Background

Abiotic stresses such as high salinity and drought have profound negative impacts on plant development and biomass formation, resulting in significant reductions in crop yield worldwide [1]. To adapt to these abiotic stresses, plants have evolved complex mechanisms for physiological and biochemical mitigation of stress-associated damage, such as the release of reactive oxygen species (ROS) [2, 3]. Previous studies have shown that hydrogen peroxide (H2O2) pretreatment can improve the salt tolerance of wheat by modulating antioxidant enzyme activity, mineral uptake, and proline levels [4]. In the cells of higher plant, ROS exists in many forms, including H2O2, superoxide radicals (O2.−), and hydroxyl radicals (OH−). ROS are generated under abiotic conditions, and cause rapid cell damage by damaging membrane lipids, nucleic acids [5]. Plants have established a complex system to scavenge ROS for maintaining the steady-state level of ROS by activating the antioxidant system. In this case, the antioxidant system refers mainly to free-radical scavenging by several endogenous antioxidant enzymes, such as glutathione peroxidase (GPX), ascorbate peroxidase (APX), catalase (CAT), and superoxide dismutase (SOD) [6,7,8]. It has been reported that antioxidant enzymes (e.g., APX, CAT) are also altered when plants experience salt stress [3, 9, 10]. In addition, PRXs have been reported to protect cells against ROS by catalyzing redox reactions [11].

PRXs exist in many species, such as microorganisms, animals, and plants [12,13,14,15]. These compounds are divided into three superfamilies based on different molecular structures and catalytic properties. The first superfamily includes animal enzymes, such as eosinophil PRX and GPX. The second superfamily is widely distributed in many species (bacteria, animals, fungi, plants, and yeast). The third superfamily is found in plants, bacteria, and fungi [14, 15]. According to differences in primary protein structure, PRXs are divided into three classes-class I PRXs found intracellularly, class II PRXs found extracellularly, and class III PRXs comprises large multigene families [16]. Class I PRXs play key roles in scavenging excess H2O2 [16,17,18], class II PRXs (found in fungi) are involved in the degradation of soil debris [16, 19], and class III PRXs are plant-specific [20]. More than 110 class III PRXs have been identified in allohexaploid wheat [21]. Oryza sativa comprises 138 class III PRXs [22]. Seventy-three sequences encode class III PRXs of Arabidopsis thaliana, and 119 class III PRXs have been identified in maize [12, 23]. Populus trichocarpa contains 93 class III PRXs [24].

Class III PRXs have various functions in plant development processes, including cell wall hardening, crosslinking of cell wall components, defense against pathogen infection, H2O2 removal, and wounding [12, 16, 25]. A large number of PRXs have been studied in A. thaliana and their functions have been demonstrated. For example, AtPRX72 plays an important role in lignification [26], whereas AtPRX33 and AtPRX34 were identified to play a role in cell elongation [16]. Some studies have demonstrated that AtPRX21, AtPRX62, and AtPRX71 are secreted as part of a response to wounding and fungal stresses [27, 28]. The Gossypium hirsutum gene GhPOX1 may cause cotton fiber cell elongation through ROS production [29]. Some PRXs have been reported to play central roles in host plant defenses against necrotrophic or biotrophic pathogens by coordinating salicylic acid (SA), jasmonic acid (MeJA), and ethylene (ET) [20]. For example, TaPRX111, TaPRX112, and TaPRX113 are involved in plant response to nematode infection in wheat [30]. In rice, the expression patterns of PRXs revealed important functional diversity, particularly in response to stresses [31]. The Zea mays PRXs ZmPRX26, ZmPRX42, ZmPRX71, ZmPRX75, and ZmPRX78 are involved in the response to various abiotic stress conditions [12]. Furthermore, GPXs are members of the PRX family, and they have important functions in plants. The A. thaliana GPX gene, AtGPX3, acts as a general scavenger and signal transducer under drought stress and abscisic acid (ABA)-mediated signaling [32]. Six GPXs in Cucumis sativus were found to respond to ABA treatments and abiotic stress. Moreover, five rice GPXs were known to play roles in response against H2O2 and cold stress [33]. Several wheat PRXs have also been discovered to play a role in drought resistance, as revealed by a microarray experiment [34]. Two wheat GPXs, W69 and W106, have been shown to improve salt tolerance in transgenic Arabidopsis [7].

