Characterization of Banana SNARE Genes and Their Expression Analysis under Temperature Stress and Mutualistic and Pathogenic Fungal Colonization

Abstract

SNAREs (soluble N-ethylmaleimide-sensitive-factor attachment protein receptors) are engines for almost all of the membrane fusion and exocytosis events in organism cells. In this study, we identified 84 SNARE genes from banana (Musa acuminata). Gene expression analysis revealed that the expression of MaSNAREs varied a lot in different banana organs. By analyzing their expression patterns under low temperature (4 °C), high temperature (45 °C), mutualistic fungus (Serendipita indica, Si) and fungal pathogen (Fusarium oxysporum f. sp. Cubense Tropical Race 4, FocTR4) treatments, many MaSNAREs were found to be stress responsive. For example, MaBET1d was up-regulate by both low and high temperature stresses; MaNPSN11a was up-regulated by low temperature but down-regulated by high temperature; and FocTR4 treatment up-regulated the expression of MaSYP121 but down-regulated MaVAMP72a and MaSNAP33a. Notably, the upregulation or downregulation effects of FocTR4 on the expression of some MaSNAREs could be alleviated by priorly colonized Si, suggesting that they play roles in the Si-enhanced banana wilt resistance. Foc resistance assays were performed in tobacco leaves transiently overexpressing MaSYP121, MaVAMP72a and MaSNAP33a. Results showed that transient overexpression of MaSYP121 and MaSNPA33a suppressed the penetration and spread of both Foc1 (Foc Race 1) and FocTR4 in tobacco leaves, suggesting that they play positive roles in resisting Foc infection. However, the transient overexpression of MaVAMP72a facilitated Foc infection. Our study can provide a basis for understanding the roles of MaSNAREs in the banana responses to temperature stress and mutualistic and pathogenic fungal colonization.

1. Introduction

Vesicle trafficking, a conserved intracellular transport mechanism in eukaryotic cells, is crucial for the plant growth and development and plant adaptations to various external and internal stresses [1,2]. It mainly involves four steps, i.e., budding, transporting, tethering and the fusion of vesicles with acceptor membranes [3]. Soluble N-ethylmaleimide-sensitive-factor attachment protein receptors (SNAREs) participate and function in almost all of the membrane fusion processes in cells [4,5,6,7]. According to their localization during membrane fusion, SNAREs can be divided into vesicle-SNARE (v-SNAREs) and target-SNARE (t-SNAREs) types [8]. However, since some SNARE protein displayed variable localizations, somewhile this classification would be inaccurate [9]. SNARE proteins contain conserved SNARE domains composed of about 60 amino acids. According to whether an arginine (R) or a glutamine (Q) exists in the core sequence of this domain, SNAREs can be classified into R-SNAREs and Q-SNAREs [10]. Moreover, the Q-SNAREs can be further classified into Qa-, Qb-, Qc- and Qbc-SNAREs [11].

Since the first discovery of VAMP-1 (synaptic vesicle-associated membrane protein 1; a neuron-specific synaptic vesicle-associated internal membrane protein) in the electric organ of Torpedo in 1988 [12], many core SNAREs and their homologous proteins, such as SNAP-25 [13] and Syntaxin [14], have been successively identified. Generally, SNAREs exist as a large and dispersed multigene family with more than 20 members per genome. For example, there are 38, 26 and 25 SNARE genes in the genome of Homo sapiens [15], Drosophila melanogaster [16] and Saccharomyces cerevisiae [17], respectively. Due to the evolution and increasement of multicellularity, the numbers of SNARE members in green plant genomes became more abundant [18]. In the model plant, Arabidopsis thaliana, 64 SNARE members are identified. Rice (Oryza sativa L.), black cottonwood (Populus trichocarpa), Tomato (Solanum lycopersicum L.) and wheat (Triticum aestivum L.) have 60 [19], 69 [20], 63 [21] and 173 [22] SNARE members, respectively.

SNAREs display miscellaneous functions in regulating substance transport, participating in cytoplasmic division and mediating ion channel stability [23]. The Arabidopsis SYP111 deletion mutants displayed abnormal vesicle trafficking and inhibited cell wall formation and cytokinesis [24]. Arabidopsis VAMP721 and VAMP722 are involved in the vesicular trafficking between the plasma membrane and the cis/trans-Golgi network (CGN/TGN), playing roles in facilitating the formation of cell plates and the transportations of proteins and many other molecules during cell division [25,26]. VAMP721 can interact with and inhibit potassium channel-related proteins (such as KAT1 and KC1) and thereby influence potassium ion transport [27].

SNAREs play important roles in plant stress responses. It has been reported that SNARE proteins function in regulating the Ca²⁺ signaling pathway and indirectly influence the regulation of ABA signaling on K⁺ and Ca²⁺ channels to enhance the salt, alkali and drought resistances of plants [28]. Recently, their participations in the plant immunity responses to pathogens, especially fungal pathogens, have been reported. The accumulation of pathogenesis-related protein PR1 in SYP132-silenced transgenic tobacco plants was significantly suppressed [29]. The SNARE complex, consisting of PEN1 (SYP121), SNAP33 and VAMP721/VAMP722, is reported to be indispensable for the defense responses against powdery mildew fungi infection in Arabidopsis [7]. This complex can assist cell exocytosis in the powdery mildew infection sites, and thereby resist the invasion of pathogens [30]. Positive functions of Barley (Hordeum vulgare L.) HvROR2 [31] and tomato (Solanum lycopersicum L.) SlPEN1 [32] in the plant-powdery mildew interactions have also been confirmed. In addition, AtSEC11 was reported to have the ability of improving the SNARE complex stability by competing with VAMP721 and SNAP33 for PEN1 binding, thereby improving the powdery mildew resistance of Arabidopsis [33].

