1. Introduction
Tomatoes are one of the most economically important food crops, and belong to
Solanaceae family. The dietary consumption of tomatoes is linked to many health benefits. Tomatoes contain lycopene, folate, and vitamins C and K, and are rich in minerals; it has been reported that the consumption of tomatoes can reduce the risk of heart disease and cancer [
1,
2]. These health benefits, coupled with their distinctive taste, make tomatoes desirable and subject to high consumer demand. However, tomatoes are a perishable, seasonal fruit. Worldwide tomato production in 2018 was more than 180 million tons (t), of which only about 1/5 were consumed without processing; the largest proportion was used for making canned tomatoes and tomato concentrates (
https://www.globenewswire.com, accessed on 10 September 2019). Preserving the quality of fresh tomatoes and extending their shelf-life remains a big challenge in the food industry, even though various strategies have been developed, including the use of carbon dioxide [
3,
4].
Ethylene is a plant hormone that regulates climacteric fruit ripening, and a number of studies have been conducted that analyze the ethylene profile during tomato ripening and its response to adverse environmental conditions like low oxygen, high temperature, etc. [
3,
5,
6]. Among the promising technologies developed to regulate ethylene production or to maintain food quality, the use of CO
2 has attracted much attention due to its easiness to obtain, low cost [
7,
8], and long history of safe use for producing luscious and ripe fruits to be sold in grocery stores [
9,
10,
11]. Research has found that CO
2 can either promote or inhibit ethylene production. It has been reported that the treatment of tomatoes, from the breaker to the turning stage of the ripening process, with 80% CO
2 flow for 24 h stimulated ethylene production, but delayed the color change [
12]. Other reports showed that the application of 20% CO
2 flow to tomatoes at the pink stage for 24 h [
13], or at the light red stage for 5 d [
14], reduced ethylene production and suppressed color development. Moreover, studies on climacteric plants that used modified air with CO
2 content in the range of <1% to 100% showed that the life cycle of the tomato plant could be affected by elevated external CO
2 from the biochemical aspects of biomass synthesis, sugar signaling pathways, and hormonal crosstalk [
12,
13,
14,
15,
16]. Based on the results of these previous studies, it is clear that the presence of CO
2 affects ethylene production and the downstream molecular mechanisms that regulate fruit growth and ripening. Yet, it is still not clear how CO
2 exerts this effect on cellular activity. The use of pure CO
2 may provide detailed information on the extent and magnitude of this effect at the molecular level.
In this study, transcriptomics was applied in order to obtain insight into the effect of CO
2 stress, under post-harvest conditions, on tomato ripening at the genomic level. Tomatoes were conditioned with gaseous CO
2 at room temperature, and their differential gene expression patterns were identified and the alteration of transcription during different developmental stages was analyzed. The application of this advanced genome analysis to elucidate the fruit ripening process at the molecular level provides a more detailed understanding of fruit ripening; in addition, this information may also be utilized to generate non-transgenic plants with improved fruit quality [
17,
18,
19,
20].
2. Materials and Methods
2.1. Tomato Fruit and CO2 Treatment
Tomatoes (Solanum lycopersicum L.) in the mature green stage with uniform shapes and an average weight of 160–200 g per fruit were obtained from Coastal Sunbelt Produce, LLC (Savage, MD, USA). The tomatoes were randomly sorted into three groups. Group 1 was treated with a mixed gas (5% CO2 and 95% air) at a flow rate of 75–100 mL/min for 14 d (T1). Group 2 was treated with 100% CO2 gas at a flow rate of 2 L/min for 3 h (T2). Group 3 was used as a control (CT; airflow at 75–100 mL/min for 14 d). For each group, 6–7 tomatoes were loaded into an airtight jar (3.5 L) connected with a CO2 cylinder (Air Products, Allentown, PA, USA) and an in-house air supply; gas filters, a gas mixer, and gas flow meters were used to adjusted gas composition and maintain flow rate. At each designated time point, two jars from each group were disconnected from the gas supply lines, and the tomatoes were removed and placed on a storage rack maintained at 22–24 °C and 65–75% relative humidity (RH) for subsequent physiological examination, RNA extraction, and characterization.
