Introduction

Plants have evolved adaptive mechanisms to defense adverse environments as they are nonmotile, including regulating the specific genes’ expression and producing a large number of stress-related proteins to response various stress (Zhai et al. 2016). DnaJs/Hsp40s (heat-shock protein 40) is a large ubiquitous family which functions in a myriad of cellular processes as a co-chaperone component for Hsp70s machinery (Kampinga and Craig 2010; Qiu et al. 2006; Silver and Way 1993; Vitha et al. 2003). DnaJ proteins contain a highly conserved 70-amino-acid consensus sequence known as J domain, and they can bind and regulate Hsp70 proteins by J domain (Hennessy et al. 2005; Jiang et al. 2007). DnaJ has conservatism throughout evolutionary process and facilitates the protein folding and refolding, and also play important roles for protein translation, translocation, and degradation (Chen et al. 2010; Wang et al. 2004). Expression of heat-shock proteins in planta can be induced by a variety of environmental stresses, such as soil salinity, high temperature, cold, osmotic and oxidative stresses (Dekker et al. 2015). For example, BIL2, encoding mitochondria-localized heat shock protein, enhanced resistance against salt and high light stress in brassinosteroid signaling if overexpressed in Arabidopsis (Bekh-Ochir et al. 2013). AtDjA3 gene is expressed in various tissues, and its expression could be induced by heat, cold, drought, osmotic shock stresses and ABA (Salas-Munoz et al. 2016), and also under saline conditions with alkaline pH (Yang et al. 2010). Therefore, DnaJ proteins are widely considered as cellular stress adapter (Piippo et al. 2006; Rajan and D’Silva 2009; Scarpeci et al. 2008).

Alkaline soils are widely spread in the world and severely affect the terrestrial ecology, and subsequently limit agricultural productivity globally. Previously reports have clearly defined salt stress and alkaline stress, salt stress is mainly caused by neutral salts such as NaCl and/or Na2SO4, while alkaline stress is mainly caused by alkaline salts like NaHCO3 and/or Na2CO3 (Shi and Wang 2005). Salt stress in soil generally involves osmotic stress and ion-induced injuries (Munns 2002); there are similar problems for alkaline stress but with additional high-pH influence. Alkaline stress can severely affect soil structure, interfere with essential micro nutrients uptake and upset intracellular ion balance in plant (Yang et al. 2007). Plant grown in alkaline soil showed reduced leaf area, leaf length and leaf width, consequently organism shoot biomass is decreased. The diminished leaf area is mainly contributed to decreased photosynthetic rate and stomatal conductance in alkaline-induced leaf chlorosis (Bie et al. 2004). Alkaline environment surrounding the root can inhibit root respiration and affect accumulation and compartmentation of organic acids in root cells (Lee and Woolhouse 1969; Yang et al. 1994). It can also lead metal ions and phosphorus to precipitate, with loss of the normal physiological functions of root and destruction of the root cell structure (Li et al. 2009). Those inhibitory effects would cause reduced root growth and elongation. Plants exposed to NaHCO3 produced more reactive oxygen species (ROS, O2 −, H2O2), and exhibited increased activities of antioxidant enzymes (SOD, POX) at a low concentration of NaHCO3 (Gong et al. 2014; Guan et al. 2016).

In previous reports, PROTEIN KINASE5 (PKS5) and the chaperone DNAJ HOMOLOG3 (J3) in Arabidopsis thaliana are responsive to salt at alkaline pH via regulating the interaction between PM H+-ATPase and 14-3-3 proteins (Fuglsang et al. 2007; Yang et al. 2010). Similar, 14-3-3 protein TFT4 regulates PKS5-J3 signaling pathway and significantly enhanced H+ efflux and the activity of PM H+-ATPase in the root tips and increased plant alkaline resistance in tomato (Xu et al. 2013). As wild soybean (Glycine soja) G07256 is an ideal candidate for exploring resistant genes and breeding of transgenic legume crops with superior salt-alkaline tolerance (Chen et al. 2013). We have identified and characterized a couple of genes including a chaperone 14-3-3 gene from wild soybean (Liu et al. 2015; Sun et al. 2015, 2016; Yu et al. 2016). These genes are initially identified from a transcriptome analysis of G. soja roots under NaHCO3 stress on the Illumina Genome Analyzer IIx (GAIIx) platform (DuanMu et al. 2015).

Here we systematically compared the evolutions and diversities of DnaJ family genes and investigated their expression profiles in wild soybean based on the transcriptome data of G. soja under NaHCO3 stress, and identified one DnaJ gene, GsJ11, which was significantly responsive to NaHCO3 treatment. Moreover, we found the gain-of-function of GsJ11 and loss-of-function mutants of its homologous gene in Arabidopsis significantly altered plant alkaline resistance and physiological indices, providing the clues to understand the novel functions of DnaJ proteins and the mechanism of plant alkaline resistance.