Wheat is an important commercial crop worldwide, but its yield is often restricted by abiotic stresses [35, 36]. The roles of some PRXs in tolerance against salinity stress have been reported; however, the molecular mechanisms of wheat PRXs underlying these responses remain to be fully understood. In this study, we cloned a PRX gene TaPRX-2A from wheat (Triticum aestivum) and investigated tolerance against salt stress conferred by TaPRX-2A in transgenic wheat. Evolutionary analysis revealed that some exon fusion events and positive selection might have occurred during TaPRX-2A evolution. Gene expression pattern analysis demonstrated that TaPRX-2A expression was upregulated by drought, salt, H2O2, and ABA treatments. Our results showed that TaPRX-2A improved the tolerance of wheat against salt by improving antioxidative stress ability and regulating stress-related genes. Our work will give the researchers with new insights into abiotic stress tolerance mechanisms in plants.

Results

Isolation and evolution of TaPRX-2A

To obtain further insights into the evolutionary conservation or divergence of genes among class III PRXs, we identified, classified, and described the gene structures of class III PRXs. PRXs of 12 plants (T. aestivum, Triticum dicoccoides, Triticum urartu, Aegilops tauschii, Brachypodium distachyon, O. sativa, Z. mays, A. thaliana, Vitis vinifera, Selaginella moellendorffii, Physcomitrella patens, and Chlamydomonas reinhardtii) were identified by HMMER 3.1 and Pfam 32.0 in batch mode with the PRX domain (peroxidase.hmm, PF00141.23) (Additional file 1: Table S1, and Additional file 2: Table S2). We excluded the atypical PRXs of these 12 plants that showed < 50% alignment with the PRX domain (Additional file 3: Table S3). The classification of these PRXs was based on two methods, HMMER3.1 scan and neighbour-joining (NJ) phylogenetic reconstruction (Additional file 2: Table S2 and Additional file 4: Figure S1). The exon-intron structures within the PRXs domain were also examined in the 12 plants (Additional file 5: Figure S2).

Among them, we cloned one member (named TaPRX-2A) of the PRXs obtained from the wheat cultivar “Sumai 3.” The predicted TaPRX-2A open reading frame (ORF) is 1026 bp, and the deduced TaPRX-2A protein comprises 342 amino acid residues. BLAST (basic local alignment search tool) results from the National Center for Biotechnology Information (NCBI) showed that a PRX gene (GenBank: AJ878510.1) in the T. aestivum cultivar “Cheyenne” contained the minimum E value. Our local BLAST against the identified PRXs of the 12 plants showed that the T. aestivum PRX TraesCS2A02G573900.1.cds1 from subfamily VI contained 100% sequence similarity with TaPRX-2A. To investigate the evolution of this clone, we reconstructed a small NJ phylogenetic tree only containing subfamily VI PRXs from these 12 plants and compared their structural features (Fig. 1a, b). As shown, the exon–intron structure of the T. aestivum clone (TraesCS2A02G573900.1.cds1) was a one-exon structure, whereas the other four wheat and Ae. tauschii homologous PRXs (Tdi_TRIDC2AG080470.2, Ata_AET2Gv21275100.1, Tae_TraesCS2B02G613900.1.cds1, and Tdi_TRIDC2BG088710.2) in this clade also had a one-exon structure, suggesting that this one-exon structure originated in these PRXs before the Triticum–Aegilops split (Fig. 1b).

Fig. 1
figure 1

Phylogenetic tree and gene structures of TaPRX-2A and related PRXs in wheat Ae. tauschii, and other plants. a The neighbor-joining tree elaborated in this study. Amino acid sequences of the PRX domain were used to construct the neighbor-joining tree using the MEGA-CC 7.0 software with the p-distance model. Most sequences belong to subfamily VI of class III PRX and some branches are compressed. Detailed information is showed in Additional file 4: Figure S1. b The exon–intron structures of some PRXs examined in this study. Red boxes represent the PRX domain; white boxes represent other exon regions; black boxes represent untranslated regions (UTRs); lines represent the PRX introns; numbers 0, 1, and 2 represent the exon phases. The long introns are shortened by “//.” Our investigated PRX (TraesCS2A02G573900.1.cds1) in T. aestivum is circled by a red or cyan box

