The character analysis of small RNA (sRNA) in the sterile line 05ms and fertile line S63 of eggplant. (A) The length distribution of sRNAs. (B) The first nucleotide bias analysis of known sRNAs. (C) The microRNA (miRNA) families analysis of known miRNAs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). sRNA frequency represents the percentage of sRNA among S63-summer, S63-spring, 05ms-summer, and 05ms-spring under different length. First nucleotide represents the percentage of A/U/C/G in the first base of sRNA of this length. The numbers above each bar represent the total numbers of sRNA of this length. Different miRNA families were detected in the Solanaceae crops. Family numbers represent the numbers of different miRNA families.
Analysis of differential expression levels and secondary structures of microRNAs (miRNAs) in the sterile line 05ms and fertile line S63 of eggplant. (A) Numbers of differentially expressed miRNAs (DEMs) in different comparisons. (B) Venn diagram showing the overlaps in different comparisons. (C) Heatmap showing the hierarchical cluster analysis results from samples and DEMs. (D) Secondary structures of DEMs. DEMs represents the numbers of DEMs. “Upregulated” indicates upregulated DEMs. Downregulated indicates downregulated DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). The red star indicates the key DEMs. Each column indicates sample, and the colored bar indicates the relative expression level from high (red) to low (blue). Red highlights indicate the locations of mature sequence of miRNA.
Comparison of results obtained via small RNA sequencing (sRNA-seq) and quantitative real-time PCR (qRT-PCR) for the six differentially expressed microRNAs (DEMs) at different periods in the sterile line 05ms and fertile line S63 of eggplant. miR168b-3p, miR397–5p, and miR408, Novel_116, Novel_119, and Novel_97 were the six DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). Error bars indicate the SE of three biological replicates.
Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enrichment analysis of the 892 target genes in the sterile line 05ms and fertile line S63 of eggplant. (A) The top 20 significantly enriched GO terms. (B) The top 20 significantly enriched KEGG pathways.
Regulatory networks comprising Gene Ontology terms, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent GO terms, DEMs, and target genes, respectively.
Regulatory network comprising Kyoto Encyclopedia of Genes and Genomes pathways, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent KEGG pathways, DEMs, and their target genes, respectively.
The contents comparison of abscisic acid (ABA), hydrogen peroxide (H2O2), and catalase (CAT) between 05ms-spring and 05ms-summer in the sterile line 05ms. (A) ABA contents in 05ms-spring and 05ms-summer. (B) H2O2 contents in 05ms-spring and 05ms-summer. (C) CAT contents in 05ms-spring and 05ms-summer (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms). Error bars indicate the standard error of three biological replicates. Statistical significance was analyzed by a t test to compare 05ms-summer to 05ms-spring for each category in all statistical. **Significant at P < 0.01.
Morphological comparison of flowers in the sterile line 05ms and fertile line S63 of eggplant. (A) The phenotype of flower in 05ms. (B) The phenotype of flower in S63.
Quantitative real-time polymerase chain reaction confirmation of target genes at different periods in the sterile line 05ms and fertile line S63 of eggplant: 05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63. Sme2.5_00016.1_g00020.1, Sme2.5_01800.1_g00006.1, Sme2.5_05438.1_g00004.1, and Sme2.5_02906.1_g00004.1 are the four target genes. Different letters above the bars represent significant differences at P < 0.05 according to Duncan’s multiple range test.
Click on author name to view affiliation information
Thermosensitive genic male sterile (TGMS) lines are the core of two-line hybrid systems. MicroRNAs (miRNAs) play critical roles in plant growth and development. However, knowledge of regulation of anther development by miRNAs in TGMS eggplant (Solanum melongena) is largely unexplored. To investigate the mechanism underlying miRNA regulation of male sterility, we employed high-throughput small RNA sequencing in anther samples from the reverse TGMS line 05ms and the temperature-insensitive line S63 in eggplant, under high temperature and low temperature conditions. The 05ms line is sterile at low temperature and fertile at high temperature. A total of 166,273,427 raw reads were obtained, 143 known miRNAs from 42 miRNA families and 104 novel miRNAs were detected. Further, six differentially expressed miRNAs (DEMs) were identified, including three known (miR168b-3p, miR397–5p, and miR408) and three novel miRNAs (Novel_116, Novel_119, and Novel_97), which might be related to anther development. Moreover, the six DEMs were validated by quantitative real-time polymerase chain reaction and 892 target genes of which were predicted. Gene Ontology analysis of target genes revealed significant enrichment in the “copper ion binding,” “oxidation-reduction process,” and “oxidoreductase activity” terms. Kyoto Encyclopedia of Genes and Genomes analysis revealed that “plant hormone signal transduction” and “other glycan degradation” were enriched. In addition, we constructed regulatory networks comprising miRNAs, target genes, and important terms/pathways and found the miR397-5p was the most linked miRNA, down-regulated under low temperature. Our findings contribute to understanding of the roles of miRNA during anther development and provide the theoretical foundation for two-line hybrid breeding of eggplant.
