Interaction of BPC1 and ALDH2 Affects Natural De-Astringency in Chinese PCNA Persimmon (Diospyros kaki)

Abstract

Pollination-constant non-astringent (PCNA) persimmons have significant commercial value due to their natural de-astringency trait. The Chinese PCNA (C-PCNA) type is particularly valuable for genetic improvement because this trait is controlled by dominant genes. However, the regulatory mechanism underlying this trait remains unclear. Our previous research identified ALDH2 (acetaldehyde dehydrogenase 2, a key gene downstream of acetaldehyde metabolism) as being negatively correlated with natural de-astringency in C-PCNA persimmon, and revealed its interaction with DkBPC1, a transcription factor from the BBR/BPC (BARLEY B RECOMBINANT/BASIC PENTA CYSTEINE, abbreviated as the BPC protein family) family whose function had not been experimentally validated. The full-length cDNA of DkBPC1 was isolated from ‘Luotian Tianshi’ (C-PCNA type). A dual-luciferase assay demonstrated that DkBPC1 significantly enhances the promoter activity of DkALDH2b. Subcellular localization conducted in tobacco confirmed that DkBPC1 is localized in the nucleus. Transient overexpression of DkBPC1 in C-PCNA leaves and fruit discs resulted in a significant increase in soluble tannin content, a significant decrease in insoluble tannin content, and a notable upregulation of DkALDH2b gene expression. Conversely, transient knockdown of DkBPC1 expression in C-PCNA leaves led to a dramatic drop in soluble tannin content and a significant increase in insoluble tannin content. These results indicate that BPC1 can increase the conversion of soluble tannins to insoluble tannins by downregulating the expression of the DkALDH2 gene, thereby promoting natural de-astringency in C-PCNA persimmon.

1. Introduction

Persimmon (Diospyros kaki Thunb.), a species of the genus Diospyros in the family Ebenaceae, is a globally significant economic fruit tree [1,2]. Its fruits are valued for fresh consumption, processing, and medicinal applications, and its industry has experienced rapid development in recent years [3]. Based on whether the fruit astringency is lost or not at maturity, cultivated persimmon varieties are primarily classified into two major types, PCNA and non-PCNA [4]. The former hold significant commercial value because their fruits undergo a natural de-astringency trait, and it is subdivided into Chinese PCNA (C-PCNA) and Japanese PCNA (J-PCNA) according to the differences in genetic characteristics [5]. Furthermore, since the natural de-astringency trait is controlled by dominant genes in C-PCNA [6], it possesses considerable research value for the genetic improvement of PCNA persimmons, which has demonstrated enormous commercial value due to its popularity in the market [7]. It is noteworthy that the C-PCNA, as a unique genetic resource native to the Dabie Mountains in China, possesses a distinct genetic basis for its natural de-astringency, endowing it with an irreplaceable value as a gene pool for global PCNA persimmon breeding [8,9]. Investigating the regulatory network of natural de-astringency in C-PCNA and subsequently identifying key genes within this network are of great importance for persimmon cultivar improvement.

Studies have revealed that the natural de-astringency process in C-PCNA involves both a “dilution effect” (the decrease in soluble tannin concentration during fruit expansion) and a “coagulation effect” (the conversion of soluble tannins into their insoluble form), with the latter potentially playing a more critical role [9]. Acetaldehyde is instrumental in the coagulation effect. We discovered that the acetaldehyde dehydrogenase ALDH2 family, key enzymes downstream of acetaldehyde metabolism, inhibits natural de-astringency in C-PCNA. Furthermore, a yeast-one hybrid system using the DkALDH2b promoter identified a BBR/BPC (BARLEY B RECOMBINANT/BASIC PENTA CYSTEINE, abbreviated as the BPC protein family) transcription factor, DkBPC1 [10]. However, the function of this transcription factor had not been experimentally validated.