Full size image

Based on the phylogenetic and exon-intron structure analysis (Additional file 5: Figure S2), we proposed an evolutionary model to determine the origin of TaPRX-2A (TraesCS2A02G573900.1.cds1), which was involved in the processes of exon fusion (Fig. 2a). This model suggests that two rounds of exon fusion events occurred during Angiosperm and Gramineae emergence. The first exon fusion event (four exons became three) occurred during Angiosperm emergence. An ancestral sequence resembling P. patens PRX (Pp3c19_20780V3.3) contained a conserved exon–intron structure within four exons and the “001” exon phases near the PRX domain. This four-exon structure within the “001” exon phase was retained in the ancestral sequences resembling two S. moellendorffii PRXs (Smo_EFJ32905 and Smo_EFJ15769). However, the exon–intron structures of PRXs in A. thaliana (Ath_AT1G71695.1), V. vinifera (Vvi_VIT_18s0072g00160.t01), and O. sativa (Osa_Os04t0688200–01) changed into the three-exon structure within the “00” exon phases, suggesting the occurrence of an exon fusion event in the last two exons of the four-exon structures within the “001” exon phases before the monocot–eudicot evolutionary split. The second exon fusion event (three exons changed into two or one exon) occurred during Gramineae emergence. As shown in Fig. 2a, the first two exons in the three-exon structure within the “00” exon phases may have fused and changed into the two-exon structure within the “0” exon phase (B. distachyon, KQJ85452). Similarly, the last two exons could have fused (T. aestivum, TraesCS2A02G574400.1; T. urartu, TRIUR3_03591-P1) or all the three exons could have fused, thus merging into a single-exon structure (Ae. tauschii, Ata_AET2Gv21275100.1; T. dicoccoides, Tdi_TRIDC2AG080470.2, Tdi_TRIDC2BG088710.2; T. aestivum, TraesCS2A02G573900.1.cds1, Tae_TraesCS2B02G613900.1.cds1). The alignments of these PRXs within the breakpoints of exon fusion events supported our proposed evolutionary model (Fig. 2b).

Fig. 2
figure 2

Evolutionary model of subfamily VI PRXs, including TaPRX-2A. a The exon fusion events of some subfamily VI PRXs in evolution. b The alignment of some subfamily VI PRXs. Exon phases are circled by red boxes and arrows. The PRX we investigated (TraesCS2A02G573900.1.cds1) in T. aestivum is circled by a cyan box

Full size image

To confirm these PRX sequences for the TaPRX-2A evolutionary model, we determined the cDNA-level evidences in RNA-sequencing (RNA-seq) data from seven plants (P. patens, A. thaliana, V. vinifera, B. distachyon, Ae. tauschii, T. dicoccoides, and T. aestivum) (Additional file 6: Table S4). We did not determine these evidences in S. moellendorffii and T. urartu because their GFF3 annotation files were just in scaffolds and not in chromosomes. The results showed that most PRX sequences (except VIT_18s0072g00160.t01 and TraesCS2A02G574400.1) from seven plants were detected in RNA-seq data (FPKM and coverage values in the “information” column of Additional file 6: Table S4), suggesting that the occurrence of exon fusion events during plant evolution.

We also detected positive selection of PRX domain sequences in TaPRX-2A and 12 other homologous PRXs using PAML 4.9 (Table 1). According to the likelihood ratio test of site-specific models, the M2a (selection) model was significantly higher than M1a (neutral) (degrees of freedom (df) = 2, 2ΔlnL = 68.4, P < 0.005), indicating that some amino acid sites underwent positive selection during evolution. The M7–M8 comparison (df = 2, 2ΔlnL = 7.47, P < 0.025) also supported the hypothesis of positive selection. These positively selected sites were identified using Naive Empirical Bayes and Bayes Empirical Bayes analyses (Additional file 7: Figure S3a,b). Two (95 E and 185 K, refer to sequence: Smo_EFJ32905) and six positively selected sites (95 E, 110 S, 117 Q, 135 E, 185 K, and 212 R) were found in the M2a and M8 models, respectively. Ancestral sequences in evolutionary nodes were also inferred by PAML 4.9 and MEGAX (Additional file 7: Figure S3).

Table 1 Detection of positive selection of TaPRX-2A and other 12 homologous PRX genes in plants
Full size table

Expression patterns of TaPRX-2A in various tissues and stress treatments

To detect the expression patterns of TaPRX-2A in response to stress-related signaling, we performed qRT–PCR in different tissues (leaf, stem, and root) and with different stress treatments (PEG6000, NaCl, H2O2, SA, MeJA, IAA, and ABA). The results showed that TaPRX-2A was differentially expressed in the roots, stems, and leaves, with significantly higher expression levels in root tissues than in leaf and stem tissues (Fig. 3a). Then, we checked the expression patterns of TaPRX-2A by qRT–PCR in treatments of PEG 6000, NaCl, and H2O2. The results showed that the expressions of TaPRX-2A were induced by PEG 6000, NaCl, and H2O2 treatments and the expression levels reached a peak at 6 h after treatments (Fig. 3b, c, and d). We also examined the expression patterns of four phytohormones. As shown in Fig. 3e, TaPRX-2A exhibited approximately 2.5-fold upregulation at 1 h after SA treatment (Fig. 3e). Similarly, the expression levels of TaPRX-2A reached a peak at 6 h after JA and ABA treatments (Fig. 3f, h). However, the expression levels of TaPRX-2A remained relatively unchanged throughout 0–6 h after IAA treatment but exhibited an approximate 1.5-fold upregulation at 12 h (Fig. 3g). These results suggested the involvement of TaPRX-2A in various abiotic stress responses.

