Cracking of jujube fruits is associated with differential expression of metabolic genes

Here, we identified differentially expressed genes (DEGs) between cracked and normal jujube fruits, of which 785 DEGs were up‐regulated and 251 DEGs were down‐regulated. Several DEGs encode proteins with roles in metabolic processes, including growth hormones and surface wax production.

highly used because of its simple technique and rapid turnaround. Transcriptome sequencing can determine altered gene expression in response to different conditions. It can reveal the dynamics of gene networks and their respective functions, as well as the steady-state level of all expressed transcripts in each particular state [17]. In 2014, Liu et al. [18] completed the entire genome sequencing of winter jujube for the first time and preliminarily revealed the molecular mechanism of drought resistance, fruit branch detachment, and additional fruit properties through comparative genome and transcriptome analysis. The aim of this study is to profile differentially expressed genes (DEGs) in cracked and normal jujube fruits.

Materials and methods
Test materials and sample processing 'Huping' jujube was used for this experiment. Healthy jujube plants obtained from the resource garden of Shanxi Agricultural University consisted of both normal and cracked fruit at their mature stage (Fig. 1). Three biological replicates were used, and these were quickly stored in liquid nitrogen before being transported to Beijing Baimaiker technology Co., Ltd. (Beijing, China) for RNA extraction and transcriptome sequencing analysis. Three samples of normal fruit were numbered T01, T02 and T03; three samples of cracked fruit were numbered C01, C02 and C03. All plant tissues were immediately frozen in liquid nitrogen and stored at −80°C for later research.

Extraction and library construction of total RNA
Total RNA was extracted from the normal and cracked jujube fruit using the RNA extraction kit (Polysaccharides Pure Polyposis-rich) from Tiangen Biochemical Technology (Beijing) Co., Ltd. Capture beads for mRNA were added to the total RNA samples. After two rounds of binding, mRNA was eluted with Tris-HCl and incubated with the first strand synthesis reaction buffer and random primers. These short sequences were used as templates, consisting of six bases of random hexamers, to synthesize the first cDNA strand. Next, the buffer, dNTPs, RNase H and DNA polymerase I were added to synthesize the second strand of cDNA. cDNA was purified by AMPure XP beads. End repair reaction buffer and end repair enzyme mix were added to cDNA for end repair. Polyadenylation was performed through addition of poly(A) tails. The sequencing joints were subsequently connected, and USER enzyme was added to open the linker. The reaction products were supplemented with nuclease-free water, reaching a volume of 50 μL. They were then transferred to a 1.5-mL centrifuge tube for fragment selection. PCR amplification and product purification were performed, and finally, the cDNA library was enriched by PCR amplification.

Transcriptome sequencing and comparison with reference genomes
The high-throughput sequencing of the transcriptome was conducted by Beijing Baimaiker technology Co., Ltd. The cDNA library was sequenced on the Illumina X-TEN PE150 (San Diego, CA, USA), thereby generating the maximum 300-bp pair-end reads. Raw data in fastq format were first processed through in-house perl scripts to obtain clean data by removing the adapter, ploy-N and low-quality sequences (Q < 20). At the same time, Q20, Q30, GC content and the sequence duplication level of the clean data were calculated. All of the downstream analyses were based on clean data with high-quality clean reads. The clean reads were aligned with HISAT2 (http://ccb.jhu.edu/software/ hisat2) with default parameters to the Ziziphus jujuba (assembly ZizJuj_1.1) [18] on National Center for Biotechnology Information. Transcripts were assembled and quantified with STRINGTIE (http://ccb.jhu.edu/software/ stringtie).

Gene annotation and Kyoto Encyclopedia of Genes and Genomes functional enrichment analysis
To analyze the functional annotation and pathway enrichment of DEGs, we performed differential expression analysis based on the expression levels of genes in different samples, according to the Gene Ontology (GO) database (http://www.geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG; http://www.genome.jp/kegg/).

