Report

A peptide hormone required for Casparian strip diffusion barrier formation in Arabidopsis roots

See allHide authors and affiliations

Science  20 Jan 2017:
Vol. 355, Issue 6322, pp. 284-286
DOI: 10.1126/science.aai9057

Keeping roots water-tight

The Casparian strip provides a waterproofing function to plant roots, protecting them against unregulated influxes of water and minerals. The integrity of the Casparian strip depends on a receptor-like kinase. Doblas et al. and Nakayama et al. now identify the peptide ligands in the core of the root (the stele) that help regulate Casparian strip formation. The receptor is expressed on the outward-facing surface of the root endodermal cells that surround the stele. When the endodermal layer is sealed by the Casparian strip, the peptide ligands cannot reach their receptors. When the endodermal layer is breached, whether by damage or during development, the peptides reach their receptors and activate signaling that encourages lignin deposition, shoring up the strips.

Science, this issue p. 280, p. 284

Abstract

Plants achieve mineral ion homeostasis by means of a hydrophobic barrier on endodermal cells called the Casparian strip, which restricts lateral diffusion of ions between the root vascular bundles and the soil. We identified a family of sulfated peptides required for contiguous Casparian strip formation in Arabidopsis roots. These peptide hormones, which we named Casparian strip integrity factor 1 (CIF1) and CIF2, are expressed in the root stele and specifically bind the endodermis-expressed leucine-rich repeat receptor kinase GASSHO1 (GSO1)/SCHENGEN3 and its homolog, GSO2. A mutant devoid of CIF peptides is defective in ion homeostasis in the xylem. CIF genes are environmentally responsive. Casparian strip regulation is not merely a passive process driven by root developmental cues; it also serves as an active strategy to cope with adverse soil conditions.

Nutrient uptake by roots is a fundamental process for plant growth and development. Mineral ions enter into the cytoplasm of epidermal cells and the cortex or outer face of endodermal cells primarily via transporter-mediated pathways. These ions then move across the endodermis by symplastic transport through plasmodesmata and are ultimately secreted into the xylem for transfer to the shoot. Because nutrients often accumulate against a concentration gradient in the xylem vessels, vascular plants have evolved a ring like physical barrier that prevents passive apoplastic diffusion of ions and water across the endodermal cells that surround the vascular bundles. This hydrophobic barrier embedded within the walls of endodermal cells is called the Casparian strip (1).

The Casparian strip consists primarily of a hydrophobic, water-impermeable lignin polymer that seals the extracellular spaces between neighboring endodermal cells. Molecular and genetic screening studies have identified several key factors required for Casparian strip formation (26), including members of the Casparian strip membrane domain protein (CASP) family, which act as scaffolds (7), and the GASSHO1/SCHENGEN3 (GSO1/SGN3) leucine-rich repeat receptor kinase (LRR-RK) (8). Of these proteins, the LRR-RK GSO1/SGN3 expressed in root endodermal cells is of particular interest, as accumulating evidence suggests that LRR-RKs of subfamily XI act as receptors for small peptide hormones (9). Loss-of-function mutations in GSO1/SGN3 result in incorrect patch-like localization of the CASPs, leading to the formation of a repeatedly interrupted, discontinuous Casparian strip. This finding led to the hypothesis that peptide ligand-mediated GSO1/SGN3 receptor activation is crucial for contiguous Casparian strip formation in roots.

In most cases, genes encoding the precursors of small peptide hormones exist as multiple paralogous copies in the genome (9). Individual precursor polypeptides share a family-specific conserved domain close to the C terminus, from which mature functional peptide hormones are generated through posttranslational modification followed by proteolytic processing. In the course of screening novel peptide hormone candidates in Arabidopsis according to these empirical rules (10), we identified two closely related paralogous genes, At2g16385 and At4g34600, each of which encodes an ~80–amino acid polypeptide characterized by a conserved domain at the C terminus (fig. S1A). Phylogenetic analyses indicate that possible orthologs of these polypeptides are widely distributed among land plants (fig. S1B). Nano–liquid chromatography and tandem mass spectrometry analyses of the peptides secreted from Arabidopsis plants overexpressing At2g16385 determined that the mature form is a 21–amino acid tyrosine-sulfated peptide derived from the conserved domain (Fig. 1, A and B, and fig. S1, C and D).

