Effects of GA₃ Treatments on Fruit Vascular Structure and Water Transport of Grape

ABSTRACT

Exogenous gibberellic acid was applied to regulate fruit growth of grapes to acquire high quality seedless berries. The vascular bundles provide the pathway of water and nutrient transport into the fruit and play an important role in berry growth and sugar accumulation. Therefore, changes in berry size and quality are closely related to the structure and function of the vascular tissue. In this study, 10-year-old “Shine Muscat” grape vines were treated with GA₃ in grapevine production. Based on previous research and experimental environmental conditions, four treatments were applied: elongating cluster（5 mg L⁻¹ GA₃ on Apr. 28）+ seedless（25 mg L⁻¹ GA₃ on May. 24）+ expanding（50 mg L⁻¹ GA₃ on Jun. 7） (T1), seedless（25 mg L⁻¹ GA₃ on May. 24）+ expanding（50 mg L⁻¹ GA₃ on Jun. 7） (T2), expanding（50 mg L⁻¹ GA₃ on Jun. 7） (T3) and water (CK). Compared to CK, all fruits treated with GA₃ had greater berry size. In addition, GA₃ treated grapes had greater vascular bundle areas, including the areas of both phloem and xylem, as well as a greater number of vessels compared to control fruit. During the first rapid growth stage, water transport into the grape was rapid and did not differ among treatments. During veraison, the water transport rate among the four treatments differed, and the rank of speed was T2>T3>CK>T1. Therefore, we found that GA₃ treatment can significantly promote the differentiation of vascular tissues and improve water transport capacity and phloem sugar unloading; T2 promoted the development of vascular bundles, enhanced water transport speed and phloem sugar unloading of fruit after veraison, and increased berry size.

KEYWORDS:

• Grape
• GA₃
• vascular bundle
• water transport
• sugar unloading

Introduction

Grape is one of the most widely grown fruit species in the world and is therefore one of the most economically important crops. Grapes are used fresh, dried, and processed in food and beverage industries. Large berry size for table grapes is desired by consumers, and therefore, fruit size is increased with the use of gibberellic acid (GA₃) applications during production (Harrell and Williams, 1987).

Endogenous gibberellins mediate many physiological processes in plants regulating plant growth and development including seed germination, internode elongation, flower and berry development, leaf extension, and xylem cell differentiation (Davière et al., 2008; Sun, 2011; Sun and Gubler, 2004; Vera-Sirera et al., 2016). As early as the middle of the last century, gibberellins were found to induce the expansion of seedless grapes. With the progress of science, technology, and the modernization of agriculture, gibberellins started to be widely used to elongate the inflorescence prior to anthesis (Gao et al., 2020). Application to grapes near or at bloom causes pollen and ovule abortion resulting in seedlessness (Coombe, 1960). Gibberellins can induce endogenous auxin production, promote nutrient absorption, strengthen fruit competition with shoots, and promote berry expansion (Teszlák et al., 2005; Ben-Tal, 1990; Li et al., 2003; Pérez and Gómez, 2000).

Fruit expansion requires the consumption of large amounts of nutrients and increased energy metabolism. Vascular bundles in the grape are channels for the transport of water, hormones, and various nutrients(Xie et al., 2012). Understanding the structural and functional changes in the fruit vascular system during fruit development can provide new insights for advancing our understanding of the hormonal regulation of grape growth. It can also provide additional insights into the use of hormones in grape production.

The vascular system of plants consists of two main tissues, the xylem and phloem, which are important channels for the transport of water, minerals, and photosynthetic products (Dettmer et al., 2009). Studies have shown that the expansion of the grape berry is related to the uptake of water, which occurs through the vascular bundles that change as the fruit develops (Coombe and Bishop, 1980; Delrot et al., 2001; Tyerman et al., 2004). Bian et al. (2015) proposed that the vascular bundle area is coordinated with berry cell growth by comparing their development. Vascular tissue development is a complex process that is affected by many factors. Recently, the relationship between plant hormones and vascular tissue development has received considerable attention. However, the effects of GA₃ on the development and function of grape vascular tissue were rarely reported. Wang et al. (2015) found that GA₃ treatment promoted xylem development in carrot roots. Guzmán et al. (2021) found that spraying “Superior Seedless” grapes with IBA and GA₃ at full bloom increased berry size and, therefore, bunch weight. The above studies provide the theoretical basis and methodology for this study to investigate the water transport function of grape vascular bundles. Recent research on GA₃ treatments focused primarily on the effect and mechanism of berry seedlessness and expansion. However, few studies have been conducted on the structure and function of vascular bundle, the conduit of water and nutrients in the grape. Whether GA₃ affects the transport function of berries by affecting the vascular bundle structure has not yet been reported.

