Effects of Plant Growth Regulators on Plum ( Prunus domestica L.) Grown on Two Rootstocks at Harvest and at the Postharvest Period

: Plant growth regulators (PGRs), such as cytokinins (6-benzyladenine; BA) and gibberellins (GAs), are widely used in fruit production. This study focused on the plum cultivar “ ˇCaˇcanska rodna” ( P. domestica L.) grafted on vegetative rootstock “WaVit” and generative rootstock Prunus cerasifera , with interstock Prunus spinosa . PGR treatments included 50 mg L − 1 and 100 mg L − 1 of BA and 200 mg L − 1 of gibberellin A3 (GA 3 ) and non-treated control. Pomological characteristics of plum fruits were examined at harvest, while physicochemical properties were analyzed at harvest and after 28 days of cold storage and 4 days of shelf life. GA 3 and BA application changed the fruit morphological traits, improved skin strength, and increased carotenoid, anthocyanin and sucrose content while decreasing the titratable acidity at harvest. The beneﬁcial effects of higher sucrose, anthocyanin and carotenoid levels persisted in all PGR-treated fruits after cold storage and shelf life. GA 3 -treated fruits had ﬁrmer ﬂesh, stronger skin and higher total soluble solids (TSS) content, while in BA-treated plums, these effects were rootstock-dependent. The physical properties and chemical composition of plum fruit in the postharvest period suggest beneﬁcial effects of the applied PGR treatments. Moreover, these chemical treatments might have prolonged the beneﬁcial impact on fruit storability, nutritional proﬁle and sensory properties. Based on our results, GA 3 preharvest treatment can be included in standard cultivation practices within contemporary production systems of European plums not only to improve fruit quality at harvest but also to improve the storage potential and nutritional value, regardless of the rootstock used.


Introduction
Unlike Japanese plums (Prunus salicina L.), which are mostly eaten fresh, European plums (Prunus domestica L.) are usually consumed in processed form, such as dried or canned whole fruit, jams, jellies, purée or brandy [1]. The recently published review provides an overview of the beneficial health effects attributed to plum consumption, including bone health, antioxidant and anti-inflammatory activity, enhanced cognitive ability, reduced risk of cardiovascular diseases, laxative effects, and anti-allergic and anti-microbial properties [2]. To increase the consumption of fresh plums, preharvest and postharvest measures should be encouraged and supported. Among the preharvest measures which are increasingly drawing the attention of both researchers and producers is the application of plant growth regulators (PGRs). The impact of PGR treatments on ments included 50 and 100 mg L −1 of 6-benzyladenine (BA50 and BA100, respectively), 200 mg L −1 of gibberellin A3 (GA 3 ) and non-treated control, prepared from Gerba 4 LG (4% active ingredient of BA) and Gibrelin (1.8% active ingredient of GA 3 ), respectively, both purchased from "L-Gobbi", Italy. Treatments were applied with a backpack sprayer (Stihl SR-420) until run-off. The total volume of each PGR solution was 2.4 L per treatment. Fruits were harvested as commercially ripe with appropriate total soluble solids (TSS, above 15%) and fruit firmness (105 to 140 N) values typical for "Čačanska rodna".
For postharvest analysis, approximately 5 kg of plum fruits were distributed in wooden crates of 50 × 30 × 8 cm dimensions and were placed in a cooling chamber (at 1 ± 1 • C; 80 ± 10% RH). After 28 days, fruits were removed from cold storage and subjected to shelf-life conditions (24 ± 2 • C) for 4 days. Pomological properties were analyzed only at harvest, while fruit physicochemical properties were analyzed at harvest, after cold storage and after shelf life.