Eggplant (Solanum melongena) is an important vegetable globally. With rising labor costs, it is estimated that the utilization of eggplant with male sterility will play a greater role in their cultivation. Hybrid seed production using male sterile lines mainly involves the three- and two-line hybrid systems. The three-line hybrid system requires male sterile, restorer, and maintainer lines, whereas the two-line system involves an environmentally sensitive genic male sterile (EGMS) line, such as a photoperiod-sensitive or thermosensitive genic male sterile (PGMS or TGMS) line, which is used to produce hybrid seeds, thus eliminating the need for maintainer and restorer lines (Sun et al., 2021). Hence, the two-line system is more convenient for hybrid seed production (Zhou et al., 2016). The EGMS lines have been reported in some vegetable crops, including tomato [Solanum lycopersicum (Sheoran et al., 2009)] and eggplant (Li et al., 2019).
miRNAs are a class of small endogenous non-coding RNAs. Since the first report of miRNAs in rabidopsis [Arabidopsis thaliana (Llave et al., 2002)], there has been considerable research aimed at understanding their molecule mechanism (Voinnet, 2009). The miRNAs comprise ≈21 to 24 nucleotides and function as negative regulators of gene expression by degrading mRNA or inhibiting translation (Betti et al., 2021). Extensive research has demonstrated critical roles for miRNAs in regulation of various biological processes (Feng et al., 2019; Shi et al., 2019), including organ development and hormone signaling, as well as responses to abiotic stresses (Lin et al., 2020). Plants are more vulnerable to environmental stresses, such as low temperature or high temperature at the reproductive growth stage than during the vegetative growth phase (Tang et al., 2012). Numerous studies have identified the roles of miRNAs in regulating plant male sterility primarily by inhibiting target genes that affect anther development (Dong et al., 2020) and regulate fertility (Jiang et al., 2021). The male sterility in eggplant was induced by suppressing TATA box-binding protein associated factors using artificial miRNA technology (Toppino et al., 2011). Thirteen pairs of miRNA/target genes that regulate male sterility in PGMS/TGMS rice (Oryza sativa) by responding to temperature change were identified by miRNA, transcriptome, and degradome sequencing, and miR156, miR5488, and miR399 were found to affect male sterility in PA64S by influencing anther wall lignin synthesis and the flavonoid metabolism pathway (Sun et al., 2021).
Reactive oxygen species (ROS) and phytohormones are two important signaling systems involved in plant responses to abiotic stress, and drive changes that guide plant adaptation and survival (Devireddy et al., 2021). Cellular reduction/oxidation (redox) homeostasis plays an important role in balancing temperature tolerance and plant development (Concetta et al., 2015). High temperature during the anthesis stage causes a significant rise in ROS content accumulation in anthers, triggering male sterility due to differential antioxidant enzyme activity-mediated anther damage (Dwivedi et al., 2019). Phytohormones also play essential roles during anther development processes (Zhang et al., 2021). Research into a wheat (Triticum aestivum) TGMS line indicated that six miRNAs and one transacting small interfering RNA regulated the auxin signaling network, which are linked with male sterility during cold stress (Tang et al., 2012). Further, arabidopsis transgenic phenotypes of auxin response factor 6 (arf6) plants mutated miR167 target sites suggest that miR167 may repress ARF6 translation, preventing proper pollen release from anthers (Zheng et al., 2019). Abscisic acid (ABA) is a key phytohormone that regulates plant development and responses to biotic and abiotic stresses (Yoshida et al., 2015). MYB33 is a major target of miR159, which promotes ABA Insensitive 5 (ABI5) transcription through directly binding to its promoter. The ABI5 promotes vegetative phase development in arabidopsis by affecting expression of the miR156-SPL pathway (Guo et al., 2021). MYB33 and SPL were genes that regulate early development of tapetum in rice (Huang et al., 2011). miR397a was up-regulated in ABA-deficient mutant rice (Tian et al., 2015). Moreover, miRNAs and their target genes have vital functions in male fertility, lignin formation, and adaptation to abiotic stress in rice (Tian et al., 2015). miR408 and miR397 were both negatively regulated by copper, an essential mineral required for plant development, and their target genes encoded copper-containing proteins, including laccase (LAC) in arabidopsis (Abdel-Ghany and Pilon, 2008).
