Regulation of a vacuolar proton-pumping P-ATPase MdPH5 by MdMYB73 and its role in malate accumulation and vacuolar acidification

As the main organic acid in fruits, malate is produced in the cytoplasm and is then transported into the vacuole. It accumulates by vacuolar proton pumps, transporters, and channels, affecting the taste and flavor of fruits. Among the three types of proton pumps (V-ATPases, V-PPases, and P-ATPases), the P-ATPases play an important role in the transport of malate into vacuoles. In this study, the transcriptome data, collected at different stages after blooming and during storage, were analyzed and the results demonstrated that the expression of MdPH5, a vacuolar proton-pumping P-ATPase, was associated with both pre- and post-harvest malate contents. Moreover, MdPH5 is localized at the tonoplast and regulates malate accumulation and vacuolar pH. In addition, MdMYB73, an upstream MYB transcription factor of MdPH5, directly binds to its promoter, thereby transcriptionally activating its expression and enhancing its activity. In this way, MdMYB73 can also affect malate accumulation and vacuolar pH. Overall, this study clarifies how MdMYB73 and MdPH5 act to regulate vacuolar malate transport systems, thereby affecting malate accumulation and vacuolar pH. Supplementary Information The online version contains supplementary material available at 10.1007/s42994-023-00115-7.


INTRODUCTION
Organic acids affect the acidity of fleshy fruits and play an important role in the regulation of osmotic pressure, pH homeostasis, stress resistance, and sensory quality of fruits (Huang et al. 2021).Malate is one of the most important organic acids in fruits, and accounts for 90% of total organic acids in apple.Malate can generate ATP, resist oxidation, and relieve oxidative stress (Etienne et al. 2013).Additionally, it can also relieve angiosclerosis, facilitate absorption of calcium, iron, and other elements, and stimulate secretion by digestive glands.
In fruit cells, malate is mainly synthesized in the cytoplasm, via phosphoenolpyruvate carboxylation (PEP), and is typically catalyzed by phosphoenolpyruvate carboxylase (PEPC) and malate dehydrogenase (MDH) (Sweetman et al. 2009).Malate is then transported into vacuoles for storage.Although malate accumulation in cells is controlled by both metabolism and vacuolar storage, the transport process from the cytoplasm into vacuoles may have a determining effect on malate accumulation.Therefore, it is necessary to thoroughly investigate this transport process.As a matter of fact, this process involves multiple vacuolar transporters, ion channels and carriers, among which vacuolar transporters and channels play a dominant role in the transport of organic acids, such as the tonoplast dicarboxylate transporter (tDT) (Emmerlich et al. 2003) and an aluminium-activated malate transporter (ALMT) (such as ALMT6 and ALMT9/Ma1) (Emmerlich et al. 2003;Meyer et al. 2011;Cohen et al. 2014;Martinoia 2018).In addition, some proton pumps on the tonoplast also play a key role in this process by transporting H ? into vacuoles and acidifying vacuoles.In this way, the proton electrochemical gradients that serve as the driving force for the transport of malate into vacuoles are generated.
To date, three types of proton pumps, including vacuolar H ? -ATPase (V-ATPase), vacuolar H ? -PPase (V-PPase), and vacuolar P-type ATPase (P-ATPase), have been identified in fruits (Hurth et al. 2005;Kovermann et al. 2007;Martinoia et al. 2007).As a novel proton pump thoroughly investigated in recent years, the P-ATPase was first identified in petunia and may be associated with vacuole acidification.In plants, P-ATPase consists of five major evolutionarily related subfamilies (P1-P5).Among them, ATPases in the P3 subfamily are responsible for energizing the electrochemical gradient serving as the driving force of secondary transport (Pedersen et al. 2012).The P-type ATPases (PhPH5 and PhPH1) found in petunia are involved in the vacuolar acidification of petal cells, whereas only PhPH5 shows independent proton transport activity (Verweij et al. 2008;Faraco et al. 2014;Li et al. 2016).Moreover, their homologous genes were also identified in fruits.In citrus, the homologous gene, CitPH5, is closely related to the generation of superacidified fruit (Strazzer et al. 2019).In pear, PbPH5 is localized to the vacuolar membrane and can promote malate accumulation in fruits (Song et al. 2022).As a homologue of PhPH5 in apple, MdPH5 is slightly upregulated in transgenic MdMYB73 callus, which is responsible for malate accumulation and vacuolar acidification (Hu et al. 2017), suggesting that MdPH5 may also be associated with malate accumulation in apple.Nevertheless, the specific function of MdPH5 in apple fruits remains unclear.
The regulation of organic acid transporters and proton pumps involves complex gene regulatory networks.Indeed, transcriptional regulation is one of the most common events.Among them, MYB transcription factor (TF) families comprise plant-specific R2R3-MYB TFs, many of which are involved in the transport of organic acids by activating or inhibiting gene expression of transporters and proton pumps.Remarkably, MdMYB1, MdMYB44, and MdMYB73, which are present in apple, can affect the transcriptional activity of malate transporters and proton pumps to regulate malate accumulation and vacuolar acidification (Hu et al. 2016(Hu et al. , 2017;;Jia et al. 2021).CrMYB73, which is homologous to MdMYB73, leads to increased citrate accumulation in citrus plants, whereas its downstream target genes remain unclear (Li et al. 2015).In addition, PhPH4 in petunia is an R2R3-MYB TF that plays a similar role to VvMYB5a and VvMYB5b in grape for the regulation of citrate accumulation.Specifically, they both can activate expression of the downstream genes PH1 and PH5, thus acidifying vacuoles (Cavallini et al. 2014;Kasajima et al. 2016;Amato et al. 2019).In soybean petals, GmPH4, which is an R2R3-MYB TF, is also involved in vacuolar acidification as it directly regulates the expression of GmPH5, a P 3A -type ATPase gene (Sundaramoorthy et al. 2020).However, the specific regulators of MdPH5 in apple have not yet been identified.
In this study, the role of MdPH5 in regulating malate accumulation and vacuolar acidification in apple was investigated and its transcriptional regulation by MYB TF MdMYB73 was verified.This study provides a reference for understanding the molecular factors involved in fruit quality.