Fig. 1 Structures and receptors of the mature At2g16385 and At4g34600 peptides.

(A) Structure of mature At2g16385 peptide (later named CIF1). O denotes hydroxyproline. (B) Sequence alignment of the mature peptide domains of At2g16385 and At4g34600. The residues conserved in both peptides are shaded in black. Amino acid abbreviations: D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; N, Asn; P, Pro; R, Arg; S, Ser; V, Val; Y, Tyr. (C) Exhaustive photoaffinity labeling using [125I]ASA-At2g16385 against membrane fractions derived from individual receptor kinase expression lines. (D) Competitive displacement of [125I]ASA-At2g16385 binding by 300-fold excess unlabeled peptide.

To identify receptors that directly interact with members of this peptide family, we chemically synthesized a cross-linkable derivative of At2g16385 in which photoactivatable 4-azidosalicylic acid (ASA) is incorporated into the 11th residue from the N terminus, a residue not conserved among paralogs (fig. S2A). After radioiodination, we performed an exhaustive binding assay by photoaffinity labeling against an Arabidopsis receptor kinase expression library (11) and found that two related LRR-RKs, GSO1/SGN3 and GSO2, in subfamily XI directly interact with the [125I]ASA-At2g16385 peptide (Fig. 1C and fig. S2B). This binding was competitively inhibited by excess unlabeled At4g34600 as well as At2g16385 but not by an unrelated peptide, RGF1 (12); such a result indicates that the At2g16385 and At4g34600 peptides act as specific ligands for GSO1/SGN3 and GSO2 receptor kinases (Fig. 1D).

β-Glucuronidase (GUS) reporter-aided histochemical analyses revealed At2g16385 promoter activity in the stele, especially at the phloem pole, of the mature region of the primary roots (Fig. 2A). No signal was detected in shoots. At4g34600 promoter activity was also confined to the root stele in the elongation and differentiation zones of both primary and lateral roots (Fig. 2B). GUS activity disappeared at the sites of lateral root initiation (Fig. 2B). In the root tip, the GUS signal became visible at and above approximately 10 cells after the onset of elongation (Fig. 2C).

Fig. 2 Promoter activities of the At2g16385 (CIF1) and At4g34600 (CIF2) genes.

(A) Promoter activity of the At2g16385 gene. From left to right: Whole view of a plant, a close view of the mature region of the primary roots, and root cross section. (B) Promoter activity of the At4g34600 gene. From left to right: Whole view, views at the sites of lateral root initiation and emergence, and root cross section. (C) Promoter activity of the At4g34600 gene in the root tip. Arrowhead marks the first elongating cortical cell. Scale bars, 1 mm [(A), left and center, and (B), left], 100 μm [(B), two central panels, and (C)], 20 μm [(A), right, and (B), right].

To determine whether loss of the At2g16385 and At4g34600 peptides phenocopies the gso1/sgn3 single or gso1/sgn3 gso2 double mutants in roots, we identified T-DNA insertion lines that carry mutations in the 5′-untranslated region of the At2g16385 gene and in the intron of the At4g34600 gene (Fig. 3A and fig. S3A). Although the At2g16385 At4g34600 ligand double mutant germinated normally on agar plates without ectopic adhesion of cotyledons caused by embryonic cuticle defects reported in gso1/sgn3 gso2 double mutants (13) (Fig. 3B and fig. S3, B and C), we observed that the mutant displayed an obvious defect in endodermal barrier formation in the roots, as visualized by penetration of the hydrophilic dye propidium iodide into the vasculature (Fig. 3C). This phenotype was comparable to that observed in the roots of the gso1/sgn3 gso2 receptor double mutant. The root defect in the ligand double mutant was restored when the mutant was treated with synthetic At2g16385 peptide. In contrast, the gso1/sgn3 gso2 receptor double mutant was completely insensitive to the peptide.

Fig. 3 Phenotypes of T-DNA insertion mutants of the At2g16385 (CIF1) and At4g34600 (CIF2) genes.

