Effects of Combined Treatment with Gibberellin and Prohydrojasmon on Volatile and Dimethyl Sulfide Content in Juice Sacs of Satsuma Mandarin (Citrus unshiu Marc.) Fruit During Maturation

Abstract

Combined treatment with gibberellin (GA) and prohydrojasmon (PDJ) is effective at reducing peel puffing in satsuma mandarin (Citrus unshiu Marc.). However, the effect of this treatment on volatiles remains unknown. In satsuma mandarin, dimethyl sulfide (DMS) is a key flavor compound. In this study, the effects of combined treatment with GA and PDJ on the content of volatiles, DMS, and S-methylmethionine (SMM), a precursor of DMS, in the juice sacs of satsuma mandarin were investigated. Mixtures of GA (1, 3.3, or 5 mg·L⁻¹) and PDJ (25 or 50 mg·L⁻¹) were applied to ‘Okitsu-wase’ and ‘Aoshima’ in mid-September. Gas chromatography-mass spectrometry analysis showed that among the volatiles, DMS content was reduced by the treatment, and was below the DMS odor threshold in several treatments. In ‘Okitsu-wase’, the DMS content in treated fruit was 1/3 to 1/17 of the control. In the treatment with GA and 50 mg·L⁻¹ PDJ, a reduction in DMS content was observed until mid-November, irrespective of GA concentration, whereas in the treatment with 1 mg·L⁻¹ GA and 25 mg·L⁻¹ PDJ, the reduction was observed only until early November. In ‘Aoshima’, the DMS content was also reduced by the treatments, but in the treatment with 1 mg·L⁻¹ GA and 50 mg·L⁻¹ PDJ, the reduction was observed only until mid-November. Although the applied treatments delayed color development by approximately 1–2 weeks, this delay was not related to a reduction in DMS content. The effects on SMM and methionine content were small or negligible. As DMS has a strong odor, with trace amounts affecting fruit flavor, controlling DMS content is important. This study revealed that combined treatment with GA and PDJ reduced the accumulation of DMS in the juice sacs of satsuma mandarin fruits.

Introduction

Gibberellin (GA) and prohydrojasmon (PDJ) are growth regulators (Koshiyama et al., 2003; McAtee et al., 2013). In satsuma mandarin (Citrus unshiu Marc.), Kuraoka et al. (1966) reported that spraying GA was effective in reducing peel puffing (physiological disorder); however, further improvements were necessary for practical use because the treatment strongly delayed color development. Recent studies revealed that combined spraying of GA and PDJ was effective in reducing peel puffing, and the delay in color development (1–2 weeks) was acceptable in terms of practical use (Makita and Yamaga, 2004; Sawano, 2010; Sato et al., 2015). The studies also showed that the effects on soluble solid content (SSC) and acidity were small or negligible in most cases. Therefore, the combined spraying of GA and PDJ has been widely used in the cultivation of satsuma mandarin. We noticed that the flavor of fruit treated with GA and PDJ was different from the control; the flavor of the treated fruit was young and fresh, whereas that of the control was deep and ripe, although differences in SSC and acidity were negligible.

Volatiles are essential for citrus fruit flavor. Among the volatile compounds, dimethyl sulfide (DMS) is one of the key compounds responsible for producing off-flavors, which arise from the thermal processing of satsuma mandarin juice (Manabe, 1975; Sawamura et al., 1977; Cheng et al., 2020). Volatile sulfur compounds (e.g., DMS) have a strong odor and contribute to the agreeable and disagreeable flavors of food (Shankaranarayana et al., 1974). DMS has a cabbage-like or seaweed-like odor, with an extremely low odor threshold (Guadagni et al., 1963). As a result, trace levels of DMS alter the flavor of fruits and vegetables (Scherb et al., 2009; Morisaki et al., 2014; Cannon and Ho, 2018). In satsuma mandarin fruit, Kwak et al. (1990) reported that the DMS content in the central cavity of the fruit was associated with off-flavor during storage, and the content increased at high temperatures but not at low temperatures. Cheng et al. (2020) reported that DMS content in the unheated juice from satsuma mandarin cultivars was markedly higher than in juice from other mandarin cultivars, and the DMS content in the satsuma mandarin fruit has odor activity as the content is above the odor threshold for DMS.

DMS is formed from the decomposition of its precursor, S-methylmethionine (SMM), and heating of satsuma mandarin juice stimulates the reaction (Sawamura et al., 1978). To suppress the emergence of DMS in juice processing, methods to reduce SMM content, such as ion-exchange of juice and low-temperature storage of fruit, have been studied (Osajima et al., 1985; Kwak et al., 1990). However, methods to reduce the accumulation of DMS during the maturation of satsuma mandarin fruit have not been elucidated.

