A TIR-NBS-LRR Gene MdTNL1 Regulates Resistance to Glomerella Leaf Spot in Apple

Glomerella leaf spot (GLS), caused by the fungus Colletotrichum fructicola, is one of the most devastating apple diseases. Our previous study reported that the GLS resistance locus was defined on the chromosome 15 region. Here, we further found a single-nucleotide polymorphism (SNP) site (SNP7309212) in the GLS resistance that was able to distinguish resistant cultivars (lines) from susceptible ones. On the basis of the SNP site, we cloned a TNL gene from the GLS resistant locus and named it MdTNL1 (NCBI Accession Number: ON402514). This gene contains a toll/interleukin-1 receptor transmembrane domain (TIR), nucleotide-binding sites (NBS), and leucine-rich repeat (LRR) domain. Subcellular location indicated that MdTNL1 was expressed in the nucleus and cell membrane. Ectopic overexpression of MdTNL1 in Nicotiana benthamiana caused cell death. We further demonstrated allelic polymorphisms in MdTNL1. It is noteworthy that NBS and LRR domains of the MdTNL1 protein serve as the repository for generating allelic diversity. Quantitative real-time PCR (qRT-PCR) assay revealed that MdTNL1 was highly expressed in resistant apple cultivar ‘Fuji’ after inoculation with C. fructicola, whereas susceptible cultivar ‘Golden Delicious’ exhibited low expression after inoculation. Over-expression of MdTNL1-1 in susceptible apple fruits and leaves improved disease resistance, while in ‘Orin’ calli, silencing the MdTNL1-1 gene conversely decreased GLS resistance. In conclusion, we identified a GLS associated with SNP7309212 and demonstrated that a TIR-NBS-LRR gene MdTNL1-1 positively regulates GLS resistance in apple.


Introduction
Glomerella leaf spot (GLS) is a severe apple disease caused by Colletotrichum fructicola, which has affected the industrial development of apple (Malus domestica) in the last decades [1,2]. GLS mainly occurred in some major apple cultivars, such as 'Gala' and 'Golden Delicious'. Some other widely cultured apple cultivars, such as 'Fuji' and its derivative cultivars, exhibited high resistance to GLS [3]. GLS was first reported in the United States in 1970, but it did not come to attention for the less damage to the apple production [4]. In China, GLS was first found in Henan province, in the middle area of China, in 2008 [2]. Climate conditions of high temperature and high humidity in this area are suitable for GLS occurrence. Once the disease occurs, it develops rapidly, leading to necrosis spots in the leaves followed by defoliation of leaves [5]. In 2013, the 'Gala' orchard in Xianyang City, China, suffered from leaf necrosis with an area of 1309 hm 2 , which is a typical manifestation of GLS. In 2018, more than 70% of GLS diseases occurred in Wugong County, China. GLS also weakens tree vigor, causes fruit necrotic lesions, decreases fruit quality, and affects the fruit growth and the fruit setting rate of the next year [6,7]. For its rapid development and the difficulty of prevention, GLS has become an epidemic fungal disease that seriously threatens the apple industry in China [8]. For effective management of this fungal disease, it is necessary to explore the defense mechanism of resistant apple cultivars. Previous studies explored the apple defense mechanism against GLS. MicroRNA Md-miRln20 negatively regulates the resistance to GLS by suppressing Md-TN1-GLS expression [9]. MKK4-MPK3-WRKY17-mediated salicylic acid degradation is also the regulation of the GLS resistance [10]. Although these studies revealed the regulation of apple resistance to GLS, the resistance gene specific to GLS remains to be identified.
Nucleotide binding sites and leucine-rich repeats genes (NBS-LRRs or NLRs) constitute a large family that plays a crucial role in plant resistance to bacterial and fungal pathogens [11]. By interacting with effectors of pathogens, NLRs trigger effector-triggered immunity (ETI), leading pathogen-specific immune responses [12,13]. NLR proteins can be divided into two classes on the basis of their N-terminal domains, including the toll/interleukin-1 receptor (TIR) for NLRs proteins (TNL) and the coiled-coil (CC) for NLRs proteins (CNL) [14]. TNL have been identified as resistant genes in Arabidopsis thaliana [15], grape (Vitis vinifera) [16], rice (Oryza sativa) [17], and cotton (Gossypium hirsutum). In apple plants, the resistance locus Rvi15 (Vr2) for apple scab has been found to contain three TNL genes [18]. The quantitative trait locus for resistance to fire blight in apple is linked to NLR gene [19], and powdery mildew resistance gene Pl1 is also linked to TNS genes [20]. However, whether the TNL gene is the key factor regulating apple GLS resistance remains to be known.
Our previous research has reported that apple GLS resistance is controlled by a recessive single gene on chromosome 15 [3], and the resistance gene loci (Rgls) are located between SNP 4208 and SNP 4257 , as determined by high-resolution melting (HRM) technique [21]. However, the major recessive gene has not been cloned. In the present study, we cloned the gene responsible for GLS resistance on the basis of the SNP analysis and further addressed its function in GLS resistance.