Pomological Properties, Respiration Rate and Ethylene Production
Fruit weight (g) was measured on 30 fruits per treatment using a technical balance (Kern 572-35, Kern & Sohn, GmbH, Balingen, Germany). Fruit size (mm) was determined by measuring three linear dimensions (length, width, and thickness) of each fruit with a digital caliper gauge (0.01 mm) (Mitutoyo, CD-6"CX, Tokyo, Japan). The fruit shape index was calculated by applying the following formula: length × length/width × thickness [14].
Respiration rate and ethylene production were determined on approx. 300 g of fruit (equivalent to 10 plums), placed in a 700 mL container and hermetically sealed with multilayer foil at 24 ± 2 • C. Ethylene production and respiration rate were measured on harvested fresh fruit and stored fruit once a day for seven consecutive days.
Ethylene detection was carried out from 2 mL of gas sampled by a plastic syringe and injected into a 10 mL headspace vial sealed with silicone septa. The detection was performed by gas chromatography (GC7890, Agilent, Santa Clara, CA, USA), equipped with an FID detector (Agilent, Santa Clara, CA, USA) and an autosampler (COMBIPAL, CTCAnalytics AG, Zwingen, Switzerland), according to the protocol described by Mandić et al. [53]. A DB-WAX column was used for separation. The temperature gradient varied from 60 • C to 150 • C, and the flow rate was set to 30 mL min −1 , with nitrogen (N 2 ) as the carrier gas. The injection was in split mode (10:1). The ethylene content was calculated from the calibration developed using different injections of 4% ethylene in N 2 . The ethylene production (nL g −1 h −1 ) was calculated from ethylene concentration, taking into account the weight of fruit in the dish, its volume, the volume of the dish and the exact time that elapsed from the moment the dish was sealed until the sampling was performed.
Respiration rate was determined by measuring carbon dioxide (CO 2 ) concentration directly in the atmosphere surrounding fruits by puncturing multiple sealed foils with a sampling needle from an OXYBABY 6.0 instrument (WIT-Gasetechnik GmbH & Co KG T, Witten, Germany). CO 2 production (µL g −1 h −1 ) was calculated from the difference in CO 2 concentration before sealing the dish and after 4 h by taking into account the weight of fruit in the dish, its volume, the volume of the dish and the exact time period from the moment the dish was sealed until the sampling was performed.

Physicochemical Properties
Fruit skin and flesh color were measured in CIELAB color space using a CR-400 Chroma Meter (Konica-Minolta, Osaka, Japan). After removing the wax layer, the skin color was measured on 20 randomly selected specimens, with two measurements on the opposite sides of each fruit, in the equatorial region. After removing the stone, the flesh color was measured on 20 randomly selected specimens on both halves of each fruit.
Texture analysis was performed using TA.XT Plus Texture Analyzer (Stable Micro Systems, Godalming, UK). The skin and flesh textural properties were analyzed on 20 randomly chosen plum fruits. Skin strength and elasticity were measured by conducting the penetration test, whereby a 2 mm diameter stainless steel flat cylinder probe (P/2) was inserted into each specimen at a constant penetration speed of 3 mm s −1 to the pre-set penetration distance of 10 mm. The load cell mass was 30 kg, while the trigger force was set to 5 g. The skin elasticity was measured for each fruit at two opposite points in the equatorial region. Skin elasticity is the distance (mm) to which the skin inflects prior to probe penetration into the fruit, while skin strength is the force (N) needed for puncturing the fruit skin. The fruit flesh firmness was measured using a penetration test with a stainless-steel rounded cylinder probe of 8 mm diameter. Prior to the measurements, the skin was removed from the equatorial region with a sharp peeler (15-20 mm diameter). Penetration was performed to the pre-set distance of 4 mm at a 10 mm s −1 penetration speed with a 25 g trigger force. Fruit firmness was recorded as the force (N) needed for penetration.
Total soluble solids (TSS; %) and pH were measured directly from the previously homogenized sample at 20 • C using a refractometer (ATR-ST plus, Schmidt + Haensch, Berlin, Germany) and pH meter (AMT12, Amtast, Phoenix, AZ, USA), respectively. The titratable acidity (TA; %) was measured from 3 g of sample diluted in 30 mL of deionized water. After homogenization, the sample was centrifuged at 13,776× g for 5 min (Centrifuge 5804 R, Eppendorf, Germany). Titration was carried out on 10 mL of supernatant with 0.1 M NaOH.
The anthocyanin content was determined according to Lee et al. [54] with minor modifications. Briefly, anthocyanins were extracted from approx. 2 g of homogenized sample in 20 mL 0.1% HCl in methanol. The sample extract (1 mL) was transferred to 9 mL of buffer (pH 1.0 or pH 4.5). After stabilization for 2 h at 4 • C, absorbance was measured at 515 and 700 nm using a spectrophotometer (Cintra 303, GBC Scientific Equipment, Braeside, VIC, Australia). A blank was prepared by dissolving 1 mL of 0.1% HCl in methanol in 9 mL of buffers. The total anthocyanin content was calculated using measured absorbance, molar extinction coefficient and molar mass of cyanidin-3-glucoside and was expressed as cyanidin-3-glucoside (mg kg −1 of fresh weight).
Carotenoids were extracted from 0.5 g of plant material with 20 mL of acetone. The carotenoid content was determined from the supernatant obtained after centrifugation for 5 min at 13,776× g (Centrifuge 5804 R, Eppendorf, Germany) by measuring the absorbance at 470, 653 and 666 nm using a spectrophotometer (Cintra 303, GBC Scientific Equipment, Braeside, VIC, Australia). The carotenoid content (mg kg −1 of fresh weight) was calculated using the molar extinction coefficient [55].
The composition of sugars and organic acids was determined using liquid chromatography according to the method described by Milenković et al. [56].

