Pan-genome cloning and expression analysis of S-RNase homologous gene in SI and SC of different tomato species Authors: Chunwei Li, Panfeng Zhao Published: 2026 ABSTRACT Flowering plants encode polymorphic ribonucleases (S-RNases) that determine the self-incompatibility (SI) phenotype of the style. The pan-genome cloning and expression analysis of the S-RNase homologous genes in SI and self-compatibility (SC) lines of different tomato species have not been reported. In this study, partial sequences of cDNA and DNA of the S-RNase homologous gene (designated ZS) were obtained from S. pimpinellifolium, S. Lycopersicum, S. habrochaites and S. chilense which included SI lines and SC lines. Tissue-specific expression of the S-RNase gene was detected using semi-quantitative RT-PCR and quantitative real-time PCR. We selected 6 ZS genes (two come from the SI lines and four come from SC lines) which come from different tomato species mentioned in this paper to research. ZS genes were exclusively in styles and there were significant differences in expression level of the different tomato species. The expression level of the ZS genes in the styles of the SI tomato clade S. chilense was higher than that in the SC tomato clades S. pimpinellifolium and S. lycopersicum. And, the SC tomato clade S. habrochaites, in which the expression level of the ZS genes was even higher than that in the SI tomato clade S. chilense. The outcomes presented in this research provide the basis for future deeper investigation of the molecular mechanism of SI in tomato.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-57781-y. Introduction Self-incompatibility in flowering plants prevents inbreeding and promotes outcrossing to generate genetic diversity1. To date, SI has attracted much attention and become one of the hot topics in the field of plant molecular biology2–7. SI is genetically controlled by a single S-locus that has multiple S-alleles enabling the recognition and rejection of ‘self’ pollen8. In flowering plants, SI can be classified into sporophytic self-incompatibility (SSI) and gametophytic self-incompatibility (GSI). Among them, GSI is the most widespread SI type, and it is mainly determined by the S-RNase gene in styles. Currently, much work has been done on the isolation and detection of S-RNase genes in commercial fruit species, such as apple (Malus pumila)12–14, pear (Pyrus serotina)11,15,16, sweet cherry (Prunus avium)13,17–19(almond (Prunus dulcis)20,21,22 Japanese plum (Prunus salicina)11,23 and apricot (Prunus armeniaca)24,25. This work has provided useful information for breeders and growers to control pollination effectively. The interaction between style S-RNase allelic genes and pollen S-allele products has not been well understood to date. However, it has been reported that S-RNase plays an important role in the mutual recognition and rejection between pistil and pollen26–29. The ability to transfer the S-RNase gene into petunia (Petunia inflata) and tobacco (Nicotiana alata) provided direct evidence that the biochemical mechanism of GSI involves the ribonuclease activity of S proteins29. In the past decades, great progress has been made in S-RNase genes associated with SI in Japanese pear (P. serotina), apple (M. pumila), sweet cherry (P. avium), almond (P. dulcis), apricot (P. armeniaca) and other Rosaceae species. The S-RNase genes play a crucial role in the self - incompatibility (SI) system of many flowering plants, including those in the Solanaceae and Rosaceae families30. The S-RNase gene was first identified as the female component of the S-locus in the Solanaceae, and then in the Rosaceae and Scrophulariaceae9–11. S-RNases exert their function through cytotoxic activity28. S-RNases result in SI through degradation model in Antirrhinum, Rosaceae