Recently, the effect of growth regulator treatment on volatiles during cultivation has been studied in fruits and crops such as grapes and beans. Treatment with GA, cytokinin, and cytokinin-like compounds affected the metabolism of volatiles and altered their content and composition in grapes with a potential impact on flavor (Wang et al., 2017, 2020; Tyagi et al., 2021). Treatment with PDJ increased the content of aldehydes, esters, and terpenes in grape (Wang et al., 2015), and increased esters and terpenes in lima beans (Uefune et al., 2014).

Considering these results, we think that combined treatment with GA and PDJ may also affect the content and composition of volatiles in satsuma mandarin fruit. In fact, we noticed that the flavor of the treated fruit and the control was different. However, the effect of combined treatment with GA and PDJ on volatiles in satsuma mandarin fruit has not been studied. In the present study, the effect of combined treatment with GA and PDJ on changes in the content of volatiles, DMS, and its precursors (SMM and Met) in the juice sacs of satsuma mandarin fruit was investigated during maturation of ‘Okitsu-wase’ and ‘Aoshima’ satsuma mandarin fruit.

Materials and Methods

Plant materials and treatments

‘Okitsu-wase’ and ‘Aoshima’ satsuma mandarin trees (C. unshiu Marc.), grafted onto trifoliate orange (Poncirus trifoliate (L.) Raf.) at the NARO Institute of Fruit Tree and Tea Science, Citrus Research Division, Okitsu (Shizuoka, Japan) were used for this study. The combined treatment with GA and PDJ was conducted in 2015 and 2016. Agrochemicals GA₃ (3.1% granule) and PDJ (5% solution) obtained from Meiji Seika Pharma Co., Ltd. (Tokyo, Japan) were used for the treatments. On each tree, two branches were selected, of which one branch was used for a treatment (single concentration) and one branch was used as a control. A mixture of GA and PDJ in water was prepared and sprayed onto the fruits on a branch using a hand sprayer. Four trees were used per treatment. Two to three fruits of uniform size were collected from each branch. In 2015, ‘Okitsu-wase’ was used and three combined treatments with different concentrations were conducted on September 15. The three treatments were conducted as independent experiments for each concentration, using different trees. The concentrations of each experiment were as follows: 1 mg·L⁻¹ GA and 25 mg·L⁻¹ PDJ, 1 mg·L⁻¹ GA and 50 mg·L⁻¹ PDJ, 3.3 mg·L⁻¹ GA and 50 mg·L⁻¹ PDJ. Fruit was collected on October 21, November 4, 11, and 18. In 2016, both ‘Okitsu-wase’ and ‘Aoshima’ were used. In ‘Okitsu-wase’, the concentration of PDJ was set at 50 mg·L⁻¹ and three combined treatments with different concentrations of GA (1, 3.3, and 5 mg·L⁻¹) were conducted on September 15. The three treatments were conducted as independent experiments, as in 2015. Fruit was collected on October 19 and 26, November 2, 9, and 14. In ‘Aoshima’, the concentration of PDJ was set at 50 mg·L⁻¹ and two combined treatments with different concentrations of GA (1 and 3.3 mg·L⁻¹) were conducted on September 16. The two treatments were conducted as independent experiments, as in 2015. Fruit was collected on November 22 and 28 and December 5. In both years, the treatments were conducted in September because previous studies had reported that treatment in September was effective in reducing peel puffing and the delay in color development was acceptable in terms of practical use (Makita and Yamaga, 2004; Sawano, 2010; Sato et al., 2015).

Measurement of color, peel puffing, soluble solids content, and acidity

Peel color was assessed objectively on the basis of the ratio of orange colored area on the fruit surface and expressed as peel color score (0–10). Peel color was further measured by a colorimeter (NF-333; Nippon Denshoku Industries Co., Ltd., Tokyo, Japan) at two locations around the equatorial plane of each fruit, and was expressed as the Hunter a/b ratio (Stewart and Wheaton, 1971). The a/b ratio is negative for green fruit, approximately zero for yellow fruit (color break stage), and positive for orange colored fruit. The degree of peel puffing was evaluated objectively and expressed as peel puffing score: none (0), low (1), intermediate (2), and expensive (3), as reported previously (Kawase, 1987; Imai et al., 2017). The results are presented as the means of 8–12 fruit derived from four trees per treatment.

Soluble solids content (SSC) and acidity were measured in the juice, which was prepared by blending 2–3 fruit collected from each branch. SSC was determined using a PAL-1 digital refractometer (Atago Co., Ltd., Tokyo, Japan). Acidity was measured by titration with 0.1 M NaOH using a phenolphthalein indicator. The results are presented as the means of four biological replicates (four trees).