Fig. 3
figure 3

Transcript analysis of TaPRX-2A. a TaPRX-2A expression levels in the leaves, roots, and stems in normal conditions and under different stress treatments: (b) 20% (w/v) PEG 6000, (c) 200 mM NaCl, (d) 10 mM H2O2, (e) 2 mM SA, (f) 100 μM MeJA, (g) 100 μM IAA, and (h) 100 mM ABA. 18SrRNA was used as an endogenous control. The gene relative expression was calculated by the cycle threshold (Ct) values using formula 2–ΔΔCT. Data are presented as mean ± standard deviation calculated from triplicates

Full size image

Subcellular localization of the TaPRX-2A protein

To characterize the function of TaPRX-2A, the ORF of TaPRX-2A was fused to a pBIN35S-eGFP vector under the control of a CaMV 35S promoter (Additional file 8: Fig. S4a). The pBIN35S:eGFP empty vector control and the pBIN35S:TaPRX-2A:eGFP recombinant vector construct were transformed into tobacco leaf cells by Agrobacterium infiltration. We observed the epidermal cells of injected N. benthamiana leaves by confocal microscopy and found that TaPRX-2A:eGFP was mainly localized in the cell nuclei (Additional file 8: Fig. S4b1-d2). In addition, the pBIN35S-TaPRX-2A-eGFP and pBIN35S-eGFP vector were transformed into onion epidermal cells. Consistent with the localization results observed in tobacco epidermal cells, the TaPRX-2A:eGFP was also mainly localized in the nuclei of onion epidermal cells (Additional file 8: Fig. S4b3-d4). Moreover, the prediction of web server cNLS showed that five nuclear localization signal (NLS) sequences were present in TaPRX-2A (Additional file 9: Figure S5).

TaPRX-2A enhanced salt tolerance in transgenic wheat

To further confirm the function of TaPRX-2A in responses against salt stress in plants, we transformed the wheat cultivar “KN199” with TaPRX-2A overexpression and constructed three independent transgenic lines (TaOE1, TaOE2, and TaOE3). The expression profile of TaPRX-2A was analyzed in TaPRX-2A transgenic lines through qRT–PCR. The results showed that the transgenic lines exhibited a higher expression level than wild-type (WT) plants (Additional file 10: Fig S6a). We subsequently measured the PRX activity in three independent transgenic lines and WT and found that the activity was higher in transgenic lines than in WT (Additional file 10: Fig. S6b). Taken together, we concluded that TaPRX-2A overexpression caused high PRX activity in transgenic lines.

Then, we measured the phenotypic differences between transgenic lines (three independent lines (TaOE1, TaOE2, and TaOE3) and WT under salt stress conditions. Under non-stress conditions, no visibly phenotypic difference was observed between TaOE1–3 and WT. Under salt stress conditions, transgenic lines showed stronger growth compared with WT. In addition, the WT leaves turned yellow and wilted under salt stress, whereas the TaOE leaves still remained green (Fig. 4a). We also found that the survival rate of WT plants was only 40% after salt treatment, whereas the survival rates among TaOE1, TaOE2, and TaOE3 plants were 63.6, 57.6, and 63%, respectively (Fig. 4b). We then compared the shoot lengths, relative water content (RWC), and root lengths between WT and TaOE plants under salt treatment (Fig. 4c, d, and e). The results showed that transgenic lines exhibited longer shoot length and higher RWC than WT plants. However, no significant difference in root lengths was observed between WT and transgenic lines. Taken together, these results indicated that TaPRX-2A overexpression drastically enhanced salt tolerance in wheat.

Fig. 4
figure 4

TaPRX-2A enhanced the salt tolerance of wheat. a Salt stress tolerance responses of TaPRX-2A-overexpressing transgenic and WT wheat (the cultivar “KN199”). (b) Survival rates, (c) shoot length, and (d) root length of TaPRX-2A-overexpressing and WT plants. (e) Relative water content of TaPRX-2A-overexpressing and WT plants. Data are presented as mean ± standard deviation calculated from triplicates. * and ** above each column indicate a significant difference compared with WT plants (*P < 0.05; **P < 0.01)

Full size image

To further explore mechanisms underlying TaPRX-2A-mediated response to salt stress, we measured physiological–biochemical indices between TaOE and WT plants under non-stress and salt stress conditions (Fig. 5a-d). Under salt treatment, TaOE lines contained significantly lower malondialdehyde (MDA) content than WT; they also contained higher soluble sugar, proline, and soluble protein contents. Moreover, the proline contents of transgenic lines were approximately 2-fold greater than those of the WT plants (Fig. 5c). These results suggested that overexpression of TaPRX-2A increased the contents of metabolites that were necessary for osmotic and oxidative stress tolerance in wheat cultivar “KN199”, thus resulting in improving tolerance to salt.

