The complete genome and protein sequences of the potato DM variety were obtained from the SpudDB database (https://spuddb.uga.edu/dm_v6_1_download.shtml, accessed 4 January 2025). Candidate StTBL genes were identified using the HMM profile for the PC-Esterase (PF 13839), with protein domain downloaded from the Pfam database (https://pfam.xfam.org/, accessed 30 October 2024). Physicoche-mical properties of the encoded membrane proteins were analyzed with TBtools v2.210 (Chen et al., 2020), and subcellular localization was predicted using Cell-PLoc 2.0 (http://www.csbio.sjtu.edu.cn/bioinf/Cell-PLoc-2/, accessed 24 April 2024). Genes’ positions on chromosomes were visualized with TBtools, and members of the StTBL family were renamed according to their chromosomal order.

Protein sequences of Arabidopsis thaliana were retrieved from TAIR (https://www.arabidopsis.org/, accessed 2 June 2024) and extracted using the Fasta Extract Filter (Quick) function in Tbtools v2.210 (Bischoff et al., 2010a; Chen et al., 2020; Gao et al., 2017). Multiple sequence alignments of potato and Arabidopsis TBL proteins were performed with ClustalW using default parameters. Phylogenetic analysis of the StTBL gene family was conducted in MEGA 7.0 software (Kumar et al., 2016) using the maximum likelihood (ML) method with 1,000 bootstrap replicates and default settings. The resulting phylogenetic tree was visualized and refined with Evolview (https://www.evolgenius.info/evolview/#/treeview, accessed 6 June 2025).

Exon-intron structures of StTBL genes were analyzed by aligning cDNA sequences with their corresponding genomic DNA sequences. Conserved motifs were identified using MEME (Multiple Em for Motif Elicitation; https://meme-suite.org/meme/tools/meme, accessed 28 April 2025) (Bailey et al., 2006), with the maximum number of motifs set to 10 and other parameters left at default. Conserved domains were annotated using the NCBI CDD database (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi, accessed 28 April 2025). Gene structures were analyzed and visualized with TBtools v2.210 (Chen et al., 2020) using GTF/GFF3 files. Finally, the phylogenetic tree, conserved motifs, and gene structures were integrated using the Gene Structure View (Advanced) function in TBtools v2.210 (Chen et al., 2020).

Genome, genome annotation, and CDS sequence files of Solanum lycopersicum, A. thaliana, O. sativa, Triticum aestivum, and Zea mays were downloaded from Ensemblplants (https://plants.ensembl.org/index.html, accessed 11 May 2025). The ge-nome and annotation data of Nicotiana benthamiana were obtained from the Sol Genomics Network (https://solgenomics.net/, accessed 11 May 2025). CDS sequences of N. benthamiana were extracted from the genome and annotation files using Linux commands.

Gene duplication and collinearity analyses of the TBL gene family were performed using the Advanced Cricos plug-in in TBtools v2.210 (Chen et al., 2020). Interspecies collinearity of TBL genes of S. tuberosum with S. lycopersicum, N. benthamiana, A. thaliana, O. sativa, T. aestivum, and Z. mays was analyzed using TBtools v2.210 (Chen et al., 2020) and MCScanX (http://chibba.pgml.uga.edu/mcscan2/, accessed 11 May 2025) (Wang et al., 2012). The Ka and Ks substitution rates of duplicated gene pairs were calculated using TBtools v2.210 (E-value cut-off < 1 × 10^-10 and num of BlastHits with 5) (Chen et al., 2020; Yuan et al., 2024), and the Ka/Ks ratio was used to infer the evolutionary patterns of StTBL genes. Ka/Ks values within and between potato species were plotted using OriginPro 2024 (OriginLab Corporation, Northampton, MA, USA).

The 2000 bp upstream promoter sequences of potato StTBL coding regions were extracted using TBtools v2.210 (Chen et al., 2020). Cis-regulatory elements related to plant growth and development, hormone response, and stress response were predicted using the PlantCARE database (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed 15 April 2025). Elements lacking clear annotation or biological relevance were excluded. The distribution of cis-acting elements was visualized as heatmaps using TBtools v2.210 (Chen et al., 2020).