Analysis of aroma volatiles

To analyze the volatiles in the juice sacs, fruit were peeled carefully to avoid contamination with peel oil. The juice sacs were separated from the 2–3 fruit collected per branch, blended, immediately frozen in liquid nitrogen, and stored at −80°C until use. The frozen juice sacs were homogenized in liquid nitrogen, and the aliquots (1.0 g) were placed in a headspace vial (volume size, 10 mL). As an internal standard, 5 μL of 1-pentanol (1:1000 v/v in water) was firstly added, and then 1 mL of a saturated aqueous solution of sodium chloride was added. The vial was sealed immediately and stored at −80°C until analysis. The volatiles were manually extracted from the headspace of the vials through solid-phase microextraction (SPME) using StableFlex fiber coated with a 50/30 μm layer of Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS), supplied by Supelco (Bellefonte, PA, USA). Before each extraction, the fiber was conditioned for 5 min at 250°C using a gas chromatograph injector. Before analysis, the samples stored at −80°C were thawed in a water bath at 40°C and equilibrated at 40°C for 5 min. Subsequently, to extract volatiles, the fiber was exposed to the headspace of the sample vial for 10 min at 40°C. After extraction, the fiber was desorbed in a split/splitless injector at 250°C for 1 min and analyzed using gas chromatography-mass spectrometry (GC-MS).

GC-MS analysis was performed using a Shimadzu GC2010 coupled to a model QP2010 mass spectrometer (Shimadzu Corp., Kyoto, Japan). The aroma volatiles were separated on an Agilent HP-5 capillary column (30 m × 0.25 mm i.d., 0.5 μm film thickness) (J&W Scientific Inc., Folsom, CA, USA). The column oven temperature was held at 40°C for 5 min, then increased to 190°C at a rate of 5°C·min⁻¹, then increased again to 250°C at a rate of 20°C·min⁻¹, and held at 250°C for 3 min. The helium carrier gas was set at 0.81 mL·min⁻¹. The mass detector was in electron impact (EI) mode at 70 eV with a source temperature of 200°C, interface temperature of 250°C, and mass scan range of m/z 35–350.

The aroma volatiles were identified by comparing the GC-MS spectrum and retention index (RI) of each compound with those in the US National Institute of Standards and Technology (NIST) Mass Spectral Library and Chemistry WebBook. The RI for each compound was calculated using the retention characteristics of n-alkanes (C5–C20). Acetaldehyde, ethanol, dimethyl sulfide (DMS), ethyl acetate, hexenal, α-pinene, β-pinene, p-cymene, limonene, γ-terpinene, terpinolen, linalool, nonanal, terpinen-4-ol, α-terpineol, decanal, and valencene were identified by analyzing authentic chemical standards. Except for DMS, the content of each volatile compound was represented as the peak area ratio (analyte peak area/internal standard peak area × 10³). The peak area was determined using the corresponding quantification ion (QI) for each compound in the extracted-ion chromatogram. The QI values for each compound are presented in Table 1. The QI for the internal standard (1-pentanol) was m/z 55, and that for DMS was m/z 62.

Table 1

Effect of combined treatment with GA and PDJ on relative content of volatiles during maturation of ‘Okitsu-wase’ in 2015.

Quantification of DMS was conducted by comparing the peak area ratio (DMS/internal standard) with that of authentic standards. The defined amount of DMS was dissolved in ethanol, and 5 μL of each solution with different concentrations was mixed with a synthetic citrus juice solution (7% sucrose, 1.3% glucose, 2% fructose, and 1% citric acid in water), and then used as a dilution series of the standard solution. The DMS calibration curve was determined in the range of 0.925–185 μg·L⁻¹. The results are presented as the means of four biological replicates (four trees). Each replicate was prepared from 2–3 fruit, that is, a total of 8–12 fruit per treatment.

S-Methylmethionine and amino acid analysis

The S-methylmethionine (SMM) and 20 amino acids (Ala, Arg, Asn, Asp, GABA, Gln, Glu, His, Ile, Leu, Lys, Met, ornithine, Phe, Pro, Ser, Thr, Trp, Tyr, Val) in the juice sacs were extracted using 80% ethanol and quantified using LC/MS/MS (AB SCIEX API2000 triple-stage quadrupole tandem mass spectrometer) with a TurboIon spray source (AB SCIEX, Foster City, CA, USA) as described previously (Matsumoto and Ikoma, 2012). A multiple-reaction monitoring (MRM) mode was used for quantification, using an internal standard method. The MRM transitions (m/z) of SMM and its internal standards, SMM-d3, were as follows: precursor-product ion, 164-102 for SMM and 167-102 for SMM-d3. The total amino acid content was the sum of the 20 amino acids. The results are presented as the mean of four biological replicates (four trees, a total of 8–12 fruit per treatment).

Statistical analysis

Statistical significance was analyzed using a t-test at the 5% level, using the application software JMP 8 (JMP release 8.0; SAS Institute Inc., Cary, NC, USA).

