Introduction

Plants are protected from a wide range of pathogens by specialized immune responses, including the hypersensitive resistance (HR), systemic acquired resistance (SAR), and induced systemic resistance (ISR). SAR is an important part of plant defenses, providing long-term, broad-spectrum resistance that is effective against a wide variety of fungal, viral, and bacterial pathogens in tissues distal to the initial site of infection (An and Mou 2011; Friedrich et al. 1996; Robert-Seilaniantz et al. 2011; Sticher et al. 1997). The accumulation of SA or its functional analogs, such as 2,6-dichloroisonicotinic acid, benzothiadiazole S-methyl ester, can activate the SAR response and pathogenesis-related (PR) gene expression (Metraux et al. 1991; Ward et al. 1991; Lawton et al. 1995). ISR, normally induced by Jasmonic acid (JA) and ethylene, always confers resistance to insects and some necrotrophic pathogens (Durrant and Dong 2004; Gfeller et al. 2006; Kessler and Baldwin 2002).

The NPR1 gene was discovered in Arabidopsis mutants with compromised resistance against pathogens (Cao et al. 1994). The AtNPR1 mutants lost or had reduced transcriptional expression of some PRs, including PR-1 and lost the capacity for SAR, suggesting that the AtNPR1 gene could prevent secondary infection of virulent pathogens. NPR1 is also involved in JA/ET-dependent ISR, which is triggered by certain Pseudomonas fluorescens strains. NPR1 has been shown to modulate cross talk between SA- and JA-dependent defense pathways (Dong 2004; Iavicoli et al. 2003; Pieterse and Van Loon 2004). The AtNPR1 gene encodes a protein that contains two domains, a BTB/BOZ domain and an ankyrin repeat domain (Cao et al. 1997; Ryals et al. 1997). These domains mediate protein-protein interactions and are present in a wide range of proteins with diverse functions (Aravind and Koonin 1999). NPR1 shares structural features with the transcriptional regulator IκB, which mediates innate immune responses in animals (Baldwin 1996). A recent review on NPR1 described its mechanisms of action and indicated that NPR1 functions as a bona fide SA receptor (Pajerowska-Mukhtar et al. 2013). During the induction of SAR, Arabidopsis NPR1 protein oligomers in the cytoplasm depolymerize and release active monomeric NPR1, which is translocated into the cell nucleus via a C-terminal bipartite nuclear localization signal (NLS) (Kinkema et al. 2000; Somssich 2003). In the nucleus, NPR1 protein interacts with the TGA (TGACG motif-binding factor) family of basic leucine zipper (bZIP) transcription factors (Despres et al. 2000; Fan and Dong 2002; Kim and Delaney 2002; Subramaniam et al. 2001; Zhang et al. 1999; Zhou et al. 2000). These bZIP transcription factors are involved in SAR and SA-dependent activation of PR genes (Niggeweg et al. 2000). In the absence of pathogen challenge, NPR1 is continuously cleared from the nucleus by the proteasome, preventing activation of SAR. SA accumulation promotes NPR1 phosphorylation, which resulted in its recruitment by a Cullin3-based ubiquitin ligase. Turnover of phosphorylated NPR1 is required for full induction of target PR genes and the establishment of SAR in plant defense (Mukhtar et al. 2009; Spoel et al. 2009).

The Arabidopsis NPR1 gene homologs have been extensively studied. Besides AtNPR1, five other homologs have been found in Arabidopsis. Phylogenetic analysis of these six homologs revealed three distinct groups (Hepworth et al. 2005; Zhang et al. 2006). AtNPR1 and AtNPR2 are in the first group. The function of AtNPR2 has not been reported, but it is supposed to participate in the SAR and SA signaling pathway as AtNPR1. Proteins encoded by the second group, which includes AtNPR3 and AtNPR4, are thought to directly bind to SA with different binding affinity (NPR3, K d  = 981 nM; NPR4, K d  = 46 nM) and play a role in mediating the degradation of NPR1 and decreasing or keeping its function in basal defense (Fu et al. 2012). The members in the third group, encoding Arabidopsis NPR5 and NPR6, also known as BLADE-ON-PETIOLE 1 (BOP1) and BOP2, are associated with the development of lateral organs (Hepworth et al. 2005). A recent report found that the Arabidopsis BOP proteins are responsible for abiotic resistance induced by methyl jasmonate (MeJA) in the wild type (Canet et al. 2012). With the development of genome sequencing, NPR1 gene homologs have been identified from the whole-genome sequences of many plants, such as Vitis vinifera (Le Henanff et al. 2009), Populus trichocarpa (Shao et al. 2013), Carica papaya (Peraza-Echeverria et al. 2012), and Persea americana (Backer et al. 2015). Their functions in disease resistance have been widely reported. Overexpression of the AtNPR1 gene has been shown to result in enhanced bacterial and fungal resistance in many plants. Furthermore, overexpression of an endogenous NPR1 orthologue (named MdNPR3 in our research) in apple increased resistance to fire blight and two other major fungal pathogens (Venturia inaequalis and Gymnosporangium juniperi-virginianae) of apple (Malnoy et al. 2007). A study on two banana NPR1-like genes, MNPR1A and MNPR1B, found that they could induce PR-1 gene expression and restore pathogen resistance as in the Arabidopsis npr1 mutant (Yocgo et al. 2012). As master regulatory genes in defense responses, the functions of NPR1 genes have been reported in many studies, but the full complement of gene members and their functions in disease resistance in apple have not been described.

