Introduction

Pear (Pyrus L.) is one of the most common and popular fruits in the world, and red pears are favored by consumers for their beautiful appearance and abundant anthocyanins1. In plant tissues, anthocyanins are widely present; these molecules are important flavonoids with multiple physiological functions that aid pollination, seed dispersal, and resistance to adverse environmental conditions2. In addition, anthocyanins have significant antioxidant activity and potential benefits for human health, such as reducing the risk of cancer, inflammation, and coronary arteriosclerosis3,4,5.

Anthocyanin biosynthesis is coordinately regulated by structural genes and TFs in many species. The structural genes consist of seven enzyme genes, from phenylalanine ammonia-lyase (PAL) to UDP-glucose: flavonoid 3-O-glucosyltransferase (UFGT). In addition, synthetic anthocyanins are transported and stored in vacuoles. There are three mechanisms for anthocyanin transport: glutathione S-transferase (GST)-mediated transport, membrane transport, and vesicle trafficking6,7. GST can mobilize anthocyanins by acting as a carrier of these molecules, and anthocyanins are transported from the cytoplasm to the tonoplast through the ABC transmembrane transporter8,9. Many secondary transporters and channels, such as malate transporter and MATE-type transports and ABC transporters, were reported to exist on the tonoplast, and the activities of these transporters are directly or indirectly dependent on the proton gradient generated by the pyrophosphate-energized vacuolar membrane proton pump9–11.

By forming the MBW transcriptional complex, the MYB, bHLH, and WD40 proteins can regulate structural genes12,13,14. MYBs, as central factors regulating anthocyanin biosynthesis, have been widely researched. In horticultural crops, apple MdMYB10, MdMYB1, and MdMYB110a; strawberry FaMYB10; bayberry MrMYB1; and pear PyMYB10, and PyMYB114 were reported to regulate anthocyanin biosynthesis in succession by forming the MBW transcriptional complex15,16,17,18. However, in different species, the regulatory mechanism of the MBW complex is different. In Arabidopsis and Myrica rubra, the three proteins in the MBW complex can interact with each other to form a transcriptional complex19,20. In apple, MdMYB10 interacts with MdbHLH3 and MdbHLH33 to enhance the synthesis of anthocyanins15. In pears, PyMYB114 or PyMYB10 interacts with PybHLH3 to significantly enhance anthocyanin biosynthesis18. Recently, overexpression of the SlMYB75 gene was shown to promote anthocyanin biosynthesis in tomato21.

In addition to the MBW complex, other TFs, such as NACs, ERFs, HY5, BBX22, and WRKY, were found to be involved in regulating anthocyanin biosynthesis22,23,24. These TFs could regulate anthocyanin biosynthesis by indirectly or directly binding to the MBW complex. The WRKY TFs belong to one large gene family that regulates a series of physiological processes, including development and senescence, and resistance to adverse environments24. WRKY TFs were identified by one or two conserved WRKY domain(s) usually followed by a zinc-finger motif. The WRKY TF contains a specific nucleic acid sequence (C/T)TGAC(T/C) named the W-box, which regulates the defense response to various stresses by self-regulation, and it could recognize and bind to the W-box or other promoters of the WRKY TFs to generate biological effects by achieving crosstalk of different WRKY25. Compared with the MYB TFs, WRKY is an emerging player in the plant signaling regulation network. The interaction between the upstream regulator of the WRKY TF and the downstream target gene constitutes a complex regulatory network25,26.

Recently, several reports have shown that WRKY proteins have an obvious correlation with the regulation of anthocyanin biosynthesis. For example, GbWRKY1 in Gossypium barbadense was proven to have a positive correlation with anthocyanin accumulation when expressed in Arabidopsis thaliana27. In a previous study, AtWRKY75 responded to low phosphate (Pi) stress by decreasing anthocyanin accumulation in A. thaliana seedlings28. Moreover, the Atwrky41 mutation resulted in increased anthocyanin content in A. thaliana rosette leaves29. AtWRKY6 promotes PR1 promoter activity, which is related to senescence and pathogen defense, and the plant responds to abiotic and biotic stresses by decreasing anthocyanin accumlation30. Recently, it was reported that the WRKY TF PhPH3 in petunia correlates with changes in the color of petals by playing a role downstream of the MBW complex31. Amato et al. also proved that VvWRKY26, a homologous gene of PhPH3 in Vitis vinifera, induces the accumulation of flavonoids23. In apple, MdWRKY40 is a key modulator in wounding-induced anthocyanin biosynthesis32. Yang et al. reported that the WRKY family was related to anthocyanin biosynthesis in red-skinned pear33. However, whether WRKYs are involved in anthocyanin biosynthesis by interacting with the TFs PyMYB114 and PybHLH3 in red pears is still unclear.