Quantitative RT-PCR verification of DEGs
Quantitative RT-PCR (qRT-qPCR) was used on nine DEGs with significantly different expression. The fluorescent quantitative PCR (qPCR) primers (Table 1) were designed using PRIMER PREMIER 5.0 (Primier company, Toronto, Canada). The jujube endogenous gene GATA binding protein 6 (Gene ID: 29300) with highly stable expression was used for reference [forward (F): 5 0 -TGGCTGGAAGATGGAAGATG-3 0 , reverse (R): 5 0 -ATGAAGTCTATCCCCAATCGC-3 0 ]. The TransScript II All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (One-Step gDNA Removal) reverse transcription kit was used for reverse transcription into cDNA. The components and conditions for the reverse transcription reaction system were as follows: Components: Total RNA, 1 μg; 5× TransScript II All-in-One SuperMix for qPCR, 4 μL; gDNA Remover, 1 μL; RNase-free water, up to 20 μL. Reverse transcription conditions: 55°C for 30 min; 85°C for 5 s; addition of 80 μL RNase-free water after the reaction to dilute the cDNA.
Two samples of normal fruit and cracked fruit were detected by real-time fluorescence qPCR, and the detection of each gene in each sample was repeated three times. The qPCR system consisted of forward primer (10 μM), 0.4 μL; reverse primer (10 μM), 0.4 μL; 2× TransStart Top Green qPCR SuperMix, 10 μL; template (diluted cDNA), 2 μL; and ddH 2 O to a final volume of 20 μL. The PCR conditions were as follows: 94°C for 2 min, 94°C for 5 s, 45 cycles of 60°C for 15 s and 72°C for 10 s. For the dissociation stage, the 2 ÀΔΔCT method was used to determine the relative expression of the sample to be tested. Results are expressed as the mean AE standard deviation. The experimental data were analyzed with the Student's t-test using SPSS (V18.0) statistical software (IBM SPSS, Amonk, NY, USA).

Transcriptomic analysis of jujube fruits
To identify molecular events possibly associated with fruit cracking, we performed comparative transcriptome analysis on cracked and normal jujube fruits ( Fig. 1). cDNA libraries were generated and subsequently sequenced using the Illumina X-TEN platform. After quality control, 51.81 Gb of clean data was obtained, yielding approximately 26 269 299-34 063 751 clean reads. In the jujube transcriptome, GC content made up 45.52% (Table 2). A fold change of 2 (false discovery rate < 0.01) was the cutoff for the detection of DEGs. Our analysis collectively identified 1036 DEGs, including 785 up-regulated and 251 down-regulated genes, that were differentially expressed between cracked jujube fruits and normal jujube fruits ( Fig. 2A).
To validate the RNA-seq results, we used qRT-PCR to assess DEGs. Eight up-regulated genes and one down-regulated gene found in cracked fruits were Table 1. Primers for qRT-PCR verification of DEGs in jujube fruits.

Primer
Primer sequence randomly selected. These DEGs were mainly involved in the biosynthesis of pectin methylesterase, swollenin and Xyloglucan endotransglucosylases (XET). The expression levels of these DEGs were significantly different between cracked and normal fruits, which was consistent with the results of transcriptome analysis (Fig. 2B). GO annotation was performed to explore the possible functions of DEGs. As shown in Fig. 3, DEGs were significantly assigned to the metabolic process, cellular process and single-organism process in the biological process category. In the cell component category, DEGs were significantly enriched in cell and cell parts, followed by organelles and membranes. In the molecular function category, DEGs were mainly assigned to catalytic activity and binding. Two to three times as many genes that were up-regulated than those that were down-regulated were associated with most categories, including metabolic process, cellular process, single-organism process, response to stimulus, biological regulation, cell and cell parts, and nucleic acid binding transcription factor activity ( Fig. 3 and Table S1).

Classification analysis of KEGG metabolic pathways
To further examine genes possibly involved in fruit cracking, we carried out significant enrichment analysis of the KEGG pathway, which allows systematic analysis of the metabolic pathways of gene products in cells. Among them, there were 13 metabolic pathways exhibiting significant differences between cracked and normal jujube (P < 0.05) ( Table 3 and Fig. 4). Notably, the biosynthetic metabolic pathways annotated as plant-pathogen interactions had the most DEGs. Moreover, amino acid and nucleotide sugar metabolism were annotated by 14 biosynthetic pathways. Twelve DEGs were enriched in the metabolism of α-linolenic acid associated with the biosynthesis of jasmonic acid (JA), seven DEGs in fructose and mannose metabolism, and five DEGs in plant endogenous hormone abscisic acid-related carotenoid biosynthesis pathways (Fig. 5A). The allene oxide cyclase (AOC) gene K10525 (EC: 5.3.99.6), allene oxide synthase (AOS) gene ko1723 (EC: 4.2.1.92) and   Table 4). The pathways with the highest number of up-regulated genes were those involved in plant-pathogen interactions, amino acid and nucleotide sugar metabolism, and alpha-linolenic acid metabolism (Fig. 4). Additional top-ranked pathways included metabolism of riboflavin, arginine and proline, cysteine and methionine, vitamin B 6 and thiamine; biosynthesis of keratin, subepithelium and wax, as well as phenylalanine, tyrosine and tryptophan; and synthesis and degradation of ketone bodies (Fig. 4).
For genes down-regulated in cracked fruits, the top 20 metabolic pathways were enriched ( Table 5). The pathways with the highest numbers of down-regulated genes included antenna proteins involved in photosynthesis, porphyrin and chlorophyll metabolism, fructose and mannose metabolism, thiamine metabolism, carotenoid biosynthesis and metabolism of ubiquinone and  other terpenoids. Quinone biosynthesis, photosynthesis and the amount of gene expression of the pentose phosphate pathway were restrained, and the expression level was decreased.