2.2. Characterization of Physical and Chemical Properties
Tomatoes from the T1, T2, and CT groups were examined for ethylene production, color change, firmness, and soluble carbohydrate right before treatment and on days 3, 5, 7, and 14 (d1, d3, …).
2.2.1. Ethylene Production
A tomato from groups T1, T2, and CT of known weight was placed in an airtight jar (330 mL) on a lab bench. The sealed jar was subject to an ambient room temperature of 22–24 °C for 60 min, and then, 100 μL of gas was withdrawn from the jar using an airtight syringe and used to calculate ethylene production. The analysis was carried out on a gas chromatograph (GC) (Hewlett-Packard 5890; Hewlett-Packard, Cupertino, CA, USA) equipped with a flame ionization detector (FID) and a capillary column (30 m × 0.25 mm) coated with 5% phenyl methyl silicon (0.25 μm in thickness). Samples were injected under splitless conditions. The GC was programmed at an isothermal temperature of 30 °C, with an injection temperature of 50 °C. The detector was operated at 230 °C. Helium was used as the carrier gas at a 1.5 mL/min column flow. The amount of ethylene produced was calculated against a standard curve, which was obtained with the same instrument and operating conditions, using a known amount of ethylene gas. The production of ethylene at each time point was divided by the fruit weight and expressed as ng/g/h. Each sample was measured in triplicate.
2.2.2. Measurements of Color Change
The change in the color of the tomatoes over time was determined using images taken via photography, and also measured using a color difference meter (ColorQuest XE; HunterLab, Reston, VA, USA). The value of a*, representing the development from the green to the red axis in the CIE color system, was used to determine surface color change during the experiment, based on an average of 20 measurements around the circumference (equatorial diameter) of each tomato.
2.2.3. Firmness Test
The firmness of the tomatoes was measured as resistance to compression using a texture analyzer (Model TA-XT2; Stable Micro System, Godalming, UK). The whole tomato was placed on a stationary steel plate with the stem end down. A 5 kg load cell was used in conjunction with a round-headed probe (P/0.25 S, ¼ spherical stainless) to compress the tomato at a crosshead speed of 3 mm/min. For each tomato, 7 separate determinations of force required to push the probe to a depth of 5 mm were obtained.
2.2.4. Total Soluble Carbohydrate Determination
The total soluble carbohydrates in the tomato juice were determined using a modified phenol–sulfuric acid method [
21]. Tomato juice was centrifuged at 8700 rpm for 10 min at 22–24 °C. A fraction (0.5 mL) of the supernatant and 0.5 mL phenol solution (5%,
w/
v) were added to a round bottom glass tube (10 mL) followed by the addition of 2.5 mL of concentrated sulfuric acid. The tube was immediately capped, mixed well via vortex at the highest speed for 5 sec, and boiled for an additional 15 min. The tubes were cooled down to room temperature before reading the absorbance at 490 nm (UV-2600; Shimadzu, Columbia, MD, USA). Blank samples were prepared via the same procedure using distilled water. The carbohydrate content was calculated against a calibration curve prepared from glucose of different concentrations.
2.2.5. RNA Extraction
RNA sample preparation and sequencing were performed as previously reported [
22]. Briefly, RNA samples were extracted from the pericarps of tomatoes from the T1, T2, and CT groups on days 0, 1, 3, and 7, and RNA extracted using an RNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA). RNA concentrations were measured using a Nanodrop Spectrophotometer (ThermoFisher, Foster City, CA, USA). RNA purity and integrity were determined using the RNA Nano 6000 Assay Kit and an Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA) with an RNA integrity number (RIN) between 9.0 and 9.7 for all samples.
2.3. RNA-Seq and Data Analysis
RNA sequencing and subsequent bioinformatic analysis were performed by Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). Sequencing libraries were constructed using 3 µg of RNA per sample with the NEBNext Ultra RNA Library Prep Kit and quantified using a Qubit 2.0 fluorimeter. The insert sizes were determined using an Agilent 2100 Bioanalyzer. The library preparations were then sequenced on an Illumina Sequencing System (Illumina HiSeq 2000; Illumina, San Diego, CA, USA) and 150 bp paired-end reads were generated. Each sample treatment (T1, T2, and CT groups) was sequenced in duplicate.