GLS Resistance Evaluation of Apple Cultivars
For KASP analysis, we evaluated the GLS resistance of 56 apple germplasms, including 22 common cultivars and 34 F1 hybrids from a cross of 'Fuji' with 'Golden Delicious'. The spore of C. fructicola were dotted on the detached leaves of the 56 apple germplasms for inoculation. The lesion diameters of the inoculated leaves were measured 3 days after inoculation. Five cultivars that exhibited obvious lesions including 'Golden Delicious', 'Non-russet sport of Golden Delicious', 'Ruihong', 'Shuangyanghong', and 'Gala' were susceptible to GLS. The other 17 cultivars showed high resistance to C. fructicola (Figures 1A and S1; Table S1). Among the 34 F1 hybrids, 14 lines exhibited susceptible disease and 21 lines were resistant to GLS (Figures 1A and S1; Table S1). By spraying conidia suspension of C. fructicola on the leaves of susceptible cultivar 'Golden Delicious', visible lesions appeared on the leaves of 'Golden Delicious' 36 h after inoculation. There are about 40 necrosis spots per cm 2 of leaf area 72 h after inoculation. No visible necrotic lesions were observed on the leaves of 'Fuji' 72 h after inoculation, indicating that 'Fuji' is highly resistant to GLS ( Figures 1B and S2A).

SNP7309212 Sites Show Polymorphism in Resistant and Susceptible Apple Germplasms
On the basis of the previous finding that the resistance gene was located at the 4.1-4.6 Mb candidate region of chromosome 15 by whole-genome resequencing [1], we looked for SNP sites in this region by KASP sequencing using the susceptible and resistant cultivars evaluated above. According to Liu's results [1], we selected 17 SNP sites in the 4.1-4.6 Mb candidate region (Table S2). KASP detection on 17 SNP sites followed the random using of resistant and susceptible materials for 17 SNP sites (Tables S3 and S4). We found that among these SNP sites, only one SNP site was different between resistant and susceptible leaves. This SNP (designate as SNP7309212) was inferred to be closely associated with GLS resistance. Among the examined apple germplasms, there are two genotypes at SNP7309212, namely, homozygous genotype AA and heterozygous genotype TA, which correspond exactly to resistant and susceptible germplasms. It is noted that all AA genotypes were clustered together, while TT genotypes were clustered together with heterozygous genotype TA ( Figure 1C; Table S4). PCR amplification and sequencing confirmed the resistant/sensitive pattern represented by A/T (isoleucine/phenylalanine) ( Table S5). The SNP7309212 locus appeared different between disease-resistant and susceptible apple germplasms. In addition, high-resolution melting (HRM) analysis showed that obviously

SNP 7309212 Sites Show Polymorphism in Resistant and Susceptible Apple Germplasms
On the basis of the previous finding that the resistance gene was located at the 4.1-4.6 Mb candidate region of chromosome 15 by whole-genome resequencing [1], we looked for SNP sites in this region by KASP sequencing using the susceptible and resistant cultivars evaluated above. According to Liu's results [1], we selected 17 SNP sites in the 4.1-4.6 Mb candidate region (Table S2). KASP detection on 17 SNP sites followed the random using of resistant and susceptible materials for 17 SNP sites (Tables S3 and S4). We found that among these SNP sites, only one SNP site was different between resistant and susceptible leaves. This SNP (designate as SNP 7309212 ) was inferred to be closely associated with GLS resistance. Among the examined apple germplasms, there are two genotypes at SNP 7309212 , namely, homozygous genotype AA and heterozygous genotype TA, which correspond exactly to resistant and susceptible germplasms. It is noted that all AA genotypes were clustered together, while TT genotypes were clustered together with heterozygous genotype TA ( Figure 1C; Table S4). PCR amplification and sequencing confirmed the resistant/sensitive pattern represented by A/T (isoleucine/phenylalanine) ( Table S5). The SNP 7309212 locus appeared different between disease-resistant and susceptible apple germplasms. In addition, high-resolution melting (HRM) analysis showed that obviously different melting curves were observed between resistant and susceptible germplasms at SNP 7309212 ( Figure 1D). These results indicated that SNP 7309212 marker is closely associated with GLS resistance in apple.