Fig. 5
figure 5

Physiological and biochemical indices between TaPRX-2A-overexpressing transgenic lines and WT plants. a MDA content, (b) soluble sugar content, (c) proline content, and (d) soluble protein content of TaPRX-2A-overexpressing and WT plants. Data are presented as mean ± standard deviation calculated from triplicates. * and ** above each column indicate a significant difference compared with WT plants (*P < 0.05; **P < 0.01)

Full size image

TaPRX-2A regulates ROS scavenging and the expression of stress-related genes in transgenic wheat

Previous studies have demonstrated that the tolerance to oxidative stress is associated with the physiological response of the plant to abiotic stresses [35, 37, 38]. Therefore, we examined the function of TaPRX-2A in reducing ROS levels in transgenic lines under salt stress. As major indicators of the ROS level, we assayed the accumulation of O2− and H2O2 for comparison between TaOE and WT lines using nitroblue tetrazolium (NBT) and 3-diaminobenzidine (DAB) staining. Under salt treatment, we found that the levels of O2− (stained blue with NBT) and H2O2 (stained brown by DAB) were significantly lower in transgenic lines than in WT plants (Fig. 6a-d). In addition, the activities of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT) antioxidant enzymes were higher in the transgenic plants than in the WT plants (Fig. 6e-g).

Fig. 6
figure 6

TaPRX-2A confers ROS-scavenging capacity by improving antioxidant enzymes activity. a Tissue localization of O2− accumulation, (b) O2− content, (c) tissue localization of H2O2 generation, (d) H2O2 content, (e) SOD activity, (f) CAT activity, and (g) POD activity of TaPRX-2A-overexpressing and WT plants. Data are presented as mean ± standard deviation calculated from triplicates. * and ** above each column indicate a significant difference compared with WT plants (*P < 0.05; **P < 0.01)

Full size image

To determine whether stress-responsive genes were associated with enhancing salt tolerance through TaPRX-2A, we determined the expression patterns of various stress-related genes in TaOE plants using qRT–PCR (Fig. 7). These stress-related genes (encoding dehydration-responsive protein, RD22; thaumatin-like protein, TLP4; ABA-inducing protein, ABAI; germin-like protein, GLP4; glutathione S-transferase, GST22; and the genes encoding the ROS-scavenging enzymes FeSOD, CuSOD, and CAT) were reported to be involved in the response to various abiotic stresses. Our results showed that the majority of these stress-related genes were more highly expressed in the TaOE lines than in the WT plants under salt stress, except CuSOD expression, which was not significantly different between the WT and transgenic lines. We also found that the expression of some stress-related genes, including RD22, ABAI, and CAT, was lower in the WT plants under non-stress conditions than in the transgenic plants. Taken together, these results suggested that TaPRX-2A overexpression may improve salt tolerance in wheat by enhancing the transcription levels of stress-responsive genes.

Fig. 7
figure 7

Analysis of the transcript profile of stress-related genes. Expression levels of stress-related genes in TaPRX-2A-overexpressing transgenic lines and WT plants under salt stress by qRT–PCR. The gene 18SrRNA was as an endogenous control. Each treatment had three independent biological repeats. The gene relative expression was calculated by the cycle threshold (Ct) values using formula 2–ΔΔCT. The data are presented as mean ± standard deviation calculated from triplicates. * and ** above each column indicate a significant difference compared with WT plants (*P < 0.05; **P < 0.01)

Full size image

Discussion

Evolution of TaPRX-2A in T. aestivum

The objective of this study was to characterize the role of the wheat PRX gene TaPRX-2A in the plant’s response to salinity stress, in light of the severe reduction in crop yield associated with this form of abiotic stress [39]. Based on the classification of the NJ phylogenetic tree and the HMM scan, TraesCS2A02G573900.1.cds1 was found to belong to subfamily VI PRXs, which can be found in S. moellendorffii but not in P. patens, suggesting that subfamily VI PRXs have appeared in fern-resembling ancestors. Subfamily VI PRXs contain only one member in two investigated eudicots A. thaliana and V. vinifera, whereas subfamily VI PRXs contains various members of the investigated monocots (Additional file 5: Figure S2), suggesting that subfamily VI has experienced monocot-specific duplication events after the monocot–eudicot split.