After filtering low-quality reads from the raw data, clean reads with a quality score over Q20 were mapped against the S. Lycopersicum genome using TopHat v2.0.12 software. Gene expression levels were determined based on reads per kilobase million (RPKM) of mapped reads. Genes with q-values < 0.05 were considered differentially expressed genes (DEGs). The comparison of DEGs between different groups was carried out using the DESeq R package. The DEGs were used for Gene Ontology (GO) (Version 2.12.) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) (V2.0) enrichment analysis using the KOBAS software.
2.4. Real-Time Quantitative Reverse Transcription Polymerase Chain Reaction (qRT-PCR) Assays
The synthesis of cDNA was performed using an Applied Biosystems GeneAmp PCR System 9700 (ThermoFisher, Foster City, CA, USA) as described previously [
22]. Primers designed using Primer3 (v.0.4.0) software based on the gene sequences are listed in
Supplementary Table S1.
2.5. Statistical Analysis
The Student’s t-test was used to determine whether there were significant difference between samples for all measurements.
4. Discussion
Tomatoes are climacteric fruits characterized by a surge in ethylene biosynthesis at the onset of ripening. Tomato ripening occurs due to the activation of a series of molecular pathways that determine changes in appearance and nutrition, pigmentation levels, the production of volatiles, sweetness and acidity, and the promotion of tissue softening [
34,
35,
36].
Figure 8 illustrates the current understanding of the correlation between ethylene biosynthesis during a tomato’s growth and ripening and its morphophysiological steps. System 1 (S1, auto-inhibitory) is responsible for producing basal ethylene levels during fruit growth, while system 2 (S2, auto-catalytic) operates during climacteric ripening. The pathway of ethylene biosynthesis in both S1 and S2 can be simplified into 3 steps: Step I, Methionine (precursor)
→ S-adenosylmethionine (intermediate, SAM); Step II, SAM
→ 1-aminocyclopropane-1-carboxylate (intermediate, ACC); and Step III, ACC
→ Ethylene. These three steps are catalyzed by methionine adenosyltransferase (MAT), 1-aminocyclopropane-1-carboxylate synthase (ACS), and 1-aminocyclopropane-1-carboxylate oxidase (ACO), respectively [
9,
17,
34]. In the present study, tomato ripening was investigated via RNA-Seq analysis to capture accurate information on the process and mechanism at the molecular and genetic levels, which simple genome sequence analysis cannot provide. Tomato fruits in late mature green stage (MG3, right before the breaker stage) were chosen for this study because at this growth stage, the seeds are mature and ready for dispersal. This development drives the fruit to undergo ripening, and thus, the regulation of ethylene production transitions from system 1 to system 2 [
10]. The tomatoes in this study were subjected to one of two CO
2 conditions, T1 (5% CO
2, 14 d) or T2 (pure CO
2, 3 h), in order to further understand the role of CO
2 stress on the ripening process.
Although 14 ACS genes and 6 ACO genes have been identified in the tomato genome [
17,
36,
37,
38], only 6 ACS and 6 ACO genes exhibited significantly different expression levels in response to CO
2 treatment in the present study (
Figure 5,
Table 2). For the T1 group, all genes encoding for ethylene synthesis and signal transduction were significantly downregulated. These results matched well with the ethylene production profile shown in
Figure 1a, except for ACO4 synthesis. The higher expression level of the gene Solyc12g099000 revealed that the formation of MAT, step I of ethylene biosynthesis, was the driving force of ethylene production for the tomatoes treated with the T1 method. This was consistent with the finding shown in
Figure 4c, where the biosynthesis of amino acids was the most enriched KEGG pathway for tomatoes in the T1 group.
For the T2 group, two important genes encoding for ACS6 and ACS2 were significantly upregulated on d3 and d7, respectively. It is generally accepted that ACS6 is expressed before the onset of tomato ripening and inhibited by ethylene [
10,
11,
39]. In the present experiment, the high expression level of ACS6 at the early stage was accompanied by a large amount of ethylene production (
Figure 1a), implying that another mechanism was used to regulate Solyc08g008100 gene expression. Perhaps the 100% CO
2 increased stress so much that there was subsequent overcompensation in ethylene production upon CO
2 removal. On d7, highly expressed ACS2 carried out the transition from S1 to S2; furthermore, ACS2 in combination with ACS4 promoted peak ethylene production. Ethylene regulates ripening by binding to the ethylene receptor and activating signal transduction pathways [
10,
22]. In comparison with the T1 method, the CO
2 induced by the T2 method further suppressed Solyc09g075440 gene expression and activated the positive regulatory molecule ethylene insensitive 2 (
Table 3); all of these factors together impacted ethylene production during the entire process of tomato ripening.