In this study, according to the transcriptome data of ‘Starkrimson’ and its green mutant pear at three fruit development stages, we screened two WRKY TFs and performed bioinformatics analysis. Furthermore, PyWRKY26 or PyWRKY31 with PyMYB114 and PybHLH3 were cotransformed into tobacco leaves and strawberry receptacles by a transient expression system. In addition, RT-qPCR analysis and dual-luciferase reporter system assays revealed the regulatory pattern of cotransformed PyWRKY26 with its partners to activate the activity of anthocyanin biosynthesis- and transport-related structural genes. Furthermore, firefly luciferase complementation and Y2H assays confirmed the interaction of PyWRKY26 with PybHLH3. Our research reveals a potential mechanism of regulating anthocyanin biosynthesis in red-skinned pears, which will help elucidate the regulatory network to clarify anthocyanin accumulation in other species.

Results

Screening of the candidate WRKY genes by transcriptome data and bioinformatics analysis

A previous search by Yang et al. reported that the WRKY family was related to anthocyanin biosynthesis by RT-qPCR analysis in red-skinned pear33. To identify the function of WRKY genes in controlling anthocyanin biosynthesis in red pears, we screened 66 differentially expressed genes (DEGs) of the WRKYs and analyzed the transcriptome data of ‘Starkrimson’ pears and its green mutant at 40, 55, and 85 day after full bloom (DAFB). By analyzing the transcript abundance of 66 DEGs via a heat map, we found that Pbr000122.1, Pbr032698.1, Pbr013092.1, and Pbr026903.1 were more highly expressed in the three developmental stages of ‘Starkrimson’ fruits than in the green-skinned fruits (Fig. 1a). Furthermore, a phylogenetic tree was constructed using the neighbor-joining method and bootstrap analysis (1000 replicates) and MEGA7 software. The results indicated that the gene Pbr013092.1 (named PyWRKY26) has the most similar predicted protein sequences with homologous genes among all anthocyanin-related genes by a detailed phylogenetic analysis, which included the TFs AtWRKY4, FvWRKY44, NtWRKY2, PhPH3, PtWRKY1, and Pbr000122.1 (named PyWRKY31), which have high homology with AtWRKY6 (Fig. 1b). In this phylogenetic analysis, other PyWRKY sequences are part of different clusters and groups, further indicating the gene function of PyWRKY26 and PyWRKY31. Furthermore, analysis of the derived polypeptide alignment of PyWRKY26, PyWRKY31 and other genes involved in anthocyanin biosynthesis, such as WRKY TFs, revealed the presence of WRKY motifs in all these genes (Fig. 1c).

Fig. 1: Screening and analyzing the candidate WRKY genes from transcriptome data of red-skinned pear ‘Starkrimson’ and its green mutant by bioinformatics.
figure 1

a WRKY gene transcript abundance in ‘Starkrimson’ (Pyrus communis L.) and its green mutant at different developmental stages. R1, R2, and R3 indicate the 40, 55, and 85 DAFB samples of the red-skinned pear of ‘Starkrimson’; G1, G2, and G3 indicate the 40, 55, and 85 DAFB samples of the green mutant of ‘Starkrimson’. b WRKY genes of different species are analyzed in the phylogenetic tree. The analysis of the alignments of the C-terminal WRKY domain was performed using the neighbor-joining method by the MEGA7 Program. c Multiple sequence alignment of AtWRKY4 (At1g13960.1), PhPh3 (AMR43368), FvWRKY44 (XM_004302784.1), NtWRKY2 (AB063576), PtWRKY (XM_002326290.1), PyWRKY26 (Pbr013092.1), PyWRKY31 (Pbr000122.1) and AtWRKY6 (At3g58710). Identical, conserved and similar residues are shown in black, red and blue. WRKY-DNA domains are shown in squares with β2, β3, and β4. d Chromosomal localization of the WRKY genes in pear

Full size image

Moreover, the members of the WRKY gene family were widely distributed on seventeen chromosomes in pear (Fig. 1d). As shown in Fig. 1d, most WRKYs were closely located to each other. This finding might indicate that tandem duplication events occurred in the WRKY gene family. The genes PyWRKY26 (Pbr13092.1) and PyWRKY31 (Pbr000122.1) located at Chr 3 and Chr 5 were explored in this study.