Results and Discussion

Effect of combined treatment with GA and PDJ on color, peel puffing, SSC, and acidity

In 2015, ‘Okitsu-wase’ was used, and three combined treatments with 1 mg·L⁻¹ GA and 25 mg·L⁻¹ PDJ (1–25 treatment), 1 mg·L⁻¹ GA and 50 mg·L⁻¹ PDJ (1–50 treatment), and 3.3 mg·L⁻¹ GA and 50 mg·L⁻¹ PDJ (3.3–50 treatment) were conducted as independent experiments for each concentration, using different trees. For all the treatments, the peel color score and Hunter a/b ratio in the treated fruit were lower than in the control throughout the experimental period (Fig. 1). The peel color score or the a/b ratio of the control (e.g., October 21) were almost similar to the treated fruit after two weeks (e.g., November 4). Thus, these treatments delayed color development by approximately two weeks. Peel puffing was not observed until November 11 in all treatments. However, on November 18, peel puffing was observed only in the control, and the degree of peel puffing (0.13–0.17) was higher in the control than in the treated fruit (0) in all treatments.

Fig. 1

Effect of combined treatment with GA and PDJ on peel color score and Hunter a/b ratio during maturation of ‘Okitsu-wase’ and ‘Aoshima’. Concentrations of the combined treatments are described as follows: 1–25, 1 mg·L⁻¹ GA + 25 mg·L⁻¹ PDJ; 1–50, 1 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ; 3.3–50, 3.3 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ and 5–50, 5 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ. Each value indicates the mean ± SE (n = 8–12 fruits). * Indicates significant differences at P < 0.05 by t-test.

In 2016, both ‘Okitsu-wase’ and ‘Aoshima’ were used, and combined treatments with GA (1, 3.3, or 5 mg·L⁻¹) and PDJ (50 mg·L⁻¹) were conducted as in 2015. As observed in 2015, the peel color score and a/b ratio in the treated fruit were lower than in the control in all treatments (Fig. 1). In ‘Okitsu-wase’, the peel color score and a/b ratio of the control were almost similar to those of the treated fruit after 1–2 weeks in the 1–50 treatment, and after two weeks in the 3.3–50 and 5–50 treatments, respectively. In ‘Aoshima’, the peel color score and a/b ratio of the control were almost the same as the treated fruit after one week in the 1–50 treatment and after more than two weeks in the 3.3–50 treatment, respectively (Fig. 1). Thus, these treatments delayed color development approximately 1–2 weeks in both cultivars. In ‘Aoshima’, the degree of peel puffing on December 5 was higher in the control (1.25–1.38) than in the treated fruit (0.13–0.25). With respect to the SSC and acidity, differences between the control and treated fruit were not observed in any treatment in both 2015 and 2016.

Combined treatment with GA and PDJ is effective in reducing peel puffing in satsuma mandarin fruit, while the treatment delays fruit color development during maturation for more than one week, although the effects on SSC and acidity are generally small or negligible (Makita and Yamaga, 2004; Sawano, 2010; Sato et al., 2015). As reported in previous studies, the treatment in the present study also delayed color development by approximately 1–2 weeks and reduced the degree of peel puffing without any clear changes in SSC or acidity.

Effect of combined treatment with GA and PDJ on dimethyl sulfide and volatiles contents during maturation

The changes in the volatiles contents and dimethyl sulfide (DMS) in the juice sacs were analyzed using headspace SPME GC-MS analysis. In the three experiments conducted in 2015, 30 volatile compounds, including aldehydes, alcohols, esters, terpenes and DMS, were detected in the juice sacs of ‘Okitsu-wase’ (Table 1; Fig. 2). Among the volatiles, the DMS content was reduced by the treatments (Fig. 2). In the 1–50 and 3.3–50 treatments, the DMS content in the treated fruit remained at low levels (0.37–0.97 μg·kg⁻¹), and was significantly lower than in the control (1.23–2.96 μg·kg⁻¹) during the experimental period until mid-November. The content of the control was 3- to 5-fold higher than in the treated fruit. In contrast, in the 1–25 treatment, the DMS content in the treated fruit collected on October 21 and November 4 (0.61 and 0.7 μg·kg⁻¹) was significantly lower than in the control (2.0 and 1.8 μg·kg⁻¹), and the content in the control was 3-fold that in the treated fruit. However, after November 11, the DMS content in the treated fruit increased, and the difference in the content between the control and treated fruit became insignificant.

Fig. 2

Effect of combined treatment with GA and PDJ on DMS content during maturation of ‘Okitsu-wase’ and ‘Aoshima’. Concentrations of the combined treatments are described as follows: 1–25, 1 mg·L⁻¹ GA + 25 mg·L⁻¹ PDJ; 1–50, 1 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ; 3.3–50, 3.3 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ and 5–50, 5 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ. Each value indicates the mean ± SE (n = 4 trees). * Indicates significant differences at P < 0.05 by t-test.

