Effects of Climatic Conditions and Genotypes on Vitamin C Levels in Stone Species and Apple Cultivars

Abstract

Genetic and climatic factors influence the nutritional content of fruit, with vitamin C being a key component. Using HPLC, we quantified the amount of vitamin C in cherries, apricots, plums, and apples from 2022 to 2024. Contents ranged from 1.6 to 24.6 mg/100 g fresh weight basis (FW), with apples and plums displaying the highest coefficient of variation (32.53% and 45.25%). The highest content was consistently found in accession ‘HL827’, which exceeded 20 mg/100 g FW. Cherries reached up to 12.1 mg/100 g FW in 2023 (‘13590′), but decreased to 1.9 mg/100 g FW in 2024 (‘Jacinta’). Apricots showed high fluctuation, with ‘Betinka’, ‘Candela’, and ‘HL08-052’ exceeding the 30% variance coefficient. Accessions that remained stable (‘HL96-266’) maintained a low variance only. Plums were the most sensitive, experiencing low vitamin C content under hot and dry conditions. Regression analysis identified temperature (NTavg-20) as the dominant climatic driver in plums and cherries (R² = 0.999, p < 0.05 and R² = 0.995, p < 0.05) respectively, whereas apples and apricots showed negligible responses (R² ≤ 0.210). These findings underscore the importance of genotype/environment interactions at the local level and highlight the value of stable accessions as valuable resources for breeding cultivars with high and resilient vitamin C content.

1. Introduction

Vitamin C exists primarily as L-ascorbic acid (AA) and its oxidized form L-dehydroascorbic acid (DHA). Vitamin C has strong antioxidant properties and plays an essential role in several physiological functions, including collagen synthesis, supporting the immune support, and enhancing the absorption of non-heme iron [1]. In plants, vitamin C biosynthesis involves the Smirnoff–Wheeler pathway, which converts GDP-mannose to L-galactono-1,4-lactone. This compound is then oxidized to AA by L-galactono-1,4-lactone dehydrogenase [2]. In fresh fruit, AA often represents over 85% of total vitamin C [3]. However, postharvest storage at ambient temperature can reduce AA by around 50% within a week due to its oxidation to DHA. Although DHA is biologically active, it is more labile and prone to irreversible hydrolysis [4]. In addition to its direct antioxidant action, AA regenerates other antioxidants, such as α-tocopherol, and modulates plant responses to oxidative stress [5].

Recent studies have detailed how the balance between AA and DHA homeostasis is regulated at multiple metabolic nodes. This regulation includes feedback inhibition by downstream products and the upregulation of GDP-L-galactose phosphorylase under stress [6]. These findings help explain how climatic conditions can alter the balance between AA and DHA in harvested fruit.

Climatic conditions prior to harvest, particularly rainfall patterns and intra-seasonal temperature fluctuations during critical phenological stages, significantly impact vitamin C biosynthesis and stability [7]. One study demonstrated that temperature fluctuations during ripening decreased AA levels by 25% compared to stable conditions. Lee and Kader [8] found that excessive rainfall during ripening diluted soluble solids, including AA. In contrast, drought stress can alter ascorbate metabolism, thereby increasing or decreasing AA levels depending on the severity and timing of the stress. High night-time temperatures can also downregulate key enzymes in the Smirnoff–Wheeler pathway, resulting in reduced ascorbate accumulation [6].

The vitamin C content of fruits can vary widely across species and accessions. Sweet cherries (Prunus avium L.) range from 3 to 16 mg vitamin C per 100 g FW, with darker-skinned cultivars reaching approximately 18–24 mg/100 g FW [9,10,11,12]. Apricots (Prunus armeniaca L.) may contain up to 20 mg/100 g FW vitamin C [13], whereas some cultivars, such as ‘Shalakh’ may reach markedly higher concentrations, up to approximately 80 mg 100 g FW under specific growing conditions [14]. Plums (Prunus domestica L.) generally contain 1–25 mg/100 g FW vitamin C [15,16], while wild or selected germplasm accessions may reach 20–54 mg/100 g FW [17,18]. Apples (Malus domestica Borkh.) range from 1 to 32 mg/100 g FW depending on the cultivar and climatic conditions [19,20,21].

Abiotic and biotic stresses, such as extreme temperatures, pest pressure, and orchard management practices (including irrigation, pruning, and harvest timing) affect vitamin C content. Lee and Kader [8] reported that fully mature fruit contained about 30% less AA than fruit harvested slightly earlier. Phillips et al. [4] demonstrated that storing homogenates at 4 °C maintained over 90% of the initial vitamin C content for up to 14 days. Furthermore, differences in canopy microclimate within the same orchard can lead to intra-tree variation in AA content exceeding 15% [22]. The influence of weather conditions prior to harvest on the stability of AA content in fruits of different species and cultivars has not yet been sufficiently researched. This study provides a thorough comparison of vitamin C contents in commercial and newly bred cultivars of sweet cherries, apricots, plums, and apples cultivated under temperate Central European conditions. It highlights the importance of newly developed accessions for breeding programs. The study also evaluates the influence of seasonal climatic factors, such as temperature variability and rainfall distribution, on vitamin C content. The selected species are widely cultivated in Central Europe and are the focus of ongoing breeding efforts at our institution, providing unique access to both commercial cultivars and newly developed accessions.

Our integration of biochemical insights with cultivar and climate comparisons addresses a key gap in the literature, offering breeders and growers robust data to guide the selection of accessions with superior and stable nutritional profiles. Although the study was conducted under Central European orchard conditions over a three-year period, the identified genotype–environment patterns are likely relevant to temperate production systems with comparable climatic variability. While extrapolation to contrasting climatic zones requires caution, the dataset provides a structured comparative framework for evaluating species-specific stability and climatic sensitivity in fruit antioxidant metabolism.

2. Results

2.1. Differences in Vitamin C Content Among Accessions

Substantial genotypic variability in vitamin C content was observed across the four fruit species evaluated and all years studied, ranging from 1.58 to 21.60 mg/100 g FW. These differences underscore the species- and genotype-specific control of ascorbic acid accumulation.

In cherries, the mean value ranged from 6.85 to 14.70 mg/100 g FW (Figure 1) expressing a moderate coefficient of variance (16.68%) in the vitamin C content of fruits. Accession ‘13420’ had the highest average (14.70 mg/100 g). This value was more than twice that of ‘Justyna’ (6.85 mg/100 g) and about 35% higher than that of the commercial cultivar ‘Early Korvik’ (10.89 mg/100 g). These numerical differences were statistically significant (p < 0.001) and biologically meaningful, indicating a nearly twofold genotypic range within commercially relevant accessions. Conversely, accession ‘16755’ (7.81 mg/100 g) clustered with the lowest group and differed significantly from mid-range cultivars such as ‘Kordia’ (10.53 mg/100 g; p < 0.01) and ‘Amid’ (10.32 mg/100 g; p < 0.01).

Figure 1.

Mean vitamin C content (mg/100 g fresh weight) of cherry fruit across 2022–2024. Values are presented as the mean ± confidence interval (CI, n = 9). Different letters above the bars indicate statistically significant differences among accessions according to Tukey’s HSD test at p less than 0.05.

