The SMXL8-AGL9 module mediates crosstalk between strigolactone and gibberellin to regulate strigolactone-induced anthocyanin biosynthesis in apple

Abstract

Although the strigolactone (SL) signaling pathway and SL-mediated anthocyanin biosynthesis have been reported, the molecular association between SL signaling and anthocyanin biosynthesis remains unclear. In this study, we identified the SL signal transduction pathway associated with anthocyanin biosynthesis and the crosstalk between gibberellin (GA) and SL signaling in apple (Malus × domestica). ELONGATED HYPOCOTYL5 (HY5) acts as a key node integrating SL signaling and anthocyanin biosynthesis, and the SL-response factor AGAMOUS-LIKE MADS-BOX9 (AGL9) promotes anthocyanin biosynthesis by activating HY5 transcription. The SL signaling repressor SUPPRESSOR OF MAX2 1-LIKE8 (SMXL8) interacts with AGL9 to form a complex that inhibits anthocyanin biosynthesis by downregulating HY5 expression. Moreover, the E3 ubiquitin ligase PROTEOLYSIS1 (PRT1) mediates the ubiquitination-mediated degradation of SMXL8, which is a key part of the SL signal transduction pathway associated with anthocyanin biosynthesis. In addition, the GA signaling repressor REPRESSOR-of-ga1-3-LIKE2a (RGL2a) mediates the crosstalk between GA and SL by disrupting the SMXL8-AGL9 interaction that represses HY5 transcription. Taken together, our study reveals the regulatory mechanism of SL-mediated anthocyanin biosynthesis and uncovers the role of SL-GA crosstalk in regulating anthocyanin biosynthesis in apple.

Figure 1.

The role of MdHY5 in SL-induced anthocyanin biosynthesis in apple. A and B) Effect of GR24^5DS on anthocyanin biosynthesis in apple peel. Debagged fruits at 130 DAFB were treated with 5 μM GR24^5DS for 30 min. Apple fruits without GR24^5DS treatment were used as the control. The treated fruits were exposed to light for 6 d. Anthocyanin content in the peel was measured, and the value for the control was set to 1. Each treatment was performed in triplicate and each replicate contained 8 to 10 fruits. RAC, relative anthocyanin content. A representative photograph is shown here. Error bars denote standard deviations. Asterisks indicate significant differences based on the t-test. **P < 0.01. C) RT-qPCR analysis of MdHY5 expression in the peel of apple fruits treated with 5 μM GR24^5DS. Untreated fruits were used as the control. REL, relative expression level. The value for the control at 0 h was set to 1. Error bars denote standard deviations. Means and standard deviations from three biological replicates are shown (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. D to –F) Functional analysis of MdHY5 in anthocyanin biosynthesis through transient silencing in apple fruits. Anthocyanin content and gene expression around the infiltration sites indicated by the arrows were measured 8 d after infiltration. TRV, TRV1 + TRV2; MdHY5-TRV, TRV1 + MdHY5-TRV2. Anthocyanin content and gene expression level in the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01; *P < 0.05. G and H) Anthocyanin accumulation phenotype and anthocyanin content in wild-type (WT) and MdHY5-suppressed apple calli (asMdHY5) after GR24^5DS treatment. Fifteen-day-old apple calli were transferred to MS medium containing 5 μM GR24^5DS and exposed to light for 14 d. Apple calli without GR24^5DS treatment were used as the control. Anthocyanin content in wild-type calli grown on GR24^5DS medium at 0 d was set to 1. Each treatment was performed in triplicate, and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Each time point was analyzed separately. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 2.

Effect of MdAGL9 on the transcription of MdHY5. A) Identifying regulatory factor(s) that bind to the MdHY5 promoter using DNA affinity capture assays. Total proteins were extracted from GR24^5DS-treated apple peel. B) Schematic diagram of the MdHY5 promoter. The CArG motifs are indicated by blue boxes. W1–W3 represent DNA fragments used for the ChIP-PCR assay, and CArG-3 m is a mutated CArG motif that contains six nucleotide substitutions in the CArG-3 motif. DNA sequences of CArG-1, CArG-2, CArG-3, and the mutated CArG-3 are highlighted in red. C) EMSA showing the binding of MdAGL9 to the MdHY5 promoter. Unlabeled probes were used as competitors. The experiment was repeated three times with similar results. A representative photograph is shown here. D) ChIP-PCR assay showing the binding of MdAGL9 to the MdHY5 promoter. Chromatin was immunoprecipitated from MdAGL9-overexpressing apple calli, and WT was used as the control. The W1, W2, and W3 fragments were examined by qPCR. The value of each region for the control was set to 1. The experiment was repeated three times with similar results. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. E) Dual LUC assay showing the transcriptional regulation of the MdHY5 promoter by MdAGL9. The experiment was repeated three times with similar results and each replicate contained 3 to 5 Nicotiana benthamiana leaves. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. F) GUS staining and activity in apple calli showing transcriptional regulation of the MdHY5 promoter by MdAGL9. The MdAGL9-pXSN-GFP overexpression vector was transiently expressed in ProMdHY5-GUS transgenic calli. The value for the empty vector (EV) was set to 1. The experiment was repeated three times with similar results and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 3.

