Chemical versus mechanical dwarfing of M7 apple rootstock: contrasting pathways induced by paclobutrazol and bonsai techniques

Abstract

Background

Controlling vegetative vigour is a key part of modern high-density apple orchards that use semi-dwarf rootstocks like M7. Genetic dwarfing is widely employed; however, additional management through chemical growth retardants or physical constraints is frequently necessary to preserve optimal canopy architecture. It is still not clear if these two different approaches stop growth by using the same metabolic pathways or different stress-mediated mechanisms. The objective of this study was to elucidate the physiological, metabolic, and hormonal profiles in three-year-old M7 apple trees exposed to three different treatments: mechanical dwarfing (bonsai techniques), chemical dwarfing (paclobutrazol/PBZ), and untreated controls.

Results

Even though the two treatments used different methods, they both worked to keep shoot elongation to less than 12% of the control. Hormonal regulation was the main cause of PBZ-induced dwarfing. This was shown by a marked decline in gibberellin A3 (GA3; 63% drop) and a rise in zeatin. This change in hormones was significant starch accumulation and better absorption of macronutrients with few signs of stress. In contrast, dwarfing through bonsai techniques caused a systemic stress response, as shown by a 284% rise in the stress osmolyte sorbitol and a 190% rise in abscisic acid (ABA). This group also had low nutrient levels, a sharp drop in the GA/ABA ratio, and clear signs of physiological stress, like lower photosynthetic efficiency and higher oxidative markers.

Conclusions

Our results show that PBZ acts as a precise biochemical regulator that produces a sink-limited state without imposing substantial physiological stress on the plant, whereas bonsai techniques control growth by causing a resource-limited, stress-driven physiology. These contrasting signatures highlight paclobutrazol as a more controlled dwarfing tool and provide a physiological framework for developing stress-resilient dwarfing rootstocks.

Background

A vital component of the world’s temperate fruit production, apple trees (Malus domestica Borkh.) are prized for both their nutritional value and economic importance. Vegetative rootstocks are essential to the intensification of contemporary orcharding systems because they regulate tree architecture, optimizing light interception, increasing labor efficiency, and facilitating better input management. The Malling 7 (M7) semi-dwarf rootstock is one of the most popular genotypes due to its demonstrated capacity to provide a fruitful balance between high fruit yield and manageable vigor [1]. The basic physiological mechanisms that differentiate its natural semi-dwarfism from externally induced dwarfing are still not fully understood, despite its significance in horticulture. As a result, thorough research into distinct chemical and mechanical growth control methods is required.

In this context, M7 is the perfect intermediate model system because it has enough vegetative plasticity to clearly show the difference between stress-induced and regulated dwarfing pathways. Understanding these specific mechanisms gives us useful information for improving canopy management in high-density systems and finding breeding targets that separate dwarfing traits from bad stress responses. The complex interactions of phytohormones, which control the developmental flexibility of shoot architecture, are essential for controlling vegetative vigor. Particularly, the homeostatic balance between growth-inhibiting abscisic acid (ABA), a crucial signaling molecule mediating adaptive responses to environmental stress, and growth-promoting gibberellins (GAs), which drive both cell division and internodal elongation, determines a tree’s ultimate stature. As a result, the majority of horticultural treatments, whether chemical or cultural, intended to reduce apple shoot growth work by essentially upsetting this crucial GA/ABA hormonal axis, albeit via wildly disparate and frequently conflicting physiological cascades [2].

The purpose of this study was to examine two such opposing paradigms: systemic, stress-mediated control using mechanical bonsai techniques and targeted biochemical regulation using paclobutrazol (PBZ). By specifically targeting the enzyme ent-kaurene oxidase, paclobutrazol, a persistent growth regulator of the triazole class, effectively inhibits GA biosynthesis, halting internode elongation and causing a compact phenotype [3]. By means of pot confinement and root pruning, bonsai techniques, on the other hand, impose physical constraints. According to Sun et al., these mechanical interventions cause systemic stress signals that lead to the noticeable buildup of ABA by lowering the root system’s absorptive capacity and impairing hydraulic conductance. Although the effectiveness of each approach is known, there is a substantial knowledge gap because their relative physiological signatures within a single experimental setup have never been methodically compared [3].

Since these hormonal changes are closely related to the nutritional status of the plant, this knowledge gap is especially important [4]. The metabolic basis for all growth and developmental processes is nutrient homeostasis, which is supported by the uptake and distribution of essential macronutrients like potassium (K), phosphorus (P), and nitrogen (N) [5, 6]. By changing sink strength in shoot tissues, controlling transporter activity, or changing root architecture, hormonal changes brought on by dwarfing agents can significantly alter nutrient dynamics. A key, unsolved question is how this highly regulated hormonal-nutritional nexus is affected differently by the systemic, ABA-driven stress model of bonsai versus the targeted, GA-suppression model of PBZ.

Unlike previous inter-genotypic comparisons (e.g., M9 vs. M26), where dwarfing arises from inherent differences in auxin transport and vascular anatomy, the present study imposes two exogenous dwarfing stimuli on the same clonal M7 rootstock, thereby isolating treatment-specific effects and enabling direct mechanistic attribution of GA/ABA shifts and sorbitol-mediated osmotic adjustment to either targeted biochemical inhibition or systemic stress signaling. Thus, the purpose of this study is to compare and contrast these two dwarfing paradigms in M7 apple rootstocks in a methodical manner. We aim to distinguish the unique, integrated physiological architecture linked to chemical versus mechanical growth control by combining quantitative analyses of phytohormonal profiles, nutritional landscapes, and important physiological stress indicators. A strong scientific foundation for developing precision horticulture and breeding new, climate-resilient fruit tree cultivars will be provided by the findings, which are expected to offer fundamental insights into the hormonal-nutritional crosstalk governing plant form.