The vitamin C content of apricots ranged from 5.88 to 11.25 mg/100 g FW giving a moderate coefficient of variance of 17.29%. The top accessions ‘VOJ5/150’ (11.25 mg/100 g), ‘HL08-052’ (10.90 mg/100 g), and ‘HL08-018’ (10.88 mg/100 g) significantly exceeded the standard cultivars such as ‘Sophinka’ (5.88 mg/100 g) and ‘Betinka’ (6.01 mg/100 g). However, they did not exceed ‘Harogem’ (9.39 mg/100 g) (Figure 2). The accession ‘HL97-052’ (8.05 mg/100 g) did not differ significantly from ‘Sophinka’ or ‘Betinka’, but it had a significantly lower content compared to ‘Harogem’ (p < 0.05).

Figure 2.

Mean vitamin C content (mg/100 g fresh weight) in the fruits of apricot accessions across 2022–2024. Values are presented as the mean ± confidence interval (CI, n = 9). Different letters above the bars indicate statistically significant differences among accessions according to Tukey’s HSD test at p less than 0.05.

The plum accessions exhibited the highest coefficient of variance 45.25% in vitamin C, ranging from 1.58 to 9.60 mg/100 g FW. The nearly sixfold difference between ‘HL9900004’ (1.58 mg/100 g) and ‘HLT1-10’ (9.60 mg/100 g) indicates strong genotypic control (Figure 3). Notably, ‘HLT1-10’ had values that were more than three times higher than ‘Tophit’ (2.71 mg/100 g) and ‘HL0635’ (2.50 mg/100 g). The coefficient of variance for vitamin C in apple accessions was 32.53%, giving a range overall, from 7.24 to 21.60 mg/100 g FW. ‘HL827’ (21.60 mg/100 g) greatly surpassed ‘Frosta’ (13.87 mg/100 g) by about 55% and was nearly three times higher than ‘Rubinstep’ (7.24 mg/100 g) (Figure 4.). Mid-performers such as ‘HL1579’ (13.09 mg/100 g) and ‘HL1311’ (12.73 mg/100 g) clustered with commercial cultivars. Meanwhile, ‘HL1343’ (8.56 mg/100 g) and ‘HL1651’ (8.41 mg/100 g) formed the low tier.

Figure 3.

Mean vitamin C content (mg/100 g fresh weight) in the fruits of plum accessions across 2022–2024. Values are presented as mean ± confidence interval (CI, n = 9). Different letters above the bars indicate statistically significant differences among accessions according to Tukey’s HSD test at p less than 0.05.

Figure 4.

Mean vitamin C content (mg/100 g fresh weight) in the fruits of apple accessions across 2022–2024. Values are presented as mean ± confidence interval (CI, n = 9). Different letters above the bars indicate statistically significant differences among accessions according to Tukey’s HSD test at p less than 0.05.

2.2. Effect of Year and Climatic Conditions on Vitamin C Content

Across the three experimental years (2022–2024), significant interannual variability in climatic conditions was observed. The year 2022 was characterized by the highest daily maximum temperatures (up to 36.9 °C in August) and low July rainfall (34 mm), as shown in Table 1. The year 2023 featured elevated rainfall totals (115 mm in August) and moderate maximum temperatures (Table 2). The year 2024 exhibited intermediate patterns. This natural gradient provided a useful framework for assessing the interaction between year-to-year climatic variation and genotype under Central European orchards conditions, which experience both extremes and moderate seasons.

Table 1.

ANOVA results for vitamin C content (mg/100 g FW) in apples, apricots, cherries, and plums. F-values, p-values are presented for the factors “Accession,” “Year,” and their interaction. For “Year,” mean values and Tukey HSD grouping as letters a, b, and c are shown.

Species	Factor	Df	SS	MS	F-Value	p-Value	Means 2022	Means 2023	Means 2024
Cherries	Accession	23	565.1	24.57	8.453	<0.001	9.775 b	11.609 a	8.327 c
	Year	2	389.6	194.79	46.46	<0.001
	Accession × Year	23	153.6	6.68	2.298	<0.01
Apricots	Accession	16	352.3	22.02	12.284	<0.001	10.057 a	8.635 b	8.442 b
	Year	2	79.4	39.70	8.092	<0.001
	Accession × Year	16	183.1	11.44	6.385	<0.001
Plums	Accession	16	548.1	34.26	20.251	<0.001	4.256 ab	4.958 a	3.722 b
	Year	2	39.2	19.606	3.469	<0.05
	Accession × Year	16	130.2	8.14	4.812	<0.001
Apples	Accession	15	1722.2	114.81	65.214	<0.001	12.572 a	10.215 b	10.148 b
	Year	2	183.0	91.48	6.076	<0.01
	Accession × Year	15	245.5	16.36	9.295	<0.001

Df degrees of freedom, SS sums of squares, MS mean squares.

Table 2.

Mean vitamin C content (mg/100 g FW) of apricot accessions from the years from 2022 to 2024. Values are expressed as mean ± standard deviation (SD = 3). For particular cultivars and accessions, the coefficient of variance was calculated for the years 2022–2024.

Apricot Accession	Vitamin C Content (mg/100 g Fresh Weight)			Coefficient of Variance
	2022	2023	2024
HL01-091	10.32 ± 0.60	8.66 ± 0.72	9.65 ± 0.24	9.14
HL02-024	8.51 ± 0.16	11.41 ± 0.21	10.62 ± 0.90	13.56
HL08-002	9.52 ± 0.20	10.69 ± 0.15	11.08 ± 1.01	8.40
HL08-009	10.70 ± 0.46	6.58 ± 0.38	8.04 ± 0.20	21.77
HL08-018	12.24 ± 0.25	9.05 ± 0.46	11.36 ± 0.82	13.84
HL96-266	9.06 ± 0.20	10.20 ± 0.13	9.77 ± 0.70	6.42
HL96-599	10.48 ± 0.37	8.14 ± 0.15	6.92 ± 0.39	18.70
Betinka	8.71 ± 0.21	4.87 ± 0.12	4.44 ± 0.02	33.97
Candela	12.85 ± 0.20	6.35 ± 0.04	4.92 ± 0.24	45.57
Harogem	10.74 ± 0.35	9.83 ± 0.71	7.61 ± 0.21	15.47
HL08-015	7.12 ± 0.04	9.28 ± 0.85	7.85 ± 0.30	13.01
HL08-052	15.84 ± 0.87	6.43 ± 0.12	10.42 ± 0.22	37.76
HL96-288	12.07 ± 1.14	8.68 ± 0.34	8.69 ± 0.14	18.31
HL97-052	7.03 ± 0.11	9.34 ± 0.21	7.78 ± 0.45	13.05
Sophinka	7.38 ± 0.16	4.93 ± 0.44	5.34 ± 0.06	19.74
VOJ 5/147	7.72 ± 0.08	9.65 ± 0.30	8.69 ± 0.33	9.96
VOJ 5/150	10.68 ± 1.21	12.72 ± 0.50	10.33 ± 0.29	11.60

The year effect was statistically significant in all species, with peak values occurring in 2022 for apples (12.57 mg/100 g FW, F = 6.08, p < 0.01) and apricots (10.06 mg/100 g FW, F = 8.09, p < 0.001). Peak values occurred in 2023 for cherries (11.61 mg/100 g FW, F = 46.46, p < 0.001) and plums (4.96 mg/100 g FW, F = 3.47, p < 0.05). A consistent decline was observed across all species in 2024, with plums reaching a minimum of 3.72 mg/100 g FW (Table 1).

