QTL mapping of fruit sugar and acid traits in the F1 generation of jujube (Ziziphus jujuba Mill.) Authors: Yujia Luo, Zhi Luo, Mengjun Liu, Cuiyun Wu, Fenfen Yan Published: 2026 ABSTRACT AbstractBackgroundJujube (Ziziphus jujuba Mill.) is one of China’s important specialty fruit trees with high nutritional and medicinal values. With the growing demand for high-quality jujube fruit, genetic improvement of fruit quality has become a key focus of the jujube industry. Fruit flavor is critical for market competitiveness, and soluble solids, soluble sugar and titratable acid are the main components determining fruit flavor quality.ResultsTo analyze the genetic inheritance of sugar and acid traits in jujube fruits, the F1 generation of 140 progeny from the cross between jujube ‘Yuhong’× ‘Jiaocheng 5’ was established and used for the measurement of sugar and acid contents for a period of 3 years. Thereafter, based on a well-constructed, high-density genetic map of jujube, quantitative trait locus (QTL) localization analysis was carried out. The results showed that a total of 11 QTLs were detected across the sugar and acid traits, which distributed across 6 chromosomes, with LOD values between 3.01 and 4.64, and phenotypic variance explained (PVE) ranged from 12.84% to 23.34%. When the LOD was ≥ 3.5 and PVE ≥ 10%, a total of 5 major QTLs were detected. In the LG1 chromosome, there was one overlapping interval for the traits of soluble solid content and soluble sugars between 35.33 cM and 43.02 cM. In the region from 162.86 cM to 217.25 cM, one overlapping interval for the titratable acidity trait was detected in both 2021 and 2023, with PVE values all above 15%. In addition, within the stable QTL intervals, the candidate genes related to the accumulation of sugar and acid were first screened. Quantitative real-time PCR (qRT-PCR) was employed to analyze the expression patterns of these candidates, leading to the identification of two genes (LOC107416407, LOC107409930) potentially associated with sugar content trait. Bioinformatics analysis revealed that LOC107416407 contains a conserved PFK domain, suggesting it encodes phosphofructokinase, a key regulatory enzyme in glycolysis. LOC107409930 harbors a TPT domain and belongs to the triose-phosphate transporter family. These findings provide a crucial theoretical basis for further investigating the functions of these two genes in sugar accumulation of jujube fruit.ConclusionsIn summary, eleven QTLs and two stable overlapping intervals were detected on the LG1 chromosome. The sugar-related QTL interval is unreported previously and may contain novel genes for sugar accumulation. 2 candidate genes were identified, whose functions are related to sugar metabolism. This study provides theoretical basis, molecular markers and gene resources for exploring sugar metabolism mechanisms and quality-oriented breeding of jujube.Keywords: Ziziphus jujuba Mill., Sugar and acid traits, QTL mapping, Sugar metabolism Introduction Jujube (Ziziphus jujuba Mill.) belongs to the Ziziphus genus of the Rhamnaceae family, which is one of China’s important specialty fruit trees, boasting a long cultivation history and a wide area of cultivation, especially in the middle and lower reaches of the Yellow River [1, 2]. The main characteristics of jujube are that it adapts the main abiotic stresses and its fruit has significant nutritional value and medicinal effects [3, 4]. For example, the jujube fruits are rich in vitamin C, sucrose, cyclic adenosine monophosphate, and other biologically active substances, as well as containing calcium, iron and other mineral elements and dietary fiber [5]. With ongoing improvements in living standards and heightened health awareness, the market’s demand for the superior flavor quality of jujube fruits has increased significantly. Consequently, the genetic improvement of jujube fruit quality has emerged as the core issue of jujube industrial development. Fruit flavor is a core indicator that directly affects and determines consumer preference and market competitiveness [6]. While the contents of soluble solid, soluble sugar and titratable acid are the main components that determines the fruit flavor, and the soluble solid-acid and sugar-acid ratios are the key factor to assesses fruit flavor quality. However, the above traits exhibit quantitative genetic characteristics regulated by multiple genes, which makes it difficult to achieve rapid and effective varietal improvement during breeding [7–11]. Quantitative Trait Locus (QTL) analysis is a genetic method based on molecular marker technology that enables the localization of the key genes controlling quantitative traits to specific regions of chromosomes [12, 13]. Through this method, numerous studies have shown that the quantitative genetic characteristics regulated by multiple genes have been identified and which can be applied for fruit quality improvement. For example, the QTL associated with soluble solid content in the hybrid progeny of ‘Yuluxiangli’בShinseiki’ was identified. The related QTL was found on LG07 chromosome, specifically located between 687,965 and 971,813 base pairs, with LOD value as 3.85 and the PVE value as 11.7%. At the QTL mapping interval, a total of 38 genes were identified [14]. In the hybrid progeny of ‘Gala’בXiahongrou’, four QTLs linked to soluble solid content were detected on LG7 and LG8 chromosomes, with LOD values ranging frome 3.72 to 4.61. One QTL associated with titratable acid content localized on LG7 chromosome, with a LOD value of 3.05 [15]. In the hybrid progeny of ‘Red Fuji’ × ‘Hongrou’ six QTLs related to titratable acid were identified on LG1、LG6 and LG7 chromosomes, with LOD values between 2.45 and 7.90. One QTL associated with soluble sugar was identified on LG2 chromosome, with a LOD value of 2.12. Additionally, two QTLs for soluble solid content were mapped to LG1 and LG7 chromosomes, with LOD values of 2.95 and 2.81, respectively [16]. In sweet cherry (Prunus avium L.), the QTL analysis was performed on five groups of hybrid progeny, it showed that the soluble solid content (SSC) trait was stably mapped to the qP-SSC4.1 m locus on LG4 chromosome over four consecutive years’ study, with each PVE value exceeding 30%. For titratable acid (TA), the stable qP-TA6.1 m locus was consistently detected, and its PVE values ranged from 15.0% to 21.57% [17]. In strawberries (Fragaria × ananassa Duch.), two stable QTLs associated with the SSC trait localized on chromosomes 3B and 6 A, with PVE values ranging from 1.2% to 2.8% [18]. However, the QTL mapping analysis of sugar and acid traits in jujube hybrid progeny is lagged relatively behind, with only a limited number of loci having been identified [19, 20]. Specifically, 7 QTLs for SSC, 31 QTLs for SS, and 5 QTLs for TA were identified in ‘Dongzao’ × ‘Jinsi 4’ hybrid progeny, with LOD values ranging from 2.52 to 5.73 and PVE values between 15.3% and 31.0% [19]. Using the same hybrid progeny, more QTLs associated with SSC were found and these QTLs exhibited LOD values ranging from 3.04 to 4.04 and 17.1% to 22.0% of the PVE values. Additionally, more QTLs linked to TA were identified, with LOD values between 3.05 and 4.01 and PVE values spanning 17.7% to 22.6% [20]. Moreover, in terms of genes related to sugar and acid traits in jujube fruits, even some genes have been identified, such as ZjHXK5 and ZjHXK6, ZjSWEET11, ZjSWEET18 and ZjACO3 [21–23], the systematic mining of more key genes related to the improvement of fruit sugar-acid traits remains insufficient, which has hindered the genetic improvement of jujube fruit quality to a certain extent. To further elucidate the genetic variation patterns and identify the key genes underlying sugar-acid traits in jujube, in the current study, 140 F1 individuals from ‘Yuhong’ × ‘Jiaocheng 5’ were conducted to measure the soluble solid content, soluble sugar and titratable acid content over 3 consecutive years. In addition, the QTLs associated with sugar and acid traits based on a high-density genetic map constructed by whole-genome resequencing were identified which could be applied to dig out more candidate genes which might regulate the sugar and acid synthesis in jujube fruits. The results can provide a solid theoretical foundation for the genetic improvement of jujube sugar-acid traits and be used for the related molecular marker-assisted breeding. Result Genetic variation and correlation analysis of sugar-acid traits The soluble solid content (SSC), soluble sugar (SS) and titratable acid (TA) content were measured over 3 consecutive years in the ‘Yuhong’ × ‘JiaoCheng 5’ (J5) hybrid progeny from 2021 to 2023. As shown in Fig. 1, the contents of SSC, SS and SA exhibited continuous variation with normal or skewed normal distributions, which might demonstrate that these 3 traits are quantitative traits controlled by micro-efficient polygenes. Frequency distribution analysis of SSC, SS and TA traits in the F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023 In addition, the mean values of SSC in F1 hybrid progeny were 36.18%, 34.62%, and 41.01%, respectively, from 2021 to 2023, which were in the middle of female ‘Yuhong’ and male ‘J5’, but skewed toward the higher parent (Table 1). The SSC of ‘Yuhong’ was recorded as 27.87% in 2021, and increased to 40.40% and 41.50% in 2022 and 2023, respectively. Meanwhile, the SSC of ‘J5’ ranged consistently between 37.60% and 39.67% over the same period, indicating that the male parent has the stable behavior. The coefficients of variation ranged from 8.88% to 12.42%, indicating a low dispersion, and the genetic transmission ability was consistently above 88%, suggesting that the trait of SSC could be stably passed on to the hybrid progeny. The mid-parent heterosis rates were 7.15%, -11.24% and 2.56%, respectively. The negative value in 2022 indicated that the F1 hybrid progeny was lower than that of the mid-parent value. Furthermore, the mean values of SS ranged from 29.33% to 32.08%. The values of F1 hybrid progeny were lower than the mid-parent level in both 2021 and 2023,suggesting a weak genetic tendency. However, in 2022 the mean values surpassed the mid-parent threshold, demonstrating a stronger genetic tendency. The maximum values ranged between 37.41% and 49.93%, while the minimum values of F1 hybrid progeny ranged from 13.43% to 18.70%, both exceeding parental ranges. This significant segregation pattern indicated broad genetic variation existed in SS content. Coefficients of variation were 14.57%, 19.44% and 22.44%, and the hereditary transmission ability surpassed 95%. These metrics collectively confirmed the efficient genetic transmission characteristics of SS. The mean values of TA ranged from 0.30% to 0.48%, consistently below the mid-parent value (0.52% in 2021 and 0.41% in 2022). Notably, a reversal occurred in 2023 when the TA mean value surpassed the mid-parent value. The minimum values ranged from 0.17% to 0.28%, while the maximum values ranged from 0.65% to 0.78%. Coefficients of variation spanned 24.53%-33.08%, demonstrating substantial genetic variation in TA expression patterns. The genetic transmission abilities were quantified