Alternative splicing (AS) is a process that generates two or more different splice variants from the same pre-mRNA molecule using different splice sites. These splice sites are recognized by a large RNA-protein complex, the spliceosome. Four main types of such splicing are known: exon skipping, alternative 5′ and alternative 3′ splice sites, and intron retention. AS not only contributes to proteome diversity but also generates truncated proteins that are potentially regulatory or detrimental to the cell. It also plays a role in gene expression by regulating transcript levels through the production of isoforms that are degraded by the nonsense-mediated decay pathway (Mastrangelo et al. 2012; Syed et al. 2012). In Citrus clementina, a defense-related gene, acidic chitinase II, can produce an additional transcript containing a premature termination codon after the first 135 amino acid that is inducible by pathogen infection and MeJA treatment (Del Carratore et al. 2011).

Apple (Malus domestica Borkh.) is one of the most important fruit crops cultivated in China and worldwide. Fungal diseases, such as Marssonina coronaria, Glomerella cingulata, and Valsa canker, pose a great threat to the growth and production of apple trees. In this study, eight NPR1 gene homologs were identified from the apple genome and their full complementary DNA (cDNA) sequences were cloned; furthermore, AS of some NPR1 homologs were found. The tissue- and organ-specific expression of the NPR1 homologs and their differential expression in response to M. coronaria, SA, and MeJA induction between resistant and susceptible cultivars were investigated. Some specific members and AS variants of NPR1 were only found in the resistant cultivar. This implies that these AS forms may be related to disease resistance responses in apple. Our findings provide an invaluable resource for further study and functional characterization of the NPR1 homologs in apple.

Materials and methods

Identification and annotation of apple NPR1 homologs

Six Arabidopsis NPR1 homolog protein sequences including AtNPR1–6 were downloaded from The Arabidopsis Information Resource (TAIR, http://www.arabidopsis.org/) with the accession numbers AT1G64280.1, AT4G26120.1, AT5G45110.1, AT4G19660.1, AT2G41370.1, and AT3G57130.1. Their amino acid sequences were used as queries for BLASTp searches at the Genome Database for Rosaceae (GDR, https://www.rosaceae.org/) with an E-value cutoff of 1e − 003. The results were submitted to the Simple Molecular Architecture Tool (SMART, http://smart.emblheidelberg.de) and Pfam protein family databases (http://pfam.xfam.org/) to detect the presence of BTB/POZ and ankyrin domains. Sequences containing both domains were selected as putative NPR1 homologs. Afterward, the coding sequences identified from the apple genome data were inspected and revised according to the M. domestica and M. pumila expressed sequence tags (ESTs), which are available from the NCBI web server (http://www.ncbi.nlm.nih.gov/).

Plant materials and treatments

Five-year-old apple trees of the “Qinguan” and “Pacific Rose” cultivars and a wild apple species Malus baccata Borkh were grown in the field under a rain-shelter at the experimental station of Northwest A&F University (Yangling, Shaanxi, China) and used as materials throughout the study. The apple cultivars Qinguan and Pacific Rose were shown to have different levels of resistance to M. coronaria, which causes apple brown spot. Pacific Rose was previously reported as susceptible, while Qinguan and M. baccata are resistant and highly resistant, respectively, to the disease (Yin et al. 2013).