Evaluating the correlation of PyWRKY26/PyWRKY31 with anthocyanin accumulation and other factors regulating anthocyanin biosynthesis in pears

To confirm the correlation of PyWRKY26/PyWRKY31 with anthocyanin and anthocyanin-related TFs, we determined the anthocyanin contents in red-skinned ‘Starkrimson’ and green-skinned ‘Jinzheng No. 1’ pears at different developmental stages. The results showed that the anthocyanin contents increased with fruit development in the ‘Starkrimson’ pear and were higher than those of the ‘Jinzheng No. I’ pear (Fig. 2a). Furthermore, the expression levels of PyDFR, PyANS, PyUFGT, PyMYB10, PyMYB114, PybHLH3, PyWRKY26, PyWRKY31, and PyGST were largely higher in the ‘Starkrimson’ pear than in the ‘Jinzheng No. 1’ pear (except PyABC transporter, PyAVP1 and PyAVP2) (Fig. 2b, a–h). There was an obvious positive correlation between the anthocyanin contents and the anthocyanin biosynthesis structural genes and TFs, such as PyUFGT, PyGST, PyWRKY26, PyWRKY31, PyMYB114, and PybHLH3, and a negative correlation between the anthocyanin contents and PyABC transporter, PyAVP1 and PyAVP2 (Fig. S1).

Fig. 2: Evaluating the correlation of the candidate WRKY genes with anthocyanin biosynthesis and transport-related genes in ‘JinZheng No. 1’ and ‘Starkrimson’ pear cultivars.
figure 2

a The appearance of the ‘JinZheng No. 1’ and ‘Starkrimson’ pears at 30, 60, and 90 DAFB. b Determination of the anthocyanin contents in the ‘Starkrimson’ and ‘JinZheng No. 1’ pears at 30, 60, and 90 DAFB. c The expression level was analyzed by RT-qPCR for anthocyanin metabolic structural and regulatory genes. Significant differences are indicated at the level of P < 0.05 with lowercase letters. Highly significant differences are shown with uppercase letters (P < 0.01). Error bars are shown with the three biological replicates

Full size image

Heterologous expression of PyWRKY26/PyWRKY31 and other TFs induces anthocyanin accumulation in tobacco leaves

To verify the function of PyWRKY31 and PyWRKY26 in anthocyanin synthesis, we transiently transformed these genes into tobacco leaves. As shown in Fig. 3a, the empty vector pSAK277 was injected as a negative control, pigmentation was not observed when PyWRKY31 or PyWRKY26 alone was transformed with PyMYB114, and little pigmentation could be observed after cotransforming PyMYB114 or PyMYB10 with PybHLH3. The pigmentation was largely enhanced when PyWRKY26 or PyWRKY31 was cotransformed with PyMYB114, PyMYB10, and PybHLH3. Moreover, the anthocyanin contents in the tobacco leaves were analyzed by a colorimeter, and the changes in the L*, a*, and b* values were consistent with our expectations (Fig. 3b). When PyWRKY26/PyWRKY31 was involved in cotransformation, the anthocyanin content in tobacco was significantly higher than that when only PyMYB114, PyMYB10, and PybHLH3 were cotransformed (Fig. 3c) (P < 0.01). The above results indicated that cotransformation of PyWRKY26/PyWRKY31 with PyMYB10, PyMYB114, and PybHLH3 could significantly promote anthocyanin synthesis.

Fig. 3: Cotransformation of PyWRKY31/PyWRKY26 with its partners induced anthocyanin biosynthesis, as shown by transient expression assays in tobacco leaves.
figure 3

a The appearance of tobacco leaves 5 d after injection: a, pSAK277; b, PyMYB10; c, PyMYB114; d, PybHLH3; e, PyWRKY31; f, PyWRKY26; g, PyMYB10 + PybHLH3; h, PyMYB114 + PybHLH3; i, PyMYB114 + PyWRKY31; j, PyMYB114 + PyWRKY26; k, PyMYB10 + PyMYB114 + PybHLH3; l, PyMYB114 + PybHLH3 + PyWRKY31; m, PyMYB114 + PybHLH3 + PyWRKY26; n, PyMYB10 + PyMYB114 + PybHLH3 + PyWRKY31; o, PyMYB10 + PyMYB114 + PybHLH3 + PyWRKY26. b Differences in color are shown with the values of L*, a*, and b*. Significant differences are indicated at the level of P < 0.05 with lowercase letters. Highly significant differences are shown at the level of P < 0.01 with uppercase letters. Error bars are shown for six biological replicates. c Determination of the total anthocyanin contents in tobacco leaves with induced coloration

Full size image

Overexpression of PyWRKY26/PyWRKY31 with other related TFs results in anthocyanin accumulation in strawberry receptacles