Materials and methods

Database searches, identification and gene classification of DnaJ family genes in soybean

Previously identified ten different species and three types of DnaJ protein sequences were used as queries to establish a Hidden Markov model. Although the genome of wild soybean (Glycine soja) was sequenced, the well-annotated database is still not available (Qi et al. 2014). Since the cultivated soybean (Glycine max) is the close relative of G. soja, here, we searched for DnaJ members by using soybean (Glycine max Wm82.a2.v1) database with the HMM file. All information about soybean DnaJ genes, including sequences ID, gene locations on chromosomes, sequences lengths of DNA, cDNA, coding sequence (CDS) and protein were obtained from Phytozome (https://phytozome.jgi.doe.gov/pz/portal.html). ExPASy online software (http://web.expasy.org/protparam/) was used to calculate physical parameters such as the molecular mass (kD), and isoelectric point (pI) of all the predicted DnaJ proteins (Feng et al. 2015). Positions of J domain were obtained from NCBI Conserved Domain Search program (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). All potential candidate proteins were analyzed to verify the presence of the J domain using SMART (Simple Modular Architecture Research Tool) tools (http://smart.embl-heidelberg.de/) (Letunic et al. 2015) and Pfam (http://pfam.sanger.ac.uk/) database. In addition, SMART tool was also subjected to analysis of the obtained sequences structures (Schultz et al. 2000).

Multiple alignment and phylogenetic analysis

Multiple sequence alignment of all the members of DnaJ family protein was performed using ClustalX 2.0 software with default parameters (Larkin et al. 2007). Based on the alignment, the phylogenetic tree was constructed with MEGA 5.0 software, by using the maximum likelihood (ML) method and 500 replications for the 500 bootstrap resampling (Li et al. 2015).

Exon/intron organization, chromosomal location and gene duplication

A schematic diagram of the gene Exon/intron organization of DnaJ genes was executed through the Gene Structure Display Server (GSDS: http://gsds.cbi.pku.edu.cn/) (Hu et al. 2015). Based on the position information obtained from Phytozome database (http://www.phytozome.net), the chromosomal location image of the DnaJ family genes was generated by using the MapInspect software (http://www.plantbreeding.wur.nl/uk/software_mapinspect.html) (Wu et al. 2015).

Microarray analysis

The microarray data for gene expression in root and leaf of wild soybean (G. soja) under alkaline stress was downloaded from the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) database under accession numbers GSE17883 and GSE20323. By comparing the variation of gene expression levels between treated and control samples, differential expression genes were selected (|log2Ratio|>1,p < 0.05) and a heat map was generated using the MATLAB software.

Plant materials, growth conditions, and stress treatments

To illustrate the expression patterns of the target gene under bicarbonate treatment, the seeds of wild soybean (G. soja 07256) line were obtained from the Jilin Academy of Sciences (Changchun, China). For the purpose of surface sterilization, the G. soja seeds were firstly shaken for 15 min in 98% sulfuric acid and then washed by sterilized distilled water for 3 times, and followed by incubation in dark for 24 h to break seeds dormancy. The geminating seeds were grown in 1/4 Hoagland solution at 24–28 °C and 16 h light/8 h dark cycles for 3 weeks. The nutrient solution should change every 3 days. 21-day-old seedlings were transferred into 1/4 Hoagland solution with 50 mM NaHCO3. The root samples were collected at six independent time points (0, 1, 3, 6, 12 and 24 h) and were stored in liquid nitrogen.

Columbia ecotype (Col-0) strain of Arabidopsis thaliana was used in this study as wild type plant, and also used for generating overexpressing lines. T-DNA insertion mutant line of atj11 (SALK_015630 C) were obtained from The Arabidopsis Information Resource center (TAIR) and identified by PCR and semi-quantitative RT-PCR. The gene specific primer pairs were used for PCR: LP 5′-TTATGGCTGCATCCCTAATTG-3′ and RP 5′-TTCTTCTCCGCCTCTATCTCC-3′ and left border T-DNA primer LBb1.3 5′-ATTTTGCCGATTTCGGAAC-3′. The specific primers used for semi-quantitative RT-PCR were as follows: 5′-GAAGATCCATGCCGCTTACTG-3′ and 5′-CGGAGGAAACACAGAATACCC-3′.