at 90.71%, 72.22% and 118.66%, suggesting significant environmental modulation. Mid-parent heterosis rates transitioned from negative values (-9.29% in 2021 and − 27.78% in 2022) to a positive 18.66% in 2023, revealing interannual phenotypic divergence. Finally, both soluble solids-acid ratio and sugar-acid ratio are important indicators for evaluating fruit quality, reflecting the sweetness and sourness of the fruit’s taste. The soluble solids-acid ratio and sugar-acid ratio were higher over the three years, suggesting that the fruit flavor of the F1 hybrid progeny was sweeter, with sensory profiles shifting toward preferred consumer preferences. The degree of genetic variation varied across different traits and years, indicating that the sugar-acid trait is more influenced by the environmental conditions in addition to its own genetic factors. Table 1The genetic variation of fruit SSC, SS and TA traits in the F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023TraitYearParentsF1MaleFamaleVMPMean ± SDMinMaxCV%RHM%Ta%SkewnessKurtosisSoluble solid substance /%202127.8739.6733.7736.18 ± 4.0524.0744.3311.197.15107.15-0.460.21202240.437.63934.62 ± 4.3024.844.5312.42-11.2488.76-0.23-0.33202341.538.4739.9941.01 ± 3.6429.9748.138.882.56102.56-0.710.43Soluble sugar /%202125.5134.2629.8829.33 ± 4.2718.737.4114.57-1.8698.14-0.13-0.48202224.0936.430.2532.08 ± 6.2413.4344.9719.446.07106.07-0.730.77202334.3429.7832.0630.65 ± 6.8817.5849.9322.44-4.4195.590.560.20Titratable acid /%20210.360.680.520.48 ± 0.120.280.7824.53-9.2990.710.40-0.4720220.470.350.410.30 ± 0.100.170.6533.08-27.7872.221.351.8020230.310.390.350.42 ± 0.120.170.7729.9018.66118.660.410.40Soluble solids-acid Ratio202176.7557.9167.3379.81 ± 21.8244.34138.3127.3418.52118.520.67-0.14202285.45107.0696.25124.94 ± 34.9945.95223.3928.0129.80129.800.33-0.062023132.9598.46115.71110.06 ± 39.6949.39255.7136.06-4.8895.121.663.78Sugar-acid Ratio202170.2550.0260.1465.49 ± 16.2933.09114.7724.878.90108.900.430.02202250.96103.6377.30117.83 ± 37.3322.91197.4631.6852.44152.44-0.26-0.252023110.0076.2393.1282.04 ± 39.0033.49270.8447.53-11.8988.112.237.39 The genetic variation of fruit SSC, SS and TA traits in the F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023 Correlation analysis between sugar-acid traits (SSC, SS and TA) averaged over the 3 years varied from − 0.824 to 0.601 (Fig. 2). SS and sugar-acid ratio showed strong, significant positive correlations (p < 0.05) during 2022 and 2023, with correlation values of 0.420 and 0.601, indicating higher SS values are associated with lower sugar-acid ratios. TA and sugar-acid ratio showed a significant negative correlation (p < 0.05) in 2022, but none significant correlation was fund in 2021 and 2023. Other traits showed none significant correlations. Correlation analysis of fruit SSC, SS and TA traits in F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023 Identification of QTLs for sugar-acid traits According to the high-density genetic mapping, a total of 11 QTLs for SSC, SS and TA were identified using the interval mapping method from 2021 to 2023 in the ‘Yuhong’בJ5’ hybrid progeny (Table 2), which were distributed across LG1, LG3, LG6, LG8, LG10, and LG12 chromosomes (Fig. 3). Table 2QTL mapping analysis of SSC, SS and TA traits in F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023TraitsYearsQTLsLGsPeakStart(cM)End(cM)Lod ValuePhenotypicvarianceexplained (%)MarkerPositionSSC 2023 qSSC23-1 1 chr1-10413496 35.33 31.68 43.02 4.64 23.34 SS2021qSS21-11chr1-1097032543.0235.3360.363.0116.37 2023 qSS23-2 12 chr12-2205899 14.68 4.99 18.69 3.83 15.02 2023 qSS23-3 8 chr8-14520802 76.24 73.51 76.42 4.06 16.66 TA2021qTA21-11chr1-47942098262.19143.15262.193.2314.67 2021 qTA21-2 3 chr3-4502227 9.30 0.00 24.58 4.15 15.70 2021qTA21-36chr6-8802171146.62124.91146.623.3514.28 2022 qTA22-4 10 chr10-16713280 121.86 118.45 131.19 3.99 16.20 2022qTA22-53chr3-23391276112.6332.93147.693.3812.842023qTA23-61chr1-42179453194.87162.86217.253.3917.372023qTA23-712chr12-230670133.7526.2437.003.0413.80Note: Values in bold indicate QTLs with LOD > 3.5 QTL mapping analysis of SSC, SS and TA traits in F1 hybrid progeny of ‘Yuhong’ × ‘J5’ from 2021 to 2023 Note: Values in bold indicate QTLs with LOD > 3.5 Distribution of QTLs for SSC, SS and TA traits on jujube chromosome One QTL for SSC was detected in LG1 chromosome, explaining up to 23.34% of the total phenotypic variance explained (PVE). Three QTLs for SS were identified in LG1, LG8 and LG12 chromosomes, accounting for 16.37% to 16.66% of the PVE value. Seven QTLs for TA were mapped to LG1, LG3, LG6, LG10 and LG12 chromosomes, with PVE values ranging from 12.84% to 17.37%. When limit of detection (LOD) threshold ≥ 3.5, five major QTLs were detected (Table 3), including the SSC QTL (qSSC23-1) on LG1 chromosome, which showed the largest LOD value of 4.64 in 2023. Two SS QTLs (qSS23-2 and qSS23-3) were detected in LG8 and LG12 chromosome, s with LOD values of 4.06 and 3.83, respectively. Additionally, two QTLs (qTA21-2 and qTA22-4) for TA were identified in LG3 and LG10 chromosomes, exhibiting LOD values of 4.15 and 3.99, respectively. Table 3Information of QTL clusters for SSC, SS and TA traits in jujubeTraitsCorrespond QTLOverlapping interval(cM)Overlappingposition(bp)LGGene numberSSC SSqSSC23-1qSS21-135.33–43.029,984,393–10,970,3251117TAqTA21-1qTA23-6162.86-217.2539,813,607–42,943,0621281 Information of QTL clusters for SSC, SS and TA traits in jujube SSC SS qSSC23-1 qSS21-1 qTA21-1 qTA23-6 Considering the QTL could be stable if it was identified for more than 2 years in the same