Functional diversification of the potato R2R3 MYB anthocyanin activators AN1, MYBA1, and MYB113 and their interaction with basic helix-loop-helix cofactors

Abstract

In potato (Solanum tuberosum L.), R2R3 MYBs are involved in the regulation of anthocyanin biosynthesis. We examined sequences of these MYBs in cultivated potatoes, which are more complex than diploid potato due to ploidy and heterozygosity. We found amino acid variants in the C-terminus of the MYB StAN1, termed R0, R1, and R3, due to the presence of a repeated 10-amino acid motif. These variant MYBs showed some expression in both white and pigmented tubers. We found several new alleles or gene family members of R2R3 MYBs,StMYBA1 and StMYB113, which were also expressed in white potato tubers. From functional analysis in tobacco, we showed that the presence of a C-terminal 10-amino acid motif is optimal for activating anthocyanin accumulation. Engineering a motif back into a MYB lacking this sequence enhanced its activating ability. Versions of StMYBA1 and StMYB113 can also activate anthocyanin accumulation in tobacco leaves, with the exception of StMYB113-3, which has a partial R2R3 domain. We isolated five family members of potato StbHLH1, and one StJAF13, to test their ability to interact with MYB variants. The results showed that two alleles of StbHLH1 from white skin and red skin are non-functional, while three other StbHLH1s have different co-regulating abilities, and need to be activated by StJAF13. Combined with expression analysis in potato tuber, results suggest that StbHLH1 and StJAF13a re key co-regulators of anthocyanin biosynthesis, while the transcripts of MYB variants StAN1,StMYBA1, and StMYB113 are well expressed, even in the absence of pigmentation.

Keywords: Anthocyanin; MYB transcription factors; cofactors; diversification; interaction; potato..

Fig. 1.

Characterization of the four genotypes used in this study. (A) Skin and flesh colour of Xin Daping (XD), Hei Meiren (HM), Gannongshu No.5 (GN) and Qingshu No.9 (QS). (B) Total anthocyanin content of skin, flesh, and red vascular ring of the four genotypes. The data represent the means±SE of three biological replicates. Statistical significance was determined by one-way ANOVA; significant differences between means [Least Significant Difference (LSD), P<0.05] are indicated where letters above the bars differ. WS, white skin; WF, white flesh; PS, purple skin; PF, purple flesh; RS, red skin; RFV, red vascular ring.

Fig. 2.

Protein sequence alignment of anthocyanin MYB regulators from potato. The R2 and R3 repeat domains are indicated by arrows. Box (A) indicates the conserved region of the bHLH interacting motif ([DE]Lx2[RK]x3Lx6Lx3R). Box (B) indicates a conserved motif [A/S/G]NDV in the R2R3 domain for dicot anthocyanin-promoting MYBs. Box (C) indicates the perfect repeat TIAPQPQEGI. Box (D) indicates a C-terminal-conserved motif [R/K]Px[P/A/R]xx[F/Y] for anthocyanin-regulating MYBs (Hichri et al., 2010). NCBI protein accession numbers: StAN1-R0, AKA95391; StAN1-R1, AKA95392; StAN-R3, AKA95392; StAN1_777, AAX53089; StAN1_816, AAX53087; StMYB113-1, ALA13583; StMYB113-2, ALA13584; StMYBA1-1, ALA13581; StMYBA1-2, ALA13582; StMTF1, ABY40370; StMTF2, ABY40371.

Fig. 3.

Phylogenetic relationship analysis of potato MYBs and known anthocyanin MYB regulators from other species. Sequences were aligned using Geneious v.6.1.6 (Drummond et al., 2011) with a cost matrix of 65%, a gap open penalty of 12, and a gap extension penalty of 3. Phylogenetic and molecular evolutionary analysis was conducted using MEGA version 6.0. The evolutionary history was inferred using the neighbour-joining method and 1000 bootstrap replicates; bootstrap values less than 50 are not shown. The predicated proteins of StAN1-R0, StAN1-R1, StAN1-R3, StMYBA1-1, StMYBA1-2, StMYB113-1, and StMYB113-2 are indicated by black oval dots.

Fig. 4.

Expression analysis of StAN1s, StMYBA1s, and StMYB113s in four potato cultivars. (A) Schematic of the StAN1 gene. PCR 1–4 represent the different regions. (B) qPCR expression of the different regions of StAN1 in skin, flesh, and red vascular ring of the four genotypes, XD, HM, GN, and QS. Red skin of GN was set as a calibrator. (C) qPCR expression of StMYBA1s and StMYB113s in skin, flesh and red vascular ring of the four genotypes. Red skin of GN was set as a calibrator. The data represent the means±SE of three biological replicates. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. WS, white skin; PS, purple skin; RS, red skin; WF, white flesh; PF, purple flesh; RVF, red vascular ring.

Fig. 5.

Quantitative analysis of transcript levels of StbHLH1 and StJAF13 in skin, flesh and red vascular ring of four genotypes. Purple skin and purple flesh of HM were set as calibrators. The data represent the means±SE of three biological replicates. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. WS, white skin; PS, purple skin; RS, red skin; WF, white flesh; PF, purple flesh; RVF, red vascular ring.

Fig. 6.

Transient assays of StAN1-R0, StAN1-R1, StAN1-R3, StMYBA1-1, StMYBA1-2, StMYB113-1, StMYB113-2, and StMYB113-3. (A) Activation of three alleles of potato DFR promoters and F3′5′H promoter. Error bars are the SE of three independent experiments with four replicate reactions. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. (B) Patches of anthocyanin production in tobacco leaves induced by infiltration with StAN1-R0, StAN1-R1, StAN1-R3, StMYBA1-1, StMYBA1-2, StMYB113-1, StMYB113-2, StMYB113-3, and GUS.

Fig. 7.

Phenotypes and anthocyanin content of transgenic tobacco plants transformed with empty vector, StAN1-R0, StAN1-R1, and StAN1-R3. (A) Visible reddening was seen in leaves (i) and flowers (ii) of plants transformed with StAN1-R0, StAN1-R1, and StAN1-R3. (B) Anthocyanin content from five independent transgenic lines of each construct showed the highest concentration in two out of the five StAN1-R1 lines and the lowest concentration in three out of five StAN1-R3 lines. C represents empty vector controls. Error bars are the SE for three replicate extracts per line. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. * indicates no detectable levels of anthocyanin content.

Fig. 8.

Quantitative analysis of transcript levels of StMYB AN1, NtAN1a-bHLH, and NtAN1b-bHLH in tobacco transgenic lines. The data represent the means±SE of three biological replicates. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. Genotypes denoted by * showed no detectable levels of expression.

Fig. 9.

Transient activation of anthocyanic responses by StAN1, StMYBA1, and StMYB113 combined with StbHLH1 and with StJAF13. (A) Flowers of transgenic tobacco plants transformed with empty vector, NtAn1 RNAi, or NtJAF13 RNAi. (B) Patches of anthocyanin production in NtAn1 RNAi tobacco leaves infiltrated by MYBs alone or combined with five variants of StbHLH1 and StJAF13.

Fig. 10.

Transient activation of anthocyanic responses in tobacco by simulating transcription factors present in potato skin. (A) Patches of anthocyanin production in NtAn1 RNAi tobacco leaves induced by the combinations as shown. (B) Anthocyanin was extracted from patches of each combination. Error bars are the SE of three biological replicates. Statistical significance was determined by one-way ANOVA; significant differences between means (LSD, P<0.05) are indicated where letters above the bars differ. E represents empty vector control. Genotypes denoted by * showed no detectable levels of anthocyanin content.