Introduction

Strawberry (Fragaria ananassa) is an economically important fruit crop. Sugar content and composition (sucrose, glucose, and fructose) are critical to fruit development, and are controlled by sugar transport, metabolism, and accumulation. Sucrose is unloaded into strawberry fruit through an apoplast pathway (Li et al. 2012). It is also suggested that sucrose synthase (SS/Sus, EC 2.4.1.13) might play a dominant role in sucrose accumulation during the fruit ripening (Tian et al. 2012). However, substantial evidence regarding the molecular mechanism underlying sucrose accumulation in strawberry fruit development is lacking.

Sus, which catalyzes the reversible conversion of sucrose to monosaccharose, is a key enzyme in sucrose metabolism. In plants, its main physiological function is to cleave sucrose to generate UDP-glucose, which plays an important role in the biosynthesis of storage products (Huber and Akazawa 1986; Sung et al. 1988). In Arabidopsis seed, sucrolysis (cleaving sucrose) by Sus is required for carbon partitioning and accumulation, which may provide energy for phloem transport in the silique wall, to a large extent, AtSus2 serves as a marker during the seed maturation (Baud et al. 2004; Fallahi et al. 2008; Núñez et al. 2008; Angeles-Núñez et al. 2010). The enzyme’s key role, serving as a regulator in sink strength, has also been confirmed in cotton fiber formation (Jiang et al. 2012). In most grain crops, such as wheat (Hou et al. 2014), maize (Li et al. 2013), barley (Sreenivasulu et al. 2004), broad bean (Borisjuk et al. 2002), and rice (Karrer et al. 1992), the main function of Sus is played in starch biosynthesis.

Juicy fleshy fruit development is a distinctive process in plants. In citrus, Sus is a key enzyme for sugar accumulation, and of the six genes (Sus1–6), CitSus1 and CitSus2 were predominantly expressed in fruit juice sacs (Islam et al. 2014). In tomato fruit, the higher accumulation of hexose is correlated with an increased expression levels of both vacuolar invertase and Sus genes (Bastías et al. 2011), and interestingly, downregulation of the Sus activity inhibits the unloading capacity of sucrose (D’Aoust et al. 1999). Tian et al. (2012) suggested that sucrose accumulation during grape berry and strawberry fruit ripening might be controlled predominantly by Sus activity. In most fleshy fruits, such as pineapple (Zhang et al. 2012), grape (Martínez-Esteso et al. 2011), litchi (Yang et al. 2013), muskmelon (Wen et al. 2010), and tomato (Kortstee et al. 2007), the sucrolysis by Sus contributes to hexose accumulation and fruit quality formation.

As an important enzyme in plant growth and development, the regulation mechanisms of Sus expression were also studied in several plants. In grape berries, the Sus transcripts were repressed by glucose (Wang et al. 2014). In Arabidopsis seeds, AtSUS2 expression could be induced by exogenous glucose but not by exogenous sucrose (Angeles-Núñez et al. 2012). The Sus activity may be regulated by tomato sucrose non-fermenting-1-related protein kinase 1 (Wang et al. 2012) and a rice calcium-dependent protein kinase (Asano et al. 2002).

Although strawberry is a model plant for study of non-climatic fruit ripening, the function and regulation of the Sus remains unclear. In the present study, a strawberry Sus gene, FaSS1, was cloned and silenced through virus-induced gene silencing. Additionally, its expression and regulation were also analyzed, finding that FaSS1, an abscisic acid/sucrose-inhibited gene, plays an important role in the regulation of strawberry fruit ripening.

Results

Transcriptome and SqRT-PCR analysis of Sus gene family in strawberry fruit development

Based on the description by Jia et al. (2011), four-stage fruits, including large green, white, initial red, and partial red fruits around the onset of ripening, were used for RNA-sequencing (RNA-seq). A total of eight Sus gene-like contigs were found and classified into three groups based on their predicted log2 expression levels (Fig. 1a). The contig comp64823_c0_seq1 and its complementary strand comp65427_c0_seq1 were expressed highest in large green fruits, then declined sharply and maintained relative high levels from white to partial red stages compared with other contigs, and thus the contig was named FaSS1 (loc101306372 in https://strawberry.plantandfood.co.nz/index.html). To further confirm the mRNA expression levels of FaSS1 in developmental fruits, semiquantitative RT-PCR (SqRT-PCR) was also performed using the four-stage fruits. A similar trend of FaSS1 expression was observed around the onset of the ripening (Fig. 1b), suggesting that the processes of the fruit ripening might be required for a high and stable expression of FaSS1.