The BBR/BPC is a class of plant-specific transcription factors discovered and first studied in 2002 [11,12]. Studies in model plants such as Arabidopsis [13], tobacco [14], and tomato [15] have shown that BPC proteins can influence gene expression by interacting with GA-rich motifs in target gene promoters. Furthermore, BPCs are implicated in playing diverse roles in regulating a multitude of genes during plant growth and development [16]. It can both co-regulate gene expression with other transcription factors and influence various plant endogenous hormone signalling pathways [17,18]. Current functional studies of BPC proteins are predominantly concentrated on model plants, with relatively limited research in fruit trees, and no relevant studies have been reported in persimmon.

In this study, through the functional characterization of the candidate transcription factor DkBPC1, we aim to construct a molecular regulatory network underlying natural de-astringency in Chinese PCNA persimmon. This work will facilitate the identification of key genes responsible for natural de-astringency and provide a foundation for the genetic improvement of PCNA persimmon at the molecular level.

2. Materials and Methods

2.1. Plant Materials and Sampling

The experimental materials primarily consisted of three persimmon varieties with different de-astringency types: the C-PCNA varieties ‘Luotian Tianshi’ (D. kaki Thunb. ‘Luotian Tianshi’) and ‘Eshi 1’ (D. kaki Thunb. ‘Eshi 1’), the J-PCNA variety ‘Youhou’ (D. kaki Thunb. ‘Youhou’), and the non-PCNA variety ‘Mopanshi’ (D. kaki Thunb. ‘Mopanshi’). Fruit samples from different varieties collected at 2.5–27.5 weeks after bloom (WAB) and ‘Eshi 1’ fruit at 16 WAB were all obtained from the Persimmon Orchard of Huazhong Agricultural University. Each sample was collected with three biological replicates. Immediately after collection, samples were transported to the laboratory, peeled, and the equatorial pulp was rapidly frozen in liquid nitrogen and stored at −80 °C.

2.2. Analysis of Tannin Content

One gram of fruit pulp was ground into powder for tannin content determination. Soluble tannins were extracted using a 70% acetone solution (containing 1% VC), and insoluble tannins were extracted using an HCl-butanol solution [19]. Absorbance was measured on an Evolution 220 UV-vis spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at 640 nm and 550 nm, respectively.

2.3. Phylogenetic Analysis

Amino acid sequences of BPC proteins from four species were downloaded from NCBI. A phylogenetic tree was constructed using MEGA 6 software with the neighbour-joining (NJ) method. Multiple sequence alignment of the amino acids was performed using the Clustal Omega (v1.2.2) online website (http://www.clustal.org/omega/, accessed on 14 February 2023).

2.4. Extraction of RNA and Synthesis of cDNA

Total RNA was extracted from frozen fruit pulp using the Hipure HP Plant RNA Mini Kit (Magen, Guangzhou, China). The quantity and quality of the RNA were assessed by gel electrophoresis and a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Three biological replicates were performed for each sample. First-strand cDNA was synthesized using a PrimeScript RT Kit with gDNA Eraser (TaKaRa, Beijing, China).

2.5. qRT-PCR Expression Analysis

qRT-PCR was performed on a LightCycler^® 480 II Real-Time PCR System (Roche, Mannheim, Germany). The PCR reaction mixture (total volume 10 μL) contained TB Green Premix Ex Taq II (TaKaRa, Beijing, China), primers, ddH₂O, and cDNA. The PCR protocol consisted of 95 °C for 30 s, followed by 45 cycles of 95 °C for 5 s, 58 °C for 10 s, and 72 °C for 15 s, then 95 °C for 60 s and 40 °C for 30 s. Four technical replicates were performed for each gene, and the relative gene expression was calculated by the Ct (2^−∆∆Ct) method. D. kaki Thunb. Actin (GenBank Accession No. AB219402) was used as the reference gene, and the relevant primers are listed in Table S1.