Cloning and Characterization of GLS-Resistant Gene
We cloned a MdTNL1 gene from 'Golden Delicious' and 'Fuji' and obtained seven variants (MdTNL1-1-MdTNL1-7). Most nucleotides of the seven variants were identical, except several single nucleotide sites and a fragment of 3 terminal, which showed sequence polymorphism between the seven variants ( Figure 2A). A and T base substitutions occurred at the SNP 7309212 sites of MdTNL1-7, resulting in nonsynonymous amino acid substitutions (I/F) at this site ( Figure 2A). Among them, MdTNL1-1 exists only in 'Fuji', and MdTNL1-7 appears only in 'Golden Delicious', whereas MdTNL1-2-MdTNL1-6 sequences were found in both cultivars. Domain analysis revealed that MdTNL1 protein, which was similar to the reported TNLs in different species, contains three major conserved domains: N-terminal TIR, NB-ARC, and C-terminal LRR ( Figure 2B). MdTNL1-1 sequence (NCBI Accession Number: ON402514) was chosen as the target sequence MdTNL1 for further research. It encoded an 880 amino acid protein, and the full-length CDS consisted of 2640 bp. qRT-PCR analysis demonstrated that in apple, the expression level of MdTNL1-1 was highest in leaves, followed by fruits, and the lowest in flowers ( Figure 3A). Subcellular localization showed that the MdTNL1-1-GFP fusion protein was localized at the nucleus and plasma membrane ( Figure 3B). In addition, MdTNL1 expression was significantly upregulated in 'Fuji' after inoculation but was suppressed in 'Golden Delicious' ( Figure S2B). These results suggest that the level of MdTNL1 expression is associated with GLS resistance in apple. In addition, overexpression of MdTNL1-1 in the leaves of Nicotiana benthamiana for 7 days caused tissue necrosis in tobacco leaves, which was confirmed by the trypan blue staining. DAB staining showed the H 2 O 2 accumulation at the necrotic sites of tobacco ( Figure 3C). different melting curves were observed between resistant and susceptible germplasms at SNP7309212 ( Figure 1D). These results indicated that SNP7309212 marker is closely associated with GLS resistance in apple.

Cloning and Characterization of GLS-Resistant Gene
We cloned a MdTNL1 gene from 'Golden Delicious' and 'Fuji' and obtained seven variants (MdTNL1-1-MdTNL1-7). Most nucleotides of the seven variants were identical, except several single nucleotide sites and a fragment of 3′ terminal, which showed sequence polymorphism between the seven variants ( Figure 2A). A and T base substitutions occurred at the SNP7309212 sites of MdTNL1-7, resulting in nonsynonymous amino acid substitutions (I/F) at this site ( Figure 2A). Among them, MdTNL1-1 exists only in 'Fuji', and MdTNL1-7 appears only in 'Golden Delicious', whereas MdTNL1-2-MdTNL1-6 sequences were found in both cultivars. Domain analysis revealed that MdTNL1 protein, which was similar to the reported TNLs in different species, contains three major conserved domains: N-terminal TIR, NB-ARC, and C-terminal LRR ( Figure 2B). MdTNL1-1 sequence (NCBI Accession Number: ON402514) was chosen as the target sequence MdTNL1 for further research. It encoded an 880 amino acid protein, and the full-length CDS consisted of 2640 bp. qRT-PCR analysis demonstrated that in apple, the expression level of MdTNL1-1 was highest in leaves, followed by fruits, and the lowest in flowers ( Figure 3A). Subcellular localization showed that the MdTNL1-1-GFP fusion protein was localized at the nucleus and plasma membrane ( Figure 3B). In addition, MdTNL1 expression was significantly upregulated in 'Fuji' after inoculation but was suppressed in 'Golden Delicious' ( Figure S2B). These results suggest that the level of MdTNL1 expression is associated with GLS resistance in apple. In addition, overexpression of MdTNL1-1 in the leaves of Nicotiana benthamiana for 7 days caused tissue necrosis in tobacco leaves, which was confirmed by the trypan blue staining. DAB staining showed the H2O2 accumulation at the necrotic sites of tobacco ( Figure 3C).