region, that two stable QTLs regions for sugar-acid traits were identified in LG1 chromosome (Fig. 3; Table 3). The stable region contained two QTLs (qSSC23-1 and qSS21-1) associated with SSS and SS, located between 35.33 cM and 43.02 cM, explaining 23.34% and 16.37% of the total variance, respectively. For TA, two QTLs (qTA21-1 and qTA23-6) colocalized in a 162.86-217.25 cM region, accounting for 14.67% and 17.37% of the PVE value. According to the jujube reference genome, 117 genes were identified within the SSS and SS regions, while 281 genes were found in the TA region. However, only 3 candidate genes (LOC107416407, LOC107409930, LOC107416052) were considered as the promising candidates based on their functional annotations which might be closely associated with sugar metabolism. They were ATP-dependent 6-phosphofructokinase 2, nucleotide-sugar uncharacterized transporter 2, probable galactinol-sucrose galactosyltransferase 2 genes (Table 4). Table 4Functional annotation of the candidate genes for SSC and SS traits in jujubeTraitsCorrespond QTLGene IDFunctionSSCSSqSSC23-1qSS21-1 LOC107416407 ATP-dependent 6-phosphofructokinase 2 LOC107409930 nucleotide-sugar uncharacterized transporter 2 LOC107416052 probable galactinol–sucrose galactosyltransferase 2 Functional annotation of the candidate genes for SSC and SS traits in jujube SSC SS qSSC23-1 qSS21-1 The expression profiles of the candidate genes at fruits developmental stages To investigate the regulatory roles of above 3 candidate genes in sugar accumulation in jujube cultivars, the relative expression levels of the candidate genes at different developmental stages in low-sugar genotype (J100) (Fig. 4A) and high-sugar genotype (3J1) were analyzed (Fig. 4B). Using J100 as the control, the differences in expression levels of the candidate genes in 3J1 at the same developmental stage were compared (Fig. 4C). The expression level analysis of LOC107416407, LOC107409930, LOC107416052 at different developmental stages. ns represents none significant difference (p > 0.05), * indicates significant difference (p < 0.05), ** indicates highly significant difference (p < 0.01), *** indicates extremely significant difference (p < 0.001), **** indicates ultra-significant difference (p < 0.0001) It showed that LOC107416407 and LOC107409930 exhibited a highly consistent expression trend. In low-sugar genotype (J100), the expression levels of LOC107416407 and LOC107409930 showed a gradual upward trend with fruit development (from stage A to E), peaking at the full-red stage (E), maintaining a low expression level. In high-sugar genotype (3J1), the expression levels of these two genes also showed a continuous upward trend with fruit development, ultimately reaching the highest value at the stage E. Among them, both genes maintained at a low expression level at the stage A, while remaining in a high expression state from stage B to E. When low-sugar genotype (J100) was used as the control, the relative expression levels of LOC107416407 and LOC107409930 in high-sugar line (3J1) showed a gradual increasing trend. The relative expression levels of them increased from stage B and reached the expression peak level at stage D. In contrast, the expression pattern of LOC107416052 did not show a significant correlation with the changing pattern of sugar content in jujube fruits. Characteristics of the candidate genes Protein analysis showed that the LOC107416407 protein contain a PFK domain at amino acid position of 31–471 (Fig. 5A), indicating that it belongs to phosphofructokinase, a key regulatory enzyme in glucolysis that catalyzes the phosphorylation of fructose-6-phosphate to fructose-1, 6-bisphosphate. The LOC107409930 protein harbored a TPT domain at the position of 84–339 aa (Fig. 5B), suggesting that it is a member of the triose-phosphate transporter family involved in specific transmembrane transport of triose-phosphates. Conservative domain analysis of LOC107416407 and LOC107409930 protein Physicochemical property prediction analysis indicated that LOC107416407 was a stable acidic hydrophilic protein, whereas LOC107409930 was a stable alkaline hydrophobic protein. Both proteins were composed of 20 amino acids with different dominant components, and showed favorable stability in common expression systems (Table 5). Charge distribution analysis suggested a net negative charge for LOC107416407 and a net positive charge for LOC107409930 under physiological conditions, which was consistent with their hydrophilic and hydrophobic properties, respectively. Table 5Amino acid composition of LOC107416407 and LOC107409930 proteinAmino acidsLOC107416407LOC107409930AmountPercentageAmountPercentageAla (A)326.80%338.70%Arg (R)265.50%71.80%Asn (N)265.50%154.00%Asp (D)275.70%112.90%Cys (C)81.70%41.10%Gln (Q)183.80%133.40%Glu (E)265.50%92.40%Gly (G)418.70%287.40%His (H)163.40%30.80%Ile (I)285.90%277.10%Leu (L)398.30%5013.20%Lys (K)183.80%236.10%Met (M)71.50%102.60%Phe (F)173.60%225.80%Pro (P)234.90%133.40%Ser (S)285.90%3910.30%Thr (T)275.70%246.30%Trp (W)81.70%82.10%Tyr (Y)163.40%71.80%Val (V)418.70%338.70% Amino acid composition of LOC107416407 and LOC107409930 protein Hydrophobicity analysis confirmed the strong hydrophilicity of LOC107416407 and obvious hydrophobicity of LOC107409930, matching the results of physicochemical prediction. Both proteins contained abundant high-confidence phosphorylation sites, implying that their biological activities might be dynamically regulated by phosphorylation and dephosphorylation. In addition, various N-glycosylation and O-glycosylation sites were