In this paper, we investigate the changes in the structure and function of grape vascular bundles by observing the morphological structure of vascular tissues and the transport of extracellular dyes in vascular bundles among different GA₃ treatments.

Materials and Methods

Plant Materials

The experiment was conducted in 2021 at the Shandong Fruit Research Institute at the base of Jinniu Mountain (36.127 N, 117.004 E). The whole vineyard was sheltered from rain, and ten-year-old “Shine Muscat” grapevines were trained on a V-shaped trellis on their own roots (row spacing 1 m, north-south direction).

Thirty healthy grapevines were selected. Randomly selected branches with moderate vigor and a consistent number of leaves were chosen; one inflorescence on each branch was left and selected as the test object. Each inflorescence was trimmed one week before flowering to comprise only 5 cm of the apex. Trimmed inflorescences were assigned to one of three treatments or one control.

Three GA₃ treatment schemes were applied to the trimmed inflorescences (Table 1). In treatment 1 (T1), GA₃ was applied approximately 20 d before bloom when the rachis was elongating, 1 d after full bloom, and 14 d after bloom. In treatment 2 (T2), GA₃ was applied 1 d after full bloom and 14 d after bloom. In treatment 3 (T3), GA₃ was applied 14 d after bloom. The control check (CK) was dipped in water at the same time as T1.

Table 1. Experimental design of different gibberellin treatments.

No.	Treatment
CK	Water
T1	Elongating rachis+ Seedless+ Fruit expansion
T2	Seedless+ Fruit expansion
T3	Fruit expansion

During the treatment at the time of elongating rachis, inflorescences of the same size were randomly selected, and the inflorescences were quickly dipped in 5 mg L⁻¹ GA₃ aqueous solution. After bloom the inflorescences were dipped in 25 mg L⁻¹ GA₃ aqueous solution and at fruit expansion the inflorescences were dipped in 50 mg L⁻¹ GA₃ aqueous solution. For each of the four treatments, 30 inflorescences were treated.

Berry Growth Measurements

Berries were sampled at the second rapid growth stage (84 days after flowering; DAF) and taken back to the laboratory in an ice box immediately after harvest for index determination. For each treatment, 30 berries with uniform size were randomly selected, and the vertical and horizontal diameters of each berry was measured with vernier calipers (accurate to 0.01 mm). Berry shape index was calculated according to Equation (1)(1) $\mathrm{Shape}\mathrm{index}=\mathrm{longitudinal}\mathrm{diameter}/\mathrm{transverse}\mathrm{diameter}$(1) .

(1) $\mathrm{Shape}\mathrm{index}=\mathrm{longitudinal}\mathrm{diameter}/\mathrm{transverse}\mathrm{diameter}$(1)

Thirty berries were randomly selected from each treatment, and the berry volume was measured by the difference in water level when the berry was immersed in water. The mass of single berries was measured using an electronic balance (accurate to 0.01 g), and the content of soluble solids and total acid of berries were measured by a hand-held sugar acidity meter (ATAGO PAL-1, Atago, Tokyo). Ten randomly selected berries from each treatment were used to measure firmness with a durometer (ATAGO GY-4, Atago, Tokyo) with a 3.5 mm probe.

Paraffin Section

Grape berries were sampled at the first rapid growth stage (34 DAF), veraison (54 DAF) and the second rapid growth stage（84 DAF） by selecting representative berries at the middle and outer edges of the bunch. Sampled berries were cut into 1 cm × 1 cm × 1 cm pieces with a blade from the middle of the berry and immediately fixed with FAA. Fixed materials were selected according to Hardie et al. (1996) to make paraffin sections. Slice thickness was 8–10 μm, stained with safranine and fast green, sealed with neutral gum, and photographed under an optical microscope.

Distribution and Transportation of Berry Vascular Bundles

Dye tracing was introduced in vitro according to the methods of Bondada et al. (2005) and Rogiers et al. (2001) with modification. The dye preparation method used 5 ml of 10% Na₂S₂O₄ solution mixed with 100 ml of 0.5% basic fuchsin solution. “Shine Muscat” berries were sampled at the first rapid growth stage, veraison, and the second rapid growth stage. At each sampling time, 15 representative berries were selected from the middle and outer edges of the bunch for each treatment and immediately returned to the laboratory. The base of the peduncle was submerged under water and then inserted into the prepared solution. After 8 hours, the berries were taken out of the solution, peeled and the stained part of the berry vascular bundle was observed. Photos were taken.

