Salinity impairs photosynthetic capacity and enhances carotenoid-related gene expression and biosynthesis in tomato (Solanum lycopersicum L. cv. Micro-Tom)

Plant material, growth conditions and salt treatments

Solanum lycopersicum cv. Micro-Tom seeds were imbibed at 4 °C in darkness for 48 h and then sown in plates with Murashige and Skoog culture medium (0.5×). Plates were placed in a growth chamber under controlled conditions for 1 week at 60–80% relative humidity, 24 ± 1 °C, 16/8 h light/dark photoperiod, and 120 mol m⁻² s⁻¹ of light intensity. Thereafter, 7-day-old seedlings were transferred into pots with a peat and vermiculite mixture (2:1), irrigated twice per week with deionized water, and fertilized with 0.5× Hoagland solution (Sigma-Aldrich) for 120 days under the same environmental conditions.

For salt treatments, plants with five to six leaves were irrigated twice per week with 0, 40, 80, 120, and 160 mM NaCl solutions (100 ml of each salt solution per plant individually cultivated in 300 cm³ pot) in addition to Hoagland fertilization, for 14 weeks (until harvest). The experimental design was randomized into a complete block with at least three biological replicates for each treatment.

Plant sampling and phenotypic analysis

Vegetative and reproductive organs of tomato plants exposed to continuous NaCl irrigation were evaluated for 14 weeks. Leaves samples were taken at 2 and 8 weeks of salinity treatment (wst) for analyzing photosynthetic parameters. Fruit analyses were carried out at 11 and 14 wst, which represent the mature-green (at 40 days post-anthesis (dpa)) and red-ripe (at 60 dpa) developmental stages, respectively. For phenotypic and physiological analysis of fruits, the number of clusters, fresh weight, caliber, and total soluble solids were measured. Fruits were also evaluated using the CIEL*a*b* system with a colorimeter that measured L*, a*, and b* values. Briefly, colors are defined by an achromatic component L*, representing relative darkness or lightness, and the chromatic descriptors a* and b*, indicating the red and yellow color intensity, respectively (CIE, 2004).

Photosynthetic parameters

Net photosynthesis (Pn) and stomatal conductance (gs) were measured in the first and second leaves of three randomly selected plants exposed to 2 and 8 wst, using a portable CO₂ infrared gas analyzer (Licor LR6400, Lincoln, NE, USA), equipped with a cuvette that controlled irradiance (600 µmol photons m⁻²s⁻¹), temperature (20 °C), humidity (80%), CO₂ concentration (400 µmol mol⁻¹), and flow rate (300 cm³ min⁻¹).

Leaf chlorophyll fluorescence was measured with a portable pulse amplitude modulated fluorimeter (FMS 2; Hansatech Instruments, Norfolk, UK), in the same leaves used for measuring photosynthetic parameters and previously exposed to darkness for 20 min. Maximum quantum yield of PSII (Fv/Fm), effective quantum yield of PSII (ΦPSII), and electron transport rate (ETR) were estimated according to Genty, Briantais & Baker (1989), whereas non-photochemical quenching (NPQ) was estimated as described by Maxwell & Johnson (2000).

Carotenoid levels analysis

Carotenoids were extracted from 5 g of tomato fruits, considering a mixture of exocarp, mesocarp and endocarp (pericarp of the fruit). Samples were homogenized in a mortar with 1 g of quartz sand and 0.5 g of ascorbic acid. The initial extraction procedure was carried out according to Biacs & Daood (1994), followed by the extraction of carotenoids with 1,2-dichloroethane. Phase and carotenoid separations, as well as carotenoid quantification by high-performance liquid chromatography with diode array detector (HPLC-DAD), were performed as described by Daood et al. (2014). Total carotenoids were measured by spectrophotometry at 474 nm. Specific carotenoids were analyzed using HPLC-DAD equipped with a Hiperst C18 column (Merck). This system was operated with a mobile phase (1 ml min⁻¹), acetonitrile: methanol: methylene chloride (85:10:5 v/v) at room temperature with isocratic conditions. The carotenoids were identified according to their absorption spectra, retention time and comparison with specific pigment standards. Carotenoid levels were expressed in lg/g of dry weight (lg/g DW). Extractions were performed and analyzed in triplicate, on ice and dark conditions. The conversion factor for fresh to dry weight in each experimental condition was determined by desiccation of 1g of fresh weight in an oven at 65 °C for 72 h.

Gene expression analysis

Fruits belonging to the first and second clusters were sampled from each plant, and total RNA was extracted using a CTAB-Spermidine method as described in Reid et al. (2006). RNA was treated with RQ1 DNAse (Promega) and quantified by spectrophotometry using a NanoDrop® ND-1000 spectrophotometer (Thermo Scientific™). For cDNA synthesis, RNA was reversely transcribed using oligo(dT)25 and RevertAid Reverse Transcriptase (Thermo Fisher™), following the manufacturer’s instructions.