Tissue- and organ-specific expression of the apple NPR1 homologs was investigated in Pacific Rose. New shoots (near the apices of newly-growing shoots and 3–4 mm in diameter), leaves (the third to fifth fully expanded young leaves beneath the shoot apices when newly-growing shoots were 40–60 cm in length, without petioles), flowers (with receptacle), pericarp, pulp (from mature fruits of about 7 cm in diameter), and seeds of mature fruit were collected for total RNA isolation and expression analysis.

The induction of NPR1 expression by SA and MeJA was examined in Qinguan and Pacific Rose, and by M. coronaria inoculation in Pacific Rose and M. baccata. SA and MeJA treatment was performed by spraying leaves with 100 mM SA or 50 mM MeJA solutions once at 8:00 a.m. on whole sample plants. Leaves sprayed with sterile distilled water were used as controls. Each treatment and control were performed on 15 trees. All treated plants were sampled at 0, 1, 3, 6, 12, and 24 h postinoculation (hpi) for subsequent RNA isolation and NPR1 homolog expression analysis. M. coronaria inoculation was conducted on 90~120 detached leaves according to Li et al. (2014) under a 24-h light/0-h dark photoperiod and 15~20 sample leaves were collected at 6, 24, 48, 72, 96, and 144 hpi. All treatments were performed in triplicate.

RNA extraction and cDNA synthesis

Total RNA was extracted using a modified SDS-LiCl method (Hou et al. 2011). All reagents were prepared with 1 ‰ DEPC treated water. Potassium acetate buffer (pH 5.0) was added to precipitate the SDS and macromolecule material. The chloroform/isoamyl alcohol extraction step was repeated three or four times until the protein phase was completely removed and the water phase became clean. RNA integrity was assessed on 2 % TAE agarose gel under non-denaturing conditions. The first-strand cDNA was synthesized from mature mRNA using the PrimeScriptII RT reagent Kit (Takara, Dalian, China) with Oligo(dT) primers according to the manufacturer’s guidelines. Intron-spanning primers were used to assess genomic DNA contamination.

Molecular cloning of NPR1 homologs in apple

To confirm the identified apple NPR1 homolog loci in the apple genome, eight pairs of gene-specific primers targeting the exons were designed based on the putative NPR1 homolog gene sequences from the apple genome database and the apple EST sequences (Supplementary Table S1). Putative full-length coding sequences of the apple NPR1 homolog genes were amplified by RT-PCR from both Qinguan and Pacific Rose leaves with cDNA templates instead of genomic DNA because some of the gene fragments were too long to amplify. PCR was performed with pfu Taq DNA polymerase (Thermo, Shanghai, China), using the manufacturer recommended reaction conditions. The PCR products were retrieved and purified with a DNA Fragment Purification Kit (Tiangen, Beijing, China). A tails were added to the PCR fragments using rTaq (Takara) at 72 °C for 35 min, and the fragments were inserted into the pMD-19T cloning vector (Takara). Positive clones were verified by blue/white colony assays and sequencing. The absence or presence of an AS variant was judged by agarose gel or sequencing of 10 independent clones from the RT-PCR product of NPR1s.

Sequence alignment, phylogenetic analysis, and exon-intron structure analysis of apple NPR1 homologs

Because the eight deduced apple NPR1 homolog cDNA sequences and most transcript variants were detected in Pacific Rose, the cDNA sequences of the apple homologs in Pacific Rose were used for subsequent sequence alignment analyses. The apple NPR1 protein sequences deduced from the cDNA fragments were aligned using the program ClustalX with BLOSUM30 as the protein weight matrix. A phylogenetic tree of apple NPR1 protein sequences was constructed with 28 other putative NPR1-like proteins sequences downloaded from the NCBI database from eight plant species (Table 1) except for CpNPR1–4 (the sequences were obtained from the supplementary material of Peraza-Echeverria et al. (2012)). Phylogenetic trees were constructed using the neighbor-joining method by the program MEGA5 (Molecular Evolutionary Genetics Analysis) with the p-distance and complete deletion option parameters selected (Saitou and Nei 1987; Tamura et al. 2011). A diagram of the exon/intron organization was generated by the online Gene Structure Display Server (GSDS, http://gsds.cbi.pku.edu.cn). A gene structure map was generated based on the coding sequence of the apple cultivar Pacific Rose and corresponding genome sequence of apple (Ver 1.0, https://www.rosaceae.org/).