(A) Schematic representation of T-DNA insertion sites in the mutants. (B) Two-week-old wild-type and mutant plants. (C) Absence of endodermal diffusion barrier in the ligand and receptor mutants visualized by penetration of propidium iodide into the vasculature. Seedlings were treated with 100 nM At2g16385 (CIF1) peptide for 24 hours. (D) Casparian strip in the ligand and receptor mutants visualized by autofluorescence. (E) Quantitative analysis of the number of holes per 100 μm in the Casparian strips shown in (D). Data are means ± SD (*P < 0.05, Student’s t test, n = 4 or 5; ns, not significant; n.d., not detected). (F) Dose response of contiguous Casparian strip formation to peptide treatment for 24 hours. Significant differences are indicated by different letters (P < 0.05, one-way analysis of variance, n = 3 to 7). Data are means ± SD. (G) CASP1-GFP localization in wild-type and mutant plants. (H) Time-lapse imaging of the fusing dynamics of the Casparian strip domain in the cif1-1 cif2-1 mutant after treatment with CIF1 peptide. (I) Quantitative analysis of the number of holes in the Casparian strips after CIF1 treatment (n = 4 to 6). Data are means ± SD. (J) CASP1-GFP localization pattern in cif1-1 cif2-1 mutant after CIF1 deprivation. Scale bars, 5 mm (B), 50 μm (C), 10 μm [(D), (G), (H), and (J)].

We next directly observed the Casparian strip by lignin autofluorescence and confirmed that loss of either the ligands or the receptors resulted in the formation of a repeatedly interrupted, discontinuous Casparian strip (Fig. 3, D and E). Peptide treatment restored the defective Casparian strip in the ligand double mutant but not in the receptor double mutant. Dose dependence experiments demonstrated that both the At2g16385 and At4g34600 peptides were effective at concentrations as low as 1 nM (Fig. 3F). These results unambiguously indicate that the At2g16385 and At4g34600 peptides are critical for integrity of the Casparian strip, and we therefore named the peptides Casparian strip integrity factor 1 (CIF1) and CIF2, respectively.

A cif1-1 single mutant showed no detectable phenotype, but a cif2-1 mutant was defective in endodermal barrier formation (fig. S3, D to F). We confirmed that defects in the Casparian strip and rosette leaf size in the cif1-1 cif2-1 double mutant were fully complemented by a 3.2-kb DNA fragment containing the CIF2 gene (fig. S3, G and H). Additionally, we found that the gso1/sgn3 receptor single mutant that is defective in Casparian strip formation was still sensitive to the CIF1 peptide (fig. S3, I and J). Because the GSO2 transcript has been detected in roots (14), we concluded that, in addition to GSO1/SGN3, the GSO2 receptor kinase plays a certain—albeit minor—role in perception of CIF peptides in roots. The cif1-1 cif2-1 ligand double mutant, however, differs from the gso1/sgn3 gso2 receptor double mutant in that the former displays no cuticle defects in cotyledons, which suggests that additional ligands other than CIF family peptides may exist for GSO1/SGN3 and GSO2.

The Casparian strip defects in the cif1-1 cif2-1 double mutant were further characterized by live-cell imaging using a CASP1-GFP (CASP1 fused to green fluorescent protein) that had been used to visualize the Casparian strip domain (7). We observed that CASP1-GFP accumulated in discontinuous patches in the cif1-1 cif2-1 mutant (Fig. 3G), in a pattern similar to that reported for the gso1/sgn3 receptor mutant (8). Endogenous expression of the CASP1 and CASP2 genes was also down-regulated in the cif1-1 cif2-1 mutant (fig. S4A). When treated with CIF1 peptide, the discontinuous CASP1-GFP patches fused into contiguous bands comparable to those in wild-type plants, accompanied by up-regulation of CASP1 and CASP2 expression (Fig. 3, H and I, and fig. S4B). This fusing response was detected after 5 hours of peptide treatment and completed by 6 to 9 hours (Fig. 3, H and I). We also observed that when the cif1-1 cif2-1 double mutant that had been treated with CIF1 peptide for 24 hours was transferred to new medium devoid of the peptide, small breaks occasionally formed in the CASP1-GFP bands 24 hours after transfer, accompanied by reduced expression of CASP1 and CASP2 (Fig. 3J and fig. S4, C and D). These results suggest that the CIF peptides are required for formation and maintenance of the contiguous Casparian strip.