The reverse-transcription quantitative real-time PCR (RT-qPCR) analysis was performed using the SensiMix™ Plus SYBR commercial kit (Quantace) and the StepOne™ Real-Time PCR System (Applied Biosystems™, Foster City, CA, USA). Primers for the analyzed carotenoid-pathway genes (Table S1) were designed with the Primer-BLAST Tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Relative gene expression levels were calculated using the ΔΔC_T method (Livak & Schmittgen, 2001), employing three biological replicates. For mRNA levels normalization, the actin gene (Solyc04g011500.3) was used. The expression of this reference gene was stable in our experimental conditions.

Statistical analysis

The data presented in this study are shown as the mean ± SE (n = 3 − 16) for each treatment, depending on the evaluated variable. Differences between treatments were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test, considering P ≤ 0.05.

Photosynthetic capacity is hampered by salt stress in tomato plants

To determine the impact of long-term salt-stress on plant growth and fruit development, 3-week-old S. lycopersicum cv. Micro-Tom plants were exposed to saline irrigation twice per week with 0, 40, 80, 120, and 160 mM NaCl solutions (see Methods). Foliar photosynthetic parameters were evaluated after 2 and 8 weeks of salinity treatment (wst). The net photosynthesis (Pn) rate (as indicated by CO₂ assimilation) decreased with the increase in salt concentration at both sampling dates (Figs. 1A, 1C). Since Pn reduction is known to be partly due to a reduced stomatal conductance (gs), as described in several plant species (Munns, James & Läuchli, 2006; Chaves, Flexas & Pinheiro, 2009), we also determined gs under our conditions. At 2 wst, plants showed a 43% reduction in gs only at 160 mM NaCl, while at 8 wst a decreased gs was observed from 40 mM salt treatment onwards, reaching a 63% reduction at the highest salt-treatment concentration (Figs. 1B, 1D).

To further evaluate the damage of salt stress in the photosynthetic apparatus, maximum quantum yield of PSII (Fv/Fm), effective quantum yield of PSII (ΦPSII), electron transport rate (ETR), and non-photochemical quenching (NPQ) were measured. For Fv/Fm no significant differences between treatment and time were observed (Fig. S1). At 2 wst, ΦPSII and ETR in all salt-treated plants were similar to those in the non-saline control treatment (Figs. 2A and 2C). However, our NPQ analysis shows that 120 and 160 mM NaCl lead to a significant increase of NPQ compared to control plants (Fig. 2E). On the other hand, significant differences were observed at 8 wst in all three measured parameters from 80 mM NaCl upwards (Figs. 2A, 2C and 2E), indicating that the exposure time is an important factor modulating the response to salinity.

Because salt-stress treatments were imposed before the flowering of the earlier tomato fruit clusters, we evaluated the effects of salt in fruit-set and fruit development. The total number of clusters per plant were quantified at flowering and ripening stages. In the first stage (between 8 to 9 wst), only plants treated with 160 mM NaCl showed a significant reduction (by 25%) in the number of clusters per plant relative to non-stressed plants (Fig. S2A). Likewise, in the case of clusters with ripe fruits at 14 wst, plants treated with 120 and 160 mM NaCl showed reductions of 27 and 50% compared to control treatment, respectively (Fig. S2B). No differences in the number of initial flowers from first and second truss between salt-treated and control plants were found, but fruit retention decreased by 48% in 160 mM NaCl treated-plants, measured as the number of fruits that did not abscise at 14 weeks of saline irrigation (Fig. S3).

According to the USDA color classification chart (USDA, 1976), fruit quality was analyzed at two developmental stages. These were defined in control conditions as matured-green (at 40 days post-anthesis (dpa) from plants with 11 wst) and red-ripe (at 60 dpa from plants with 14 wst). For each saline condition tested, matured-green and red-ripe developmental stages were defined at 40 and 60 dpa. We chose to use a dpa-based approach since a color-based strategy could be influenced by the salinity treatments. Since no visual differences based on the color classification chart were observed at 60 dpa between fruits from control or different salt-treatments, we evaluated the color of the fruit surface at the red-ripe developmental stage (60 dpa) using the CIEL*a*b* system based on the Opponent-Color Theory. In this procedure, the color is represented as three values or indexes: L* for the lightness from black (0) to white (100), a* from green (−) to red (+), and b* from blue (−) to yellow (+) (CIE, 2004). Regarding the a* index color parameter (that is indicative of red color intensity), tomato fruits from salt-stressed plants irrigated with 80, 120, and 160 mM NaCl presented higher values than the control plants. No significant differences compared to non-stressed plants were observed when a moderate salt concentration (40 mM NaCl) was used (Fig. 3). The salinity effect on b* index (indicative of yellow color intensity) was similar to the effect observed in a* values. On the contrary, the L* index was only significantly reduced in tomato fruits from plants under 80 mM NaCl treatment (Fig. 3). On the aggregate, these findings indicate that the salinity conditions (80 mM NaCl and over) increased red and yellow pigments in a concentration-dependent manner, suggesting an increase in the carotenoids level.