Banana (Musa spp.) is one of the most important fruit trees in the world. As a tropical fruit tree, it is very sensitive to high and low temperature stresses [34]. Moreover, the banana industry has been severely threatened by the devastating Fusarium wilt (FW) disease caused by soil-borne fungus Fusarium oxysporum f. sp. cubense (Foc) [35,36]. Previously, we found that Serendipita indica (Si), a culturable arbuscular mycorrhizal fungus (AMF)-like endophytic fungus, could promote banana plant growth and improve its resistance to both temperature stress and FW disease [37,38,39]. Although SNARE proteins have been extensively reported to play roles in plant stress responses, till now, there are still few reports on banana SNAREs. In this study, to explore the possible contributions of SNAREs to temperature stress and mutualistic and pathogenic fungal colonization, genome-wide identification and characterization of banana SNARE genes was performed, and their expression patterns in different organs, in leaves treated with low temperature and high temperature, and in roots treated with S. indica and Foc tropical race 4 (FocTR4), were investigated. Furthermore, the effects of three roots that were highly expressed and S. indica- and FocTR4-responsive MaSNAREs (MaSYP121, MaVAMP72a and MaSNAP33a) on the infection of Foc1 and FocTR4 were studied using the tobacco leaf transient expression method. The results obtained in this study will be helpful in clarifying the roles of MaSNAREs in banana stress responses.

2. Results

2.1. Physicochemical Properties of the Identified MaSNAREs

In total, 84 SNARE genes were identified from the M. acuminata genome (Supplemental Table S1). They were named according to their closest Arabidopsis homologous genes. For the banana SNARE genes that shared the same homologous gene, they were orderly named based on their chromosome location information. For example, the homologous genes of Arabidopsis SYP121, VAMP721 and SNAP33 in banana were named as MaSYP121, MaVAMP72a-c and MaSNAP33a-e, respectively.

The CDS lengths of MaSNAREs ranged from 159 bp to 3336 bp, and their encoded proteins contained 52 aa~1111 aa, with a molecular weight of 6057.79 kDa~121,494.46 kDa and the isoelectric point (PI) of 4.71~9.93. Their protein instability coefficient ranged from 29.68 to 76.84 and their lipid solubility index ranged from 60.87 to 123.48, respectively. Among them, MaSYP125a, MaVTI1b, MaNPSN13b, MaNPSN13c, MaBET1b and MaVAMP71b were hydrophobic proteins, and the others were hydrophilic proteins.

2.2. Subcellular Localization Prediction of MaSNAREs

Subcellular localization prediction results showed that there were 30, 14, 13, 13, 8 and 5 MaSNAREs localized in the plasma membrane, vacuole, endoplasmic reticulum, Golgi apparatus, nucleus and cytoplasmic matrix (Table S1), respectively. In consistence with the Arabidopsis SNAREs [1], the plasma membrane-localized MaSNAREs take the largest part.

2.3. Phylogenetic Analysis of MaSNARE Proteins

By using the 84 MaSNAREs and the 64 Arabidopsis SNAREs sequences, a phylogenetic tree was constructed. Results showed that MaSNAREs could be classified into five subfamilies, including Qa (containing 29 members), Qb (22 members), Qc (14 members), Qbc (5 members) and R (14 members) subfamilies (Figure 1 and Supplemental Figure S1). The banana Qa subfamily could be further divided into seven groups (SYP11, SYP12, SYP13, SYP2, SYP3, SYP4 and SYP81, with 2, 13, 4, 4, 3, 2 and 1 MaSNARE members, respectively). The number of the banana SYP12 members is much larger than that of the Arabidopsis SYP12 (5). The banana Qb subfamily could be divided into GOS1, SEC20, MEMB, USE1, VTI1 and NPSN1 groups, containing 3, 2, 2, 2, 4 and 9 members, respectively. The banana Qc subfamily could be divided into SFT1, BET1, SYP5, SYP6 and SYP7 groups, containing 3, 4, 2, 2 and 3 members, respectively. The banana Qbc subfamily contained a single SNAP33 group with five MaSNARE members. And the banana R subfamily could be divided into TYN1, VAMP71, VAMP72, YKT6 and SEC22 groups, containing 2, 2, 3, 3 and 4 members, respectively.

2.4. Conserved Motifs in MaSNAREs and Gene Structures of Their Encoding Genes

A total of 20 conserved motifs were identified from MaSNARE proteins (Figure 2A). Among the Qa subfamily MaSNAREs, all SYP13 group members contained Motif8, 1, 4, 2 and 13; all SYP2 group members contained Motif16, 18 and 19; all SYP3 group members contained Motif18 and Motif3; and all SYP4 group members contained Motif3 and Motif9. Among the Qb subfamily members, all GOS1 and VTI1 group members contained Motif18; SEC20 and SYP4 group members contained the same motifs; MEMB11, MEMB12 and SYP13 group members contained the same motifs in the same order; USE1 group members contained only Motif20; of the NPSN11 group members, MaNPSN11a, b and d had four motifs ordering in Motif7-5-6-3-9, while MaNPSN11c and MaNPSN11e had four other motifs in the order of Motif14-11-12-10; of the MaNPSN13 group members, MaNPSN13a had three motifs ordering in Motif19-3-9, while MaNPSN13b and MaNPSN13c had four motifs ordering in Motif1-15-2-13. Among the Qc subfamily MaSNAREs, Motif18 and Motif20 were found in SFT1 group members, and all BET1, SYP5 and SYP7 group members contained Motif3. In the Qbc subfamily MaSNAREs, there were two types of motif orders for SNAP33 group, i.e., Moitf12-3 and Motif7-5-6-3-9. In the R subfamily, the motif ordering of VAMP71 group members was Motif1-15-2-13 and that of VAMP72 group members was Motif5-1-4-2-13. In addition, no conserved motif was identified in MaSYP122, MaSYP124c, MaSYP125a and MaTYN11.