MdAGL9 promotes GR24^5DS-induced anthocyanin biosynthesis. A) RT-qPCR analysis of MdAGL9 expression after GR24^5DS treatment. Apple fruits were treated with 5 μM GR24^5DS for 4 h, and untreated fruits were used as the control. REL, relative expression level. The value for the control at 0 h was set to 1. Error bars denote standard deviations. Means and standard deviations from three biological replicates are shown (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. B) GUS staining and relative GUS activity of apple calli transformed with the MdAGL9 promoter:GUS fusion gene. Control, without GR24^5DS treatment; GR24^5DS, GR24^5DS treatment for 2 h. The GUS activity of the untreated apple calli was set to 1. The experiment was repeated three times with similar results, and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Asterisks indicate significant differences based on the t-test. **P < 0.01. C and D) Analysis of the regulatory role of MdAGL9 in anthocyanin biosynthesis using transient overexpression assays in apple fruits. Anthocyanin concentration and gene expression related to anthocyanin biosynthesis in the infiltrated sites indicated by arrows were measured 2 d after infiltration. A representative photograph is shown here. pIR, IL60-1 + IL60-2; MdAGL9-pIR, IL60-1 + MdAGL9-IL60-2. RAC, relative anthocyanin content. Anthocyanin content and gene expression level at the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. E and F) Analysis of the regulatory role of MdAGL9 in anthocyanin biosynthesis using transient suppression assays in apple fruits. Anthocyanin concentration and gene expression related to anthocyanin biosynthesis in the infiltrated sites indicated by arrows were measured 6 d after infiltration. TRV, TRV1 + TRV2; MdAGL9-TRV, TRV1 + MdAGL9-TRV2. Anthocyanin content and gene expression level in the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01; *P < 0.05. G and H) Analysis of the regulatory role of MdAGL9 in anthocyanin biosynthesis using stable transformation in apple calli. Apple calli were incubated on MS medium containing 5 μM GR24^5DS or without GR24^5DS for 5 d under constant light conditions. WT, wild-type; MdAGL9, MdAGL9 overexpression; asMdAGL9, suppression of MdAGL9. The value of WT grown on medium without GR24^5DS was set to 1. Each treatment was performed in triplicate and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Asterisks indicate significant differences based on the t-test. **P < 0.01. I) Analysis of the association between MdHY5 and MdAGL9-mediated anthocyanin biosynthesis using transient expression in apple fruits. pIR + TRV, IL60-1 + IL60-2 + TRV1 +TRV2; MdAGL9-pIR + MdHY5-TRV, IL60-1 + MdAGL9-IL60-2 + TRV1 + MdHY5-TRV2. Anthocyanin content of pIR was set to 1. Each treatment was performed in triplicate and each replicate contained 8 to 10 apple fruits. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. J and K) Analysis of the association between MdHY5 and MdAGL9-mediated anthocyanin biosynthesis using stable transformation in apple calli. Apple calli were exposed to light for 5 d. MdAGL9/asMdHY5, suppression of MdHY5 in the MdAGL9-overexpressing calli. Anthocyanin content of WT was set to 1. Each treatment was performed in triplicate and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 4.

MdAGL9 interacts with MdSMXL8. A) Schematic diagram of the constructs containing different domains of the MdAGL9 protein. MdAGL9, 1 to 239 aa; MdAGL9-N, 1 to 128 aa; MdAGL9-C, 129 to 239 aa. The orange boxes represent the DNA binding domain of MdAGL9. B) Y2H assay showing the interaction between MdAGL9 and MdSMXL8. The full-length or truncated coding sequence of MdAGL9 was cloned into the pGAD vector, and the full-length coding sequence of MdSMXL8 was cloned into the pGBD vector. −T/−L, SD/−Trp/−Leu; −T/−L/−H/−A, Sd −Trp/−Leu/−His/−Ade. C) Pull-down assay showing the interaction between MdAGL9 and MdSMXL8. D) Confirmation of the interaction between MdAGL9 and MdSMXL8 in nuclei using a BiFC assay in onion epidermal cells. Scale bars = 10 μm. E) Co-IP assay. The MdAGL9-GFP protein was immunoprecipitated from MdAGL9-GFP and MdAGL9-GFP/MdSMXL8-GUS transgenic apple calli with anti-GFP antibody, and the eluent was detected with anti-GUS antibody.

