Systematic analyses of the MIR172 family members of Arabidopsis define their distinct roles in regulation of APETALA2 during floral transition

Abstract

MicroRNAs (miRNAs) play important roles in regulating flowering and reproduction of angiosperms. Mature miRNAs are encoded by multiple MIRNA genes that can differ in their spatiotemporal activities and their contributions to gene regulatory networks, but the functions of individual MIRNA genes are poorly defined. We functionally analyzed the activity of all 5 Arabidopsis thaliana MIR172 genes, which encode miR172 and promote the floral transition by inhibiting the accumulation of APETALA2 (AP2) and APETALA2-LIKE (AP2-LIKE) transcription factors (TFs). Through genome editing and detailed confocal microscopy, we show that the activity of miR172 at the shoot apex is encoded by 3 MIR172 genes, is critical for floral transition of the shoot meristem under noninductive photoperiods, and reduces accumulation of AP2 and TARGET OF EAT2 (TOE2), an AP2-LIKE TF, at the shoot meristem. Utilizing the genetic resources generated here, we show that the promotion of flowering by miR172 is enhanced by the MADS-domain TF FRUITFULL, which may facilitate long-term silencing of AP2-LIKE transcription, and that their activities are partially coordinated by the TF SQUAMOSA PROMOTER-BINDING-LIKE PROTEIN 15. Thus, we present a genetic framework for the depletion of AP2 and AP2-LIKE TFs at the shoot apex during floral transition and demonstrate that this plays a central role in floral induction.

Fig 1. Expression patterns of the MIR172 gene family in the shoot apex.

(A–F) Graphs depicting RT-qPCR results of (A) AP1, (B) MIR172A, (C) MIR172B, (D) MIR172C, (E) MIR172D, and (F) MIR172E expression in materials derived from the shoot apex of LD-grown plants that were harvested at the indicated times after germination. Expression was normalized to 7d samples. Error bars indicate SEM of 3 independent biological replicates. Data underlying panels A–F are provided in S1 Data. (G, H) Confocal laser scanning micrographs of the shoot apices of (G) MIR172A-NVG, (H) MIR172B-NVG, and (I) MIR172D-NVG transgenic plants grown in LD conditions and harvested at the indicated times after germination. (H) Note fluorescence in L1 at 21d MIR172B-NVG^11.3 meristem (inset, arrowheads). Fluorescence from the Venus protein is artificially colored in green, and the fluorescence from the Renaissance dye is artificially colored in magenta. LD, long-day; NVG, NLS-Venus-GUS; RT-qPCR, quantitative reverse transcription PCR; SEM, standard error of the mean.

Fig 2. Functional analysis of single mir172 mutants LDs and SDs.

(A) Schematic depicting the structure of a representative “reference” miR172 precursor and the mutant forms generated using CRISPR-Cas-9 DNA editing. The orange, red, and green bars represent the coding sequences for miR172, miR172*, and the biogenesis site required for DCL1 cleavage, respectively. Truncation of these bars and opaque sections of the mutant mir172 precursor indicates the regions that were deleted via CRISPR-Cas-9 in mir172a-2, mir172b-3, mir172c-1, mir172d-2, and mir172e-1. (B) A graph depicting the TLN of first generation transformants in a Col-0 background harboring a T-DNA containing only the UBQ10pro-3′OCS promoter-terminator cassette or expressing reference Col-0 or mutated versions of the indicated MIR172 gene under the control of the UBQ10pro-3′OCS cassette. (C, D) Graphs depicting the TLN and bolting times of WT and plants harboring single mir172 mutations in (C) LD and (D) SD conditions. Data underlying panels B–D are provided in S2 Data. (E–G) Confocal laser scanning micrographs of the shoot apices of (E) MIR172A-NVG, (F) MIR172B-NVG, and (G) MIR172D-NVG transgenic plants grown in SD conditions and harvested at the indicated times after germination. Fluorescence from the Venus protein is artificially colored in green, and the fluorescence from the Renaissance dye is artificially colored in magenta. Cas-9, CRISPR associated protein-9; CRISPR, clustered regularly interspaced short palindromic repeats; LD, long-day; NVG, NLS-Venus-GUS; SD, short-day; TLN, total leaf number; WT, wild-type.

Fig 3. Flowering time of higher-order mir172 mutants in LDs and SDs.

(A, B) Graphs depicting DTB and TLN or RLN of Col-0, mir172a-2, mir172a-2 b-3, mir172a-2 b-3 c-1, mir172a-2 b-3 c-1 d-3, and mir172a-2 b-3 c-1 d-3 e-1 mutant combinations in LD (A) and SD (B) conditions. The difference between the means of the indicated genotypes is shown above the box plots. Data underlying panels (A) and (B) are provided in S2 Data. (C) Photographs of mir172a-2 b-3 c-1 d-3 e-1 plants after approximately 6 months of growth under SD conditions. Plant initiated flowering shoots from axillary meristems (white arrowheads) rather than the primary meristem. (D, E) Graphs depicting DTB and total or RLN of Col-0, mir172a-2 b-3 d-2, mir172a-2 b-3 c-1 d-3 e-1 mutant combinations and 35Spro:MIM172 in LD (D) and SD (E) conditions. Data underlying panels (D) and (E) are provided in S2 Data. (F) A photograph of Col-0, mir172a-2 b-3 d-2, mir172a-2 b-3 c-1 d-3 e-1 mutant combinations and 35Spro:MIM172 grown under SD conditions. DTB, days to bolting; LD, long-day; RLN, Rosette leaf number; SD, short-day; TLN, total leaf number.