predicted in both proteins, suggesting potential post-translational modification regulatory mechanisms (Fig. 6). Prediction of Hydrophilic and hydrophobic, phosphorylation and N-glycosylation/O-glycosylation sites of LOC107416407 and LOC107409930 protein. a: Hydrophobic/hydrophilic prediction; b: Phosphorylation sites prediction; c: N-glycosylation sites prediction Transmembrane structure and signal peptide analysis revealed that LOC107416407 had no transmembrane domain or signal peptide and was predicted to localize in the cytoplasm. In contrast, LOC107409930 contained 10 complete transmembrane helices without a signal peptide, and was likely located in the cell membrane, consistent with its putative transporter function (Fig. 7). Transmembrane region analysis and signal peptide prediction of LOC107416407 and LOC107409930 protein. a: Transmembrane region analysis; b: Signal peptide prediction Secondary structure analysis showed that both proteins were mainly composed of alpha-helix and random coil. High-reliability tertiary structures were further obtained by homology modeling, which supported the accuracy of secondary structure prediction and provided a structural basis for their functional execution (Fig. 8). Prediction of secondary (a) and tertiary (b) structures of LOC107416407 and LOC107409930 protein Cis-elements analysis of the promoter of these two genes showed that abundant core cis-elements and various regulatory motifs were identified. The promoter of LOC107416407 was enriched in stress-responsive elements, while light-responsive elements were dominant in that of LOC107409930, indicating that their transcription might be regulated by stress and light signaling pathways, respectively (Fig. 9). Prediction and analysis of cis-elements anlysis of the promoter of LOC107416407 and LOC107409930 Phylogenetic analysis revealed that LOC107416407 was closely clustered with tomato 6-phosphofructokinase (XP_004236642.1) and shared high sequence similarity. Sequence alignment results showed that the amino acid sequence similarity between LOC107416407 and ten tomato proteins annotated as 6-phosphofructokinase (function not yet experimentally validated) ranged from 33.76% to 75%, with the highest similarity (75%) observed with tomato XP_004236642.1 (Fig. 9A(a)), and both contained the conserved PFK domain. Similarly, LOC107409930 exhibited close evolutionary relatedness with tomato nucleotide-sugar transporter (XP_004251941.1) and possessed the typical TPT domain; sequence alignment with five tomato proteins annotated as nucleotide-sugar uncharacterized transporter 2 (function not yet experimentally validated) showed similarity ranging from 50.40% to 70.98%, with the highest similarity (70.98%) to tomato XP_004251941.1 (Fig. 9B(a)). These results supported that the two candidate genes encode a functional glycolytic enzyme and a sugar phosphate transporter, respectively, which might participate in sugar metabolism and transport in jujube fruit (Fig. 10). Amino acid sequence alignment (a) and phylogenetic analysis (b) of LOC107416407 and LOC107409930 protein Discussion The SSC, SS and TA traits are crucial factors determining the commercial value of fruits, and directly influencing fruit flavor and quality [9, 24]. Consequently, sugar-acid traits are the key target traits for fruit tree breeding, and understanding their genetic mechanisms is essential for improving fruit quality. In the study, SSC, SS and TA exhibited continuous variation and normal distributions in the ‘Yuhong’ × ‘J5’ hybrid progeny, consistent with sugar-acid traits being typical quantitative characteristic [11, 25–27]. Sugar-acid contents are significantly influenced by external environmental factors such as temperature, light, soil conditions, which can affect the accumulation of sugar and acid in the fruit [28–31]. However, when analyzing the sugar-acid traits in the apricot F1 generation, it was found that variations in sugar-acid traits were more susceptible to genetic effect than environment factors [32]. In our study, after measuring and analyzing fruit sugar-acid traits over 3 years, the performances of the F1 hybrid progeny were relatively stable, despite some differences between the parents. This stability may be attributed to the high genetic diversity between the parents being reflected in the F1 hybrid progeny, with genetic laws constraining phenotypic variation to a certain extent and avoiding extreme deviations from parental traits. Additionally, the interaction between environmental and genetic factors may also fine-tune the variation in sugar-acid contents, which is consistent with the complex genetic regulation of quantitative traits. QTL localization based on high-density genetic maps provides an important molecular tool for genetic studies of sugar-acid traits in jujube. Indeed, Tang et al. [20] detected 22 QTLs for sugar-acid traits in 103 F1 ‘Dongzao’ × ‘Jinsi 4’ hybrid progeny using the composite interval mapping method, including a total of eleven SSC QTLs, three SS QTLs and eight TA QTLs, with LOD values ranging from 2.65 to 4.56, explaining 15.60–24.40% of the total PVE values. Notably, 4 SSC QTLs were mapped to LG1 and LG11 chromosomes, and 3 TA QTLs to LG7 and LG10 chromosomes. Xie et al. [19] identified 43 QTLs for sugar-acid traits, with LOD values ranging from 2.51 to 5.73 and PVE values from 15.20% to 31%, including 7 major SSC QTLs on LG3 and LG11 chromosomes, and 4 SS QTLs on LG8 chromosome. In this study, 11 QTLs associated with sugar-acid traits were identified, distributed across LG1, 3, 6, 8, 10 and 12 