Statistical Analysis
The univariate analysis consisted of determining the significance of the applied treatments using ANOVA followed by Duncan's multiple range test. The Tukey's HSD (honestly significant difference) test was adopted for ethylene production and respiration rate, while multivariate explanatory analysis was performed using principal component analysis (PCA). The data were analyzed using Statistica (TIBCO Software Inc., Palo Alto, CA, USA, 2018, version 13) commercial software.

Results
Obtained results were presented in the form of a comparison of the average values of determined plum properties accompanied by an ANOVA-based analysis of the significance of effects and in the form of multivariate PCA.

Univariate Results
All plums treated with PGRs were characterized by higher weight than the controls, but the increase was statistically significant only in BA50-treated fruits grown on "WaVit" rootstock ( Figure 1A). PGR treatments affected fruit width and thickness, while fruit length remained unchanged ( Figure 1C). Differences in the fruit shape index (due to fruits being more round) were noted only in fruits grown on "WaVit" and treated with BA ( Figure 1B). Rootstocks did not exhibit an observable effect on fruit weight or shape index ( Figure 1).

Univariate Results
All plums treated with PGRs were characterized by higher weight than the controls, but the increase was statistically significant only in BA50-treated fruits grown on "WaVit" rootstock ( Figure 1A). PGR treatments affected fruit width and thickness, while fruit length remained unchanged ( Figure 1C). Differences in the fruit shape index (due to fruits being more round) were noted only in fruits grown on "WaVit" and treated with BA ( Figure 1B). Rootstocks did not exhibit an observable effect on fruit weight or shape index ( Figure 1).  Both at harvest and after cold storage, ethylene production was characterized by an increasing trend regardless of rootstock or PGR treatment, with approximately the same quantities produced regardless of applied treatments ( of gibberellin A3; BA50-50 mg L −1 of 6-benzyladenine; BA100-100 mg L −1 of 6-benzyladenine. Marks in one graph followed by the same letter were not significantly different (p < 0.05). The main factors are presented, and their significance is annotated by: NS-not significant, **-significant at 0.01. The error bars on the columns represent the standard deviation.
Both at harvest and after cold storage, ethylene production was characterized by an increasing trend regardless of rootstock or PGR treatment, with approximately the same quantities produced regardless of applied treatments (  Ethylene production and respiration rate in fresh and stored plums grown on different rootstocks ("WaVit" and P. cerasifera/P. spinosa) and subjected to PGR treatments. (A-D) Ethylene production (nL g -1 h -1 ): "WaVit" at harvest (A) and after cold storage (C). P. cerasifera/P. spinosa at Figure 2. Ethylene production and respiration rate in fresh and stored plums grown on different rootstocks ("WaVit" and P. cerasifera/P. spinosa) and subjected to PGR treatments. (A-D) Ethylene production (nL g −1 h −1 ): "WaVit" at harvest (A) and after cold storage (C). P. cerasifera/P. spinosa at harvest (B) and after cold storage (D). (E-H) Respiration rate (µL g −1 h −1 ): "WaVit" at harvest (E) and after cold storage (G). P. cerasifera/P. spinosa at harvest (F) and after cold storage (H). The main factors are presented, and their significance is annotated by: *-significant at 0.05 and **-significant at 0.01. The error bars represent the standard deviation.
Rootstock type did not exert any effect on the textural properties (Table 1). PGR application induced significant changes, although, at harvest, flesh firmness was significantly lower only in BA100-treated fruits grown on P/P compared to the controls. After cold storage and shelf life, GA 3 -treated plums were firmer than the controls, and for plums grown on "WaVit," this difference was statistically significant. BA50 application on fruit grown on "WaVit" increased the skin strength at harvest and after cold storage, while after shelf life, greater skin strength was recorded in fruits grown on P/P and treated with GA 3 . Application of BA100 on fruits grown on "WaVit" rootstock significantly reduced fruit skin elasticity after both cold storage and shelf life. Table 1. Changes in flesh firmness and skin strength, and elasticity of fresh and stored plum grown on two different rootstocks ("WaVit" and P. cerasifera/P. spinosa) influenced by PGR treatment. Values designated by the same letter were not significantly different (p > 0.05). Main factors and their interactions are presented, and their significance is annotated as follows: NS-not significant, *-significant at 0.05 and **-significant at 0.01; PGR treatment: control-untreated plums; GA 3 -200 mg L −1 of gibberellin A3; BA50-50 mg L −1 of 6-benzyladenine; BA100-100 mg L −1 of 6-benzyladenine; storage: 0-at harvest; 28-28 days of cold storage; 28 + 4-28 days of cold storage followed by 4 days of shelf life; rootstock: WaVit, P. cerasifera/P. spinosa. Standard deviation of data presented in Tables 1-3 are presented in  Supplementary Table S1.

