Fig 1.
ARF2A expression in tomato fruit.
Relative expression levels of ARF2A analyzed by qRT-PCR, in (A) WT fruit at five developmental stages; (B) rin and nor mutants; (C) TAGL1 over-expressing fruit (35S:TAGL1); and (D) nr mutant. Relative expression levels of ARF2A in fruit at three developmental stages, treated with (E) 1-MCP; (F) ethylene; (H) NAA; and (J) ABA. Relative expression levels of ARF2A in fruit at the MG stage, at 0, 2, 4 and 6 days post-treatment with (G) ethrel; (I) NAA; and (K) ABA. Error bars represent SE. Statistical significance was evaluated using a student’s t-test, *p-value<0.05, **p-value<0.01 and ***p-value<0.001; dpa: days post anthesis; IG: immature green; MG: mature green; Br: breaker; Or: orange; and R: red.
Fig 2.
Phenotype and expression levels of ARF2 genes in ARF2as transgenic lines.
Fruit of ARF2as lines were analysed for relative expression levels by qRT-PCR of (A) ARF2A; and (B) ARF2B. (C-D) The ripening of ARF2as fruit is delayed as compared to WT. (D-E) ARF2as fruit were parthenocarpic or nearly parthenocarpic with reduced number of seeds. (F) Principle component analysis (PCA) plot from untargeted analysis of metabolites. DPA: days post anthesis; error bars represent SE; Statistical significance was evaluated using a student’s t-test, **p-value<0.01.
Fig 3.
Metabolic analysis of ARF2as transgenic fruit.
Accumulation in WT and ARF2as fruit of metabolites from the classes: (A) phenylpropanoids and glycoalkaloids (in red stage fruit); (B) isoprenoids (in red stage fruit, undetectable in green fruit); and (C) phytohormones (abscisates are presented in the two left graphs, and cytokinins in the middle and right panels; at both green and red stages of fruit ripening). WT samples are displayed as black bars and ARF2as as grey bars; error bars represent SE; statistical significance was evaluated using a student’s t-test, *p-value<0.05, **p-value<0.01, ***p-value<0.001; and DPA: days post anthesis.
Fig 4.
Over-expression of ARF2A in transgenic tomato plants.
(A) Silent mutations in the tasi-RNA recognition site introduced into the ARF2A open reading frame; (B) days from anthesis to breaker stage in WT and ARF2-OX fruit (Error bars represent SE); (C) blotchy ripening phenotype observed in ARF2-OX fruit (DPA: days post anthesis); and (D) blotchy ripening occasionally observed in WT fruit which eventually all turn red in comparison with the ARF2-OX lines which have regions which remain yellow; (E) A scheme representing the sampling of tissues from WT and ARF2-OX fruit at 39, 42 and 53 dpa, in this study; when patches were visible at 42 and 53 dpa, they were harvested and treated separately. Relative expression levels of ARF2A variants were analyzed by qRT-PCR in WT and patches of ARF2-OX fruit at 39 and 42 dpa, using oligonucleotides specific to (F) the transgene (trans-ARF2A); and (G) the endogenous ARF2A gene (endo-ARF2A). Black bars represent WT, hatched bars ARF2-OX at 39 dpa, white bars ARF2-OX green patches at 42 dpa and grey bars ARF2-OX red patches at 42 dpa. Error bars represent SE. DPH: days post-harvest; n.d.: not detected; dpa: days post anthesis.
Fig 5.
Altering ethylene signaling in ARF2-OX transgenic fruit.
ARF2-OX fruit at the mature green (MG) stage, before the visual appearance of patches, were treated with either (A) ethrel, or (B) 1-MCP, and phenotypes were observed at (A) 10 and 16, or (B) 7 and 10 DPT. (C) Ethylene emission was measured from WT and ARF2-OX fruit harvested at the MG stage, every 1–3 days for 16 days, the red bars and arrows indicate the breaker stage. Error bars represent SE. DPT: days post treatment.
Fig 6.
Microarray analysis of ARF2-OX fruit.
Gene expression was analyzed in WT at 42 dpa (mature green stage; WT-42G) and 53 dpa (red stage; WT-53R) and green (ARF2-OX-42G) and red (ARF2-OX-42R) patches from ARF2-OX fruit at 42 dpa by microarray analysis. Results are displayed as (A) a principal component analysis (PCA) and (B) a Venn diagram of the differentially expressed genes in the comparisons: ARF2-OX-42G to WT-42G; ARF2-OX-42R to WT-42G; and WT-53R to WT-42G. P-value<0.01 and FDR<0.05; dpa: days post anthesis.
Fig 7.
Expression of ripening-related genes in ARF2-OX fruit.