and Rutaceae31–34. In this model, SLFs (S-locus F-box proteins) specifically recognize non- self S-RNases and mediate their polyubiquitination, leading to degradation via the 26 S proteasome. In contrast, self S-RNases evade ubiquitination and retain cytotoxic activity, thereby enforcing self-incompatibility (SI)35. Building on the established ubiquitin-proteasome degradation model, a phase separation-driven condensation model has recently been proposed in Petunia. This model posits that S-RNases undergo liquid-liquid phase separation (LLPS) with specific scaffold proteins, forming cytotoxic condensates that disrupt pollen tube cellular homeostasis35,36. In Nicotiana, an alternative theory, known as the compartmentalization model, has been put forward. This model places emphasis on the sub - cellular trafficking process of S-RNases and underscores the regulatory function of HT-B within Nicotiana35,37. GSI is the most widespread SI type and the SI of tomato species is GSI, which is mainly determined by the S-RNase gene in styles. In tomato species, some species exhibit SI, while others display SC. The pan-genome cloning and expression analysis of the S-RNase homologous genes in SI and self-compatibility (SC) lines of different tomato species have not been reported. Moreover, the molecular mechanism of SI in tomato has not been deeply explored further. Therefore, quantitative study of the S-RNase gene(s) in tomato is essential to an understanding of the relationship between S-RNase gene(s) and SI /SC. Results Cloning and sequence analysis of ZS genes in SI and SC of different tomato species The protein alignment (Fig. 1) shows a typical distribution of conserved and variable sequences as well as conserved regions C2, C3, C4 and C5. Genomic PCR products contained intron sequences in a position consistent with all known S-RNase genes, suggesting a conserved genomic structure among these genes. Accession number of S-hab-1 is GU361144 and accession numbers of ZS1-ZS18 genes are JN226557, JN226558, JN226559, JN226560, JN226561, JN226562, JN226563, JN226564, JN226565, JN226566, JN226567, JN226568, JN572893, JN581380, JN581381, JN581382, JQ040814, JQ040815. These sequences exhibit sequence diversity as one notable feature of S-RNases which variable sites scattered throughout the protein. Fig. 1Amino acid sequences encoded by the different tomato species ZS cDNAs aligned for the representative S-RNase sequences. Amino acid sequences encoded by the different tomato species ZS cDNAs aligned for the representative S-RNase sequences. The nucleotide sequences of ZS genes from different tomato species were aligned with sequences of the S - RNase genes from Rosales, Solanales, Gentianales and RNase T2. The biochemical nature of S-proteins was revealed when the sequence of RNase T2 of Aspergillus oryzae was determined and it was unexpectedly found to share sequence similarity with S-proteins. This finding led to the subsequent confirmation that S-proteins have RNase activity in vitro. So, RNase T2 was also included in the Maximum-parsimony (MP) tree and I designated RNase T2 as the root of MP tree. Figure 2 shows the MP tree produced from the DNA alignment. Bootstrap analysis provides strong support for almost all nodes, and distance trees generated using the same alignment had an identical topology (results not shown). The Rosales S-RNase sequences are in a single cluster, the Gentianales S-RNase sequences are in a single cluster, the Solanales S-RNase sequences are in a single cluster, and ZS genes are at the base of the Solanales cluster, with nodes at the base of the Solanales clusters having strong bootstrap support. To further strengthen the conclusions drawn from the MP and Neighbor-Joining (NJ) tree, Maximum Likelihood (ML) tree was