The Casparian strip is thought to play roles in environmental adaptation by acting as a physical barrier that prevents unfavorable inward and outward leakage of ions between the xylem and the soil (15). Under nutrient-limiting conditions, for example, the gso1/sgn3 receptor mutant exhibits symptoms of potassium deficiency (8). In the course of testing the response of the cif1-1 cif2-1 double mutant to excess essential minerals, we found that this mutant is highly sensitive to excess iron (Fig. 4A). Although wild-type Arabidopsis plants tolerated excess iron at concentrations as high as 500 μM, the cif1-1 cif2-1 double mutant showed a noticeable growth defect at 300 μM and exhibited severely stunted growth accompanied by bronzing of the leaves at 500 μM after 7 days of treatment. Excessive intake of iron by plants has been reported to induce the production and accumulation of toxic reactive oxygen species, causing damage to the plants (16). These growth defects were fully restored by supplementation of the medium with CIF1 peptide (Fig. 4A). These results indicate that mutants exhibiting defects of the Casparian strip cannot adapt to fluctuating iron levels.

Fig. 4 The cif1-1 cif2-1 double mutant is hypersensitive to excess iron.

(A) Growth of wild-type and mutant plants under excess iron conditions for 7 days in the presence or absence of CIF1 peptide. (B) Comparison of iron levels in the culture medium, wild-type xylem sap, and mutant xylem sap, as determined by synchrotron radiation x-ray fluorescence spectrometry. (C) Expression levels of CIF1 and CIF2 in wild-type plants treated with excess mineral ions for 24 hours at pH 5.5 or 4.5. Data are means ± SD (*P < 0.05, Student’s t test, n = 3); CTL, control. (D) Promoter activity of CIF2 in roots treated with 75 μM iron at pH 5.5 as a control or 500 μM iron at pH 4.5 for 24 hours. Scale bars, 5 mm (A), 100 μm (D).

To confirm whether excess iron intake into the xylem occurred in the mutants, we collected xylem sap from the hypocotyl of decapitated Arabidopsis plants and analyzed the mineral ion concentrations by synchrotron radiation x-ray fluorescence spectrometry. At the normal iron concentration (75 μM), the iron content in the xylem sap of the cif1-1 cif2-1 mutant was virtually the same as that of the wild type (Fig. 4B). In contrast, the iron level in the xylem sap of the cif1-1 cif2-1 mutant cultured under excess iron conditions was considerably higher than that of the wild type, probably as a result of inward leakage. Notably, we found that expression of both CIF1 and CIF2 was up-regulated by excess iron and was further synergistically regulated by lowering the medium pH (Fig. 4, C and D, and fig. S5A). Because iron is more soluble in acidic soils than in neutral soils, this response may reflect an adaptation to ensure the growth and survival of the plants under unfavorable mineral conditions. In addition, we found that the cif1-1 cif2-1 mutant exhibited retarded growth under low-potassium conditions, and the potassium level in the xylem sap of the mutant was lower than that of the wild type, probably because of concentration-dependent outward leakage (fig. S5, B and C). These findings suggest that Casparian strip mutants are defective in ion homeostasis in the xylem associated with inward or outward leakage of ions (depending on the ionic concentration gradient across the endodermis cell layer), leading to pleiotropic phenotypes in unfavorable mineral environments.

As sessile organisms, plants have evolved sophisticated machinery to acquire nutrients without interference from a complex array of fluctuating environmental conditions. The Casparian strip, which forms an essential diffusion barrier between the roots and soil environment, is one such example. Our biochemical and physiological analyses revealed that the stele-expressed peptide hormones CIF1 and CIF2 mediate the formation and maintenance of a functional, contiguous Casparian strip by activating endodermis-expressed GSO1/SGN3 and GSO2 receptor kinases. Given that CIF peptides are environmentally responsive, Casparian strip regulation is likely not just a passive process driven by root developmental cues but also serves as an active strategy to cope with adverse soil conditions.

Supplementary Materials

www.sciencemag.org/content/355/6322/284/suppl/DC1

Materials and Methods

Figs. S1 to S5

Table S1

References (1724)

References and Notes

  1. Acknowledgments: Supported by Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Scientific Research (S) 25221105 and Grant-in-Aid for Scientific Research on Innovative Areas 15H05957 (Y.M.) and by JSPS Grants-in-Aid for Scientific Research on Innovative Areas 26113520 and 16H01234 and Grant-in-Aid for Young Scientists (B) 25840111 (H.S.). The x-ray fluorescence spectrometry experiments were performed at the BL5S1 of Aichi Synchrotron Radiation Center. The supplementary materials contain additional data.
View Abstract

Stay Connected to Science

Navigate This Article