Relative expression levels of ripening-related genes analyzed by qRT-PCR in WT and ARF2-OX fruit at 39, 42 and 53 dpa. (A) GLK2; (B) ETR2; (C) ACS4; (D) AP2a; (E) NR; (F) PSY; (G) RIN; (H) ACS2; Black bars represent WT, hatched bars ARF2-OX at 39 dpa, white bars ARF2-OX green patches at 42 and 53 dpa and grey bars ARF2-OX red patches at 42 and 53 dpa. Error bars represent SE. Statistical significance was evaluated using a student’s t-test, *p-value<0.05; dpa: days post anthesis.
Table 1.
Summary of expression profiles of ripening-related genes in red regions of the ARF2-OX fruit at 42 days post anthesis.
Fig 8.
Metabolic analysis of ARF2-OX fruit.
WT and ARF2-OX fruit were analyzed at 42 and 53 dpa by UPLC-qTOF-MS in positive mode, (A) results are visualized by a principle component analysis (PCA) plot; and displayed as histograms for (B) targeted flavonoids (upper row), and targeted glycoalkaloids (lower row). (C) Isoprenoids were analyzed in WT and ARF2-OX fruit at 42 and 53 dpa by HPLC. Grey bars represent WT, black bars ARF2-OX green/yellow patches and white bars ARF2-OX red patches. Error bars represent SE. Statistical significance was evaluated using a student’s t-test, *p-value<0.05; dpa: days post anthesis.
Fig 9.
Hormone profiling in ARF2-OX transgenic fruit.
WT and ARF2-OX fruit, at 39, 42 and 53 dpa were analyzed for levels of (A) salicylic acid; (B) abscisic acid; (C-G) cytokinins of the trans-zeatin biosynthesis branch and (H-J) cytokinins of the cis-zeatin biosynthesis branch, using UPLC-ESI-MS/MS. Relative expression level of the cytokinin-related genes (K) CKX7-like; (L) LOG8-like was analysed by qRT-PCR. Black bars represent WT, hatched bars ARF2-OX at 39 dpa, white bars ARF2-OX green patches at 42 and 53 dpa and grey bars ARF2-OX red patches at 42 and 53 dpa. Error bars represent SE. Statistical significance was evaluated using a student’s t-test, *p-value<0.05 and **p-value<0.01; dpa: days post anthesis.
Fig 10.
Dimerization of the ARF2A protein and its interaction.
ARF2A was cloned downstream of the DNA-binding domain (DB-ARF2A) and co-transformed into yeast with either (A) ARF2A cloned downstream of the activation domain (AD-ARF2A); or (B) ASR1 cloned downstream of the activation domain (AD-ASR1), yeast growth on media lacking leucine, tryptophan, histidine and adenine indicated positive protein-protein interactions. (C) Relative expression levels of ASR1 in WT cv. MicroTom fruit at five developmental stages (IG: immature green; MG: mature green; Br: breaker; Or: orange; and R: red), error bars represent SE; statistical significance was evaluated using an ANOVA test (JMP software, SAS) with three biological repeats based on the average of three technical replicates, values indicated by the same letter (a,b,c) are not statistically significant, p-value<0.05. (D) A Bimolecular Fluorescence Complementation assay (BiFC) was carried out by transient expression in tobacco leaves; ARF2A was cloned downstream of the amino-terminal region of YFP (yellow fluorescent protein; YN-ARF2A) and ASR1 was cloned downstream of the carboxy-terminal region of YFP (YC-ASR1); leaf regions were examined for fluorescent signal by light and confocal fluorescence microscopy. Inset zoom region shows that the ARF2A-ASR1 interaction is nuclear localized. Scale bars in the light and confocal fluorescence microscopy represent 50 μm and 10 μm, respectively.
Fig 11.
A general scheme depicting the activity of ARF2A in the regulatory network controlling fruit ripening.
Data from this study and the complementary work [33] as well as previous reports concerning ripening control and information on tomato ARF2A and the ARF protein family was integrated and presented in a general scheme. ARF2A exerts its activity through the ethylene-dependent ripening pathway by impacting ripening regulators such as ones of the MADS-Box and AP2 protein families, genes associated with ethylene biosynthesis and signaling, carotenoid as well as other ripening metabolic pathways. The induction of ripening is likely indirect through an additional factor which functions as a ripening-repressor. ARF2A functions as a negative regulator, reducing the expression of the unknown ripening-repressor and thus activating the expression of several ‘master’ regulators and downstream ripening genes. It appears that ARF2A obtains signals from at least three hormone pathways, including ABA, auxin and ethylene while its activity impacts ABA, cytokinins and salicylic acid, at least at the level of hormone biosynthesis. The ASR1 protein likely interacts with ARF2A and together they fine tune the sensitivity of the fruit tissue to ethylene and to the capacity to ripen.