Materials and methods

Preparation of plant materials

This study employed three-year-old acclimatised M7 apple rootstocks to guarantee established vegetative architecture. The plant materials were initially propagated through tissue culture, guaranteeing clonal uniformity and a disease-free condition, and were acquired from Royan Pazhohesh Azarbayejan Co. (Urmia, Iran). Before treatments began, the plantlets were successfully acclimatised and cultivated in a controlled greenhouse environment for three years to achieve the necessary developmental stage.

In this study, we used 15 identical M7 plants. The experiment took place in a greenhouse and used a Completely Randomised Design (CRD) with three different treatments (cultivation techniques) and five independent biological replicates for each treatment (n = 5). At the time of the evaluation, all of the plants were about three years old. Sampling took place in mid-summer (mid-July) to get the most active plants. The first group (control) consisted of 20-liter plastic pots filled with fertile loamy soil (pH 6.5–7.0, high organic matter content, and good drainage optimized for apple growth). The plants were fertigated using a balanced NPK fertilizer (10-10-10) at a rate of 2 g/L every two weeks.

Similar pot, soil, and horticultural management conditions were used to grow and care for the second group of plants. Cultar^® (25% SC paclobutrazol formulation, Syngenta Crop Protection AG, Switzerland) was dissolved in two liters of distilled water to a concentration of 0.5 g/L and applied as a soil drench at a fixed rate of 1 g active ingredient per pot (or 4 mL of Cultar^®). In order to sustain the dwarfing effects over two growing seasons, the treatment was applied once a year in the early spring (usually in mid-April, when new shoot extension reached about 5 cm). This was followed by a second application in the early spring of the following year. To improve root uptake, pots were pre-irrigated to field capacity prior to application. To reduce leaching, no additional irrigation was given for 24 h after treatment.

Uniform clonal M7 plants were carefully selected for the bonsai treatment in the third group. Shallow unglazed ceramic pots (20 cm diameter × 10 cm depth) were used in conjunction with standard bonsai tools, such as sterilized concave cutters, root hooks, pruning shears, and aluminum wire (1–3 mm gauge), to aid in root restriction and aesthetic development. To maximize aeration, nutrient uptake, and microbial activity—thereby promoting phytochemical processes like flavonoid biosynthesis and preventing root pathologies—a specialized soil mixture consisting of 40% akadama, 30% pumice, and 30% lava rock (particle size 2–5 mm) was prepared. It was amended with 5% composted pine bark and brought to a pH of 6.0–6.5 using dolomitic lime. To create a compact fibrous system, 50–70% of the root mass was pruned in the early spring. The trunk and branches were then styled in the informal upright (moyogi) manner using spiral wiring with minimal constriction to prevent vascular impairment. Finally, 30–50% of the apical buds were removed to encourage back-budding and ramification for dwarfed morphology. Nebari roots had to be aesthetically positioned, the soil depth had to be kept between 8 and 10 cm, and tie-down wires had to be used for stability. In order to maintain physiological balance, post-establishment maintenance involved biennial repotting with 30% root pruning and biweekly half-strength NPK (10-10-10) fertilization enhanced with trace elements throughout the growing season. This treatment was specifically designed to impose three simultaneous constraints to simulate mechanical dwarfing: (1) Physical root restriction (limited container volume), (2) Mechanical root pruning (limiting absorptive capacity), and (3) Nutrient limitation (via porous substrate). These elements together set apart “stress-induced” mechanical dwarfing from the biochemical control of PBZ.

The research greenhouse where the experiment took place is near Tabriz, Iran (38°01′54″ N, 46°23′36″ E). All plants were kept in controlled environments with temperatures between 18 and 25 °C and relative humidity between 65 and 80%. There was no use of photoperiod manipulation or artificial lighting. Every day, the plants were exposed to full sunlight for four hours, after which they were shaded to reduce the light intensity by 30%. Standard procedures for managing pests and diseases that were suitable for apple plants were followed.

Phytohormone extraction and purification

Extraction of abscisic acid (ABA) and Indole-3-Acetic Acid (IAA)

For phytohormone analysis, fully expanded leaves (excluding senescent basal leaves and immature apical leaves) were harvested from the middle third of the current season’s shoots. 1.0 g of tissue powder was extracted using liquid-liquid extraction (LLE) for ABA and IAA analysis. To avoid photodegradation, all processes for IAA were carried out in low light. Ten milliliters of 80% (v/v) aqueous methanol with 0.1% formic acid were used to extract ABA for twenty-four hours while being continuously shaken in the dark. Ten milliliters of 80% (v/v) aqueous methanol supplemented with 0.1% (w/v) butylated hydroxytoluene (BHT) were used to extract IAA for four hours in the dark on an orbital shaker. After extraction, the homogenates were centrifuged at 10,000 ×g for 15 min at 4 °C, and the supernatants were gathered. According to Chiwocha et al., methanol was extracted under vacuum using a rotary evaporator for ABA or a mild stream of nitrogen gas for IAA [7].