For surface sterilization, the seeds of Arabidopsis were shaken for 6 min in 5% NaClO and then washed with distilled water for 6 times to remove residues completely. After that, the sterilized seeds were incubated at 4 °C for 3 days to break dormancy, and then the seeds were sown on 1/2 MS solid medium under controlled environmental condition (21–23 °C, 100 μmol photons m−2 s−1, 60% relative humidity, 16 h light/8 h dark cycles). To analyze the expression of stress-responsive marker genes, 14-days-old seedlings of WT, GsJ11 OX and atj11 lines were treated with water or 50 mM NaHCO3 solution and then were sampled at various time points (0, 3, 6 and 12 h) from 3 biological replicates after treatments.

Quantitative real-time PCR

Total RNA was extracted as per described by EasyPure Plant RNA Kit protocol (Transgen Biotech, China) from G. soja or Arabidopsis seedlings. RNA samples were used for reverse transcription with SuperScriptTM III Reverse Transcriptase kit (Invitrogen, Carlsbad, CA, USA). Quantitative real-time PCR was performed using SYBR Green Master Mix (TaKaRa, Shiga, Japan) to detect transcript levels. GADPH (Glycine soja) and ACTIN2 (Arabidopsis thaliana) were used as internal references. All experiments were carried out with three independent biological replicates for statistical analyses. The relative expression levels were calculated by using 2− ΔΔCT method (Steibel et al. 2005). The gene specific primers used for quantitative real-time PCR were listed in Online Resource 1.

Vector construction and generation of transgenic Arabidopsis thaliana

The full-length GsJ11 coding DNA sequence was cloned by USER cloning method with primer pairs: 5′-GGCTTAAUATGATTTCTTCCGTGTCC-3′ and 5′-GGTTTAAUCTACCAGCACTGATCCGT-3′ (Nour-Eldin et al. 2006). The PCR products were inserted into pCAMBIA330035Su USER vector and sent for sequencing. The recombinant vector pCAMBIA330035Su-GsJ11 was transformed in to Agrobaterium tumefaciens strain LBA4404 and transformed into wild-type Arabidopsis by floral dip method. The T0 generation seeds were germinated and selected on 1/2 MS medium supplemented with 50 mg/L phosphinothricin ammonium. The transformants were selected and identified by PCR with the above gene cloning primers. Homozygous OX lines were selected by genotyping of T2 generations on screening medium and identified by PCR with gene specific primers and semi-quantitave RT-PCR with primer pairs: 5′-ATGATTTCTTCCGTGTCCTTTCC-3′ and 5′-CTACCAGCACTGATCCGTTTCC-3′.

Phenotypic analysis of Arabidopsis plants

For germination analysis, the seeds of GsJ11 OX, WT and mutant lines were germinated on 1/2 solid medium supplemented with 0, 6, 7 or 8 mM NaHCO3. The germination profiles were recorded according to emergence of seed radicals for consecutive 7 days after sowing, and photos were taken to display the growth performance of each line. For each experiment, 120 seeds were used in total and experiments were repeated at least in three times.

To investigate the stress resistance at seedling stage, 7-days-old seedlings of GsJ11 OX, WT and mutant lines were grown on normal 1/2 MS solid medium and then transferred to medium supplemented 6 or 7 mM NaHCO3 for 6 days. After that, we measured and recorded the primary root length of every seedling for statistics analysis. For each experiment, 15 seedlings were used in total and all experiments were also repeated at least three times.

To characterize the phenotype at adult stage, the seeds of each line were sown in the mixture of garden soil: peat moss: vermiculite (volume ratio, 1:1:1). 4-week-old plants were irrigated with150 mM NaHCO3 after every 3 days for a total of 14 days. The photographs were taken before and after 14-day treatment.

Statistical analysis

Each group experiments were repeated at least three times. Data were expressed as mean ± SD. Data were analyzed statistically using Duncan’s multiple range tests or Student’s t test. Results were considered statistically significant different when P < 0.05.

Results

Identification and classification of DnaJ family genes

DnaJ proteins are important components for protein translation, folding, unfolding, translocation, and degradation in cells (Yang et al. 2010). In this case, we aimed to identify all DnaJ family members from soybean genome. Totally 214 non-redundant DnaJ genes were identified by using the soybean database. All the obtained protein sequences were subjected to SMART and Pfam online tools to ensure that the potential candidates contain J-domain; those lacking the conserved domain were ignored. Finally, we obtained 196 DnaJ family members, and the detailed information of them was exhibited in Table 1. Interestingly, dissimilar to other Hsp protein families, the average molecular mass and isoelectric point of J-proteins are significantly large. The polypeptides range from 68 to 2589 amino acids long, with calculated molecular mass between 7.72 and 283.51 and isoelectric point values of 4.21–11.09.