Eggplant is a thermophilic plant; hence, low temperature is an important abiotic stress that influences its growth and development (Yang et al., 2017). The reverse TGMS (rTGMS) eggplant line 05ms is sterile at low temperature and fertile at high temperature (Li et al., 2019). Abundant evidence supports key roles for miRNAs in the regulation of anther development in many crops. However, there have been no reports regarding miRNA expression in eggplant anther. To address this, we performed small RNA sequencing (sRNA-seq) of samples from anthers (meiosis stage) of the 05ms and S63 eggplant lines cultivated under low temperature and high temperature. Two coexpression networks, comprising Gene Ontology (GO) terms, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, miRNAs, and target genes, were constructed, and genes with pivotal roles were discovered. Further, a candidate miRNA with a role in regulating rTGMS was predicted, and a mechanism by which it can potentially result in temperature-dependent fertility conversion was discovered. Our data provide a valuable information for miRNA research in eggplant for hybridization breeding, which broaden our understanding of the mechanisms underlying fertility conversion in rTGMS lines in plants.
Two inbred eggplant lines, rTGMS line 05ms and the fertile line S63, were used as the experimental materials (Supplemental Fig. 1). The rTGMS line was a natural mutation from S63. Plant seeds were sown on 3 Jan. 2020 in a greenhouse in Shijiazhuang (lat. 38.03°N, long. 114.26°E), Hebei province, China. When the plants grew to five to six true leaves (15 Apr.), seedlings of each line were planted in plastic greenhouse, where a self-measuring thermometer was placed. The rTGMS line 05ms exhibits male sterility at <18.0 °C and is fully fertile at >19.5 °C. Anthers samples at meiosis stage were collected at the sterile period of spring low temperature from 05ms (05ms-spring), the fertile period of summer high temperature from 05ms (05ms-summer), and the fertile period of spring low temperature from S63 (S63-spring), the fertile period of summer high temperature from S63 (S63-summer). Samples were taken in late April and early June. Each time, ≈10 to 15 flower buds were taken from five plants as a sample, and quickly peeled off the ovary and petals, collected the anthers, placed them in a 1.5-mL centrifuge tube, and rapidly transferred into liquid nitrogen (–196 °C). There were 12 samples in total (05ms-spring, 05ms-summer, S63-spring, and S63-summer, three biological replicates). Samples were taken back to the laboratory and stored in an ultra-low temperature freezer (–80 °C) for RNA extraction.
Total RNA was extracted from anther by using the DP441 RNA extraction Kit (Tiangen Biotech Co., Ltd., Beijing, China) according to the manufacturer’s instructions. RNA degradation and contamination was monitored on 1% agarose gels. RNA purity was checked using a spectrophotometer (NanoPhotometer; IMPLEN, Westlake Village, CA). RNA concentration was measured using RNA Assay Kit in a flurometer (Qubit 2.0; Life Technologies, Carlsbad, CA). RNA integrity was assessed using the RNA Nano 6000 Assay Kit of an automated electrophoresis tool (Bioanalyzer 2100; Agilent Technologies, Palo Alto, CA).
Twelve samples from four treatments (05ms-spring, 05ms-summer, S63-spring, and S63-summer, three biological replicates) were sent to Novogene Bioinformatics Technology Co. Ltd. (Beijing, China) for sRNA-seq analysis. A total amount of 3 μg RNA per sample was used as input material for the sRNA library. Sequencing libraries were generated using NEBNext Multiplex sRNA Library Prep Set for Illumina (NEB, Ipswich, MA) following the manufacturer’s recommendations, and index codes were added to attribute sequences to each sample. Polymerase chain reaction (PCR) products were purified on a 8% polyacrylamide gel (100 V, 80 min). DNA fragments corresponding to 140 to 160 bp (the length of small noncoding RNA plus the 3' and 5' adaptors) were recovered and dissolved in 8-μL elution buffer. Library quality was then assessed on the automated electrophoresis tool (Bioanalyzer 2100) using DNA High Sensitivity Chips (Agilent Technologies, Santa Clara, CA). The clustering of the index-coded samples was performed on a cBot Cluster Generation System using TruSeq SR Cluster Kit v3-cBot-HS (Illumia, San Diego, CA) according to the manufacturer’s instructions. After cluster generation, the library preparations were sequenced on an Illumina HisEq. 2500 platform and 50 bp single-end reads were generated. The raw data were deposited into the National Center for Biotechnology Information Sequence Read Archive with accession number PRJNA791470.