Storage
Fruit color analysis revealed that the PGRs and rootstocks exerted a significant impact on lightness (L*) in plum skin and flesh and on the intensity of red (a*) and yellow tone (b*) in the flesh (Table 2, Supplementary Figure S1).   Tables 1-3 are presented in  Supplementary Table S1. At harvest, a darker appearance was noted in the control fruit grown on "WaVit" ( Table 2). Applied PGRs reduced the difference in skin L* between fruits grown on different rootstocks. BA50 application on fruits grown on P/P significantly reduced lightness at harvest, while the L* value did not change during cold storage or shelf life. Application of PGRs showed a tendency to increase the lightness of plum flesh at harvest. After cold storage and shelf life, applied PGRs changed the flesh lightness, but their effect was rootstock-dependent. GA 3 application increased red color intensity (a*), while the effect of PGRs and rootstocks on b* did not exhibit a discernible pattern.
Based on the ANOVA analysis results, TSS and TA in plums were significantly affected by all treatments applied in the study: PGRs, rootstocks, as well as cold storage ( Table 3). Application of GA 3 resulted in higher TSS regardless of the rootstock type. At harvest, TA in plums treated with PGRs was lower in comparison to the respective controls. However, the recorded differences were less pronounced after cold storage and shelf life, especially for fruit grown on "WaVit" rootstock. Table 3. Changes in pigment, sugar and acid content in fresh and stored plums grown on two different rootstocks ("WaVit" and P. cerasifera/P. spinosa) influenced by PGR treatment. Values designated by the same letter were not significantly different (p > 0.05). Main factors and their interactions are presented, and their significance is annotated as follows: NS-not significant, **-significant at 0.01; PGR treatment: control-untreated plums; GA 3 -200 mg L −1 of gibberellin A3; BA50-50 mg L −1 of 6-benzyladenine; BA100-100 mg L −1 of 6-benzyladenine; storage: 0-at harvest; 28-28 days of cold storage; 28 + 4-28 days of cold storage followed by 4 days of shelf-life; rootstock: "WaVit", P. cerasifera/P. spinosa. Standard deviation of data presented in Tables 1-3 are presented in Supplementary Table S1.

Storage
Regardless of the significance of PGR treatments, no pattern could be established for monosaccharide (glucose + fructose) content in treated plums (Table 3). At harvest, sucrose content was not influenced by PGR treatment either, but after shelf life, it was significantly higher in treated plums. Plums grown on P/P had higher sucrose content than those grown on "WaVit" rootstock. No significant differences in succinic or malic acid content were observed between the PGR-treated plums and the controls.
The anthocyanins content was the only parameter that was not affected by cold storage but was influenced by both rootstock and PGR treatment type (Table 3). It was notable that the anthocyanin content was higher in the fruits grown on P/P compared to those grown on "WaVit". Additionally, the PGR-treated plums had higher anthocyanin content in comparison to the respective controls. Carotenoid content increased significantly during storage and shelf life and was higher in fruits grown on P/P and in fruit treated with GA 3 and higher BA concentration regardless of the storage period.