To further identify the function of PyWRKY26 and PyWRKY31 in anthocyanin synthesis, we performed transient transformation of strawberry receptacles. As shown in Fig. 4a, no pigmentation was observed in the empty vector pSAK277-transformed receptacles. The pattern of change was similar to that of tobacco leaves; PyMYB114 + PyWRKY31 or PyMYB114 + PyWRKY26 were cotransformed, and no pigmentation was observed. Some pigmentation could be observed when PyMYB10 + PybHLH3 or PyMYB114 + PybHLH3 were cotransformed. A deeper color change could be observed when PyMYB114, PyMYB10, and PybHLH3 were coinjected. Meanwhile, the accumulation of anthocyanin was significantly higher when three TFs, PyMYB114, PyMYB10, and PybHLH3, were coinjected with PyWRKY26/PyWRKY31. In addition, the changes in the L*, a*, and b* values were influenced by the color change of the strawberry receptacles (Fig. 4b). When PyWRKY26 was coinjected with PyMYB114, PyMYB10, and PybHLH3, the total anthocyanin content in strawberry was significantly higher than that with only the cotransformation of PyMYB114, PyMYB10, and PybHLH3 (Fig. 4c) (P < 0.01). Meanwhile, when PyWRKY31 was coinjected with PyMYB114, PyMYB10, and PybHLH3, the similar results for the anthocyanin contents were obtained. Overall, cotransformation of PyWRKY26/PyWRKY31 with PyMYB10, PyMYB114, and PybHLH3 can significantly enhance the anthocyanin biosynthesis in strawberry receptacles.

Fig. 4: The functional analysis of PyWRKY31/PyWRKY26 cotransformed with its partners, which resulted in anthocyanin biosynthesis, as shown by transient expression assays in strawberry receptacles.
figure 4

a The appearance of the strawberry receptacles 5 d after infiltration: a, pSAK277; b, PyMYB10; c, PyMYB114; d, PybHLH3; e, PyWRKY31; f, PyWRKY26; g, PyMYB10 + PybHLH3; h, PyMYB114 + PybHLH3; i, PyMYB114 + PyWRKY31; j, PyMYB114 + PyWRKY26; k, PyMYB10 + PyMYB114 + PybHLH3; l, PyMYB114 + PybHLH3 + PyWRKY31; m, PyMYB114 + PybHLH3 + PyWRKY26; n, PyMYB10 + PyMYB114 + PybHLH3 + PyWRKY31; o, PyMYB10 + PyMYB114 + PybHLH3 + PyWRKY26. b Differences in color are shown with the values of L*, a* and b*. Significant differences are indicated at the level of P < 0.05 with lowercase letters, and highly significant differences are shown at the level of P < 0.01 with uppercase letters. Error bars are based on six biological replicates. c Determination of the total anthocyanin contents. Error bars are based on three biological replicates. d–i RT-qPCR analysis of the expression levels of the genes FvDFR, FvANS FvUFGT, FvGST, FvABC transporter and FvAVP. Significant differences are indicated at the level of P < 0.05 with lowercase letters. Highly significant differences are indicated with uppercase letters at P < 0.01. Error bars show the SEs of the means (n = 3)

Full size image

RT-qPCR analysis of the expression levels of anthocyanin biosynthesis- and vacuolar transport-related genes in strawberry receptacles

To investigate the mechanism of PyWRKY26/PyWRKY31 in regulating anthocyanin metabolism, we evaluated six structural genes of anthocyanin biosynthesis and transport in strawberry receptacles by RT-qPCR analysis. FvDFR, FvANS, and FvUFGT have been previously reported to control anthocyanin biosynthesis. The other three genes, FvGST, FvABC transporter, and FvAVPs, were identified as key genes involved in anthocyanin transport. As shown in Fig. 4d–f, cotransformation of PyWRKY26/PyWRKY31 with PyMYB10, PyMYB114, and PybHLH3 significantly enhanced the expression levels of the FvDFR, FvANS, FvUFGT, and FvGST genes. Meanwhile, PyWRKY26 had a stronger upregulation than PyWRKY31. Compared with pSAK277 alone, the FvAVP and the FvABC transporter genes were significantly repressed when four TFs were cotransformed into the strawberry receptacles (Fig. 4h–i). RT-qPCR analysis suggested that cotransformation of four TFs, PyMYB10, PyMYB114, PybHLH3, and PyWRKY26/PyWRKY31, enhanced the anthocyanin synthesis and transport by upregulating the expression of the FvDFR, FvANS, FvUFGT, and FvGST genes and downregulating the expression of the FvAVPs and FvABC transporter genes in the strawberry receptacle.