Table 1 The detailed information of DnaJ gene family members in soybean
Full size table

For a better understanding of DnaJ family in soybean, we performed structural classification of J proteins. As shown in Online Resource 2, the DnaJ proteins could be clustered into three types (I, II and III). Type I J proteins, as the typical DnaJ molecular chaperone model, contain four domains, an N-terminal J-domain was connected with a glycine-rich domain (G/F domain), and followed by a zinc-finger domain and a distal carboxy-terminal (C-terminal) domain, whereas type II J proteins lack the zinc-finger domain. Type III J proteins only have a J domain but lack the other sequence features that are found in type I and II members of the family.

Multiple sequence alignments and phylogenetic analyses of DnaJ family

To further clarify the characteristics of the conserved domain architectures, we aligned all soybean J proteins by ClustalX2.0 (Fig. 1). Coinciding with the classification results above, the 196 DanJ proteins were able to be clustered into three groups with 20 type I, 23 type II and 153 type III, respectively. All type I J-proteins possess 4 characteristic domains including a compact helical J domain, a proximal G/F domain which is rich in Gly/Phe, a (CXXCXGXG)4 zinc-finger domain and a less conserved carboxy-terminal (Fig. 1a). Type II J proteins are similar to type I except they do not have zinc-finger domain (Fig. 1b) and type III just have J-domain (Fig. 1c). Furthermore, we found a highly conserved tri-peptide composed of His, Pro and Asp (HPD motif) in all of J domains which may be crucial for functions of J-proteins.

Fig. 1
figure 1

Multiple sequence alignments of soybean DnaJ family members. Multiple sequence alignments of type I (a), type II (b) and type III (c) DnaJ proteins were performed by ClustalX2.0 program. Conserve domains of each type of DnaJ were marked with colored box. (Color figure online)

Full size image

To further clarify the complexity of DnaJ proteins in soybean, we then explored the evolutionary relationships among DnaJ proteins of soybean. A phylogenetic tree was constructed by maximum likelihood (ML) method with 500 bootstrap replicates using all the full-length J-proteins from soybean (Fig. 2). The data showed that these DnaJ sequences fall into three clusters, consistent with above the sequence alignment results. This work provides a considerable basis for structural and functional characterization of soybean J proteins.

Fig. 2
figure 2

Phylogenetic analysis of DnaJ family proteins. A phylogenetic tree was constructed by using MEGA 5.0 to determine evolutionary distance among the members of DnaJ protein family of Glycine max. The maximum likelihood (ML) method was used for construction of the tree and the reliability of the branches was inferred from a bootstrap analysis of 500 replicates. Each cluster was indicated by a specific color, type I, type II and type III were marked in green, pink and blue, respectively. (Color figure online)

Full size image

Exon/intron organization analysis and chromosomal locations of DnaJ family genes

It is widely believed that exon/intron structural divergences have a crucial impact on the evolution of multiple gene families (Xu et al. 2012). Paralogous genes always have near resemblance in exon/intron organization (Rogozin et al. 2005). By analyzing gene structure, we can get some information about the evolutionary mechanism underlying the genesis of gene families (Zhao et al. 2016). Therefore, to further understand the structural diversity of DnaJ family genes, we investigated the exon/intron organization by comparing the CDS sequences with their corresponding genome sequences of individual DnaJ genes in soybean (Fig. 3). According to the analysis result, we found that the majority of gene pairs exhibited highly conservation, such as the exon numbers and exon length. The exon numbers in type I and type III have considerable variation between these two groups, meaning these two types of DnaJ genes have different tendency from origin by gaining or losing intron. In general, the similar exon/intron distributions were presented in the same phylogenetic group, suggesting soybean DnaJ had undergone gene duplications during the evolutionary processes.

Fig. 3
figure 3

Exon/intron organization structure of soybean DnaJ family genes. Schematic diagram for exon/intro structures of 196 DnaJ family genes identified from soybean. Diagram was generated by GSDS online tool using coding sequences and corresponding genome sequences. Yellow boxes, blue boxes and lines represent exons, upstream/downstream and introns, respectively. The sizes of exons and introns can be estimated using the scale at the bottom. (Color figure online)

Full size image

As shown in Fig. 4, 195 of the 196 soybean DnaJ genes are widely but biasedly distributed on a total of twenty soybean chromosomes, only one gene (Glyma.U012100) is mapped on the unattributed scaffold. The number of DnaJ genes in each chromosome is various. Chromosome 17 has only 4 soybean DnaJ genes while chromosomes 8, 12 and 13 have 18, 15 and 14 DnaJ genes, respectively. Moreover, we found that some chromosomal regions have dense distribution of DnaJ genes, such as the end of chromosome 3, 13 and the center of chromosome 8. However, no substantial clustering of soybean DnaJ genes was exhibited on the map.