Plant materials and growth conditions
'Orin' apple calli obtained from young embryos were subcultured at 3-week intervals on MS medium supplemented with 1.5 mg/L 2,4-dichlorophenoxyacetic acid (2,4-D) and 0.4 mg/L 6-benzylaminopurine (6-BA) at 25 °C, in the dark.Subsequently, these calli were subcultured, at half-month intervals, three times before being used for further studies.
'Royal Gala' apples were collected at 41, 70, 94, 128 days post-flowering.'Royal Gala' apple were collected at 120 days post-flowering and stored in airconditioned tanks.Batch of apples of similar size, color, maturity, disease-free, insect-free, and without mechanical damage were selected for storage for 120 days, and samples were taken every 30 days.These apples were frozen in liquid nitrogen.

Phylogenetic analysis of PH5 proteins
The MEGA_X based on the neighbor-joining method and bootstrap analysis with 1000 replications was used to construct the phylogenetic tree of MdPH5 and Arabidopsis thaliana P 3A subfamily.

Analysis of the MdPH5 promoter
The cis element in the MdPH5 promoter (1500 bp upstream of the transcription initiation site) was analyzed with the online software PlantCARE (http://bioin formatics.psb.ugent.be/webtools/plantcare/html/).

Construction of the MdPH5 gene expression vector and genetic transformation of apple calli
The ORF sequence of MdPH5 was cloned into pRI-101 to obtain the overexpression vector, and the non-conserved regions of MdPH5's ORF were reversely cloned into pRI-101 to obtain the antisense vector.Transgenic apple calli were obtained by Agrobacterium-mediated transformation.

Quantitative real-time-PCR (RT-qPCR) analysis
Plant RNA was extracted with an RNA extraction kit (TIANGEN, Beijing, China) and single-stranded cDNA was obtained with a reverse transcription kit (TaKaRa, Shiga, Japan).The RT-qPCR analyses were executed with three biological and technical replications to test the expression levels of MdPH5, which were performed with the methods as described by Hu et al. (2016).The quantitative analysis of results used the 2 -DDCT method.