During the dyeing, two of the berries per treatment were taken every hour for peeling and observations, and the staining length of the main veins of the peripheral vascular bundle and the central vascular bundle was measured with linen tape, and the staining rate was calculated according to the staining time. After staining, the remaining berries were peeled and observed, and the number of staining strips of peripheral vascular bundles and the main veins of peripheral vascular bundles of each berry was recorded and photos were taken.

Xylem Hydraulic Conductivity

To estimate the frequency distribution of receptacle xylem vessels size classes and their potential contribution to the area-specific hydraulic conductivity (K, kg m⁻¹ MPa⁻¹ s⁻¹), xylem region(s) were selected in the cross-sectional images of the central and peripheral vascular bundles using ImageJ. Their corresponding area (A_c-xylem, µ m²), vessel number (n), and density (VD, n mm⁻²) were then calculated. Using hand-drawn ellipses in ImageJ (Schneider et al., 2012), semi-major (a₁, µm) (half of Feret diameter) and semi-minor (b₁, µm) (half of minimum-Feret) axes were determined for the single xylem vessel in the corresponding xylem area. Idealized vessel diameter (d₁, µm), as the equivalent circle diameter for the lumen, was obtained using Equation (2)(2) $d_1={\left(32\times \frac{{\left(a_1\times b_1\right)}^3}{\left({a_1}^2+{b_1}^2\right)}\right)}^{\frac{1}{4}}$(2) (White and Majdalani, 2006).

(2) $d_1={\left(32\times \frac{{\left(a_1\times b_1\right)}^3}{\left({a_1}^2+{b_1}^2\right)}\right)}^{\frac{1}{4}}$(2)

Assuming the selected xylem region(s) used for the image segmentation was representative of each cross-sectional xylem vessel, the potential area-specific hydraulic conductivity (K, kg m⁻¹ MPa⁻¹ s⁻¹) of total vessels and each vessel size class (i.e., 2 µm class interval in d₁) were estimated according to the Hagen-Poiseuille equation, Equation (3)(3) $\mathit{K}=\left(\frac{\mathit{\pi \rho }}{128\mathit{\eta Ac}-\mathit{xylem}}\right)\sum \left({d_1}^4\right)$(3) (Tyree and Ewers, 1991), and then the percentage of K for each vessel size class was determined，

(3) $\mathit{K}=\left(\frac{\mathit{\pi \rho }}{128\mathit{\eta Ac}-\mathit{xylem}}\right)\sum \left({d_1}^4\right)$(3)

where η is the viscosity of water (1.002 × 10⁻⁹ MPa s) and ρ is the density of water (998.2 kg m⁻³), both at 20°C.

Sugar Unloadin

The “New Berry Cup” was prepared according to the method of (Wang et al., 2006). At veraison (August 19th) and at the second rapid growth stage (September 28th) of “Shine Muscat” grapes, a randomly selected berry was sampled from a randomly selected cluster with consistent berry development for each treatment. A 1 cm cut was gently made at the equator of the berry with a dissecting knife, with the depth of the cut not cutting the flesh tissue of the berry. The peel was gently removed to the base of the berry using ophthalmic forceps, and then the peel was dissected using scissors and rinsed with a prepared Mes buffer consisting of 5 mmol L⁻¹ Mes (2- (N-morpholine) ethane sulfonic acid), 2 mol L⁻¹ CaCl₂, 100 mmol/L mannitol (D-mannitol), and 0.2% (w/v) polyvinyl pyrrolidone (PVPP). The peeled grape was carefully placed into a 20 mm infusion tube, which was pre-filled with Mes buffer. The opening of the tube was sealed with parafilm and the berry cup was attached to the cluster. Finally, aluminum foil was wrapped on the outer layer of the “Berry Cup” to prevent excessive changes in heat inside the cup from affecting the test results and extending the service life of the “Berry Cup.” The experiment started at 9:30 and collected buffer solution for 0.5 hours, followed by another collection every 2 hours until 18:00, during which time the buffer solution was discharged from the cup and then replaced with fresh buffer using a syringe. The collected buffer solution was then immediately frozen in liquid nitrogen at −80°C for later determination of glucose, fructose, and sucrose.

Statistical Analysis

SPSS 26.0 software was used for the statistical analysis of the test data. A single-factor ANOVA was conducted for each factor, and Duncan’s new multiple-range tests were used for multiple comparisons, with a significance level of p < .05.

Results

Berry Ripening

Effect of GA₃ Treatments on Fruit External Quality

The shape of “Shine Muscat” grapes is naturally long and elliptical (berry shape index > 1). The GA₃ treatments increased the longitudinal diameter of the berry, but there was no significant difference in the transverse diameter. The longitudinal diameter of fruit subjected to T2 (31.04 ± 0.92 mm) was significantly higher than the other GA₃ treatments (mean of the others = 28.95 ± 1.17 mm), which was 12.2% higher than the control (CK) (Figure 1a). In addition, the berry shape index was the highest under T2 (1.3 ± 0.04) (Figure 1b).