Gene structure analysis results showed that the exon numbers of Qa subfamily MaSNAREs ranged from 1 to 12, with MaSYP121 having the most exons (12), and MaSYP123a and MaSYP123b having a single exon (Figure 2B). MaSYP131a-c all contained five exons, MaSYP2a-d contained seven exons, MaSYP3a-c contained five exons, and MaSYP4a-b contained eight exons. The exon numbers of Qb subfamily MaSNAREs ranged from 1 to 13. MaNPSN11a, MaNPSN11b and MaNPSN11d had the most exons (13), and MaGOS1a had one exon. The exon numbers of Qc subfamily MaSNAREs ranged from 2 to 9, MaSYP7a, MaSYP7b and MaSYP7c had the most exons (9), and MaSYP6b had the least exons (2). The exon numbers of Qbc subfamily MaSNAREs ranged from 1 to 6. MaSNAP33a-e had 5, 5, 6, 1, and 1 exons, respectively. The exon numbers of R subfamily MaSNAREs ranged from 5 to 24, with MaTYN11 and MaTYN12 having the most exons (24), and MaYKT6c and MaSEC22d having the least exons (5).

2.5. Promotor Cis-Acting Elements Analysis Results of MaSNAREs

A large number of light-, phytohormone-, and stress-responsive, and growth and development-related cis-acting elements were identified in the promoters of MaSNAREs (Figure 3). Moreover, there were also many cis-acting elements related to some unknown functions (Supplemental Table S2).

In total, we identified 17 types of light-responsive elements in the promoters of MaSNAREs. Notably, 68 (80.95%) MaSNAREs contained the light-responsive Box4 elements in their promoters, and the MaSYP121 promoter contained the largest number of Box4 element (in total of 10). In addition, 50 (59.52%) MaSNARE promoters contained CTT-motif elements.

Five types of growth and development-related elements (CAT-box, O2 (Opaque2)-site, GCN4_motif, MSA-like and circadian) were identified in the promoters of MaSNAREs. Among them, zein metabolism regulation O2-site element was found in the promoters of 37 (44.05%) MaSNAREs; the meristem expression-related elements CAT-box was identified in the promoters of 33 (39.29%) MaSNAREs; the circadian clock regulation-related element circadian was found in the promoters of 18 (21.43%) MaSNAREs; the endosperm expression-related element GCN4_motif was found in the promoters of 16 (19.05%) MaSNAREs; and the cell cycle regulation-related MSA-like motif was found in the promoters of 5 (5.95%) MaSNAREs.

The MaSNAREs promoters contained cis-acting elements related to the responses to six kinds of phytohormones, including gibberellin (GA), methyl jasmonate (MeJA), salicylic acid (SA), auxin, abscisic acid (ABA) and ethylene (ET). There were three kinds of GA responsive elements (P-box, TATC-box and GARE-motif), two kinds of MeJA responsive elements (TGACG-motif and CGTCA-motif), two kinds of SA responsive elements (TCA-element and TCA), three kinds of auxin responsive elements (TGA-box, TGA-element and AuxRR-core), one kind of ABA responsive element (ABRE) and one kind of ethylene responsive element (ERE) in the MaSNARE promoters. Interestingly, the promoters of MaNPSN13b, MaVAMP71b and MaVAMP72a contained elements related to all the six phytohormones. Moreover, there were 66 (78.57%), 65 (77.38%) and 50 (59.52%) MaSNARE contained MeJA responsive TGACG-motif and CGTCA-motif elements, ABA responsive elements and SA responsive elements in their promoters, respectively. It was noted that MaSYP121 contained eight SA-responsive TCA elements. In addition, there were 44 (52.38%), 36 (42.86%) and 28 (33.33%) MaSNAREs contained auxin-, GA-, and ET-responsive elements in their promoters, respectively.

The promoters of MaSNAREs also contained seven types of stress-responsive elements, including anoxic specific inducibility-, anaerobic induction-, low temperature-, high temperature-, drought inducibility-, wound-, defense- and stress-related elements. Notably, the promoters of MaSYP122, MaSYP4b, MaNPSN11a and MaSEC22d contained all the seven types of stress-responsive elements. Except for MaSYP2d, MaMEMB12, MaNPSN13c, MaSYP7a and MaSNAP33a, all the other 79 (94.05%) MaSNAREs contained drought inducibility-related MYC element in their promoters. High-temperature inducibility-related STRE element was found in the promoters of 76 (90.48%) MaSNAREs. Anaerobic induction-related ARE element was predicted in the promoters of 71 (84.52%) MaSNAREs. Additionally, there were 58 (69.05%), 52 (61.90%) and 41 (48.81%) MaSNAREs contained MYB, W-box and LTR element in their promoters, respectively.

2.6. Chromosome Localization and Collinearity Analysis Results for MaSNAREs

Chromosome localization analysis results showed that MaSNAREs were unevenly distributed on all of the 11 chromosomes of M. acuminata (Figure 4). The accounts of MaSNARE members on Chr6 and Chr10 were the largest (12 members for each chromosome). Chr9 had nine members; Chr1 and 11 both had eight members; Chr2, 4 and 5 had seven members; Chr7 and 8 both had six members; and Chr3 had one member. In addition, MaSYP124a was found to be localized on chrUn_random. The accounts of Qa subfamily MaSNAREs on Chr6 and Chr10 were both the largest (five members for each chromosome). Qb subfamily MaSNAREs were distributed on all the chromosomes except Chr3 and Chr5. Qc subfamily MaSNAREs were unevenly distributed on Chr2, 5, 6, 7, 9, 10 and 11. Qbc subfamily MaSNAREs had one member each on Chr4, 5, 6, 8 and 9. And the R subfamily MaSNAREs were distributed on Chr2, 4, 5, 6, 8, 9, and 11.

By using MCScanX software, the gene duplication events that occurred in the MaSNAREs were analyzed. A total of 23 segmental duplication gene pairs involving in 34 MaSNAREs were identified, but no tandem duplication gene pair was found. Of the 23 segmental duplicated gene pairs, six gene pairs were from the R subfamily (MaSEC22a, 22c; MaVAMP71a, 71b; MaVAMP72a, 72c; MaVAMP72b, 72a; MaYKT6a, 6b; MaVAMP72b, 72c), five gene pairs were from the Qa subfamily (MaSYP123a, 123b; MaSYP2a, 2d; MaSYP3a, 3b; MaSYP3a, 3c; MaSYP3b, 3c), three gene pairs were from the Qb subfamily (MaNPSN11c, 11e; MaNPSN13b, 13c; MaUSE1a, 1b), three gene pairs were from the Qc subfamily (MaSYP5a, 5b; MaSYP7a, 7b; MaSYP7b, 7c), and the remaining six gene pairs were from different subfamilies (MaKNOLLE/MaSEC22b, MaNPSN12/MaVAMP72c, MaSYP125a/MaVTI1c, MaSYP4b/MaSEC20b, MaVAMP72a/MaNPSN12 and MaVAMP72b/MaNPSN12). No duplicated gene pair was identified from the Qbc subfamily. Notably, two duplicated gene pairs were found to be involving three MaSNARE members (MaSYP3a, 3b and 3c; MaVAMP72b, 72a and 72c).