Validation of the interaction of the transcriptional regulatory complex by a dual-luciferase reporter system

A dual-luciferase reporter assay was used to verify the interaction of our candidate TFs with the structural genes, including PyDFR, PyANS, PyUFGT, PyGST, PyABC transporter, and PyAVP1/2 in Nicotiana tabacum. The results showed that PyMYB10 and PyMYB114 cotransformed with PybHLH3 could induce the activity of these promoters, and the additional TF PyWRKY26/PyWRKY31 could significantly enhance the transactivation activity of the PyDFR, PyANS, PyUFGT, and PyGST promoters (Fig. 5a–d). Moreover, compared with that of pSAK277, the transactivation of the PyABC transporter and PyAVP1/2 promoters was not obviously activated when PyMYB114, PyMYB10, PybHLH3, and PyWRKY26/PyWRKY31 were cotransformed (Fig. 5e–g). These results indicated that PyWRKY26/PyWRKY31 coregulates anthocyanin accumulation through coordination with the PyMYB10 + PyMYB114 + PybHLH3 complex.

Fig. 5: The transactivate activity of TFs PyMYB10, PyMYB114, PybHLH3, and PyWRKYs for anthocyanin biosynthesis and transport related genes was verified by a dual-luciferase reporter assay.
figure 5

The dual-luciferase reporter assay verified that PyMYB10, PyMYB114, PybHLH3 and PyWRKY26/PyWRKY31 cotransformation induces the activity of the PyDFR (a), PyANS (b), PyUFGT (c), PyGST (d), PyABC transporter (e) and PyAVP1/2 (f, g) promoters. The LUC/REN value was used to express promoter activity. Error bars represent the SEs for three replicate reactions

Full size image

Activation activity of PyWRKY26 and PybHLH3 on the PyMYB114 promoter was verified

To further explore the relationship of the TFs PyMYB114, PybHLH3, and PyWRKY26/PyWRKY31, we cloned the upstream 2 kb promoter of PyMYB114 and analyzed it by a dual-luciferase reporter assay in tobacco leaves. As shown in Fig. 6a, cotransformation with PyWRKY26 and PybHLH3 had a stronger activation effect on the promoter sequences of PyMYB114 than other factors, including PyWRKY31. The binding of PybHLH3 and PyWRKY26 to the PyMYB114 promoter was further verified by yeast one-hybrid technology. Promoter structure analysis revealed that multiple cis-regulatory elements were predicted by PlantCARE (Fig. 6b). The promoter segment baits were fused to the prey vectors pGADT7-PybHLH3 and pGADT7-PyWRKY26 and introduced into the Y1HGold yeast strain, and the results suggested that PyWRKY26 could bind to the S2 fragments. However, we did not detect a direct association between PybHLH3 and the promoter of PyMYB114, although several binding sites were located in different individual promoter regions (Fig. 6c).

Fig. 6: Verifying the regulatory pattern of TFs PyMYB114, PybHLH3, and PyWRKY26 in vivo.
figure 6

a The dual-luciferase reporter assay verifying PybHLH3 and PyWRKY26 cotransactivation activity. b A schematic of the PyMYB114 promoter. The bar segments were analyzed by using the PLACE and PlantCare databases and are divided by solid lines (S1–S3). c Yeast one-hybrid assays between PybHLH3/PyWRKY26 and the PyMYB114 promoter. Positive colonies indicated strong specific interactions with the PyMYB114 promoter segments. The colony name indicates the bait segment, and prey vectors were used. d Model of the NLuc, CLuc and NLuc/CLuc constructs. e Firefly luciferase complementation assays in young tobacco leaves. Error bars are shown with six biological replicates. **P < 0.01. f Verification of the interaction of PyWRKY26 and PybHLH3 in vivo

Full size image

Verification of the interaction of PybHLH3 with PyWRKY26

For further identification of the interaction of PybHLH3 with PyWRKY26, a tobacco-based firefly luciferase complementation assay was performed. PybHLH3 was inserted into the N-terminal region of the firefly luciferase (NLuc), whereas PyWRKY26 was linked to the C-terminal region of the firefly luciferase (CLuc) (Fig. 6d). Coexpression of the NLuc-PybHLH3 and CLuc-PyWRKY26 constructs showed the strongest capability to rescue intense luciferase enzyme activity (P < 0.01). Nevertheless, for the control constructs, including NLuc-PybHLH3 with C and CLuc-PyWRKY26 with Nluc, no obvious luciferase enzyme activity was observed (Fig. 6e).

Then, the interaction of PyWRKY26 with PybHLH3 was further verified by Y2H assays. The full-length coding sequences of PyWRKY26 and PybHLH3 were inserted separately into pGBKT7 and pGADT7. The results indicated that cotransformed PyWRKY26 and PybHLH3 in AH109 yeast cells resulted in healthy growth on the medium (Fig. 6f). Above all, the results suggested that PyWRKY26 interacts with PybHLH3 in vivo.

