The six A. thaliana TIR1/AFB protein sequences were retrieved using the Arabidopsis Information Resource website (https://www.arabidopsis.org/). The protein sequence FASTA files of Cuiguan, Shanxi Duli, Zhongai 1, Nijisseiki, Yunhong No.1, d’Anjou, Bartlett v2.0, and Dangshansuli’ v.1.1 were used as subject sequences to run the BLAST+ command-line tool. First, local databases for the eight Pyrus proteomes were created, and then the BLASTp search was conducted against each Pyrus proteome sequence database using TIR1/AFB protein sequences as the queries. Then, the resulting BLAST hits were filtered by steps including isoforms and duplicate removal.

The candidate Pyrus TIR1/AFB sequences were searched for the F-box protein family domains using the NCBI conserved domain database (CDD) (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) (Marchler-Bauer et al., 2011) and the InterPro (https://www.ebi.ac.uk/interpro/) (Hunter et al., 2009) database to identify the final protein family sequences. Physicochemical characteristics, including molecular weight, their isoelectric point (pI), aliphatic index (AI), instability index II), and the grand average of hydropathicity (GRAVY) values were predicted by using the ExPASy ProtParam tool (https://web.expasy.org/protparam/) (Gasteiger et al., 2005). The subcellular localization of these TIR1/AFB proteins was predicted by using the WoLF PSORT tool (https://wolfpsort.hgc.jp/) (Horton et al., 2007).

Phylogenetic analysis was conducted to evaluate the intraspecific evolutionary links among Pyrus TIR1/AFBs. A multiple sequence alignment of seven Cuiguan, six Shanxi Duli, six Zhongai 1, seven Nijisseiki, six Yunhong No.1, nine d’Anjou, five Bartlett v2.0, four Dangshansuli’ v.1.1, 6 A. thaliana (Kepinski and Leyser, 2005b), 18 Brassica juncea (Cai et al., 2000), and eight Populus (Shu et al., 2015) protein sequences was performed using ClustalW (Thompson et al., 2003). A phylogenetic tree was constructed using the IQTREE Web Server (http://iqtree.cibiv.univie.ac.at/) (Trifinopoulos et al., 2016) using the maximum likelihood (ML) method performing model selection via ModelFinder and bootstrap replicates of 1000. The tree was visualized and edited using the online Interactive Tree of Life server, iTOL (https://itol.embl.de/) (Letunic and Bork, 2007).

To evaluate the conserved common motifs present in all eight Pyrus TIR1/AFB sequences, the multiple Expectation Maximization for Motif Elicitation (MEME, https://meme-suite.org/meme/) (Bailey et al., 2009) tool was used. The conserved motif number was set to 20 to search for each sequence. The gene structures were constructed by using CDS and genomic sequences through the Gene Structure Display Server (GSDS 2.0, https://gsds.gao-lab.org/) (Hu et al., 2015). The identified motifs and gene structures were visualized using TBtools (Chen et al., 2018).

The chromosomal positions for each Pyrus TIR1/AFB were retrieved from the GFF/GFF3 files and mapped to chromosomes by using the gene location visualization tool of the TBtools software (Chen et al., 2018). Depending on whether the shorter gene’s length covered 70% of the longer gene and whether the two aligned genes’ similarity was equal to or greater than 70%, TIR1/AFB gene duplication occurrences were identified (Tahir ul Qamar et al., 2023). The genes were also checked to determine whether the duplication pattern was segmental or tandem. To predict the selection pressure for the duplicated genes, Ka/Ks values were also predicted using DnaSP v.6 software (Le et al., 2011; Rozas et al., 2017). Depending on whether the Ka/Ks ratio was more than, equal to, or less than one, purifying, neutral, or positive selection was analyzed (Zia et al., 2022). Furthermore, the divergence time for the duplicated gene pairs was also calculated by using the formula t = Ks/2λ×10⁻⁶, where the λ value for dicots, 1.5 × 10⁻⁸, calculates the time of duplication in million-year units (Fatima et al., 2023a).

The STRING database (von Mering et al., 2003) was used to analyze the PPIs among the Pyrus TIR1/AFB proteins using the amino acid sequences. The top 10 interactions were set to perform the prediction, and the interactions threshold was kept at 0.4. Furthermore, the PPI network was visualized by using Cytoscape software (Shannon et al., 2003). The GO enrichment analysis components, biological processes (BPs), cellular components (CCs), and molecular functions (MFs) were predicted using the PANNZER database (pannzer2 (helsinki.fi)) (Törönen et al., 2018).