The aqueous phase was partitioned twice against an equal volume of n-hexane after being brought to pH 8.5 with 1 M NaOH for ABA purification. After using 1 M HCl to acidify the aqueous phase to pH 2.5, it was partitioned three times against an equivalent volume of ethyl acetate. Dry evaporation was performed on the combined ethyl acetate fractions. In contrast, the aqueous residue was acidified to pH 2.5 using 1 M HCl for IAA Purification, and it was then partitioned three times against an equivalent volume of diethyl ether. After that, half a volume of 5% (w/v) NaHCO₃ was used to back-extract the combined ether phases. After collecting the aqueous bicarbonate phase, it was re-acidified to pH 2.5 using 1 M HCl and then partitioned three more times against diethyl ether. Dry evaporation was performed on the final pooled ether fractions. Before High-Performance Liquid Chromatography (HPLC) analysis, the dried residues of ABA and IAA were reconstituted in 500 µL of their respective initial mobile phases and filtered through a 0.22 μm PTFE syringe filter [8].

Extraction and Solid-Phase Extraction (SPE) of zeatin

Cytokinins were extracted from 1.5 g of powdered tissue overnight using 10 mL of a pre-chilled (−20 °C) modified Bieleski buffer (methanol/water/formic acid, 15:4:1, v/v/v), following a modified protocol by Novák et al. (2003). After centrifugation (15,000 ×g, 20 min, 4 °C), the supernatant was dried under vacuum. The residue was redissolved in 1 mL of 1% (v/v) formic acid and purified using an Oasis MCX mixed-mode cation-exchange SPE cartridge (60 mg, Waters). The cartridge was conditioned with 3 mL of methanol and equilibrated with 3 mL of 1% formic acid. After sample loading, the cartridge was washed sequentially with 3 mL of 1% formic acid and 3 mL of 50% methanol. Cytokinins were eluted with 3 mL of 5% (v/v) NH₄OH in 60% (v/v) methanol. The eluate was evaporated to dryness under N₂ and reconstituted in 200 µL of the initial mobile phase for HPLC analysis [9].

Extraction, SPE, and derivatization of gibberellins (GAs)

Gibberellins were extracted from 2.0 g of powdered tissue with 10 mL of 80% (v/v) aqueous methanol containing 0.1% formic acid and 0.1% BHT for 12 h. The extract was centrifuged (15,000 ×g, 20 min, 4 °C), and the methanol was removed from the supernatant under reduced pressure. The remaining aqueous phase was loaded onto a C18 SPE cartridge (500 mg) pre-conditioned with methanol and water. The cartridge was washed with deionized water, and GAs were eluted with 80% methanol. The eluate was evaporated to dryness under N₂.

For derivatization, the dried residue was redissolved in 100 µL of anhydrous acetonitrile. 1 mg of p-bromophenacyl bromide (PBPB) was added along with catalytic amounts of 18-crown-6 and 2 mg of anhydrous K₂CO₃. The mixture was heated at 70 °C for 45 min in a sealed vial. After cooling, the sample was filtered (0.22-µm PTFE membrane filter) for HPLC analysis [10].

HPLC analysis of phytohormones

Chromatographic analyses were performed on an HPLC system (Knauer, Berlin, Germany) comprising a mixing chamber, two Model 64 pumps, and a UV/VIS detector. Separations were conducted on a Knauer Eurospher I 100–5 C18 column (250 × 4.6 mm, 5 μm particle size) protected by a compatible C18 guard column. The injection volume for all analyses was 20 µL. Specific chromatographic conditions for each phytohormone are detailed in Table 1.

Table 1.

Sample Preparation and chromatographic analysis of plant hormones

Analyte	Mobile Phase A	Mobile Phase B	Gradient Program (Flow Rate)	λ (nm)
ABA	0.1% Acetic Acid in H₂O	0.1% Acetic Acid in ACN	(1.0 mL/min) 0–5 min, 30% B; 5–20 min, 30→80% B; 20–25 min, 80→100% B; 5 min wash (100% B); 5 min re-equilibration (30% B)	254
IAA	0.1% Acetic Acid in H₂O	0.1% Acetic Acid in MeOH	(1.0 mL/min) 0–5 min, 35% B; 5–25 min, 35→90% B; 25–30 min, 90% B; 5 min re-equilibration (35% B)	280
Zeatin	50 mM Ammonium Acetate (pH 4.0)	ACN	(0.8 mL/min) 0–5 min, 5% B; 5–30 min, 5→50% B; 5 min wash (95% B); 5 min re-equilibration (5% B)	269
GA₃	Deionized H₂O	ACN	(1.0 mL/min) 0–20 min, 50→100% B	260

Phytohormones were identified by comparing retention times with those of authentic standards (Sigma-Aldrich). Quantification was based on external standard calibration curves constructed with at least six concentration points for each analyte. All calibration curves demonstrated high linearity (R² > 0.99). For GA analysis, standards were subjected to the identical derivatization procedure as the samples to generate the calibration curve. All experiments were performed using three independent biological replicates, and each biological replicate was analyzed in three technical replicates.

Analysis of non-structural carbohydrates

Fully expanded, sun-exposed leaves of 3-year-old M7 apple rootstock were sampled mid-season (10 Aug–10 Sep) between 06:00 and 07:00 AM, flash-frozen in liquid N₂ and stored at − 80 °C. Lyophilized tissue was ground to a fine powder and 50 mg DW was extracted with 3 × 5 mL 80% (v/v) ethanol at 80 °C (xylitol, 10 µg, was added as an internal standard prior to extraction); pooled extracts were evaporated to near dryness, reconstituted in 1.0 mL HPLC-grade water, filtered (0.22 μm) and analyzed by HPLC–RID (Aminex HPX-87 C column, 300 × 7.8 mm, Bio-Rad) at 85 °C with water as the mobile phase (0.6 mL min⁻¹; 20 µL injection). Residual pellets were gelatinized (0.2 M KOH, 95 °C, 30 min), neutralized, digested with heat-stable α-amylase and amyloglucosidase to glucose, which was quantified by HPLC–RID and converted to starch using a factor of 0.90. Sorbitol, sucrose, glucose and fructose were identified by retention time vs. authentic standards and quantified using external calibration curves (≥ 5 levels, R² ≥ 0.99) with IS normalization. Method validation included LOD/LOQ (S/N 3/10 or 3.3·SD/slope and 10·SD/slope), matrix-spike recoveries (low/med/high), intra- and inter-day precision (CV%), and pooled QC injections every 8–12 samples. Data are reported as mg g⁻¹ DW (FW conversions provided using measured moisture) [11].