Based on the analysis of exon–intron structures of the 12 plants investigated in this study, we proposed an evolutionary model involving two rounds of exon fusion events to infer the origin of TaPRX-2A (TraesCS2A02G573900.1.cds1) (Fig. 2). Among these exon fusion events, we focused on one of the second round of exon fusion events, wherein a three-exon structure changed into a one-exon structure before the Triticum–Aegilops split (formed into TaPRX-2A ancestor). The possible mechanism of this one-exon structure emergence might be “retroposition” (as a result of retrotransposition, the newly duplicated paralogs lack introns), which was reported in the origin of the gene jingwei in Drosophila species [40] and ATP synthase PGAM3 [41]. For instance, PGAM1 contains three introns, whereas PGAM3 has none. In plants, 69 retroposons and 1235 primary retrogenes were identified in A. thaliana and O. sativa, respectively [42, 43]. We also detected the positive selection among these 13 PRXs of evolutionary model by PAML4.9, and found 2 or 6 positively selected sites. It was also reported that 7 gene pairs among 24 retrogenes in Oryza species were identified to be under positive selection [44].

TaPRX-2A enhanced the antioxidative stress ability and improved salt tolerance in wheat

In higher plants, class III PRXs comprise a large gene family, the members of which have been reported to participate in plant’s response to abiotic stresses [16]. For example, the class III PRX gene OsPRX38 in O. sativa reportedly improved Arabidopsis arsenic (As) tolerance by activating the antioxidant system (SOD, PRX, and GST) and scavenging H2O2 [45]. In tobacco, it has been shown that the overexpression of the class III PRX gene AtPrx64 in A. thaliana improved the plant’s aluminum (Al) tolerance by increasing root growth and scavenging the accumulation of Al and ROS [46]. In A. thaliana, overexpression of AtPRX3 was shown to alleviate dehydration and increase salt tolerance. However, inhibition of AtPRX3 expression decreased the tolerance to dehydration and salt [47]. Furthermore, the class III PRX genes CrPrx1 and CrPrx in Catharanthus roseus have been reported to improve germination rate under salinity stress in Nicotiana tabacum [48]. Consistent with the results of these reports, our results also proved a positive regulator in the salt tolerance of wheat by TaPRX-2A.

Among the physiological issues that accompany abiotic stresses in plants, the excessive accumulation of ROS (particularly O2− and H2O2) and high concentrations of ROS can damage cell membrane permeability and integrity as well as cell compartmentation [49,50,51]. The class III PRXs in plants can catalyze H2O2 reduction in the peroxidative cycle by transferring electrons from different donors [13, 16]. Previous studies have shown that class III PRXs can improve stress tolerance by regulating the ROS balance in plants. For example, OsPRX38 improves Arabidopsis As tolerance by activating the antioxidant system and scavenging H2O2 [45]. AtPrx64 improves the plant’s Al tolerance by scavenging ROS accumulation [46]. To maintain ROS balance through scavenging of free radicals, plants have evolved a complex antioxidative system to protect cells from damage [49, 51]. Notably, SOD, CAT, and PRX expression reportedly contributes to enhanced salt tolerance [9, 10]. In our study, we found that TaPRX-2A overexpression improved the antioxidant activity through CAT and PRX, thereby reducing ROS levels. In addition, the expression of antioxidant genes (FeSOD, CuSOD, and CAT) was altered in the TaPRX-2A-overexpressed transgenic plants compared with that observed in the WT plant, suggesting that TaPRX-2A regulates the expression of these genes, thereby affecting salt tolerance. Future research will explore the mechanisms through which TaPRX-2A regulates other antioxidant-encoding genes.

Interestingly, we found that TaPRX-2A was localized in the nucleus with NLSs. Some reports have shown that PRX genes such as TaPRXs, AtGPX8, and GPX in mammals and LjGpx1 were located in the cell nuclei [52,53,54,55,56]. It was reported that a barely PRX possess a putative nuclear localization signal, which is located in the nucleus [57]. Previous studies have shown that ROS could cause DNA damage by activating endonucleases and damaging important biological macromolecules (such as nucleic acids) [5, 58, 59]. In Arabidopsis, AtGPX8 is localized in the nucleus and can protect nuclear DNA from ROS damage by maintaining cellular redox homeostasis [53]. Based on the above evidence, we propose possible regulatory mechanisms: one explanation is that TaPRX-2A is expressly conserved for inhibiting ROS-mediated damage to genomic DNA in the nucleus and that other enzymes are responsible for scavenging ROS in or adjacent to organelles. Our second explanation is that TaPRX-2A is coexpressed with other ROS-scavenging enzymes and that its transcriptional upregulation leads to upregulation of enzymes modulating ROS levels outside the nucleus during exposure to salinity stress (Fig. 7).

