To screen the candidate gene sequences of AS2/LOB in physic nut, the identified AS2/LOB protein sequences of Arabidopsis were downloaded from the NCBI database. Subsequently, the Arabidopsis AS2/LOB protein sequences were subjected to homologous comparison with the protein sequences of the physic nut genome using TBtools software, with the E-values set at < 0.01. A total of 28 candidate AS2/LOB proteins of physic nut were identified. Furthermore, the hidden Markov model (HMM) file of the AS2/LOB protein domain (PF00010) was downloaded from the Pfam database (http://pfam.xfam.org/). The HMMER v3.0 software was utilized to search for the AS2/LOB protein sequence in physic nut. The conserved domains of the 28 candidate physic nut AS2/LOB protein (JcAS2/LOB) sequences were confirmed using the SMART tool (http://smart.embl-heidelberg.de). All of them encompassed the AS2/LOB domain.

The identified physic nut JcAS2/LOB protein sequence was subjected to a tBlastN similarity search against the physic nut genome database (accessible from DDBJ/EMBL/GenBank under the accession number AFEW00000000). Subsequently, the corresponding gene sequence, open reading frame sequence, mRNA sequence, and coding sequence were downloaded from the NCBI database. The online tool ProtParam (http://web.expasy.org/protparam/) provided by ExPaSy was utilized to analyze the basic parameters of the physic nut AS2/LOB protein sequence, such as the number of amino acids, theoretical molecular weight (Mw), isoelectric point (pI), and amino acid composition.

The CDS sequence of JcAS2/LOB genes was compared with the corresponding genomic sequences of these genes by employing the Spidey tool of GenBank (https://www.ncbi.nlm.nih.gov/) to determine the structure of introns and exons of these genes, and the gene structure diagram was drawn using GSDS (Gene Structure Display Server, https://gsds.gao-lab.org/) (Hu et al., 2015). The conserved motifs of the AS2/LOB gene family in physic nut were analyzed using the MEME online website (Multiple EM for Motif Elicitation) (https://meme-suite.org/meme/) with the following parameters: the site distribution was set as zero or one occurrence (of a contributing motif site) per sequence, the number of predicted motifs was set at 14, and the motif width was set within the range of 6 to 60 amino acids. The results of the conserved motifs analysis were visualized by means of Tbtools (Chen et al., 2023).

The amino acid sequences of the AS2/LOB gene family in physic nut were compared and analyzed using DNAMAN software, with all parameters set to default values. The amino acid sequences of the AS2/LOB gene family in physic nut and Arabidopsis were compared by means of ClustalW in MEGA 10 software (Kumar et al., 2008), and the phylogenetic tree was constructed via the Neighbor-Joining (NJ) method, with the bootstrap value set at 1000 and all other parameters set to default values. The chromosome location was determined based on the genetic linkage map of physic nut constructed by Wu et al., and the gene location map was drawn using MapChart (Version 2.1).

The promoter sequences of 2000 bp upstream of the start codon ATG of each JcAS2/LOB gene were acquired through BlastN similarity search of the physic nut genome based on the CDS sequence of AS2/LOB. The cis-acting elements of the JcAS2/LOB genes (JcASLs) promoter were predicted by means of the PlantCARE online website (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). The results of the promoter cis-acting element analysis were visualized using TBtools.

The protein interaction analysis of the 28 identified physic nut AS2/LOB proteins was conducted using the STRING website (https://cn.string-db.org/). Arabidopsis was selected as the model plant. It was verified that the 28 Jatropha curcas AS2/LOB protein sequences corresponded to the proteins in Arabidopsis thaliana, and the protein interaction results were inputted.