Table 1 Additional NPR1-like proteins from other plants used to construct the phylogenetic tree
Full size table

Gene expression detection by qRT-PCR

Real-time quantitative PCR (qRT-PCR) was performed on a Bio-Rad iQ5.0 platform to detect the expression levels of apple NPR1 homologs in different tissues and in response to different treatments. The reactions were performed in 20-μL volumes containing 10 μL of SYBR® Premix ExTaq II (Takara), 2.0 μL of cDNA template (50 ng/μL), 1.6 μL of gene-specific primers (10 pmol/L), and 6.4 μL of sterile distilled water. The PCR amplification conditions were 95 °C for 1 min, and then 40 cycles of 95 °C for 10 s, 58 °C for 30 s, and 72 °C for 20 s, followed by melting curve analysis in which the products were heated from 58 to 95 °C by 0.5 °C per cycle. Fluorescence was measured during the extension step of each cycle. Specific primers were designed by primer premier 5.0 for each gene, and each pair was subjected to a local blast search to ensure its specificity (Supplementary Table S1). The apple actin gene (GQ339778) was selected as an internal standard (Fan et al. 2011). Three biological replicates were used for each assay. The relative expression level of each gene was calculated according to the 2−ΔΔCT method (Livak and Schmittgen 2001).

Results

The published apple genome contains eight NPR1 gene homologs

A total of eight NPR1-like genomic sequences containing both BTB/POZ and ankyrin repeat domains were identified from the apple genome database (Ver 1.0) and named MdNPR1 to MdNPR8. The locations of these apple NPR1 gene homologs in the apple genome were listed in Table 2. The eight genes were located in seven different chromosomes; two in chromosome 2 and one each in chromosomes 5, 9, 10, 12, 14, and 17. Their location was scattered over different chromosomes and did not show any cluster distribution clues to the expansion of the gene family. Blast analysis using the predicted amino acid sequence of the apple NPR1 homologs showed that MdNPR1 was most similar to AtNPR1 (49.7 % identity) (Table 3), MdNPR2, MdNPR3, MdNPR4, and MdNPR5 were most similar to AtNPR3 (54.7, 59.9, 59.7, and 53.8 % identity, respectively), and MdNPR6 was most similar to AtNPR4 (51.9 % identity). Furthermore, MdNPR7 and MdNPR8 were most similar to AtNPR5 (77.8 and 77.8 % identity, respectively). No homologs in apple were found to be similar to AtNPR2 or AtNPR6.

Table 2 NPR1 homologs’ distribution in genome and their alternative splicing variants in apple cultivars
Full size table
Table 3 Identity of putative amino acid sequences of NPR1 homologs between apple and Arabidopsis
Full size table

Different transcripts of the NPR1 homologs were found in resistant and susceptible apple cultivars

The full cDNA fragments of all eight NPR1 gene homologs were cloned and sequenced in the apple cultivars Qinguan and Pacific Rose. A comparison of all the NPR1 homolog sequences showed that the coding sequences of MdNPR1, 3, 5, and 6 were the same in both varieties (Supplementary Fig. S1~S8). The MdNPR4, 7, and 8 gene sequences of the two varieties differed in several bases, but the protein sequences were unchanged. The MdNPR2 gene sequences of the two cultivars had several different bases leading to two amino acid mutations in the conserved ankyrin domain and two outside the conserved domain region. Two different MdNPR2 transcripts with several base differences at different positions were cloned from Qinguan, implying that gene editing may have occurred. A total of 12 transcripts of the eight NPR1 homologs were found, showing differential splicing of NPR1 homologs. MdNPR1, 2, 4, and 5 had two different splicing variants each, while MdNPR3, 6, 7, and 8 had only one. The variation of spliced size varied from 77 bp which only could be seen by sequencing to 754 bp which could be seen by the agarose gel (Fig. 1). The sequences of the different splicing variants were submitted to GenBank, and their presence or absence in the resistant and susceptible apple cultivars is shown in Table 2. According to the sequences (Supplementary Fig. S1~S8), different forms of alternative splicing were found for the NPR1 homologs (Fig. 2). An alternative 3ʹ splicing site in the last exon was found in MdNPR1, an alternative 5ʹ splicing site in the second exon and an alternative 3ʹ splicing site in the last exon was found in MdNPR2, intron retention in the third intron was found in MdNPR4, and exon skipping in the second exon was found in MdNPR5. The splicing variants MdNPR1a and MdNPR4b were only found in Pacific Rose, but other splicing variants were found in both cultivars.