Analysis of subcellular localization
The full-length coding sequences of MdPH5 were fused to the GFP protein, to construct the fusion expression vector 35S:MdPH5-GFP, and the resulting plasmid was transformed into Agrobacterium strain LBA3101.The constructed vector was injected into tobacco (Nicotiana benthamiana) epidermal cells and cultured in the dark for 3 days.The AtCBL-red fluorescent protein (RFP) was used as a vacuolar membrane marker (Ma et al. 2019) and was co-transformed with 35S:MdPH5-GFP.Fluorescence images were obtained at 488 nm with a highresolution laser confocal microscope (LSM880, Zeiss, Meta, Jena, Germany).

Measurement of vacuolar pH
Isolation of protoplasts from apple calli and measurement of vacuolar pH were carried out, as previously described by Hu et al. (2016).The vacuolar pH was detected with the cell-permeant and pH-sensitive fluorescent dye, 2 0 ,7 0 -bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF)-AM, while vacuolar pH was quantified by the ratio of pH-dependent (488 nm) and pH-independent (458 nm) excitation wavelengths from a calibration curve, and ratio images were generated with the ion concentration tool of Zeiss LSM confocal software.

Determination of malate content
Malate content was measured by high-performance liquid chromatography, as previously described by Hu et al. (2016).

EMSA
EMSA was conducted according to Xie et al. (2012).MdMYB73 was cloned into the expression vector pGEX4T-1.The MdMYB73-GST recombinant protein was expressed in Escherichia coli strain BL21.An oligonucleotide probe of the MdMYB73 promoter was labeled using an EMSA probe biotin labeling kit (Beyotime) according to the manufacturer's instructions.The recombined protein of MdMYB73-GST was incubated with 10 9 binding buffer, 1 lg/lL poly (dI-dC), and 400 fmol of biotin-labeled double-stranded binding consensus oligonucleotides (total volume 20 lL) using a LightShift Chemiluminescent EMSA Kit (Thermo Scientific).The binding reaction was performed at room temperature for 20 min.The DNA-protein complexes were separated on 6.5% non-denaturing polyacrylamide gels, electrotransferred, and detected following the manufacturer's instructions.The binding specificity was also examined by competition with a fold excess of unlabeled oligonucleotides.

LUC assay
The apple MdPH5 promoter fragments were amplified by PCR and cloned into the pGreenII 0800-LUC vector to construct the LUC reporter vector (MdPH5pro-LUC).The full-length coding sequences of MdMYB73 were cloned into the effector vector pGreenII 62-SK.Individual combinations of reporter vectors and effector vectors were transformed into Agrobacterium strain GV3101 cells alongside the pSOUP vector.The Agrobacterium strains were used to tobacco (Nicotiana benthamiana) epidermal cells.A live-imaging apparatus was used to measure luminescence after 2 days.

Chromatin immunoprecipitation qPCR analysis
35S:MdMYB73-GFP and 35S::GFP transgenic apple cultures were used for the ChIP-qPCR analysis.The anti-GFP antibody (Beyotime) was used for chromatin immunoprecipitation (ChIP), as described by Xie et al. (2012).The resultant samples were used as templates for qPCR assay.

Data presentation and statistical analysis
The data obtained in this study were analyzed by DPS Software (Enfield, UK), with P \ 0.05 considered as indicative of significant differences.