The Gene Ontology (GO) enrichment file (DM_1-3_516_R44_potato.v6.1. working_models.go.txt) was downloaded from SpudDB (https://spuddb.uga.edu/dm_v6_1_download.shtml, accessed 21 May 2025). GO enrichment analysis was performed using the clusterProfiler package in R, and results were ranked by enrichment factor (Yu et al., 2012).

CDS of 72 StTBLs were extracted from the SpudDB file (DM_1-3_516_R44_ potato.v6.1.working_models.cds.fa, https://spuddb.uga.edu/dm_v6_1_download.shtml, accessed 4 January 2025) using the seqkit command in Linux. Potential miRNAs targeting StTBL genes were predicted with psRNATarget (https://www.zhaolab.org/ps-RNATarget/, accessed 18 May 2025) using default parameters (The expectation is set to 5). Interaction networks between miRNAs and StTBL genes were visualized with Cytoscape 3.9.1.

The expression patterns of StTBL genes in various tissues of DM potato were comprehensively analyzed using RNA-seq data retrieved from the PGSC database (https://spuddb.uga.edu/dm_v6_1_download.shtml, accessed on 4 January 2025). Ba-sed on the publicly available transcriptome data of potato tissues, the expression status of potato TBL family genes in different tissues and organs was studied by FPKM (Fragments Per Kilobase of transcript per Million mapped reads) value analysis method. The log2 (FPKM + 1) formula is used for calculation and the heat map is drawn. Raw sequences are available in the National Center for Biotechnology Information Sequence Read Archive under BioProject PRJNA753086 (Brose et al., 2025).

Stem segments of potato (Solanum tuberosum L. cv. Atlantic) were cultured on MS solid medium (TQ-AL-PL361, Techisun, Shenzhen, China) for 35 days under controlled conditions (22 ± 2°C, light intensity 25.0-37.5 μmol m^-2 s^-1, 16 h light/8 h dark). MS solid medium containing 200 mM NaCl (4.41 g MS solid medium and 11.688 g NaCl (HDM-7647-14-5A5, TIANJINKEMAO, Tianjin, China) were added to 1L of distilled water), MS solid medium containing 200 mM mannitol (4.41 g MS basic medium and 36.4 g mannitol (CM7091-500g, Coolaber, Beijing, China) were added to 1L of distilled water) and ordinary MS solid medium (4.41 g solid medium was added to 1L of distilled water) were prepared (Garner and Blake, 1989; Zhu et al., 2020). The pH was adjusted to 5.8 and sub-packed in a test tube with a bottom diameter of 2.4 cm and a height of 18 cm. The height of the solid medium was 5 cm. High pressure sterilization 121°C 15min. The potato tissue culture seedlings with consistent growth were selected, and the roots were rinsed with sterile water, dried with filter paper, and transferred to MS test tube medium containing 200 mM mannitol or 200 mM NaCl. The potato tissue culture seedlings in the control group were transferred to ordinary MS solid medium. Drought stress was simulated using 200 mM mannitol. Nine replicate samples were set for each of the six groups (0, 12, 24, 36, 72 and 96 h). Samples were collected at 0,12,24,36,72 and 96 h after the start of stress. Three potato tissue culture seedlings with consistent growth were selected, and their leaf tissues were taken, frozen in liquid nitrogen and stored at − 80°C for further analysis. This experiment was repeated three times. Atlantic was gifted by the Inner Mongolia Potato Virus-free Seed Potato Breeding Center, China.

Total RNA was extracted from potato leaves using standard protocols. First-strand cDNA was synthesized with the All-in-One First-Strand Synthesis MasterMix (with dsDNase) for qPCR (Cat#EG15133S, BestEnzymes, Nanjing, Jiangsu, China). Primers were designed using SnapGene software (https://www.snapgene.com). qRT-PCR was performed on a QuantStudio^® 3 Real-Time PCR Instrument (96-well 0.2 ml Block; Cat#A28567, Thermo Fisher Scientific, USA) using F488 SYBR qPCR Mix (Universal) (Cat#EG23111L, BestEnzymes, Nanjing, Jiangsu, China). The housekeeping gene StTubulin (PGSC0003DMC400020469) served as the internal control. Amplification conditions were 95°C for 30 s followed by 40 cycles of 95°C for 10 s, and 60°C for 30 s. Relative expression levels of target genes were calculated using the 2^-ΔΔCt method. Ct values were obtained from four biological replicates, with two technical replicates. The average relative expression of each group of genes was used to draw the heat map of cluster analysis using the HeatMap plug-in of TBtools v2.210 (Chen et al., 2020). The standard curves of cDNA (1, 2, 4, 10,10²,10³,10⁴,10⁵ × dilutions) diluted in a series of control groups were constructed, and the amplification efficiency of each pair of primers was calculated. The formula was E (%) = [10 ^{(− 1/slope)} − 1] × 100. The average cycle threshold (Ct) of each gene was obtained by four biological repeats, and each biological repeat was composed of two technical repeats (Fan et al., 2022; Livak and Schmittgen, 2001; Radonic et al., 2004). The primer sequences and amplification efficiency of 72 StTBL genes and reference genes were shown in Supplementary Table S8. The result of qRT-PCR melting curve analysis showed that there was a single smooth curve with a TM of the target gene around 80°C (Supplementary Figure S1).