Lignin content analysis

The acid-insoluble lignin content was determined using a modified Klason lignin procedure. Prior to hydrolysis, approximately 1 g of milled wood powder was rendered extractive-free by exhaustive Soxhlet extraction, typically with a 2:1 (v/v) toluene-ethanol mixture for 8 h, followed by drying. A known mass of the extractive-free sample (200 mg) was then subjected to a two-step acid hydrolysis. First, the sample was treated with 3.0 mL of 72% (w/w) sulfuric acid and maintained at 30 °C for 1 h with periodic stirring to hydrolyze the polysaccharides. Subsequently, the mixture was diluted with 84 mL of deionized water to a final acid concentration of 3% (w/w) and autoclaved at 121 °C for 1 h. The acid-insoluble residue was collected by vacuum filtration through a pre-weighed, medium-porosity sintered glass crucible, washed with hot deionized water until the filtrate was neutral, and dried at 105 °C to a constant weight. The lignin content was calculated gravimetrically and expressed as a percentage of the initial extractive-free dry sample mass [12].

Leaf mineral nutrient analysis

Total nitrogen (N) was determined by the micro-Kjeldahl method. Briefly, dried and ground leaf samples were digested with concentrated sulfuric acid (H₂SO₄) in the presence of a catalyst to convert organic nitrogen into ammonium sulfate. The resulting digest was then steam-distilled after alkalinization with sodium hydroxide (NaOH), and the liberated ammonia was quantified by titration. For phosphorus (P) and potassium (K) analysis, leaf samples were acid-digested using a microwave digestion system (CEM Corporation) with HNO₃ and H₂O₂. Elemental concentrations in the digest were quantified by inductively coupled plasma optical emission spectrometry (ICP-OES) (Agilent 5110), as described by Kalra [13].

Guaiacol peroxidase (GPX) activity assay

Crude enzyme extracts for GPX (EC 1.11.1.7) activity were prepared from 0.5 g of fresh leaf tissue in an ice-cold extraction buffer (potassium phosphate buffer, EDTA, and PVP). GPX activity was assayed by monitoring the rate of guaiacol oxidation to tetraguaiacol by the increase in absorbance at 470 nm (ε = 26.6 mM⁻¹cm⁻¹) in the presence of H₂O₂ [14].

Chlorophyll a fluorescence measurement

The maximum quantum efficiency of photosystem II (Fv/Fm) was measured on intact leaves after a 30-minute dark-adaptation period using a portable chlorophyll fluorometer (MINI-PAM-II, Walz, Germany). This parameter was used as an indicator of the photosynthetic apparatus’s integrity and potential efficiency [15].

Data analysis

All statistical analyses were performed using IBM SPSS Statistics for Windows (Version 26.0; IBM Corp., Armonk, NY, USA). Prior to analysis, data were assessed for the assumptions of normality and homogeneity of variances using the Shapiro-Wilk and Levene’s tests, respectively. A one-way analysis of variance (ANOVA) was employed to determine significant differences among treatments. Mean separation was subsequently conducted using Tukey’s Honestly Significant Difference (HSD) post hoc test.

To investigate the interrelationships among key physiological and biochemical variables, Pearson correlation coefficients were calculated. This analysis included shoot cumulative increase, phytohormone levels (ABA, IAA, zeatin, GA3), nutrient concentrations (N, P, K), lignin content, GPX activity, chlorophyll fluorescence (Fv/Fm), and carbohydrate profiles (fructose, glucose, sorbitol, starch, sucrose). Given the exploratory nature of this analysis, no adjustments for multiple comparisons were made. For all statistical tests, significance was accepted at p < 0.05.

Results

Shoot growth dynamics

Over the course of three months, from March 21 to June 22 (2023), the shoot growth increment (%) was tracked in three treatment groups of apple rootstock M7. At the end of the experiment, visual evaluation showed that the treatment groups had remarkably different phenotypes (Fig. 1). Strong vegetative growth was demonstrated by the control plants, which had a canopy of healthy, well-expanded leaves and extensive shoot elongation. On the other hand, the application of paclobutrazol resulted in a highly compact shoot architecture with a significantly reduced internodal length; these plants’ dense foliage frequently displayed slight morphological aberrations like leaf crinkling. The most severe dwarfing effect was caused by the bonsai regimen, which led to a significant decrease in both the overall stature of the plant and the area of the lamina on individual leaves. This phenotype is consistent with a systemic, stress-induced suppression of growth.

Fig. 1.

