Morphological, Transcriptomic and Hormonal Characterization of Trimonoecious and Subandroecious Pumpkin (Cucurbita maxima) Suggests Important Roles of Ethylene in Sex Expression

Abstract

: Sex expression is a complex process, and in-depth knowledge of its mechanism in pumpkin is important. In this study, young shoot apices at the one-true-leaf stage and 10-leaf stage in Cucurbita maxima trimonoecious line '2013-12' and subandroecious line '9-6' were collected as materials, and transcriptome sequencing was performed using an Illumina HiSeq^TM 2000 System. 496 up-regulated genes and 375 down-regulated genes were identified between shoot apices containing mostly male flower buds and only female flower buds. Based on gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, the differentially expressed genes were mainly enriched in the ethylene and auxin synthesis and signal transduction pathways. In addition, shoot apices at the 4-leaf stage were treated with the ethylene-releasing agent 2-chloroethylphosphonic acid (Ethrel), aminoethoxyvinyl glycine (AVG), AgNO₃ and indoleacetic acid (IAA). The number of female flowers up to node 20 on the main stem of '2013-12' increased significantly after Ethrel and IAA treatment and decreased significantly after AVG and AgNO₃ treatment. The female flowers in '9-6' showed slight changes after treatment with the exogenous chemicals. The expression of key genes in ethylene synthesis and signal transduction (CmaACS7, CmaACO1, CmaETR1 and CmaEIN3) was determined using quantitative RT-PCR, and the expression of these four genes was positively correlated with the number of female flowers in '2013-12'. The variations in gene expression, especially that of CmaACS7, after chemical treatment were small in '9-6'. From stage 1 (S1) to stage 7 (S7) of flower development, the expression of CmaACS7 in the stamen was much lower than that in the ovary, stigma and style. These transcriptome data and chemical treatment results indicated that IAA might affect pumpkin sex expression by inducing CmaACS7 expression and indirectly affecting ethylene production, and the ethylene synthesis and signal transduction pathways play crucial roles in pumpkin flower sex expression. A possible reason for the differences in sex expression between pumpkin lines '2013-12' and '9-6' was proposed based on the key gene expression. Overall, these transcriptome data and chemical treatment results suggest important roles for ethylene in pumpkin sex expression.

Keywords: Cucurbita maxima; chemical treatment; ethylene signal synthesis and transduction; floral sex expression; transcriptome sequencing.

Figure 1

Different flower phenotype of two materials. (A) Flower phenotype of ‘9–6’. (B) Flower phenotype of ‘2013–12’. ‘9–6’ and ‘2013–12’ produced female flowers 4% and 51% respectively when observed up to node 20 on the main stem. Blue arrows indicated male flowers and red arrows indicated female flowers.

Figure 2

Correlation analysis and principal component analysis of samples. (A) Correlation analysis of gene expression between samples. Comparisons of gene expression between replicates on the x-axis and those on the y-axis. The correlation coefficients between replicate samples were approximately 1. The correlation coefficient was show as a color value (blue: less similar, and red: more similar). (B) Principal component analysis plot of RNA sequencing (RNA-seq) data in samples. Three biological replicates for each sample type were clustered together. Blue plots indicated samples of ‘9–6’ at one-true-leaf stage (LS1) (MS1), purple plots indicated samples of ‘9–6’ at 10-leaf stage (LS10) (MS10), pink plots indicated samples of ‘2013–12’ at LS1 (FS1), and green plots indicated samples of ‘2013–12’ at LS10 (FS10).

Figure 3

Heatmap visualization of expression profiles of MS1, MS10, FS1 and FS10. (A) Heatmap visualization of expression profiles between FS1 and FS10. (B) Heatmap visualization of expression profiles between FS1 and MS1. (C) Heatmap visualization of expression profiles between FS10 and MS10. (D) Heatmap visualization of expression profiles between MS1 and MS10. Red arrow, > two-fold up-regulation; green arrow, > two-fold down-regulation.

Figure 4

Number of differentially expressed genes among MS1, MS10, FS1 and FS10. (A) Number of down-regulated expressed genes. (B) Number of up-regulated expressed genes. MS1 indicated samples of ‘9–6’ at LS1, MS10 indicated samples of ‘9–6’ at LS10, FS1 indicated samples of ‘2013–12’ at LS1, and FS10 indicated samples of ‘2013–12’ at LS10.

Figure 5

The gene ontology classification of up-regulated genes (A) and down-regulated genes (B) between young shoot apices containing male flower buds and female flower buds.

Figure 6

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway scatterplot of 871 differentially expressed genes (DGEs). The x-axes indicate enrichment score. The bigger the bubble, the more the DGEs. The smaller the p-value, the more significant the KEGG enrichment.

Figure 7

Expression comparison of 17 selected genes using RNA-seq and quantitative reverse transcription polymerase chain reaction (qRT-PCR). (A) Relative expression of 17 selected genes in FS1 and FS10 using qRT-PCR. The CmaACO1 expression at FS10 was assumed as 1. Data are displayed as the ratio of expression to CmaActin with three biological replicates. Error bars represent standard error (SE). The qRT-PCR of primers used are given in Table S3. (B) Comparison of the expression ratios of 17 selected genes in ‘2013–12’ at LS1 and LS10 stage using RNA-seq and qRT-PCR.

Figure 8

Expression changes of CmaACS7 (A), CmaACO1 (B), CmaETR1 (C) and CmaEIN3 (D) after Ethrel (Eth), aminoethoxyvinyl glycine (AVG), AgNO₃ and indoleacetic acid (IAA) treatment. The expression level of CmaACS7, CmaACO1, CmaETR1 and CmaEIN3 were detected using qRT-PCR. Gene expression of control (CK) ‘2013–12’ was assumed as 1 in chemical treatment ‘2013–12’, and control (CK) ‘9–6’ was assumed as one in chemical treatment ‘9–6’. Data are displayed as the ratio of expression to CmaActin with three biological replicates. Error bars represent standard error (SE). The qRT-PCR of primers used are given in Table S3. Phytohormone levels followed by the same lowercase letter are not significantly different at p < 0.05 using SSR’s test.

Figure 9

Relative expression of CmaACS7 at different stages (S1-S7) of ‘2013–12’ flower development using qRT-PCR analyses. The CmaACS7 expression at S1 ovary was assumed as unity. The relative expression level of CmaACS7 in ovary, stigma and style was significantly high, while less expression was observed in stamen. Data are displayed as the ratio of expression to CmaActin with three biological replicates. Error bars represent standard error (SE). The qRT-PCR of primers used are given in Table S3. Phytohormone levels followed by the same lowercase letter are not significantly different at p < 0.05 using SSR’s test.