Genome data of R570 BAC and R570 STP were obtained from the sugarcane genome hub (https://sugarcane-genome.cirad.fr/; Garsmeur et al., 2018). Genome data of A. thaliana TAIR10, Oryza sativa (IRGSP-1.0), S. bicolor (NCBIv3), and S. spontaneum AP85-441 were obtained from the Ensembl Plants database (http://plants.ensembl.org/index.html; Yates et al., 2022).

The hidden Markov model (HMM) file of the PRX domain (PF00141) was downloaded from the Pfam database (https://pfam-legacy.xfam.org/; Sonnhammer et al., 1998; Blom et al., 1999; Mistry et al., 2021). The HMMER software (version 3.1b1; http://www.hmmer.org/) was used to search against the whole genome protein file of R570, A. thaliana, O. sativa, S. bicolor, and S. spontaneum under the condition of e-value < 1×10⁻²⁰. Multiple sequence alignment was performed using ClustalW (http://www.clustal.org/), and a new HMM matrix file was constructed. The PRX domain-containing proteins were searched on the new HMM matrix file and screened by e-value < 0.001. The PRX protein Ref-seq of all plants was downloaded to build a library from the NCBI database (https://www.ncbi.nlm.nih.gov/). The putative PRX genes were searched using BLASTP and screened by e-value < 1×10⁻¹⁰ and identity > 75%. The protein-conserved domains were verified by Pfam and NCBI CDD (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi), and the genes without the PRX domain were deleted (Lu et al., 2020; Mistry et al., 2021). In addition, the PRX gene family was identified based on the sugarcane transcriptome data.

The physicochemical properties of the ShPRX family proteins were calculated using ExPASy-ProtParam (https://web.expasy.org/protparam/; Wilkins et al., 1999). The subcellular localization of the ShPRX family proteins was predicted using Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/; Chou and Shen, 2010). The transmembrane domain of the ShPRX family proteins was predicted using TMHMM 2.0 (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0; Sonnhammer et al., 1998). The signal peptide of the ShPRX family proteins was predicted using SignalP 6.0 (https://services.healthtech.dtu.dk/service.php?SignalP; Teufel et al., 2022). The phosphorylation sites of the ShPRX family proteins were predicted using NetPhos 3.1 (https://services.healthtech.dtu.dk/service.php?NetPhos-3.1; Blom et al., 1999). The secondary structure of the ShPRX family proteins was analyzed using SOPMA (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_sopma.html; Geourjon and Deleage, 1995).

The multiple protein sequences of the PRX gene family in R570 STP, A. thaliana, O. sativa, S. bicolor, and S. spontaneum were compared using the MEGA-X MUSCLE software (https://www.megasoftware.net/; Kumar et al., 2018). The phylogenetic tree was constructed using the neighbor-joining method with the Jones–Taylor–Thornton model, 1,000 bootstrap replications, gamma distributed (G), and partial deletion gaps by MEGA-X. The phylogenetic tree was visualized using the online tool iTOL (https://itol.embl.de/; Letunic and Bork, 2021).

The positional and structural information of ShPRX genes on the R570 STP chromosomes was extracted from the Generic Feature Format Version 3 (GFF3) (Garsmeur et al., 2018). The chromosomal locations of the ShPRX family genes were drawn using the online tool MG2C v2.1 (http://mg2c.iask.in/mg2c_v2.1/index.html). The conserved motifs of the ShPRX family proteins were identified using the online MEME with optimized parameters: minimum width, 6; maximum width, 50; and the number of motifs, 26 (Bailey and Elkan, 1994). The phylogenetic tree, gene structure, and conserved motifs of the ShPRX family genes were visualized using TBtools (Chen et al., 2020).