Discussion

Anthocyanin, a common secondary metabolite in plants that exhibits antioxidant properties, gives plants a red or purple color. At present, an extensive body of literature has clearly demonstrated that the anthocyanin biosynthetic pathway is regulated at the transcriptional level by an MYB-bHLH-WD40 (MBW) regulatory complex in plants, and MYB in particular plays a vital role in anthocyanin synthesis. Compared with MYB, the bHLH TF has a significantly lower functional specificity, similar to AtEGL3, AtGL3, and AtTT8, and the three bHLH TFs have functional redundancy in the regulation of anthocyanin biosynthesis and trichome development15,16,17. In addition, WD40 plays an important role in enhancing the stability of the MBW complex, and the interaction between MYB and bHLH TFs is a prerequisite for specific recognition of DNA sequences19,34. In peach, the expression of PpMYB10.1 was activated with partner TF helix-loop-helix proteins and PpNAC1 and coregulated anthocyanin biosynthesis35. In pear, Yao et al. showed that PyMYB114/PyMYB10, which interacts with PybHLH3 in tobacco leaves, strawberry receptacles and pears, could promote the biosynthesis of anthocyanins. Moreover, PyERF3 promoted anthocyanin biosynthesis by coexpression with PyMYB114 and PybHLH318. In this study, we also demonstrated that PyWRKY26/PyWRKY31 could be coexpressed with PyMYB114 and PybHLH3, resulting in anthocyanin biosynthesis in tobacco leaves and strawberry receptacles (Figs 3, 4). Thus, the regulatory networks of color formation in red-skinned pears are highly complex, and whether and how the interaction of PyWRKY26/PyWRKY31 with PyERF3 regulates anthocyanin biosynthesis still need further study.

The important TF regulatory family WRKY plays important roles in various plant processes, such as plant growth and metabolism and senescence35,36,37. WRKYs were shown to exert their biological functions and regulate metabolite biosynthesis, especially that of secondary metabolites, through physical interactions with different proteins38. Verweij et al. reported that PH3, which encodes a WRKY protein of petunia, is a target gene of the AN11-AN1-PH4 complex, could bind to AN11 and is required together with the AN11-AN1-PH4 complex for the transcription of PH5 (which encodes an MYB protein), which regulates hair development, tannin accumulation, and mucilage production in Arabidopsis31. In our study, PyWRKY26 largely enhanced anthocyanin accumulation after transient cotransformation with PyMYB114 and PybHLH3 compared with the two cofactors PyMYB114 + PybHLH3 (Figs. 3 and 4). This evidence proved that WRKY needs to interact with other partner factors to carry out its biological functions. The regulatory complex was consistent with Verweij et al.31, although PyWRKY26 could regulate anthocyanin accumulation, while PH3 regulates tannin accumulation and mucilage production. Thus, TF WRKYs are involved in the formation of different secondary metabolites. In addition, we further explored the target genes of PyWRKY26 and PyWRKY31, and cotransformation with PyWRKY26 and PybHLH3 had a stronger activation effect on the promoter sequences of PyMYB114 than other factors. PyWRKY26 directly activated the promoter sequences of PyMYB114 (Fig. 6b, c), but PyWRKY31 could not bind to the PyMYB114 promoter (Fig. 6a). Further, the interaction of PyWRKY26 with PybHLH3 was proven by firefly luciferase complementation and Y2H assays (Fig. 6d–f). Therefore, PyWRKY26 and PyWRKY31 could regulate color formation, but the downstream target genes were different.

Generally, TFs forming regulatory complexes regulate anthocyanin accumulation by a series of anthocyanin biosynthesis- and transport-related structural genes. In this study, the expression levels of PyDFR, PyANS, PyUFGT, and PyGST in red-skinned pear were generally higher than those in green-skinned pear, and the expression patterns of ABC transporters and AVPs were opposite (Fig. 2). Furthermore, cotransformation of PyWRKY26 with its partners in strawberry receptacles showed that the expression of FvDFR, FvANS, FvUFGT, and FvGST was upregulated and FvABC transporter and FvAVPs were downregulated (Fig. 4). The dual-luciferase reporting system also demonstrated a similar conclusion (Fig. 5). It has been widely reported that DFR, ANS, and UFGT catalyze anthocyanin biosynthesis. GST has been shown to be involved in anthocyanin transport in strawberry39. Recently, the ATP-binding cassette transporter AtABCC2 was reported to be involved in the vacuolar transport of anthocyanins and other flavonoids in Arabidopsis40. In a previous study, GST acted as a carrier of anthocyanins and mobilized anthocyanins from the cytoplasm to the tonoplast by the ABC transmembrane transporter8,9. Many secondary transporters and channels, such as malate transporter, MATE-type transporters and ABC transporters, exist on the tonoplast, and the activities of these transporters are directly or indirectly dependent on the proton gradient generated by V-ATPase and V-Ppase10,11,41,42,43. The H+-pumping activities of vacuolar H+-ATPase (VHA) and pyrophosphatase (VHP) have an extremely important role in the transport of anthocyanins, malate and other metabolites44. These reports were consistent with the findings of our study showing involvement in the vacuolar transport of anthocyanins, but previous reports did not present evidence of the relationship between transport-related genes and the transcription regulatory complex. Moreover, we confirmed that the regulatory mode of the PyABC transporter and PyAVPs is different from that of PyGST, and we speculate that there is a difference in the timing between the anthocyanin synthetic process and transport process in plant cells. When the synthetic process was significantly activated, which results in anthocyanin accumulation, PyGST participates in transport over time, but the PyAVP and PyABC transporter genes are not activated. Anthocyanin biosynthesis and anthocyanin transport may be independent processes, although they are regulated by the same transcription complex; furthermore, they are not expressed simultaneously. The molecular mechanism by which transporter genes participate in anthocyanin transport still requires further research.