Figure 5.

MdSMXL8 negatively regulates MdAGL9-induced MdHY5 transcription. A and B) Analysis of the regulatory role of MdSMXL8 in anthocyanin biosynthesis using transient overexpression assays in apple fruits. MdSMXL8-overexpressing fruits were exposed to light for 6 d. pIR, IL60-1 + IL60-2; MdSMXL8-pIR, IL60-1 + MdSMXL8-IL60-2. RAC, relative anthocyanin content; REL, relative expression level. Anthocyanin content and gene expression level in the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01; *P < 0.05. C and D) Analysis of the regulatory role of MdSMXL8 in anthocyanin biosynthesis using transient suppression assays in apple fruits. MdSMXL8-suppressed fruits were exposed to light for 2 d. TRV, TRV1 + TRV2; MdSMXL8-TRV, TRV1 + MdSMXL8-TRV2. Anthocyanin content and gene expression level in the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. E and F) Analysis of the regulatory role of MdSMXL8 in anthocyanin biosynthesis using stable transformation assays in apple calli. The calli were transferred to MS medium containing 5 μM GR24^5DS or without GR24^5DS under constant light conditions for 5 d. WT, wild-type; MdSMXL8, overexpression of MdSMXL8; asMdSMXL8, suppression of MdSMXL8. The value of WT grown on medium without GR24^5DS was set to 1. Each treatment was performed in triplicate and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Asterisks indicate significant differences based on the t-test. *P < 0.05. G) Evaluation of the relationship between MdAGL9 and MdSMXL8 in regulating anthocyanin biosynthesis via transient expression assays in apple fruits. pIR, IL60-1 + IL60-2; MdAGL9-pIR, IL60-1 + MdAGL9-IL60-2; MdAGL9-pIR + MdSMXL8-pIR, IL60-1 + MdAGL9-IL60-2 + IL60-1 + MdSMXL8-IL60-2. Anthocyanin content of pIR was set to 1. Each treatment was performed in triplicate and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. H and I) Evaluation of the relationship between MdAGL9 and MdSMXL8 in regulating anthocyanin biosynthesis via stable transformation assays in apple calli. Apple calli were exposed to light for 5 d. WT, wild-type; MdAGL9, overexpression of MdAGL9; MdAGL9/MdSMXL8, overexpression of MdSMXL8 on the basis of MdAGL9-overexpressing calli; MdAGL9/asMdSMXL8, suppression of MdSMXL8 based on MdAGL9-overexpressing calli. The anthocyanin content of WT was set to 1. Each treatment was performed in triplicate and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. J) EMSA showing the effect of MdSMXL8 on the binding of MdAGL9 to the MdHY5 promoter. The experiment was repeated three times with similar results. A representative photograph is shown here. K) Dual LUC assay showing the effect of MdSMXL8 on MdAGL9-activated transcription of MdHY5. The experiment was repeated three times with similar results, and each replicate contained 3 to 5 Nicotiana benthamiana leaves. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. L) Detection of GUS staining and activity in apple calli showing the effect of MdSMXL8 on MdAGL9-activated transcription of MdHY5. MdSMXL8-pXSN-MYC and MdAGL9-pXSN-GFP overexpression vectors were transiently expressed in ProMdHY5-GUS transgenic calli. The value for the empty vector (EV) was set to 1. The experiment was repeated three times with similar results, and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 6.