The sequence 2000 bp upstream from the start codon of the ShPRX family gene was obtained using the hybrid cultivar R570 reference genome (Garsmeur et al., 2018). The cis-acting elements of ShPRX family genes in promotor regions were predicted using the PlantCAR (Lescot et al., 2002). The phylogenetic tree and cis-acting elements of the ShPRX family genes were visualized using GSDS 2.0 (http://gsds.cbi.pku.edu.cn; Hu et al., 2015). The codon usage bias of the ShPRX family genes was analyzed using CodonW (https://codonw.sourceforge.net/) and EMBOSS: chips (https://www.bioinformatics.nl/cgi-bin/emboss/chips; Rice et al., 2000). The effective number of codon (ENc) plots and Parity rule 2 (PR2) plot analysis was performed as described by Wright (1990) and Chakraborty et al. (2020).

The syntenic relationships of the PRX family genes were analyzed and visualized using MCScanX and Circos, respectively (Krzywinski et al., 2009; Wang et al., 2012). The frequencies of synonymous (Ks) and nonsynonymous (Ka) mutations, along with their ratios, were calculated to analyze the selection pressure of PRX duplicated gene pairs using TBtools (Chen et al., 2020). The divergence time (T) of class III PRX family genes was calculated as T = Ks/(2 × λ) × 10⁻⁶ Mya (λ = 6.5 × 10⁻⁹ for grasses; (Gaut et al., 1996). The divergence time values were estimated using TimeTree (http://www.timetree.org/; Kumar et al., 2022).

The RNA-seq expression data of the growth and development of S. spontaneum were downloaded from the Saccharum Genome Database (SGD; http://sugarcane.zhangjisenlab.cn/sgd/html/index.html; Zhang et al., 2018). The transcriptome data of sugarcane were obtained by sequencing (Akbar et al., 2021). The gene expression data of rice under Cd (GSE35502) and salt (GSE60287) stresses were downloaded from the GEO database (https://www.ncbi.nlm.nih.gov/geo/; Ishimaru et al., 2012; Garg et al., 2015; Clough and Barrett, 2016; Shankar et al., 2016). The expression pattern of the PRX family genes was drawn based on FPKM values or log₂ fold change (log₂FC) using the R software (v4.0.5).

Badila and B48, grown to leaf stages 6–8 in the field, was inoculated using SCMV crude extract following Yao’s method (Yao et al., 2017), and the −3 and +1 leaves were collected at day 21 post-inoculation. The control plants were rubbed using 0.1 M phosphate buffer (pH 7.0). Sugarcane plants of Zhongzhe 1 were watered using 4.3-mM Cd solution or 0.5-L 1.0% sodium chloride solution when Zhongzhe 1 grew to leaf stage 3 in barrels containing 16 kilograms of soil. The control plants were watered using 0.5-L double-distilled water. The +1 leaves of sugarcane were collected at different time after treatment (0, 4, 8, 12, and 24 h).

Total RNA was extracted using Eastep^® Super Total RNA Extraction Kit (Promega (Beijing) Biotech Co. Ltd., Beijing, China) following the manufacturer’s instructions. The first strand of cDNA was synthesized using PrimeScript™ II 1^st Strand cDNA Synthesis Kit (Takara Biomedical Technology Co. Ltd., Beijing, China). The qRT-PCR experiment was conducted in three independent replicates using ChamQ Universal SYBR qPCR Master Mix (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China) with the Bio-Rad CFX96 fluorescence quantitative PCR Instrument. The primers for qRT-PCR of 10 PRX genes were designed using Primer Premier 6. The relative expression levels of the PRX family genes were calculated using the 2^−ΔΔCt method, and statistical significance was analyzed using ordinary one-way ANOVA in GraphPad Prism 7.

Eighty-two PRX proteins with a conserved PRX domain were characterized as members of the class III PRX gene family in R570 STP and named ShPRX1–ShPRX82 based on their respective locations on the chromosomes (Tables S1 , S2). The total number of PRX family genes in the sugarcane R570 cultivar was lower than that in the S. spontaneum (113), sorghum (150), and rice (126), but slightly higher in A. thaliana (80) (Table S2).