2.6. Isolation of Genes and Promoters

Based on the full-length sequence information of the DkBPC1 gene, its complete cDNA was amplified using the SMART RACE cDNA Kit (Clontech, San Jose, CA, USA). Utilizing the previously obtained promoter sequence of the DkALDH2b gene via chromosome walking, the promoter of DkALDH2b was amplified using the Universal GenomeWalker Kit (Clontech, San Jose, CA, USA). Primers used for gene isolation and promoter isolation are listed in Table S2.

2.7. Dual Luciferase Activity Assay

The tobacco dual-luciferase transient expression system utilized the pGreen II 62-SK and pGreen II 0800-LUC expression vectors [20,21]. A 1200 bp promoter sequence of the ALDH2b gene (Figure S1) was cloned upstream of the LUCIFERASE (LUC) reporter gene, while the full-length CDS of BPC1 was cloned into the pGreen II 62-SK vector. The constructed vectors, along with the empty vector control, were individually transformed into Agrobacterium tumefaciens competent cells GV3101 containing the pSoup plasmid [20]. Agrobacterium-mediated transient transformation was performed by infiltrating the abaxial side of tobacco leaves. Each combination was repeated three times. After infiltration, the leaves were marked and returned to the light incubator for continued cultivation. The dual-luciferase activity in the infiltrated leaves was measured 2–3 days later.

2.8. Subcellular Localization

The full-length CDS of DkBPC1 without the termination codon was fused with the YFP101 vector to construct the 35S::DkBPC1-YFP vector. This vector, along with the control vector 35S::YFP, was transformed into Agrobacterium competent cells for transient transformation in tobacco. The infiltrated tobacco plants were cultured under low light for two days. Tobacco leaves were harvested, mounted on slides, and protein localization was observed using laser scanning confocal microscopy (Nikon C2-ER, Tokyo, Japan).

2.9. Vector Construction and Plant Transformation

The full-length of the DkBPC1 gene was amplified by PCR. Firstly, it was cloned into the pDONR207 vector using the Gateway BP Clonase™ II Enzyme Mix (Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, it was recombined into the binary vector pK2GW7 using the LR Clonase™ II Enzyme Mix (Thermo Fisher Scientific, Waltham, MA, USA) to construct the pK2GW7-DkBPC1 overexpression vector. The VIGS-DkBPC1 interference vector was constructed by recombining the DkBPC1 gene with the pTRV2 vector using homologous recombination. Both vectors, along with their corresponding empty vectors, were individually transformed into Agrobacterium tumefaciens competent cells GV3101 for subsequent transient transformation of persimmon leaves and fruit discs.

Referring to the method of Mo et al. [22], the Agrobacterium-mediated transformation was used to infiltrate the bacterial suspension into live leaves of the ‘Eshi 1’ persimmon. Leaves were collected 10 days after infiltration, frozen in liquid nitrogen, and stored at −80 °C. For the transient transformation of persimmon fruit discs, discs with a diameter of 1 cm and a thickness of 0.5 cm were prepared using a puncher, placed on MS solid medium, and dark-cultured for 2 days. Subsequently, the fruit discs were frozen in liquid nitrogen and stored at −80 °C.

3. Results

3.1. Changes in Tannin Content During Fruit Development

The tannin content during fruit development in the three persimmon types was determined (Figure 1). It indicates that while the soluble tannin content exhibited a consistent declining trend across all types, the magnitude of decrease varied substantially (Figure 1a). The soluble tannin content in C-PCNA decreased at a significantly slower rate throughout the developmental period compared to J-PCNA, reaching relatively low levels around 25 WAB. J-PCNA showed a sharp decline between 2.5 and 5 WAB, reaching the de-astringency threshold by 10 WAB and maintaining this level until fruit maturity. In contrast, non-PCNA maintained high soluble tannin levels throughout development, failing to reach the de-astringency standard even at 27.5 WAB.