chromosomes. A stable major QTL region for both SSC and SS on LG1 (35.33–43.02 cM) chromosome that was consistently detected in 2021 and 2023, with no overlap with the QTL intervals reported by Tang et al. and Xie et al., which suggests that this region may harbor novel genes regulating sugar accumulation in jujube. Furthermore, two QTLs for TA (qTA21-1 and qTA23-6) were found within a stable region (162.86-217.25 cM) on LG1 chromosome in 2021 and 2023, indicating these QTLs have higher genetic stability and potential utility as targets for marker-assisted selection (MAS) in jujube breeding. Notably, the co-localization of SSC and SS QTLs on LG1 chromosome implies potential pleiotropic regulation or tight linkage of genes controlling these two traits, which could explain the significant positive correlation between SSC and SS observed in the phenotypic data. However, since the sugar-acid traits of jujube change with year and environmental changes, the QTLs identified so far only represent part of the genetic basis, longer data analysis is required to further validate the functions of these loci. Additionally, the F1 mapping hybrid progeny contained only 140 individuals, which is relatively modest for QTL mapping. Although this hybrid progeny was sufficient to detect major and stable QTLs for fruit sugar and acid traits, the relatively small hybrid progeny individuals led to limited statistical power. Consequently, some minor-effect QTLs with small phenotypic contributions may not have been detected. In addition, loci with weak genetic effects could be easily overlooked under the current mapping strategy. Future studies will expand the individuals of the mapping hybrid progeny and integrate higher-density molecular markers to improve mapping accuracy. Two candidate genes (LOC107416407 and LOC107409930) were identified as potentially associated with SSC and SS traits, supported by both expression pattern analysis and bioinformatic prediction. LOC107416407 encoded ATP-dependent 6-phoshofructokinase 2 (PFK2), a key rate-limiting enzyme in the glycolytic pathway. The PFK2 isoform may coordinate carbon metabolism with energy supply in plants or microorganisms through allosteric regulation in response to intracellular ATP/AMP ratios [33]. In pomegranate (Punica granatum), the expression level of PFK2 shows a significant positive correlation with fructose and glucose content, demonstrating a substantial and direct regulatory role in fructose and glucose synthesis via the glycolysis pathway [34]. Meanwhile, in apple, MdPFPβ (a pyrophosphate-dependent phosphofructokinase gene) modulates soluble sugar accumulation by enhancing photosynthetic ability and sugar metabolite enzyme activity. In addition, the transcription factor MdbHLH3 binds to the promoter of MdPFPβ and activates its expression, thereby increasing photosynthetic rate and soluble sugar accumulation [35]. In our study, bioinformatic analysis showed that LOC107416407 had no transmembrane domain and was predicted to localize to the cytoplasm, which is consistent with the cytoplasmic localization characteristics of glycolytic enzymes. This structural feature provides a structural basis for its efficient catalysis of the phosphorylation of intracellular fructose-6-phosphate and participation in the key step of glycolysis. In addition, the promoter region of this gene contained multiple stress-responsive elements, suggesting that its expression may be regulated by environmental stresses such as drought and salinity, thereby affecting sugar metabolism in jujube fruit. LOC107409930 encoded nucleotide-sugar uncharacterized transporter 2 (NST), which may participate in nucleotide-sugar transport and indirectly regulate O-glycosylation processes [36]. Glycosylation could directly or indirectly influence fruit sugar accumulation. This process relies on nucleotide sugars such as UDP-glucose and GDP-mannose as precursors [37, 38]. The synthesis of cell wall polysaccharides (pectin, hemicellulose) requires substantial amounts of UDP-galacturonic acid, potentially reducing the conversion of free monosaccharides into sucrose or starch [39, 40]. Bioinformatic analysis further supported the functional prediction of this gene. LOC107409930 contained a TPT domain, possessed 10 transmembrane helices, and was localized to the cell membrane, which conforms to the typical structural characteristics of membrane transporters. This is closely related to its function in mediating the transmembrane transport of nucleotide sugars and participating in glycosylation modification. The promoter region was rich in light-responsive elements, indicating that its transcription may be regulated by light signal. By responding to light signal, it regulates the transport efficiency of nucleotide sugars in jujube fruit, thereby affecting the glycosylation process and sugar accumulation. Notably, the expression patterns of LOC107416407 and LOC107409930 were consistent with the dynamic changes in sugar accumulation during fruit development, with significantly higher expression levels in high-sugar genotypes. These findings collectively support the hypothesis that these two genes are involved in sugar metabolism and transport in jujube fruit, which could provide critical support for further elucidating and validating the functions of key genes regulating sugar-related traits in jujube. Conclusion QTL mapping for fruit sugar and acid traits was conducted based on an F₁ hybrid progeny derived from the cross between jujube cultivars ‘Yuhong’ and ‘J5’, using phenotypic data collected over three consecutive years