The Ka/Ks values of these gene pairs ranged from 0.0235 to 0.4168 (Table S3), indicating that the evolution of this banana gene family was mainly influenced by strong purification selection pressure. The Ka and Ks values of the duplicated gene pairs were further used to calculate their divergence times. It was revealed that these gene duplication events occurred at 55.27~115.62 Mya (Table S3).

2.7. Gene Expression Analysis Results

2.7.1. Expression Analysis Results of MaSNAREs in Four Different Organs

Based on our transcriptome data, the spatial expression variations of MaSNAREs in banana root, corm, leaf and fruit were investigated (Figure 5). There were 69, 58, 81 and 9 members expressed in root, corm, leaf and fruit, respectively. MaSEC22a, MaBET1b, MaMEMB11 and MaSYP122 expressed in all the four banana organs, while MaSNAP33d showed no expression in any of these organs. The expression level of MaVAMP72a in root was the highest among all of the MaSNAREs, followed by MaSNAP33e and MaSYP121. The expression level of MaVAMP72c was the highest in corm among all of the MaSNAREs, followed by MaVAMP72a. The expression level of MaVAMP72c was the highest in leaf among all of the MaSNAREs. And the expression level of MaSYP6b ranked the first in fruit.

The expression levels of different group members from the same subfamily in different organs also varied a lot. In the Qa subfamily, the expression level of MaSYP121 in root was the highest, followed by MaSYP2b. The expression level of MaSYP2b in leaf was the highest, followed by MaSYP121. The expression of MaSYP125b and MaSYP125a was the highest in corm and fruit, respectively. In the Qb subfamily, the expression of MaVTI1c was the highest in both root and leaf; the expression level of MaNPSN12 was the highest in corm, followed by MaVTI1d; the expression of MaMEMB11 was the highest in fruit. In the Qc subfamily, MaBET1a showed the highest expression in root among all the Qc members, followed by MaSYP5b; the expression level of MaBET1a was the highest in leaf, followed by MaSFT1a; the expression level of MaSYP7b in corm was the highest, followed by MaSYP5b; the expression level of MaSYP6b was the highest in fruit. In the Qbc subfamily, MaSNAP33e expressed the highest in both root and corm, followed by MaSNAP33a; the expression level of MaSNAP33a was the highest in leaf; but none of the five members expressed in fruit. In the R subfamily, MaVAMP72a expressed the highest in roots; the expression of MaVAMP72c was the highest in both corm and leaf; and MaSEC22a was the only R subfamily member that expressed in fruit.

2.7.2. Expression Analysis of MaSNAREs in Banana Leaves under Low and High Temperature Treatments

We further studied the influences of low and high temperature treatments on the expression of MaSNAREs in banana leaves (Figure 6). The expression levels of MaVAMP72a, MaNPSN11a, MaSYP121 and MaBET1d increased more than twofold by low temperature treatment, accounting for 5.62-, 3.33-, 2.58-, and 2.1-fold of CK, respectively. Among them, only the MaSYP121 promoter did not contain a low-temperature responsive element LTR. The expression of MaNPSN13c and MaNPSN13b in low temperature treated leaves, however, significantly decreased to 39.74% and 41.52% of CK, respectively. The expression levels of MaYKT6a, MaSYP81, MaSYP131c, MaSYP2d and MaBET1d increased by high temperature treatment, accounting for 7.05-, 5.74-, 5.63-, 3.38- and 2.92-fold of CK, respectively. Among them, only the MaBET1d promoter did not have a high-temperature responsive element STRE. The expression of MaTYN12, MaNPSN11a and MaNPSN12 in high temperature treated leaf decreased to 16.1%, 17.25% and 33.64% of CK, respectively. Interestingly, the expression of MaNPSN11a was up-regulated by low temperature but down-regulated by high temperature.

Obvious expression change pattern differences were found among different subfamily members and members from the same subfamily under temperature stress. In the Qa subfamily, MaSYP121 expressed the highest in low temperature treated leaf, followed by MaSYP2d and MaSYP2a; the expression level of MaSYP131c was the highest in the high temperature treated leaf, followed by MaSYP121 and MaSYP81; the expression level of MaSYP121 in low temperature and high temperature treated banana leaf accounted for 2.58- and 1.78-fold of CK, respectively; the expression of MaSYP2d in low temperature and high temperature treated banana leaf accounted for 1.77- and 3.38-fold of CK, respectively. In the Qb subfamily, the expression level of MaVTI1c was the highest in both high and low temperature treated leaves among all the Qb subfamily members; its expression in high temperature treated leaf was about 3.15-fold of CK. In the Qc subfamily, the expression of MaBET1a was the highest in both high and low temperature treated leaves, followed by MaSFT1a; the expression level of MaSFT1a in low temperature treated banana leaf decreased to 64.62% of CK. In the Qbc subfamily, the expression of MaSNAP33a was the highest in low temperature treated leaves, followed by MaSNAP33b and MaSNAP33e; the expression level of MaSNAP33b in high temperature treated leaves was the highest among the Qbc members, followed by MaSNAP33a and MaSNAP33e; the expression levels of MaSNAP33a, MaNAP33b, MaNAP33c and MaSNAP33e were up-regulated by both high and low temperature treatments. In the R subfamily, MaVAMP72c expressed the highest expression in both high and low temperature treated leaves, followed by MaVAMP72a and MaYKT6a; the expression of MaVAMP72a and MaVAMP72b in low temperature treated banana leaf increased to 5.62- and 4.12-fold of CK, respectively; the expression of MaYKT6a and MaVAMP72b in high temperature treated banana leaf was, respectively, up-regulated to 7.05- and 3.35-fold of CK, while the expression of MaTYN12 was down-regulated to only about 16.1% of CK.