Figure 1. Berry longitudinal diameter (a) and shape index (b) of ‘Shine Muscat’ fruit subjected to different GA₃ treatments. Different lower-case letters indicate significant differences among treatments (n= 3, Duncan’s test, p<.05).

The GA₃ treatments increased the berry size of the grape. By comparing the berry weight and volume between the different treatments, it was found that T2 treated fruit had significantly higher berry weight (12.05 ± 0.98 g) than the other GA₃ treatments (mean of the others = 9.85 ± 0.24 g), which was 142.8% of the control (Figure 2a). At the same time, the berry volume of T2 grapes (10.63 ± 0.25 cm³) was significantly higher than that of T1 grapes and control (CK) fruit (Figure 2b).

Figure 2. Berry weight (a) and volume (b) of ‘Shine Muscat’ fruit subjected to different GA₃ treatments. Different lower-case letters indicate significant differences among treatments (n= 3, Duncan’s test, p<.05).

Effects of GA₃ Treatments on Fruit Internal Quality

During the second rapid growth stage, the total soluble solids (TSS) content of T2 fruit was the highest (18.97 ± 0.86%), and no significant difference in TSS content was found between T3 and control fruit (Figure 3a). In addition, there was little difference in the sugar/acid ratio (Figure 3b) and firmness (Figure 4) between GA₃ and control treatments.

Figure 3. TSS (a) and the sugar/acid (b) of ‘Shine Muscat’ fruit subjected to GA₃ treatments. Different lower-case letters indicate significant differences among treatments (n = 3, Duncan’s test, p <.05).

Figure 4. Firmness of ‘Shine Muscat’ fruit subjected to GA₃ treatments. Different lower-case letters indicate significant differences among treatments (n= 3, Duncan’s test, p <.05).

Anatomical Structure

Density of Flesh Cell

GA₃ treatments significantly reduced the vessel density of flesh cells (Figure 5). The vessel density of flesh cells in the control fruit was highest (84 ± 6 vessels mm⁻²), and the vessel density of flesh cells in the T2 fruit was lowest (62 ± 4 vessels mm⁻²). The vessel density of flesh cells T1 and T3 fruit were similar (71 ± 5 vessels mm⁻²). Therefore, the T2 treatment was most effective in increasing flesh cell expansion.

Figure 5. Berry cell density in grapes subjected to GA₃ treatments. Different lower-case letters indicate significant differences among treatments (n= 3, Duncan’s test, p <.05).

Anatomical Structure of the Vascular Bundle

Paraffin sectioning was performed on the equatorial part of “Shine Muscat” berries during the first rapid growth stage, at veraison, and during the second rapid growth stage to observe the cross-sectional microstructure of the central (Figure 6) and peripheral (Figure 8) vascular bundles. The vascular bundle structure of “Shine Muscat” grapes was similar under different treatments. The xylem vessels and parenchyma cells in the vascular bundles were regularly distributed and arranged radially. The shape of the peripheral vascular bundle was nearly oval, and the parenchyma cells were less damaged by lateral expansion so the boundary between vascular tissues and flesh tissues was obvious, and the structure was nearly complete (Figure 8). The shape of the central vascular bundle was oval. The expansion of the parenchyma cells distorted the central vascular bundle, resulting in damage or even breaking of the vascular bundle structure (Figure 6f,l). The area of the single peripheral vascular bundles was about 1/4–1/3 of that of the central vascular bundles. The areas of the vascular bundle, phloem, and xylem were significantly greater in GA₃ treated fruit compared to control fruit (Figure 7). The area of the central vascular bundle (A_v), phloem (A_p), and xylem (A_x) were largest in the T2 treated fruit, which were significantly larger than those of the T3 fruit and the control. At the first rapid growth stage, the T2 treatment resulted in a small difference in the area of the vascular bundle compared to that of the T1 treated fruit, which was about twice that of the control. By veraison, the difference in the area of the vascular bundle, phloem, and xylem between each treatment decreased. At the second rapid growth stage, the area of the vascular bundle, phloem, and xylem in the equatorial of berries under T2-treated was greatest (A_v: 5 5273 ± 3571 μm², A_p: 2 6149 ± 97 μm², A_x: 1 8602 ± 208 μm²). The vascular bundle area of berries in three stages of growth was analyzed, and it was found that the vascular bundle area was largest at the second rapid growth stage, and the vascular bundle area in the first rapid growth stage was similar to that at veraison.