Analysis of transcriptome and multiple metabolites in apple fruit at different developmental stages
To clarify the trends of different metabolites in apple fruits after flowering, the contents of major carbohydrates and malate in apple fruits, 41, 70, 94, and 128 days after blooming (DAB), were determined by gas chromatography-mass spectrometry (Fig. S1A).These results showed that the malate content decreased gradually and regularly after flowering (Fig. 1A).The levels of both galactose and sorbitol were minimized by 70 DAB, whereas the level of starch was maximized by 70 DAB (Fig. S1B).
A total of 12 samples, collected at 4 stages were investigated, using transcriptome analysis.The number of upregulated and downregulated genes, in 41 DAB vs. 70 DAB, 70 DAB vs. 94 DAB, and 94 DAB vs. 128 DAB, were 1746and 4426, 1289and 1876, and 2435 and 4939, respectively (Fig. S2A).Figures S2B and S2C are Venn diagrams reflecting the overlap of differential genes in different cases.As observed, the malate content decreased regularly and the expression of genes that might be associated with malate showed changes according to the transcriptomic data, as shown in the heat maps.Figure S2D shows the expression levels of 32 malate-related genes, at different developmental stages.
According to the heat maps of the expression levels of malate-related genes, the malate content decreased significantly from 41 to 71 DAB.As observed, genes such as MD03G1155400, MD13G1044200, and MD17G1155800 were significantly downregulated (Fig. S3A).To visualize changes of malate-related genes and identify genes with the highest correlation with malate, three volcano maps (41 DAB vs. 70 DAB, 41 DAB vs. 94 DAB, and 41 DAB vs. 128 DAB) were developed.The number of downregulated genes in the volcano maps were 4, 6, and 6, respectively.Among these genes, MD17G1155800 was the only malate-related gene present in all data sets (Fig. 1B; Fig. S3B, C).As malate was significantly downregulated in 41 DAB vs. 70 DAB, 41 DAB vs. 94 DAB, and 41 DAB vs. 128 DAB, MD17G1155800 may be positively correlated with malate.It has been reported that Md17G1155800 is a P-type proton pump (MdPH5), and its homologous gene, PbPH5, in petunia can acidify vacuoles and change petal color (Faraco et al 2014).

Correlation of MdPH5 expression with malate content during apple fruit ripening and postharvest
To further clarify the role of MdPH5 in apple, the level of MdPH5 expression in the transcriptome dataset was monitored.The results showed that its expression level decreased (Fig. 1C).Correlation analysis of MdPH5 and different carbohydrates and acid substances revealed that MdPH5 expression was highly correlated with malate (R 2 = 0.8787; P \ 0.05) (Fig. 1D).Unlike for the acids, the contents of galactose, starch, sorbitol, maltose, and glucose had low correlations with the expression of MdPH5 (Fig. S4A-E).Although the correlation between MdPH5 expression and fructose content was relatively high (R 2 = 0.8159) (Fig. S4F), unlike the correlation between MdPH5 expression and malate content, the correlation between them was negative.These results indicated that MdPH5 is positively correlated with malate accumulation in apple plants.
To clarify the impacts of MdPH5 on malate content, during storage, Gala apple fruit stored for 0, 30, 60, 90, and 120 days were employed as test samples.These results demonstrated that the malate content decreased gradually with the extension of storage (Fig. 1E).Subsequently, RT-qPCR analysis of samples, at different post-harvest stages, showed that the expression of MdPH5 also decreased gradually with increasing storage time (Fig. 1F), indicating a positive correlation of MdPH5 expression with the malate content (R 2 = 0.8421; P \ 0.05) (Fig. 1G).

Bioinformatics analysis and subcellular localization of MdPH5 protein
The cDNA of MdPH5 was 2850 bp in full length and encodes for 950 amino acids.As shown in Fig. 2A, the cDNA of MdPH5 comprised three conserved domains.Herein, the secondary protein structure of MdPH5 was predicted.The results showed that random coils (69.88%), alpha-helices (30.00%), extended-strands (17.05%), and beta-turns (5.58%) were dominant in the secondary protein structure of MdPH5 (Fig. 2B).
A phylogenetic tree was established, using MEGA-X, to investigate the genetic correlation of MdPH5 with the P 3A subfamily ATPases in Arabidopsis thaliana.These results indicated that MdPH5 (MD17G1155800) had the closest genetic correlation with At1G17260 (Fig. 2C).
As MdPH5 encodes a P-ATPase, pCaMV35S::MdPH5-GFP fusion vectors were developed and visualized by transient expression in leaves of N. benthamiana, with pCaMV35S::GFP as a negative control, so as to determine the subcellular localization of MdPH5.Herein, different profiles of the vacuolar membrane-labeled red fluorescence and the green fluorescence of the MdPH5-GFP protein were observed.Additionally, the fluorescence signal of MdPH5-GFP overlapped with that of the AtCBLlabeled vacuolar membrane marker (Fig. 2D).Therefore, it could be concluded that MdPH5 is localized to the vacuolar membrane.