In contrast, the increased rainfall in 2023 likely resulted in higher vitamin C levels in cherries (up to 12.14 ± 0.07 mg/100 g FW in accession ‘13590’) and plums (‘HLT1-10’ at 15.17 ± 0.30 mg/100 g FW) (Table 3 and Table 4).

Table 3.

Mean vitamin C content (mg/100 g FW) of plum accessions from the years from 2022 to 2024. Values are expressed as mean ± standard deviation (SD = 3). For particular cultivars and accessions, the coefficient of variance was calculated for the years 2022–2024.

Plum	Vitamin C Content (mg/100 g Fresh Weight)			Coefficient of Variance
Accession	2022	2023	2024
HL0400011	2.84 ± 0.13	2.85 ± 0.01	2.53 ± 0.10	6.50
HL0600012	3.74 ± 0.08	3.65 ± 0.22	2.02 ± 0.04	26.95
HL0624	4.24 ± 0.05	5.12 ± 0.08	2.26 ± 0.02	32.73
HL0635	2.51 ± 0.18	2.55 ± 0.11	2.45 ± 0.07	4.76
HL0653	4.18 ± 0.11	8.89 ± 0.59	7.05 ± 0.46	31.16
HL0800011	3.61 ± 0.01	3.74 ± 0.20	1.66 ± 0.10	33.71
HL0900045	6.96 ± 0.24	7.28 ± 0.29	5.70 ± 0.17	11.35
HL0900090	3.06 ± 0.17	5.86 ± 0.38	4.83 ± 0.23	27.27
HL0900097	3.42 ± 0.24	3.97 ± 0.17	7.29 ± 0.04	37.26
HL0900134	3.52 ± 0.01	3.22 ± 0.20	2.47 ± 0.04	15.70
HL0900151	4.69 ± 0.41	3.21 ± 0.22	1.81 ± 0.10	39.12
HL0900208	3.82 ± 0.03	3.57 ± 0.10	3.15 ± 0.15	8.69
HL9900004	3.29 ± 0.17	2.85 ± 0.09	1.59 ± 0.06	30.00
HLT1-10	9.01 ± 0.07	15.17 ± 0.30	4.61 ± 0.43	47.95
Presenta	5.43 ± 0.12	4.55 ± 0.34	3.07 ± 0.26	24.35
Tophit	3.00 ± 0.32	3.13 ± 0.19	1.98 ± 0.05	21.31
Toptaste	5.03 ± 0.08	4.68 ± 0.31	8.79 ± 0.40	32.34

Table 4.

Mean vitamin C content (mg/100 g FW) of cherry accessions from 2022 to 2024. Values are expressed as mean ± standard deviation (SD, n = 3). For particular cultivars and accessions, the coefficient of variance was calculated for the years 2022–2024.

Cherries	Vitamin C Content (mg/100 g Fresh Weight)			Coefficient of Variance
Accession	2022	2023	2024
10364	8.58 ± 0.33	13.70 ± 0.53	7.50 ± 0.31	29.12
13420	15.23 ± 0.11	15.50 ± 0.48	13.36 ± 0.59	7.35
13467	8.94 ± 0.45	12.16 ± 0.60	10.68 ± 0.34	13.75
13477	12.64 ± 0.36	11.66 ± 0.18	7.31 ± 0.25	23.43
13590	7.57 ± 0.02	12.14 ± 0.07	7.41 ± 0.12	25.74
13976	9.26 ± 0.21	10.73 ± 0.34	7.85 ± 0.18	13.65
14580	11.28 ± 0.51	12.18 ± 0.11	9.11 ± 0.20	12.87
15361	8.10 ± 0.45	11.90 ± 0.25	5.49 ± 0.17	33.02
15585	10.03 ± 0.43	9.64 ± 0.09	9.33 ± 0.69	5.27
16678	8.08 ± 0.29	11.31 ± 0.23	7.85 ± 0.37	18.65
16686	12.75 ± 0.72	12.90 ± 0.48	10.01 ± 0.70	12.73
16705	6.38 ± 0.26	10.26 ± 0.23	7.40 ± 0.14	21.87
16735	8.15 ± 0.37	10.10 ± 0.17	8.42 ± 0.28	10.67
16736	9.70 ± 0.31	15.24 ± 0.57	10.06 ± 0.92	23.53
16755	7.86 ± 0.43	7.35 ± 0.25	8.20 ± 0.12	5.78
16764	10.05 ± 0.41	12.54 ± 0.30	8.77 ± 0.47	16.22
16772	8.67 ± 0.58	13.04 ± 0.30	9.93 ± 0.29	18.77
16862	11.21 ± 0.09	10.96 ± 0.17	8.89 ± 0.78	11.34
16865	11.01 ± 0.89	10.39 ± 0.47	8.40 ± 0.65	13.30
Amid	9.30 ± 0.11	12.52 ± 0.32	9.13 ± 0.53	16.32
Early Korvik	12.38 ± 0.40	12.54 ± 0.20	7.74 ± 0.09	21.79
Jacinta	9.06 ± 0.05	11.19 ± 0.36	1.89 ± 0.04	57.24
Justyna	7.57 ± 0.09	7.20 ± 0.07	5.78 ± 0.41	12.33
Kordia	10.79 ± 0.67	11.47 ± 0.29	9.34 ± 0.14	9.59

Upon examining each species individually, it was found that cherries are highly sensitive to temperature and moisture. NTavg-20 exhibited a statistically significant negative relationship with vitamin C content (R² = 0.995, t = −14.77, p < 0.05) (Figure 5A). Relatively strong, though non-significant, negative associations were also shown by rainfall (R² = 0.940, t = −3.968, p > 0.05) and NTmax-30 (R² = 0.883, t = −2.751, p > 0.05). RH (R² = 0.754, t = −1.750, p > 0.05), NTdif-12 (R² = 0.496, t = 0.992, p > 0.05), and NTmin-10 (R² = 0.390, t = −0.800, p > 0.05) exhibited weaker associations (Table 5). Despite the limited strength of the models due to the small sample size, these results suggest that seasonal mean temperature is an important the primary climatic driver of cherries, while other variables have weaker, more indirect effects.

Figure 5.

Relationship between mean vitamin C content in cherry (A), apricot (B), plum (C), and apple (D) accessions and the number of days with mean temperature higher than 20 °C (2022–2024). Linear regression results are shown; significant relationships (p less than 0.05) are marked with an asterisk.

Table 5.

Effect of climatic variables on mean vitamin C content (mg/100 g FW) of four fruit species (2022–2024). Results are from linear regression; R², t, and p-values are reported (df = 1). Significance: * p < 0.05.