Fig. 4
figure 4

Chromosomal locations of DnaJ family genes in soybean. Chromosomal locations of soybean DnaJ family genes on all 20 chromosomes were performed by using MapInspect software. The chromosomes are represented by green bars and the numbers of chromosomes are shown at the top of each bar. The scale on the left of the image represents the length of the chromosome. (Color figure online)

Full size image

Differential expression profiles of G. soja DnaJ genes under alkaline stress

Abiotic stresses can obviously inhibit the growth, development and yield of most crop plants (Jose and Thomas 2015). Previous studies reported that the expression of DnaJ genes could be induced by multiple environmental stresses, such as salinity, high and low temperature, and oxidative stresses. However, we still knew a little about the functions of DnaJ proteins in alkaline stress. For this reason, we explored the whole DnaJ gene family expression profiles under 50 mM NaHCO3 (pH 8.5) in G. soja G07256 using our previous RNA-seq data (Fig. 5a). Amongst them, 27 DnaJ genes (no sequence in type I, Glyma.18G127500 in type II and the others 26 sequences in type III) exhibited differential expressions upon alkaline stress. Notably, we noticed from heat map that the majority of differently expressed genes were distributed to type III DnaJ genes. Thus, we inferred type III J-proteins were mostly involved in response to alkaline stress. We also found that four type III DnaJ genes had similar expression profiles and showed significantly increased transcript levels at early time point (3 h) after NaHCO3 treatment. Coincidentally, those genes were neighbors with each other according to the phylogenetic tree (Fig. 5a). Therefore, we chose one representative gene, GsJ11 (Glyma.11G077400), which was highly induced by NaHCO3 treatment for further functional dissection.

Fig. 5
figure 5

Differential expression profiles of GsJ11and characteristics of OX lines and mutant lines. a Expression profiles of DnaJ genes in G. soja roots under 50 mM NaHCO3 treatment according to the RNA-seq data. The color scale represents genes log2 fold changes (LogFC) to 0 h. Red and green squares represent high or low levels of transcript abundance compared with 0 h, respectively. b The spatial expression of GsJ11 was detected in various tissues of wild soybean. c Relative expression analysis of GsJ11 induced by NaHCO3 revealed by qRT-PCR. d Identification of atj11 T-DNA insertion mutant. Lane 1 was WT sample and lane 2 was distilled water; lane 3–12 was atj11 mutant samples. e Expression levels of endogenous AtJ11 and GsJ11 in WT, atj11 mutant and three GsJ11 overexpressed lines, respectively. (Color figure online)

Full size image

Spatial expression pattern of GsJ11 and its response to bicarbonate alkali stimuli in G. soja

To determine the spatial expression patterns, we measured the endogenous GsJ11 expression levels in root tip, stem, young leaf, old leaf, seed, pod and flower of G. soja using quantitative real-time PCR (Fig. 5b). As shown in Fig. 5b, GsJ11 was widely expressed in all vegetative and reproductive tissues and organs. GsJ11 had a high expression levels in young leaf and flower but moderate expression levels in root tip, seed and pod.

To examine the response of GsJ11 to bicarbonate alkali, qRT-PCR were carried out using the RNA samples extracted from the root tips of 21-d-old G. soja 07256 seedlings treated with 50mM NaHCO3 for different time durations. Consistent with our previous RNA-seq data, the transcript accumulation of GsJ11 was obviously increased after treatment (Fig. 5c). Specifically, the transcript level of GsJ11 maximized with more than twofold increase at 3 h and then dropped down to normal level, implicating that GsJ11is actively responsive to bicarbonate alkaline stress at early stage.

Ectopic expression of GsJ11 enhances plant tolerance to NaHCO3

To investigate the physiological function of GsJ11 gene,we ectopically expressed GsJ11 controlled by CaMV35S promoter in Arabidopsis. Three homozygous T3 OX transgenic Arabidopsis lines (#1, #2, #3) were generated and selected for further study. By contrast, we obtained one AtJ11 (At4G36040) T-DNA insertion lines from TAIR (SALK_015630 C). The homozygous mutant line, atj11, was authenticated by T-DNA border primers and gene-specific primers (Fig. 5d). We confirmed that expression of endogenous AtJ11 was abolished in mutant line whereas it was present in WT and GsJ11 OX lines using semi-quantitative RT-PCR using AtJ11 gene-specific primers (Fig. 5e up). Meanwhile, we identified the transcription abundance of each GsJ11 OX line by semi-quantitative RT-PCR using GsJ11 gene-specific primers (Fig. 5e down). The analysis result confirmed that these transgenic lines were independent and had high enough exogenous GsJ11 expression levels to function. However, the WT and mutant plants did not show any GsJ11expression.