Role of MdPH5 in regulating malate accumulation and vacuolar pH in apple
Transgenic calli with overexpression or silencing of MdPH5 were acquired to investigate the functions of the gene (Fig. 3A, B).The expression of MdPH5 in the overexpression or silencing groups was significantly higher and lower than that in the control group, respectively, indicating successful preparation of the gene overexpression or silencing materials (Fig. 3C).Compared with the control group, the overexpression or silencing groups exhibited increased and decreased malate contents (Fig. 3D), respectively, indicating that MdPH5 favors malate accumulation in calli.The effects of MdPH5 on vacuolar pH were investigated on the basis of BCECF [2 0 ,7 0 -Bis(2-carboxyethyl)-5(6)-carboxyfluorescein], which is a ratiometric fluorescent pH indicator.The average vacuolar pH value of WT apple calli was 3.66, whereas that of MdPH5 overexpression or silencing apple calli was 3.96 and 3.41, respectively (Fig. 3E, F).Overall, it could be inferred that MdPH5 regulates malate accumulation and vacuolar pH in apple calli.
To further verify the effects of MdPH5 on malate, the expression of MdPH5 in apple was tuned by a virusvector-based transformation method.Two virus constructors (MdPH5-IL60 and MdPH5-TRV) were injected into the fruit by agrobacterium transformation, with empty vectors as the control (Fig. 3G).The expression level of MdPH5 was aligned with expectation by a 7-day incubation in darkness (Fig. 3H).Then the malate content in the MdPH5-IL60 group was measured and compared with that in the control group.As indicated, the average malate content in the MdPH5-IL60 group increased by 0.671 mg/g.Similarly, the average malate content in the MdPH5-TRV group decreased by 0.670 mg/g, which was consistent with that in the transgenic calli (Fig. 3I).

Role of MdMYB73 in MdPH5 expression
Previous studies have shown that some TFs can directly activate the expression of several vacuolar proton pump subunit genes (Hu et al. 2016(Hu et al. , 2017)).To elucidate the transcriptional regulatory mechanism of MdPH5 in apple, cis-acting element analysis of its promoter was conducted and the results demonstrated the presence of abundant MYB-binding cis elements in the MdPH5 promoter.This may be attributed to the presence of some MYB TFs that are located upstream of MdPH5 and regulate its expression.After that, the MYB TFs interacting with MdPH5 were identified by yeast one hybrid assays.These results demonstrated that MdMYB73, which was previously identified as a major TF regulating malate, is indeed a candidate partner.
Chromatin immunoprecipitation (ChIP)-PCR assays were employed to investigate the in vivo binding of MdMYB73 to the MdPH5 promoter.Herein, 35S::MdMYB73-GFP and 35S::GFP transgenic apple cultures were used.The MYB-binding site element (CAA-CAG) in Region S4 of the MdPH5 promoter was enriched in the 35S::MdMYB73-GFP transgenic cultures, though the MYB-binding element in other three regions was not enriched (Fig. 4A, B).The results provided in vivo evidence for the binding of MdMYB73 to the MdPH5 promoter.
An electrophoretic mobility shift assay (EMSA) was performed to investigate the in vitro binding of MdMYB73 to the MdPH5 promoter.As observed, MdMYB73 bound to the MdPH5 promoter fragments containing the CAACAG motifs, and the level of a specific DNA-MdMYB73 protein complex decreased with the increase in the number of unlabeled MYB-competing probes with the same sequence.However, these complexes were not observed if the CAACAG motif was changed to the AAAAAA motif (Fig. 4C).These results The luciferase (LUC) transactivation assay was employed to clarify the effects of MdMYB73 on the activity of the MdPH5 promoter.As indicated, promoter-LUC reporter plasmids were expressed, transiently, in the leaves of N. benthamiana, via transfection with Agrobacterium tumefaciens strain GV3101.The inflitrated leaves containing the sequences of MdPH5 promoter showed higher LUC activity (Fig. 4D).In summary, these findings support the notion that MdPH5 transcription is activated by MdMYB73.
In addition, the expression levels of MdMYB73 and MdPH5 in 35S::MdMYB73-GFP transgenic apple plantlets were investigated.As indicated, the expression levels of both genes in the transgenic apple plantlets were significantly higher than those in the control group (Fig. 4E).