Figure 6. Cross-sectional microstructure of central vascular bundles (e-h) of ‘Shine Muscat’ grapes under CK (a-c), T1 (d-f), T2 (g-i) and T3 (j-l) treatments at the first rapid growth stage、veraison and the second rapid growth stage. X: xylem; P: phloem; V: vessel; Scale bar =100 μm.

Figure 7. Areas of the central vascular bundle (Cvb) of ‘Shine Muscat’ grapes under different treatments during three stages of growth.A_v: vascular bundles area, A_p: phloem area, A_x: xylem area. Different lower-case letters indicate significant differences among treatments (n= 3, Duncan’s test, p <.05).

Figure 8. Cross-sectional microstructure of peripheral vascular bundles (e-h) of ‘Shine Muscat’ grapes under CK (a-c), T1 (d-f), T2 (g-i) and T3 (j-l) treatments at the first rapid growth stage (a、d、g、j)、veraison(b、e、h、k) and the second rapid growth stage (c、f、i、l). (X: xylem; P: phloem; Ve: vessel) Scale bar =100 μm.

During the first rapid growth stage of the fruit, the area of the equatorial vascular bundle and phloem in T1-treated fruit was significantly higher than that of other treatments (A_v: 60174 ± 525 μm²; A_p :4,225 ± 2059 μm²) (Figure 9). At this point, the maximum area of the xylem in fruit under T2-treated was 9431 ± 2316 μm². At veraison, there was a small difference in the area of the equatorial vascular bundle and phloem of berries under gibberellin treatment, which was significantly higher than that of the control. At this time, the vascular bundle subjected to the T2 treatment was the largest, and the area of the equatorial xylem was significantly higher than that of the other three treatments (7910 ± 973 μm²). At the second rapid growth stage, there was a small difference in the area of the equatorial phloem and xylem subjected to gibberellin treatment, which remained significantly higher than that of the control. By analyzing the area of peripheral vascular bundles in the three periods, it was found that they were different from the central vascular bundles.

Figure 9. Areas of the peripheral vascular bundles (pvb) of ‘Shine Muscat’ grapes under different treatments during three stages of growth.A_v: vascular bundles area, A_p: phloem area, A_x: xylem area. Different lower-case letters indicate significant differences among treatments (n = 3, Duncan’s test, p <.05).

The number of vessels in the vascular bundles of fruit at the first rapid growth stage, veraison, and the second rapid growth stage treated with gibberellin was significantly higher than the control, with the number of vessels being greatest in fruit subjected to T2, with T2>T1>T3>CK (Figure 10).

Figure 10. Number of vessels in the xylem of the central vascular bundles (Cvb) and peripheral vascular bundles (Pvb) of ‘Shine Muscat’ under different treatments during three stages of growth. Different lower-case letters indicate significant differences among treatments (n = 3, Duncan’s test, p <.05).

Water Transport

Water absorption of berries at different stages could be seen by observing the distribution of dye in the berry vascular bundles (Figure 11). Almost all the main veins and branches of all peripheral vascular bundles under GA₃-treated and control fruit were stained during the first rapid growth stage of “Shine Muscat” grapes, and the central vascular bundles and some of the embryonic vascular bundles were also stained; this indicated that water could be transported to the stained cells (Figure 11a-h). During veraison, the staining area of peripheral vascular bundles was significantly reduced in all treatments. Only a small part of some peripheral vascular bundles’ main veins were stained in T2 and T3 fruit (Figure 11k,l), and only part of the central vascular bundles were stained in T3 fruit (Figure 11p). The dye transport in fruit subjected to the T1 treatment and the control (CK) only reached the pedicel (Figure 11i,j). During the second rapid growth stage, the staining of vascular bundles was less than during the first rapid growth stage but higher than during veraison. The main veins of peripheral vascular bundles were stained in control, T2, and T3 fruit (Figure 11q,s,t).

Figure 11. Dye transport distribution in ‘Shine Muscat’ grapes in the first rapid growth stage (A-H), veraison (I-P), and the second rapid growth stage (Q-X) subjected to CK(a、e、i、m、q、u)、T1(b、f、j、n、r、v)、T2(c、g、k、o、s、w) and T3(d、h、l、p、t、x) treatments.