TaPRX-2A’s effects on salt tolerance via the ABA-dependent pathway

In plants, the ABA signaling pathway regulates various abiotic stress responses [60, 61]. For example, dehydrin and thaumatin-like proteins (TLP) genes, which are essential for tolerance to abiotic stresses, may be induced by ABA during stresses [62, 63]. In Arabidopsis, ABA reportedly mediates the transcriptional upregulation of the dehydration-responsive gene RD22 [61]. Similar to TaPRX-2A, several class III PRX genes are reported to mediate tolerance to abiotic stresses via the ABA signaling pathway [64]. For example, the expression of AtPRX3 in A. thaliana is induced by both salinity stress and exogenous ABA treatment [47]. Seven class III PRXs genes from Tamarix hispida are controlled by the ABA signaling pathway [65]. In our study, the expression of TaPRX-2A was highly upregulated by both NaCl and exogenous ABA treatments (Fig. 3). Moreover, we observed that the overexpression of TaPRX-2A in transgenic wheat led to the transcriptional upregulation of the stress-related genes RD22, TLP4, ABAI, GLP4, and GST22 under salinity stress (Fig. 7). Our work demonstrated that TaPRX-2A enhances salt tolerance of wheat through activating the downstream stress-related genes and the ABA signaling pathway. Further study is warranted to explore the regulatory mechanisms through which TaPRX-2A regulates stress-related genes.

Conclusions

In this study, we identified and characterized the role of the PRX gene TaPRX-2A in response to salinity stress in wheat. Phylogenetic analysis revealed the occurrence of some exon fusion events and positive selection during TaPRX-2A evolution. The overexpression of TaPRX-2A enhanced salt tolerance in transgenic wheat through activation of the ABA pathway and antioxidant enzymes, resulting in lower ROS accumulation and increased levels of osmotic metabolites. This work and its findings have strong future application value in the cultivation of salt-tolerant wheat varieties, which is especially relevant given the anticipated crop losses associated with the future impacts of climate change.

Methods

Isolation and cloning of TaPRX-2A and its transformation

Leaves of the harvest wheat cultivar “Sumai 3” were used to extract total RNA using TRIzol reagent (Transgen). cDNA was synthesized to amplify TaPRX-2A. The full-length cDNA sequence of TaPRX-2A (Genbank No. AJ878510.1) was obtained from NCBI (https://www.ncbi.nlm.nih.gov/), ligated into the PC186 vector, and transformed into “KN199” using particle gun-mediated gene transformation [66].

Plant materials and abiotic treatments

Bread wheat (T. aestivum cultivars “KN199” and “Sumai 3”) seedlings were sourced from our laboratory (State Key Laboratory of Crop Biology, College of Agronomy, Shandong Agricultural University). The TaPRX-2A-overexpressing transgenic wheat lines and the WT “KN199” were grown at 20 °C–25 °C with a photoperiod of 16/8 h. When the plants grew to a period of one leaf and one heart, the transgenic plants and “KN199” were treated with 200 mM NaCl treatment. Concerning salt treatment, the control and transgenic seedlings were cultured in 200 mM NaCl solution for 10 days.

Identification and classification of class III PRXs in wheat, Ae. tauschii, and other plants

The genomes and proteomes of 12 plants (S. moellendorffii, Z. mays, B. distachyon, T. aestivum, Ae. tauschii, T. dicoccoides, V. vinifera, T. urartu, O. sativa, A. thaliana, P. patens and C. reinhardtii) were downloaded from the Ensembl Plants 42 (http://plants.ensembl.org/) and analyzed. To identify PRXs, we scanned all the proteomes of the 12 plants in batch mode using our local server with Hmmer 3.1 (pfam profile PF00141.23, peroxidase.hmm, PRX domain). Then the pfam 32.0 website (http://pfam.xfam.org/) with an E value of 0.01 was used. Typical PRXs with a PRX domain covering > 50% alignment were retained and analyzed. Those covering < 50% of the PRX domain alignment were considered atypical PRXs and excluded from further analysis. PRX alignment of truncated sequences in the PRX domain was performed using ClustalW v2.0 [67]. We used the MEGA-CC 7.0 software to construct the NJ phylogenetic tree in our local server [68]. The PRX subfamily classification was performed using HMMER 3.1, and the models were generated based on maize PRX alignments [12].