Discussion
In the market, consumers prefer juicy, crispy, and large jujube fruits. Not only is the overall quality reduced by cracks over the fruit, but consumer satisfaction is reduced as well. As noted, Liu et al. [18] have previously annotated the genome and gene families of jujube [21][22]. However, little is known about the change in gene expression that contributes to fruit cracking. Using RNA-seq and bioinformatic analysis, we collectively identified 785 up-regulated genes and 251 down-regulated genes that were differentially expressed in cracked jujube fruits compared with normal jujube fruits. To  our knowledge, we provide the first transcriptome dataset and comparative gene expression analysis of cracked and normal jujube fruits. The results of the GO term and KEGG pathway enrichment analysis demonstrated that several genes related to metabolic and hormone signaling pathways exhibit differential expression between normal and cracked jujube fruits. Previous results indicated that phytohormones, especially JA, play essential roles in fruit cracking in plants. Our data indicate that genes related to α-linolenic acid metabolism are significantly enriched in cracked fruits. As noted, α-linolenic acid is the original metabolic precursor of JA. JA, a new type of growth hormone, and its precursors and derivatives, referred to as jasmonates (JAs), play important roles in the regulation of many physiological processes and synthesis of metabolites in plant growth and development, and especially the mediation of plant responses to biotic and abiotic stresses [23,24]. In the biogenesis of JA, OPR3 is formed via AOS and subsequently AOC. The intermediate produced is further catalyzed by OPR3 to form (+)-7-iso -JA. Methyl jasmonate is formed when hydrogen (-H) on the carboxyl group of JA is replaced by methyl (-CH3) [25]. We observed that the AOC gene K10525 (EC: 5.3.99.6), AOS gene ko1723 (EC: 4.2.1.92) and OPR3 gene ko5894 (EC: 1. 3.1.42) are elevated, whereas LOX2 gene (ko0454, EC: 1.13.11.12) is down-regulated in a time-course manner with fruit cracking development. The dynamic changes of these genes in this study suggest that phytohormone regulation may be related to fruit cracking. However, further metabolic analysis of hormone production is required to confirm a potential link between JA synthesis and fruit cracking.
KEGG analysis also revealed that the cutin, suberin and wax biosynthesis pathways were significantly enriched (P = 0.017563046; ko00073) for genes upregulated in cracked fruits. It is important to note that surface waxes play protective roles against pathogen infection, herbivorous insects and environmental stresses, such as drought, UV damage and frost. Plant very longchain fatty acids (VLCFAs) are known to be involved in the process of biofilm membrane lipid synthesis and serve as precursors for the biosynthesis of stratum corneum waxes. The synthesis of VLCFAs is catalyzed by fatty acyl-CoA elongase, which is a multienzyme system composed of KCS, 3-ketoacyl-CoA reductase, 3-hydroxyacyl-CoA dehydratase (HCD) and trans-2,3-enoyl-CoA reductase. The synthesized VLCFAs enter the stratum corneum waxy synthesis pathway by decarbonylation and acyl reduction to form various waxy components.
KCS is a rate-limiting enzyme in the endoplasmic reticulum that catalyzes the first step of the condensation reaction in the synthesis of VLCFAs. It has been studied in the fruit-setting stage of sweet cherry fruit [26] that PaKCS6 exhibits higher expression in rip-prone compared with rip-resistant varieties. In this study, we support this idea by providing evidence that the gene encoding KCS (EC: 2.3.1.199) (ko00062; ko01100; ko01110) is dramatically changed in cracked fruits (Fig. 5B), which might lead to altered synthesis of cuticle wax and consequently jujube cracking. Our findings from transcriptome analysis are in congruence with a previous study regarding the outer skin and pulp tissue of tomatoes by Mintz-Oron Mintz-Oron et al. [27].

Conclusions
Our study provides an atlas of DEGs between cracked and normal jujube fruits. Our data may serve as a valuable resource for investigation into the mechanisms by which jujube fruits undergo cracking.