constructed. The ML tree yielded highly comparable results (ML tree in Supplementary Information file). Fig. 2Maximum-parsimony tree (5087 steps) derived from aligned S-RNase and ZS DNA sequences. Bootstrap values (1000 pseudoreplicates) are shown at the nodes. Sequences in the tree and their accession numbers are RNase T2 (XM_001817972), ErJ S4 (EU442289), Ma x D S-4 (AF327223), PyC S21 (EU477839), CoC Ccan02 S3 (JN035305), CoR Cres94 S29 (JN035365), CoR Cres92 S28 (JN035364), PsE Pebrac 122 PS1 (JN035312), CoR Cres91 S53 (JN035366), LyHG LhgSRN-2 (AB072479), Pe x H So-RNase (FJ490180), LyB S-BARB9 (HM195031), SoCa N-SC (L40551), SoP SP17 (HM357228), SoS 47 D S (HQ699470), PhyC S16 (L46668), WiS S23 (AY454121). Maximum-parsimony tree (5087 steps) derived from aligned S-RNase and ZS DNA sequences. Bootstrap values (1000 pseudoreplicates) are shown at the nodes. Sequences in the tree and their accession numbers are RNase T2 (XM_001817972), ErJ S4 (EU442289), Ma x D S-4 (AF327223), PyC S21 (EU477839), CoC Ccan02 S3 (JN035305), CoR Cres94 S29 (JN035365), CoR Cres92 S28 (JN035364), PsE Pebrac 122 PS1 (JN035312), CoR Cres91 S53 (JN035366), LyHG LhgSRN-2 (AB072479), Pe x H So-RNase (FJ490180), LyB S-BARB9 (HM195031), SoCa N-SC (L40551), SoP SP17 (HM357228), SoS 47 D S (HQ699470), PhyC S16 (L46668), WiS S23 (AY454121). The tissue-specific expression patterns of the ZS genes in SI and SC of different tomato species We selected 6 ZS genes which come from several tomato species mentioned in this paper to research. Specifically, the ZS9 gene comes from the tomato clade of S. chilense LA0130, the ZS16 gene comes from the tomato clade of S. chilense LA2405, the ZS17 gene comes from the tomato clade of S. habrochaites LA0094, the ZS25 gene comes from the tomato clade of S. pimpinellifolium LA1585, the ZS18 gene comes from the tomato clade of S. lycopersicum LA3475 and the ZS7 gene comes from the tomato clade of S. lycopersicum P86. There were significant differences in the expression of the ZS genes between different tissues in SI and SC of different tomato species using semi-quantitative RT-PCR and quantitative real-time PCR (Fig. 3, Supplementary Information file and 4). The highest expression level was detected in the styles, while there was almost no expression in leaves, petals and anthers. In SI tomato species, the ZS genes were highly and specifically expressed in styles. In contrast, in SC tomato species, there was almost no expression in leaves, petals, and anthers, and low expression in styles. However, the SC tomato clade LA0094 was an exception, in which the ZS genes were highly and specifically expressed in styles. The results presented in Figs. 3 and 4 indicated that a relatively high expression level of ZS genes in styles has the potential to impede the growth of pollen tubes. Fig. 3Expression of the ZS genes in different tissues in SI and SC of different tomato species. The grouping of gels cropped from different parts of the same gel, or from different gels (full-length gels in Supplementary Information file). The red dashed box indicated the cropping boundary of the original image. Expression of the ZS genes in different tissues in SI and SC of different tomato species. The grouping of gels cropped from different parts of the same gel, or from different gels (full-length gels in Supplementary Information file). The red dashed box indicated the cropping boundary of the original image. Fig. 4Relative expression of the ZS genes in different tissues in SI and SC of different tomato species (n = 3 biological replicates, each with triplicate technical assays **p < 0.01). Relative expression of the ZS genes in different tissues in SI and SC of different tomato species (n = 3 biological replicates, each with triplicate technical assays **p < 0.01). Relative