Comparative morphology of M7 apple rootstocks following chemical (paclobutrazol) and mechanical (bonsai) dwarfing treatments. Control plants (left) exhibit vigorous growth with normal shoot elongation and leaf expansion. Paclobutrazol application (center) induced a characteristic compact phenotype with severely compressed internodes and dense foliage. The bonsai regimen (right) resulted in the most profound dwarfing, evident by a drastic reduction in height, smaller leaf size, and a stressed appearance linked to the visibly confined root system

The control group’s shoot growth increased gradually and linearly, as shown in Fig. 2, starting at a nearly zero increment in late March and reaching about 22% by June 22. This growth was consistent across replicates. With a first plateau phase until mid-April and a subsequent slow increase to roughly 11% by the end of the period, the paclobutrazol-treated group showed a delayed and attenuated growth pattern. With the smallest error bars, which indicate uniform suppression, and a nearly flat trajectory with few increments, the bonsai group showed the most controlled growth, reaching about 5% by June 22. Significant differences between groups were confirmed by statistical analysis (one-way ANOVA). At all-time points after April 21, the control group’s shoot growth increments were significantly higher than those of the paclobutrazol-treated and bonsai-treated groups, according to the post-hoc Tukey’s HSD test. Furthermore, starting on May 22, the group treated with paclobutrazol showed noticeably higher growth than the bonsai group. The control and paclobutrazol groups demonstrated increasing significance over time, whereas within-group temporal variations for the bonsai treatment were non-significant.

Fig. 2.

Cumulative increase in shoot growth (%) of M7 rootstock under different treatments over a 90-day period from spring emergence. Data points represent the mean of 5 replicates, and error bars indicate the standard error of the mean

Photosystem II quantum efficiency

Significant photoinhibitory stress was found in dwarfed M7 apple trees based on the maximum quantum efficiency of photosystem II (PSII), which is measured by the Fv/Fm ratio. The control plants had a healthy baseline Fv/Fm value of about 0.79 and a strong photosynthetic apparatus. Conversely, both dwarfing treatments caused a statistically significant drop in Fv/Fm, which reached its lowest value of ~ 0.68 in the bonsai group and dropped to ~ 0.70 in plants treated with paclobutrazol. A clear connection between dwarfing phenotypes and physiological stress is established by this decrease in photosynthetic efficiency, which is in good agreement with our metabolic and hormonal data (Fig. 3).

Fig. 3.

Photosystem II Maximum Quantum Efficiency (Fv/Fm Ratio) in Dwarfed M7Apple Trees Treated with Paclobutrazol or Subjected to Bonsai Cultivation

Phytohormone concentrations

When paclobutrazol (PBZ) was applied and bonsai was grown, the endogenous hormonal profile of M7 apple rootstock showed clear, treatment-specific changes (Table 2). Abscisic acid (ABA) rose quantitatively from 35.0 ± 0.8 ng·g⁻¹ in control plants to 60.1 ± 1.6 ng·g⁻¹ after PBZ treatment and to 101.5 ± 3.2 ng·g⁻¹ in bonsai conditions. The levels of indole-3-acetic acid (IAA) decreased from 35.6 ± 0.8 ng·g⁻¹ (control) to 29.2 ± 1.3 ng·g⁻¹ (PBZ) and 21.3 ± 0.8 ng·g⁻¹ (bonsai). In bonsai plants, zeatin showed divergent responses, rising to 24.2 ± 0.7 ng·g⁻¹ with PBZ but falling to 11.4 ± 0.5 ng·g⁻¹ (control: 19.7 ± 1.7 ng·g⁻¹). The concentrations of gibberellic acid (GA) were significantly reduced from 27.2 ± 0.9 ng·g⁻¹ in controls to 10.0 ± 0.8 ng·g⁻¹ (PBZ) and 11.3 ± 0.4 ng·g⁻¹ (bonsai). Figure 4, which depicts the dynamic changes in phytohormone concentrations (ABA, IAA, zeatin, GA) and the GA/ABA ratio across control, PBZ, and bonsai treatments, provides a visual context for these hormonal changes. The figure notably shows the sharp drop in the GA/ABA ratio, which is a major factor in the growth-dormancy transition, from 0.78 in controls to 0.17 (PBZ) and 0.11 (bonsai).

Table 2.

Alterations in the endogenous hormonal profile of M7 Apple rootstock in response to Paclobutrazol and bonsai cultivation

	Abscisic Acid	Indole-3-acetic Acid	Zeatin	Gibberellic Acid
Control	35.0 ± 0.8	35.6 ± 0.8	19.7 ± 1.7	27.2 ± 0.9
Paclobutrazol	60.1 ± 1.6	29.2 ± 1.3	24.2 ± 0.7	10.0 ± 0.8
Bonsai	101.5 ± 3.2	21.3 ± 0.8	11.4 ± 0.5	11.3 ± 0.4

Fig. 4.

Phytohormone Dynamics and GA/ABA Ratio Modulation in Plants under Paclobutrazol and Bonsai Treatments

Non-structural carbohydrates concentration

The amounts of non-structural carbohydrates in the leaf tissues of apple trees subjected to control, paclobutrazol (PBZ), and bonsai treatments are compiled in Table 3. While bonsai treatment produced intermediate starch levels (56.6 ± 2.3 mg g⁻¹ DW), PBZ-treated plants showed significantly higher starch accumulation (80.8 ± 1.3 mg g⁻¹ DW) compared to controls (36.2 ± 2.1 mg g⁻¹ DW). Under bonsai conditions, the concentration of sorbitol increased significantly (31.5 ± 1.2 mg g⁻¹ DW), roughly four times higher than controls (8.2 ± 0.6 mg g⁻¹ DW) and three times higher than plants treated with PBZ (10.5 ± 0.7 mg g⁻¹ DW). In comparison to untreated controls (12.3 ± 1 mg g⁻¹ DW), sucrose levels decreased in both dwarfing treatments, with PBZ causing the most severe reduction (5.6 ± 0.5 mg g⁻¹ DW) and bonsai causing a moderate decrease (7.7 ± 0.8 mg g⁻¹ DW). Both glucose and fructose concentrations increased in both treatments; however, the increases linked to PBZ were more noticeable for glucose (12.8 ± 1.2 mg g⁻¹ DW vs. 5.9 ± 0.7 mg g⁻¹ DW in controls) and fructose (12.7 ± 0.8 mg g⁻¹ DW vs. 4 ± 0.5 mg g⁻¹ DW in controls). Bonsai-treated plants had the highest levels of total soluble sugars (61.7 mg g⁻¹ DW), followed by PBZ (41.6 mg g⁻¹ DW) and controls (30.4 mg g⁻¹ DW). In comparison to controls (1.19), the starch-to-soluble sugar ratio rose in PBZ-treated plants (1.94), but fell in bonsai (0.92).