To predict the cis-regulatory elements, the 2-kb sequences upstream of the translation start site of Pyrus TIR1/AFB genes were retrieved and used to search the PlantCARE online tool (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Rombauts et al., 1999).

To gain insights into the expression pattern of Pyrus Dangshansuli TIR1/AFBs, transcriptomic RNA-seq data of different developmental stages of pear fruit (BioProject: PRJNA309745), under drought stress (BioProject: PRJNA655255), and fruit hardening disease (BioProject: PRJNA763913) were obtained from the SRA-NCBI database (https://www.ncbi.nlm.nih.gov/sra). The Dangshansuli genome and annotation (GFF) files were downloaded from the Pear Genomics Database (PGDB) (http://pyrusgdb.sdau.edu.cn/) (Chen et al., 2023). The quality of the reads was evaluated by using the FastQC tool (Wingett and Andrews, 2018). The indexes of the Pyrus genome used were built by using HISAT (Kim et al., 2019), and the high-quality paired-clean reads were then mapped onto the indexed genome. The abundance estimation of gene family members was performed by using StringTie (Pertea et al., 2016). Finally, the heatmap was generated by using the fragments per kilobase of transcript per million mapped reads (FPKM) values.

A protein requires a three-dimensional (3D) structure to perform its functions properly. The 3D structures of four Dangshansuli TIR1/AFBs (DaAFB1, DaAFB3a, DaAFB3b, and DaAFB5) were predicted by using the AlphaFold2 (https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb) (Jumper et al., 2021). The predicted 3D structures were evaluated by using the SAVES (https://saves.mbi.ucla.edu/) (Zameer et al., 2022) server and MolProbity (http://molprobity.biochem.duke.edu/) (Davis et al., 2007). Finally, the UCSF ChimeraX software (Goddard et al., 2018) was used to visualize these structures (Neupane et al., 2018; Zia et al., 2024).

A total of seven genes from the Cuiguan genome (CuTIR1/AFB), six from Shanxi Duli (ShTIR1/AFB), six from Zhongai1 (ZhTIR1/AFB), seven from Nijisseiki (NiTIR1/AFB), six from Yunhong No.1 (YuTIR1/AFB), nine from d’Anjou (AnTIR1/AFB), five from Bartlett v2.0 (BrTIR1/AFB), and four from the Dangshansuli’ v.1.1 genome (DaTIR1/AFB) were identified. All of these members were confirmed for the presence of the F-box_5 superfamily domain (Supplementary Table S3). The protein names of each member were given according to their phylogenetic relationships (Table 1).

The length of the Cuiguan TIR1/AFB proteins ranged from 584aa–629aa and MW: 63 kDa–69 kDa. The CuTIR1/AFBs were located in the nucleus, chloroplast, and cytoplasm. For Shanxi Duli, the length of ShTIR1/AFB proteins ranged from 568aa–639aa and MW: 62 kDaa–71 kDa. The ShTIR1/AFBs were located in the nucleus and chloroplast. For Zhongai 1, the length of ZhTIR1/AFB proteins ranged from 568aa–632aa, MW: 63 kDa–70 kDa. The ZhTIR1/AFBs were located in the nucleus, chloroplast, and cytoplasm. For Nijisseiki, the length of NiTIR1/AFB proteins ranged from 572aa–639aa, MW: 63 kDa–71 kDa. The NiTIR1/AFBs were located in the nucleus and chloroplast. For Yunhong No. 1, the length of YuTIR1/AFB proteins ranged from 570aa–639aa and MW: 63 kDa–71 kDa. The YuTIR1/AFBs were located in the nucleus and chloroplast. For d’Anjou, the length of AnTIR1/AFB proteins ranged from 568aa–639aa and MW: 65 kDa–70 kDa. The AnTIR1/AFBs were located in the nucleus, cytoplasm, and chloroplast. For Bartlett, the length of BrTIR1/AFB proteins ranged from 568aa–639aa and MW: 41 kDa–70 kDa. The BrTIR1/AFBs were located in the nucleus, cytoplasm, and chloroplast. For Dangshansuli, the length of DaTIR1/AFB proteins ranged from 568aa–639aa and MW: 63 kDa–71 kDa. The DaTIR1/AFBs were located in the nucleus and chloroplast.