Role of MdMYB73 as an upstream gene of MdPH5 in regulating malate accumulation
To clarify the regulatory role of MdMYB73 and MdPH5 in malate accumulation, four virus builders (IL60 ?TRV, MdPH5-TRV, MdMYB73-IL60, and MdPH5-TRV ?MdMYB73-IL60) were used for fruit injection tests (Fig. 5A).The results of RT-qPCR showed that, compared with IL60 ?TRV, the relative transcription levels of MdMYB73 and MdPH5 were consistent with expectations (Fig. 5B, C) and the transcription level of MdPH5 decreased with downregulated expression of MdMYB73.Subsequently, the malate content was determined.As indicated, the malate concentration in the injection area of MdMYB73-IL60 was higher than that in the control group, whereas that in the injection area of MdPH5-TRV was lower than that in the injection area of the control group.Additionally, the malate content in the injection area of MdMYB73-IL60 ?MdPH5-TRV was higher than that in the injection area of MdPH5-TRV, but slightly lower than that in the injection area of the control group.In other words, MdPH5-TRV may partially offset the effects of MdMYB73-IL60 in the injection area of MdMYB73-IL60 ?MdPH5-TRV (Fig. 5D).In summary, MdMYB73 is genetically upstream of MdPH5 in the regulation of malate accumulation in apple fruit.