Fig. 1
figure 1

The RT-PCR result of NPR2 and NPR5 gene homologs from apple. QG and PR indicate the lane from ‘Qinguan’ and ‘Pacific Rose’ respectively; DL2000 is a DNA lander including 2000, 1000, 750, 500, 250, and 100-bp stripes; a, b indicates different variants from alternative splicing

Full size image
Fig. 2
figure 2

Exon-intron structures of NPR1 gene homologs from apple and Arabidopsis. Alt 3’ss alternative 3′ splice sites, Alt 5’ss alternative 5′ splice sites, ES exon skipping, IR intron retention

Full size image

Conserved domain and cysteine site mutations were found in some NPR1 homologs

Multiple sequence alignment was carried out to examine the conservation of the domains in the apple NPR1 proteins at each residue position. The results showed that the apple NPR1 proteins all contain the BTB/POZ and ankyrin repeat domains, except MdNPR5b, which had lost the ankyrin repeat domain (Fig. 3). The 12 apple NPR1 homolog proteins all harbored conserved cysteine residues at C82, C150, C155, and C160, but only the MdNPR1a and MdNPR1b proteins had a conserved cysteine at C216. A key cysteine at C521/529 in AtNPR1 was not conserved in any of the apple NPR1 proteins. These cysteine sites are thought to be important for the formation of functional oligomers, and the mutation or deletion of these cysteine sites may result in functional changes of the NPR1 proteins. The normal transcripts of the MdNPR1-MdNPR6 contained a NPR1_like_C region and NLS in the C-terminus of their protein. However, the alternative splicing variants MdNPR1b, MdNPR2b, MdNP4b, MdNPR5b, MdNPR7, and MdNPR8 did not have the NLS in their C-terminus; their function and localization need further investigation.

Fig. 3
figure 3

Multiple alignment of amino acid sequences of M. domestica and Arabidopsis NPR1 homologs

Full size image

Gene structure and phylogenetic analysis of the apple NPR1 homologs

Gene structure analysis of the apple NPR1 homolog sequences indicated that they had similar exon-intron structures to the NPR1 genes of Arabidopsis. The homologs in group I and II had four exons, while those in group III had two. In addition, the lengths of corresponding exons in each group were almost the same (Fig. 2).

To study the evolutionary relationships among the NPR1 homologs, a phylogenetic tree was constructed with 28 putative NPR1-like proteins from eight plant species (Fig. 4). According to the results, all the selected NPR1 homologs were clustered into three groups, which was consistent with previous findings in Arabidopsis. In group I, only MdNPR1 was located in the same cluster as AtNPR1 and AtNPR2. Notably, Group II had the most numbers of apple NPR1 homologs (five homologs) compared with only two from Arabidopsis, two from Carica papaya, three from Populus trichocarpa, and one from Persea Americana. MdNPR2, MdNPR3, and MdNPR4 were closely clustered with PpNPR1.1, AtNPR3, and AtNPR4. MdNPR5 and MdNPR6 were clustered closely with CpNPR3, PtNPR3, and PtNPR4. Two members (MdNPR7 and MdNPR8) were classified into group III, which was called the BOP group in reference to Arabidopsis.

Fig. 4
figure 4

Phylogenetic analysis of the NPR1 homolog proteins from M. domestica and other plant species. The tree was generated with the MEGA software Ver. 5.0 using the neighbor-joining method. A total of 1000 bootstrap replicates were performed, and values are indicated above the branch points

Full size image

Differential expression levels of MdNPR1 homologs in apple tissue and organs

Investigating the specific expression of a gene in different tissues and organs can help infer its major function. The expression of all eight apple NPR1 homologs was examined in stems, flowers, leaves, pericarp, pulp, and seeds of Pacific Rose (Fig. 5). The results showed that their expression levels in different tissues were significantly different. The expression level of MdNPR1 was highest in pulp (2.09-fold), followed by leaves and relatively low in other tissues. MdNPR2, MdNPR3, and MdNPR4 were expressed at a relatively high level in pericarp compared with leaves, stems, pulp, and seeds. The expression levels of MdNPR2 and MdNPR3 were lowest in flowers, while the expression of MdNPR4 was highest in flowers. MdNPR5 was mainly expressed in pericarp and pulp. MdNPR6 expression was also highest in pulp (18.52-fold) like the MdNPR1 but was moderate in pericarp and seeds and low in leaves, flowers, and stems. Interestingly, the transcription levels of both MdNPR7 and MdNPR8 were relatively high in flowers and relatively low in stems, leaves, and seeds, but the expression of MdNPR7 in fruits was high, while the expression of MdNPR8 in fruits was very low.