Multivariate Analysis
In order to identify and compare the directions and intensities of rootstock-and PGRdependent changes, as well as enable comparison of the magnitude of PGR and rootstock influence, multivariate principal component analysis (PCA) was performed ( Figure 3). monosaccharide (glucose + fructose) content in treated plums (Table 3). At harvest, sucrose content was not influenced by PGR treatment either, but after shelf life, it was significantly higher in treated plums. Plums grown on P/P had higher sucrose content than those grown on "WaVit" rootstock. No significant differences in succinic or malic acid content were observed between the PGR-treated plums and the controls.
The anthocyanins content was the only parameter that was not affected by cold storage but was influenced by both rootstock and PGR treatment type (Table 3). It was notable that the anthocyanin content was higher in the fruits grown on P/P compared to those grown on "WaVit". Additionally, the PGR-treated plums had higher anthocyanin content in comparison to the respective controls. Carotenoid content increased significantly during storage and shelf life and was higher in fruits grown on P/P and in fruit treated with GA3 and higher BA concentration regardless of the storage period.

Multivariate Analysis
In order to identify and compare the directions and intensities of rootstock-and PGRdependent changes, as well as enable comparison of the magnitude of PGR and rootstock influence, multivariate principal component analysis (PCA) was performed (Figure 3). Values designated by the same letter were not significantly different (p > 0.05). Main factors and their interactions are presented, and their significance is annotated as follows: NS-not significant, **-significant at 0.01; PGR treatment: control-untreated plums; GA3-200 mg L −1 of gibberellin A3; BA50-50 mg L −1 of 6-benzyladenine; BA100-100 mg L −1 of 6-benzyladenine; storage: 0-at harvest; 28-28 days of cold storage; 28 + 4-28 days of cold storage followed by 4 days of shelf-life; rootstock: "WaVit", P. cerasifera/P. spinosa. Standard deviation of data presented in Tables 1, 2 and 3  are presented in Supplementary Table S1. Regardless of the significance of PGR treatments, no pattern could be established for monosaccharide (glucose + fructose) content in treated plums (Table 3). At harvest, sucrose content was not influenced by PGR treatment either, but after shelf life, it was significantly higher in treated plums. Plums grown on P/P had higher sucrose content than those grown on "WaVit" rootstock. No significant differences in succinic or malic acid content were observed between the PGR-treated plums and the controls.
The anthocyanins content was the only parameter that was not affected by cold storage but was influenced by both rootstock and PGR treatment type (Table 3). It was notable that the anthocyanin content was higher in the fruits grown on P/P compared to those grown on "WaVit". Additionally, the PGR-treated plums had higher anthocyanin content in comparison to the respective controls. Carotenoid content increased significantly during storage and shelf life and was higher in fruits grown on P/P and in fruit treated with GA3 and higher BA concentration regardless of the storage period.

Multivariate Analysis
In order to identify and compare the directions and intensities of rootstock-and PGRdependent changes, as well as enable comparison of the magnitude of PGR and rootstock influence, multivariate principal component analysis (PCA) was performed ( Figure 3). The findings indicate that the first two principal components jointly explain >65% of the variability, with 52.71% attributed to the first and 13.60% to the second principal component. Clear segregation of the samples with respect to the first principal component in relation to the storage duration can also be noted, differentiating fruits analyzed at harvest, after cold storage and after shelf life. Moreover, fruits after cold storage can be distinguished from other samples concerning the second principal component. In each cluster of samples stored for different time periods, differentiation based on the rootstock type can also be made for the first principal component. The first principal component is highly correlated with monosaccharide content (with a 0.93 correlation coefficient obtained for both fructose and glucose), texture properties (skin strength −0.86, fruit firmness −0.96 and skin elasticity 0.96) as well as color properties (skin L*, a* and b*, −0.81,−0.87 and −0.66, respectively, and flesh L*, a* and b*, −0.93, 0.86 and −0.83, respectively).
Differentiation of non-treated fruits from those treated with PGRs, particularly with BA, within clusters of samples stored for different time periods can also be made with respect to the second principal component, confirming the physicochemical properties and composition of the studied fruit are influenced by PGR treatment. Still, in comparison to the changes induced by rootstock type, and particularly by prolonged cold storage, this influence is minor (PC1 = 52.71%; PC2 = 13.60%). The second principal component is negatively correlated with sucrose (−0.68) and acid (−0.73 and −0.68 for malic and succinic acid, respectively) content, as well as positively correlated with anthocyanin and carotenoid content (0.63 and 0.32, respectively).