MdTNL1-1 Was Required for the Resistance of Apple Plant to C. fructicola
To investigate whether MdTNL1-1 was involved in the resistance to GLS, overexpression was performed in leaves and fruits of susceptible 'Gala' using the agroinfiltration method. Quantitative RT-PCR showed that the significantly increased expressions of MdTNL1-1 were observed in leaves and fruits subjected to gene overexpression (Figures 4 and 5, respectively). Overexpression of MdTNL1-1 improved the resistance of apple leaves to C. fructicola ( Figures 5B and C). The similar phenomenon was also observed in fruits. That is, overexpression of MdTNL1-1 enhanced fruit resistance to C. fructicola (Figures 4B and C). Conversely, overexpression of MdTNL1-7 in fruits enhanced pathogen infection ( Figure S2). These results indicate that MdTNL1-1 plays a positive role in GLS resistance in apple. In order to investigate whether MdTNL1-1 expression affects the growth of C. fructicola, we further knocked down the expression of MdTNL1-1 in 'Orin' calli using RNAi technique. Compared with wild type (WT) and empty vector (EV: pCAMBIA1300), MdTNL1-1-silencing calli exhibited significantly reduced MdTNL1-1 expression and increased the growth of the pathogenic fungi ( Figure 6). These results revealed that MdTNL1-1 expression affected the immune response capacity of apple to C. fructicola.

MdTNL1-1 Was Required for the Resistance of Apple Plant to C. fructicola
To investigate whether MdTNL1-1 was involved in the resistance to GLS, overexpression was performed in leaves and fruits of susceptible 'Gala' using the agroinfiltration method. Quantitative RT-PCR showed that the significantly increased expressions of MdTNL1-1 were observed in leaves and fruits subjected to gene overexpression (Figures 4 and 5, respectively). Overexpression of MdTNL1-1 improved the resistance of apple leaves to C. fructicola ( Figure 5B,C). The similar phenomenon was also observed in fruits. That is, overexpression of MdTNL1-1 enhanced fruit resistance to C. fructicola ( Figure 4B,C). Conversely, overexpression of MdTNL1-7 in fruits enhanced pathogen infection ( Figure S2). These results indicate that MdTNL1-1 plays a positive role in GLS resistance in apple. In order to investigate whether MdTNL1-1 expression affects the growth of C. fructicola, we further knocked down the expression of MdTNL1-1 in 'Orin' calli using RNAi technique. Compared with wild type (WT) and empty vector (EV: pCAM-BIA1300), MdTNL1-1-silencing calli exhibited significantly reduced MdTNL1-1 expression and increased the growth of the pathogenic fungi ( Figure 6). These results revealed that MdTNL1-1 expression affected the immune response capacity of apple to C. fructicola.  (A) Expression of MdTNL1-1 in the 'Gala' apple 9 days after infiltration. WT, wild type fruits with no treatment before inoculation. EV, infiltrated with Agrobacterium containing empty vectors, pCAMBIA1300; OE-MdTNL1-1, infiltrated with Agrobacterium harboring pCAMBIA1300-MdTNL1-1. Data are presented as the mean ± SD of three biological replicates. Asterisks indicate significant difference compared to WT (** p < 0.01, one-way ANOVA test by Tukey's test). (B) Apple fruits were inoculated with C. fructicola 7 days after infiltration. Photographs were taken at 0, 4, and 7 dpi. (C) Apples' lesion diameter measured at indicated time points. (D) H 2 O 2 accumulation analyzed at 36 hpi. Data on (C,D) are presented as the mean ± SD of three biological replicates. Different asterisks indicate a significant difference (** p < 0.01, two-way ANOVA followed by Tukey's post hoc test). Asterisks indicate significant difference compared to WT in the same hours.  are presented as means ± SD of three biological replicates. Different asterisks indicate a significant difference. Asterisk indicates significant difference compared to WT (* p < 0.05, ** p < 0.01; one-way ANOVA test by Tukey's test).