expression of the ZS genes in SI and SC of different tomato species style relative expression of the ZS genes in SI and SC of different tomato species style There were significant differences in the expression of the six ZS genes between the styles of SI and SC tomato species using quantitative real-time PCR (Fig. 5). The expression level of the ZS genes in the styles of the SI tomato clade S. chilense was higher than that in the SC tomato clades S. pimpinellifolium and S. lycopersicum. However, the SC tomato clade S. habrochaites LA0094 was an exception, in which the expression level of the ZS genes was even higher than that in the SI tomato clade S. chilense. Fig. 5Relative expression of the ZS genes in SI and SC of different tomato species styles (n = 3 biological replicates, each with triplicate technical assays **p < 0.01). Relative expression of the ZS genes in SI and SC of different tomato species styles (n = 3 biological replicates, each with triplicate technical assays **p < 0.01). ZS gene structure Partial sequences for ZS genes are isolated from the genomes and style cDNAs of SI and SC different tomato species by a degenerate PCR approach. Figure 6a (Supplementary Information file) shows that the length of the genomic sequences is different from that of the style cDNA sequences. Moreover, the alignment result of the 6 ZS nucleotide sequences from the genome with those from the style cDNA shows that the 6 ZS genes have an intron within the cloned sequences, and the intron is located in the C2 and C3 regions (Figs. 2 and 6b). Fig. 6Genome PCR and RT-PCR of the ZS genes in SI and SC of different tomato species styles and analysis of ZS Gene structure. (a) Genome PCR and RT-PCR (Normal-contrast gel image in Supplementary Information file); (b) ZS Gene structure. Genome PCR and RT-PCR of the ZS genes in SI and SC of different tomato species styles and analysis of ZS Gene structure. (a) Genome PCR and RT-PCR (Normal-contrast gel image in Supplementary Information file); (b) ZS Gene structure. Southern blot analysis To examine the copy number of the ZS gene in the tomato genome, Southern blot analysis was performed. As shown in Fig. 8, one hybridization band was observed in the DNA blots of the 3 tomato accessions using KpnI, SacI (Fig. 8a-c) and EcoR I (Fig. 8d-f) digestions since none of these restriction sites are present in the ZS gene specificity probe genomic DNA. This result suggests that the ZS gene is likely to be a single-copy gene. Fig. 7The details of materials. The details of materials. Fig. 8Southern blot analysis of the S-RNase homologous gene in tomato species. (a–c) using KpnI, SacI digestion; (d–f) using EcoRI digestion; (a,d): LA3863; (b,e): LA3864; (c,f): LA1255. The arrows indicate the different size hybridized bands. The grouping of gels cropped from different parts of the same blot, or from different blots (full-length blots in Supplementary Information file). The red dashed box indicated the cropping boundary of the original image. Southern blot analysis of the S-RNase homologous gene in tomato species. (a–c) using KpnI, SacI digestion; (d–f) using EcoRI digestion; (a,d): LA3863; (b,e): LA3864; (c,f): LA1255. The arrows indicate the different size hybridized bands. The grouping of gels cropped from different parts of the same blot, or from different blots (full-length blots in Supplementary Information file). The red dashed box indicated the cropping boundary of the original image. Discussion Since the discovery of the S-RNase gene almost two decades ago, much of what we have learned about the Solanaceae type of SI is limited to this female determinant of the SI interaction, However, there are many questions that needed to answer. One question is whether and how S-RNase gene exists and is expressed in SC