Through the BLAST plug-in of TBtools v2.210 (Chen et al., 2020), the sequences of 72 StTBL proteins and 46 AtTBL proteins were compared. Plotting using DNAMAN (version 8.0, Lynnon Biosoft, Quebec, Canada).

Based on the potato DM genome, 102 candidate StTBL proteins were initially identified using the Hidden Markov Model (HMM) profile of the PC-Esterase domain (PF13839). After domain validation with SMART and Pfam, redundant sequences (When the amino acid sequence similarity of the protein encoded by the gene is > 90% and the coverage length is > 90%, it is judged to be a redundant gene.) were removed, yielding 72 StTBL proteins (Figure 1, Supplementary Table S1). Phylogenetic analysis was conducted using 72 StTBL and 46 AtTBL protein sequences Supplementary Table S2. The neighbor-joining tree divided the proteins into three groups: Group I (31 potato, 22 Arabidopsis), Group II (29 potato, 15 Arabidopsis), and Group III (12 potato, 9 Arabidopsis). Detailed protein sequence data are provided in Supplementary Table S2.

Analysis of physicochemical properties of StTBL proteins revealed that StTBL proteins ranged from 98 (StTBL39) to 846 (StTBL30) amino acid residues, with molecular weights from 10727.21 Da (StTBL39) to 98974.08 Da (StTBL30). Theore-tical isoelectric points (pI) ranged from 5.43 (StTBL7) to 10.04 (StTBL43), with 67% of proteins being basic (pI > 7.0). Instability index varied from 20.15 (StTBL45) to 59.8 (StTBL25), with 41 proteins classified as stable (index < 40). The aliphatic index ranged from 62.14 (StTBL65) to 98.56 (StTBL7). All proteins were hydrophilic, with GRAVY values ranging from –0.742 (StTBL21) to –0.107 (StTBL67). Subcellular localization predicted 24 proteins in the plasma membrane, 8 in the cell wall, 33 in the chloroplast, 1 in the cytoplasm, 1 in the mitochondria, and 5 in the nucleus (Supplementary Table S1). Chromosomal mapping showed that the 72 StTBL genes were unevenly distributed across 12 chromosomes (Figure 2). Chr01, Chr02, and Chr07 each contained 11 genes, while Chr04 and Chr08 had only one gene each. Genes were named StTBL1–StTBL72 according to their chromosomal order (Supplementary Table S1).

Ten conserved motifs (motifs 1 − 10) were identified using the MEME program (Supplementary Figure 2). The predominant motif arrangement was motif 8 → motif 9 → motif 1 (GCD) → motif 6 → motif 4 → motif 7 → motif 5 → motif 10 → motif 3 → motif 2 (Asp-X-X-histidine, DXXH). Notably, Group III contained both the GCD and motif 4 (Figures 3A, B). The amino acid sequences of GCD, motif 4, and DXXH were DYLKWRWQPNDCELPRFBAKQFLEKQFLEKLRGKRJMFVGDSLNRNQ WZSLVCLL, WKGADVLIFNTGHWWW, and QDCSHWCLPGVPDTWNELLYAL L, respectively (Supplementary Figure 2). Conserved domain analysis showed that 57 genes carried a PC-esterase domain, and 43 of these also contained a PMR5N domain. Group I included 26 PC-esterase and 20 PMR5N domain genes; Group II comprised 26 PC-esterase and 16 PMR5N domain genes; and Group III contained 5 PC-esterase and 6 PMR5N domain genes. StTBL30 had the highest exon count, 15, whereas StTBL39 contained only one exon. Most StTBL genes carried 3 − 6 exons (Figure 3D).