Species	Variables	R2	t-Value	p-Value
Cherries	RH	0.754	−1.750	0.331
	Rainfall	0.940	−3.968	0.157
	NTmin-10	0.390	−0.800	0.571
	NTmax-30	0.883	−2.751	0.222
	NTdif-12	0.496	0.992	0.503
	NTavg-20	0.995	−14.77	0.043 *
Apricots	RH	0.283	−0.629	0.643
	Rainfall	0.041	0.208	0.870
	NTmin-10	0.651	−1.364	0.403
	NTmax-30	0.145	−0.412	0.751
	NTdif-12	0.462	−0.927	0.524
	NTavg-20	0.012	−0.110	0.930
Plums	RH	0.838	−2.276	0.264
	Rainfall	0.881	−2.722	0.224
	NTmin-10	0.494	−0.989	0.504
	NTmax-30	0.942	−4.026	0.155
	NTdif-12	0.392	0.802	0.570
	NTavg-20	0.999	−26.61	0.024 *
Apples	RH	0.210	−0.515	0.697
	Rainfall	0.082	0.299	0.815
	NTmin-10	0.567	−1.145	0.457
	NTmax-30	0.090	−0.315	0.806
	NTdif-12	0.547	−1.099	0.470
	NTavg-20	0.001	−0.024	0.985

Climatic variables: RH = relative humidity; Rainfall = monthly precipitation during ripening (mm); NTmin-10 = number of nights with minimum temperature ≤10 °C; NTmax-30 = number of days with maximum temperature ≥30 °C; NTdif-12 = number of days with a diurnal temperature range ≥12 °C; NTavg-20 = number of days with mean temperature ≥20 °C.

The high but non-significant R² value for rainfall suggests that an adequate water supply may partially offset this stress. However, genotype-specific responses also play a significant role. This pattern is reflected in the performance of the accessions: ‘13590’ remained relatively stable (7.41–12.14 mg/100 g FW, highest in 2023 at 12.14 ± 0.07 mg/100 g FW), whereas ‘Jacinta’ fluctuated widely (1.89–11.19 mg/100 g FW, lowest in 2024 at 1.89 ± 0.04 mg/100 g FW) (Table 4). Standard cultivars such as ‘Early Korvik’ (9.85–11.98 mg/100 g FW, p < 0.05), ‘Kordia’ (9.62–11.22 mg/100 g FW, p < 0.05), and ‘Amid’ (9.38–11.09 mg/100 g FW, p < 0.05) exhibited moderate yet significant interannual changes.

Apricots exhibited similar strong year effects, albeit with distinct patterns. The strongest association, though still non-significant, was with NTmin-10 (R² = 0.651, t = −1.364, p > 0.05). Moderate yet weak associations were observed for NTdif-12 (R² = 0.462, t = −0.927, p > 0.05) and RH (R² = 0.283, t = −0.629, p > 0.05). Very weak effects were found for NTmax-30 (R² = 0.145), rainfall (R² = 0.041), and NTavg-20 (R² = 0.012) (Table 5; Figure 5B). In our dataset, ‘Betinka’, ‘HL08-052’, and ‘Candela’ showed sharp fluctuations with coefficient of variation = 33.97%, 37.76%, and 45.57%, respectively, whereas ‘HL96-266’ remained stable (coefficient of variation = 6.42%) (Table 2). These results illustrate the dominant role of genetics over climate.

Plums demonstrated the extremes in year-to-year variation, and confirmed that average seasonal temperature was a key driver. Among the climatic variables, only NTavg-20 showed a statistically significant association with vitamin C content (R² = 0.999, t = –26.61, p < 0.05) (Figure 5C). Other variables displayed high yet non-significant correlations: NTmax-30 (R² = 0.942, t = −4.026, p > 0.05), rainfall (R² = 0.881, t = −2.722, p > 0.05), and RH (R² = 0.838, t = −2.276, p> 0.05). Lower associations were found for NTmin-10 (R² = 0.494, t = −0.989, p > 0.05) and NTdif-12 (R² = 0.392, t = 0.802, p > 0.05) (Table 5). These patterns suggest that mean temperature likely has the strongest effect among the analyzed climatic variables, with moisture potentially acting as a secondary buffer. Our results mirrored these patterns. The high-vitamin accession ‘HLT1-10’ peaked at 15.17 ± 0.30 mg/100 g FW in 2023, then declined sharply to 4.61 ± 0.43 mg/100 g FW in 2024 (p < 0.001). In contrast, the stable accession ‘HL0400011’ remained within a narrow range of 2.53–2.84 mg/100 g FW (p > 0.05) (Table 3). Standard cultivars such as ‘Presenta’ (3.07–5.43 mg/100 g FW) and ‘Tophit’ (1.98–3.00 mg/100 g FW) exhibited moderate fluctuations.

Apples demonstrated significant, yet controlled, fluctuations. Under the studied central European conditions, the climate had little influence. These results emphasize the predominance of genetic factors in apples, though responses may vary in other production regions. None of the evaluated climatic variables had a significant effect (all p > 0.05). The strongest associations were found with NTmin-10 (R² = 0.567, t = −1.145, p > 0.05) and NTdif-12 (R² = 0.547, t = −1.099, p > 0.05), followed by RH (R² = 0.210, t = −0.515, p > 0.05). Rainfall (R² = 0.082, t = 0.299, p > 0.05), NTmax-30 (R² = 0.090, t = −0.315, p > 0.05), and NTavg-20 (R² = 0.001, t = −0.024, p > 0.05) had negligible influence (Table 5; Figure 5D). In our dataset, the elite accession ‘HL827’ consistently outperformed other cultivars (18.24 ± 0.52 → 24.57 ± 1.11 mg/100 g FW across years, p < 0.001). The cultivars and accessions with low variance in vitamin C content were ‘Reluga’, ‘HL1282’, and ‘HL2350’ expressing a coefficient of variance of 6.42%, 7.19%, and 7.49%, respectively (Table 6). The rest of the standard cultivars showed rather moderate variance: ‘Frosta’ 13.73%, ‘Idared’ 17.04%, and ‘Rubinstep’ 21.13%. The highest variance in vitamin C content among the observed years was found with accessions ‘HL1343’ and ‘HL1651’.

Table 6.

Mean vitamin C content (mg/100 g FW) of apple accessions from 2022 to 2024. Values are expressed as mean ± standard deviation (SD, n = 3). For particular cultivars and accessions, the coefficient of variance was calculated for the years 2022–2024.