lines. In this study, we showed that S-RNase gene is not only expressed in the styles of SI tomato species but also is expressed in the styles of SC tomato species. An attempt to amplify the S - RNase sequence using primer pairs for conserved regions was unsuccessful when using genomic DNA and cDNA from different tomato cultivars38. We successfully amplified the S-RNase gene using genomic DNA and cDNA from the tomato cultivars LA3475 and P86. As a high level of S-RNase expression in the style has previously been shown to be essential for the gametophytic self-incompatibility response in the Solanaceae, the lack of S-RNase expression may be why tomato styles fail to reject self-pollen38. However, we found that at the RNA expression level, a high level of S-RNase expression in the style of SC in some tomato species (for example, LA0094) (Figs. 4 and 5). The unusual behavior of SC tomato LA0094 needs further in-depth research. In genome gene structure, we have found that S-RNase genes are the same in SI and SC tomato species, but they are somewhat different in cDNA gene structure. Deletion of one nucleotide can result in a reading frame shift and a truncated protein, which can result in natural stylar-part mutants lacking S-RNase activity39,40. Here, in several SC tomato species, we have found that not only is there a deletion of one nucleotide in the S-RNase gene cDNA sequence, but there is also an insertion of one nucleotide in the sequence, which results in a reading frame shift. This alteration would theoretically render the messenger RNA untranslatable into a functional protein. Further research is required to verify whether these mutations actually result in truncated or non-functional proteins at the translational level. Structural analysis of S-RNase sequences indicates a conserved structural framework across Solanaceae, Rosaceae, Plantaginaceae, and Rutaceae35,41. This framework comprises five conserved domains (designated C1 to C5) and two visually distinct hypervariable regions (termed Hva and Hvb)35,41. As shown in Fig. 1, the protein alignment of S-RNase in tomato displays a typical pattern of how conserved and variable sequences are distributed. It also clearly depicts the conserved regions C2, C3, C4, and C5. Although it is important to focus on the S-RNase role to the SI, additional proteins such as HTA, HTB, and SLF are required for the SI response. However, most of these candidate proteins identified to date do not appear to be specific to the SI system42. Physiological studies have demonstrated that pollen tube growth in solid and hollow styles is heterotrophic and occurs at the expense of stylar reserves43. It has been proposed that the extracellular matrix of the style assists pollen tube migration and provides nutritive support for the growing pollen tubes43. Therefore, it is worth exploring whether there are other nutrients related to SI. Future research is required to elucidate how many types of nutrients are related to SI, and how these molecules include in S-RNase, HTA, HTB, SLF interact with each other and the specific roles they play in SI. Materials and methods Plant material SI lines for LA2405 (S. chilense), LA0130 (S.chilense), LA2951 (S. lycopersicoides), LA3863 (S. habrochaites), LA3864 (S. habrochaites) and SC lines for LA3475 (S. lycopersicum), LA1585 (S. pimpinellifolium), LA0094 (S. habrochaites), LA1255 (S. habrochaites) were supplied by TGRC (the C.M. Rick Tomato Genetics Resource Center), and SC lines for P86 (Cultivar) was kept in ourselves laboratory (Seeds are stored in a seed storage cabinet at temperature/humidity-controlled (4 °C, 15% RH), in Nanchang Normal University). The details of materials are shown in Fig. 7. Primer design and PCR PCR amplification of S-RNase genes was performed as described