The physical and chemical properties analysis revealed that the ShPRX family genes were predicted to encode polypeptides from 235 to 472 amino acids, with predicted molecular weights ranging from 25.11 to 49.76 kD. The theoretical pI ranged from 4.59 to 10.08, and the grand average of the hydropathicity values of 52 ShPRX proteins was negative, ranging from −0.51 to −0.002, indicating a hydrophilic characteristic. The grand average of the hydropathicity values of 30 ShPRX proteins was positive, ranging from 0.004 to 0.297, indicating a hydrophobic characteristic. The predicted number of negatively charged residues (Asp + Glu) in the ShPRXs was 14–59, and the number of positively charged residues (Arg + Lys) was 20–54. The instability index of 46 ShPRX proteins was less than 40, ranging from 25.38 to 39.76, indicating a stable characteristic. The instability index of 36 ShPRX proteins was greater than 40, ranging from 40.36 to 56.05, indicating an unstable characteristic (Table S1).

The subcellular localization prediction analysis showed that most ShPRX genes were located in the cytoplasm, and some were located in the vacuole, chloroplast, peroxisome, mitochondrion, cell membrane, and nucleus. Most ShPRX genes had signal peptides, 42 ShPRX genes had no transmembrane domain, and other ShPRX genes had only one transmembrane domain. The number of phosphorylation sites ranged from 16 to 47, while four ShPRX genes had no tyrosine phosphorylation site. The secondary structure prediction analysis showed that ShPRX proteins were mainly composed of alpha helix and random coil (Table S1).

The 82 ShPRXs were unevenly mapped onto the eight chromosomes of R570 STP (see Figure 1) based on the annotation information of the R570 STP genome. The ShPRX genes distributed in Sh01 (16), Sh02 (12), Sh03 (15), Sh04 (11), Sh06 (7), and Sh09 (10), while chromosomes Sh07 and Sh08 had only 4 and 3 ShPRX genes, respectively. Four ShPRX genes were distributed on scaffolds (Sh_011C11, Sh_025L09, and Sh_232D06).

The phylogenetic tree from five plants was constructed on PRX amino acid to clarify the evolutionary relationship of the PRX gene family of sugarcane, as shown in Figure 2. The ShPRX family genes were divided into six groups. The phylogenetic tree based on the amino acid sequences of ShPRX indicated that the ShPRX family genes in groups 1, 2, 3, 4, 5, and 6 from five plants were clustered into groups I, II, VI, V, III and IV from sugarcane, besides ShPRX1 (group 3), ShPRX34 (group 3), ShPRX22 (group 4), and ShPRX62 (group 2). ShPRX1 and ShPRX34 were divided into group V, and ShPRX22 and ShPRX62 were divided into group IV (Figure 2; Figure S1). Group 1 comprised the lowest number of PRX genes. The PRX family genes of sugarcane shared high homology with the PRX family genes in sorghum and rice (Figure 2). The ShPRX family genes in other groups shared high homology to S. spontaneum or sorghum (Figure 2).

The conservative structure of the PRX gene family was deciphered through the evolutionary relationship, motif, and structure of the PRX family genes in sugarcane (Figure 3). MEME was used to analyze the motif distribution within the ShPRX gene family, and 26 motifs were identified (p < 0.05) (Table S3). The ShPRX genes in the same group had the same motifs: the genes in Group I had motifs 3, 8, 18, and 19; Group II had motifs 3, 4, 5, 8, 10, and 12; Group III had motifs 3, 8, and 11; Group IV had motifs 2, 4, 5, 6, and 7; group V had motifs 1, 4, 7, and 8; and Group VI had motif 3, and most ShPRX genes have top 14 motifs (Table S4; Figures 3A, B). The ShPRX family genes had motif 3 except for ShPRX10 and ShPRX43, and most genes had motif 8 except for ShPRX43 and ShPRX44 (Figure 3B).

According to the structural analysis of the PRX family genes, the length of these genes ranged from 723 to 7812 bp, where ShPRX67 had the most considerable length and 11 coding regions, ShPRX47 had the most petite length (Table S1; Figure 3C, S2). The respective structures of these genes were found to be similar in the same group, and 10 PRX genes had no introns. Most PRX genes contained 1–3 introns, but ShPRX57, ShPRX64, and ShPRX67 in Group I contained 13, 8, and 10 introns, respectively, and ShPRX20 (Group IV), ShPRX39 (Group VI), and ShPRX69 (Group VI) had long introns (Figures 3A, C).