Apple	Vitamin C Content (mg/100 g Fresh Weight)			Coefficient of Variance
Accession	2022	2023	2024
Frosta	15.08 ± 0.27	15.17 ± 0.30	11.35 ± 0.31	13.73
HL1194	11.30 ± 0.50	10.34 ± 0.07	12.74 ± 0.86	10.10
HL1282	8.80 ± 0.08	7.53 ± 0.35	8.24 ± 0.21	7.19
HL1311	13.90 ± 0.48	11.20 ± 0.35	13.08 ± 0.13	9.70
HL1343	11.44 ± 0.06	9.40 ± 0.39	4.84 ± 0.17	34.30
HL1579	16.41 ± 0.26	10.09 ± 0.09	12.77 ± 0.38	21.07
HL1651	11.69 ± 0.84	9.10 ± 0.55	4.42 ± 0.11	38.45
HL2010	9.00 ± 0.49	6.61 ± 0.12	9.46 ± 0.22	16.24
HL2350	9.62 ± 0.53	9.73 ± 0.42	8.58 ± 0.54	7.49
HL308	17.13 ± 0.69	12.18 ± 0.50	8.54 ± 0.16	29.78
HL53	9.49 ± 0.41	7.07 ± 0.15	5.21 ± 0.19	25.82
HL601	13.27 ± 0.37	8.58 ± 0.79	10.44 ± 0.62	19.66
HL827	21.98 ± 0.39	18.24 ± 0.52	24.57 ± 1.11	13.10
Idared	14.41 ± 0.28	12.09 ± 0.93	9.81 ± 0.45	17.04
Reluga	10.05 ± 0.15	10.75 ± 0.73	9.54 ± 0.12	6.42
Rubinstep	7.58 ± 0.41	5.35 ± 0.21	8.78 ± 0.20	21.13

3. Discussion

3.1. Genotypic Determinants of Vitamin C Accumulation

The strong genotypic effect observed in sweet cherries, as demonstrated in Section 2, is consistent with previous reports demonstrating substantial variability in ascorbic acid content among Prunus avium cultivars [23]. Nevertheless, the present study did not assess gene expression, enzyme activities, or allelic variation; therefore, the observed differences cannot be directly attributed to specific changes in individual biosynthetic genes such as GDP-L-galactose phosphorylase (GGP). Current understanding of vitamin C metabolism indicates that ascorbate accumulation is controlled by a highly integrated regulatory network rather than by single rate-limiting steps. Recent syntheses emphasize that regulation of the Smirnoff–Wheeler pathway operates at multiple levels, including feedback control of GGP translation via upstream open reading frames, transcriptional and post-transcriptional modulation, light-dependent signaling, subcellular compartmentalization, and tight coupling with mitochondrial respiration and broader metabolic processes [24,25]. These studies also show that attempts to increase ascorbate content through simple overexpression of individual pathway genes frequently result in only moderate changes due to multilayered feedback and metabolic constraints. Within this conceptual framework, the twofold differences observed among cherry accessions in our dataset most likely reflect genotype-specific integration of biosynthesis, recycling, transport, and developmental regulation, rather than being attributable to variation in a single biosynthetic enzyme or isolated regulatory step. Together, these processes—biosynthesis, recycling, degradation, and intracellular transport—form an integrated regulatory network that likely underpins genotypic differences not only in cherries but also in the other fruit species examined in this study.

Accessions such as ‘13420’ could reflect differences in regulatory control that lead to the higher expression of GDP-L-galactose phosphorylase (GGP). GGP is the rate-limiting step in the Smirnoff–Wheeler pathway [26,27]. Beyond GGP, other enzymes of the Smirnoff–Wheeler pathway, such as GME (GDP-mannose 3,5-epimerase), GMP (GDP-mannose pyrophosphorylase), and GalLDH (L-galactono-1,4-lactone dehydrogenase), also exert regulatory control. Allelic variation in these steps may contribute to cultivar-specific accumulation patterns [28]. In contrast, cultivars such as ‘Justyna’ may exhibit reduced biosynthetic efficiency or increased ascorbate turnover via enzymes such as ascorbate peroxidase (APX), which contributes to ascorbate oxidation [27]. Beyond biosynthesis, recycling processes are critical for maintaining steady-state pools. Enzymes such as monodehydroascorbate reductase (MDHAR) and dehydroascorbate reductase (DHAR) regenerate ascorbate from its oxidized forms, thereby sustaining intracellular contents. Genotypic differences in MDHAR/DHAR activity have been reported in fruits and are considered important contributors to variability in ascorbate stability [27,29]. Transport processes, including vacuolar sequestration and plastidial ascorbate transporters, also influence ascorbate distribution within fruit tissues affecting whole-fruit content [28].

The mechanistic framework described above for cherries likely also applies to plums, where genotype-dependent variation in vitamin C accumulation has been reported [26]. Nisar et al. [26] documented substantial differences in vitamin C content and antioxidant activity among plum genotypes. Within this integrative regulatory context, high-performing accessions such as ‘HLT1-10’ may reflect more efficient coordination of biosynthetic and recycling pathways, whereas low-performing genotypes may indicate reduced metabolic flux or enhanced turnover [27,29]. Overall, this integrated network of biosynthesis, recycling, degradation, and compartmentalization provides a coherent explanation for the broad genotypic range observed across species. In addition, anatomical characteristics may contribute: genotypes with thicker or more intensely pigmented peels (e.g., ‘Toptaste’, 6.17 mg/100 g; ‘HL0653’, 6.71 mg/100 g) may exhibit higher whole-fruit vitamin C concentrations due to differential tissue distribution.

Variation in pigmentation and secondary metabolism also contributes to these differences. Dark-fleshed cultivars often accumulate both anthocyanins and ascorbate in a coordinated manner, which enhances their antioxidant buffering capacity [17,30]. For example, Gündoğdu [30] reported that darker-skinned cherry cultivars contained approximately 12 mg/100 g FW vitamin C, whereas lighter-skinned cultivars contained 7–9 mg/100 g FW. These results suggest that the elevated vitamin C levels in ‘13420’ may be partially explained by its phenolic profile, which protects against oxidative degradation during ripening. The ranges observed in this study (6.85–14.70 mg/100 g FW) align closely with published values. Mertoğlu [31] reported values of 6–11 mg/100 g FW across Turkish cultivars and Gündoğdu [30] confirmed similar distributions.

The results for vitamin C in apricots reflect pronounced genotypic variation, which is consistent with reports that the ascorbic acid contents of Prunus armeniaca germplasm can vary by severalfold [32]. Published ranges extend from ~3 mg/100 g FW to over 20 mg/100 g FW depending on genotype and region [9,32,33]. Compared to this broader spectrum, our accessions fall within a moderate range. Top performers, such as ‘VOJ5/150,’ approached values characteristic of high-vitamin cultivars, such as ‘Preventa’ (>16 mg/100 g FW) [32].

Higher levels of vitamin C in apricot accessions can be interpreted within the same integrated regulatory framework described above, where genotype-specific control of metabolic flux and antioxidant turnover determines steady-state ascorbate levels [27,29]. Environmental and developmental factors also modulate these patterns. Iordanescu et al. [33] demonstrated that early-ripening cultivars frequently had higher vitamin C levels, likely due to their shorter exposure to oxidative stress during development. Tissue distribution also plays a role. Apricot peel contains more vitamin C than the flesh. However, its contribution to the whole fruit is smaller than that in apples due to the thinner peel proportion [34].

Compared to the global diversity, our observed content and range of vitamin C in plums (1.58–9.60 mg/100 g FW) is low. Kuru et al. [18] reported values of 20.3–54.4 mg/100 g FW in wild and cultivated accessions. These results suggest that, while our collection captures variability, it only represents part of the broader genetic potential. Pigmentation and secondary metabolism are closely related to vitamin C accumulation. Mitic et al. [35] demonstrated that darker-colored accessions had higher antioxidant activity. Ruiz-Rodríguez et al. [17] found that blackthorn (P. spinosa) typically contains up to 15.35 mg/100 g FW. Peel tissue also contributes significantly. Drogoudi et al. [36] discovered that the peel can account for 45–50% of the total antioxidant capacity having vitamin C contents two to three times higher than those in the flesh.