The insoluble tannin content patterns differed markedly among the three types (Figure 1b). In C-PCNA, the insoluble tannin content continuously decreased from 2.5 to 15 WAB but demonstrated a distinct pattern during later development compared to the other two types, showing two different levels of increase between 15 and 27.5 WAB. J-PCNA showed a gradual decline in insoluble tannin content during fruit development, maintaining low levels from 10 to 27.5 WAB. The trend in non-PCNA was generally similar to J-PCNA, though the decrease occurred more gradually.

3.2. Sequence Analysis of the DkBPC1 Gene

In our previous study, a transcription factor belonging to the BBR/BPC family, which interacts with DkALDH2, was identified through the yeast one-hybrid system and designated as DkBPC1. Based on the previously obtained DkBPC1 gene sequence information, its full-length sequence was amplified from the C-PCNA variety ‘Luotian Tianshi’. Sequence analysis revealed that the DkBPC1 gene has an ORF length of 633 bp, encoding 210 amino acids. The sequence exhibits high conservation, contains a BPC conserved domain, and belongs to the BBR/BPC transcription factor protein family, hereafter referred to as the BPC protein family.

It is known that BPC transcription factors in Arabidopsis are classified into three categories, each playing distinct roles in plants. Phylogenetic analysis of the persimmon DkBPC1 protein indicated that DkBPC1 clusters with Class I BPCs. Most BPC family members belong to either Class I or Class II, with very few members classified as Class III (Figure 2).

Furthermore, amino acid sequence alignment was performed between DkBPC1 and selected genes belonging to Type I: AtBPC1 (NP_973398.1), AtBPC2 (NP_001321304.1), AtBPC3 (NP_176979.3), PpBPC2 (XP_007222357.1), and VvBPC1 (XP_019076625.1) (Figure 3). The results showed that DkBPC1 shares the highest homology (79.14%) with the VvBPC1 protein from grape. The amino acid sequence homology with the other four transcription factors was 53.62%, 56.22%, 50.83%, and 72.33%, respectively.

3.3. Expression Patterns of DkBPC1 During Fruit Development

The expression patterns of the DkBPC1 gene during fruit development in three astringency types of persimmons are shown in Figure 4. Overall, the expression level of DkBPC1 in C-PCNA was higher than that in J-PCNA and non-PCNA. During the early fruit development stage (2.5–10 WAB), the expression of this gene in C-PCNA remained relatively high, approximately two-fold greater than that in J-PCNA and non-PCNA. At 10 WAB, the expression level in C-PCNA reached its relative peak, about four-fold higher than in J-PCNA and non-PCNA. As the fruit developed further toward natural de-astringency (10–27.5 WAB), the expression in C-PCNA gradually declined to levels comparable to the low expression observed in J-PCNA and non-PCNA. However, between 20 and 25 WAB, a modest increase in expression was observed in C-PCNA, a trend not seen in J-PCNA or non-PCNA.

3.4. DkBPC1 Is Located in the Nucleus

It is known that the fusion protein contains a Yellow Fluorescent Protein (EYFP) tag, which emits a yellow, fluorescent signal. Analysis of the subcellular localization of DkBPC1, as shown in Figure 5, revealed that under YFP excitation, distinct yellow fluorescence signals were observed in the nucleus in both the control and experimental groups. However, the fluorescence signal detected in the cell membrane in the control group was absent in the experimental group. The results indicate that the DkBPC1 protein is localized in the nucleus.

3.5. DkBPC1 Enhances the Activity of the DkALDH2b Promoter

To verify the effect of DkBPC1 on the transcriptional activity of the DkALDH2b promoter, a dual-luciferase assay was performed in transiently transformed tobacco leaves. The Agrobacterium cultures containing the Reporter (pGreen II-0800-LUC-DkALDH2b) and the Effector (pGreen II-62-SK-DkBPC1) were co-infiltrated into tobacco leaves. Fluorescence was observed, and the activities of both LUC and REN enzymes were measured separately. The LUC/REN ratio was used to assess the regulatory effect of the transcription factor on promoter activity.