by Kondo et al. (2002)44. PCR was performed with ~ 100ng of genomic DNA or 1 µL of cDNA template. The 50 µL of reaction volume contained 0.5 µL ExTaq, 5ul 10× PCR buffer (Mg2+ Plus) (TaKaRa), 4 µL dNTP (2.5 mM), 1 µL of each primer (20 µM), 37.5 µL ddH2O and 2 µL DNA (100 ng). Two sets of degenerate outer and inner primers were used to amplify S-RNase sequences: outer C2 primer LP-5 A: AA(T/C)TT(T/C)AA(C/G/A)AT(T/C/A)CA(T/C)GG, inner C2 primer LP-5B: AT(T/C/A)CA(T/C)GG(C/G)(G/C)T(T/A/G)TGGCC(A/T/G)GAT, outer C5 primer LP-3 A: GGT (T/C)(G/T/A)A(A/G)ACATAT(G/A/T)CC, and inner C5 primer LP-3B: ATAT(G/A/T)CC(T/A)A(T/C)(T/C)TC(A/C)(G/T/A)(T/G/A) (T/A)AGTT CAA. Outer primer (LA-5 A, LA-3 A; 35 cycles: 95 °C, 60 s; 50 °C, 60 s; 72 °C, 60 s) PCR products were diluted (1:20) and amplified using the inner set (i.e., LA-5B, LA-3B; 35 cycles: 95 °C, 60 s; 50 °C, 60 s; 72 °C, 60 s). Cloning and DNA sequencing PCR products were purified using the GENOMED quick kit (GENOMED GmbH, Löhne, Germany) and ligated into the pMD18-T vector (TaKaRa). The recombinant plasmid pMD18-T/ZS was then transformed into E. coli DH5ཁ competent cells and recombinant clones were selected for a PCR-based screen using vector specific primers (M13+/M13-). And then the Plasmid DNA from clones was sequenced commercially (Invitrogen, San Diego). Sequences of different tomato species ZS genes have been deposited in GenBank with the accession nos. JN226557, JN226558, JN226559, JN226560, JN226561, JN226562, JN226563, JN226564, JN226565, JN226566, JN226567, JN226568, JN572893, JN581380, JN581381, JN581382, JQ040814, JQ040815, JQ040816, JQ040817, JQ040818, JQ040819, JQ040820, JQ040821, JQ040822. And JQ040816, JQ040817, JQ040818, JQ040819, JQ040820, JQ040821, and JQ040822 were identified as pseudogenes due to frameshift mutations caused by single-nucleotide insertions or deletions. Protein domain analysis was done using the DNAMAN software. All evolutionary analyses were performed using the Phylip package of programs. DNA alignments were generated using TRANALIGN (http://bioweb.pasteur.fr/docs/EMBOSS/tranalign), based on ClustalW (v1.8) protein alignments of the corresponding sequences. MP tree and NJ tree offer the distinct advantage of being straightforward to implement. MP tree was estimated from the DNA alignment using DNAPARS with default settings. NJ tree was generated using NEIGHBOR, with a DNA distance matrix computed using DNADIST. Bootstrap analyses were carried out using 1000 bootstrap pseudoreplicates. And we constructed the ML tree to strengthen the conclusions. RT-PCR analysis At the stage of full anthesis, total RNA was isolated from leaves, petals, anthers, and pistils. Total RNA was prepared using RNAprep pure Plant Kit (TIANGEN). DNA contamination was removed by treating isolated RNA (1 mg) with 3 units of DNase I (TIANGEN). First-strand cDNA synthesis was carried out using RevertAidTM First Strand cDNA Synthesis Kit (#K1621, Fermentas). Less than 200 bp PCR products were separated by 3% gel using the Gel electrophoresis. Sequences of the gene-specific primers used in this experiment are shown in Table 1. Table 1Preferred PCR primer pairs in this experiment.Tomato accessionTarget genePrimer namePrimer sequence 5′ → 3′PCR typeAnnealing temperature (°C)Extension time (s)Product size (bp)LA1585ZS10ZS10FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS10RGGCACGACTGATTTGGTCACCGGZS10TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS10TRGGCACGACTGATTTGGTCACCGGLA3475ZS18ZS18FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS18RGGCACGACTGATTTGGTCACCGGZS18TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS18TRGGCACGACTGATTTGGTCACCGGLA0094ZS17ZS17FTGGGCTTTGGCCGGATATAAAGGGTRT-PCR /5060109ZS17RGGGCCAGTGCTTGTTCCGCTZS17TFTGGGCTTTGGCCGGATATAAAGGGTReal-time