Fig. 1
figure 1

Transcriptome data and SqRT-PCR analysis of sucrose synthase gene expression. a Eight contigs related to sucrose synthase were found and classified into three groups based on their log2 gene expression levels gained from RNA-seq data. b SqRT-PCR analysis of FaSS. Actin was used as internal control. SG small green fruits, LG large greening fruits, Wt white fruits, IR initial red fruits, PR partial red fruits, FR full red fruits

Full size image

Cloning of the FaSS1 gene

To isolate the FaSS1 gene, the sequence gained from RNA-seq was used as a query to search the strawberry gene library (https://strawberry.plantandfood.Co.nz/index.html) using the BLAST algorithm. A highly homologous sequence was found and the specific primers were designed. Finally, a 2421-bp sequence encoding 806 amino acids was cloned by RT-PCR. A protein domain search in the NCBI website (http://structure.ncbi.nlm.nih.gov/structure/cdd/wrpsb.cgi) was performed using the amino acid sequence, in which a sucrose metabolism domain [GT1-sucrose-synthesis (Supplimentary Fig. 1A); red underlined (Supplimentary Fig. 1B)] and a sucrose-synthesis domain [sucrose-synth (Supplimentary Fig. 1A); green underlined (Supplimentary Fig. 1B)] were predicted. Thus, the putative strawberry Sus gene, FaSS1, was isolated successfully.

Functional analysis of FaSS1 in strawberry fruit by virus-induced gene silencing

To investigate the role of FaSS1 in developmental fruit, we generated tobacco rattle virus (TRV)-based pTRV1- and pTRV2-FaSS500 RNAi fruits using de-greened fruits according to the description by Jia et al. (2011). 2 weeks after inoculation, the surface of the control fruits inoculated with empty TRV vector only, turned fully red; in contrast, the inoculated sector on the surface of the RNAi fruits remained white (Fig. 2a).

Fig. 2
figure 2

Downregulation of FaSS1 transcripts using virus-induced gene silencing in strawberry fruits. a 2-week-old large green fruits still attached to the plants were used for inoculation. The control fruits were inoculated with Agrobacterium containing the TRV only (control), while treated fruits were inoculated with Agrobacterium containing the TRV carrying a FaSS1 fragment (RNAi1 and RNAi2). 2 weeks after inoculation, phenotypes were observed for the control fruits and RNAi fruits. b, c FaSS transcripts of the control and RNAi fruits were detected by SqRT-PCR analyses. Actin was used as internal control. Red bars 1.0 cm

Full size image

To confirm the suppression of FaSS1 expression at the molecular level, a SqRT-PCR analysis was performed using the RNAi and control fruits. The FaSS1 transcripts were downregulated in the RNAi fruits in comparison with the control fruits (Fig. 2b, c). These results indicated that FaSS1 plays a role in the regulation of strawberry fruit ripening.

Understanding the role of FaSS1 in the regulation of fruit ripening

To further understand the role of FaSS1 in the regulation of fruit ripening, we analyzed the soluble sugar contents, including sucrose, glucose, and fructose, together with the ripening-related events, including fruit firmness and anthocyanin contents in the RNAi and control fruits. We also assessed the transcripts of the ripening-related genes polygalacturonase 1 (PG1), pectate lyase 1 (PL1), and xylitol dehydrogenase 2 (XYL2) for fruit firmness, and chalcone synthase (CHS), chalcone isomerase (CHI), and dihydroflavonol 4-reductase (DFR) for anthocyanin (Jia et al. 2011, 2013; Tian et al. 2012). The results showed that the sucrose contents were lower in RNAi fruits than in control fruits; however, the glucose and fructose contents were higher in RNAi fruits than in control fruits (Fig. 3a). Fruit firmness increased but the anthocyanin contents decreased in RNAi fruits compared with the control fruits (Fig. 3b). The SqRT-PCR analysis showed that the transcripts of the ripening-related genes, including PG1, PL1, XYL2, CHS, CHI, and DFR, were all significantly downregulated in RNAi fruits compared with the control fruits (Fig. 3c). The changes in the ripening-related events and gene transcript levels in RNAi fruits are consistent with a role of FaSS1 in the regulation of the fruit ripening.