Clean data (clean reads) were obtained by removing reads using Cutadapt software. The sequence quality was verified using FastQC (Wingett and Andrews, 2018). Then, unique sequences (18 to 30 nt long) were chosen from clean reads to do all the downstream analyses. To prevent every unique sRNA mapping to multiple noncoding RNA, we used the following priority rule: known miRNA > ribosomal RNA (rRNA) > transfer RNA (tRNA) > small nuclear RNA (snRNA) > small nucleolar RNA (snoRNA) > repeat > gene > novel miRNA; this ensured that every unique sRNA mapped to only one annotation. sRNA tags were mapped to reference sequences (Hirakawa et al., 2014) using bowtie-0.12.9 (Langmead et al., 2009). The known and novel miRNAs and secondary structures drawn were identified by BLAST search with the miRBase22.1 database, modified software mirdeep2 (Friedlander et al., 2011), and sRNA-tools-cli. To remove tags originating from protein-coding genes, repeat sequences, rRNA, tRNA, snRNA, and snoRNA, sRNA tags were mapped to RepeatMasker, Rfam database, or those types of data from the specified species itself. In our analysis pipeline, families of known miRNAs were determined using miFam.dat (Kozomara et al., 2019), and novel miRNA precursors were submitted to Rfam (Nawrocki et al., 2015) to search for Rfam families. miRNA expression levels were estimated as transcripts per million, using previously described criteria (Zhou et al., 2010).
Analysis of differential expression between two groups was performed using the DESeq2 package ver. 3.0.3 (Love et al., 2014). The following criteria were used to identify significantly upregulated and downregulated miRNAs: | log2 (fold change) | > 1 and adjusted P < 0.05. miRNA target genes were predicted using psRobot_tar in psRobot for plants (Wu et al., 2012).
GO enrichment analysis was used on the target gene candidates of differentially expressed miRNAs (DEMs) in the following. GOseq based Wallenius noncentral hypergeometric distribution (Young et al., 2010), which could adjust for gene length bias, was implemented for GO enrichment analysis. KEGG is a database resource for understanding high-level functions and utilities of the biological system, such as the cell, the organism and the ecosystem, from molecular-level information, especially large-scale molecular datasets generated by genome sequencing and other high-throughput experimental technologies (Kanehisa et al., 2008). We used KOBAS (Mao et al., 2005) software to test the statistical enrichment of the target gene candidates in KEGG pathways.
Total miRNA samples were extracted from frozen anthers using a DP501 kit (Tiangen Biotech Co., Ltd.). miRNA First-Strand cDNA was synthesized using a KR201 Kit (Tiangen Biotech Co., Ltd.). An miRcute miRNA Detection Kit (FP401; Tiangen Biotech Co., Ltd.) was then used to quantify known miRNAs following the manufacturer’s instructions. Six DEMs were analyzed by quantitative real-time (qRT)-PCR with U6 as a reference. The miRNA forward primers are listed in Supplemental Table 1, and the reverse primer was a universal primer provided in the FP401 kit. Assays were performed on a CFX ConnectTM instrument (Bio-Rad Laboratories, Hercules, CA) using the following program: 94 °C for 2 min, followed by 40 cycles at 94 °C for 20 s and 60 °C for 30 s. Target genes of miRNAs were analyzed by qRT-PCR following a previously published method (Li et al., 2019). The primers are listed in Supplemental Table S, and actin was used as a reference. Data were analyzed using the 2-ρρCT method (Schmittgen and Livak, 2008).
ABA concentrations were detected on the testing platform (QTRAP 6500 LC-MS/MS; AB Sciex, Framingham, MA). Hydrogen peroxide and catalase (CAT) concentrations were measured using the H2O2-1-Y and CAT-1-Y kits (Comin Biotechnology Co., Ltd., Suzhou, China) with a testing tool (Cytation 1; BioTek Instruments Inc., Winooski, VT). Statistical analysis software (IBM SPSS Statistics ver. 18.0; IBM Corp., Armonk, NY) was used to conduct analysis of variance and the t test to compare 05ms-summer to 05ms-spring for each category. P < 0.05 was considered statistically significant in all analyses. Figures were prepared with an analysis software (Prism 8.0.2; GraphPad Software Inc., San Diego, CA). All data were averaged over three biological replicates.