and a high‑density genetic map. A total of 11 related QTLs were detected, distributed across 6 chromosomes. Two stable overlapping intervals were identified that the 35.33 ~ 43.02 cM region on LG1 chromosome is a major co-localized QTL for soluble solids/sugar content, and the 162.86 ~ 217.25 cM region on LG1 chromosome is a stable QTL for titratable acidity. Notably, the sugar-related QTL interval does not overlap with previously reported regions, suggesting it harbors novel genes associated with sugar accumulation in jujube fruit. Furthermore, two key candidate genes (LOC107416407 and LOC107409930) were screened within the sugar‑related interval, encoding phosphofructokinase (a key glycolytic enzyme) and triose-phosphate transporter, respectively, with expression patterns consistent with the dynamic changes in fruit sugar accumulation. These findings not only lay an important foundation for elucidating the genetic regulatory mechanism of sugar metabolism in jujube fruit but also provide stable molecular markers and candidate gene resources for jujube quality breeding, featuring significant theoretical and practical values. Materials and methods Plant materials The female parent was the male sterile jujube cultivar ‘Yuhong’, and the male parent was the superior genotype ‘Jiaocheng 5’ of ‘Junzao’ jujube. Hybridization experiments were conducted over two consecutive years from 2015 to 2016 using net isolation combined with controlled bee pollination techniques. Hybrid progeny were obtained in 2017. In 2018, 140 selected F1 hybrid progeny and two parental cultivars were top-grafted into the orchard of the 10th Regiment, First Division of the Xinjiang Production and Construction Corps. In April of 2021, scions from the three-year-old parental and F1 hybrid progeny in this orchard were collected and again top-grafted onto six-year-old ‘Junzao’ rootstocks at the Red Jujube Experimental Base of the 12th Regiment, First Division of the Xinjiang Production and Construction Corps. This area is located in a warm temperate extreme continental arid desert climate zone, characterized by abundant sunshine, long daylight hours, rich light and heat resources, providing suitable environmental conditions for jujube growth. Finally, 140 F1 individuals and the two parents were used as plant materials. The plants were spaced at 1.5 m×0.5 m, and uniform field management practices were implemented, following conventional irrigation and fertilization methods. Three sugar-acid traits were evaluated during 3 years (2021, 2022 and 2023) at the full red stage of fruit maturity, 30 healthy fruits were collected from the central portion of jujube tree. Flesh tissues from the upper, middle, and lower sections of each fruit were excised, homogenized after dicing, and stored at -20℃ for subsequent analyses. Measurement of soluble solid content, soluble sugar and titratable acid Soluble Solid Content (SSC) was measured using a portable digital refractometer. Fruit flesh was homogenized with a juicer and extracted juice was dipped onto the refractometer prism and data were presented in °Brix. Soluble Sugar (SS) was determined by the anthrone colorimetric method [41]. The glucose standard curve was used for calculation. Titratable Acid (TA) was estimated by acid-base titration [42]. Acid content was calculated based on NaOH titration volume. For these data, calculated the soluble solids-acid ratio (the ratio of SSC to TA) and sugar-acid ratio (the ratio of SS to TA), which reflecting fruit sweetness-acidity balance. Statistical analysis of soluble solid content, soluble sugar and titratable acid Soluble solid content, soluble sugar and titratable acid were tabulated, and descriptive statistics including mid-parent value (VMP), mean, standard deviation (SD), minimum (Min), maximum (Max), coefficient of variation (CV%), heterosis rate (RHm%), transmission ability (Ta%), skewness and kurtosis were calculated using Excel 2010. Phenotypic normality was examined using the Shapiro-Wilk test implemented in Origin 2026 software. Trait correlations within years were analyzed in SPSS 26.0 by Spearman’s rank correlation coefficient (p < 0.001), with correlation strength categorized per standard statistical definitions. QTL mapping A high-density genetic map for the jujube hybrid progeny was constructed by the method of our previous study [43]. Using the T2T genome as a reference [44], QTL analysis for SSC, SS and TA traits was performed with the R/QTL software. Interval mapping was employed for QTL detection. The permutation test was used to determine the limit of detection (LOD). The threshold of the LOD value for significance (p = 0.05) was determined using 1,000 permutations. The calculation of the percentage of soluble solid content, soluble sugar and titratable acid variance explained by each QTL was based on the population variance found within the segregation population. A LOD threshold ≥ 3 was used to declare significant QTLs. For major QTLs, we applied a stricter threshold of LOD ≥ 3.5 and PVE ≥ 10% [45]. QTL was named followed as ‘q+trait abbreviation+year+serial number’, e.g., qSSS23-1 denotes the first QTL for soluble solid substance identified in 2023. And a QTL was considered stable if it was identified for more than 2 years or more than two traits in the same chromosome region. Genomic regions corresponding to QTL intervals were analyzed using the jujube genome assembly (https://www.ncbi.nlm.nih.gov/genome/15586?genome_assembly_id=219393) and functional annotation data. Genes potentially regulating fruit sugar and acid traits were identified