The results regarding vitamin C content in apples align with prior studies indicating that wild Malus sieversii accessions typically contain 18–28 mg/100 g FW, compared to 7–12 mg/100 g FW in many commercial cultivars [20]. Bianchi et al. [37] confirmed cultivar-specific stability across seasons, which is consistent with our observation that ‘HL1311’ remained high and ‘Rubinstep’ remained low across years.

Ascorbate accumulation in apples similarly reflects genotype-dependent regulation within the broader metabolic network outlined above [27,29]. However, in apples genetic regulation appears to dominate environmental modulation in apples. Moreover, tissue-specific distribution further modulates whole-fruit vitamin C concentrations. Li et al. [34] demonstrated that the peel contains up to three times more vitamin C than the flesh, due to higher biosynthetic and recycling activity. Thus, whole-fruit analyses reflect tissue composition and peel-to-flesh ratios. This gives genotypes with thicker or more pigmented peels (e.g., ‘HL827’) an advantage. Environmental factors further interact with genotype. Li et al. [38] showed that altitude and orchard conditions modulate accumulation, particularly in wild versus cultivated apples. The ranges observed here (7.24–21.60 mg/100 g FW) are consistent with broader surveys. Bassi et al. [19] quantified cultivar-specific variation and confirmed strong tissue effects, with peel containing 2.7–56.0 mg/100 g FW and pulp 0.1–13.9 mg/100 g FW; whole-fruit values reported elsewhere range up to ~25.6 mg/100 g FW, and Kumar et al. [21] confirmed that wild or traditional cultivars often exceeded 20 mg/100 g FW up to 32 mg/100 g FW. Overall, within the conditions of the present study, apples exhibited comparatively greater stability across years than the evaluated stone fruit species.

In interpreting these interspecific differences, it is important to note that vitamin C concentrations were expressed on a fresh-weight basis to ensure nutritional relevance, as fruit is predominantly consumed fresh. Although dry-matter normalization may facilitate compositional comparisons, it can distort practical dietary interpretation, particularly in species differing in water content.

While this section emphasized genetic determinants, climatic variability represents the second major axis shaping species-specific patterns of ascorbate accumulation.

3.2. Climatic Influences and Genotype–Environment Interactions

Climate influences plastidial and vacuolar transporters that shape tissue-specific accumulation, indicating that ascorbic acid levels are sensitive to interannual weather variability. Environmental conditions affect not only biosynthetic flux but also ascorbate transport and compartmentalization [28]. Lee and Kader [8] reported that vitamin C content in citrus fruits grown under high summer temperatures (30–35 °C) was reduced by up to 25% compared with fruit grown under cooler conditions (12–14 mg/100 g versus 16–18 mg/100 g FW). Similarly, Davey et al. [3] documented interannual variation in strawberry ascorbate content (22–45 mg/100 g FW) depending on rainfall and temperature, and Fenech et al. [22] observed approximately 30% lower vitamin C levels in apples during hot, dry seasons compared with wetter years (6–8 mg/100 g FW versus 9–12 mg/100 g FW).

Consistent moisture-related trends have been reported in stone fruits. Ivanova et al. [23] reported vitamin C values ranging from approximately 7.3 to 14.5 mg/100 g FW across cultivars and years, with significant contributions of precipitation-related variables during the ripening period. Usenik et al. [10] confirmed that wetter sites produced cherries with 20–30% higher vitamin C concentrations (6–8 to 9–12 mg/100 g FW). Additionally, Hegedűs [32] recorded seasonal differences of up to 50% in apricots, with wetter years yielding 9–12 mg/100 g FW and drier years 6–8 mg/100 g FW. These published patterns are consistent with our 2023 observations, in which cherries averaged 11.61 mg/100 g FW and plums 4.96 mg/100 g FW across accessions.

High temperatures, especially above 30 °C, have repeatedly been associated with reductions in fruit vitamin C content. For example, one study demonstrated that grape berries exposed to temperatures above 32 °C during ripening lost up to 40% of their ascorbate content compared to berries ripening at moderate temperatures (4 mg/100 g vs. 7 mg/100 g FW) [39]. This effect was evident in our apricots in 2022, where some accessions such as ‘HL08-052’ still reached 15.84 ± 0.87 mg/100 g FW but this value declined sharply to 6.43 ± 0.12 mg/100 g FW in 2023. This result highlights the sensitivity of certain accessions to heat stress, although species-level linear models were not significant (all p > 0.05). Within the limitations of the three-year dataset, this suggests that genotype-by-year interactions may play a stronger role than direct effects of single climatic parameters. This interpretation is consistent with findings by Hegedűs et al. [32] and Iordanescu et al. [33], who reported that harvest timing and cultivar identity strongly influence ascorbic acid content, with early cultivars often exhibiting greater stability. Rather than implying complete buffering, these results indicate that apricot vitamin C levels may be regulated by complex physiological mechanisms, including antioxidant recycling capacity [40], which warrant further investigation. Several studies suggest that high temperature stress during fruit development and ripening can negatively impact ascorbic acid accumulation, implying that extreme summer heat may reduce vitamin C content in sensitive genotypes under field conditions [41]. This trend was mirrored in our ‘HLT1-10’accession, which decreased from 15.17 ± 0.30 mg/100 g FW in 2023 to only 4.61 ± 0.43 mg/100 g FW in 2024.

Overall, the observed species-specific patterns illustrate the interaction between genotype and climate interact within Czech orchards. These findings provide valuable guidance for breeding and cultivation in temperate Central Europe. However, they should be viewed as exploratory and not automatically extrapolated to other climatic zones.

4. Materials and Methods

4.1. Experimental Design

This study evaluated newly bred accessions and commercial cultivars of sweet cherries, apricots, plums, and apples from the Research and Breeding Institute of Pomology (RBIP) breeding program over a three-year period (2022–2024). This timeframe was chosen to capture variation under different climatic conditions. The selected commercial cultivars, which are commonly used in integrated fruit production systems in the Czech Republic, served as control variants for comparison with the breeding accessions. The multi-year design allowed us to assess genetic differences and the effects of environmental variations on vitamin C content from one year to the next.

4.2. Orchard Conditions

The experimental orchards of the RBIP (Holovousy, Eastern Bohemia, Czech Republic) are located in a moderately warm macro-region with a moderately humid sub-region. The orchards (50.383629° N, 15.576902° E, 360 m a.s.l.) have a long-term average annual temperature of 8.4 °C and average annual rainfall of 663.5 mm. Climatic data from 2022 to 2024 were obtained from an AMET automated weather station (AMET, Velke Bilovice, Czech Republic) located within the orchards. This data includes the monthly means and maximums of daily temperatures, rainfall, and mean relative humidity during the specific ripening and harvest periods for each fruit species (June–October), (Table 7 and Table 8).

Table 7.

The average monthly air temperatures (°C) with means for the five months (A), and the highest temperatures (°C) with means for the five months (B) recorded from the beginning of sweet cherry ripening and harvest through to the final apple harvest, covering the full seasonal span for all tested fruit species during the experimental period 2022–2024.