Table 3.

Concentrations of Non-structural carbohydrates in Apple leaf tissues under Control, Paclobutrazol, and bonsai treatments (Units: Mg g⁻¹ dry Weight)

	Control	Paclobutrazol	Bonsai
Starch	36.2 ± 2.1	80.8 ± 1.3	56.6 ± 2.3
Sorbitol	8.2 ± 0.6	10.5 ± 0.7	31.5 ± 1.2
Sucrose	12.3 ± 1	5.6 ± 0.5	7.7 ± 0.8
Glucose	5.9 ± 0.7	12.8 ± 1.2	8.1 ± 0.9
Fructose	4 ± 0.5	12.7 ± 0.8	14.4 ± 1.5
Total Soluble Sugars	30.4	41.6	61.7
Starch: Soluble Sugar Ratio	1.19	1.94	0.92

Guaiacol peroxidase activity

Spectrophotometric analysis of the guaiacol peroxidase (EC 1.11.1.7; GPX) activity in one-year-old M7 apple tree branches revealed a baseline activity of roughly 0.6 µmol min⁻¹ mg⁻¹ protein in control samples. In bonsai-cultivated plants, this activity peaked at 1.2 µmol min⁻¹ mg⁻¹ protein after paclobutrazol treatment, with minimal standard errors highlighting methodological reliability. It then increased to 1.0 µmol min⁻¹ mg⁻¹ protein. These increases, which amounted to 67% and 100% increases over controls, respectively, demonstrated that dwarfed plants had more pronounced oxidative stress responses and lignification (Fig. 5).

Fig. 5.

Guaiacol Peroxidase (GPX) Activity in One-Year-Old Branches of M7Apple Trees Under Control, Paclobutrazol, and Bonsai Treatments

Lignin concentration

The Klason method of quantifying lignin in one-year-old M7 apple tree branches showed that the control groups had a baseline concentration of about 25% dry weight (DW). The lignin concentration increased to 40% DW under paclobutrazol treatment, which was 60% higher than controls. In comparison to controls, bonsai cultivation produced an intermediate elevation to 35% DW, or a 40% increase. ANOVA analysis validated statistically significant differences among treatments, and minimal standard errors across experimental replicates confirmed high precision. In dwarfed phenotypes, these increases highlighted improved secondary cell wall fortification (Fig. 6).

Fig. 6.

Lignin Concentration (%) Dry Weight in One-Year-Old M7Apple Branches Under Control, Paclobutrazol, and Bonsai Treatments

Primary macronutrients concentration

Nitrogen (N), phosphorus (P), and potassium (K) concentrations varied significantly among the three treatment groups, according to leaf tissue analyses (Table 4). Baseline values for the control group, which was fed optimally without growth retardants, were 2.21 ± 0.09% N, 0.24 ± 0.011% P, and 1.81 ± 0.07% K. Conversely, plants treated with paclobutrazol, which exhibited noticeable signs of dwarfing (compact canopy, decreased internode length), had higher levels of nutrients: 2.71 ± 0.07% N (22.6% increase compared to control), 0.33 ± 0.01% P (37.5% increase), and 2.13 ± 0.1% K (17.7% increase). On the other hand, bonsai-grown plants that were subjected to wire training, root pruning, and pot volume restrictions showed lower levels of nutrients: 1.76 ± 0.06% N (20.4% decrease), 0.17 ± 0.009% P (29.2% decrease), and 1.66 ± 0.08% K (8.3% decrease). These variations demonstrated different physiological reactions to chemical versus mechanical dwarfing techniques and were statistically significant (P < 0.05; one-way ANOVA followed by Tukey’s HSD test).

Table 4.

Macronutrient concentrations (% dry weight) in leaf tissues of M7 Apple rootstocks under different dwarfing treatments. Values are means ± SE (n = 10). Different letters indicate significant differences (P < 0.05)

	Nitrogen	Phosphorus	Potassium
Control	2.21 ± 0.09	0.24 ± 0.011	1.81 ± 0.07
Paclobutrazol	2.71 ± 0.07	0.33 ± 0.01	2.13 ± 0.1
Bonsai	1.76 ± 0.06	0.17 ± 0.009	1.66 ± 0.08

Covariation of phytohormonal, biochemical, and growth parameters under dwarfing treatments