Discussion
Gibberellins and cytokinines are involved in the regulation of numerous plant development processes. These plant growth regulators (PGRs) accumulate in the early phases of fruit development, during which cell division and cell expansion occur. Empirical evidence further indicates that higher concentrations of these compounds during the fruit ripening phase inhibit the ripening process [46]. Gibberellins are known to increase cell elongation, while cytokinines influence cell division [57]. Thus, their application in the early phase of fruit development, as was done in the present study, induces beneficial changes to the morphological properties of the fruit (fruit weight, size, diameter, dimensions and volume) and thus to fruit production [45,46,[58][59][60][61][62]. However, several authors have challenged this assertion, arguing that these effects are not significant and are dependent on the production season or variety [35,59,63].
In our investigation, the application of cytokinine (BA) and gibberellin (GA 3 ) induced changes in at least one of the measured fruit shape parameters, indicating modifications in fruit morphology. These findings confirm that the expected alterations occurred in the fruit development phase due to increased concentration of PGRs ( Figure 1). The most prominent effects were exerted by BA50, as fruit width and thickness changed irrespective of the rootstock used. The observed differences in the morphological properties of fruit grown on different rootstocks are consistent with those published in pertinent literature, where the influence of rootstock on plum yield, fruit and stone dimensions and weight is well documented for both European and Japanese plum [49,51,[64][65][66][67][68][69][70][71]. Numerous findings confirm that GA 3 and BA significantly impact the physical properties and chemical composition of harvested fruit. Still, the majority of extant research focusing on the effects of PGR application on plums was carried out on Japanese plums [4,45,[58][59][60][61]63], while the European plums are rarely studied [46,47].
The majority of plum varieties, with rare exceptions, are categorized as climacteric [72]. In these fruits, the ripening processes, accompanied by the changes in fruit composition that leads to decay, are closely related to increased ethylene production [28,73,74]. Recent findings [75] related to the effects of both gibberellins and cytokinins on the ripening process in different climacteric fruit species emphasize the inhibitory role of higher concentrations of these PGRs, which may be associated with the inhibition of ethylene production. However, the underlying molecular mechanisms which control the ripening processes, the complete set of ripening-related genes and signal transduction have yet to be fully discovered and understood [75]. Our results indicate that ethylene production increased in fresh fruit as well as after prolonged cold storage. Thus, it can be assumed that the inhibiting mechanisms of GA 3 or BA on ethylene production were not triggered in our study, probably due to the PGR application in the early phase of fruit development, i.e., long before the ripening stage.
The intensity of fruit respiration, as a series of biochemical processes resulting in the oxidation of sugars to CO 2 [76], is another factor contributing to the postharvest quality losses [77]. Increased CO 2 production shortens shelf life and leads to the degradation of traits that determine fruit acceptability for consumers [78][79][80]. As in the present study, PGR application during the early fruit development phase did not affect the respiration rate of ripe fruit after harvest. The differences in respiration cannot be correlated with PGR use.
The main plum composition parameters affected by PGR application-as indicated by the PCA findings ( Figure 3)-were pigment content (including both anthocyanins and carotenoids), as well as sugar (i.e., sucrose) and acid content. Higher sucrose, anthocyanin and carotenoid contents were noted in PGR-treated plums at harvest, and these values increased after prolonged cold storage and shelf life (Table 3). These findings indicate that PGR application might have significant positive effects on the fruit's sensory properties, visual appeal and taste, as well as the nutritional profile (in terms of higher content of bioactive compounds) during the postharvest period when most fruit is consumed. A significant increase in the polyphenol (especially anthocyanin) content in stored fruits was also reported for multiple treatments of plums with salicylates [81]. In the present study, anthocyanin and carotenoid content increased irrespective of the rootstock type; however, higher values were obtained for both pigments on fruits grown on P/P.