In the three experiments conducted in 2016, similar to in 2015, 30 volatile compounds were detected in the juice sacs of ‘Okitsu-wase’ (Table 2; Fig. 2). Among the volatiles, DMS content was reduced by the treatments. The DMS content (0.09–0.45 μg·kg⁻¹) in the treated fruit was significantly lower than in the control (0.97–2.15 μg·kg⁻¹) during the entire experimental period until mid-November (Fig. 2). The content in the control was 3- to 4-fold that in the treated fruit in the 1–50 treatment, 5- to 7-fold in the 3.3–50 treatment, and 4- to 17-fold in the 5–50 treatment, respectively.

Table 2

Effect of combined treatment with GA and PDJ on relative content of volatiles during maturation of ‘Okitsu-wase’ in 2016.

In the juice sacs of ‘Aoshima’, 27 volatile compounds were detected in the two experiments conducted in 2016 (Table 3; Fig. 2). Among the volatiles, as observed in ‘Okitsu-wase’, the DMS content was decreased by the treatments. In the 3.3–50 treatment, the content in the treated fruit (0.02–0.09 μg·kg⁻¹) was significantly lower than in the control (0.56–0.89 μg·kg⁻¹) during the entire experimental period (Fig. 2). The content in the control was 10- to 23-fold higher than in the treated fruit. In contrast, in the 1–50 treatment, the DMS content in the treated fruit was lower than in the control on November 22; however, thereafter, the content increased, and no difference was observed after November 28.

Table 3

Effect of combined treatment with GA and PDJ on relative content of volatiles during maturation of ‘Aoshima’.

Regarding the other volatiles, in ‘Okitsu-wase’, the content of ethanol and ethyl esters (ethyl acetate, ethyl propanoate, and ethyl 2-methyl butanoate), which have a fruity and sweet odor, was higher in the treated fruit than in the control at several time points in each treatment (Tables 1 and 2). The content of hexanal, which has a green and glassy odor, was higher in the treated fruit than in the control in the 1–50 and 3.3–50 treatments in 2015. In ‘Aoshima’, differences in ethanol and ethyl esters levels were not observed, whereas hexanal content was higher in the treated fruit than in the control at several time points (Table 3).

These results revealed that combined treatment with GA and PDJ in September affected the content and composition of volatiles, and especially reduced the DMS content in the juice sacs of the satsuma mandarin fruit during maturation. In ‘Okitsu-wase’, the combined treatment with GA (1, 3.3, or 5 mg·L⁻¹) and PDJ (50 mg·L⁻¹) was effective in reducing DMS content from late October to mid-November. However, in the combined treatment with 1 mg·L⁻¹ GA and 25 mg·L⁻¹ PDJ, reduction in DMS content was observed only until early November. In ‘Aoshima’, the combined treatment with GA (1 or 3.3 mg·L⁻¹) and PDJ (50 mg·L⁻¹) was also effective in reducing DMS content. However, in the 1–50 treatment, reduction in DMS content was only observed until mid-November.

The odor threshold of DMS is low (0.33 μg·kg⁻¹) (Cheng et al., 2020), and trace amounts of DMS strongly affect the flavor of many foods (Cannon and Ho, 2018). We noticed that the flavor of GA-PDJ-treated fruit was different from that of untreated fruit; the flavor of treated fruit was young and fresh while the flavor of untreated fruit was deep and ripe. In the present study, as shown in Figure 2, the DMS content of the control was above the odor threshold in all treatments in ‘Okitsu-wase’ (1.23–2.96 μg·kg⁻¹ in 2015, 0.97–2.15 μg·kg⁻¹ in 2016) and ‘Aoshima’ (0.56–0.89 μg·kg⁻¹). In contrast, in 2016, in ‘Okitsu-wase’, the content in most treated fruit was below the odor threshold (0.09–0.32 μg·kg⁻¹), with the exception of the DMS content on November 14 (0.45 μg·kg⁻¹) in the 1–50 treatment. In ‘Aoshima’, the DMS content in all treated fruit was below the odor threshold (0.05–0.26 μg·kg⁻¹). In 2015, in ‘Okitsu-wase’, the content in the treated fruit (0.37–0.97 μg·kg⁻¹ in 1–50 and 3.3–50 treatment) was higher than the odor threshold but lower than in the control (1.23–2.96 μg·kg⁻¹). Therefore, differences in the DMS content as a result of the treatments may contribute to the difference in flavor between the treated and untreated fruit. However, it is difficult to determine whether the reduction in DMS content during maturation is beneficial for the flavor, because flavor preference differs from person to person.