The results showed that the fluorescence intensity driven by the DkALDH2b promoter was significantly higher in the presence of the transcription factor DkBPC1 compared to the control. Furthermore, when the transcription factor DkBPC1 was co-transformed with the DkALDH2b promoter, the relative luciferase activity (LUC/REN) was significantly stronger compared to the control group, indicating that DkBPC1 significantly enhances the activity of the DkALDH2b promoter (Figure 6).

3.6. Transient Expression of DkBPC1 in Persimmon Leaves In Vivo

To further elucidate the function of DkBPC1, transient expression of the DkBPC1 gene was performed in live leaves of the C-PCNA variety ‘Eshi 1’ using Agrobacterium-mediated transformation. Compared to the wild-type and empty vector controls, transient overexpression of DkBPC1 in persimmon leaves resulted in a significant increase in the expression levels of both DkBPC1 and DkALDH2b, which rose approximately 2.5-fold and 2.8-fold, respectively (Figure 7a,b). Tannin content analysis revealed that soluble tannin content in the wild-type and empty vector control leaves was about 100 mg/g, while insoluble tannin content was about 1 mg/g. In the experimental group, soluble tannin content increased to approximately 115 mg/g, and insoluble tannin content decreased to about 0.8 mg/g, with both differences reaching statistical significance (Figure 7c).

Compared to the wild-type and empty vector controls, transient knockdown of DkBPC1 expression in persimmon leaves resulted in an approximately 40% decrease in its expression level, a difference that was statistically significant (Figure 8a). In contrast, the expression level of DkALDH2b showed no significant change (Figure 8b). Tannin content analysis revealed that soluble tannin content in the wild-type and empty vector control leaves was approximately 100 mg/g, while insoluble tannin content was about 1.3 mg/g. In the experimental group, soluble tannin content significantly decreased to around 90 mg/g, and insoluble tannin content significantly increased to approximately 1.7 mg/g (Figure 8c).

3.7. Transient Overexpression of DkBPC1 in Persimmon Fruit Discs

To further investigate the effects of DkBPC1 on tannin content and the expression of related genes in persimmon fruit, transient overexpression was performed in fruit discs of the C-PCNA variety ‘Eshi 1’ at 16 WAB. Compared to the empty vector control, transient overexpression of DkBPC1 in persimmon fruit discs resulted in a significant increase in the expression levels of both DkBPC1 and DkALDH2b, by approximately 2-fold and 1.5-fold, respectively, although both genes exhibited relatively low basal expression levels (Figure 9a,b). Tannin content analysis revealed that, compared to the empty vector control, soluble tannin content in the experimental fruit discs increased significantly from about 14 mg/g to approximately 18 mg/g, while insoluble tannin content decreased significantly from about 1.2 mg/g to approximately 1.0 mg/g (Figure 9c).

4. Discussion

Persimmon tannins are the key determinants of fruit astringency. As important secondary metabolites, their biosynthesis is transcriptionally regulated by various transcription factors [23,24]; however, the precise regulatory mechanism remains incompletely understood. Several transcription factors, such as DkWRKY13, DkWRKY15 [25], and DkMYB14 [26], have been identified to participate in the natural de-astringency of C-PCNA persimmon. In addition, this study reveals that BPC1, a member of the BPC family, is a novel regulator involved in this process. The BPC protein family can be divided into Class I, Class II, and Class III based on functional differences [16]. DkBPC1 belongs to Class I (Figure 2). Studies have reported that in Arabidopsis, the Class I member BPC1 can upregulate the expression of LEAFY COTYLEDON2 (LEC2), INNER NO OUTER (INO), and SEEDSTICK (STK) genes [13,27,28] or interact with FIS-PRC2 to form complexes [29], thereby influencing reproductive development or seed development. The natural de-astringency trait of C-PCNA is primarily characterized by the conversion of soluble tannins to insoluble tannins during fruit development (Figure 1). Acetaldehyde, which originates from seeds [30], is a factor contributing to this conversion. Given the role of Class I BPC1 in seed development in Arabidopsis [29], a plausible hypothesis is that BPC1 affects de-astringency in C-PCNA persimmon by influencing the accumulation of acetaldehyde in seeds. Based on the known mechanism that BPC proteins bind GA-rich motifs to regulate genes [13], we analyzed the DkALDH2b promoter and identified five GC-rich regions as potential BPC1 binding sites (Figure S1), indicating a strong structural basis for this interaction. Based on our finding that DkBPC1 activates the promoter of the key acetaldehyde metabolic gene DkALDH2b (Figure 6), and in line with findings from Yang et al. [9] and Chen et al. [26], we establish that DkBPC1 influences natural de-astringency by integrating into the acetaldehyde-mediated regulatory network of persimmon tannins (Figure 10).