The sterilized physic nut seeds were planted in blue pots with a sand:vermiculite mass ratio of 3:1 and placed in a temperature-controlled greenhouse for growth under the following conditions: a temperature of 30°C, a relative humidity of 60%, and a 16 h light/8 h dark cycle. When the physic nut seedlings reached the 6-leaf stage, the roots, stems, leaves, and seeds 16 d and 35 d after pollination were selected for tissue-specific expression analysis. For the salt stress treatment, the 6-leaf-stage physic nut seedlings were directly watered with Hoagland solution containing 150 mM NaCl, while the control group was watered with Hoagland solution. However, for the drought stress treatment, irrigation was ceased for the 6-leaf-stage physic nut seedlings. Samples were collected at 0 h, 2 h, 2 d, and 4 d for the salt treatment and 2 d, 4 d, and 7 d for the drought stress, respectively, with three biological replicates. The leaves were harvested from the fourth leaf from the top to the bottom, immediately frozen in liquid nitrogen, and stored in a -80°C ultra-low temperature refrigerator.

The RNA from various tissues of physic nut was extracted using the Plant RNA Dual Column Kit (Polysaccharide Polyphenol) (Magen, http://magentec.com.cn/), and the extracted RNA was reverse transcribed into cDNA using the FastKing cDNA First Strand Synthesis Kit. Fluorescence quantitative PCR analysis was conducted using the LightCycler 480 quantitative PCR system and a 20 μL real-time PCR reaction system (10 μL of 2×SYBR Green Abstart PCR Mix, 2 μL of cDNA, 1 μL of each 10 μmol·L-1 forward and reverse primers, and 6 μL of ddH₂O). The quantitative PCR program was as follows: 95°C for 10 min; 95°C for 15 s, 53°C for 30 s, for 40 cycles, and fluorescence signals were collected during annealing. The relative expression of genes was calculated by employing the 2^-ΔΔCt method. All primers used in this study were listed in Supplementary Table S1 . All gene-specific primers were designed by using the Primer Premier 5.0 software package (http://www.premierbiosoft.com/primerdesign/) in accordance with the following criteria: PCR amplicon lengths ranging from 100 to 200 bp, Tm of 60 ± 1°C, and GC contents ranging from 45% to 60%, based on the sequences of the C-terminal regions of JcASL genes.

Duncan tests were used to assess the significance of differences in measured variables between the materials with the SAS software package version 9.

To identify the members of the AS2/LOB family in physic nut, 42 protein sequences of AS2/LOB from Arabidopsis were employed as query sequences for BLAST alignment against the genome of physic nut. Additionally, the HMM AS2/LOB gene model was also utilized to detect AS2/LOB family members that might have remained undetected by BLAST alignment. After eliminating sequences lacking the AS2/LOB conserved domain, a total of 28 AS2/LOB proteins were identified in physic nut. According to the location order of the 28 AS2/LOB proteins on the physic nut chromosome, these proteins were named JcASL01 to JcASL28. The physicochemical properties of all identified AS2/LOB (JcASL) proteins, encompassing the number of amino acid sequence, the length of the open reading frame length (ORF), the isoelectric point (pI), and the molecular weight (MW), were analyzed and presented in Supplementary Table S2 . The lengths of the open reading frames of the 28 JcASL genes ranged from 133 bp (JcASL8) to 349 bp (JcASL4), corresponding to the range of encoded amino acids from 402 to 1050 amino acid. In addition, the molecular weights of these proteins ranged from 15.02 kDa (JcASL8) to 39.22 kDa (JcASL4), and the theoretical isoelectric points ranged from 5.1 (JcASL23) to 9.13 (JcASL18). The isoelectric points of 17 JcASL proteins were less than 7.0, indicating that the majority of JcASL proteins were acidic proteins.