MdPRT1 promotes ubiquitination-mediated degradation of MdSMXL8. A) GUS staining showing the effects of GR24^5DS and MG132 on MdSMXL8 protein stability. 1 g apple calli were added to each 2 mL centrifuge tube, and 1 mL GUS staining solution was added for staining. Control, without GR24^5DS treatment; GR24^5DS, 5 μM GR24^5DS treatment for 2 h; GR24^5DS + MG132, 5 μM GR24^5DS and 100 μM MG132 treatments for 2 h. B) Y2H assay showing the interaction between MdPRT1 and MdSMXL8. The full-length coding sequences of MdPRT1 and MdSMXL8 were cloned into the pGAD or pGBD vectors, respectively. −T/−L, SD/−Trp/−Leu; −T/−L/−H/−A, Sd −Trp/−Leu/−His/−Ade. C) Pull-down assay showing the interaction between MdPRT1 and MdSMXL8. D) Confirmation of the interaction between MdAGL9 and MdSMXL8 in nuclei using a BiFC assay in onion epidermal cells. Scale bars = 10 μm. E) Co-IP assay. The MdPRT1-GFP protein was immunoprecipitated from MdPRT1-GFP and MdPRT1-GFP/MdSMXL8-GUS transgenic apple calli with anti-GFP antibody, and the eluent was detected with anti-GUS antibody. F) Ubiquitination assay in vitro. MdPRT1-GST was tested for E3 ubiquitin ligase activity in the presence and absence of ATP, ubiquitin, E1, E2, MdPRT1-GST, and MdSMXL8-HIS. G) Ubiquitination analysis in vivo. MdSMXL8-GUS was immunoprecipitated from MdSMXL8-GUS and MdSMXL8-GUS/MdPRT1 transgenic apple calli with anti-GUS antibody, and the protein sample was detected with anti-GUS and anti-Ubi antibodies. H) Protein-degradation assay in vitro. Total proteins extracted from wild-type and transgenic apple calli with overexpression or repressed expression of MdPRT1 were incubated with the purified MdSMXL8-HIS fusion protein. I) GUS staining and protein abundance determination of MdSMXL8-GUS, MdSMXL8-GUS/MdPRT1, and MdSMXL8-GUS/asMdPRT1 transgenic apple calli. 1 g apple calli were added to each 2 mL centrifuge tube, and 1 mL GUS staining solution was added for staining. J) Effect of GR24^5DS on ubiquitination level of MdSMXL8. MdSMXL8-GUS was immunoprecipitated using anti-GUS antibody from MdSMXL8-GUS and MdSMXL8-GUS/asMdPRT1 transgenic apple calli with or without 5 μM GR24^5DS treatment for 2 h. Immunoblotting using anti-Ubi antibody is shown on the left, and that using anti-GUS antibody is shown on the right. Control, without GR24^5DS treatment; GR24^5DS, with 5 μM GR24^5DS treatment. K) Effect of GR24^5DS on the stability of MdSMXL8 protein. MdSMXL8-GUS and MdSMXL8-GUS/asMdPRT1 transgenic apple calli were treated with or without 5 μM GR24^5DS treatment for 2 h. MdSMXL8-GUS protein abundance was assessed by immunoblotting using anti-GUS antibody. Control, without GR24^5DS treatment; GR24^5DS, with 5 μM GR24^5DS treatment.

Figure 7.

MdPRT1 alleviates the inhibition of anthocyanin biosynthesis by MdSMXL8. A) RT-qPCR analysis of relative MdPRT1 expression after GR24^5DS treatment. Apple fruits were treated with 5 μM GR24^5DS for 4 h, and untreated fruits were used as the control. REL, relative expression level. The value for the control at 0 h was set to 1. Error bars denote standard deviations. Means and standard deviations from three biological replicates are shown (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. B and &C) Identification of the role of MdPRT1 in regulating anthocyanin biosynthesis via transient overexpression assays in apple fruits. Apple fruits overexpressing MdPRT1 were exposed to light for 2 d. pIR, IL60-1 + IL60-2; MdPRT1-pIR, IL60-1 + MdPRT1-IL60-2. RAC, relative anthocyanin content. Anthocyanin content and gene expression level at the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01. D and E) Identification of the role of MdPRT1 in regulating anthocyanin biosynthesis via transient suppression assays in apple fruits. Apple fruits with suppressed MdPRT1 expression were exposed to light for 6 d. TRV, TRV1 + TRV2; MdPRT1-TRV, TRV1 + MdPRT1-TRV2. Anthocyanin content and the gene expression level at the sites infiltrated with the empty vector were set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. The qPCR data are means ± standard deviations of three biological replicates (n = 3). The MdACTIN gene served as an internal control. Asterisks indicate significant differences based on the t-test. **P < 0.01; *P < 0.05. F and G) Identification of the role of MdPRT1 in regulating anthocyanin biosynthesis via stable transformation assays in apple calli. Apple calli were transferred to MS medium containing 5 μM GR24^5DS or without GR24^5DS and exposed to light for 5 d. WT, wild-type; MdPRT1, overexpression of MdPRT1; asMdPRT1, suppression of MdPRT1. The value of WT grown on medium without GR24^5DS was set to 1. Each treatment was performed in triplicate, and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Asterisks indicate significant differences based on the t-test. **P < 0.01. H) Evaluation of the relationship between MdSMXL8 and MdPRT1 in regulating anthocyanin biosynthesis via transient expression in apple fruits. MdPRT1-pIR, IL60-1 + MdPRT1-IL60-2; MdSMXL8-TRV, TRV1 + MdSMXL8-TRV2; MdPRT1-pIR + MdSMXL8-TRV, IL60-1 + MdPRT1-IL60-2 + TRV1 + MdSMXL8-TRV2. Anthocyanin content of TRV was set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 8.