These results show that the members did not show significant differences in aa and MW values except for some proteins that showed deviant behavior. The Pyrus TIR1/AFB proteins had pI values ranging from 5–8, indicating their acidic as well as basic behavior. The II values of all these proteins were above 40, which indicates that these are unstable in the test tube. The AI values of all the Pyrus TIR1/AFB proteins indicated that these proteins are thermally stable. Furthermore, the negative GRAVY values of all these proteins indicated that these proteins are hydrophilic (Table 1).

To analyze the evolutionary relationships among the TIR1/AFB members from eight Pyrus genomes, a phylogenetic tree was constructed by using 78 amino acid sequences from 11 species: eight Pyrus, A. thaliana, B. juncea, and Populus. All TIR1/AFB proteins were clustered into six groups: TIR1, AFB1, AFB2, AFB3, AFB4, and AFB5. TIR1 comprised the largest clade containing one AtTIR1, and the remaining members belonged to Pyrus genomes. AFB1 contained members from all eight Pyrus genomes. The AFB2 clade contained members from seven genomes but no members from the Zhongai1 genome. The AFB3 clade contained members from all seven genomes and no members from the Bartlett genome. The AFB4 clade contained members from six genomes, except the Shanxi Duli and Dangshansuli genomes. The AFB5 clade contained members from seven Pyrus genomes with no members from the Zhongai1 genome (Figure 1).

Each group is represented by a particular color, and specific colors are used for each species’ genome.

The conserved motifs and gene structures were analyzed to reveal the evolutionary pattern among Pyrus TIR1/AFBs. All members of each subfamily shared highly conserved motifs. Members of TIR1, AFB1, AFB2, and AFB3 have exactly similar motif patterns (conserved motifs 3–16). Meanwhile, only one member of the Bartlett genome, BrTIR1, had a different conservation pattern, with only motifs 3–10 being conserved. Members belonging to groups AFB4 and AFB5 showed a similar pattern as the groups mentioned earlier with some other motifs (motifs 1 and 2) (Figure 2A,C). This high conservation of motifs implies no major differences in the structure and functions of Pyrus TIR1/AFBs.

Similarly, the gene structure was also found to be highly conserved among the members of the same subfamily. Almost all members contained at least three exons and two introns. A few members of the AFB2 subfamily (AnAFB2a, AnAFB2b, and AnAFB3) contained four conserved exons with three introns (Figure 2B). This conservation of motifs and exons among the Pyrus TIR1/AFBs indicated that these genes are conserved across eight Pyrus genomes.

The chromosomal gene localization was identified to evaluate the gene distribution pattern of Pyrus TIR1/AFBs on 17 chromosomes of each Pyrus genome. This analysis showed that all genes were unevenly distributed on chromosomes. In the Cuiguan genome, seven CuTIR1/AFBs were found to be localized on seven of seventeen chromosomes (Chr3, 5, 8, 11, 15, 16, and 17). The remaining chromosomes had no CuTIR1/AFB genes present on them. In Shanxi Duli, six ShTIR1/AFBs were scattered on six of seventeen chromosomes. In the Zhongai1 genome, six ZhTIR1/AFBs were distributed on five of seventeen chromosomes. In the Nijisseiki genome, seven NiTIR1/AFBs were distributed on seven chromosomes. In the Yunhong No.1 genome, six YuTIR1/AFBs were unevenly distributed on six chromosomes. Seven AnTIR1/AFBs in d’Anjou genomes were distributed on six chromosomes. Five Bartlett v2.0 BrTIR1/AFBs were distributed on four chromosomes (Supplementary Figures S1–S7). Four DaTIR1/AFB members from the Dangshansuli’ v.1.1 genome were distributed on 3 of 17 chromosomes (Figure 3).

Gene duplication events were also analyzed among members of each Pyrus TIR1/AFBs (Table 2). The Cuiguan genome contained four duplicated pairs of genes, with all of them being segmentally duplicated. The genome of Shanxi Duli contained five segmentally duplicated pairs of ShTIR1/AFBs. The Zhongai1 genome contained four duplicated pairs of ZhTIR1/AFBs. Of these four pairs, one (ZhAFB4a/ZhAFB4b) exhibited tandem duplication, while the rest showed segmental duplication. The Nijisseiki genome contained five pairs of segmentally duplicated NiTIR1/AFBs. The Yunhong No.1 genome contained four pairs of YuTIR1/AFB genes that originated through segmental duplication. The d’Anjou genome exhibited eight pairs of duplicated genes. Of these two pairs, AnAFB2a/AnAFB3 and AnAFB2b/AnAFB3 had tandem duplication, while the rest were segmentally duplicated. The Bartlett genome contained two pairs of segmentally duplicated genes, BrTIR1/BrAFB2 and BrAFB4/BrAFB5. The Dungshanxuli genome exhibited only one segmentally duplicated gene pair, DaAFB1/DaAFB3a.