Cis-acting elements can participate in gene expression and regulation, and members of the ShPRX gene family play a key role in biotic and abiotic stress. The function of the ShPRX family genes can be predicted by analyzing the cis-acting elements in the 2-kb upstream region of the ShPRX family genes. Thirty-one cis-acting elements were identified and involved in hormones, abiotic stress, tissue-specific cell cycle, and circadian control by analyzing and selecting cis-acting elements of the ShPRX family genes (Table S5). The differences were detected in the variety and number of cis-acting elements across the ShPRX family genes (Table S5). ShPRX15, ShPRX41, ShPRX48, and ShPRX62 had the maximum variety of cis-acting elements (15), but the number of cis-acting elements in ShPRX52 was the greatest (37). ShPRX22, ShPRX38, ShPRX49, and ShPRX62 contained plant hormone-responsive elements, including IAA-, GA-, ABA-, SA-, and MeJA-responsive elements. We inferred that plant hormones might regulate these ShPRX family genes. ShPRX31 and ShPRX66 did not contain ABA and light-responsive elements but contained seed-specific regulation elements and MYB binding sites involved in drought inducibility (Table S5). The cis-acting regulatory elements of most ShPRX family genes were involved in ABA, MeJA, light responsiveness, anaerobic induction, and drought inducibility (Figure 4A).

The cis-acting elements in Groups III, IV, V, and VI, were involved in the meristem-specific activation and expression. Over half of those in the Group I, II, and V were involved in the defense and stress responsiveness (Figures 4B, C). The cis-acting regulatory elements in Group 1 involved in the IAA, ABA, SA, MeJA, light, low-temperature, and anaerobic induction, and group IV in the seed-specific regulation (Table S6). These results suggested that the ShPRX family genes might participate in the response of sugarcane to biotic and abiotic stresses and tissue-specific responses.

The codon usage bias and base composition analysis of coding sequences of the PRX family genes were calculated (Table S7, S8). The mean values of the codon composition at the third position from high to low were C3s (0.589), G3s (0.440), T3s (0.093), and A3s (0.080), and the mean content of the GC (65.1%) was also higher than AT (34.9%), suggesting a GC-rich composition of coding sequences of the PRX family genes (Table S7). The ENC values of coding sequences ranged from 28.25 to 58.62, with a mean of 38.059 (ENC < 40), and most ENC values of coding sequences were below 40, indicating a solid codon usage bias (Table S7). The RSCU revealed that 22 codons of 28 high-frequency codons (mean RSCU value > 1.0) were over-represented (mean RSCU value > 1.6). In comparison, 31 codons of 34 low-frequency codons (mean RSCU value < 1.0) were under-represented (mean RSCU value < 0.6) (Figure 4D; Table S8). The ENc plot (Figure S3A) and PR2 plot (Figure S3B) analyses showed that the codon use of the class III PRX family genes in sugarcane was affected by mutation and selection pressure.

The syntenic relationships of the ShPRXs were analyzed to explore the genomic expansion of the PRX gene family in sugarcane. In total, 32 of the 82 ShPRXs had syntenic relationships, and 10 ones in five syntenic pairs underwent segmental duplication, while 22 genes in 13 syntenic pairs underwent tandem duplication (Figure 5A; Table S9). The Ka/Ks ratios of 17 of the 18 syntenic pairs were < 1, which might have undergone purifying selection, indicating that the evolution of these pairs was slow (Figure 5A; Table S9).