The RNA-seq data of P. patens (SRR11434644, SRR11434645, and SRR11434646), A. thaliana (SRR11308184, SRR11308187, and SRR11308188), V. vinifera (SRR11249050, SRR11249059, and SRR11249060), B. distachyon (SRR10380965, SRR10380966, SRR10380967, and SRR10380968), Ae. tauschii (SRR9657462 and SRR9657463), and T. dicoccoides (SRR9657450 and SRR9657451) were downloaded from the NCBI SRA transcriptome database (https://www.ncbi.nlm.nih.gov/sra/). The RNA-seq data of T. aestivum (ERR1201797, ERR1201798, and ERR1201799) were downloaded from EBI ArrayExpress (https://www.ebi.ac.uk/arrayexpress/). Mapping of sample reads to the reference genome (the Ensembl Plants 42) was conducted using Hisat2 (version 2.2.0, https://daehwankimlab.github.io/hisat2/download/#version-hisat2-220). Conversion (sam to bam) and sorting were performed using Samtools (version 1.10, https://github.com/samtools/samtools/releases/). The transcripts were assembled using Stringtie (version 2.1.1, https://ccb.jhu.edu/software/stringtie/index.shtml).

Domain and intron–exon structure diagram of PRXs

We used Perl and R scripts to generate PRXs intron–exon structures and domain diagrams for each of the 12 plants included in this study, based on the corresponding GFF file information from Ensembl Plants 42 (http://plants.ensembl.org/). The domain information of PRXs was batched from Pfam 32.0 (http://pfam.xfam.org/).

Analysis of selective pressure

The truncated amino acid PRX domain sequences in TaPRX-2A and the other 12 homologous PRXs were aligned using Clustal X2. Based on the information of Pfam 32, the corresponding truncated cDNA of the PRX domain was generated from our Perl scripts. The codon alignment was generated by the web server PAL2NAL [69]. PAML 4.9 (CODEML) [70] and graphical interface PAMLX [71] were used to detect the selective pressure. Site-specific models M0 (one ratio), M1a (neutral), M2a (selection), M7 (beta), and M8 (beta & ω) were generated. Log likelihood (lnL) value of each model was calculated by CODEML. Models were compared using 2 Δ lnL = 2(lnL1 − lnL0) in accordance with the χ2 distribution with df. Ancestral sequences were inferred by CODEML (rst file of results) and MEGAX (using the ML method and JTT matrix-based model) [72].

Expression pattern of TaPRX-2A in different abiotic stress treatments

The wheat leaf tissues of three-leaf stage seedlings were harvested at 0, 6, 12, 24, 48, and 72 h after 20% (w/v) PEG 6000 treatment and 200 mM NaCl treatment. The wheat leaf tissues were harvested at 0, 2, 6, 12, 24, 48, and 72 h after 10 mM H2O2 treatment. We harvested the wheat leaf tissues at 0, 1, 3, 6, 12, 24, and 48 h after treatments with 2 mM SA, 100 μM MeJA, 100 μM IAA, and 100 mM ABA. Total RNA of all harvested samples was extracted using TRIzol reagent (Invitrogen). First-strand cDNA synthesis and qRT–PCR were performed using the Roche LightCycler®480 system (Roche, Germany). The wheat gene 18SrRNA was used as an endogenous control. Relative mRNA expressions were calculated using the 2–ΔΔCT method. All qRT–PCR primers are summarized in Additional file 11: Table S5.

Subcellular localization of the TaPRX-2A protein

According to the ORF of TaPRX-2A, we cloned this gene without a stop codon and constructed it into a pBIN35S-eGFP vector using the CaMV 35S promoter. Subsequently, pBIN35S-TaPRX-2A-eGFP and pBIN35S-eGFP (control) were transformed into Agrobacterium EHA105. The Agrobacterium EHA105 was resuspended in the suspension (10 mM MgCl2, 10 mM 4-morpholineethane-sulfonic acid hydrate at pH 5.6, 200 mM acetosyringone). The Agrobacterium suspension was adjusted to an optical density 600 value of 0.6, injected into tobacco leaves, and cultured for 3 days. The epidermal cells of the injected tobacco leaves were observed using a confocal microscope (Zeiss LSM880 Meta Confocal Microscope). In addition, we also transformed the pBIN35S-TaPRX-2A-eGFP and pBIN35S-eGFP vectors into onion epidermal cells through gene gun-mediated transformation [73]. The transformed epidermal cells were cultured in the dark at 28 °C for 8–12 h and observed using a confocal microscope. The NLS sequences of TaPRX-2A were predicted using the web server cNLS (http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi) [74].

Measurements of physiological–biochemical parameters

We collected the leaves of TaPRX-2A-overexpressed and “KN199” plants at 10 days after salt treatment. We used the formula to measure the leaf RWC: RWC = (FW − DW)/(TW − DW) × 100% where RWC is relative water content, FW is fresh weight, TW is turgid fresh weight, and DW is dry weight [75]. MDA content was measured using the thiobarbituric acid method [76]. Proline content was measured using the ninhydrin reaction method [77]. Soluble total sugars were determined by the anthrone method [78]. We used NBT and DAB staining to visualize O2− and H2O2 levels, as reported previously [79, 80]. The SOD, CAT and PRX activity were detected using previously described methods [81,82,83].