Environmental regulation of lateral root emergence in Medicago truncatula requires the HD-Zip I transcription factor HB1

Abstract

The adaptation of root architecture to environmental constraints is a major agricultural trait, notably in legumes, the third main crop worldwide. This root developmental plasticity depends on the formation of lateral roots (LRs) emerging from primary roots. In the model legume Medicago truncatula, the HD-Zip I transcription factor HB1 is expressed in primary and lateral root meristems and induced by salt stress. Constitutive expression of HB1 in M. truncatula roots alters their architecture, whereas hb1 TILLING mutants showed increased lateral root emergence. Electrophoretic mobility shift assay, promoter mutagenesis, and chromatin immunoprecipitation-PCR assays revealed that HB1 directly recognizes a CAATAATTG cis-element present in the promoter of a LOB-like (for Lateral Organ Boundaries) gene, LBD1, transcriptionally regulated by auxin. Expression of these genes in response to abscisic acid and auxin and their behavior in hb1 mutants revealed an HB1-mediated repression of LBD1 acting during LR emergence. M. truncatula HB1 regulates an adaptive developmental response to minimize the root surface exposed to adverse environmental stresses.

Figure 1.

HB1 Encodes an HD-Zip I TF Regulated by ABA, NaCl, and Mannitol. (A) HB1 gene structure and promoter region highlighting the location of the ABA response elements (ABREs) and drought response elements (DREs). TSS, transcription start site. Arrows indicate position of hb1-1 and -2 alleles obtained through TILLING and the nucleotide substitutions. Numbers indicate distances from the transcription start site. (B) Protein alignment of HB1 with Arabidopsis HB7 and 12. Identity is indicated on the left side as a percentage relative to the HB1 sequence used as reference. The alignment was performed with MAFFT 6.0 (Katoh and Toh, 2008) and the visualizing tool MView (Brown et al., 1998). The hb1-1 and -2 mutations obtained by TILLING are indicated using arrowheads, and the protein sequence modifications are shown below the alignment. (C) Real-time RT-PCR analysis of HB1 expression in response to different concentrations of NaCl, ABA, and mannitol. Bars express the fold change on a linear scale. H3L was used as reference gene, and ratios were normalized against the control (nontreated) condition. Error bars represent sd of four biological replicates.

Figure 2.

Plants Affected in HB1 Function Exhibit Perturbed Root Architecture. (A) Two representative composite M. truncatula plants transformed with 35S:GUS and 35S:HB1 constructs after 4 weeks of growth in sand/perlite (greenhouse). (B) and (C) Root dry weight per plant (B) and per centimeter (C) of primary root was quantified under control and salt stress (100 mM NaCl) conditions. The letters indicate mean values significantly different between groups (Kruskal-Wallis test P < 0.05, n > 20 in every case). (D) Primary root length was measured for wild-type and hb1-1 mutant plants. (E) to (G) Quantification of LR initiation and emerged LRs per centimeter of primary root in hb1-1 mutants and the wild type. (E) Total number of initiated LR per centimeter in plants growing in nitrogen-containing medium. (F) and (G) Percentages of emerged LR per centimeter of primary roots in the wild type and hb1-1 mutants in the same medium (F) and 7 d after inoculation with S. meliloti in a nitrogen-poor medium (symbiotic conditions) (G). In (B) to (F), n > 20 per construction and condition per experiment. The asterisks indicate mean values significantly different between the wild type and hb1 mutants (for [D] to [G], Student’s t test, P < 0.01, error bars represent sd between biological replicates). (H) Segregation analysis of the hb1-2 mutation. The number of emerged LRs per centimeter was determined in heterozygous, homozygous mutant, and wild-type sibling plants from a 3-week-old segregating hb1-2 population (n = 12 homozygous, 20 heterozygous, and 10 wild-type siblings; Kruskal-Wallis test, P < 0.05). Letters indicate mean values significantly different between groups.

Figure 3.

HB1 Regulates LR Emergence in M. truncatula. Total initiated LR per centimeter of primary root (A) and percentage of emerged roots (relative to the total number of initiated roots; [B]) for the wild type and hb1-1 mutants in presence of different concentrations of ABA and NaCl. In all treatments, the percentage of emergence in hb1-1 mutants was similar (n > 20 per experiment; Kruskal-Wallis test, P < 0.05). The letters indicate mean values significantly different between groups. Error bars represent sd between three biological replicates.