Studies have shown that AS2/LOB proteins are characterized by the presence of AS2/LOB conserved domains at the N-terminus. Hence, the conserved characteristics of the AS2/LOB domain of the deduced JcASL proteins were further analyzed using DNAMAN software, as depicted in Figure 1 . The results of our comparative analysis indicated that the CX2CX6CX3C sequence (where X represents an unconserved residue, and Arabic numerals represent the quantity of non-conserved residues), named the C-motif, was completely conserved in the conserved domains of all identified physic nut AS2/LOB proteins. Besides the C-motif, the leucine-zipper-like sequence was also highly conserved among the 28 identified JcASL proteins. More than half of the physic nut AS2/LOB proteins contained the following conserved sequences: PCAACKXLRRKCX3C within the C-motif; FAPYFX5PX2FAXVHKVF between the C-motif and the leucine-zipper-like; YGCXG and LQXQ within the leucine-zipper-like (where X represents an unconserved residue, and Arabic numerals represent the quantity of non-conserved residues). Additionally, amino acid residues K7, F23, G39, N43, K45, P73, G76, L84 (where Arabic numerals represent the positions of amino acid residues) were completely conserved within the conserved domains of the 28 identified physic nut AS2/LOB proteins.

To investigate the phylogenetic relationship between the AS2/LOB proteins of physic nut and Arabidopsis, a phylogenetic tree of 42 Arabidopsis AS2/LOB proteins and 28 physic nut AS2/LOB proteins was constructed using MEGA10.0 software based on the Neighbor-Joining method. Based on the previous grouping of Arabidopsis AS2/LOB proteins (Iwakawa et al., 2002), 70 AS2/LOB proteins from Arabidopsis and physic nut were classified into two groups, designated as Class I and Class II, and Class I was further subdivided into Class Ia and Class Ib ( Figure 2 ). The phylogenetic results demonstrated that all 28 JcASL proteins were categorized into Class I, while no JcASL proteins were found in Class II. Four JcASL proteins (JcASL2, JcASL4, JcASL11, and JcASL27) were assigned to the Class Ia group, and 24 JcASL proteins (JcASL1, JcASL3, JcASL5, JcASL6, JcASL7, JcASL8, JcASL9, JcASL10, JcASL12, JcASL13, JcASL14, JcASL15, JcASL16, JcASL17, JcASL18, JcASL19, JcASL20, JcASL21, JcASL22, JcASL23, JcASL24, JcASL25, JcASL26, and JcASL28) were allocated to the Class Ib group. Additionally, within the phylogenetic tree, some AS2/LOB proteins from Arabidopsis and physic nut formed related sister pairs, such as ASL29 and JcASL11, ASL30 and JcASL2, ASL32 and JcASL4, ASL23 and JcASL18, ASL20 and JcASL19, ASL21 and JcASL13, ASL18 and JcASL5, ASL24 and JcASL23, ASL6 and JcASL12, ASL5 and JcASL24, ASL25 and JcASL12, ASL11 and JcASL26, ASL10 and JcASL7, ASL4 and JcASL10, and AS2 and JcASL28.

In order to clarify the intron-exon structure of the JcASL genes, the coding region sequences of 28 JcASL genes and their corresponding genomic sequences were submitted to the GSDS website for gene structure analysis ( Figure 3 ). In Class Ib, eight JcASL genes (JcASL3, JcASL10, JcASL15, JcASL16, JcASL17, JcASL20, JcASL25, JcASL28) did not contain introns and were clustered on one branch, while in Class Ia, only one gene (JcASL4) did not contain introns. Among the 28 physic nut JcASL genes, 18 JcASL genes contained 1 to 2 introns. For instance, JcASL22 and JcASL26 contained two introns, and the remaining 16 JcASL genes contained one intron.

We further employed MEME online software to perform protein motif analysis on the amino acid sequences of 28 JcASL genes in physic nut. The results revealed that 14 conserved motifs were identified in the 28 deduced JcASL proteins ( Figure 4 ). Motif 1, Motif 2, and Motif 3 were present in all 28 deduced JcASL proteins. Motif 2 corresponded to the C-motif, while Motif 3 corresponded to a leucine zipper-like sequence. Motif 4, Motif 5, and Motif 6 were only existent in JcASL15, JcASL16, JcASL17, and JcASL20, whereas Motif 7 was only found in JcASL16, JcASL17, and JcASL20. Additionally, JcASL proteins in the same branch of the phylogenetic tree exhibited similar conserved motifs. For instance, motif 10 only existed in JcASL7 and JcASL26, and motif 11 only existed in JcASL9 and JcASL12. Furthermore, Motif 8, Motif 9, Motif 10, Motif 11, Motif 12, Motif 13, and Motif 14 were only detected in Class Ib but not in Class Ia.