MdTNL1-1 Influenced the Expression of PR Genes, and SA and JA Signaling
To determine whether MdTNL1-1 influences the defense responses, we investigated the expression of PR genes, MdEDS1, MdPAD4, MdPAL, and MdPDF1.2. MdEDS1, Md-PAD4, and MdPAL are important for SA synthesis [22,23]. MdPR1 and MdPR5 are involved in SA signaling pathway [24]. MdPDF1.2 is involved in JA signaling pathway [25]. MdTNL1-1-silencing calli were sampled at 0, 12, and 36 h after inoculation with C. fructicola. Quantitative RT-PCR revealed that MdEDS1, MdPAD4, MdPAL, MdPR1, MdPR5, MdPR2, and MdPR8 showed enhanced expression in WT calli after inoculation with C. fructicola (Figure 7). However, in MdTNL1-1-silencing calli, these genes exhibited lower expression compared with that in control calli. Moreover, the higher MdPDF1.2 expression was observed in MdTNL1-1-silencing calli than that in control calli after inoculation. MdPR10a expression did not show significant difference between WT and MdTNL1-1-silencing calli after inoculation. The above results indicate that C. fructicola infection influences SA and JA signaling as well as PR gene expression, suggesting that MdTNL1-1 contributes to resistance to C. fructicola by affecting SA and JA signaling pathways. Data are presented as the mean ± SD of three biological replicates. Aster indicate significant differences compared to WT (** p < 0.01; one-way ANOVA test by Tukey's t (B) Effects of MdTNL1-1 silencing on the calli resistance. The photos were taken 0, 7, or 9 days inoculation with C. fructicola. (C) Quantitation of C. fructicola growth on apple calli. (D) H2O2 a mulation of 'Orin' calli. Data on (C,D) are presented as the mean ± SD of three biological replic Different asterisks indicate a significant difference (** p < 0.01; two-way ANOVA followed Tukey's post hoc test). Asterisks indicate significant difference compared to WT in the same ho

MdTNL1-1 Influenced the Expression of PR Genes, and SA and JA Signaling
To determine whether MdTNL1-1 influences the defense responses, we investiga the expression of PR genes, MdEDS1, MdPAD4, MdPAL, and MdPDF1.2. MdED MdPAD4, and MdPAL are important for SA synthesis [22,23]. MdPR1 and MdPR5 are volved in SA signaling pathway [24]. MdPDF1.2 is involved in JA signaling pathway [ MdTNL1-1-silencing calli were sampled at 0, 12, and 36 h after inoculation with C. fru cola. Quantitative RT-PCR revealed that MdEDS1, MdPAD4, MdPAL, MdPR1, MdP MdPR2, and MdPR8 showed enhanced expression in WT calli after inoculation wit fructicola (Figure 7). However, in MdTNL1-1-silencing calli, these genes exhibited lo expression compared with that in control calli. Moreover, the higher MdPDF1.2 exp sion was observed in MdTNL1-1-silencing calli than that in control calli after inoculat MdPR10a expression did not show significant difference between WT and MdTNL1-1 lencing calli after inoculation. The above results indicate that C. fructicola infection in ences SA and JA signaling as well as PR gene expression, suggesting that MdTNL1-1 c tributes to resistance to C. fructicola by affecting SA and JA signaling pathways. Data on (C,D) are presented as the mean ± SD of three biological replicates. Different asterisks indicate a significant difference (** p < 0.01; two-way ANOVA followed by Tukey's post hoc test). Asterisks indicate significant difference compared to WT in the same hours.