through further annotation and analysis. Gene expression analysis of the candidate genes In the ‘Yuhong’ × ‘J5’ hybrid progeny, high-sugar genotype 3J1 and low-sugar genotype 100 were selected as experimental materials. Fruits were sampled at five development stages: early white ripe stage (A), white ripe stage (B), pink spotted stage (C), half-red stage (D), and full-red stage (E). Each time we collected ten fruits, removed the peel and seeds, chopped and homogenized the flesh, and then stored it at -80℃ for RNA extraction. Three biological replicates were included for each treatment. Total fruit RNA was extracted using the Universal Plant RNA kit (Polysaccharide & Polyphenol-rich) from Sanshi Biotechnology, and the concentration and purity of RNA were detected using a Nanodrop spectrophotometer. Reverse transcription was preformed using the All-in-One script RTpremix (with dsDNase) to synthesize the first strand of cDNA, following the manufacturer’s instructions. qRT-PCR primers were designed based on the sequences of candidate genes. Detailed primer information is provided as follows: LOC107416407 (F: GCGTATCTACTGCGACCCAA; R: CGAAAATCCGACGAACACCG), LOC107409930 (F: TGATTCGTTGCTGGGAGGAG; R: TCCAGTGCTCTACCTGCTTC), LOC107416052 (F: GGGTCAAAGCACAGACTCCA; R: AGACTCCACAACCTCCCACA). Using cDNA from jujube fruit samples at different developmental stages as templates, and ZjActin as the housekeeping gene (F: AGCCTTCCTGCCAACGAGT; R: TTGCTTCTCACCCTTGATGC), the temporal expression profiles of candidate genes were analyzed with samples from the early white ripe stage (A) as the control. The 2×SYBR Green qPCR Mix kit was uesd for qRT-PCR analysis. All procedures were performed strictly according to the manufacturer’s instructions. The 10 µL qPCR reaction system consisted: 5 µL of 2×SYBR Green qPCR Mix, 0.2 µL each of forward and reverse primers, 1 µL of cDNA template, and 3.6 µL of RNase-free water. The relative expression levels of genes were calculated using the 2−∆∆Ct method. Statistical analysis of data was conducted with Excel 2010 and SPSS Statistics 20.0 software, and graphs were generated using GraphPad Prism 9.5.0 software. Bioinformatics analysis of the candidate genes Multiple bioinformatics tools were applied to analyze the structural and physicochemical properties of candidate genes. Conserved protein domains were analyzed using NCBI (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi?tdsourcetag). The physicochemical properties of proteins were predicted with Expasy ProtParam (https://www.bioinformatics.org/sms2/prot_param.html). Expasy ProtScale (https://web.expasy.org/protscale/) was used to predict protein hydrophilicity and hydrophobicity. Phosphorylation sites were analyzed via NetPhos 3.1 (https://services.healthtech.dtu.dk/services/NetPhos-3.1/). N-glycosylation sites were analyzed using NetNGlyc-1.0 (https://services.healthtech.dtu.dk/services/NetNGlyc-1.0/), while O-glycosylation sites were analyzed with NetOGlyc-4.0 (https://services.healthtech.dtu.dk/services/NetOGlyc-4.0/). Transmembrane domains of proteins were predicted by TMHMM-2.0 (https://services.healthtech.dtu.dk/services/TMHMM-2.0/). Signal peptides were predicted using SignalP-6.0 (https://services.healthtech.dtu.dk/services/SignalP-6.0/). Subcellular localization was predicted with Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/). The secondary and tertiary structures of proteins were predicted via SOPMA (https://npsa.lyon.inserm.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_sopma.html) and SWISS-MODEL (https://swissmodel.expasy.org/), respectively. Cis-acting regulatory elements of promoters were analyzed using PlantCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/). Amino acid sequence alignment and phylogenetic tree analysis were performed with Tbtools and DNAMAN 9.0. Acknowledgements The authors are grateful to The National-Local Joint Engineering Laboratory of High Efficiency and Superior-Quality Cultivation and Fruit Deep Processing Technology on Characteristic Fruit Trees, Southern Xinjiang Distinctive Forestry & Pomology Technology Innovation Center, Tarim University for providing experimental conditions and technical support. Ethical consideration Not applicable Authors’ contributions All authors contributed to the article conception and design. F.Y. and Y.L. conceived and designed the experiments. Y.L. and Z.L. performed the material collection, investigation, and experiments, Y.L. analyzed the data and wrote the manuscript. Y.L., J.W. C.W and M.L. contributed to the review and editing, F.Y. and M.L performed the review, supervision, editing, conceptualization, investigation and project management and were responsible for financial support. Funding This work was funded by Xinjiang Production and Construction Corps science and technology innovation talent plan project (2023CB009-04), and National Science Foundation of China (32060656). Postgraduate Scientific Research Innovation Project of Tarim University(TDBSCX202505) in 2026. Data availability All relevant data can be found within the paper. The sequenced DNA sequence had been uploaded to the NCBI database (registration number is PRJNA877294) and were also used in our previous publication on the genetic mapping of the whole genome of jujube [43]. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All relevant data can be found within the paper. The sequenced DNA sequence had been uploaded to the NCBI database (registration number is PRJNA877294) and were also used in our previous publication on the genetic mapping of the whole genome of jujube [43].