Leaf samples from M7 apple rootstocks were subjected to a Pearson correlation analysis in order to clarify the relationships between important physiological, biochemical, and hormonal parameters under various dwarfing circumstances. The 16 measured variables were found to have a highly structured network of significant positive and negative associations, according to the correlation matrix (Fig. 7). Gibberellic acid (GA3; r = 0.989) and sucrose (r = 0.954) showed a nearly perfect positive correlation with shoot cumulative increase, the main indicator of vegetative vigor. Additionally, there were strong positive correlations with the maximum quantum efficiency of photosystem II (Fv/Fm; r = 0.632) and indole-3-acetic acid (IAA; r = 0.720). On the other hand, lignin content (r = −0.964), fructose (r = −0.940), starch (r = −0.901), guaiacol peroxidase activity (GPX; r = −0.857), glucose (r = −0.793), and abscisic acid (ABA; r = −0.682) were all strongly and negatively connected with the cumulative increase in shoots. It had a negative, but not statistically significant, correlation with sorbitol (r = −0.460). The macronutrients nitrogen (N), phosphorus (P), and potassium (K) did not significantly correlate with shoot cumulative increase (r = −0.163, −0.188, and − 0.330, respectively). There was a clear antagonistic relationship between the hormonal parameters. ABA and GA3 had a strong negative correlation (r = −0.710). Zeatin showed a very strong positive correlation with N, P, and K (r = 0.959, 0.902, and 0.909, respectively) but not with shoot growth (r = 0.035). Stress and metabolism biochemical markers were closely related. GPX activity had a negative correlation with Fv/Fm (r = −0.497, non-significant) but a positive correlation with ABA (r = 0.861) and lignin (r = 0.828). Carbohydrate profiles demonstrated a distinct dichotomy: fructose, glucose, and starch were strongly and positively intercorrelated and negatively associated with growth, while sucrose was positively associated with growth promoters (e.g., GA3, r = 0.914). Fv/Fm had a negative correlation with fructose (r = −0.612) and lignin (r = −0.611), but a positive correlation with growth-promoting hormones such as GA3 (r = 0.612).

Fig. 7.

Correlation matrix of physiological and biochemical parameters in M7 apple

Discussion

Shoot growth dynamics

The vigorous M7 rootstocks exhibited the highest shoot elongation under unrestricted conditions, driven by optimal gibberellin-mediated cell expansion in the presence of sufficient root space and nutrients [16]. Consistent with its established mode of action—specifically inhibiting ent-kaurene oxidase in the gibberellin biosynthetic pathway—paclobutrazol treatment resulted in an intermediate dwarfing phenotype (11% growth) [17]. This regulated suppression effectively mimics the physiological efficiency observed in genetically dwarfing rootstocks like M9, yet it achieves this through transient biochemical inhibition rather than fixed anatomical constraints [18]. Conversely, bonsai cultivation resulted in the most severe dwarfing (5% growth). The chronic root restriction associated with this method induces a systemic suppression response, likely limiting hydraulic conductance and root-derived growth signals. Unlike the targeted regulation by paclobutrazol, the bonsai constraints impose a prolonged growth arrest analogous to severe environmental stress. As noted by Aarif et al. [19], such moderate shoot suppression (as seen in PBZ) is highly desirable for high-density planting, whereas the severe restriction of bonsai provides a model for studying survival limits under resource scarcity.

Photosynthesis and carbon allocation

Both dwarfing treatments were associated with a decrease in Photosystem II efficiency (Fv/Fm), yet the underlying mechanisms appear distinct. The moderate decrease in PBZ-treated leaves is consistent with feedback inhibition caused by sink limitation. When shoot elongation is biochemically halted, photoassimilates accumulate in source leaves (evidenced by high starch), leading to down-regulation of electron transport to prevent oxidative damage [20]. In contrast, bonsai plants exhibited a slightly greater decline in Fv/Fm, which likely reflects stomatal closure triggered by elevated ABA levels and root confinement [21]. Additionally, the increased GPX activity and accumulated osmoprotectant sorbitol in these bonsai leaves likely serve as partial defenses [22]. This stomatal limitation restricts CO₂ availability, increasing the susceptibility of the photosynthetic apparatus to photo-inhibition. Thus, while both methods reduce photosynthetic efficiency, PBZ does so via downstream sink feedback, whereas bonsai acts via upstream stomatal limitations.

Hormonal reprogramming

Dwarfing conditions coincided with profound alterations in hormonal balance, favoring stress resilience over elongation. Growth-promoting hormones (GA and auxin) were significantly lower in both treatments, while the stress-signaling hormone ABA was elevated. While PBZ directly targets GA biosynthesis enzymes [17], the reduction of GA in bonsai plants is likely a downstream consequence of systemic stress signaling. Notably, ABA accumulation was markedly higher in bonsai plants (190% increase) compared to the PBZ group. This specific hormonal signature distinguishes the mechanical stress of bonsai from the physiology of standard dwarfing rootstocks like M9 or M26. In M9/M26, dwarfing is typically attributed to impaired basipetal auxin transport or altered tissue sensitivity, rather than the massive systemic ABA accumulation observed here under root restriction [23]. Cytokinin (zeatin) levels also showed different responses. Zeatin levels were slightly higher in shoots treated with paclobutrazol. This may have helped maintain meristem activity even though GA levels were low. Conversely, bonsai plants exhibited a significant decrease in zeatin, presumably due to reduced synthesis at the root tips [24]. These results indicate that PBZ functions mainly as a specific biochemical modulator, while bonsai techniques seem to initiate a wide-ranging stress-adaptive cascade predominantly governed by ABA. So, the GA/ABA antagonism seems to be a major factor in controlling vegetative stature, with a high ABA/low GA ratio being strongly linked to the dwarfed phenotype [25]. These findings suggest that chemical dwarfing may provide more consistent hormonal control of tree size, whereas traditional bonsai techniques probably engage a broader spectrum of stress-related signaling. More research is needed to find out how genetic differences in rootstock affect these hormone networks to improve dwarfing strategies.