To investigate the roles of miRNAs in response to temperature variation during anther development in 05ms, 12 sRNA libraries (three biological replicates respectively for 05ms-spring, 05ms-summer, S63-spring, and S63-summer) were constructed and a total of 166,273,427 raw data were obtained using the Illumina sequencing platform. After removing low-quality reads, 39,602,285 (05ms-spring), 41,921,223 (05ms-summer), 40,774,449 (S63-spring), and 39,515,432 (CH) clean reads were obtained, respectively (Supplemental Table 2). More than 90% of these sRNA could be mapped to reference sequences (Supplemental Table 3). sRNAs in all sequencing libraries were categorized. Interestingly, there were fewer rRNA, snRNA, and snoRNA in the 05ms line than in the S63 line in anthers from plants grown under both high and low temperature conditions (Table 1). sRNAs within specific length ranges were then selected for the subsequent analysis of sRNA distribution. The 18–30 nt sRNAs were dominant in all libraries. Among these, the most abundant sRNAs were 24 nt siRNAs (>40%), followed by 21–22 nt miRNA (10% to 15%; Fig. 1A). We analyzed the first nucleotide bias and found 20–23 nt miRNAs mostly starting with “U” as the first base, with “C” the first base in most 21 nt molecules from the 05ms-spring group. Further, 24-nt miRNAs mostly started with an ‘A’ residue in the S63-summer and 05ms-spring groups, whereas they began with “U” in the 05ms-summer and S63-spring groups (Fig. 1B). Next, sRNAs screened for length were matched to reference genome. A total of 143 known miRNAs and 104 novel miRNAs were detected in anthers from the four groups. These miRNAs belonged to 42 miRNA families by comparing them in Solanaceae crops, with miR156, miR166, miR395, miR169_2, miR172, and miR399 as the top six families (Fig. 1C).
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
To identify miRNAs related to anther development and fertility conversion in eggplant, we screened for DEMs in the 05ms-spring, 05ms-summer, S63-spring, and S63-summer libraries. DEMs were identified based on threshold values: false discovery rate < 0.05 and |log2 (fold-change)| > 1. Most DEMs (n = 113) were identified in the 05ms-spring vs. S63-spring groups, among which 61 and 52 miRNAs were upregulated and downregulated in 05ms-spring, respectively. Only 41 DEMs were identified in comparisons of both 05ms-spring vs. 05ms-summer and 05ms-summer vs. S63-summer (Fig. 2A). These findings suggest that some miRNAs were only expressed in the sterile line during the 05ms-spring period.
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
Venn diagram analysis of DEMs showed that six DEMs were most closely related to sterility because they did not differ significantly when comparing anther fertility groups (05ms-summer vs. S63-summer and S63-spring vs. S63-summer), but there were significant differences when comparing anther sterility groups (05ms-spring vs. S63-spring and 05ms-spring vs. 05ms-summer; Fig. 2B). Among these six DEMs, three novel miRNAs (Novel_116, Novel_119, and Novel_97) were upregulated in 05ms-spring, with Novel_97 the most significantly altered. The other three DEMs (nta-miR408, sly-miR168b-3p, and sly-miR397–5p), downregulated in 05ms-spring, were known miRNAs, with sly-miR397-5p the most significantly difference (Fig. 2C, Supplemental Table 4). The secondary structures of the six miRNAs are shown in Fig. 2D.
We performed qRT-PCR to verify further the expression of the six DEMs between 05ms and S63. The results of qRT-PCR were consistent with those of sRNA-seq (Fig. 3, Supplemental Table 5). Software analysis predicted 892 target genes for the six identified miRNAs. The target genes of miR397–5p were predicted, and 382 target genes were found, including transcription factors dysfunctional tapetum 1 (dyt1) and WRKY, and many LAC family genes. We verified the expression levels of four target genes by qRT-PCR and found the expression levels were opposite of miR397-5p in 05ms, thus further confirming that they were underlying target genes of miR397-5 (Supplemental Fig. 2). The four target genes of miR397–5p were Sme2.5_00016.1_g00020.1 (WRKY32), Sme2.5_05438.1_g00004.1 (dyt1), Sme2.5_01800.1_g00006.1 (LAC4), and Sme2.5_02906.1_g00004.1 (LAC4).
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
The regulatory function of miRNA is executed through its target genes. To further understand the biological functions of miRNAs, annotations of the 892 target genes were classified into three GO categories: molecular function (29.86%), biological process (55.43%), and cellular component (14.71%) (Fig. 4A). Significant terms included “copper ion binding” [GO: 0005507 (n = 21)], “oxidation-reduction process” [GO: 0055114 (n = 59)], and “oxidoreductase activity” [GO: 0016491 (n = 60)]. These results suggest that the redox is an important factor in rTGMS fertility conversion. The detailed information of the first 20 terms in GO functional enrichment analysis are listed in Supplemental Table S6.
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
To further explore our findings, KEGG enrichment analysis of 892 target genes was conducted. The target genes were mapped to 72 biological pathways, and the top 20 pathways of KEGG functional enrichment analysis is shown in Fig. 4B. Most target genes were enriched in “plant hormone signal transduction” [sly04075 (n = 7)], “biosynthesis of amino acids” [sly01230 (n = 6)], and “spliceosome” [sly03040 (n = 6)]. The pathways “other glycan degradation” [sly00511 (n = 3)], “valine, leucine and isoleucine degradation” [sly00280 (n = 4)], and “glycine, serine and threonine metabolism” [sly00260 (n = 3)] were also particularly noteworthy. The detailed information of the 72 terms in KEGG functional enrichment analysis is listed in Supplemental Table 7.