In potato, eight tandemly duplicated StTBL gene pairs and 23 segmental duplication events were detected across different chromosomes (Figure 4; Supplementary Table S3). These duplications likely contributed to the expansion of the StTBL family (Figure 4). We calculated Ka/Ks ratios of duplicate gene pairs, and all values were < 1 (Figure 5). Collinearity analysis was then conducted between potato and six representative species. We identified 91 orthologous pairs with S. lycopersicum, 126 with N. benthamiana, 62 with A. thaliana, 19 with O. sativa, 31 with T. aestivum, and 8 with Z. mays (Figure 6; Supplementary Table S4). Among these, 17 StTBL genes showed collinearity with 3–8 N. benthamiana genes. Similarly, 13 genes had collinearity with three S. lycopersicum genes, while 6 genes were collinear with 3–4 A. thaliana genes. Fewer relationships were observed with monocots: one potato gene was collinear with three O. sativa genes, and several had collinearity with T. aestivum. No potato gene exhibited collinearity with multiple Z. mays genes (Supplementary Table S4). Ka/Ks values for orthologous pairs between potato and A. thaliana, S. tuberosum, and S. lycopersicum were consistently < 1, with some cases showing Ka=0 or undefined Ka/Ks values, likely due to extreme evolutionary conservation. By contrast, comparisons with monocots (O. sativa, T. aestivum, and Z. mays) yielded undefined values, likely due to low sequence homology preventing reliable estimation of substitution rates (Supplementary Table S4).

The 2000 bp upstream sequences of StTBL genes were analyzed for cis-acting elements. A total of 1,218 elements were identified, grouped into four major categories: defense and stress response (51, 4.1%), growth and development (111, 9.1%), hormone response (239, 19.6%), and light response (817, 67.0%). Light-responsive elements were the most abundant, accounting for two-thirds of all elements. Among the defense and stress response category, drought-inducible elements were the most frequent (39, 3.2%), with MBS occurring 28 times; MBS is known to regulate drought-induced gene expression (Tan et al., 2025). Hormone-responsive elements included auxin response (53, 4.4%), gibberellin response (53, 4.4%), and MeJA response (130, 10.7%), with MeJA elements being most abundant. In the growth and development category, elements associated with zein metabolism regulation (31, 3.5%), meristem expression (28, 2.3%), endosperm expression (17, 1.4%), and circadian rhythm control (16, 1.3%) were relatively enriched (Figure 7; Supplementary Table S5).

GO enrichment analysis revealed eight biological processes, two molecular fun-ctions, and two cellular components associated with StTBL genes (Figure 8; Supplementary Table S6). Enriched biological processes included xylan biosynthetic process (11), plant-type cell wall modification (6), plant-type secondary cell wall biogenesis (6), pectin biosynthetic process (5), cellulose biosynthetic process (5), circadian rhythm (6), response to freezing (3), and xyloglucan metabolic process (2). Molecular functions terms were O-acetyltransferase activity (72) and xylan O-acetyltransferase activity (14). Cellular components included the trans-Golgi network (6) and Golgi trans cisterna (3). Functionally, StTBL15/16/25/31/33/50/51/61/62/63/71 exhibited xylan O-acetyltransferase activity and were involved in xylan biosynthesis. Genes such as StTBL2/64/65 were linked to cell wall modification and secondary wall biogenesis, while StTBL2/50/62/64/65 participated in both pectin biosynthesis and cellulose biosynthesis. Cold stress-related processes included StTBL25, StTBL31, and StTBL33. Localization analysis showed that StTBL/38/40/64/65/70 were enriched in the trans-Golgi network, with StTBL38/40/70 also enriched in the Golgi trans cisterna (Supplementary Table S6).

77 putative miRNAs targeting 72 StTBLs genes were identified, while no miRNA targeted genes StTBL39/58/70 (Figure 9). Among them, 90 stu-miR395 family members targeted 9 StTBL genes, 20 stu-miR5303 family members targeted 10 StTBL genes, 19 stu-miR156 family members targeted 7 StTBL genes, 16 stu-miR172 family members targeted 9 StTBL genes, 17 stu-miR1886 family members targeted 10 StTBL genes, 16 stu-miR399 family members targeted StTBL25/32/41, 11 stu-miR482 family members targeted 10 StTBL genes (Supplementary Table S7).