Non-structural carbohydrates concentration

The divergence in carbohydrate profiles provides critical insight into the osmotic status of the plants. PBZ-treated plants accumulated significant amounts of starch and hexoses, reinforcing the “sink-limited” hypothesis where carbon supply exceeds the demand for structural growth. In sharp contrast, the carbohydrate profile of bonsai plants was defined by a dramatic accumulation of sorbitol. In Rosaceae species, sorbitol acts as a primary osmolyte and compatible solute. Its accumulation under bonsai conditions serves as a vital mechanism for osmotic adjustment, allowing the plant to maintain cellular turgor and water retention despite the severe hydraulic constraints imposed by root pruning and pot confinement [11]. This accumulation suggests that bonsai plants actively metabolize carbon for stress survival (osmoprotection), whereas PBZ plants passively store excess carbon due to growth arrest.

GPX activity and lignification

“Both dwarfing strategies improved oxidative defenses and enhanced cell wall fortification, but to different degrees. The trees that were treated with paclobutrazol had the most lignin and higher GPX activity. This suggests that GA inhibition may change the flow of metabolism into the phenylpropanoid pathway, where GPX helps monolignol polymerization to make cell walls stiffer [26, 27]. In bonsai plants, the strong activation of GPX, along with higher levels of ABA and sorbitol, suggests that GPX plays a dual role: scavenging ROS that stress causes and helping lignification [28]. Therefore, PBZ appears to promote a type of “developmental” lignification that happens when plants mature, while bonsai probably starts an “active” defense response to reduce oxidative stress [29]. From a horticultural standpoint, PBZ-induced lignification may bolster branch durability, while stress-induced lignification in bonsai could confer drought resilience; however, excessive wall thickening may undermine vascular flexibility and efficiency. Overall, these results show the trade-off between structural reinforcement and hydraulic function. This means that future transcriptomic analysis of phenylpropanoid genes is needed to fully understand these regulatory networks.”

Nutrient dynamics

Nutrient acquisition patterns revealed a fundamental trade-off. Paclobutrazol-treated plants maintained higher levels of N, P, and K compared to controls. This accumulation likely results from a “concentration effect” due to reduced shoot biomass combined with sustained root activity, as ABA at moderate levels can upregulate certain ion transporters [30]. Conversely, bonsai plants exhibited significant depletion of macronutrients. The physical constraints of root pruning and limited soil volume mechanically restrict the absorptive surface area, overwhelming any compensatory uptake mechanisms [18, 31]. Consequently, bonsai plants appear susceptible to nutrient deficiencies, implying that fertilization strategies may need to be modified. These results underscore a distinct physiological divergence: paclobutrazol appears to sustain root activity and potentially enhance nutrient use efficiency, suggesting possible benefits for reducing fertilizer inputs in high-density orchard systems. Conversely, bonsai cultivation likely necessitates precise nutritional management to mitigate depletion. Future research integrating transcriptomics and nutrient profiling may further elucidate the hormonal regulation of nutrient absorption in dwarfed phenotypes.

Integrated physiological trade-offs

Correlation analysis delineated two distinct physiological states. The “growth-oriented” state (Control) showed a strong positive association between shoot elongation, GA, and sucrose, supporting the model where high GA promotes sink strength and utilization of photosynthates [25, 32]. In contrast, the “defense-oriented” state (Dwarfing) was characterized by elevated ABA, lignification, and starch/sorbitol accumulation [26]. The nearly perfect positive correlation between shoot growth and GA or sucrose suggests that GA-mediated sucrose transport to growing shoots may be a primary driver of elongation. Conversely, the strong negative correlations between growth and ABA, lignin, fructose, glucose, and starch indicate that during shoot suppression, photosynthates likely accumulate in source tissues while cells transition towards rigid secondary wall formation. This implies that dwarfing may accelerate the shift from expandable primary cell walls to lignified secondary walls [33], thereby limiting further expansion. These trade-offs suggest that dwarfed plants dynamically reallocate resources, prioritizing either growth (associated with low ABA and lignin) or stress adaptation (associated with high ABA and lignin). Consequently, shoot growth appears to serve as a reliable indicator of this physiological balance [26].

Conclusion

This study demonstrates that while both paclobutrazol and bonsai techniques effectively suppress shoot elongation in M7 apple rootstocks, they elicit distinct physiological signatures. Paclobutrazol-induced dwarfing was associated with a specific inhibition of gibberellin biosynthesis and a sink-limited metabolic state (high starch, maintained nutrients) with minimal physiological stress. In contrast, bonsai techniques imposed growth restriction through a systemic stress response, characterized by elevated ABA, significant sorbitol accumulation, and nutrient depletion.

These findings have significant implications for modern horticulture. For commercial orchard management, the use of targeted chemical regulators like PBZ offers a viable strategy to achieve ‘precision dwarfing’ in high-density systems without compromising the tree’s nutritional status or inducing detrimental stress. Conversely, the specific stress markers identified in the bonsai model—particularly high ABA sensitivity and sorbitol-mediated osmotic adjustment—highlight potential targets for breeding programs. Selecting for these traits could facilitate the development of new ‘climate-resilient’ rootstocks that decouple desirable dwarfing characteristics from negative stress symptoms, ultimately contributing to more sustainable and resource-efficient orchard systems.

Authors’ contributions

MGT grew plants, executed the experiments and collected the samples, FZN designed and supervised the research and wrote the manuscript, MD analyzed the results and edited the manuscript, MHG performed nutrient analysis, all authors read and approved the final manuscript.

Funding

This research received no external funding.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This article does not contain any studies with human participants performed by any of the authors. Ethics approval and consent to participate were therefore not required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.