DISCUSSION
Acidity is one of the most important quality-related features of apple, which directly affects fruit flavor and quality.Apple contains abundant organic acids, among which malate is the dominant species, as it accounts for more than 90% of the total organic acids (Yu et al. 2021).Widely distributed in various plant tissues and organs, malate has various important functions in the life cycle of plants (Fernie et al. 2004;Noguchi and Yoshida 2008;Sweetman et al. 2009;Bai et al. 2015;Hu et al. 2017;Dong et al. 2018).Vacuolar proton pumps and malate transporters are essential key factors influencing the transport of malate into vacuoles (Terrier et al. 2001;Bai et al. 2015;Ma et al. 2015Ma et al. , 2019;;Hu et al. 2016).In this study, MdPH5, a P-type ATPase, was shown to promote vacuolar acidification and malate accumulation in apple calli and fruit.As an upstream MYB TF of MdPH5, MdMYB73 binds to the MdPH5 promoter and activates its expression, thereby facilitating malate accumulation.In addition, the expression of MdPH5 was positively correlated with malate content, at both the developmental and post-harvest stages.
Vacuolar proton pumps have significant effects on fruit acidity, as they can deliver H ? to acidifying vacuoles, thereby facilitating the accumulation of organic acids (Emmerlich et al. 2003).As MdPH5 acidifies apple vacuoles, its homolog PhPH5 acidifies petal cell vacuoles.Nevertheless, PhPH5 affects petal color rather than the accumulation of organic acids (Faraco et al. 2014).P-type ATPases promote the accumulation of organic acids and play an important role in fruits.Ma10, a candidate gene for the acidity of apple fruit, encodes a P 3A -type ATPase proton pump that promotes the transport of malate into vacuoles and vacuole acidification (Ma et al. 2019).In pears, the PH5 gene also promotes malate accumulation (Song et al. 2021).Overall, these genes exhibit similar functions to MdPH5 in apple.However, downregulated expression of CitPH5 led to the formation of some low-acid varieties of some fruits, including lemon, orange, pummelo, and rangpur lime (Strazzer et al. 2019).Despite that citric acid is the dominant organic acid in these fruits, this study focused on the effects of MdPH5 on malate accumulation, but not citric acid accumulation in apple.In addition, the cis element of MdPH5 promoter was further analyzed using the PlantCARE.As shown in Supplemental Table 2, there are many cis elements involved in light responsiveness, drought inducibility, and hormone responsiveness (jasmonic acid, salicylic acid, and auxin).These findings further suggest that the MdPH5-mediated malate metabolism may be induced by various signals, such as plant hormones, stress, and light.
Malate content is subjected to the synergistic effects of environmental factors, developmental and metabolic signaling pathways, and corresponding TFs.TFs can regulate malate accumulation either individually or in the form of complexes (Etienne et al. 2013;Hu et al. 2016Hu et al. , 2017;;Li et al. 2020).In apple, the transcription activities of malate transporters and proton pumps can be significantly tuned by regulating MdMYB1, MdMYB44, and MdMYB73, which can effectively regulate malate accumulation and vacuolar acidification (Hu et al. 2016(Hu et al. , 2017;;Jia et al. 2021).However, MdMYB1 and MdMYB73 are positive regulators, whereas MdMYB44 is a negative regulator, and they have different downstream target genes.Indeed, the direct downstream target genes of MdMYB1 are MdVHA-B1, MdVHA-E, MdVHF1, and MdtDT, whereas MdMYB44 inhibits the promoter activity of MdVHA-A3, MdVHA-D2, Ma10, and MdALMT9.In contrast, MdMYB73 directly activates the expression of MdVHA-A, MdVHP1, and MdALMT9.In this study, MdPH5 was identified to be another downstream target gene of MdMYB73.Similarly, MdMYB73 could regulate MdPH5 to promote malate accumulation (Hu et al. 2016(Hu et al. , 2017;;Jia et al. 2021;Fig. 6).Notably, MdMYB44 has been demonstrated to regulate both V-ATPase and P-ATPase genes (Jia et al. 2021).To date, only a few downstream V-ATPase target genes for MdMYB73 TF have been reported.MdPH5 is a P 3A -type ATPase, suggesting that MdMYB73 can also simultaneously regulate both types of proton pumps.
Besides MYB TFs, WRKY, bHLH, ERF, and other TFs are also involved in the regulation of organic acid accumulation and vacuole acidification.An InDel in the SlALMT9 promoter disrupts a W-box binding site, thereby preventing the binding of the WRKY TFs (Ye et al. 2017).In apple, MdbHLH3 (bHLH TF) directly regulates the expression of MdcyMDH, which is a cytosolic malate dehydrogenase gene, to mediate carbohydrate allocation and malate accumulation (Yu et al. 2021).CitERF13 (citrus TF) regulates citrate accumulation by directly activating CitVHA-c4, which is a vacuolar proton pump gene (Li et al. 2016).Also, the formation of MBW complexes can directly regulate the expression of downstream acid-related target genes.As reported, MdbHLH3, MdbHLH49, and MdCIbHLH1 interact with MdMYB1, MdMYB44, and MdMYB73, respectively, and enhance the activities of corresponding MYB TFs to further regulate the activities of downstream genes, including malate transporters and vacuolar proton pumps (Xie et al. 2012;Hu et al. 2016Hu et al. , 2017;;Jia et al. 2021).Therefore, there may be more complex regulatory relationships in addition to the proposed MdMYB73-MdPH5 malate regulatory pathway.In addition, the cis element of MdMYB73 promoter was further analyzed using the PlantCARE.As shown in Supplemental Table 3, there are many cis elements involved in light responsiveness and drought inducibility.These results further suggest that the proposed MdMYB73-MdPH5 malate regulatory pathway may be induced by light and drought signals.Fruit quality is one of the most concerned features in fruit breeding.It has been demonstrated that organic acids significantly affect fruit flavor.Malate is one of the most important organic acids in apple, while the genetic basis for its accumulation in apple vacuoles remains unclear (Wu et al. 2007;Zhang et al. 2012).In this study, MdPH5, a P 3A -type vacuolar proton pump, was demonstrated to promote malate accumulation and vacuolar acidification, and MdMYB73, an upstream MYB TF of MdPH5, can facilitate malate accumulation by binding to its promoter and activating its expression (Fig. 6).This study provides a new method for acidity regulation of apple fruit, and may facilitate the development of new varieties with improved flavor and stress resistance.