MdSMXL8 interacts with MdRGL2a. A) Effect of GR24^5DS and GA on anthocyanin biosynthesis in apple peel. Debagged “Fuji” fruits at 120 DAFB were treated with 5 μM GR24^5DS or 20 μM GA₃ (GA) for 30 min. Untreated apple fruits were used as controls. The treated apple fruits were exposed to light for 6 d. RAC, relative anthocyanin content. Anthocyanin content was measured in apple peel, and the control was set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. B) Y2H assay showing the interaction between MdSMXL8 and MdRGL2a. Full-length MdRGL2a and MdSMXL8 were cloned into the pGAD or pGBD vectors, respectively. −T/−L, SD/−Trp/−Leu; −T/−L/−H/−A, SD/−Trp/−Leu/−His/−Ade. C) Pull-down assay showing the interaction between MdSMXL8 and MdRGL2a. D) Confirmation of the interaction between MdSMXL8 and MdRGL2a in nuclei using a BiFC assay in onion epidermal cells. Scale bars = 10 μm. E) Co-IP assay. The MdRGL2a-GFP protein was immunoprecipitated from MdRGL2a-GFP and MdRGL2a-GFP/MdSMXL8-GUS transgenic apple calli with anti-GFP antibody, and the eluent was detected with anti-GUS antibody. F) Evaluation of the relationship between MdSMXL8 and MdRGL2a in regulating anthocyanin biosynthesis via transient expression assays in apple fruits. pIR, IL60-1 + IL60-2; MdSMXL8-pIR, IL60-1 + MdSMXL8-IL60-2; MdSMXL8-pIR + MdRGL2a-pIR, IL60-1 + MdSMXL8-IL60-2 + IL60-1 + MdRGL2a-IL60-2. Anthocyanin content of pIR was set to 1. Each treatment was performed in triplicate, and each replicate contained 8 to 10 fruits. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. G and H) Evaluation of the relationship between MdSMXL8 and MdRGL2a in regulating anthocyanin biosynthesis via stable transformation in apple calli. Apple calli were exposed to light for 5 d. WT, wild-type; MdSMXL8, overexpression of MdSMXL8; MdSMXL8/MdRGL2a, overexpression of MdRGL2a in MdSMXL8-overexpressing calli; MdSMXL8/asMdRGL2a, suppression of MdRGL2a in MdSMXL8-overexpressing calli. Anthocyanin content of WT was set to 1. Each treatment was performed in triplicate and each replicate contained 3 to 5 calli masses. A representative photograph is shown here. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test. I) Competitive binding assay showing the effect of MdRGL2a on the interaction between MdSMXL8 and MdAGL9. J) LUC complementation imaging assay showing the effect of MdRGL2a on the interaction between MdSMXL8 and MdAGL9. The value of control was set to 1. The experiment was repeated three times with similar results and each replicate contained 3 to 5 Nicotiana benthamiana leaves. Error bars denote standard deviations. Different lowercase letters indicate a significant difference at P < 0.05 based on one-way ANOVA followed by Tukey's test.

Figure 9.

A proposed model for the mechanism underlying SL-induced anthocyanin biosynthesis in apple. MdAGL9 promotes anthocyanin biosynthesis by activating the transcription of MdHY5, a key regulator involved in light-induced anthocyanin pigmentation in apple. When SLs are absent, MdSMXL8 inhibits anthocyanin biosynthesis by interacting with MdAGL9 to reduce MdAGL9-activated transcription of MdHY5. When SLs are present, MdPRT1 mediates the ubiquitination and degradation of MdSMXL8 to release MdAGL9 from the MdSMXL8-MdAGL9 inhibitory complex, leading to the activation of MdAGL9-mediated anthocyanin biosynthesis. MdRGL2a weakens the inhibitory effect of MdSMXL8 on anthocyanin biosynthesis by interfering with the interaction between MdSMXL8 and MdAGL9, and the MdRGL2a-MdSMXL8 module mediates the cross-talk between GA and SL signaling.