To determine the physiological functions of GsJ11 under alkaline condition, we examined the effect of J11 on seed germination. The WT, OX and atj11 mutant seeds were germinated on 1/2 Murashige and Skoog (MS) solid media containing 0, 6, 7 and 8 mM NaHCO3, respectively. Without NaHCO3 treatment, each line seeds showed similar and rapid seed germination and young seedling growth (Fig. 6a). However, on the media containing NaHCO3, the seed germinations were generally inhibited. On medium with 6 mM NaHCO3, the germination rates of all lines were significantly decreased in early stage, but the seed germination rate of GsJ11 OX lines were significantly higher than WT and atj11 mutant lines under the same condition after 5 days of seed sowing. The germination percentages were nearly 50% for mutant line and 60% for the WT, whereas 80–90% for the OX seeds. This trend was even more pronounced in the presence of higher NaHCO3 concentrations (Fig. 6a, b).

Fig. 6
figure 6

Ectopic expression of GsJ11 enhances plant tolerance to NaHCO3 during the seed germination stage. a The growth performance of WT, GsJ11 OX and atj11 seedlings on medium containing 0, 6, 7 or 8 mM NaHCO3, respectively. b Seed germination rates of WT, GsJ11 OX and atj11 lines

Full size image

To further determine if GsJ11 overexpression can enhance plant tolerance to NaHCO3, 7-day-old seedlings of each line grown on the normal medium were transferred to media containing 0, 6 or 7 mM NaHCO3 and were continued to grow for 6 days. The seedlings on the normal medium did not show obvious difference of root elongation among WT, GsJ11 OX and atj11 mutant lines (Fig. 7a). However, on medium containing 6mM NaHCO3, the root elongation of atj11 was significant inhibited compared to WT, and this reduction in root development was more pronounced in the presence of 7 mM NaHCO3 (Fig. 7a, b). However, ectopic expression of GsJ11 (OX lines #1, #2 and #3) could significantly promote root elongation (Fig. 7a, c). These results suggested that GsJ11 can enhance plant resistance to NaHCO3.

Fig. 7
figure 7

Ectopic expression of GsJ11 promotes seedling root elongation under NaHCO3 treatment. a Phenotypes of WT, atj11 and GsJ11 OX seedlings under normal and alkaline stress. b Primary root length of WT and atj11 seedlings. c Primary root length of WT and GsJ11 OX seedlings. *P < 0.05; **P < 0.01 by Student’s t test

Full size image

To provide more evidence for our discovery, we treated the adult seedlings of WT, atj11, three GsJ11 OX lines with NaHCO3. 4-week-old plants of each line grown in soil were irrigated with 150 mM NaHCO3 solution every 3 days for total 14 days. Consistent with the observations from young seedlings, all lines exhibited similar growth in the pots irrigated with water (Fig. 8a) but their phenotypes demonstrated significant differences after irrigation with NaHCO3 solutions for the indicated time. All of the GsJ11 OX lines had stronger growth, higher survival rate and more green leaves than the WT. However, the mutant line was the worst and most plants turned yellow even dead after treatment (Fig. 8a). To further confirm our observation, the total chlorophyll contents of all lines were measured. As shown in Fig. 8b, the chlorophyll contents were generally decreased in all lines after NaHCO3 treatment, but GsJ11 OX lines still had relatively the highest pigment contents among the WT and atj11 lines. More specifically, the mean value of chlorophyll contents was 79.73 (μg/g FW) in wild type, and 70.31 (μg/g FW) in mutant line, while the value of OX lines were from 86.48 to 93.28 (μg/g FW). Moreover, it is widely accepted that malondialdehyde (MDA) is generated by peroxidation process of membrane lipid, and high MDA level can lead to damage of cell membrane, so MDA is usually seen as an indicator to measure oxidative damage (Wang et al. 2016; Weber et al. 2004). We then determined the MDA contents in all plants with normal and alkaline treatments. The data showed that MDA levels in WT and atj11 lines were much higher than GsJ11 OX lines (Fig. 8c). For example, the MDA contents in WT and mutant lines were approximately 40 (μmol/g FW) under NaHCO3 treatment, whereas the MDA contents in GsJ11 OX lines were only half or less. These results support our conclusion that GsJ11 enhances plant NaHCO3 tolerance while loss of J11 increases sensitivity to NaHCO3.