Fig. 1
Fig. 1 Correlation of MdPH5 expression with malate content in apple fruit during fruit development and storage.A Malate contents in apple fruit on different DAB.B Volcanic maps of malate-related genes in the 41 DAB vs. 70 DAB group.C The expression levels of MdPH5 on different DAB.D Correlation of the MdPH5 expression with malate content in apple at different developmental stages.E Malate contents at different storage periods.F Expression of MdPH5 in different storage periods determined by RT-qPCR.G Correlation of MdPH5 expression with malate content at different storage periods.Herein, letters indicate significant differences (P \ 0.05) among the culture media for the various parameters, based on the two-way ANOVA, followed by Duncan's multiple range test.Bars are SE (n = 3)

Fig. 2
Fig. 2 Bioinformatics analysis of the MdPH5 gene and subcellular localization of the MdPH5 protein.A Conserved sequence of the MdPH5 gene; a1 refers to ATPase-N, a2 refers to E1-E2_ ATPase, and a3 refers to hydrolase.B Predicted secondary structures of the MdPH5 protein.The numbers denote the length of amino acids.C Phylogenetic tree of the MdPH5 protein and ATPase of the Arabidopsis thaliana P 3A subfamily.D Subcellular localization of the MdPH5 protein with a AtCBL tonoplast marker.Herein, lines and boxes highlight the position of vacuoles exhibiting green and red fluorescence, respectively.35S::GFP refers to the control group, scale bar = 10 lm, and B represents a bright field

Fig. 3
Fig. 3 Functions of MdPH5.A, B Transgenic callus materials obtained.C The gene expression of MdPH5 in the wild-type (WT) and MdPH5 transgenic apple calli (overexpression and antisense) determined by RT-qPCR.D Malate contents in MdPH5-OVX and MdPH5-RNAi transgenic apple calli.E Emission intensities of protoplast vacuoles in WT and transgenic calli MdPH5-OVX and MdPH5-RNAi loaded with wit20,70-bis-(2-carbox-yethyl)-5-(6)-carboxyfluorescein at 488 nm (first column) and 458 nm (second column).The pseudo-color scale on the right indicates the fluorescence intensity.Scale bar = 10 lm.Lowercase letters indicate significant differences at P \ 0.05.All values are the mean ± SD of three independent replicates.F Quantification of the luminal pH in vacuoles of WT and transgenic calli MdPH5-OVX and MdPH5-RNAi.Error bars represent the SE of 5 measurements from 30 individual intact vacuoles.Lowercase letters indicate significant differences at P \ 0.05.G Apple fruit injected with plasmid mixtures (IL60: IL60-1 ?IL60-2; MdPH5-IL60: IL60-1 ?MdPH5-IL60-2).An empty IL60 vector was used as the control group.A solution containing agrobacterium cells (TRV: TRV1 ?TRV2; MdPH5-TRV: TRV1 ?MdPH5-TRV2) was injected into the apple tissue, with an empty TRV vector serving as the control group.Scale bar = 2 cm.Measured expression levels of MdPH5 (H) and malate contents (I) at the injection sites.Significant differences are indicated by the use of lowercase letters if P \ 0.05.All values are the mean ± SD (n = 5)

Fig. 4
Fig. 4 MdMYB73 regulates malate accumulation by binding to MdPH5.A The putative MYB-binding element on the MdPH5 promoters; the numbers represent the integrated position.B ChIP-qPCR results indicating the enrichments of the target gene promoters in the 35S::MdMYB73-GFP transgenic apple seedlings compared with the 35S::GFP transgenic apple seedlings.C Results of electrophoretic mobility shift assays of the interaction between MdMYB73 and labeled DNA probes in the MdPH5 promoters.Lane 1 of each blot shows the labeled DNA probes without the MdMYB73 protein; lane 2 shows the labeled DNA probes and the MdMYB73 protein without a competitor.Different amounts (9 5 and 9 10) of unlabeled DNA fragments were added as cold competitors.D Results of LUC activity assays indicating that MdMYB73 enhances the basal activity of the MdPH5 promoters.E The expression level of MdPH5 in MdMYB73 transgenic apple seedlings.A significant difference is indicated by distinct lowercase letters at P \ 0.05.All values are the mean ± SD (n = 3)

Fig. 5
Fig. 5 Verification of the upstream and downstream relationship of MdMYB73 and MdPH5 genes.A Injected apples kept in darkness for a week.Scale bar = 2 cm.B, C Expression of related genes in apple fruit after injection.D Malate contents in different groups.Error bars represent the SE of five measurements from at least independent biological replicates

Fig. 6
Fig. 6 Working model showing that MdMYB73 binds to the MdPH5 promoter to regulate malate accumulation in apple