To explore StTBLs gene expression across potato tissues, RNA-Seq transcriptome data were analyzed. The results showed that StTBL9/10/11/12/13/14/16/18/20/32/35/40/41/48/52/63/64/48/63/64/65/70 were expressed in all tissues, with StTBL14/20/40/48/63/64/65 showing consistently high expression. Specifically, StTBL14 was the most highly expressed in shoots, leaves, sepals, carpels, petals, and mature fruit (FPKM > 5); StTBL20 in tubers, stolons, petals, and mature fruit (FPKM > 5); and StTBL64 and StTBL66 in mature fruit (FPKM > 6). By contrast, StTBL3/6/8/43/53 were either undetected or expressed at very low levels (FPKM < 1) (Figure 10; Supplementary Table S8). The clustering results revealed multiple pairs of StTBL genes sharing identical expression patterns. These included: StTBL31 and StTBL41; StTBL35 and StTBL65; StTBL10 and StTBL48; StTBL19 and StTBL66; StTBL27 and StTBL64; StTBL11 and StTBL29; StTBL12 and StTBL32; StTBL7 and StTBL58; StTBL44 and StTBL45; StTBL18 and StTBL34; StTBL1 and StTBL28; StTBL30 and StTBL42; StTBL24 and StTBL57; StTBL4 and StTBL69; StTBL46 and StTBL62; StTBL36 and StTBL53; and StTBL51 and StTBL72. Additionally, StTBL33, StTBL25, and StTBL50 also exhibited the same expression pattern.

To evaluate the role of StTBL gene in abiotic stress response, qRT-PCR was used to analyze the gene expression of StTBL gene in leaves under salt stress and drought stress. Under NaCl stress, 57 StTBL genes showed significant expression changes rela-tive to the control, while 15 genes remained unchanged. StTBL1/3/16/20/22/28/58/59/60/68 were consistently induced, and StTBL31/42/43/56 were induced during the early stage. StTBL15/22/28/41 reached peak expression at 36 h, whereas StTBL1/3/7/13/19/44/60/69 peaked at 72 h. StTBL68 was only induced at the late stage. These genes may function as positive regulators of salt stress responses. In contrast, StTBL10/11/25/33/46/50/51/62/66/70/72 was down-regulated (Figure 11; Supplementary Table 10). Overall, these findings indicate that many TBL genes participate in salt stress response, with individual members exhibiting distinct regulatory patterns.

Under drought stress, StTBL1/2/3/12/19/21/22/28/31/33/69 were significantly un-regulated, while StTBL4/13/14/50/70 showed no significant change. StTBL7/28/34 was induced early, and StTBL43/44 were induced late. StTBL21 exhibited the highest ex-pression at 36 h. In contrast, 43 genes (StTBL5/9/10/11/15/16/18/20/24/26/29/32/35/36/37/38/39/40/41/45/46/47/48/49/51/52/53/54/55/56/57/58/59/61/62/63/64/65/66/67/68/71/72) were down-regulated, showing a strong synergistic repression under drought stress (Figure 11). This suggests that many family members may contribute to stress sensitivity or to the function of metabolic or developmental pathways influenced by osmotic stress. Across both treatments, StTBL6/8/30 showed no detectable expression, probably due to extremely low transcript abundance in leaves (0, 0, and 0.14, respec-tively; Supplementary Table 8). Cluster analysis revealed several groups with similar expression dynamics, including StTBL1/3/59, StTBL16/20/58, StTBL19/22, StTBL28/33, StTBL60/21/31, and StTBL2/12/69. StTBL68 did not cluster with other stress-responsive genes.

Protein sequence alignment revealed seven gene pairs with > 70% homology. Among them, AtTBL33–StTBL34 showed 78.261% identity (Supplementary Figure 3), followed by AtTBL29 and StTBL25 at 72.016% (Supplementary Figure 4), AtTBL33 and StTBL26 at 71.649% (Supplementary Figure 5), AtTBL28 and StTBL31 at 71.591% (Supplementary Figure 6), and AtTBL13 and StTBL11 at 71.023% (Supplementary Figure 7).