To functionally characterize DEMs and their target genes, we built regulatory networks including significant GO terms and KEGG pathways. The three GO terms included in the regulatory network comprised 64 target genes and five miRNAs (Fig. 5). Further, three KEGG pathways in the regulatory network included 11 target genes and 4 miRNAs (Fig. 6). Five genes were targeted by sly-miR397-5p, three of which (Sme2.5_31020.1_g00001.1, Sme2.5_09769.1_g00002.1, and Sme2.5_00110.1_g00020.1) were redox (dehydrogenase) enzymes. The sly-miR397-5p was the most linked in both the KEGG and GO networks, which could be a key miRNA in eggplant thermosensitive male sterility.
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
According to our previous research results, ABA and ROS had significant influence on male sterility of eggplant. Thus, ABA, H2O2, and CAT content in ‘05ms’ anthers from the 05ms-spring and 05ms-summer were detected. ABA and H2O2 were higher and CAT was lower in 05ms-spring than in 05ms-summer (Fig. 7). The content of ABA was 41.50 U/g in 05ms-spring, which was 2.16-fold higher than that in 05ms-summer. H2O2 content was 1.70-fold higher in 05ms-spring (2.52 µmol·g−1) than in 05ms-summer. In particular, the largest difference was that in CAT content, which was 4.02-fold higher in 05ms-summer (3184.18 nmol·g−1·min−1) than in 05ms-spring (792.48 nmol·g−1·min−1).
Citation: Journal of the American Society for Horticultural Science 147, 5; 10.21273/JASHS05222-22
Heterosis is a useful way to increase yield and improve eggplant quality. TGMS lines represent excellent germplasm resources because they eliminate the need for an emasculation procedure, thereby reducing the cost of hybrid seed production. With rapidly developing sequencing technology, many plant genomes have been sequenced, and the resulting omics data are excellent resources for molecular studies (Song et al., 2019). miRNAs are important in regulation of plant anther development (Dong et al., 2020). In this study, to determine the functions of miRNAs during eggplant rTGMS, miRNAs expression levels in the meiosis stage anthers from the rTGMS line 05ms and the homologous fertile line S63 were analyzed by high-throughput sequencing. This is the first report of a comprehensive comparison of miRNA expression levels between TGMS and temperature-insensitive eggplant lines.
ROS homeostasis is critical for anther development, and H2O2 is a key factor in ROS production (Hameed et al., 2012). Antioxidant enzyme-mediated redox homeostasis was a key index to estimate the deleterious effects of ROS in anthers. Superoxide dismutase content decreases and H2O2 content increases in stamens during anther development of genic male sterile mutant in cotton (Gossypium sp.; Zheng et al., 2021). Enhanced CAT activities in rice anther help to eliminate oxidative damage through scavenging of ROS (Zhao et al., 2018b). Further, a rise in temperature during anthesis can stimulate significantly increased ROS content and malondialdehyde accumulation in anthers, thereby triggering pollen mortality in wheat (Dwivedi et al., 2019). In this study, GO analysis indicated that target genes were primarily enriched in redox-related terms. Elevated ABA levels can also cause ROS release (Yu et al., 2019) and our data show that ABA and H2O2 were higher and CAT levels markedly lower in the 05ms-spring than in 05ms-summer groups (Fig. 6); therefore, we propose that low temperature stress can induce ROS burst, which may result in anther abortion in rTGMS line of eggplant.
Many studies have demonstrated that miRNAs regulate gene expression by controlling transcription factors, which may be an important feature of miRNA function (Liu et al., 2016). Some transcription factors were identified as miRNA target genes in the tomato photoperiod-sensitive male sterility mutant 7B-1, suggesting that they are associated with, or involved in, the regulation of male sterility (Omidvar et al., 2015). In this study, the expression level of miR397-5p in 05ms-spring was 29.45, whereas that in 05ms-summer was more than 6 times higher at 185.36. We speculate that low temperature severely inhibits miR397-5p expression. Target gene is key to understanding miRNA function. The four target gene ID numbers were presented by comparing the reference genome in 2019 and 2014 (Supplemental Table 8). dyt1 is a bHLH transcription factor, plays a critical role in regulating tapetum development, and its alteration can directly lead to anther abortion (Cui et al., 2016). dyt1 regulates a series of downstream genes, which are necessary for anther development, including tapetal development and function 1, Aborted microspores, and Male sterility 1, and interacts with many transcription factors (Gu et al., 2014). WRKY family genes respond to low temperature stress and affect plant stress resistance by participating in ABA signal transduction (Zhao et al., 2019). LAC genes encode key lignin biosynthesis enzymes (Xue et al., 2018). Lignin is a vital constituent part of the plant secondary cell wall and can improve mechanical support (Zhou et al., 2009). LAC was induced by ABA to positively regulated lignin synthesis and influence cell wall ductility (Li et al., 2020). Plant tissues with high LAC gene expression also have high levels of lignification (Wang et al., 2020). miR397 negatively regulated lignin content by slicing LAC transcription, thereby promoted plant defense (Wei et al., 2021). Thus, ABA participated in anther response to low temperature and formation of anther cell wall. Hence, miR397-5p likely regulates anther development and low temperature tolerance by negatively regulating these target gene expression in rTGMS line of eggplant.