Conclusion

In this study, the TFs PyWRKY31 and PyWRKY26 with their partners PyMYB10, PyMYB114, and PybHLH3 were cotransformed into tobacco leaves and strawberry receptacles and resulted in increased anthocyanin contents. Furthermore, we confirmed that this cotransformation activated the activity of PyDFR, PyANS, and PyUFGT in anthocyanin biosynthesis and PyGST in anthocyanin transport instead of the PyABC transporter and PyAVPs. Moreover, firefly luciferase reporter assays and yeast expression assays showed that the interaction of PyWRKY26 and PybHLH3 could cotarget the PyMYB114 promoter and that PyWRKY26 directly activates the promoter sequences of PyMYB114, which resulted in anthocyanin accumulation in red-skinned pear. This research provides novel insight into the regulatory network of anthocyanin accumulation and contributes to improving the appearance quality of red-skinned pears.

Materials and methods

Plant materials

The green-skinned pear ‘JinZheng No. 1’ and red-skinned pear ‘Starkrimson’ used in this study were collected from the orchard of the Institute of Pomology in Shandong Province during the 2018 growing season. Tobacco (N. tabacum) and diploid strawberry (Fragaria vesca, called ‘Yellow Wonder’ 5AF7) were used in this study for the transient transformation experiments, dual-luciferase reporter system assays and firefly luciferase complementation assays, and they were cultivated in an intelligent incubator with a 16 h photoperiod and a 21 °C/17 °C day/night temperature. Young tobacco leaves and strawberry receptacles ~2 weeks after flowering were infiltrated for the experiment, and observations at 4–6 d after injection were performed. The tobacco leaves and strawberry receptacles were collected and chopped, frozen with liquid nitrogen and stored at −80 °C.

Extraction and determination of the anthocyanins in tobacco leaves and strawberry

Anthocyanins were extracted according to the method of Yang et al.33. Approximately 0.2 g of the fruit skin of ‘JinZheng No. 1’ and ‘Starkrimson’, tobacco leaves or strawberry receptacles stored at −80 °C were ground to powder in liquid nitrogen and then homogenized with 1 ml of cold methanol containing 0.1% HCl at 4 °C for 24 h, and the homogenate was allowed to incubate in the dark at 4 °C for 24 h. Then, the supernatant was collected by centrifugation at 12,000 rpm for 20 min. The absorbance of the anthocyanins was measured at wavelengths of 530, 620, and 650 nm using a Multiskan Spectrum (Thermo Scientific Multiskan GO 1510, Finland). There were three biological replicates per sample, and the total content of anthocyanin per sample fresh weight was calculated according to the following formula: OD = (A530 − A620) − 0.1 × (A650 − A620).

Total RNA extraction and gene transcript abundance analysis by RT-qPCR

Fruit skin of ‘JinZheng No. 1’ and ‘Starkrimson’ and strawberry receptacles were ground into fine powder in liquid nitrogen. Total RNA of the powder samples was extracted in an ice bath. First strand cDNA synthesis was performed using Prime Script RT Master Mix (DRR036A, TaKaRa, China). RT-qPCR was conducted using SYBR® Premix Ex TaqTM II (DRR081A, TaKaRa, China) in a 10 μl volume. The amplification program was as follows: one cycle of 30 s at 95 °C, followed by 40 cycles of 15 s at 95 °C and 30 s at 60 °C; a strawberry housekeeping gene (gene11892) was used as an internal control. The relative expression level of the genes was calculated using the 2(-ΔΔCt) method. All analyses and error bars were determined using three biological replicates. The primer sequences for RT-qPCR are listed in Table S1.