Fig. 8
figure 8

GsJ11 enhances plant tolerance to NaHCO3 at the adult stage. a Phenotypes of WT, GsJ11 OX and atj11 plants in response to alkali at adult stage. b The total chlorophyll contents of WT, GsJ11 OX and atj11 plants. c The total MDA contents of WT, GsJ11 OX and atj11 plants. *P < 0.05; **P < 0.01 by student’s t test

Full size image

Expression patterns of abiotic stress-responsive marker genes

It is suggested that alkaline stresses induce the expression of numerous relative stress-inducible genes in the progress of plant adaptation (Alhendawi et al. 1997). For a further understanding about the role of GsJ11 in plant tolerance to NaHCO3, we measured the transcript levels of seven stress-inducible genes (NADP-ME, H + -ATPase, H + -PPase, KIN1, RD29A, COR47 and RD22). The results of qRT-PCR showed the expressions of all selected genes were rapidly and significantly induced by NaHCO3 treatment in WT and OX lines, however these marker genes showed low induction in the mutant line (Fig. 9). H + -ATPase, H + -PPase, NADP-ME, RD29A and KIN1were significantly regulated in OX lines than in wild type. Although the transcription levels of COR47 and RD22 also increased, there were no significant differences between all lines. In summary, these qRT-PCR analyses suggested that GsJ11 promoted bicarbonate resistance by regulating the expression of relative stress-induced gene, and deletion of J11 caused low responses of those genes.

Fig. 9
figure 9

Expression patterns of abiotic stress-responsive marker genes. Transcript levels of stress-inducible marker genes in WT, GsJ11 OX and atj11 seedlings determined by qRT-PCR. Actin2 gene was used as an internal control. Error bars represent standard deviations (SD) of three independent biological repeats and three technical repeats. *P < 0.05; **P < 0.01 by Student’s t test

Full size image

Discussion

Proteins are always at the risk of losing their functional conformation due to various environmental stresses in the native conditions. Molecular chaperones can keep proteins in proper conformations by facilitating the folding or refolding of the misfolded proteins. Heat shock 70 kDa proteins (Hsp70s) are known as a kind of ubiquitous chaperones which participate in a myriad of biological processes (Bukau et al. 2006; Hartl and Hayer-Hartl 2009). Many protein folding and refolding are driven by a diverse class of cofactors: DnaJ (Kampinga and Craig 2010). In this study, we focused on the diversity of this large DnaJ family in soybean genome, identified and characterized an alkaline stress-responsive gene GsJ11 based on genome-wide analysis.

Previous reports have revealed the complexity and diversity of DnaJ gene family in Arabidopsis, rice and yeast (Rajan and D’Silva 2009; Sarkar et al. 2013; Walsh et al. 2004a). In this study, we identified 196 non-redundant DnaJ genes in soybean genome (Table 1) compared to 116 DnaJ genes in Arabidopsis and 104 DnaJ genes in rice. Depending on the presence of conserved domain, DnaJ protein family could be divided into three clusters (Walsh et al. 2004b). Similar with Arabidopsis and rice, all types of DnaJ proteins were detected in soybean based on their structures (Online Resource 2). For example, the proteins in the same clusters share the same domain architectures, implicating the proteins within one cluster should have similar functions and each type of J proteins are crucial and play different roles in cellular processes.

All of the J proteins contain a highly conserved signature J domain (Cyr et al. 1992). As shown in multiple sequence alignments, the J domain are generally located in the N-termini of DnaJ proteins and are highly exhibited sequence conservation, and a highly conserved HPD motif is existed between helix II and helix III in all J domains (Fig. 1), suggesting that the J domain and HPD motif are necessary for a J protein to carry out its function.

Strikingly, according to the phylogenic tree, we found the number of each cluster members is different; the members in type III are significantly more than those in the other two types (Fig. 2). These phenomena were also found in other species, such as rice and Arabidopsis (Sarkar et al. 2013). Thus, we hypothesized that type III J proteins are widely involved in cellular physiological processes by more flexible modes. On the other hand, the exon/intron organization analysis also confirmed that type I and III DnaJ genes have considerable variation in distribution (Fig. 3). Whereas the genes in same phylogenetic branch have consistent exon numbers and lengths, suggesting DnaJ genes had undergone gene duplication in evolution.

There is plenty of evidence that DnaJ genes can be induced by multiple abiotic stresses, such as heat, cold, wounding or high-salinity (Kong et al. 2014, So et al. 2013). For example, a tomato chloroplast-targeted DnaJ protein, SICDJ2, could enhance plants tolerance to heat stress by protecting Rubisco activity (Wang et al. 2015). Three small type III DnaJ proteins were involved in optimization of photosynthetic reaction during high light condition (Chen et al. 2010). Furthermore, the Arabidopsis chaperone J3’s expression could be induced to salt at alkaline pH (Yang et al. 2010). However, only few focus on bicarbonate stress. Thus, we explored the expression profiles of G. soja DnaJ genes under 50 mM NaHCO3 based on RNA-seq data (Fig. 5a). We obtained 27 significant differential expressed genes out of 196 DnaJ genes. Notably, 26 in those differential expressed genes were distributed in type III, so we postulate that these abiotic-responsive DnaJ genes are majorly concentrated in type III and play crucial roles against abiotic stresses in plants.