The occurrence of male sterility in rTGMS line ‘05ms’ involved many biological processes and metabolic pathways. This is the first report of miRNAs in the anthers of the male sterility in eggplant using high-throughput sequencing technology. According to the analysis results of sRNA-seq, qRT-PCR, and related physiological indexes, we identified 143 known miRNAs and 104 novel miRNAs, and found that three known miRNAs (miR168b, miR397, and miR408) and three novel miRNAs (Novel_116, Novel_119, and Novel_97) probably played an important role during anther development in ‘05ms’. Target prediction and annotation showed that these target genes enriched in plant hormone signal transduction and oxidation-reduction process. The regulatory networks analysis found the miR397-5p was the most important miRNA. We speculate that the anther abortion under low temperature conditions might be caused by the hormone imbalance, the destruction of antioxidant system, and the accumulation of H2O2. Our results will facilitate understanding of the molecular mechanism underlying miRNA regulation of fertility conversion and two-line cross-breeding in eggplant.
The character analysis of small RNA (sRNA) in the sterile line 05ms and fertile line S63 of eggplant. (A) The length distribution of sRNAs. (B) The first nucleotide bias analysis of known sRNAs. (C) The microRNA (miRNA) families analysis of known miRNAs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). sRNA frequency represents the percentage of sRNA among S63-summer, S63-spring, 05ms-summer, and 05ms-spring under different length. First nucleotide represents the percentage of A/U/C/G in the first base of sRNA of this length. The numbers above each bar represent the total numbers of sRNA of this length. Different miRNA families were detected in the Solanaceae crops. Family numbers represent the numbers of different miRNA families.
Analysis of differential expression levels and secondary structures of microRNAs (miRNAs) in the sterile line 05ms and fertile line S63 of eggplant. (A) Numbers of differentially expressed miRNAs (DEMs) in different comparisons. (B) Venn diagram showing the overlaps in different comparisons. (C) Heatmap showing the hierarchical cluster analysis results from samples and DEMs. (D) Secondary structures of DEMs. DEMs represents the numbers of DEMs. “Upregulated” indicates upregulated DEMs. Downregulated indicates downregulated DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). The red star indicates the key DEMs. Each column indicates sample, and the colored bar indicates the relative expression level from high (red) to low (blue). Red highlights indicate the locations of mature sequence of miRNA.
Comparison of results obtained via small RNA sequencing (sRNA-seq) and quantitative real-time PCR (qRT-PCR) for the six differentially expressed microRNAs (DEMs) at different periods in the sterile line 05ms and fertile line S63 of eggplant. miR168b-3p, miR397–5p, and miR408, Novel_116, Novel_119, and Novel_97 were the six DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). Error bars indicate the SE of three biological replicates.
Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enrichment analysis of the 892 target genes in the sterile line 05ms and fertile line S63 of eggplant. (A) The top 20 significantly enriched GO terms. (B) The top 20 significantly enriched KEGG pathways.
Regulatory networks comprising Gene Ontology terms, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent GO terms, DEMs, and target genes, respectively.
Regulatory network comprising Kyoto Encyclopedia of Genes and Genomes pathways, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent KEGG pathways, DEMs, and their target genes, respectively.
The contents comparison of abscisic acid (ABA), hydrogen peroxide (H2O2), and catalase (CAT) between 05ms-spring and 05ms-summer in the sterile line 05ms. (A) ABA contents in 05ms-spring and 05ms-summer. (B) H2O2 contents in 05ms-spring and 05ms-summer. (C) CAT contents in 05ms-spring and 05ms-summer (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms). Error bars indicate the standard error of three biological replicates. Statistical significance was analyzed by a t test to compare 05ms-summer to 05ms-spring for each category in all statistical. **Significant at P < 0.01.
Morphological comparison of flowers in the sterile line 05ms and fertile line S63 of eggplant. (A) The phenotype of flower in 05ms. (B) The phenotype of flower in S63.