DNA methylation is carried out by DNA methyltransferases; the level of DNA methylation found here varied depending on the ripening stage of the tomatoes. During the development stage of fruit, the promotor regions of the ACS or ACO genes are hypermethylated, while they are demethylated during the ripening stage [
10,
17]. As reported, transcription starts upstream of the DNA strand adjacent to the promoter sequences; the binding affinity of transcription factors is influenced by chemical modifications of the cytosine or histone groups of DNA [
10,
17] that result in the modulation of gene transcription. As shown in
Table 4, CO
2 induced by T1 treatment activated genes encoding for epigenetic modification on d3 and d7, while the effect of T2 treatment could only be observed on d3. Based on these results, it could be proposed that the introduction of CO
2 using two different treatments resulted in different levels of gene expression due to the methylation and demethylation of DNA promoters, thereby contributing to the inhibition of ethylene production.
The red color of a tomato is indicative of its mature stage, and is due to accumulation of carotenoid metabolites and lycopene in the fruit. The metabolic pathway, which is associated with the degradation of chlorophyll and the transition from xanthophyll to lycopene and carotene, has been studied intensively. A set of genes encoding for the multi-step bioprocess has been verified [
26,
40,
41,
42,
43,
44,
45]. Phytoene formation is the first step of carotenoids biosynthesis. Phytoene is converted into ζ-carotene, and then, to lycopene by desaturase. In the present research, a set of upstream carotenoid enzymes, which catalyze the conversion of geranyl pyrophosphate to phytoene or control the differentiation of organelles, which serve as storage for the carotenoid pigment, were downregulated by T1 treatment on both d3 and d7, but their decrease was only detected on d3 for T2 treatment. Similarly, a difference between T1 and T2 was observed for the downstream genes encoding for carotenoid synthesis enzymes. Carotenoid biosynthesis is subjected to the influence of multiple intrinsic and environmental stimuli. Studies on stimulus-dependent transcriptional regulation have found that light, ethylene, and auxin may play a role in signaling interactions to control tomato carotenoid biosynthesis [
46]. Our results show less of a relationship between the ethylene produced and color change, indicating that the regulation mechanism could have been different even though the stimulus was the same, i.e., CO
2, if the stimulus was delivered via different approaches.
Tomato ripening is accompanied by cell-wall degradation and intercellular adhesion weakening, which result in fruit softening. The CO
2 treatment here, either via the T1 or T2 method, delayed tomato ripening, and thus, downregulated enzyme activities that catalyze cell-wall polysaccharide degradation or disrupt cellulose–hemicellulose association, such as galacturonase, pectinase, xyloxidase, expansin, and pectate lyase, among others [
19,
47]. However, the effect introduced by the T1 method extended to both d3 and d7, while T2 treatment only had an effect on d3, despite the two pectate lyase genes that were upregulated (
Figure 1d,e).
The abiotic stress induced by CO
2 treatment affects many aspects of the tomato lifespan, such as mature processing, protein synthesis, DNA synthesis and repair, signal transduction, metabolism and secondary metabolism, and cell differentiation. From the point of view of strength and duration, the effect of stress induced by T1 was recorded on both d3 and d7, while the effect of T2 was observed on d3 in most cases (
Figure 1b–e and
Figure 4,
Table 4,
Table 5 and
Table 6). However, ethylene synthesis is the exception (
Figure 1a,
Table 2 and
Table 4), where treating the tomatoes with pure CO
2 for only 3 h promoted ethylene production for 12 days via a distinct mechanism that was different from treatment with the T1 method. The treatments with T1 and T2 resulted in distinct gene expression changes, indicating the complexity and inter-dependency of metabolic modification in tomatoes in response to abiotic stress [
48,
49,
50,
51].