Compared with the non-treated control plants, fruit fresh weight (FW) significantly decreased (by around 50%) upon salinity treatments in plants irrigated with 120 and 160 mM NaCl. No significant changes in FW of tomato fruits from plants grown under moderate stress were observed (40 mM NaCl; Fig. S4A). Fruit caliber significantly decreased in all salt treatments (Fig. S4B). Contrary to the pattern of variation of FW and fruit diameter, total soluble solid content increased in plants under salt conditions. As shown in Fig. S4C, increases of around 28 and 47% were observed under 40 and 160 mM NaCl treatments, respectively.

Salinity-induced differential expression of carotenoid-related genes and accumulation of metabolites in tomato fruits

To study the changes in carotenoid concentrations, we performed high-performance liquid chromatography (HPLC) analysis of fruits from plants exposed to different salinity treatments in the developmental stages described above. In mature-green fruits (40 dpa, 11 wst), lycopene content significantly increased (by more than 50%) in fruits from plants treated with 120 and 160 mM NaCl compared to fruits from control plants. In red-ripe fruits (60 dpa, 14 wst), a significant increase in lycopene levels was also observed, but from 40 mM NaCl upwards (Fig. 4). In the case of lutein at 40 dpa, all salt treatments resulted in a significant increase, but the greatest lutein concentration reached a peak at 120 mM NaCl. In the red-ripe stage (60 dpa, 14 wst), fruits from salt-treated plants also showed higher lutein concentration but not when 160 mM NaCl was used (Fig. 4). The β-carotene and violaxanthin contents showed significant differences between fruits from salt and control treatments only at 40 dpa (11 wst). In both cases, the higher carotenoid concentration was also reached in fruits from plants exposed to 120 and 160 mM NaCl (Fig. 4). This latter result shows a significant effect of the two most severe salt treatments in different carotenoid contents in tomato fruits. These findings are in agreement with the total carotenoid quantification (Fig. S5) since the highest levels were observed in fruits from plants treated with either 120 or 160 mM NaCl. Overall, increasing salinity concentration in our experimental conditions led to an increase in total carotenoid levels as well as the concentration of β-carotene and lycopene in mature-green (40 dpa) and red-ripe (60 dpa) tomatoes, respectively. Since this data was normalized to tomato dry weight (DW), our findings suggest that the increase in carotenoid levels might be caused by a plant physiological or stress response, rather than being the result of a decreased water uptake by fruits.

As a starting point to understand this saline effect, we tested the gene expression of key carotenoid-pathway genes of tomato, such as phytoene synthase 1 (PSY1, Solyc03g031860), phytoene synthase 2 (PSY2, Solyc02g081330), phytoene desaturase (PDS, Solyc03g123760), ζ-carotene desaturase (ZDS, Solyc01g097810), lycopene β-cyclase (LYCB, Solyc04g040190), lycopene ε-cyclase (LYCE, Solyc12g008980) and carotenoid isomerase (CRTISO, Solyc10g081650), in order to determine whether changes in carotenoid content of tomato fruits upon salt stress was the result of a plant transcriptional response to high salt level exposure. In green fruits (40 dpa, 11 wst), PSY1, PDS, and ZDS showed a higher transcript accumulation in response to 160 mM NaCl, but no differences between control and lower salt-treated plants were observed (Fig. 5). Conversely, PSY1, PDS, and ZDS transcripts decreased in fruits from plants exposed to severe salt concentrations (120 and 160 mM) in red-ripe tomatoes (60 dpa, 14 wst). In the case of LYCB, its expression increased from 80 mM upwards compared to control treatment in green fruits (40 dpa, 11 wst), while in red-ripe fruits (60 dpa, 14 wst) an increased expression level was only observed in fruits from plants treated with 80 mM of salt (Fig. 5). Salinity treatment did not affect the gene expression of LYCE, CRTISO, and PSY2 (Fig. S6).

Since all these genes are differentially expressed depending on the stage of fruit development, we analyzed the ripening-related ethylene genes 1-aminocyclopropane-1-carboxylate oxidase 1 (ACO1, Solyc07g049530), 1-aminocyclopropane-1-carboxylic acid synthase 2 (ACS2, Solyc01g095080), ethylene-inducible CTR1-like protein kinase (CTR1, Solyc10g083610), CTR1-like protein kinase 4 (CTR4, Solyc10g085570) and three ethylene receptors (ETR3, Solyc09g075440; ETR4, Solyc06g053710; ETR6, Solyc09g089610) in mature-green (40 dpa, 11 wst) and red-ripe (60 dpa, 14 wst) fruits from control and salt-stressed tomato plants. As observed in Fig. S7, no significant differences were determined in the gene expression of these ripening-related genes upon the salinity treatment. These results suggest that the salinity treatment has an influence on the carotenoid pathway in tomato fruit but no impact on the major ethylene-related ripening genes that modulate diverse maturation processes.