The effect of phytohormones on aroma production has been widely studied in relation to ethylene; however, little is known about the effects of other growth regulators on the regulation of volatile production in fruits and crops (McAtee et al., 2013). Recent studies have reported that treatment with PDJ during cultivation increased the content of aldehydes, esters, and terpenes in grapes and bean plants (Uefune et al., 2014; Wang et al., 2015). Treatment with growth regulators, GA, cytokinin, and synthetic cytokinin-like compounds reduced the content of several volatiles and affected the volatile composition of some grape cultivars (Wang et al., 2017, 2020; Tyagi et al., 2021). However, the effect of growth regulator treatment on DMS content has scarcely been reported in plants, even though a trace amount of DMS strongly affects the flavor of fruit. The results of the present study revealed that combined treatment with GA and PDJ affects the content and composition of volatiles such as ethyl esters and hexanal, and especially reduces the accumulation of DMS in the juice sacs of satsuma mandarin fruit.

Relationship between the delay in color development and DMS content

Combined treatment with GA and PDJ delayed color development by approximately 1–2 weeks in this study (Fig. 1). To examine whether delay in color development is related to the reduction in DMS content, the DMS content of the control and the treated fruit with similar peel color (Hunter a/b ratio) were compared. In ‘Okitsu-wase’, in 2016, the peel color of the treated fruit on November 9 or 14 was similar to the control on November 2 (Fig. 1). However, the DMS content of the treated fruit on November 9 or 14 was significantly lower than the control on November 2 in all treatments (Fig. 2). Likewise, in 2015, the DMS content of the treated fruit on November 4 or 11, with a peel color similar to the control on October 21, was significantly lower than the control on October 21 in all treatments (Figs. 1 and 2). In ‘Aoshima’, the DMS content was not evaluated because the delay in color development was long in the 3.3–50 treatment, and control and treated fruit with similar peel color were not obtained. Thus, even if the peel color of treated fruit was similar to that of the control, the DMS content would be lower in treated fruit than in the control. These results show that the delay in color development caused by the treatment is not related to the reduction in DMS content; however, the combined treatment reduces the accumulation of DMS during maturation.

Effect of combined treatment with GA and PDJ on SMM, Met, and amino acid content

To examine why combined treatment with GA and PDJ reduced DMS content, the effect of the treatment on the content of SMM and Met, the precursors of DMS, was analyzed in the juice sacs. In ‘Okitsu-wase’, in 2015, SMM content in most treated fruit in the 1–50 treatment was lower than in the control (Fig. 3). However, the difference was small, and the SMM content in the control was 1.1- to 1.2-fold that in the treated fruit. In the 1–25 and 3.3–50 treatments, the SMM content was lower in the treated fruit than in the control, but was significant only at a few time points (Fig. 3). In 2016, the SMM content was slightly lower in the treated fruit than in the control at several time points in the 3.3–50 and 5–50 treatments; however, the difference was unclear in the 1–50 treatment (Fig. 3). The Met content, the precursor to SMM, was lower in the treated fruit than in the control, but was significant only at a few time points in both years (Fig. 4). In ‘Aoshima’, differences in SMM and Met contents were unclear (Figs. 3 and 4). These results showed that the effect of combined treatment with GA and PDJ on the SMM and Met content was small or negligible in satsuma mandarin fruit. Reproducible changes in the content of other amino acids as a result of the treatments were not observed in either cultivar (Tables S1, S2, and S3).

Fig. 3

Effect of combined treatment with GA and PDJ on SMM content during maturation of ‘Okitsu-wase’ and ‘Aoshima’. Concentrations of the combined treatments are described as follows: 1–25, 1 mg·L⁻¹ GA + 25 mg·L⁻¹ PDJ; 1–50, 1 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ; 3.3–50, 3.3 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ and 5–50, 5 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ. Each value indicates the mean ± SE (n = 4 trees). * Indicates significant differences at P < 0.05 by t-test.

Fig. 4

Effect of combined treatment with GA and PDJ on Met content during maturation of ‘Okitsu-wase’ and ‘Aoshima’. Concentrations of the combined treatments are described as follows: 1–25, 1 mg·L⁻¹ GA + 25 mg·L⁻¹ PDJ; 1–50, 1 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ; 3.3–50, 3.3 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ and 5–50, 5 mg·L⁻¹ GA + 50 mg·L⁻¹ PDJ. Each value indicates the mean ± SE (n = 4 trees). * Indicates significant differences at P < 0.05 by t-test.

DMS is formed from SMM via enzymatic and nonenzymatic reactions. In an aqueous solution, SMM is chemically degraded into DMS, and heating and pH affect this reaction (Williams and Nelson, 1974; Sawamura et al., 1978). In the present study, the effect of chemical degradation was negligible because SMM degradation was not observed under the analytical conditions of the present study, and differences in the pH and acidity of the juice sacs were negligible between the control and treated fruit. In higher plants, SMM is split into DMS and homoserine by methionine sulfonium lyase (SMM hydrolase) found in cabbage leaf tissue and onion seedlings (Lewis et al., 1971; Hattula and Granroth, 1974). In the present study, the reduction effect on SMM was small or negligible (content in the control was 1- to 1.3-fold that in the treated fruit), whereas that on DMS was clear (the content in the control was 3- to 23-fold that in the treated fruit) in ‘Okitsu-wase’ and ‘Aoshima’ (Figs. 2 and 3). Considering these results, it seems that combined treatment with GA and PDJ may primarily reduce the biosynthesis of DMS from SMM, and consequently, reduce the accumulation of DMS. To elucidate the mechanisms underlying the reduction in DMS, further studies on enzyme activities and mRNA levels of DMS biosynthesis are required.