2.7.3. S. indica and FocTR4 Treatments Influenced the Expression of MaSNAREs in Banana Root

The expression changes of MaSNAREs in banana roots in response to the mutualistic fungus S. indica colonization (Si group), pathogenic pathogen FocTR4 infection (Foc group) and their co-treatment (SF group, S. indica-colonized seedlings were subjected to FocTR4 infection) were studied (Figure 7). In roots of the Foc group, the expression of MaSYP121, MaSYP112, MaSYP122, MaSYP124c, MaSYP124e and MaSYP125a from the Qa subfamily was up-regulated, and MaSYP121 was the most significantly up-regulated one, which was about 2.44-fold of CK. The expression levels of MaKNOLLE and MaSYP132 decreased significantly, only accounting for 17.34% and 22.33% of CK, respectively. In the Si group, the expression levels of the six FocTR4 inducible Qa subfamily members were also up-regulated. Compared with the Foc group, however, the expression level of MaSYP121 in SF group decreased, accounting for about 62.65% of that in Foc group.

The expression of 14 Qb subfamily members were down-regulated by FocTR4, and the expression of MaNPSN13a in the Foc group was only 40.96% of CK. The expression of two Qb members, MaVTI1b and MaNPSN12, was up-regulated by FocTR4 for 8.92- and 5.25-fold, respectively. Compared with CK, the expression trend of Qb subfamily members in Si group was basically similar to that of the Foc group. Compared with Foc group, the expression levels of MaGOS1a and MaVTI1d in SF group were up-regulated to 4.71- and 2.06-fold of Foc group, respectively.

In the Qc subfamily, the expression levels of MaSYP7a and MaSYP7c increased after FocTR4 treatment, while the expression levels of the other 11 subfamily members decreased. Compared with CK, the expression level of MaSYP7c in the Si group slightly increased, while the other members decreased. Compared with the Foc group, except MaBET1a and MaSYP6b, the expression levels of other Qc subfamily members all increased in the SF group. It is worth noting that MaSFT12 and MaSYP7b showed very low expression in the Foc group, but their expression levels were much higher in the SF group than in the CK group.

In the Qbc subfamily, the expression of MaSNAP33b in the Foc group was about 4.8-fold of CK. While, the expression of MaSNAP33a was down-regulated to only 3% of the CK. After S. indica colonization, the expression level of MaSNAP33a, MaSNAP33b and MaSNAP33d increased to 4.17-, 3.29- and 2.9-fold of CK, respectively, while the expression levels of MaSNAP33a decreased to about 50% of CK. Compared with the Foc group, MaSNAP33a and MaSNAP33c were up-regulated and MaSNAP33b and MaSNAP33e were down-regulated in the SF group.

In the R subfamily, the expression of 11 members was down-regulated by FocTR4, and the expression of MaVAMP71b and MaVAMP72a in the Foc group were only about 31.77% and 22.33% of the CK group, respectively. After S. indica colonization, the expression level of MaTYN12 increased, while the other members decreased. Compared with the Foc group, except for MaVAMP72c and MaYKT6c, all the other R family MaSNARE members were up-regulated in the SF group, and the expression of MaVAMP72a was significantly up-regulated to 4.58-fold of the Foc group.

The expression changes of 10 selected MaSNAREs (MaSYP121, MaSYP122, MaVAMP72a, MaSNAP33a, MaSYP6a, MaKNOLLE, MaSYP5a and MaBET1a, MaSYP131a and MaNPSN11a) that showed high expression levels in roots and exhibited expression changes in response to S. indica or FocTR4 treatments were further validated by using quantitative real time PCR (qRT-PCR) (Figure 8). Results showed that their changing trend was basically consistent with our transcriptome data. Compared with CK, the expression level of MaSYP121, MaSYP122 and MaNPSN11a after FocTR4 treatment significantly increased, which was 2.23-, 1.45- and 1.43-fold of CK group, respectively. The expression levels of MaSYP6a, MaKNOLLE, MaSYP5a and MaSNAP33a were significantly down-regulated by FocTR4, which account for 38.78%, 71.91%, 18% and 35.79% of CK, respectively. After S. indica colonization, the expression levels of eight selected genes (except MaSYP5a and MaSNAP33a) were up-regulated, and the expression of MaNPSN11a increased to about 1.3-fold of CK. Compared with Foc group, the expression levels of MaVAMP72a, MaSYP131a, MaSYP6a, MaKNOLLE, MaSYP5a, MaBET1a, MaSNAP33a and MaNPSN11a in SF group were significantly higher, accounting for 1.31-, 1.32-, 2.71-, 1.46-, 6.4-, 1.95-, 1.08- and 4.83-fold of the Foc group, respectively.

2.8. Foc Resistance Assays in Tobacco Leaves Transiently Overexpressing MaSNAREs

The expression levels of MaSYP121 and MaVAMP72a in banana root both ranked top three among all MaSNAREs. The Qbc subfamily member MaSNAP33a was also highly expressed in roots. Both our transcriptome and qRT-PCR results showed that the expression of MaSYP121 was up-regulated by FocTR4 and S. indica. According to our transcriptome data, the expression of MaVAMP72a and MaSNAP33a was up-regulated by both fungi. Moreover, their FocTR4 responsive characteristic was greatly alleviated in the SF group. Therefore, these three members were predicted to play important roles in the S. indica–banana–Foc interactions. To verify their functions, overexpression vectors for MaSYP121, MaVAMP72a and MaSNAP33a were constructed, transiently overexpressed in tobacco leaves and subjected to Foc Race 1 (Foc1) and FocTR4 inoculation (Figure 9). After Foc1 and FocTR4 inoculation, the lesion areas in the tobacco leaves expressing MaSYP121 were significantly smaller than the empty vector controls (EV), accounting for only 3.19% and 44.18% of EV, respectively. Transient overexpression of MaSNAP33a also reduced the lesion area, and the Foc1- and FocTR4-caused lesion area in tobacco leaves overexpressing MaSNAP33a was only 65.67% and 50.66% of EV, respectively. However, transient overexpression of MaVAMP72a resulted in larger Foc1- and FocTR4-caused lesion areas in tobacco leaves, which was 2.07- and 1.72-fold of EV, respectively.

