Phenylpropanoids Accumulation in Eggplant Fruit: Characterization of Biosynthetic Genes and Regulation by a MYB Transcription Factor

Abstract

Phenylpropanoids are major secondary metabolites in eggplant (Solanum melongena) fruits. Chlorogenic acid (CGA) accounts for 70-90% of total phenolics in flesh tissues, while anthocyanins are mainly present in the fruit skin. As a contribution to the understanding of the peculiar accumulation of these health-promoting metabolites in eggplant, we report on metabolite abundance, regulation of CGA and anthocyanin biosynthesis, and characterization of candidate CGA biosynthetic genes in S. melongena. Higher contents of CGA, Delphinidin 3-rutinoside, and rutin were found in eggplant fruits compared to other tissues, associated to an elevated transcript abundance of structural genes such as PAL, HQT, DFR, and ANS, suggesting that active in situ biosynthesis contributes to anthocyanin and CGA accumulation in fruit tissues. Putative orthologs of the two CGA biosynthetic genes PAL and HQT, as well as a variant of a MYB1 transcription factor showing identity with group six MYBs, were isolated from an Occidental S. melongena traditional variety and demonstrated to differ from published sequences from Asiatic varieties. In silico analysis of the isolated SmPAL1, SmHQT1, SmANS, and SmMyb1 promoters revealed the presence of several Myb regulatory elements for the biosynthetic genes and unique elements for the TF, suggesting its involvement in other physiological roles beside phenylpropanoid biosynthesis regulation. Transient overexpression in Nicotiana benthamiana leaves of SmMyb1 and of a C-terminal SmMyb1 truncated form (SmMyb1Δ9) resulted in anthocyanin accumulation only of SmMyb1 agro-infiltrated leaves. A yeast two-hybrid assay confirmed the interaction of both SmMyb1 and SmMyb1Δ9 with an anthocyanin-related potato bHLH1 TF. Interestingly, a doubled amount of CGA was detected in both SmMyb1 and SmMyb1Δ9 agro-infiltrated leaves, thus suggesting that the N-terminal region of SmMyb1 is sufficient to activate its synthesis. These data suggest that a deletion of the C-terminal region of SmMyb1 does not limit its capability to regulate CGA accumulation, but impairs anthocyanin biosynthesis. To our knowledge, this is the first study reporting a functional elucidation of the role of the C-term conserved domain in MYB activator proteins.

Keywords: RACE; S. melongena; chlorogenic acid; gene regulation; genome walking; qRT-PCR.

FIGURE 1

LC–MS analysis of phenylpropanoids. Chlorogenic acid, delphinidin 3-rutinoside, rutin were extracted from 100 mg of lyophilized tissue in 75% MeOH and quantified by LC–MS. Values are expressed as means ± SD (n = 3). ^∗Compounds were not detected.

FIGURE 2

Relative transcript levels of SmPAL, SmC4H, Sm4CL, SmHQT, SmDFR, and SmANS in various Solanum melongena organs and at two leaf development stages (YL, young leaves and ML, mature leaves). The results were analyzed using the ^ΔΔCt method and presented as fold changes compared with the young leaves, used as internal calibrator. Data are reported as means ± SD. Means denoted by the same letter did not differ significantly at p ≤ 0.05 according to Duncan’s multiple range test.

FIGURE 3

Gene structure and phylogenetic analysis of S. melongena Myb1. (A) Representation of SmMyb1 genomic sequence (KT727965) and cDNA (KT259043). Light and dark gray boxes indicate the R2R3 domain in the SmMyb1 coding sequence, solid black line indicates the intronic regions. (B) R2R3-MYB proteins from other species were aligned using Clustal X, and the evolutionary history was inferred using the Neighbor-Joining method. The optimal tree with the sum of branch length = 2.20562095 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the p-distance method and are in the units of the number of amino acid differences per site. The analysis involved 29 amino acid sequences. All positions containing gaps and missing data were eliminated. There were a total of 94 positions in the final dataset. Evolutionary analyses were conducted in MEGA6. Protein sequences used for the phylogenetic tree have the following accession numbers: Malus domestica MdMYB1 (ADQ27443.1); Malus domestica MdMYB10 (ACQ45201.1); Malus domestica MdMYB110a (AFC88038.1); Arabidopsis thaliana AtMYB75 (AEE33419.1); Arabidopsis thaliana AtMYB113 (NP_176811.1); Arabidopsis thaliana AtMYB114 (AEE34502.1); Brassica oleracea var. botrytis BoMYB2 (ADP76651.1); Morella rubra MrMYB1 (ADG21957.1); i GhMYB10 (CAD87010.1); Vitis vinifera VvMYBA1 (BAD18977.1); Vitis vinifera VvMYBA2 (BAC07540.1); Ipomoea purpurea IpMYB1 (BAE94388.1); Nicotiana tabacum NtAN2 (ACO52470.1); Solanum lycopersicum SlANT1 (AAQ55181.1); Solanum lycopersicum SlAN2 (FJ705333.1); Antirrhinum majus AmROSEA2 (ABB83827.1); Antirrhinum majus AmROSEA1 (ABB83826.1); Solanum tuberosum StCAI (ABY40370.1); Solanum tuberosum StAN1 (AGC31676.1); Fragaria x ananassa FaMYB (ABX79947); Epimedium sagittatum EsMYBA1 (AGT39059.1); Lilium hybrid division I LhMYB6 (BAJ05399.1); Medicago truncatula MtLAP2 (ACN79539.1); Pyrus communis PcMYB10 (ABX71487.1); Solanum melongena SmMYB2 (AIP93874); Solanum melongena SmMYB1 (KT259043).

FIGURE 4

Relative transcript levels of SmMyb1, SmMyb2, SmTT8, and SmHSC-70-2-like in various S. melongena organs and at two leaf development stages (YL, young leaves and ML, mature leaves). The results were analyzed using the ^ΔΔCt method and presented as fold changes compared with the young leaves, used as internal calibrator. Data are reported as means ± SD.

FIGURE 5

The effects of over-expression of SmMyb1 in Nicotiana benthamiana leaves. (A) Leaves of N. benthamiana after agro-infiltration with SmMyb1, SmMyb1Δ9, pGWB411 (empty vector, EV). (B) Gene expression analysis of HQT, CHS, DFR, and ANS late phenylpropanoid structural genes in agro-infiltrated N. benthamiana leaves monitored by qRT-PCR. The results were analyzed using the ^ΔΔCt method and presented as fold changes compared with the young leaves, used as internal calibrator. Data are reported as means ± SD. Means denoted by the same letter did not differ significantly at p ≤ 0.05 according to Duncan’s multiple range test.

FIGURE 6

SmMYB1 interacts with bHLH TF StbHLH1 in the yeast two-hybrid assay. SmMYB1 and SmMyb1Δ9 were cloned in the prey plasmid pGADT7 and co-transformed with StbHLH1pGBKT7 (D’Amelia et al., 2014) in yeast strain AH109. The SmMyb1pGADT7/pGBKT7 and pGADT7/StbHLH1pGBKT7 combinations served as negative controls while StbHLH1/StAN1 is shown as positive control. Yeast cells grown on synthetic complete media (–W/–L, Right) and on selective media (–W/–L/–H/–A, Left) are shown. Pictures were taken 3 days after incubation at 30°C.