PCR5515109ZS17TRGGGCCAGTGCTTGTTCCGCTLA0130ZS9ZS9FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS9RGGCACGACTGATTTGGTCACCGGZS9TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS9TRGGCACGACTGATTTGGTCACCGGLA2405ZS16ZS16FTCATGGGGTTTGGCCAGATAACAAGRT-PCR /5060127ZS16RTTGAGGCCAGCGTTTGTCCAGCZS16TFTCATGGGGTTTGGCCAGATAACAAGReal-time PCR6015127ZS16TRTTGAGGCCAGCGTTTGTCCAGCP86ZS7ZS7FTGCCCGCTGGCCTCAATTGACRT-PCR /5060229ZS7RGGCACGACTGATTTGGTCACCGGZS7TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS7TRGGCACGACTGATTTGGTCACCGGActin Gen (BG140412)*AG-FACCTGATGAAGATCCTCACRT-PCR5660467AG-RATCCTCCAATCCAGACACActin gene (BT013524)*AGTFGCGAGAAATTGTCAGGGACGTReal-time PCR5515123AGTRTGCCCATCTGGGAGCTCAT*The GenBank accession nos. Table 1Preferred PCR primer pairs in this experiment.Tomato accessionTarget genePrimer namePrimer sequence 5′ → 3′PCR typeAnnealing temperature (°C)Extension time (s)Product size (bp)LA1585ZS10ZS10FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS10RGGCACGACTGATTTGGTCACCGGZS10TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS10TRGGCACGACTGATTTGGTCACCGGLA3475ZS18ZS18FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS18RGGCACGACTGATTTGGTCACCGGZS18TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS18TRGGCACGACTGATTTGGTCACCGGLA0094ZS17ZS17FTGGGCTTTGGCCGGATATAAAGGGTRT-PCR /5060109ZS17RGGGCCAGTGCTTGTTCCGCTZS17TFTGGGCTTTGGCCGGATATAAAGGGTReal-time PCR5515109ZS17TRGGGCCAGTGCTTGTTCCGCTLA0130ZS9ZS9FTGCCCGCTGGCCTCAATTGACRT-PCR/5060229ZS9RGGCACGACTGATTTGGTCACCGGZS9TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS9TRGGCACGACTGATTTGGTCACCGGLA2405ZS16ZS16FTCATGGGGTTTGGCCAGATAACAAGRT-PCR /5060127ZS16RTTGAGGCCAGCGTTTGTCCAGCZS16TFTCATGGGGTTTGGCCAGATAACAAGReal-time PCR6015127ZS16TRTTGAGGCCAGCGTTTGTCCAGCP86ZS7ZS7FTGCCCGCTGGCCTCAATTGACRT-PCR /5060229ZS7RGGCACGACTGATTTGGTCACCGGZS7TFTGCCCGCTGGCCTCAATTGACReal-time PCR5515229ZS7TRGGCACGACTGATTTGGTCACCGGActin Gen (BG140412)*AG-FACCTGATGAAGATCCTCACRT-PCR5660467AG-RATCCTCCAATCCAGACACActin gene (BT013524)*AGTFGCGAGAAATTGTCAGGGACGTReal-time PCR5515123AGTRTGCCCATCTGGGAGCTCAT*The GenBank accession nos. Preferred PCR primer pairs in this experiment. *The GenBank accession nos. Relative expression analysis Expression levels of ZS genes were quantified with a quantitative real-time PCR performed in an iQ5 real-time PCR detection system (Bio-Rad, USA) using the SYBR® Premix ExTaq TM Kit (Perfect Real Time, TaKaRa). The 20 µL of reaction volume contained 10ul SYBR Premix ExTaq (2×), 0.4 µL of each primer at 10 µM, 7.2 µL ddH2O and 2 µL cDNA. AGTF and AGTR primers (Table 1) were used to amplify the tomato actin gene (accession no.BT013524) as a loading control to normalize samples in separate tubes. Quantitative real-time PCR was performed in triplicate for each sample using the primers of real-time PCR (Table 1). Expression levels of the ZS genes were calculated using the 2−ΔΔCT method. Southern blotting Fifty microgramme genomic DNA, which are from LA2863, LA3864 and LA1255 was digested respectively by Eco R I and KpnI, SacI at 37°C for overnight, digested genomic DNA was separated on a 1% (w/v) agarose gel and transferred onto Hybond-N+ nylon membranes (Amersham, UK). A specificity probe of 229 bp DNA fragment which is a highly conserved region in ZS was amplified using primers of SF (5’-TGCCCGCTGGCCTCAATTGAC-3’) and SR (5’- GGCACGACTGATTTGGTCACCGG-3’), and then was labeled biotin-2-deoxyuridine-5༇-triphosphate (dUTP). Labeling of probe and hybridization were executed according to the manufacturer’s instruction of Gene Images random prime labeling module (Amersham AlkPhos) and CDP-Star detection module kit (GE Healthcare, UK). The signals were visualized by exposure to X-ray film (Kodak BioMax MS system) at room temperature (about 25 °C) for 3 h. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1