Comparative mutant analyses reveal a novel mechanism of ARF regulation in land plants

Abstract

The plant hormone auxin regulates a wide variety of transcriptional responses depending on the cell type, environment and species. How this diversity is achieved may be related to the specific complement of auxin-signalling components in each cell. The levels of activators (class-A AUXIN RESPONSE FACTORS) and repressors (class-B ARFs) are particularly important. Tight regulation of ARF protein levels is probably key in determining this balance. Through comparative analysis of novel, dominant mutants in maize and the moss Physcomitrium patens, we have discovered a ~500-million-year-old mechanism of class-B ARF protein-level regulation mediated by proteasome degradation, important in determining cell fate decisions across land plants. Thus, our results add a key piece to the puzzle of how auxin regulates plant development.

Fig. 1. Maize Truffula mutants have pleiotropic defects.

a–d, Normal (+/+) and Trf siblings (Trf/+, Trf/Trf) in the W22 (a,c) and Mo17 (b,d) inbred backgrounds. a,b, Mature plants. c, Midrib phenotypes in Trf leaves: midribless (i), normal (ii) and multiple midribs (iii). Scale bars, 1 cm. d, Tassel phenotypes showing ear-like traits in Trf siblings. Scale bar, 5 cm. e, Leaf number in mature plants of normal (blue) and Trf siblings (orange). **P < 0.05 determined by Kruskal–Wallis non-parametric test followed by Wilcox pairwise tests, n ≥ 6 plants. Black dot and error bars, mean ± s.d.

Fig. 2. The Truffula lesion maps to ZmARF28.

a, Plotting WGS-BSA variant mean allelic distances in 0.5-Mbp windows across the maize NAM5.0 reference genome reveals a region on chromosome 10 (Chr10) which is more associated with Trf mutants than normal siblings (red box). n = 80 per genotype. Dot and error bars, mean ± s.e.m. Colour scale: mean allelic distance −0.25:dark blue to 0.45:red. b, PCR mapping narrows the region to between 13.5 Mbp and 14.0 Mbp. Mapping primer positions are indicated (AD_# and umc#), numbers represent Mbp position on Chr10, and recombinant numbers are shown in brackets. There are 8 genes in the interval, of which 5 have an FPKM > 1 in our shoot apical meristem RNA-seq dataset (green) and 3 have an FPKM < 1 (purple). A moderate effect, Trf-specific, EMS-type SNP (*) is present in Zm00001eb408800 (ZmARF28). c, Locations of the Trf, narD72, narD120, narC29 and narC18 mutations (red arrows). The exons (blue), 5’ and 3’ UTRs (grey), CRISPR sgRNA targets for deletions (magenta arrowheads), and small RNA regulatory sites (cyan) are indicated. d, Generalized cartoon depiction of the protein domains in class-B ARFs, with the DBD, B3 and PB1 domains indicated; the relative location of the mutations identified in this study are indicated in red. e,f, The maize Trf mutation and the moss narD72, narD120, narC29 and narC18 mutations map to the same loop region in the ARF protein. e, Multiple sequence alignments of ARFs from maize, Arabidopsis and moss. f, Homology modelling of the ZmARF28, PpARFb2 and PpARFb4 DBDs for both wild-type and mutant proteins (metrics in Supplementary Table 10). (i)–(iii), Zoomed-in images of the green-boxed regions. e,f, Mutation locations (red) and amino acids previously shown to be important for ARF function (blue) are indicated.

Fig. 3. Mutations in PpARFb2 and PpARFb4 cause auxin resistance.

a, Comparison of wild type, nar and ∆pparfb4 deletion mutants after 21 days of growth on standard medium (BCD) and BCD supplemented with 5 mM ammonium tartrate or 5 µM NAA. Plants grown on NAA were imaged for RFP fluorescence from the proDR5:DsRed auxin-response reporter. On NAA-containing media, wild-type and ∆pparfb4 strains produce ectopic rhizoids and high levels of RFP fluorescence, whereas the nar mutants produce green leafy gametophores. Arrowheads indicate leafy shoots in narD120 and ∆pparfb4. Scale bar, 5 mm. b, Comparison of wild type and gene-edited point mutants grown for 21 days, as in a. c, Expression of select auxin-responsive genes (PpRSL6, ROOT HAIR DEFECTIVE SIX-LIKE 6; PpIAA2, Aux/IAA2; PpHSL1-2, HOOKLESS1-LIKE 2) in wild type, pparfb2^E266K and auxin-response mutant ppiaa2^G325S grown for 1 h on media containing 10 µM IAA or 0.01% ethanol (3 technical replicates). Bars represent mean ± s.e.m. Transcript levels were normalized to the APT gene. d, Growth comparisons of wild type, the pparfb2^E266K mutant and the pparfb2^E266K mutant with second-site mutations, as in a. brd–, L614A + F615A substitutions in the B3 repression domain; pb1K→A, K682A substitution in the PB1 domain; ∆pb1, a 5-bp frameshift-causing deletion starting at G680 codon; dbd–, P194A + R196A substitutions affecting DNA-interacting residues. n ≥ 4. Scale bars, 5 mm.

Fig. 4. Mutations in the same domain result in stabilization of ZmARF28, PpARFb and AtARF2.

a,b, PpARFb2-YFP signal in P. patens wild type, pparfb2^R269Q, pparfb2^E266K and the pparfb2^E226K,R269Q double mutants. a, Representative confocal microscopy images. Yellow, PpARFb2-YFP; magenta, chlorophyll autofluorescence. Scale bar, 25 µm. b, Nuclear fluorescence quantification. Centre black dot and error bars, mean ± s.d.; **P < 0.01, determined using Kruskal–Wallis non-parametric test followed by Wilcox pairwise tests. c, Western blot of ZmARF28 in +/+ and Trf/+ mutant siblings from a W22 backcross population. Lanes are different biological replicates. Anti-tubulin was used as a loading control. d, Quantification of ZmARF28 abundance relative to TUBULIN. **P < 0.01, Welch 2-sample t-test. Centre black dot and error bars, mean ± s.d.; 3 technical replicates, 2 biological replicates in each. e, Western blots of ZmARF28-GFP accumulation in N. bethamiana, with (+) and without (−) MG132 treatment, n = 3. f, Mean ± s.e.m. ratiometric signal of AtARF2-mNeonGreen (blue) and AtARF2-Trf-mNeonGreen (orange) in Arabidopsis protoplasts, relative to mScarlet after cycloheximide (CHX) treatment (≥2,166 cells across 6 technical replicates analysed). g, Western blots of anti-GFP immunoprecipitated samples (αGFP IP) and cleared extracts (input) using anti-GFP and anti-ubiquitin antibodies. Extracts were from bortezomib-treated untransformed control (‘C’) and stable lines expressing YFP-fused ARFb2’s DBD with and without the E266K + R269Q substitutions. Mutant IP samples were diluted 5-fold to normalize DBD-YFP amounts loaded. n = 3. Source data