The vessel diameters were divided into 6 classes from 0–12 μm using 2 μm intervals. The central vascular bundles’ xylem area-specific hydraulic conductivity (K), derived from the Hagen-Poiseuille equation, demonstrated that a significant proportion of K could be attributed to xylem vessels in the 8–10 µm diameter classes in control, T2 and T3 fruit, while the 10–12 µm diameter class of xylem vessels contribute more to K than other classes in T1-treated fruit during the first rapid growth stage (Figure 12a). During veraison, a significant proportion of K could be attributed to xylem vessels in the 8–10 µm diameter classes in fruit subjected to T2. The significant proportion of K under other treatments is mainly concentrated in 10–12 µm (Figure 12b). And a significant proportion of K could be attributed to xylem vessels in the 8–10 µm diameter classes in control fruit and those subjected to the T1-treated. In fruit from the T2 and T3-treated, a significant proportion of K was mainly concentrated in vessels with a diameter of 10–12 µm during the second rapid growth stage (Figure 12c).

Figure 12. Relative abundance (white bars) of vessels in five diameter classes in the central vascular bundles of “Shine Muscat” grapes subjected to the CK, T1, T2, and T3 treatments and their relative correlation to the potential area-specific hydraulic conductivity (K, in percentage, black bars) at the first rapid growth stage (a), veraison (b) and the second rapid growth stage (c). Different lower-case letters indicate that there are differences in vessel abundance of different classes, and different upper-case letters indicate that there are differences in K (n = 3, Duncan’s test, p < .05).

Then, the predominant vessel classes under each treatment were statistically analyzed, based on the significant proportion of K, it was found that the abundance of vessels in vessel diameter was CK>T3>T1>T2 under the significant proportion of K in the equatorial part of the central vascular bundle during the first rapid growth stage. During veraison, the abundance of vessels in the diameter class was T1>T2>T3>CK under the significant proportion of K in the equatorial part of the central vascular bundle. At the second rapid growth stage, it was T1>CK>T2>T3.

During the first rapid growth stage, a significant proportion of K could be attributed to xylem vessels in the 10–12 µm diameter class in the control fruit and those subjected to T3. The significant proportion of K is mainly concentrated in both the 6–8 µm and 8–10 µm diameter classes for fruit subjected to T1. Under T2, the significant proportion of K was mainly concentrated in the 6–8 µm diameter class (Figure 13a). In control, T2- and T3-treated fruit, a significant proportion of K could be attributed to xylem vessels in the 6–8 µm diameter classes, while the 8–10 µm diameter class of xylem vessels contribute more to K than other classes under T1-treated fruit during veraison (Figure 13b). During the second rapid growth stage, a significant proportion of K could be attributed to xylem vessels in the 6–8 µm diameter classes in control, T1, and T3 fruit, while the 8–10 µm diameter class of xylem vessels contribute more to K than other classes in fruit subjected to the T2 treatment (Figure 13c).

Figure 13. Relative abundance (white bars) of five vessel diameter classes in the peripheral vascular bundles of “Shine Muscat” grapes subjected to CK, T1, T2, and T3 and their relative correlation to the potential area-specific hydraulic conductivity (K, in percentage, black bars) at the first rapid growth stage (A), veraison (B) and the second rapid growth stage (C). Different lower-case letters indicate that there are differences in vessel abundance of different classes, and different upper-case letters indicate that there are differences in K (n = 3, Duncan’s test, p < .05).

Then, the predominant vessel classes under each treatment were statistically analyzed, based on the significant proportion of K, it was found that the abundance of vessels in the diameter classes was T1>T2>T3>CK under the significant proportion of K in the equatorial part of the peripheral vascular bundles during the first rapid growth stage and veraison. At the second rapid growth stage, it was T2>T3>CK>T1.

Sugar Unloading

During veraison, a large amount of fructose and glucose began to accumulate in the “Shine Muscat” grapes, with fructose content higher than glucose content (Figure 14). At this time, there was still a small amount of sucrose. After gibberellin treatments, the changes of fructose and glucose unloading in the fruit phloem showed a bimodal curve across the five time periods (9:30–10:00,11:30–12:00,13:30–14:00,15:30–16:00,17:30–18:00). In the morning, the fruit unloading rapidly increased to a peak at 12:00, roughly in sync with the increase in photosynthetic rate. Afterward, it rapidly decreased to a bottom at 14:00, and then sugar unloading increased again. At 16:00, it reached a small peak again and decreased again with sunset. The unloading volume is large in the morning, with a peak of about 1.1 times that of the afternoon peak. Unlike the gibberellin treatments, the control treatment showed a single peak curve, with a rapid increase in fruit unloading in the morning and reaching its peak at 14:00, followed by a rapid decrease. From different time periods, the highest fructose unloading amount (3.89 ± 0.04 mg g⁻¹) was observed in berries subjected to the T3 treatment from 9:30 to 10:00, which was significantly higher than that of control fruit (3.51 ± 0.02 mg g⁻¹). The highest fructose unloading amount (4.21 ± 0.06 mg g⁻¹) was observed in T1-treated berries from 11:30 to 12:00, and the highest fructose unloading amount was observed in control fruit from 13:30 to 14:00. From 15:30 to 16:00 and 17:30 to 18:00, the highest fructose unloading amount was observed in T2-treated berries, 3.89 ± 0.00 mg g⁻¹ and 3.30 ± 0.03 mg g⁻¹ respectively. In addition, the glucose unloading amounts of T3-treated berries were the highest at 9:30–10:00 and 11:30–12:00, respectively, at 2.98 ± 0.04 mg g⁻¹ and 3.11 ± 0.01 mg g⁻¹, similar to fructose during the 13:30–14:00 time period. The maximum glucose unloading amount in control fruit was 3.20 ± 0.03 mg g⁻¹ and the highest glucose unloading amount was still observed in fruit subjected to T3 (3.11 ± 0.04 mg g⁻¹) from 15:30–16:00, and T2 (2.91 ± 0.04 mg g⁻¹) from 17:30–18:00. Fructose and glucose unloading in fruit subjected to each treatment during the five time periods was greatest during veraison, with fruit from the T2 treatment having the highest fruit sugar unloading amount (32.25 ± 0.27 mg g⁻¹).