The syntenic relationships of the PRX family genes from S officinarum, A. thaliana, O. sativa, S. bicolor, and S. spontaneum were analyzed to better understand duplication and evolution of PRX gene. The results revealed that 52, 68, and 37 syntenic gene pairs of PRX genes were detected in S officinarum versus O. sativa, S officinarum versus S. bicolor, and S officinarum versus S. spontaneum, respectively (Figures 5B, C; Table S9). The Ka/Ks ratio of 1 of the 157 syntenic gene pairs was> 1, indicating that the syntenic gene pair had been strongly positively selected during evolutionary history (Table S9). The Ka/Ks ratios of 153 from the 157 syntenic gene pairs were < 1, which might have undergone purifying selection, indicating that the evolution of these pairs was slow. The duplication events of homologous collinearity gene pairs of the class III PRX occurred 4.094–85.36 Mya for sugarcane and rice, 21.992–83.6 Mya for sugarcane and sorghum, and 3.568–50.095 Mya for sugarcane and its wild relative of S. spontaneum. The class III PRX gene family has been identified in at least 29 plants in the whole genome. The duplication events occurred 159.9 Mya for 12 plants in monocotyledon from 21 plants in dicotyledon. Moreover, 104.7 Mya in Poaceae diverged from Bromeliaceae (Figure S3C; Table S10).

Tissue-specific expression patterns of the SsPRX genes were analyzed based on the transcriptome data of S. spontaneum to explore the functions of the PRX gene family in sugarcane further. The 113 SsPRX genes were expressed as FPKM values in leaf roll, leaf, the third, sixth, and ninth stem nodes of sugarcane at the seedling, elongation and maturity stage (Figure 6A; Table S11). Four SsPRX genes (SsPRX42, SsPRX62, SsPRX79, and SsPRX83) were highly expressed in all tissues, and 69 SsPRX genes had a low expression or no expression, and the rest of the highly expressed SsPRX genes had tissue specificity.

Sixteen PRX proteins with a conserved PRX domain were characterized as members of the sugarcane class III PRX gene family based on the transcriptome data of Badila and B48 in response to SCMV infection. We named them ShtPRX1–ShtPRX16 based on their Unigene IDs (Table S2). The expression patterns of the ShtPRX genes were analyzed on transcriptome data (Table S12). The ShtPRX genes were divided into two groups according to the clustering of expression patterns. Genes in Group I was primarily highly expressed in all samples. Genes in Group II, except for ShtPRX13 and ShtPRX16, were primarily highly expressed in some samples (Figure 6B). ShtPRX8 was only one co-differentially expressed gene (|log₂FC| ≥ 3, FDR < 0.05) in five groups (Figure 6C), indicating that ShtPRX8 might participate in SCMV stress in sugarcane.

Homologous class III PRX genes of R570 were identified using class III PRX family protein sequences of rice as a library, and screened bu BLASTP at the e-value less than 1e^-5. (Table S13). The expression profiles of rice under Cd and salt stresses were analyzed to infer the function of PRX homologous genes in sugarcane (Tables S14, S15; Figure S4). In groups #I, III, and VII, except for OsPRX10, the |log₂FC| values for (Cd/CK)-WR or (Cd/CK)-WS was more than 3, and homologous genes, including ShPRX38, ShBAC.PRX23, ShBAC.PRX41, ShBAC.PRX42, ShBAC.PRX43, ShBAC.PRX51, ShBAC.PRX52, and ShBAC.PRX75 might participate in Cd stress in sugarcane (Figure S4A; Table S13). In group 2, the |log₂FC| of OsPRX56 for PK-SS/PK-CK was more than 2, and homologous genes of ShBAC.PRX20 might participate in salt stress in Sugarcane (Figure S4B; Tables S13, S14).

qRT-PCR analysis was performed to analyze the responses of the PRX family genes in sugarcane exposed to SCMV, Cd, and salt stresses. The primer sequences were listed in Table S16. The expression levels of ShtPRX8 showed a significant decrease in Badila after SCMV infection. After applying 4.3 mM Cd²⁺ stress to sugarcane, the expression levels of ShBAC.PRX23, ShBAC.PRX41, ShBAC.PRX42, ShBAC.PRX43, ShBAC.PRX51, ShBAC.PRX52, ShBAC.PRX75 and ShPRX38 showed a significant increase from 0 to 24 h (Figure 7). After applying 1% sodium chloride solution stress to sugarcane, the expression levels of ShBAC PRX20 significantly increased from 0 to 24 h (Figure 7).