Fig. 5
figure 5

Expression of apple NPR1 homologs in various tissues of ‘Pacific Rose’

Full size image

The expression of apple NPR1 homolog responses to SA and MeJA inoculation

Since NPR1 homologs are thought to play important roles in plant defense responses, the expression of the apple genes under SA and MeJA inoculation was compared between the resistant cultivar Qinguan and susceptible cultivar Pacific Rose (Fig. 6).

Fig. 6
figure 6

Expression of apple NPR1 homologs in response to SA and MeJA in ‘Qinguan’ and ‘Pacific Rose’

Full size image

Following SA induction, MdNPR2, MdNPR3, MdNPR5, and MdNPR6 were downregulated at 1 hpi in the resistant cultivar Qinguan, and then MdNPR2~7 expression was significantly upregulated at 6 hpi, while MdNPR8 was upregulated later at 12 hpi compared with controls. Similarly, in SA-induced susceptible Pacific Rose, the expression of MdNPR2–MdNPR8 was first downregulated at 1 hpi, and then some of the genes were upregulated and reached a peak (MdNPR2 at 6 hpi; MdNPR4 and MdNPR5 at 3 hpi; MdNPR3 and MdNPR6 at 12 hpi). Considering the average expression levels between the cultivars, the apple NPR1 gene homologs were more highly expressed in Qinguan than Pacific Rose, except for MdNPR5 at its peak point. However, the MdNPR4 and MdNPR5 genes responded earlier in Pacific Rose than in Qinguan.

Following MeJA induction, the expression level of MdNPR2 was significantly upregulated at 1 hpi in Pacific Rose and at 3 hpi in Qinguan. The expression of MdNPR3 in Qinguan was first downregulated at 1 hpi and then upregulated. MdNPR3 was upregulated at 1 hpi in Pacific Rose. MdNPR4 expression was sharply upregulated in Pacific Rose at 3 hpi but only slightly upregulated at 12 hpi in Qingquan. MdNPR5 was upregulated at 0–1 hpi in both varieties and then immediately downregulated until 12 hpi in Pacific Rose, while the expression began to drop at 6 hpi in Qinguan. The expression of both MdNPR6 and MdNPR7 was upregulated at 1 hpi in Pacific Rose, while only MdNPR7 was upregulated at 6 hpi in Qinguan. MdNPR8 expression was significantly downregulated at 3 hpi in Pacific Rose but upregulated in Qinguan at 12 hpi. The expression of the MdNPR1 gene was sharply upregulated at 6 and 12 hpi with SA and MeJA, respectively, in Qinguan, but no change was observed in Pacific Rose.

Apple NPR1 gene expression in response to M. coronaria in the susceptible apple cultivar and the resistant wild species

The expression changes of the apple NPR1 homologs were compared between the susceptible apple cultivar Pacific Rose and resistant wild species M. baccata after challenge with M. coronaria (Fig. 7). The expression of MdNPR1 was downregulated in M. baccata but upregulated and reached a peak in Pacific Rose at 48 hpi. There was also a big decrease in the expression of MdNPR5 in M. baccata at 24 hpi. The MdNPR2 gene was upregulated at 48 hpi in Pacific Rose but showed no change in M. baccata. Conversely, the expression of MdNPR6 was upregulated at 48 hpi in M. baccata but showed no change in Pacific Rose. The expression of MdNPR3 in the two varieties was significantly different at 6 hpi but almost the same at all other timepoints. MdNPR4 expression was significantly higher in Pacific Rose at 96 hpi. The expression of MdNPR7 and MdNPR8 showed the same expression trends in both varieties. Interestingly, MdNPR7 was upregulated at 24 hpi, while MdNPR8 were downregulated at the same timepoint; subsequently, MdNPR7 expression returned to a normal level, but the expression of MdNPR8 was upregulated.