To analyze the evolutionary constraints of the repeated Pyrus TIR1/AFB genes, the Ka, Ks, and Ka/Ks ratios of all para-homologous gene pairs were also calculated. The Ka/Ks ratio ranged from 0.5 to 1.1 in the eight genomes, which shows both positive and negative selection events. The time of divergence of all 30 duplicated gene pairs of Pyrus TIR1/AFBs was between 0.7 and 90.83 million years ago (MYA), which suggests a significant period of evolutionary divergence (Table 2).

A PPI network of the Pyrus TIR1/AFBs was generated to understand the functional diversity among members. The three Dangshansuli members, DaAFB1, DaAFB3b, and DaAFB5, interacted with several other proteins. Most of the interactions were identified with 1AA and AUX proteins, which shows the potential roles of these Pyrus members in auxin regulation, thus their involvement in the growth and development of plants (Figure 4A).

GO enrichment analysis was carried out to ascertain the molecular roles of Pyrus TIR1/AFBs in a dynamic manner. Based on this GO analysis, Pyrus TIR1/AFB genes were classified into three different major categories: BPs, CCs, and MFs. Major BPs predicted included auxin-activated signaling pathways, defense responses, and other developmental and signaling pathways. These genes were found in CCs, including the nucleus, cytoplasm, and SCF ubiquitin ligase complex. The MFs associated with these genes included auxin and protein binding activity (Figure 4B; Supplementary Table S2). All these terms clearly indicate the functional involvement of Pyrus TIR1/AFBs in auxin signaling and drought stress responsiveness, thus resulting in growth.

To gain better insights into the diverse stress responses of Pyrus TIR1/AFBs, cis-regulatory elements in their promoter sequences were analyzed. In all genomes, cis elements associated with stress, such as light, hormones, and development-related responsiveness, were abundantly found. In relation to these elements, it was discovered that G-box, GT1-motif, and GATA-motif—cis elements Box 4—were implicated in the regulation of light stress. Hormone responsiveness was linked to five cis elements: P-box, TGA-element, ABRE, CGTCA-motif, and TCA-element. Furthermore, it was discovered that the GC-motif, LTR, TC-rich repeats, and MBS are the four cis elements associated with stress responsiveness. Developmental processes involved five elements: CAT-box, MBSI, circadian, HD-Zip 1, and o2-site. In Pyrus TIR1/AFBs, the presence of these elements shows their hormones, stress, and development-related responses (Figure 5; Supplementary Table S1).

Transcriptome expression data were used to determine the expression level of four DaTIR1/AFBs in tissues, including fruit, leaves, petals, sepals, ovaries, stems, and buds. DaAFB1 showed a decline in expression in ovary tissues. However, it was highly expressed in stem tissues. DaAFB5 exhibited a higher expression in leaves and bud tissues (Figure 6A).

Expression profiles of DaTIR1/AFBs in fruit hardening disease conditions were also performed. DaAFB1 was highly downregulated under disease conditions. DaAFB5 showed an increased expression. However, DaAFB3a/3b showed no significant change in expression (Figure 6B). Under drought stress, DaAFB1 showed a large decrease in expression. DaAFB3a/3b showed overexpression in drought stress, and DaAFB5 exhibited no specific change in expression (Figure 6C). These results show the differential expression pattern at various stages and biotic and abiotic stress conditions.

To gain further insights into the structural and functional diversity, the 3D structures of the four DaTIR1/AFB proteins were modeled. The four proteins showed high conservation in 3D structures. All these four proteins contained similar structures of helices and turns. All the proteins shared a similar helical structure at the C and N termini. This conservation of structure suggests the potential similar functions of DaTIR1/AFB proteins as the DaAFB1 showed similar expression results in ovary tissues, fruit hardening, and drought stress conditions. Similarly, DaAFB5 also showed conservation in expression levels in various tissues and disease conditions (Figure 7).