Fig. 3
figure 3

Changes of the ripening-related parameters in RNAi fruit and effects of sugar and ABA on FaSS1 transcripts in vitro. a Determination of soluble sugar contents by HPLC. The standard samples used were D-(+) glucose, D-(–) fructose, and sucrose (Sigma–Aldrich). b Determination of firmness and anthocyanin contents. c SqRT-PCR determination of transcripts of the ripening-related genes: PG1, PL1, and XYL2 for firmness; CHS, CHI, and DFR for anthocyanin. d Fruit discs were treated by exogenous sucrose, glucose, and ABA according to the description by Jia et al. (2013). FaSS1 transcripts were detected by SqRT-PCR after 2-h incubation. Actin was used as internal control. Error bars represent the SE (n = 3). Different letters indicate statistically significant differences at P < 0.05 as determined by Duncan’s test

Full size image

To explore the regulation of FaSS1 expression, effects of ABA, sucrose (Suc), and glucose (Glu) on the mRNA expression levels of FaSS1 were analyzed using a fruit-disc-incubation test in vitro. The SqRT-PCR analysis showed that, compared with the control incubated with 4 % mannitol, 100 µM ABA or 4 % sucrose significantly inhibited the mRNA expression levels of the FaSS1 gene, respectively; in contrast, 4 % glucose had no effect on the gene transcripts (Fig. 3d). The results demonstrated that FaSS1 expression could be inhibited by ABA and sucrose.

Discussion

Sugar accumulation plays an important role in fruit development and quality, and sucrose, an end product of photosynthesis, is metabolized by Sus, which is as an initial step in the biosynthesis of storage products in many plant sink organs, including starch biosynthesis (Karrer et al. 1992; Borisjuk et al. 2002; Sreenivasulu et al. 2004; Li et al. 2013; Hou et al. 2014) and hexose accumulation (Kortstee et al. 2007; Wen et al. 2010; Martínez-Esteso et al. 2011; Li et al. 2013; Zhang et al. 2012; Yang et al. 2013). Although the Sus is suggested to play a dominant role in sucrose accumulation during strawberry fruit ripening (Tian et al. 2012), substantial evidence is yet lacking.

In the present study, we have provided physiological and molecular evidence to confirm the following hypothesis: (1) FaSS1 transcripts are stably and constantly expressed during the red-color development and enlargement of the fruits (Fig. 1); (2) downregulation of FaSS1 transcripts inhibits the fruit ripening, as evidenced by changes in its mRNA expression levels (Fig. 2) and the ripening-related parameters, including fruit firmness, and the anthocyanin and soluble sugar contents, as well as the variations in transcripts of the ripening-related genes, including PG1, PL1, XYL2, CHS, CHI, and DFR (Fig. 3); and (3) FaSS1 expression could be inhibited by ABA and sucrose (Fig. 3).

Plant hormones and sugar accumulation are critical to fruit ripening (Jia et al. 2011; Chai et al. 2013; Sun et al. 2015). It was previously reported that both ABA and soluble sugar contents increase, on the whole, during strawberry fruit development, and this is especially true for sucrose, which increases slowly during de-greening and then increases rapidly during red-color formation (Jia et al. 2011). In the present study, FaSS1 transcripts were most highly expressed in large green fruits, and then expression declined sharply and was maintained at a relatively high level after the white stage (Fig. 1). The rapid increase in the sucrose contents might result from the sharp decrease in FaSS1 expression, suggesting a sucrolysis role of FaSS1 in the developmental fruits. The rapid decline in FaSS1 transcripts might result from the rapid increase in both ABA and sucrose contents, in that FaSS1 expression could be inhibited by ABA and sucrose (Fig. 3). In the FaSS1-RNAi fruits, the sucrose contents were significantly downregulated (Fig. 3), which might result from that the downregulation of FaSS1 expression could inhibit phloem transport capacity for sucrose, similar to a result reported by Fallahi et al. (2008). Given that sucrose functions as a signal involved in the regulation of strawberry fruit development and ripening (Jia et al. 2013), as a result, the decreased sucrose in RNAi fruits might lead to unripening fruits. Our results have demonstrated that a high, constant and stable expression of FaSS1 is required for the strawberry fruit ripening after white stage.

Materials and methods

Plant materials

Octaploid strawberry (F. ananassa cv. ‘Beinongxiang’) fruits were used in this study. Strawberry plants were grown under a glasshouse environment (20–25 °C, 70–85 % relative humidity, and a 14/10 h light/dark regime) in springtime during 2014–2015. Three hundred flowers on 40 strawberry plants were tagged during anthesis. Different developmental stages of fruits, including small green, big green, white, initial red, partially red and fully red fruits, were collected at 6, 15, 22, 24, 26, and 29 days after anthesis, respectively. Twenty fruits with uniform size were sampled at every stage (one replication), and quickly frozen in liquid nitrogen and stored at −80 °C until use.