Fig. 7
figure 7

Expression of apple NPR1 homologs in ‘Pacific Rose’ and M. baccata after inoculation with M. coronaria

Full size image

Discussion

Identification of NPR1 homologs in apple

A large number of studies have revealed that NPR1 homologs play key roles in various biotic or abiotic responses. However, although NPR1 genes have been isolated and characterized in several plant species, there is little information about NPR1 homologs in apple. In this study, eight novel NPR1 homologs were identified in the apple genome. The NPR1 homolog in apple is the largest community compared with estimates for other plant species (six members in Arabidopsis, four in Carica papaya (Peraza-Echeverria et al. 2012), six in Populus trichocarpa (Shao et al. 2013), and five in Persea Americana (Backer et al. 2015)). The encoded protein sequences of all eight gene homologs included the BTB/POZ and ankyrin repeat domains; these two domains enable the apple NPR1 homologs to perform their functions, possibly by interacting with other proteins, like TGA and NIMIN1/2. The NPR1-like-C domain was found in the group I and II members, and is also thought to be an essential component of NPR1 proteins of Arabidopsis (Cao et al. 1997). The NLS in the C-terminus of the NPR1 homologs is essential for nucleus translocation (Kinkema et al. 2000; Maier et al. 2011). C-terminal deletion may have led to localization and functional changes in MdNPR1b, MdNPR2b, MdNPR4b, MdNPR5b, MdNPR7, and MdNPR8. The MdNPR1 and AtNPR1 proteins harbor the conserved cysteine residues Cys82 and Cys216 in the same position, indicating that the MdNPR1 protein may form an oligomer in apple-like AtNPR1. Additionally, Cys150, Cys155, or Cys160 is also highly conserved in all the apple NPR1 homologs. Mutations in these residues reduced the accumulation of NPR1 (Durrant and Dong 2004). A recent study reported that AtNPR1 could bind SA through the copper ion associated with Cys521/529 (Wu et al. 2012). However, no other NPR1 homolog has cysteine residues in the same site. Thus, the apple NPR1 may bind to SA with other domains, such as the novel NPR1 domain LENRV or NIMINs-binding sites (Canet et al. 2010; Pajerowska-Mukhtar et al. 2013).

Sequence alignment showed that putative proteins of MdNPR3 and MdNPR4 (90.6 % identity), MdNPR7 and MdNPR8 (94.4 % identity), and MdNPR5 and MdNPR6 (85 % identity) are highly similar in amino acid sequence (Table 3). These data indicate that the NPR1 homologs in apple may have undergone an extensive gene duplication and diversification process in comparison with other plants.

Interestingly, our phylogenetic analyses indicated that the numbers of homolog members in each of the three subgroups were significantly different. The MdNPR1 gene in the first group which is highly homologous to AtNPR1 has never been reported in apple, which means it may function as a key regulator in SAR as AtNPR1. The NPR1-like gene, which has been studied before in apple (Malnoy et al. 2007), named MdNPR3 is classified into the second group. This group also included another four apple NPR1 homologs, suggesting that their roles and interactions in plant immune responses are complicated. The two members in the third group, MdNPR7 and MdNPR8, share 94.4 % identity in protein sequence, which is higher than those in Arabidopsis (78.6 % identity), indicating that MdNPR7 and MdNPR8 are possibly newly evolved in apple.

Molecular cloning of the apple NPR1 cDNA homologs sequences showed the common existence of AS of NPR1, which has never been found in other species before. This AS made the transcription regulation patterns of NPR1 complicated in apple. The AS in MdNPR1 and MdNPR4 did not remove the two conserved domains but removed the NLS at the C-terminus, their effect on the functions need further study. In particular, MdNPR1b was found in both Qinguan and Pacific Rose. MdNPR1a and MdNPR1b may form oligomers individually or together in the cytoplasm. In SAR-induced cells, more MdNPR1a may be transcribed so that more functional NPR1 can be transported to the nucleus. In non-induced cells, more MdNPR1b may be transcribed to form oligomers with MdNPR1a in the cytoplasm. Thus, AS of the MdNPR1 transcript may make nuclear transport of NPR1 proteins faster and more convenient. In addition, MdNPR5 can lose its second exon during transcription, resulting in a small protein containing only the BTB/POZ domain. The transcription variant of the MdNPR2 gene can also be translated to a shorter protein but still harbors the two conserved domains. The different sequences of MdNPR2 in the two varieties indicate alleles or RNA editing of this gene. These proteins may lose their original functions or play other roles in plants. Because we could not isolate MdNPR1a or MdNPR4b from Qinguan, it is possible that some of the transcriptional variants are specifically expressed in different apple varieties. These transcriptional variants may respond to specific signals or pathogens. Alternative splicing may be an important form of regulation for these genes.