Figure 14. Sugar offloading of “Shine Muscat” grapes during veraison and the second rapid growth stage subjected to different gibberellin treatments. Different lower-case letters indicate that there are significant differences (n = 3, Duncan’s test, p < .05).

Analysis of the components of the sugar unloading solution of nondetached fruit during the second rapid growth stage revealed that a large amount of fructose and glucose was still accumulated in the “Shine Muscat” grapes, and was similar to the rate at veraison, with fructose content higher than glucose content (Figure 14). At this time, the main sugars present in the “Shine Muscat” berries were fructose and glucose, with trace amounts of sucrose. The sugar unloading curve of grapes under different treatments during the second rapid growth stage presents a single peak curve. The fruit unloading amount reached its peak at 10:00 in the morning, then slowly decreased to the bottom at 14:00. Afterward, with the increase of photosynthetic rate, sugar unloading increased again, reaching a small peak at 16:00, and then decreasing at sunset. The unloading volume in the morning is higher than in the afternoon, with the highest value being about twice that of the afternoon. The highest amount of fructose unloading was observed in the first four time periods in fruit subjected to T2. From 17:30 to 18:00, the highest amount of fructose unloading was observed in fruit subjected to T3 (3.35 ± 0.03 mg g⁻¹). In addition, the maximum glucose unloading amounts of berries subjected to T2 were 3.09 ± 0.04 mg g⁻¹, 2.92 ± 0.00 mg g⁻¹, and 2.72 ± 0.03 mg g⁻¹ at 9:30–10:00, 13:30–14:00, and 17:30–18:00, respectively. During the time periods of 11:30–12:00 and 15:30–16:00, the glucose unloading amounts in fruit subjected to T3 were 2.91 ± 0.07 mg g⁻¹ and 3.03 ± 0.03 mg g⁻¹, respectively. Fructose and glucose unloading under each treatment during the five time periods were found to be highest during the ripening period in fruit subjected to T2 (31.26 ± 0.46 mg g⁻¹).

Discussion

GA₃, as one of the most widely applied plant growth regulators, can elongate rachis, induce seedlessness, improve fruit set, and promote fruit expansion (Aya et al., 2009; Mullins et al., 1992). There are many factors affecting the growth of grape berries and different cultivars have different sensitivities to GA₃ treatments. At present, GA₃ is applied to “Kyoho” (Lee et al., 2013), “Centennial Seedless” (Pires et al., 1997), “Shiraz” (Xie et al., 2022), “Thompson Seedless” (Kasimatis et al., 1979) and many other grape varieties to improve berry quality and marketability. In this paper, we investigated the changes in the berry quality of “Shine Muscat” grapes subjected to three different GA₃ treatments. It was found that the best berry expansion and elongation effects were evident in T2-treated fruit (Figures 1 and 2). In addition, the differences in sugar-acid ratio and firmness of berries after GA₃ treatment were small compared to that of control fruit, which suggests that GA₃ treatment did not affect the flavor of berries.