Quantitative real-time polymerase chain reaction confirmation of target genes at different periods in the sterile line 05ms and fertile line S63 of eggplant: 05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63. Sme2.5_00016.1_g00020.1, Sme2.5_01800.1_g00006.1, Sme2.5_05438.1_g00004.1, and Sme2.5_02906.1_g00004.1 are the four target genes. Different letters above the bars represent significant differences at P < 0.05 according to Duncan’s multiple range test.
Contributor Notes
This research was funded by the Scientific and Technological Innovation Talent Team Construction Project of Hebei Academy of Agriculture and Forestry Sciences (grant no. C21R0802), the Basic Scientific Research Fund of Hebei Academy of Agriculture and Forestry Sciences (grant no. 2021050202), the National Bulk Vegetable Industry Technology System Project (grant no. CARS-23-G-05), and the Third Round of the “Giant Plan” Vegetable Scientific Research and Innovation Team project in Hebei Province.
Y.W. is the corresponding author. E-mail: jzswuyanrong@163.com.
The character analysis of small RNA (sRNA) in the sterile line 05ms and fertile line S63 of eggplant. (A) The length distribution of sRNAs. (B) The first nucleotide bias analysis of known sRNAs. (C) The microRNA (miRNA) families analysis of known miRNAs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). sRNA frequency represents the percentage of sRNA among S63-summer, S63-spring, 05ms-summer, and 05ms-spring under different length. First nucleotide represents the percentage of A/U/C/G in the first base of sRNA of this length. The numbers above each bar represent the total numbers of sRNA of this length. Different miRNA families were detected in the Solanaceae crops. Family numbers represent the numbers of different miRNA families.
Analysis of differential expression levels and secondary structures of microRNAs (miRNAs) in the sterile line 05ms and fertile line S63 of eggplant. (A) Numbers of differentially expressed miRNAs (DEMs) in different comparisons. (B) Venn diagram showing the overlaps in different comparisons. (C) Heatmap showing the hierarchical cluster analysis results from samples and DEMs. (D) Secondary structures of DEMs. DEMs represents the numbers of DEMs. “Upregulated” indicates upregulated DEMs. Downregulated indicates downregulated DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). The red star indicates the key DEMs. Each column indicates sample, and the colored bar indicates the relative expression level from high (red) to low (blue). Red highlights indicate the locations of mature sequence of miRNA.
Comparison of results obtained via small RNA sequencing (sRNA-seq) and quantitative real-time PCR (qRT-PCR) for the six differentially expressed microRNAs (DEMs) at different periods in the sterile line 05ms and fertile line S63 of eggplant. miR168b-3p, miR397–5p, and miR408, Novel_116, Novel_119, and Novel_97 were the six DEMs (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63). Error bars indicate the SE of three biological replicates.
Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enrichment analysis of the 892 target genes in the sterile line 05ms and fertile line S63 of eggplant. (A) The top 20 significantly enriched GO terms. (B) The top 20 significantly enriched KEGG pathways.
Regulatory networks comprising Gene Ontology terms, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent GO terms, DEMs, and target genes, respectively.
Regulatory network comprising Kyoto Encyclopedia of Genes and Genomes pathways, differentially expressed microRNAs (DEMs), and their target genes in the sterile line 05ms and fertile line S63 of eggplant. Ellipses, triangles, and rectangles represent KEGG pathways, DEMs, and their target genes, respectively.
The contents comparison of abscisic acid (ABA), hydrogen peroxide (H2O2), and catalase (CAT) between 05ms-spring and 05ms-summer in the sterile line 05ms. (A) ABA contents in 05ms-spring and 05ms-summer. (B) H2O2 contents in 05ms-spring and 05ms-summer. (C) CAT contents in 05ms-spring and 05ms-summer (05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms). Error bars indicate the standard error of three biological replicates. Statistical significance was analyzed by a t test to compare 05ms-summer to 05ms-spring for each category in all statistical. **Significant at P < 0.01.
Morphological comparison of flowers in the sterile line 05ms and fertile line S63 of eggplant. (A) The phenotype of flower in 05ms. (B) The phenotype of flower in S63.
Quantitative real-time polymerase chain reaction confirmation of target genes at different periods in the sterile line 05ms and fertile line S63 of eggplant: 05ms-spring = spring sterile period at low temperature in 05ms; 05ms-summer = summer fertile period at high temperature in 05ms; S63-spring = spring fertile period at low temperature in S63; S63-summer = summer fertile period at high temperature in S63. Sme2.5_00016.1_g00020.1, Sme2.5_01800.1_g00006.1, Sme2.5_05438.1_g00004.1, and Sme2.5_02906.1_g00004.1 are the four target genes. Different letters above the bars represent significant differences at P < 0.05 according to Duncan’s multiple range test.