In this paper, we concentrated on a novel type III DnaJ family gene from G. soja, GsJ11, which was rapidly induced by bicarbonate stress according to our RNA-seq data. Here, we detected the inducible expression patterns of GsJ11 in G. soja root during NaHCO3 treatment as shown by qRT-PCR result (Fig. 5c), coinciding with our previous findings of the other alkaline-induced genes (Liu et al. 2015; Yu et al. 2016). To further evaluate the function of GsJ11 in response to alkaline condition, gain-of-function of GsJ11 plants and loss-of-function mutants of its homologous gene in Arabidopsis were generated for phenotypic analyses. Compared to neutral salts, alkaline salt stress has more inhibitory effect on seed germination and physiological characteristics. As expected, the plants with ectopic expression of GsJ11 exhibited greater alkali tolerance than the wild-type plants, such as higher germination rates (Fig. 6a, b), less root elongation inhibition (Fig. 7a–c), higher survival rates (Fig. 8a), higher chlorophyll contents (Fig. 8b) and much lower MDA contents (Fig. 8c). By contrast, atj11 mutant line displayed completely opposite phenotypes. Taken together, those evidences strongly suggest GsJ11 facilitates alkaline tolerance in plants. Furthermore, we also found that the transgenic lines with more transcripts (Fig. 5e) have greater tolerance (Figs. 6, 7, 8). Namely, the alkaline resistance of GsJ11 overexpressing line is positively correlated with its abundance of GsJ11 transcripts.

Evidence has been provided that most of bicarbonate-induced genes are involved in metabolism, signal transduction and transcription (Alhendawi et al. 1997). It has been reported that NADP-MEI, NADP-MEII and V-H+-PPase are highly responsive to bicarbonate stress and can prevent plant cells from the damage of environmental stress by regulating intracellular pH (Fushimi et al. 1994). In this study, we detected the transcription levels of three bicarbonate-response marker genes, including H + -ATPase, H + -ATPase and NADP-Me in GsJ11 OX, WT and atj11 lines, respectively (Fig. 9). We also explored other stress-regulated genes such as RD29A, KINI and COR47, which could be significantly induced by drought, cold and ABA (Kurkela and Franck 1990; Seki et al. 2003; Wang et al. 1995). And we found that the transcript levels of those genes in GsJ11 OX lines were rapidly up-regulated after alkaline treatment compared with WT and atj11 lines. H + -ATPase, H + -PPase, NADP-ME, RD29A and KIN1 were strongly up-regulated in GsJ11 OX lines than WT line, COR47 and RD22 did not exhibit obvious difference between GsJ11 OX and WT lines although their expression levels were also increased. These results suggested that GsJ11 may participate in alkali stress signaling transduction by regulating genes expression and subsequently enhances alkaline resistance in plants.

Based on emerging research, DnaJs participate in some essential biochemical pathways in plant cell. We have known that J3 activates PM H+-ATPase activity by repressing PKS5 kinase activity (Yang et al. 2010). A type III J-protein in Arabidopsis, J20, interacts with DXS enzyme, the first enzyme of MEP pathway and identifies unfold or misfolded form of DXS and target them to the Hsp70 system for proper folding under normal conditions or degradation upon stress conditions (Pulido et al. 2013). Two other small chloroplast-targeted J-protein, J11 and J8 participate in stabilization of PSII complexes and are involved in the folding or assembly processes of Rubisco. Moreover, LeCDJ1, plays a similar role in the maintenance of PSII activity under abiotic stress (Kong et al. 2014). A chloroplast Hsp70 could directly interact with SlCDJ2, two component function together in preventing Rubisco protein under stress conditions(Wang et al. 2015). Thus, we speculate that GsJ11 has a similar function pattern in the response to alkaline stress. The plant growth in alkaline soil is severely inhibited due to impaired chlorophyll synthesis and root respiration. GsJ11 may identify particular client protein (Hsp70) and help them recover proper conformation, sequentially improve the stability of the composites, such as PSII complexes and respiratory chain, finally maintain the normal physiological and biochemical functions in alkaline stress.

In addition, the spatial expression analysis showed that GsJ11 had highly expression levels in young leaf and flower besides root (Fig. 5b). In Arabidopsis, a type III J protein AtDjC17 caused altered root hair development (Petti et al. 2014). Those evidences provided a clue that type III DnaJ proteins may also play roles in plant development process.

In conclusion, we defined systematic diversity of soybean DnaJ gene family and found type III DnaJ proteins are involved in response to abiotic stresses. Here we identified a novel type III DnaJ gene, GsJ11, from G. soja, and confirmed it can enhance plant tolerance to bicarbonate stress. Furthermore, the precise function and the mechanism of GsJ11 were investigated at the present of NaHCO3. Our results provide references for further studies on family biological functions of DnaJ proteins.