Fig 4. Activity of MIR172 genes in leaves.

(A–C) Photographs of 14-day-old (A) MIR172A-NVG, (B) MIR172B-NVG, and (C) MIR172D-NVG transgenic plants grown in LD conditions after GUS staining and clearing. Scale bar indicates 2 mm. (D) Graphs depicting RT-qPCR results of miR172 (upper panel) and FT mRNA (lower panel) in leaves 1 and 2 of Col-0, mir172b-3 and mir172a-2 b-3 c-1 d-3 e-1 grown for 12 LDs, harvested at ZT 16. Each point represents an individual biological replicate. Each biological replicate was normalized to the mean of the 3 Col-0 biological replicates. Data underlying panels D are provided in S1 Data. (E) Micrographs of RNA in situ hybridizations using a commercial probe designed to recognize miR172 of Col-0 (upper panels) and mir172a-2 b-3 c-1 d-3 e-1 (lower panels) apices grown under LD conditions for the indicated times. FT, FLOWERING LOCUS T; GUS, β-glucuronidase enzyme; LD, long-day; NVG, NLS-Venus-GUS; RT-qPCR, quantitative reverse transcription PCR; ZT, zeitgeber time.

Fig 5. AP2 activity in the shoot apex.

(A) Confocal laser scanning micrographs of the shoot apices of AP2-Venus transgenic plants grown in LD conditions in a reference Col-0 background (upper panels) and mir172a-2 b-3 d-2 background (lower panels). Fluorescence from the Venus protein is artificially colored in green, and the fluorescence from the Renaissance 2200 dye is artificially colored in magenta. (B) Micrographs of RNA in situ hybridizations using a commercial probe designed to recognize AP2 mRNA of Col-0 (upper panels) and mir172a-2 b-3 c-1 d-3 e-1 (lower panels) apices grown under LD conditions for the indicated times. (C, D) Graphs depicting RNA-seq-derived data of the mRNA levels in apices of AP2, TOE2, and AP1 in Col-0 and miR172abd at the indicated time points after germination in (C) LDs and (D) SDs. Error bars represent the standard deviation of 3 biological replicates. Data underlying panels (C) and (D) are provided in S1 Data. (E, F) Graphs depicting the DTB and T/RLNs of Col-0, ap2-12, miR172a-2 b-3 d-2, and mir172a-2 b-3 d-2 ap2-12 plants grown under (E) LD and (F) SD conditions. Data underlying panels (E) and (F) are provided in S2 Data. (G) Photograph of Col-0, ap2-12, mir172a-2 b-3 d-2, and ap2-12 mir172a-2 b-3 d-2 plants grown in LD conditions. AP2, APETALA2; DTB, days to bolting; LD, long-day; RLN, Rosette leaf number; SD, short-day; TLN, total leaf number.

Fig 6. Interplay between Venus-rSPL15, MIR172, and FUL.

(A) Graphs depicting DTF and TLN between combinations of ful-2 and mir172a-2 b-3 c-1 d-3. Data underlying panel A are provided in S2 Data. (B, D) Photographs of (B, C) flowers arising from ful-2 mir172a-2 b-3 c-1 d-3 plants and (D) an approximately 4-month-old ful-2 mir172a-2 b-3 c-1 d-3 plant. (E–J) RT-qPCR results of (E) MIR172A, (F) MIR172B, (G) MIR172C, (H) MIR172D, (I) MIR172E, and (J) FUL expression in materials derived from the shoot apex of LD-grown Col-0 and rSPL15-Venus plants that were harvested at the indicated times after germination. Expression was normalized to 7 d Col-0 samples. Error bars indicate SEM of 3 independent biological replicates. Data from Col-0 is the same as in Fig 1, but tissue collection was originally paired with rSPL15-Venus. (K, L) Graphs depicting DTB and T/RLN between the indicated combinations of rSPL15-Venus, mir172a-2, miR172b-3, and ful-2 mutants grown in (K) LD and (L) SD conditions. Data underlying panels (K) and (L) are provided in S2 Data. DTF, days to the first flower opening; FUL, FRUITFULL; LD, long-day; RLN, Rosette leaf number; RT-qPCR, quantitative reverse transcription PCR; SD, short-day; SEM, standard error of the mean; TLN, total leaf number.

Fig 7. Genetic interactions between mir172, ful, and spl15 mutations.

(A) A box plot depicting DTB for plants carrying the indicated combinations of spl15-1, ful-2, and mir172 mutations and grown in SD conditions. Data underlying panel A are provided in S2 Data. (B) A bar chart depicting the proportion of plants from (A) that flowered from the SAM or axillary meristems. The data for combinations between ful-2, mir172a-2, and mir172b-3 are supplemented by data from S12B Fig. Data for mir172abcd(e/+) are taken from Fig 3B and are shown here for comparison. The total number, N, of plants analyzed is indicated. Data underlying panel B are provided in S2 Data. (C) Schematic to summarize the interactions coordinated by SPL15 during the floral transition. Red lines with flat caps indicate negative regulation, and blue arrows indicate positive regulation. Gray text and line color indicate inactivity at that stage of development. Black text indicates activity at that stage of development. AP2-LIKE, APETALA2-LIKE; DTB, days to bolting; FUL, FRUITFULL; SAM, shoot apical meristem; SD, short-day; SPL, SQUAMOSA PROMOTER BINDING PROTEIN-LIKE.