Year/Month	VI.	VII.	VIII.	IX.	X.	A	VI.	VII.	VIII.	IX.	X.	B
2022	20.0	19.8	21.4	13.5	11.6	17.3	33.3	35.4	36.9	29.1	22.9	31.5
2023	18.8	20.6	20.0	18.4	11.9	17.9	33.7	35.5	33.9	32.1	28.1	32.7
2024	19.4	21.1	21.6	17.5	10.9	18.1	33.7	35.5	35.9	34.1	19.2	31.7

Table 8.

The average monthly and annual rainfall (mm) (C) and average air humidity (%) (D), recorded from the beginning of sweet cherry ripening and harvest through to the final apple harvest, for the experimental period 2022–2024, capturing the full seasonal range across all tested fruit species.

Year/Month	VI.	VII.	VIII.	IX.	X.	C	VI.	VII.	VIII.	IX.	X.	D
2022	90	34	48	85	86	68	66	64	66	79	27	60
2023	20	88	115	9	84	63	65	64	74	72	65	68
2024	56	72	89	89	84	78	70	69	71	71	28	62

4.3. Orchard Management

All orchards were maintained using standard commercial practices within integrated fruit production systems. The trees were trained as free spindles, and the grass-covered inter-rows were regularly mowed. Herbicide strips were placed beneath the rows. Stone fruit trees were pruned annually before flowering, and the resulting biomass was mulched into the inter-row spaces. Apple trees were pruned during the winter dormancy period. The sweet cherry orchard, which was planted in 2008, consisted of trees on Gisela 5 rootstock spaced at 5 × 1.5 m apart. The apricot and plum trees, which were planted in 2014, were grafted onto St. Julien A rootstocks and planted at 4.5 × 1.0 m and 4.0 × 1.0 m apart, respectively. The apple trees, which were planted in 2010, were on dwarfing M9 rootstock and spaced at 4.0 × 1.0 m. Fertilization management included annually applying ammonium nitrate with lime to the soil (approx. 27% nitrogen, in both ammonium and nitrate forms) at a rate of 1.1 q/ha. Foliar feeding was provided with a liquid urea–ammonium nitrate solution at rates of 3 L/ha twice and 4.7 L/ha three times. This solution contained about 42.2% ammonium nitrate, 32.7% urea, and 25.1% water, providing a total of 30% nitrogen (approximately 25% ammonium, 25% nitrate, and 50% urea). During fruit ripening, a calcium chloride solution (12% Ca, equivalent to 16.9% CaO, fully soluble) was applied twice at 10 L/ha. No covering systems or supplemental irrigation were used.

4.4. Plant Material

The study included newly bred accessions and commercial cultivars of sweet cherries (Prunus avium L.; ‘Early Korvik’, ‘Jacinta’, ‘Kordia’, ‘Justyna’, ‘Amid’), apricots (Prunus armeniaca L.; ‘Harogem’, ‘Sophinka’, ‘Betinka’, ‘Candela’), plums (Prunus domestica L.; ‘Toptaste’, ‘Presenta’, ‘Tophit’), and apples (Malus domestica Borkh.; ‘Rubinstep’, ‘Idared’, ‘Regula’, ‘Frosta’) from the experimental orchards of research institute. A total of 24 sweet cherry accessions, 17 apricot accessions, 17 plum accessions, and 16 apple accessions were evaluated. These fruit varieties were selected based on the priorities of our institute’s breeding programs. The goal was to map the antioxidant characteristics of these varieties, specifically their vitamin C content. Full lists of breeding codes for all accessions are provided in the Supplements. The accessions were selected to represent both the genetic diversity of the breeding program and the commercial standards currently cultivated in the region. We chose these four species because they are the focus of ongoing breeding programs at our institution and are among the most economically relevant fruit crops in Central Europe. While their vitamin C levels are moderate compared with citrus fruits, these species are widely consumed and exhibit substantial genetic variability. This makes them suitable model plants for evaluating genotype-by-environment interactions in ascorbate accumulation.

To maintain consistency, all accessions were grown under the same orchard conditions described in Section 4.3. Section 4.5 details the standardized harvest protocols applied across species, adapted to the physiological and commercial maturity criteria specific to each fruit type.

4.5. Fruit Harvest

Sweet cherries were harvested at full commercial maturity and visually assessed using the PNW Dark Sweet Cherry Development Index Chart (Oregon State University, USA). This chart corresponds to shades 4–6. Apricots and plums were collected at commercial maturity using the CTIFL color chart (Centre Technique Interprofessionnel des Fruits et Légumes, Paris, France). The fruit was selected at stages 7–9 for apricots and 4–5 for plums. Apples were harvested based on standard horticultural maturity criteria, including a change in background skin color from green to yellow and starch–iodine index scores of 6–8. For each accession, fruit was harvested from three representative trees. Samples were taken from all canopy positions (upper, middle, and lower zones) and from both sun-exposed and shaded sides. Three biological replicates were prepared, each consisting of approximately 1.2 kg fresh fruit. The number of individual fruits per replicate varied according to species and size. To ensure representation of different exposure levels, the fruit was randomly collected and processed within 2 h to minimize vitamin C degradation.

4.6. Determination of Fruit Composition

Immediately after harvesting, 1000 g of seed-free fresh fruit were homogenized using a high-speed blender (Retsch GM 200, Retsch GmbH, Haan, Germany). Then, subsamples were extracted in 3% metaphosphoric acid to stabilize the ascorbic acid content. All measurements were performed in triplicate to ensure accuracy and reproducibility.

4.7. Chemicals and Standards

An L-ascorbic acid standard of at least 99.0% purity (Sigma-Aldrich, St. Louis, MO, USA) was used for calibration. Deionized water was obtained from an OmniaTap 6 UV system (STAKPURE, Niederahr, Germany). Potassium dihydrogen phosphate (Lach-Ner, Neratovice, Czech Republic) and cetyltrimethylammonium bromide (VWR International, Leuven, Belgium) were used to prepare the mobile phase. Trisodium phosphate dodecahydrate (Lach-Ner, Neratovice, Czech Republic) and L-cysteine (≥99%, Thermo Scientific, Waltham, MA, USA) were used as reducing agents. Metaphosphoric acid (VWR International, Leuven, Belgium) was used for extraction and stabilization.