Based on the high-density genetic linkage map of physic nut constructed by Wu et al., each JcASL gene of physic nut was located on the linkage groups (LGs) using genomic scaffold data, and the information was visualized using MapChart. The results demonstrated that the 28 JcASL genes were unevenly distributed across 9 LGs, and no JcASL genes were distributed on LGs 2 and 9 ( Figure 5 ). Among the nine LGs containing JcASL genes, LG10 had the largest number of JcASL genes, amounting to 6 (JcASL21, JcASL22, JcASL23, JcASL24, JcASL25, JcASL26); LG6 had the smallest number of JcASL genes, being 1 (JcASL13). LG6 and LG11 each contained two JcASL genes (JcASL1 and JcASL2 were located in LG6, and JcASL27 and JcASL28 were located in LG11). LG3, LG4, and LG8 all contained three JcASL genes, while LG5 and LG7 both contained four JcASL genes. Additionally, the JcASL genes were mainly distributed at the lower end of LGs, accounting for 71.43%. Tandem duplication of genes is an important way for gene family expansion and the evolution of protein functional diversity. Tandem duplicates, defined as tandem repeats that are located within 50 kb from each other or are separated by less than 4 non-homologous spacer genes (Andrew et al., 2004). Two pairs of tandem duplications (JcASL18 and JcASL19; JcASL15, JcASL16 and JcASL17) of the AS2/LOB gene family were identified on LGs 7 and 8 of physic nut.

The 2000 bp promoter sequences prior to the start codon of the JcASL genes in physic nut were selected, and the cis-acting elements of the JcASL gene promoter were analyzed using the PlantCARE online tool and subsequently visualized with TBtools. As depicted in Figure 6 and Supplementary Table S3 , the results indicated that in addition to the fundamental TATA-box and CAAT-box, the JcASL genes also encompassed numerous cis-acting elements related to plant growth and development, phytohormone responsive, and stress. The NON-box (cis-acting regulatory element related to meristem specific activation) was exclusively found in JcASL27, while the TGA-element (auxin-responsive element) and the AuxRR-core (cis-acting regulatory element involved in auxin responsiveness) were respectively identified only in JcASL1 and JcASL14. Stress and phytohormone responsive-related elements such as the WUN-motif (wound-responsive element), the ARE (cis-acting regulatory element essential for the anaerobic induction), TC-rich repeats (cis-acting element involved in defense and stress responsiveness), the MBS (MYB binding site involved in drought-inducibility) and the ABRE (cis-acting element involved in the abscisic acid responsiveness) were present in more than half of the JcASL gene promoters. JcASL18 had the fewest elements, namely TC-rich repeats, and JcASL19 had the maximum number of elements, amounting to 11. Furthermore, plant growth and development elements such as the CAT-box (cis-acting regulatory element related to meristem expression), the CCGTCC-box (cis-acting regulatory element related to meristem specific activation), the O2-site (cis-acting regulatory element involved in zein metabolism regulation), the GCN4-motif (cis-regulatory element involved in endosperm expression) and the circadian were also discovered in the promoter region of the JcASL genes. Phytohormone responsive elements such as the ABRE, the CGTCA-motif (cis-acting regulatory element involved in the MeJA-responsiveness), the TGACG-motif (cis-acting regulatory element involved in the MeJA-responsiveness), the GARE-motif (gibberellin-responsive element) and the P-box (gibberellin-responsive element) were also identified.