The growth of berries is closely related to the material transport channel, so the vascular bundle of fruit under each treatment was studied. Vascular bundles of grape berries are mainly responsible for transporting water, inorganic salts, and organic matter, which play an important role in berry growth and quality formation. Dye tracing studies of plant vascular bundle transport have been reported since the 1960s. Bondada et al. (2005) used basic fuchsin to observe the changes in water transport in the vascular bundle of grapes at different development stages. Davière et al. (2004) used 1% acid fuchsin to observe the changes in xylem functionality in apple fruit. Bian et al. (2015) used red ink to observe the distribution of vascular bundles in jujube. In our study, the distribution of vascular bundles was found to be the same in all treatments by dye tracing, with the central vascular bundle of the berry consisting of 4–6 vascular bundles and about 2 vascular bundles extending from the peduncle to the top of the berry. The peripheral vascular bundles are distributed in a network within the exocarp and mesocarp, with connections distributed within the mesocarp. The central vascular bundles at the top of the berry are aggregated and connected with the peripheral vascular bundle, so the vascular bundle system in grape berry is integrated, and the transport efficiency can be improved based on this kind of structure.

The differentiation and growth of the vascular system are controlled by hormones (Aloni, 1987). Gibberellin treatment has a facilitative effect on vascular tissues differentiation, which may be related to gibberellin’s ability to promote cell division and elongation (Bleecker et al., 1986; Stamm and Kumar, 2013; van der Knaap et al., 2000). In this paper, we found that the area of the vascular bundles, xylem, and phloem of the berry was significantly increased after GA₃ treatment compared with the control by counting the area of vascular bundles through sectional observation (Figures 7 and 9). Guo et al. (2015) used toluidine blue staining on the birch hypocotyl base and found that GA₃ treatment affected differentiation in the xylem of the hypocotyl and promoted xylem development. Wang et al. (2015) found that exogenous gibberellin could promote the increase of xylem in carrots; Tokunaga et al. (2006) showed that GA played an important role in the xylem vessel origination of zinnias. In addition, GA can change the growth and development of phloem by promoting the elongation of primary phloem fibers and decreasing the thickness of phloem cell walls in Coleus blumei Benth (Aloni, 1979).

The dye tracing method revealed that the vascular bundle transport capacity of the fruit was the highest at the first rapid growth stage, and the transport speed of each treatment showed CK>T3>T2>T1. At veraison, the vascular bundle transport capacity of the fruit decreased but increased at the second rapid growth stage. The transport rates between different treatments showed T2>T3>CK>T1 (Figure 11). We found that the water transport rate is highest in the control fruit, which further focuses on water transport to the vessel water transport. Currently, the significant proportion of K under water treatment has the highest vessel abundance in vessel diameter (Figure 12). Therefore, during the first rapid growth stage, water is mainly transported through the central vascular bundle xylem. Ollat et al. (2002) found that entering the second rapid growth period, the central vascular bundle becomes blocked, and water and sugar can only be transported into the berry through the peripheral vascular bundle. At veraison, although GA treatment can significantly increase the diameter of the vessels (Nishijima et al., 2005), Bondada et al. (2005) showed that the structure and function of the xylem in the grape are lost after transformation. At this time, fruit subjected to T2 has the highest water transport rate, and the area and number of vessels in the vascular bundle, xylem, and phloem are highest, however, under the significant proportion of K, the largest vessel abundance in the diameter class was observed in the fruit subjected to T1 (Figure 12), indicating that the water transport capacity of the vessel decreased at veraison, mainly through the phloem. During the second rapid growth period, fruit subjected to T2, only the significant proportion of K in the middle and lower parts of the peripheral vascular bundle was found to have the highest vascular abundance at the diameter class. The transportation capacity of the peripheral vascular bundle was higher than that of the central vascular bundle at this time (Figure 11o,p). Xie et al. (2012) showed that after veraison, the xylem vessels of the peripheral vascular bundle of grapes ruptured, and during the second rapid growth period, water was mainly transported to the berry through the phloem pathway (Choat et al., 2009). Therefore, during the second rapid growth period, water transportation to the berry was mainly completed by the peripheral vascular bundle phloem. During this stage, the accumulation of sugar in the fruit increased, and the osmotic regulation ability of the fruit improved. Under T2, the content of sugar unloading in the fruit was highest (Figure 14), which further promoted water to enter the berry through the phloem.

Conclusion

GA₃ treatment can significantly promote the differentiation of vascular tissues and improve water transport capacity and sugar unloading, thereby improving berry quality. T2 can promote the development of vascular bundles and phloem, enhance the water transport capacity of fruit after entering veraison, and promote berry expansion. During the second rapid growth stage, water transport was mainly completed by the phloem of peripheral vascular bundles.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Agricultural Variety Improvement Project of Shandong Province [2022LZGCQY019]; Agricultural Sciences Agricultural Science and Technology Innovation Project of Shandong Academy [CXGC2023A47]; Agricultural Major Technology Collaborative Promotion Project of Shandong Province [SDNYXTTG-2023-18]; the National Natural Science Foundation of China [31872050].