Discussion
In this work, we screened SNP loci using KASP technology in resistant and susceptible apple populations and obtained the GLS resistance associated SNP 7309212 in this study. The SNP 7309212 is located at the coding sequence of MdTNL1 gene in chromosome 15 (Figure 1; Tables S4 and S5). MdTNL1 gene expression was higher in the leaves of resistant cultivar after C. fructicola infection than that in susceptible one, and the expression of the MdTNL1 gene was induced by C. fructicola ( Figure S2B). KASP, HRM, and PCR sequencing revealed SNP 7309212 difference between resistant and susceptible apple cultivars ( Figure 1C,D; Tables S3-S5). SNP 7309212 in the parents 'Fuji' and 'Golden delicious' showed AA and AT, respectively, whereas in the hybrid F1, homozygous AA only existed in the resistant apple, while heterozygous AT and homozygous TT were only present in the susceptible apple varieties, which is consistent with previous reports that GLS resistance was controlled by recessive single gene [3,21]; thus, we clearly found a strong association between the MdTNL1 and GLS.
Allelic variation is an important driving force for plant co-evolution [26,27]. Previous research found that of rice BPH gene responsible for the resistance to BPH organism includes eight alleles [17]. In a study of Verticillium wilt resistance of GhDSC1 gene, it was confirmed that the SNP variation of TNL type gene led to the reduction of nucleotide binding proba-bility, which in turn affected the different resistance of G. hirsutum to cotton verticillium wilt [28]. Similarly, we found seven isomers of MdTNL1 (MdTNL1-1-MdTNL1-7) in apple, which showed polymorphisms in the LRR domain ( Figure 2A). Therefore, polymorphisms in MdTNL1 alleles are assumed to be important for apple defense against C. fructicola.
Most NLR genes play important roles in the process of plant disease resistance. They are important components of the plant immune system. A previous study showed that overexpression of NLR usually causes plant cell death [29]. Transient overexpression of MdTNL1-1 in N. benthamiana [30] resulted in tissue necrosis and H 2 O 2 accumulation ( Figure 3C). Firstly, we found that MdTNL1-1 gene triggered local resistance associated with programmed cell death. In addition, analysis of the tissue specificity of MdTNL1-1 expression in apple by qRT-PCR indicated that MdTNL1-1 functioned in leaves and fruits, two major organs infected by C. fructicola ( Figure 3A). Finally, subcellular location using onion epidermal cells showed that the MdTNL1-1 protein is expressed in the nucleus and the cell membrane ( Figure 3B). Therefore, the discovery of tissue necrosis and reactive oxygen species accumulation has a new understanding for our study of MdTNL1-1 immune research.

Pathogen Inoculation and Resistance Evaluation
The GLS pathogen C. fructicola, kindly provided by Professor Baohua Li from Qingdao Agricultural University, was grown on potato dextrose agar (PDA) at 28 • C for 7 days. Then, C. fructicola was transferred to spore culture solution containing 10 g L −1 KNO 3 , 5 g L −1 KH 2 PO 4 , 2.5 g L −1 MgSO 4 ·7H 2 O, 0.02 g L −1 Fe 2 (SO 4 ) 3 , and 50 g sugar at 28 • C for 3-5 days. Its spores were collected and suspended in deionized water at a concentration of 1×10 5 CFU mL −1 . The detached leaves of 56 apple varieties or transiently expressed apple leaves or fruits were inoculated by dropping 10 µL conidia suspensions on the surface of leaves. After inoculation, the leaves were placed in Petri dishes and kept at 28 • C and 75% humidity. The detached leaves of 'Fuji' and 'Golden Delicious' collected from a field (5 cm in length) or tissue cultured apple 'Gala3' (1 cm in length) were sprayed with spore suspensions of C. fructicola. To evaluate the disease resistance ability of leaves, we defined four ranks. Leaves with lesion diameter between 0 and 2 mm belonged to high resistance, lesion diameter between 2 and 4 mm belonged to medium resistance, lesion diameter between 4 and 8 mm belonged to medium susceptibility, and lesion diameter of more than 8 mm belonged to high susceptibility.

Cloning of MdTNL1
cDNA from the susceptible leaf 'Golden Delicious' or resistant leaves 'Fuji' was used as a template. The primers (named MdTNL1 forward or reserve) were used for PCR to clone MdTNL1. PCR amplification was performed with designed primers as previously described [34].