Genes cloning and overexpression vector construction

The gene sequences of PyWRKY26 and PyWRKY31 cloned from the cDNA of the red-skinned pear cultivar ‘Starkrimson’ were used in the following experiment. PCR amplification was conducted using TransStart FastPfu DNA Polymerase (AP221-01, Transgen, China) and the primer sequences listed in Table S2. The PyWRKY26 and PyWRKY31 genes were inserted into the pSAK277 vector under the control of the 35 S promoter with EcoRI and XhoI45. The integrated constructs were transformed into the Agrobacterium tumefaciens strain GV3101 using the chemical method, and the cells were incubated at 28 °C for 2 d. A description of the infecting is in Yao et al.18. The specific method of infiltration experiments was described by Voinnet et al.46. Tobacco leaves and strawberry receptacles were collected for anthocyanin measurement and RNA extraction at 5–7 d after infiltration. Empty vector infiltrations (pSAK277) were used as negative controls.

Dual-luciferase reporter system assays

For the dual-luciferase reporter assay, the upstream promoter sequences of PyMYB114 (2.0 kb), PyAVP1/2 (1.8 kb), PyDFR (2.0 kb), PyANS (2.0 kb), PyUFGT (1.8 kb), PyABC transporter (2.0 kb), and PyGST (2.0 kb) were cloned and inserted into the pGreen II 0800-LUC vector with the primers listed in Table S2. The recombinant plasmids were transformed into the Agrobacterium strain GV3101 (PM90) with the pSoup helper plasmid. The TFs PyWRKY26 and PyWRKY31, PyMYB114, PyMYB10, and PybHLH3 were mixed with the promoter sequences (the ratio 1:9, v/v) and then injected into young tobacco leaves for transient cotransformation expression analysis18. According to the manufacturer’s instructions, the ratio of transactivation activities of firefly luciferase and renilla luciferase was tested by the Dual-Luciferase® Reporter Assay System (E1910, Promega, USA).

Firefly luciferase complementation assay

Firefly luciferase complementation assays were performed according to the method of Chen et al.47. Gene sequences of PybHLH3 (with no stop codon) were amplified and linked with pCAMBIA1300-NLuc, and the coding sequences of PyWRKY26 were cloned and linked with the pCAMBIA1300-CLuc vector. The primer sequences are listed in Table S2. Then, the cells were transformed into Agrobacterium GV3101 and cultured using a selection medium containing kanamycin. The cells were grown to OD600 0.6, and then, bHLH3-NLuc and PyWRKY26-CLuc were mixed 1:1 by volume and infiltrated into the tobacco leaves. Leaf disks (exactly 2 cm in diameter) were punched adjacent to the infiltration site, and the firefly luciferase activity was determined by a Steady-Glo® Luciferase Assay System (E2510, Promega, USA).

Yeast one-hybrid assay

To identify transcriptional regulators of PyMYB114 by yeast one-hybrid assays, we used the ~300 bp promoter segments, corresponding to the S1–S3 sequences. The promoter fragments were inserted into the pAbAi vector, and the PybHLH3 and PyWRKY26 genes were cloned into the pGADT7 vector. In a preliminary filter, self-activation of the bait vectors was tested on SD/-ura+AbA100, SD/-ura+AbA200, and SD/-ura+AbA400 plates; the prey vectors pGADT7-PybHLH3 and pGADT7-PyWRKY26 were tested on SD/-Leu plates. The promoter segment baits were fused to the prey vectors pGADT7-PybHLH3 and pGADT7-PyWRKY26 and introduced into the Y1HGold yeast strain and tested on SD/-ura+AbA plates at 30 °C for 3 d. The primer sequences used for vector construction are listed in Supporting Information Table S1.

Yeast two-hybrid assay

According to the Matchmaker® Gold Yeast Two-Hybrid System (Clontech, HTTP:// www.clontech.com/), a Y2H assay was performed to test for protein interactions. The PyWRKY26 and PybHLH3 genes were inserted separately into pGBKT7 and pGADT7 and then cotransformed into the yeast strain AH109. The transformants were selected on SD/-Leu/-Trp medium and tested on SD/-Leu/-Trp/-His/-Ade medium. Meanwhile, pGADT7-T and pGBKT7-Lam or pGADT7-T and pGBKT7-53 were cotransformed as negative and positive controls.

Statistical analysis

All samples were assessed at least three times independently, and all data are represented as the mean ± SD. Statistical analysis was performed by Student’s t-test and one-way ANOVA. Significance was indicated by asterisks * (P < 0.05) or ** (P < 0.01) or different letters.