3. Discussion

3.1. Segmental Duplications Contributed to the Expansion of Banana SNARE Gene Family

The intracellular transportation in eukaryotic cells is largely dependent on vesicle trafficking [3]. As a key participant of vesicle trafficking, SNAREs are crucial for the membrane fusion between vesicles and target membranes [8]. The large number of SNAREs in plant genomes were reported to be closely related to the increased multicellular nature of plants [22] and their miscellaneous functions [40]. Plant SNAREs can be divided into 5 subfamilies (Qa-, Qb-, Qc-, Qbc- and R-SNARE) and 24 groups [41,42,43]. Consistently, in this study, we obtained the same classification results for MaSNAREs. The amount of MaSNARE members is approximately 1.31 times that of Arabidopsis and 1.4 times that of rice, which may be related to the multicellular and triploid nature of banana [44]. We identified 23 segmental duplicated gene pairs involving 34 MaSNAREs from banana. In tomato, segmental duplications were also reported to be main contributors for the expansion of the SNARE gene family [21]. Moreover, among these segmental duplicated tomato SNARE genes, members from VAMP groups were the most abundant, followed by SYP1 members. Consistently, in this study, MaSNARE members from the VAMP group (five members) and SYP1 group (five members) also ranked top two.

3.2. The Expression of MaSNAREs Varied a Lot in Different Organs

SNAREs are expressed in various tissues of plants and participate in almost all plant growth and development processes. Evidence has shown that their expression patterns and functions in different tissues and organs varied [6]. For example, SYP132 is expressed in all tissues and organs of Arabidopsis, while SYP124, SYP125 and SYP131 are only expressed in pollen, and SYP123 is only expressed in root hair cells [26]. In this study, we found that there were 81, 69, 58 and 9 MaSNAREs that showed expression in banana leaf, root, corm and fruit, respectively. In Vaccinium myrtillus, SYP5, SYP6, SYP7, SEC20, SEC22 and YKT6 genes were expressed in fruit [45]. Our study found that, among the nine fruit-expressing MaSNAREs, MaSYP6b (homologous to V. myrtillus SYP6) showed the highest expression level, and MaSEC22a (homologous to V. myrtillus SEC22) also showed high expression in fruit, suggesting that they might function a lot in banana fruit.

3.3. The Expression of MaSNAREs Could Be Significantly Influenced by Many Phytohormones and Environmental Factors

Evidence has revealed that phytohormones influenced greatly the expression and functions of SNARE genes [46,47,48]. The expression of Arabidopsis SYP132 could be significantly affected by auxin [46]. In this study, 52.38% of the MaSNAREs contained auxin response elements in their promoters. NtSYR1 (a homologous gene of AtSYP121 in tobacco) was proved to function in regulating the growth and development of plants by affecting the ABA response process [47]. Consistently, our study revealed that 77.38% of MaSNARE promoters contained the ABA responsive elements. SNARE proteins play important roles in plant responses to both abiotic and biotic stresses [6]. Their regulatory roles in Ca²⁺ signaling pathway and ABA signaling have been proved to be vital for enhancing the salt, alkali and drought resistance of plants [28]. In consistence with this, we found that 94.05% of MaSNAREs contained drought inducibility elements in their promoters. AtSYP121 and AtSYP122 have been proved to be negative regulators of SA, JA and ET pathways [48]. In this study, we found that 78.57% of MaSNARE promoters had MeJA responsive cis-acting elements and many SA and ET responsive elements were also identified in their promoters.

Additionally, 90.48% and 48.81% of MaSNAREs contained high-temperature and low-temperature responsive elements in their promoters, respectively. Gene expression analysis showed that many MaSNAREs were temperature stress responsive. For example, MaVAMP72a, MaNPSN11a, MaSYP121 and MaBET1d were up-regulated by low temperature more than twofold. All of them, except MaSYP121, had low-temperature responsive element LTR. The MaSYP121 promoter contained the most SA-responsive elements (accounting to eight), which might explain why it can be so significantly up-regulated by low temperature. Moreover, four high-temperature-inducible MaSNAREs, including MaYKT6a, MaSYP81, MaSYP131c and MaSYP2d, all contained high-temperature responsive element STRE in their promoters. This suggested that the presence of these elements function accordingly as in other plants, resulting in a correlation in the MaSNAREs’ expression during the encountered temperature stress conditions.

3.4. MaSYP121 and MaSNAP33a Function in S. indica-Banana-Foc Interactions

The SNARE protein complex (composed of AtVAMP721/722, AtSYP121, AtSYP122 and AtSNAP33) is necessary for Arabidopsis to resist powdery mildew [39]. In this study, MaSYP121, MaVAMP72a and MaSNAP33a were identified to be homologous genes of Arabidopsis SYP121 (PEN1), VAMP721 and SNAP33 (Supplemental Table S1), respectively. The expression levels of Arabidopsis PEN1 could be significantly up-regulated by E. cichoracearum inoculation, and PEN1 appeared to be actively recruited to the papillae during fungal attack, thereby enhancing powdery mildew resistance [49]. In rice, the expression of OsSYP121 could be significantly up-regulated after rice blast fungus Magnaporthe griseat inoculation, and transgenic plants overexpressing OsSYP121 exhibited enhanced resistance to rice blast [50]. Our study revealed that the expression of MaSYP121 was induced by FocTR4, suggesting that it functions in the banana in response to Fusarium wilt. Notably, S. indica treatment alleviated the up-regulation of MaSYP121 by FocTR4, which might be related to the enhanced disease resistance conferred by S. indica. Through pathogen resistance assays in tobacco leaves, we found that its overexpression greatly suppressed the infection of both Foc1 and FocTR4, indicating that this gene functioned positively in resisting Foc penetration.

