 1 and 5 days, respectively, but the expression of MdIAA28 was 3.22 times that of the control group at 15 days. However, some MdIAA gene expression was dramatically reduced after NaCl treatment. For example, the expression of MdIAA10 and MdIAA20 was only 14.1% and 17.7% of the control group, respectively.

In response to chilling at 4 °C, three MdIAA genes, MdIAA20, MdIAA21, and MdIAA27, increased significantly at 4 h of treatment (Fig. 8C). However, their expressions were then down-regulated over time. For example, at 8 h, MdIAA21 was only 10.82% of the control group. The other genes, such as MdIAA24 and MdIAA33, showed no alterations in expression under most conditions.

After ABA application (Fig. 8D), the expressions of MdIAA10, MdIAA14, MdIAA20, MdIAA21, and MdIAA30 decreased significantly. Among these genes, MdIAA30 changed most dramatically, and was only 7.05% of the control group at 8 h although some genes increased significantly during this time. For instance, the expression of MdIAA25 was 2.50 times that of the control group at 2 h whereas MdIAA8 was 2.42 times that of the control group at 12 h. Hence, these genes were designated as being ABA-responsive.

Overexpression of MdIAA9 enhanced plant growth of tobacco seedlings under osmotic stress

Under drought stress, MdIAA9 was significantly up-regulated, so we chose this gene to study its biological function in tobacco. Two transgenic lines were identified at the DNA level, using 35S promoter positive primers and the reverse primers of MdIAA9 (Table 2; Fig. 9). We examined the root length and fresh weights of transgenic and wild plants under osmotic stress. Our results showed that the root lengths and fresh weights of the transgenic plants were significantly higher than those of the wild types under osmotic stress. However, no significant difference was observed between the transgenic and wild plants under the control conditions (Fig. 10).

Analysis of physiological characteristics of MdIAA9-overexpressing tobacco seedlings and wild type to osmotic stress

To further study the enhancement of osmotic stress tolerance mediated by MdIAA9, the REL, chlorophyll concentration, proline, and MDA concentrations were determined. These are important markers to measure the degree of osmotic stress injury. Under normal conditions, no significant difference was observed between the transgenic and wild plants. However, the REL values of the wild lines were significantly higher than the overexpressing tobacco seedlings under 200 mM mannitol conditions (Fig. 11A). The chlorophyll levels of wild and overexpressed plants were similar under both the osmotic stress and normal conditions. The quantities of proline in transgenic plants were significantly higher than the wild line under 200 mM mannitol, but no significant differences were revealed under normal conditions between the transgenic and wild plants (Figs. 11B and 11C). Furthermore, the MDA content of the transgenic plants was significantly lower than the wild plants under the 200 mM mannitol conditions, while no significant difference was observed in the transgenic lines and wild types under normal conditions (Fig. 8D). The results suggested that the OE of MdIAA9 in tobacco seedlings leads to a higher tolerance of osmotic stress.