In conclusion, the present results show that combined treatment with GA and PDJ affects the content and composition of volatiles, and in particular, reduces the DMS content in the juice sacs of the satsuma mandarin fruit during maturation. These treatments delayed color development by approximately 1–2 weeks; however, this delay was not related to the reduction in DMS content. The reduction effect on the SMM and Met content, precursors to DMS, was much lower or negligible in comparison to that on DMS. To the best of our knowledge, a method to reduce DMS content by treatment with growth regulators has not been reported for fruits and vegetables. As DMS has a low odor threshold and trace amounts affect fruit flavor, controlling the DMS content is important. The present study demonstrated a method to control DMS content in the satsuma mandarin fruit during maturation. As DMS is known to contribute to off-flavor during storage, a treatment that reduces DMS during maturation and maintains a low DMS content during storage may be useful in preventing off-flavor during storage. Further studies are required regarding the storage period.

Acknowledgements

We wish to acknowledge Dr. Hiroshi Fujii and Dr. Fumitaka Takishita for helpful discussions and comments on the statistical analysis of the manuscript.

Literature Cited

•  Cannon,  R. J. and  C.-T.  Ho. 2018. Volatile sulfur compounds in tropical fruits. J. Food Drug Anal. 26: 445–468.
•  Cheng,  Y.,  R.  Rouseff,  G.  Li and  H.  Wu. 2020. Methanethiol, an off-flavor produced from the thermal treatment of mandarin juices: a study of citrus sulfur volatiles. J. Agric. Food Chem. 68: 1030–1037.
•  Guadagni,  D. G.,  R. G.  Buttery and  S.  Okano. 1963. Odour thresholds of some organic compounds associated with food flavours. J. Sci. Food Agric. 14: 761–765.
•  Hattula,  T. and  B.  Granroth. 1974. Formation of dimethyl sulfide from S-methylmethionine in onion seedlings (Allium cepa). J. Sci. Food Agric. 25: 1517–1521.
•  Imai,  A.,  T.  Yoshioka and  T.  Hayashi. 2017. Quantitative trait locus (QTL) analysis of fruit-quality traits for mandarin breeding in Japan. Tree Genet. Genomes 13: 79.
• Kawase, K. 1987. Kajitsuchousahou. p. 5–12. In: K. Kawase (ed.). Kankitsunochousahouhou (In Japanese). Kajusikenjou Okitsu, Shizuoka.
•  Koshiyama,  M.,  H.  Seto,  Y.  Kamuro,  K.  Watanabe,  M.  Mitomi and  K.  Imamura. 2003. A jasmonic acid analog, PDJ, comes into practical use as a plant growth regulator. Regulation of Plant Growth & Development 38. DOI: 10.18978/jscrpanb.38.0_47 (In Japanese with English abstract).
• Kuraoka, T., K. Iwasaki and K. Kadoya. 1966. Studies on peel puffing of Satsuma mandarin (Citrus unshiu Marc.). V. Effect of GA3 application. Abstr. Japan. Soc. Hort. Sci. Spring Meet.: 43–44 (In Japanese).
•  Kwak,  S.,  Y.  Ueda,  H.  Kurooka and  H.  Yamanaka. 1990. Off-odor formation during storage of Satsuma mandarin (Citrus unshiu Marc.) fruit in relation to sulfur compounds. Nippon Shokuhin Kogyo Gakkaishi 37: 799–804 (In Japanese with English abstract).
•  Lewis,  B. G.,  C. M.  Johnson and  T. C.  Broyer. 1971. Cleavage of Se-methylselenomethionine selenonium salt by a cabbage leaf enzyme fraction. Biochim. Biophys. Acta 237: 603–605.
•  Makita,  Y. and  I.  Yamaga. 2004. Autumn sprays of gibberellin and prohydrojasmon mixtures reduce the incidence of puffing rind in satsuma mandarin. J. Japan. Soc. Hort. Sci. 73 (Suppl. 2): 106 (In Japanese).
•  Manabe,  T. 1975. Relation between low-boiling components and off-flavor in heat-induced volatile gases of satsuma orange (Citrus unshiu Marc.) juice. Nippon Shokuhin Kogyo Gakkaishi 22: 337–342 (In Japanese with English abstract).