The SNARE protein complex formed by barley HvSNAP34 (homologous protein of AtSNAP33) and HvSYP121 (homologous protein of AtSYP121) is crucial in the process of resistance to barley powdery mildew [26]. Our study revealed that MaSNAP33a was down-regulated while MaSNAP33b and MaSNAP33e were up-regulated by FocTR4, by FocTR4. However, S. indica up-regulated greatly the expression of MaSNAP33a, MaSNAP33b and MaSNAP33e. Foc resistance assays in tobacco leaves revealed that, although its inhibitory effect is not strong as that of MaSYP121, the overexpression of MaSNAP33a could also suppress the infection of both Foc1 and FocTR4.

The expression of MaVAMP72a was down-regulated by FocTR4, however, its expression level was significantly up-regulated by S. indica and Si-FocTR4 treatment. Unlike MaSYP121 and MaSNAP33a, the overexpression of MaVAMP72a improved the growth and infection in both Foc1 and FocTR4. The differential expression changes the patterns of these three MaSNAREs under mutualistic and pathogenic fungal colonization and their diverse effects in resisting Foc infection suggested that they have different functions. Their functional mechanisms need to be further explored in the future research. However, it can be concluded that MaSYP121 and MaSNAP33a function positively in resisting Foc infection.

4. Materials and Methods

4.1. Fungal Strains and Plant Materials Used in This Study

The Serendipita indica, FocTR4 and Foc1 strains used in this study were provided by the Institute of Horticultural Plant Biotechnology, Fujian Agriculture and Forestry University. The 30-day-old tobacco (N. benthamiana) seedlings used for transient overexpression were provided by Shanxi Agricultural University and cultured in an artificial climate culture chamber maintained at 28 °C, 80% relative humidity, and a photoperiod of 16 h light (1500 ± 200 lx)/8 h dark.

4.2. Identification and Bioinformatic Analysis of Banana SNARE Genes

The banana (Musa acuminata) gDNA, CDS and protein sequences and genome annotation files were downloaded from the banana genome website (https://banana-genome-hub.southgreen.fr/, accessed on 4 July 2022). A total of 64 Arabidopsis SNARE family protein sequences were downloaded from TAIR (https://www.arabidopsis.org, accessed on 4 July 2022) and used as query sequences to BLASTP against the banana protein database using E ≤ 1 × e⁻⁵ as criterion. Following this, the hidden Markov model files for SNARE (PF05739), Syntaxin (PF00804), Longin (PF13774), Synaptobrevin (PF00957), SEC20 (PF03908), V-SNARE-C (PF12352), V-SNARE (PF05008) and USE1 (PF09753) were downloaded from the Pfam database (http://pfam.xfam.org/, accessed on 4 July 2022) and searched against the banana genome data using HMMER software (E-value ≤ 1 × 10⁻⁵). The identified candidate proteins were further subjected to domain validation by using CDD (http://smart.embl-heidelberg.de/, accessed on 4 July 2022), and sequences with none of the conserved domains were removed. The physicochemical properties, signal peptides, transmembrane structure and subcellular localization of remained MaSNAREs were predicted using ExPASy (https://web.expasy.org/protparam/, accessed on 5 July 2022), SignalP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP-3.0/, accessed on 5 July 2022), TMHMM Server v.2.0 (http://www.cbs.dtu.dk/services/TMHMM/, accessed on 5 July 2022) and Euk-mPLoc 2.0 (http://www.csbio.sjtu.edu.cn/bioinf/euk-multi-2/, accessed on 5 July 2022), respectively.

The conserved motifs of MaSNAREs were analyzed using the MEME (http://meme-suite.org/tools/meme, accessed on 6 July 2022) with the default value set as 20. GSDS (http://gsds.cbi.pku.edu.cn/, accessed on 6 July 2022) was used for gene structure analysis of MaSNAREs. The generated conserved motifs and gene structure results were visualized using TBtools [51].

4.3. Phylogenetic Analysis of Banana SNARE Genes

Multiple sequence alignment of SNARE proteins from banana and A. thaliana was performed using the MUSCLE method, and the phylogenetic tree was constructed using the maximum likelihood method of MEGA-X software (Jones-Taylor-Thornton (JTT) model, complete deletion and bootstrap values were 1000 replicates).

4.4. Chromosome Location and Gene Duplication Analysis of Banana SNAREs

According to the banana genome annotation information, the chromosome locations of MaSNAREs were displayed using TBtools [51]. MCscanX software was used to analyze the gene duplication events among MaSNAREs. By using Ka/Ks_Calclator 2.0 software, the synonymous substitution value (Ka) and synonymous substitution value (Ks) of the identified duplicated MaSNARE genes were calculated [52]. The divergence time (T) of gene duplication event was calculated using the formula: T = Ks/2λ × 10⁻⁶ Mya (λ: synonymous substitution rate; λ = 4.5 × 10⁻⁹) [53].

4.5. Prediction of Cis-Acting Elements in MaSNAREs Promoters

The 2000 bp sequences upstream of the initiation codon (ATG) of MaSNAREs were extracted from the banana genome data using TBtools and used as promoter sequences. PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 6 July 2022) was used to predict cis-acting elements in the MaSNAREs promoters.

4.6. Expression Analysis of MaSNAREs

The FPKM values of MaSNAREs in root, corm, leaf and fruit, in leaves treated at 4 °C for 24 h (low temperature treatment), 45 °C for 3 d (high temperature treatment) and 28 °C (control), and in roots treated by FocTR4 (Foc group, harvested at two months post FocTR4 inoculation), S. indica (Si group), S. indica and FocTR4 co-treated (SF group, harvested at two months post FocTR4 inoculation) and non-inoculated control group (CK group) were extracted from the transcriptome data. Following this, their expression values were normalized by log₂ (FPKM + 1) and used for the heatmap drawing using Heatmap embedded in TBtools.