Due to the high sugar content in soluble solids, plums can be stored at low temperatures, close to 0 • C. These storage conditions slow down postharvest ripening and decrease biochemical processes, allowing successful preservation of fruit quality for up to five weeks [77]. The changes after cold storage, regardless of the applied PGR treatment or rootstock, are followed by the expected postharvest fruit behavior, described in pertinent literature for both Japanese and European plums [27,32,33,77]. Thus, the main focus of our discussion will not be on the postharvest ripening process but rather on the differences between the fruit treated with PGRs and non-treated plums in the postharvest period.
Our findings showed that, following GA 3 and BA application, the TA values declined at harvest in comparison to non-treated fruits (Table 3). Moreover, GA 3 -treated plums in all investigated stages had higher TSS content at harvest when compared to the relevant controls. Additionally, present findings suggest that PGRs impact the sucrose, anthocyanin and carotenoid contents compared to the controls. While the observed increase in TSS, carotenoid and anthocyanin, along with the decrease in TA in PGR-treated fruits, could be attributed to increased fruit ripening, the noted improvements in texture do not support this assumption. Thus, it is likely that PGR application in the early stages of development targets specific biochemical pathways rather than the whole ripening process.
Based on the results reported in this work, it can be concluded that the application of BA and GA 3 impacts texture, flesh color, TSS, TA, sucrose and carotenoid content (Tables 1-3). The data reported in extant literature confirm that PGRs presence can induce changes in the fruit quality parameters (including TSS, TA and fruit firmness) at harvest. However, the findings obtained by other authors are inconsistent. For example, several reports on different varieties of P. domestica and P. salicina indicate that TSS increases due to GA (GA 3 ) treatment [45,47,60,62], while TA decreases [59,62,63], and fruit firmness remains unaffected [61]. On the other hand, an ample body of evidence suggests that fruit firmness increases as a result of GA 3 application [45,46,[58][59][60][61][62], which has no influence on TSS [58,59,61,63].
The rootstock-induced differences in tree productivity and fruit properties at harvest could be the reason for the different responses of the grafted cultivar "Čačanska rodna" to PGR treatments at harvest and during the postharvest period [82,83]. P. spinosa is used as the interstock with P. cerasifera as the rootstock for apricots and plums to reduce the trunk diameter and tree growth and enable high-density planting [50,84]. It induces early bearing, wide soil adaptation, frost resistance, larger fruit size, and better coloration [85]. On the other hand, "WaVit" is a rootstock selected from seedling populations of "Wangenheim" (P. domestica) with excellent uniformity. It reduces tree growth, induces early cropping, accelerates fruit ripening, and increases fruit size while producing high and regular yields [86,87].
Plum cultivar "Čačanska lepotica" did not significantly differ in terms of tree vigor when grown either on "WaVit" or P.cerasifera/P.spinosa, while the higher yields and yield efficiency were recorded on "WaVit" rootstock [50]. The rootstock did not affect fruit size in "Čačanska lepotica" or in the present research in "Čačanska rodna", while P.cerasifera/P.spinosa consistently induced the lower titratable acidity in fruits compared to "WaVit" rootstock.

Conclusions
Although rootstock type and PGR application impact the overall fruit quality at harvest as well as after cold storage and shelf life, these factors affect different quality parameters.
The effects of early application of BA and GA 3 at harvest manifest through positive changes (in terms of consumer acceptance) in fruit weight and shape, the color of skin and flesh, and TSS, TA and pigment content. However, changes to fruit weight and shape are rootstock-dependent. The initial differences between PGR treatments and rootstocks in terms of TSS, TA, and pigment content persist after cold storage. Still, additional differences in fruit texture emerge while the color difference diminishes. After shelf life, TSS and TA remain treatment-and rootstock-dependent, and the same difference starts to emerge in sucrose and pigments. Based on the obtained results, it can be surmised that PGR application in the early phases of fruit development results in different postharvest fruit biochemistry.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/horticulturae8070621/s1, Figure S1: Changes of plum fruit after cold storage and subsequent shelf life, Table S1: Standard deviation of data presented in Tables 1-3 (flesh firmness,

Conflicts of Interest:
The authors declare no conflict of interest.