4.8. Vitamin C Determination

The HPLC-DAD method with metaphosphoric acid extraction was chosen for its high specificity and sensitivity for differentiating between the reduced and oxidized forms of ascorbic acid. The method has also been proven to be reproducible in the analysis of fruit matrices [3,42]. The selected climatic variables reflect key environmental stressors known to affect vitamin C biosynthesis and stability in fruit. This enabled a targeted assessment of the interactions between genotype and climatic conditions. Typically, 20 g of freshly homogenized fruit were weighed into a 100 mL volumetric flask containing half of the 20 g/L aqueous metaphosphoric acid solution. The flask was then filled to the mark. After thorough mixing, the extracts were vacuum filtered through No. 389 filter paper. To reduce the compounds in the sample, 20 mL filtrate were combined with 10 mL L-cysteine solution (c = 40 g/L, 20 g was added to a 500 mL volumetric flask and made up to the mark with water fresh on the day of analysis). The pH was adjusted to 7.0–7.2 with trisodium phosphate solution (200 g/L) while stirring magnetically. Then, the pH was lowered to pH 2.5–2.8 with metaphosphoric acid (200 g/L). The mixture was transferred to a 50 mL volumetric flask and brought to the proper volume with distilled water. Then, it was filtered through a 0.45 µm PTFE filter into vials. The separation was carried out in isocratic elution mode using an Agilent 1260 Infinity HPLC-DAD system MassHunter with a Kinetex C18 100A column (150 × 4.6 mm, 5 µm, Phenomenex, Torrance, CA, USA) at a temperature of 25 °C. The mobile phase consisted of a methanol–water mixture containing 15 g/L potassium dihydrogen phosphate and 18 g/L cetyltrimethylammonium bromide. This mobile phase was used at a flow rate of 1.0 mL/min, with UV detection at 265 nm. A single analysis took approximately 8 min. Identification and quantification were performed using external calibration with L-ascorbic acid standard solutions (1 g/L stock solution in metaphosphoric acid reagent, diluted to 2–200 mg/L). The R² values achieved for the calibration curves (y = ax + b) ranged from 0.9991 to 1.0000. Each sample was analyzed in triplicate, which represents a commonly accepted experimental standard that minimizes random analytical error, allows estimation of measurement variability, and ensures reliable and reproducible results.

4.9. Method Validation

The analytical method for determining vitamin C was validated using high-performance liquid chromatography coupled with diode array detection (HPLC–DAD). Validation was performed according to commonly accepted analytical guidelines to ensure the reliability, precision, and accuracy of the method when applied to fruit matrices. The method was validated specifically for stone fruits (cherries, apricots, plums) and pome fruits (apples).

The limit of quantification (LOQ) for vitamin C was determined to be 2.5 mg/L, which is sufficient for analyzing vitamin C in fruits. Linearity was evaluated using matrix-matched calibration standards over a concentration range of 2.5–200 mg/L. Excellent linearity was achieved across the entire range, with a coefficient of determination (R²) of 0.999.

Method precision was assessed through repeatability studies and expressed as the relative standard deviation (RSD). The RSD values obtained for both stone and pome fruit matrices were below 10%, indicating acceptable repeatability and robustness of the analytical procedure.

The method accuracy was evaluated through recovery experiments performed on fortified samples. The average recovery of vitamin C was 88% for stone fruits (e.g., cherries) and 83% for pome fruits (e.g., apples), confirming the suitability of the method for quantitatively determining of vitamin C in complex fruit matrices.

Overall, the validated HPLC–DAD method demonstrated adequate sensitivity, linearity, precision, and accuracy, and can be considered reliable for routinely determining vitamin C content in stone and pome fruits.

4.10. Statistical Analysis

Statistical analyses were performed separately for each species using a full factorial design, with accession and year as the fixed factors in a two-way ANOVA. This approach was chosen because it allows for the evaluation of both the main effects (accession and year), as well as their interaction. This interaction was essential for detecting relationships between genotype and climatic conditions. Prior to analysis, the data were tested for normality and homogeneity of variances using the Shapiro–Wilk and Levene tests, respectively. The results of the ANOVA are reported as degrees of freedom (Df), sum of squares (SS), mean squares (MS), F-values, and their corresponding p-values. Significance levels were denoted as * p < 0.05, ** p < 0.01, and *** p < 0.001. Year effects that were significant were examined further using Tukey’s HSD post hoc test (R package agricolae), where different letters indicate significant pairwise differences among years at p < 0.05. Mean values for years and accessions are presented as mean ± standard error (SE) with calculated coefficient of variance to express the stability of the measured vitamin C content in particular species/accessions among the years. We used simple linear regression to assess the influence of individual climatic variables on vitamin C content. This approach quantified the strength and direction of the associations between the climatic variables and vitamin C content. It also provided an interpretable measure of the effects of climatic conditions. In this analysis, we regressed the average vitamin C values of all accessions of a given species in a particular year (2022–2024) against the following variables: the number of days with a daily mean temperature higher than 20 °C (NTavg-20), the maximum daily temperature higher than 30 °C (NTmax-30), the minimum daily temperature less than 10 °C (NTmin-10), the daily temperature difference higher than 12 °C (NTdif-12), the mean relative humidity (RH), and the total rainfall during the species-specific ripening and harvest period. The significance of the particular regression models was evaluated at the same levels of significance as the ANOVA tests. We acknowledge that these analyses were carried out using a relatively small dataset of three data points per cultivar. Due to the risk of overfitting the model, we could not perform multivariate analyses of the environmental factors. This also limited the interpretation of the results to some extent. However, each point represented the mean value of all cultivars of a single species lending certain robustness to the analyzed dataset. All analyses were performed in RStudio (RStudio Team, Boston, MA, USA). The F-values in the reported results reflect the ratio of explained to unexplained variance in ANOVA models, and the t-values express the standardized difference between the compared means. Degrees of freedom (Df) indicate the number of independent pieces of information available for estimating variability in each test.

5. Conclusions

This study demonstrates that vitamin C accumulation in temperate fruit species reflects a strong genotypic component, while interannual climatic variability modulates its expression in a species-dependent manner. Genetic background establishes baseline accumulation capacity, whereas seasonal thermal and moisture conditions influence realized concentrations to varying degrees.

Across species, apples exhibited comparatively lower climatic sensitivity, suggesting stronger intrinsic physiological regulation, whereas cherries and plums showed higher sensitivity to elevated seasonal temperatures. Among the evaluated climatic parameters, the number of days with a daily mean temperature ≥20 °C (NTavg-20) emerged as the strongest negative linear climatic correlate of vitamin C levels in cherries and plums, although this inference is constrained by the three-year duration of the dataset. In contrast, apricots displayed pronounced genotype-by-year interactions, and no individual climatic variable showed a statistically significant linear effect, indicating that accession-specific responses outweighed direct linear climatic drivers.

In contrast, apricots displayed pronounced genotype-by-year interactions, and no individual climatic variable showed a statistically significant linear effect, indicating that accession-specific responses outweighed direct linear climatic drivers.

These patterns suggest that breeding priorities should differ among fruit species. In climate-sensitive species such as cherries and plums, improving stability under elevated temperature conditions may enhance resilience, whereas in apples and certain apricot genotypes, maintaining high baseline accumulation appears more relevant. The distinction between resilient and high-potential genotypes provides practical guidance for cultivar selection under variable seasonal conditions.

Because the dataset spans three seasons under defined Central European orchard conditions, extrapolation to contrasting climatic zones should be approached with caution. Nevertheless, the identified species-specific genotype–environment patterns are likely relevant to temperate production systems experiencing comparable climatic variability.

Author Contributions

Conceptualization, A.B. and P.S.; methodology, P.S. and A.B.; software, A.B.; validation, A.B.; formal analysis, A.B. and D.J.; investigation, A.B., P.S., D.J. and L.Š.; data curation, A.B., D.J. and M.M.; writing—original draft preparation, A.B., P.S., D.J. and M.M.; visualization, P.S.; supervision, P.S.; resources, P.S.; founding acquisition, P.S. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors were employed by the company, Research and Breeding Institute of Pomology Holovousy Ltd. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Statement

This research was funded by the Ministry of Agriculture of the Czech Republic, grant number RO1525 and the National Agency for Agricultural Research, grant number QK21010200.