RNAprep Pure Plant Kit (TIANGEN, Beijing, China) was used to isolate RNA from CK, Foc, Si and SF root samples, and cDNA was synthesized using PrimerScript^TM RT Reagent Kit (Perfect Real Time) (TaKaRa, Dalian, China). By using Primer3 (https://primer3.ut.ee/, accessed on 10 July 2022), primers used for the quantitative real time PCR analysis of 10 selected genes (including MaSYP121, MaSYP122, MaVAMP72a, MaSNAP33a, MaSYP6a, MaKNOLLE, MaSYP5a, MaBET1a, MaSYP131a and MaNPSN11a) were designed (Table S4). A qRT-PCR analysis was performed for the validation of the expression change patterns of the 10 selected MaSNARE genes in the roots of the F, Si, SF and CK groups. The qRT-PCR reactions were performed on Roche Light-Cycler480 with the CAC as the internal reference gene. The 25 μL qRT-PCR system contained 12.5 μL of Dream Taq^TM Green PCR Master (2×), 9.5 μL of ddH₂O, 1 μL of template cDNA, and 1 μL each of the upstream and downstream primers. The reaction conditions were as follows: pre-denaturation at 95 °C for 3 min, denaturation at 95 °C for 5 s, annealing at 60 °C for 30 s, extension at 72 °C 15 s, 40 cycles. The relative expression of MaSNAREs genes in different groups was calculated using the 2^−ΔΔCT method. SPSS 26.0 software was used for significance analysis and GraphPad Prism 8 software was applied for figure drawing.

4.7. Gene Cloning and Vector Construction

DNAMAN was used to design gene-specific primers for cloning the full-length coding sequences (CDSs) of MaSYP121, MaVAMP72a and MaSNAP33a (Table S4). The cDNA was synthesized using the RevertAid First-strand cDNA synthesis Kit (Thermo Scientific, Shanghai, China). The 25 μL gene amplification system contained 1 μL cDNA, 1 μL each of forward and reverse primers, 12.5 μL 2 × Green mix and 9.5 μL ddH₂O. The amplification procedure was as follows: pre-denaturation at 95 °C for 3 min, denaturation at 95 °C for 20 s, annealing at 57 °C for 30 s, extension at 72 °C for 1 min, 35 cycles; and final extension at 72 °C for 10 min. The PCR product was gel-extracted, ligated into 18-T vector and transformed into E. coli DH5α. Positive clones were selected and sent to Sangon Biotech (Shanghai) Co., Ltd. for sequencing verification. According to the sequencing results, primers for MaSYP121, MaVAMP72a and MaSNAP33a vector constructions were further designed. By using the TA plasmid-carrying target gene as a template, target genes with the same homologous arm sequences were amplified, gel-extracted, ligated into the pBI123 vector using a ready-to-use seamless cloning kit (Sangon Biotech, Shanghai, China) and transformed into Agrobacterium GV3101 for further use.

4.8. Foc Resistance Assays in Tobacco Leaves Transiently Overexpressing MaSYP121, MaVAMP72a and MaSNAP33a

Agrobacterium GV3101 carrying pBI121-MaSYP121, pBI121-MaVAMP72a and pBI121-MaSNAP33a recombinant plasmids and empty vector (EV, as control) were shake-cultured to OD₆₀₀ = 1.5–2.0, centrifugated at 6000 rpm for 10 min to remove supernatant solutions, re-suspended with buffer (containing 10 mM/L MgCl₂ + 10 mM/L MES + 100 μM/L acetosyringone (As), pH = 5.8), adjusted to OD₆₀₀ = 0.8–1.0, and activated by shaking at 28 °C at 200 rpm for 20–30 min [54]. Leaves from 30-day-old N. benthamiana plants were gently pricked with a needle and injected with a volume of about 40 μL Agrobacterium solution using a 1 mL syringe from the abaxial side. Following this, the tobacco plants were removed to a culture chamber (25 ± 2 °C, relative humidity ~80%) for dark culture. Two days later, leaves were harvested and used for Foc inoculation.

Foc1 and FocTR4 fungi that had been cultured on PDA plates for 7 days were used as inoculation materials. The pathogen discs with a diameter of 5 mm were made and placed on the back of the tobacco leaves. The leaves were then placed on a moist filter paper in a petri dish with a diameter of 90 mm. To maintain a high humidity environment, leaf petioles were wrapped with moist cotton. The culture dishes were placed in a constant temperature incubator (28 ± 1°C, 12 h light/12 h dark photoperiod). After 48 h, leaf lesions in Foc1 inoculated tobacco leaves were photographed with UV imaging and the lesion area was calculated. As the lesions caused by FocTR4 were much smaller than that caused by Foc1, leaf lesions in FocTR4 inoculated tobacco leaves were observed at 72 h post inoculation. The lesion areas in MaSYP121, MaVAMP72a and MaSNAP33a overexpressing leaves were compared with leaves overexpressing the EV to show their inhibitory effect [55]. This experiment was performed with three biological replications, of which each was a mixed sample of six leaves.

5. Conclusions

In this study, 84 SNARE genes which can be further divided into Qa-, Qb-, Qc-, Qbc- and R-SNARE subfamilies were identified from the M. acuminata genome. The gene duplication events analysis revealed that segmental duplications contributed greatly to the expansion of this gene family. A large number of phytohormones and stress-responsive elements were identified from MaSNAREs promoters, suggesting that their expression could be influenced by phytohormones and many other factors. Consistently, the expression of some MaSNAREs containing low-temperature and high-temperature responsive elements in their promoters were influenced by high and low temperature stresses. In addition, we found that the expression of many MaSNAREs was affected by FocTR4 and S. indica infection. Pathogen resistance assays in tobacco leaves showed that MaSYP121 and MaSNAP33a overexpression could inhibit the penetration of both Foc1 and FocTR4, suggesting that they play positive roles in resisting Foc infection in banana. The functions of MaSNAREs in banana stress responses should be focused on in future studies. Our study will be helpful for understanding the roles of MaSNAREs in banana and can provide valuable gene resources for banana resistance breeding in the future.