Figure 4.

The LOB-Like LBD1 TF Gene Is Transcriptionally Repressed by HB1. (A) Real-time RT-PCR analysis of HB1 (right panel) and LBD1 (left panel) transcript levels in hb1-1, 35S:HB1, and wild-type roots. (B) Real-time RT-PCR analysis of LBD1 expression in wild-type and hb1-1 roots transformed or not with a 35S:HB1 construct. Bars in (A) and (B) express the fold change on a log2 scale. The H3L gene was used as reference gene, and ratios were normalized against the reference genotype (wild type). Error bars represent sd for three biological replicates. (C) HB1 expression pattern in different stages of lateral root primordia analyzed by detecting GUS activity in roots transformed with a ProHB1:GUS construct. ep, epidermis; c, cortex; end, endodermis; pe, pericycle; st, stele. (D) Comparative expression patterns of HB1 (top row) and LBD1 (bottom row) promoters during LR formation. The low level of expression of LBD1 in transgenic root carrying ProLBD1:GUS fusions is noteworthy. LRP, LR primordia. Bars = 250 μm. (E) Histological sections of ProHB1:GUS and ProLBD1:GUS emerged LR primordia are shown. Bars = 125 μm.

Figure 5.

The LBD1 Promoter Contains a Specific cis-Acting Element Required for HB1 Regulation. (A) Scheme of the LBD1 promoter region and gene structure highlighting the location of the cis-acting element CAATAATTG. TSS, transcription start site. (B) Analysis of ProLBD1:GUS activity in wild-type and hb1-1 roots (top and middle rows) and the effect of site-directed mutagenesis of the cis-element on the LBD1 promoter activity (Pro-MUT-LBD1) in wild-type roots (CAATAATTG was replaced by CAGGATCCG; bottom row). Bars = 250 μm. (C) ChIP-PCR analysis using LBD1 primers spanning the promoter region containing the specific cis-acting element required for HB1 regulation. The indicated results were obtained from samples precipitated with α-HA antibody. Lane 1, molecular weight marker; lane C+, LBD1 BAC clone positive control; lane C−, water negative control. ChIP-PCR assays were done on 3-week composites plants untransformed (A17) or transformed with a ProHB1:HB1:2XFLAG:2XHA construct (HA). INPUT, total nuclei sample before immunoprecipitation; ChIP-PCR, after immunoprecipitation with HA antibodies. Two independent clones expressing the HB1-HA fusion but not the A17 control give a positive signal for the LBD1 promoter.

Figure 6.

Morphological and Molecular Responses of Wild-Type Plants to Auxin, ABA, and Salt Treatments. (A) Representative plants showing the long-term physiological effects of auxin (IAA), ABA, and salt on M. truncatula root architecture. Plants (3 d old) were grown for 6 d in plates supplemented with ABA, IAA, NaCl individually, or in combination. Bars = 1 cm. (B) and (C) Total LR initiation per centimeter of primary root (B) and percentage of LR emergence (C) in wild-type M. truncatula plants after 6 d of the different treatments specified by each color. The asterisks indicate that for the short-term molecular studies, 100 μM ABA was applied (n > 20 per condition; Kruskal-Wallis test, P < 0.05). The letters indicate mean values significantly different between groups. Error bars are sd between biological replicates. (D) and (E) Real-time RT-PCR analysis of expression levels (similar barcode as [B] and [C]) for HB1 (D) and LBD1 (E) in wild-type plants in response to short-term IAA, ABA, or salt treatments (3 h). Bars express the fold change on a linear and a log2 scale, respectively. Error bars are sd in three biological replicates. Arrows indicate the most relevant results referred to in the text. (F) Proposed role for HB1 in LR emergence. Three stages for LR formation (initiation, primordial formation, and emergence) are schematized. HB1 represses LBD1 expression in the LR primordia. When M. truncatula is exposed to salt stress or high ABA concentrations, HB1 controls LR emergence, likely by repressing LBD1. HB1 mediates a developmental plasticity response to adapt root architecture to environmental constraints.