RNA extraction, library construction, and RNA-seq

RNA was extracted using the RNeasy plant mini kit (Qiagen, Dusseldorf, Germany) from mixed receptacles (n = 3) for every stage fruit above. DNase digestion with RNase-Free DNase (Qiagen) was performed to remove contaminating DNA, and the RNA samples were then processed using the RNA library prep kit (New England BioLabs, Ipswich, MA, USA) and sequenced on the Illumina HiSeq2000 platform (Illumina, USA).

RNA-seq data analysis

The raw reads were filtered with the FASTQ_Quality_Filter tool from the FASTX-toolkit, and more than 35 bp and score higher than 20 were kept (Ming et al. 2012). All valid reads were combined to perform de novo splicing by the paired-end method with Trinity software (Wang et al. 2010). The longest transcripts per locus were used as a unigene. The bowtie2-2.2.2 software was used to compare reads with unigenes using the single-end mapping method (Wang et al. 2010; Ben et al. 2012). To compare the levels of unigene expression among the four libraries, the transcript level of each expressed unigene was calculated and normalised to the reads per kilobase of the exon model per million mapped reads (RPKM, Mortazavi et al. 2008). The significance of differential unigene expression was determined by using the Chi-squared test with a threshold of P = 0.05. P values were adjusted to account for multiple testing by using the false discovery rate (FDR) and assigned error ratio Q values (<0.05) (Benjamini et al. 2001). The unigenes with an adjusted P value of <0.05 and an absolute value of log2 (expression fold change) >1 were deemed to be differentially expressed, while the unigenes with an FDR-adjusted P value of <0.05 were considered statistically significant.

Cloning of FaSS1 gene and SqRT-PCR analysis

The cDNA obtained above was used as a template for amplifying the FaSS1 gene with primers designed from a strawberry gene library (https://strawberry.plantandfood.co.nz/index.html) (forward, 5′-ACCCACATCCCATTTCTCTCT-3′; reverse, 5′-CTCAGCCAACAGATTGCTTCT-3′). PCR was performed under the following conditions: 94 °C for 5 min, followed by 35 cycles at 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 1 min, with a final extension at 72 °C for an additional 10 min. The PCR products were ligated into a T1-Sample vector and subsequently transformed into Escherichia coli DH5a. Positive colonies were selected using kanamycin (100 µg/ml), and sequenced by Huada China (Beijing, China). The primers for SqRT-PCR were designed as showed in Supplimentary Table 1. The entire process was repeated three times.

Construction of the virus vector and agrobacterium-mediated infiltration

The pTRV1 and pTRV2 vectors (Liu et al. 2002) were kindly donated by Dr. Liu Yu-le, Qinghua University. A 424-bp cDNA fragment of FaSS1 was amplified using primers (sense, 5′-CGAGTTCCACACCTACCGTG-3′; and antisense, 5′-GAACAGCCGCGTGTCCTCCTC-3′). The amplified fragment (from 114 to 537 bp) was cloned into the T1-Sample vector (Trans), digested by XbaI and SmaI, and subsequently cloned into the virus vector XbaI-SmaI-cut pTRV2. The Agrobacterium strain GV3101 containing pTRV1, pTRV2, or the pTRV2 derivative pTRV2-FaSS1 was used for RNAi. Agrobacterium-mediated TRV infiltration by syringe injection with a needle into strawberry fruits was performed as described by Fu et al. (2005).

Determination of firmness and anthocyanin and ABA contents

Three uniform strawberry fruits were used for the detection of firmness and anthocyanin and ABA contents, respectively. Flesh firmness was analyzed on two sides of each fruit using a GY-4 fruit penetrometer (Digital Force Gauge, Shanghai, China). The strength of flesh firmness was recorded as N·cm−2. The anthocyanin content was analyzed using a spectrophotometer according to the methods described by Villarreal et al. (2010). The ABA content was analyzed using HPLC (high performance liquid chromatography) according to the methods described by Jia et al. (2011). The entire process was repeated three times.

Fruit disc incubation in vitro

Incubation of fruit disc with sucrose, glucose, and ABA in vitro was done as described by Jia et al. (2013): the discs were divided into four sections, one section was incubated in equilibration buffer with 4 % mannitol as control, the other were incubated in equilibration buffer with 4 % sucrose, 4 % glucose, and 100 µM ABA, respectively. The four sections were placed in 250 ml flask and shaking at 25 °C for 2 h. After washing with double distilled water, the tissues at each time points were frozen in liquid nitrogen and kept at −80 °C until use. The entire process was repeated three times.