Transient Expression Assay in Tobacco Leaves, Apple Leaves, and Fruits
The CDS of MdTNL1-1 was cloned and ligated into pCAMBIA1300 and PK7WIWG2D vectors to generate pCAMBIA1300-MdTNL1-1-GFP for overexpression and MdTNL1-1-PK7WIWG2D for RNA interference. According to the method previously described [35], we transformed the new constructed vector into Agrobacterium strain EHA105 and infiltrated Agrobacteria into tobacco leaves, leaves of 'Gala3' apple, and fruits of 'Gala' apple 120 DAFB. The leaves should be fully infiltrated, and the fruits were required to have an injection diameter of 5 cm. The tobacco leaves were visually examined and stained by trypan blue (TPB) or diaminobenzidine (DAB) 7 days after being injected [36,37]. C. fructicola was inoculated at the apple infiltration site 9 days after Agrobacterium infiltration with 0.5 µL. Resistance was assessed 4 days or 7 days after inoculation. Through our experimental observation, we found that in the early stage of C. fruticola infecting the host plant, although the pathogen destroyed the host plant, the host plant did not show disease in a short period of time, but the outbreak time of reactive oxygen species was rapid. Therefore, we selected the pathogen to evaluate the resistance after 4 and 7 days in fruit and 7 and 9 days in calli. The content of H 2 O 2 was measured at 0, 12, and 36 h after inoculation using an H 2 O 2 content detection kit (Solarbio Biotechnology Co., Ltd, Beijing, China) according to instructions provided by the manufacturer. MdTNL1-1 expression analysis was performed 3 or 9 days after infiltration.

Subcellular Localization of MdTNL1-1 and Genetic Transformation of Apple Calli
Agrobacterium EHA105 strain containing the MdTNL1-1-GFP fusion vector or the pCAMBIA1300-GFP control vector were infiltrated into onions. The infiltrated onions were incubated in the dark for 16-24 h, and then the cut samples were observed under a laser confocal microscope (Leica, Heidelberg, Germany) [38,39]. To silence MdTNL1 in apple calli, 15-day-old calli were co-cultured with Agrobacterium carrying MdTNL1-1-PK7WIWG2D for two days on MS medium containing 1.5 mg L −1 2,4-dichlorophenoxyacetic acid (2,4-D) and 0.225 mg L −1 at 25 • C. The calli were then transferred to a selective MS medium containing 200 mg L −1 cefotaxime and 100 mg L −1 kanamycin. The transgenic calli were inoculated with C. fructicola after 7 and 9 days to evaluate the disease resistance.

Expression Analysis of PR Genes and SA-and JA-Related Genes
The apple calli were treated with 1 × 10 5 CFU ml −1 spore suspension, and the calli samples were collected at 12 h and 36 h after treatment for RT-qPCR. Genes expression analysis of MdEDS1, MdPAD4, MdPAL, MdPR1, MdPR5, MdPR2, MdPR8, MdPDF1.2, and MdPR10a were performed using RT-qPCR, as previously described with 4.5 part. The experiment was performed by using three independent biological replicates, and the significant difference was analyzed by two-way ANOVA test (Figure 7). Different asterisks indicate a significant difference (* p < 0.05, ** p < 0.01; two-way ANOVA followed by Tukey's post hoc test). Asterisks indicate significant difference compared to WT in the same hours.

Conclusions
TNL family genes are important objects for studying plant immune defense. Through this study in selected 17 SNP sites within the GLS mapping interval, we found SNP7309212 marker in MdTNL1 was closely associated with GLS resistance. The expression of MdTNL1 was significantly improved in response to specific pathogenic bacteria. Importantly, we showed that MdTNL1-1 overexpression improved resistance of apple to C. fructicola. MdTNL1-1 can play a defensive role against GLS by enhancing ROS capacity as well as regulating JA and SA signaling pathway genes and PR gene expression. Overall, the perspective presented in this study may open a new research path for understanding the complex mechanisms of resistance to GLS in apple plants.  Different asterisks indicate a significant difference (* p < 0.05, ** p < 0.01; two-way ANOVA followed by Tukey's post hoc test). Asterisks indicate significant difference compared to WT in the same hours.

Conclusions
TNL family genes are important objects for studying plant immune defense. Through this study in selected 17 SNP sites within the GLS mapping interval, we found SNP 7309212 marker in MdTNL1 was closely associated with GLS resistance. The expression of MdTNL1 was significantly improved in response to specific pathogenic bacteria. Importantly, we showed that MdTNL1-1 overexpression improved resistance of apple to C. fructicola. MdTNL1-1 can play a defensive role against GLS by enhancing ROS capacity as well as regulating JA and SA signaling pathway genes and PR gene expression. Overall, the perspective presented in this study may open a new research path for understanding the complex mechanisms of resistance to GLS in apple plants.

Data Availability Statement:
The authors declare that all data and materials supporting the findings of this study are available within the article. The data that support the findings of this study are available from the corresponding author upon reasonable request.