•  Matsumoto,  H. and  Y.  Ikoma. 2012. Effect of different postharvest temperatures on the accumulation of sugars, organic acids, and amino acids in the juice sacs of Satsuma mandarin (Citrus unshiu Marc.) fruit. J. Agric. Food Chem. 60: 9900–9909.
•  McAtee,  P.,  S.  Karim,  R.  Schaffer and  K.  David. 2013. A dynamic interplay between phytohormones is required for fruit development, maturation, and ripening. Front. Plant Sci. 4: 79. DOI: 10.3389/fpls.2013.00079.
•  Morisaki,  A.,  N.  Yamada,  S.  Yamanaka and  K.  Matsui. 2014. Dimethyl sulfide as a source of the seaweed-like aroma in cooked soybeans and correlation with its precursor, S-methylmethionine (Vitamin U). J. Agric. Food Chem. 62: 8289–8294.
•  Osajima,  Y.,  M.  Shimoda,  H.  Maeda and  S.  Ueno. 1985. Removal of S-methylmethionine sulfonium from satsuma mandarin juice. J. Agric. Chem. Soc. Jpn. 59: 501–508 (In Japanese with English abstract).
•  Sato,  K.,  Y.  Ikoma,  H.  Matsumoto and  N.  Nakajima. 2015. Effects of spray concentrations and spray times of gibberellin and prohydrojasmon on peel puffing and peel color in satsuma mandarin fruit. Hort. Res. (Japan) 14: 419–426 (In Japanese with English abstract).
•  Sawamura,  M.,  M.  Shimoda and  Y.  Osajima. 1978. Studies on off-flavor formed during heat processing of satsuma mandarin juices (III). J. Agric. Chem. Soc. Jpn. 52: 281–287 (In Japanese with English abstract).
•  Sawamura,  M.,  M.  Shimoda,  T.  Yonezawa and  Y.  Osajima. 1977. Studies on off-flavor formed during heat processing of satsuma mandarin juices (II). J. Agric. Chem. Soc. Jpn. 51: 7–13 (In Japanese with English abstract).
•  Sawano,  I. 2010. Effects of gibberellin and prohydrojasmon mixtures at different cultural practice on fruit quality of satsuma mandarin. Hort. Res. (Japan) 9 (Suppl. 1): 54 (In Japanese).
•  Scherb,  J.,  J.  Kreissl,  S.  Haupt and  P.  Schieberle. 2009. Quantitation of S-methylmethionine in raw vegetables and green malt by a stable isotope dilution assay using LC-MS/MS: comparison with dimethyl sulfide formation after heat treatment. J. Agric. Food Chem. 57: 9091–9096.
•  Shankaranarayana,  M. L.,  B.  Raghavan,  K. O.  Abraham,  C. P.  Natarajan and  H. H.  Brodnitz. 1974. Volatile sulfur compounds in food flavors. Crit. Rev. Food Sci. Nutr. 4: 395–435.
•  Stewart,  I. and  T. A.  Wheaton. 1971. Effects of ethylene and temperature on carotenoid pigmentation of citrus peel. Proc. Florida State Hortic. Soc. 84: 264–266.
•  Tyagi,  K.,  I.  Maoz,  B.  Kochanek,  N.  Sela,  L.  Lerno,  S. E.  Ebeler and  A.  Lichter. 2021. Cytokinin but not gibberellin application had major impact on the phenylpropanoid pathway in grape. Hortic. Res. 8: 51. DOI: 10.1038/s41438-021-00488-0.
•  Uefune,  M.,  R.  Ozawa and  J.  Takabayashi. 2014. Prohydrojasmon treatment of lima bean plants reduces the performance of two-spotted spider mites and induces volatiles. J. Plant Interact. 9: 69–73.
•  Wang,  S.,  H.  Takahashi,  T.  Saito,  K.  Okawa,  H.  Ohara,  M.  Shishido,  H.  Ikeura and  S.  Kondo. 2015. Jasmonate application influences endogenous abscisic acid, jasmonic acid and aroma volatiles in grapes infected by a pathogen (Glomerella cingulata). Sci. Hortic. 192: 166–172.
•  Wang,  W.,  M.  Khalil-Ur-Rehman,  J.  Feng and  J.  Tao. 2017. RNA-seq based transcriptomic analysis of CPPU treated grape berries and emission of volatile compounds. J. Plant Physiol. 218: 155–166.
•  Wang,  W.,  M.  Khalil-Ur-Rehman,  L.-L.  Wei,  N. J.  Nieuwenhuizen,  H.  Zheng and  J.-M.  Tao. 2020. Effect of thidiazuron on terpene volatile constituents and terpenoid biosynthesis pathway gene expression of shine muscat (Vitis labrusca × V. vinifera) grape berries. Molecules 25: 2578.
•  Williams,  M. P. and  P. E.  Nelson. 1974. Kinetics of the thermal degradation of methylmethionine sulfonium ions in citrate buffers and in sweet corn and tomato serum. J. Food Sci. 39: 457–460.