The natural de-astringency process in C-PCNA persimmon was found to be temporally correlated with DkBPC1 expression dynamics. The insoluble tannin content continuously decreased in the early developmental stages but increased to varying degrees later, which aligns with the “coagulation effect”. Concurrently, the expression level of DkBPC1 was significantly higher than in the other two persimmon types, particularly during the early stages. Furthermore, its expression showed a negative correlation with the insoluble tannin content in the late phase of natural de-astringency (Figure 4), indicating a potential promotive role of this gene in the de-astringency process of C-PCNA persimmon. Interestingly, our previous research demonstrated that the expression of DkALDH2b increases during this natural de-astringency process [31]. This apparent paradox suggests that while DkBPC1 interacts with DkALDH2b, it may also target other genes, such as additional members of the ALDH2 gene family, a possibility that warrants further investigation.

To bypass the current limitation of stable genetic transformation in persimmon, we transiently expressed DkBPC1 in leaves and fruit discs of C-PCNA persimmon to investigate its role in regulating tannin metabolism. After transient overexpression of DkBPC1 in C-PCNA leaves and fruit discs, the expression levels of both DkBPC1 and DkALDH2b increased significantly. Concurrently, the soluble tannin content increased significantly, while the insoluble tannin content decreased significantly (Figure 7 and Figure 9), indicating that DkBPC1 overexpression can reduce the conversion of soluble tannins to insoluble tannins. Following the transient knockdown of DkBPC1 in C-PCNA leaves, the expression level of DkBPC1 decreased significantly, and while the expression level of DkALDH2b also decreased, the change was not statistically significant. This lack of significance is potentially attributable to the inherently low expression level of DkALDH2b in persimmon leaves. However, a significant decrease in soluble tannin content and a significant increase in insoluble tannin content were observed in the leaves (Figure 8), indicating that DkBPC1 knockdown promotes the conversion of soluble tannins to insoluble tannins. It is important to note that the transient transformation assays conducted in leaves and fruit discs of C-PCNA persimmon may not fully recapitulate the role of DkBPC1 during the in vivo natural de-astringency process. Several confounding factors exist, including potential differences in tannin metabolism regulatory networks between leaves and fruits, and the altered gene expression in excised fruit discs due to environmental stresses. Consequently, the precise function of DkBPC1 in natural de-astringency warrants further validation once a stable genetic transformation system for persimmon is established.

5. Conclusions

Our study identifies DkBPC1 as a novel transcriptional regulator of natural de-astringency in C-PCNA persimmon, which functions by directly targeting the promoter of a key acetaldehyde metabolic gene, DkALDH2, to promote the conversion of soluble tannins to their insoluble form. These findings significantly advance our understanding of the de-astringency regulatory network. Looking forward, a comprehensive research pipeline—from elucidating the complete interactome via ChIP-seq and comparative genomics to applying gene editing for precision breeding—will be essential for translating these discoveries into tangible genetic gains for C-PCNA persimmon.