Expression of the apple NPR1 homologs

The Pacific Rose and Qinguan apple cultivars are generally considered to have different resistance to many diseases. The expression of the apple NPR1 homologs showed significant differences after induction with SA and MeJA. The apple NPR1 homologs in group II were more highly expressed in Qinguan than in Pacific Rose after SA induction. In banana cultivars that are resistant to Fusarium oxysporum, NPR1 (which is homologous to MdNPR1) is upregulated earlier and to a greater extent after SA treatment than in susceptible cultivars (Endah et al. 2008). The MdNPR3 gene was upregulated at 6 h after SA treatment in Qinguan. The result is the same as the previous findings that MhNPR1 (which is homologous to MdNPR3) and the SAR marker genes MhPR1 and MhPR5 can be induced by SA (Zhang et al. 2012). However, no significant expression change of MdNPR3 was found in the susceptible Pacific Rose after SA induction. Given that there was no difference in MdNPR3 sequence between the two varieties, the transcript levels of this gene may be regulated in different ways. The MdNPR4 and MdNPR5 genes responded earlier in susceptible Pacific Rose than in resistant Qinguan after SA treatment, and we speculate that these two genes may play a more important role in the susceptible cultivar Pacific Rose.

The specific expression of the NPR1 homologs was studied in six different apple tissues. The transcription level of the MdNPR1 gene was very low in all tissues, while relatively high in leaves and fruit, which were consistent with previous findings. MdNPR2, MdNPR3, MdNPR5, and MdNPR6 were highly expressed in the pericarp and pulp, with the expression of MdNPR6 in pulp being the highest. Thus, it seems that these four NPR1 homologs are more important in fruit. MdNPR4, MdNPR7, and MdNPR8 expression was highest in flowers, which may mean that these genes play important roles in flower development.

Conidia of M. coronaria are widely thought to germinate on the apple leaf surface at about 6 hpi and symptoms (disease spot) can be visually observed at 6 days after inoculation in Pacific Rose (Dang et al. 2011). MdNPR1, MdNPR3, MdNPR5, and MdNPR6 were highly expressed in M. baccata compared with Pacific Rose at 6 hpi. The expression of MdNPR2 and MdNPR4 genes fluctuated around only a little after inoculation in M. baccata. Interestingly, significant expression differences of MdNPR1, MdNPR2, and MdNPR6 were observed between Pacific Rose and M. baccata at 48 hpi. Thus, the different resistance of apple to M. coronaria could be explained in two ways. Firstly, one of genes may be the key gene affecting resistance, such as the MdNPR2 gene, which showed polymorphism in Qinguan and Pacific Rose. Secondly, the different transcription levels of the transcription variants after inoculation and their interaction may be a predominant factor in apple. The expression of MdNPR7 and MdNPR8 showed the same change trend after inoculation in both cultivars, suggesting that these two genes do not contribute to the differences in M. coronaria resistance in apple.

The functions of the NPR1 homologs are more complicated in apple than in Arabidopsis. At the protein level, it was speculated that two or more homologs may combine with each other or with signal molecules to regulate the balance of the MdNPR1. NPR1 homeostasis is controlled by SA binding to NPR3/NPR4 in a concentration-dependent manner, which have been proved in Arabidopsis (Moreau et al. 2012). At the transcription level, the apple NPR1 homolog genes can be regulated by transcription factors or transcribed to different transcripts. This complicated phenomenon should help to maintain the stability of the plant immune system.

In conclusion, this study provides evidence to help the preliminary NPR1 gene information and functional annotation of the eight newly discovered NPR1 homologs from apple. Sequence structure and homology as well as phylogenetic analysis suggest that six apple NPR1 homologs (MdNPR1–6) may be involved in defense responses, while the remaining two (MdNPR7 and 8) are most likely involved in tissue development. In particular, the expression level of MdNPR1 was very low but changed rapidly after SA and pathogen treatments. In addition to the most similarity to AtNPR1 on amino acid sequence, MdNPR1 gene may function as a key regulator of SAR. The hormone and pathogen inoculation treatment-induced expression, as well as specific expression in various tissues, of the apple NPR1 homologs support this and provide a foundation for future research. Future efforts will be focused on the intracellular interactions and localization of the NPR1 proteins. Additionally, the expression levels and functions of the different transcript variants of NPR1 homologs in apple need to be explored in depth.