Employing Arabidopsis as a species model, we further predicted the AS2/LOB gene regulatory network and the interrelated genes in the protein-protein interaction network of physic nut through the STRING database, as depicted in Figure 7 . The results indicated that there were three pairs of possible interacting proteins in the physic nut species, namely JcASL9 and JcASL7, JcASL19 and JcASL23, and JcASL19 and JcASL5. However, between physic nut and Arabidopsis, 11 physic nut JcASL proteins (JcASL1, JcASL4, JcASL5, JcASL6, JcASL7, JcASL13, JcASL18, JcASL19, JcASL21, JcASL25, JcASL26) and Arabidopsis ASL37/LBD40 proteins might interact, and 5 physic nut JcASL proteins (JcASL3, JcASL11, JcASL15, JcASL27, JcASL28) and Arabidopsis ASL36/LBD42 proteins might interact. In addition, JcASL8 may interact with both ASL39/LBD37 and ASL41/LBD39, and JcASL7 may interact with ASL28/LBD26, ASL39/LBD37, ASL41/LBD39, and ASL37/LBD40, yet the possibility of this protein interacting with ASL37/LBD40 is the highest.

In order to clarify the potential function of the JcASL genes in the growth and development of physic nut, we analyzed the expression pattern of the JcASL genes in roots, stem cortex, leaves, and seeds at different developmental stages (seeds of 19 and 35 days after pollination) through the use of RNA-Seq data ( Figure 8 ; Supplementary Table S4 ). The results revealed that five JcASL genes (JcASL4, JcASL17, JcASL20, JcASL27, and JcASL28) were not expressed in all the physic nut tissues examined, whereas 10 JcASL genes (JcASL1, JcASL9, JcASL10, JcASL12, JcASL18, JcASL19, JcASL21, JcASL22, JcASL25, and JcASL26) were expressed in all the tested physic nut tissues. JcASL1, JcASL7, JcASL12, JcASL22 and JcASL26 exhibited higher expression levels in physic nut roots compared to other tissues tested. JcASL8 and JcASL13 were expressed only in seeds among the tissues of physic nut under examination. In contrast to other tissues tested, JcASL19 was more highly expressed in seeds, JcASL25 was most highly expressed in leaves, and JcASL3 and JcASL10 were more highly expressed in stem cortex.

In order to explore the expression of JcASL genes in physic nut under abiotic stress conditions, we further analyzed the expression of 28 JcASL genes in physic nut roots under salt stress for 2 hours, 2 days, and 4 days and drought stress for 2 days, 4 days, and 7 days by utilizing RNA-Seq data (Zhang et al., 2014; Zhang et al., 2015). As illustrated in Figure 9 , ten JcASL genes (JcASL2, JcASL4, JcASL6, JcASL8, JcASL11, JcASL13, JcASL17, JcASL20, JcASL27, and JcASL28) were not detectable in roots under drought and salt stress. Sixteen JcASL genes (JcASL1, JcASL3, JcASL5, JcASL7, JcASL9, JcASL10, JcASL12, JcASL14, JcASL18, JcASL19, JcASL21, JcASL22, JcASL23, JcASL24, JcASL25 and JcASL26) responded to drought or salt stress at least under one stress condition and at least at one time point. Additionally, two JcASL genes (JcASL15 and JcASL16) did not respond to drought and salt stress throughout all the treatment time points.

In order to verify the expression of the JcASL genes under normal and abiotic stress (drought and salt stress) conditions based on the RNA-Seq data, we analyzed the expression of genes in Class Ia (JcASL1, JcASL9, JcASL12, JcASL18, JcASL22 and JcASL26) in different tissues under normal growth conditions and in the roots of physic nut under drought and salt stress conditions by qRT-PCR, since none of the genes in Class II were detected as expressed in the examined tissues. The results indicated that the expression of JcASL1, JcASL9, JcASL12, JcASL18, JcASL22 and JcASL26 in the root, stem cortex, leaf, and seed at different developmental stages in physic nut as determined by qRT-PCR was consistent with the results of RNA-seq, showing that the RNA-seq data were generally accurate ( Figure 10 ). Likewise, the expression of these genes under drought and salt stress conditions detected by qRT-PCR exhibited consistent trends with the RNA-seq results ( Figure 11 ). The above results further suggested that our RNA-seq data were reliable and provided valuable reference information for the further study of the function of the JcASL genes in physic nut.