Evolutionary Conservation of ABA Signaling for Stomatal Closure

Abstract

Abscisic acid (ABA)-driven stomatal regulation reportedly evolved after the divergence of ferns, during the early evolution of seed plants approximately 360 million years ago. This hypothesis is based on the observation that the stomata of certain fern species are unresponsive to ABA, but exhibit passive hydraulic control. However, ABA-induced stomatal closure was detected in some mosses and lycophytes. Here, we observed that a number of ABA signaling and membrane transporter protein families diversified over the evolutionary history of land plants. The aquatic ferns Azolla filiculoides and Salvinia cucullata have representatives of 23 families of proteins orthologous to those of Arabidopsis (Arabidopsis thaliana) and all other land plant species studied. Phylogenetic analysis of the key ABA signaling proteins indicates an evolutionarily conserved stomatal response to ABA. Moreover, comparative transcriptomic analysis has identified a suite of ABA-responsive genes that differentially expressed in a terrestrial fern species, Polystichum proliferum These genes encode proteins associated with ABA biosynthesis, transport, reception, transcription, signaling, and ion and sugar transport, which fit the general ABA signaling pathway constructed from Arabidopsis and Hordeum vulgare The retention of these key ABA-responsive genes could have had a profound effect on the adaptation of ferns to dry conditions. Furthermore, stomatal assays have shown the primary evidence for ABA-induced closure of stomata in two terrestrial fern species Pproliferum and Nephrolepis exaltata In summary, we report, to our knowledge, new molecular and physiological evidence for the presence of active stomatal control in ferns.

Figure 1.

Number of predicted membrane transporters and ABA reception complex proteins and families in different taxa. A and B, The number of predicted transporters and ABA reception families (A) and guard cell genes (B) are grouped into seven taxa. The GENESIS (http://genesis-sim.org/) simulation environment was used to estimate the similarity among protein sequences (A) and family (B). Candidate protein sequences were selected by BLASTP (NCBI) software searches that satisfied the criteria of E-value < 10⁻¹⁰ and query coverage > 50% for guard cell proteins and E-value < 10⁻⁵ for all proteins. C, The number of gene families in different taxa. Data are mean ± se for those with error bars.

Figure 2.

Similarity heat map for the evolution of membrane transporters and ABA reception complex proteins in different species. A and B, the GENESIS simulation environment was used to estimate the similarities among protein family (A) and sequences (B). Candidate protein sequences were selected by BLASTP software searches that satisfied the criteria of E-value < 10⁻⁵ only (A) and E-value < 10⁻¹⁰ and query coverage > 50% (B). Colored squares indicate protein sequence similarity from zero (yellow) to 100% (red). Clades are indicated by different colors on the right: angiosperm (blue), gymnosperm (pink), fern (red), lycophyte (green), moss (brown), liverwort (purple), and algae (orange). Gray squares indicate that no proteins were found that satisfied the selection criteria. KATs represent the AKTs/KATs/GORKs proteins. ABCC, ATP-binding cassette C transporter; ACA, autoinhibited Ca²⁺-ATPase; AHA, Arabidopsis plasma membrane H⁺-ATPase; AKT, Arabidopsis inwardly rectifying K⁺ channel; ALMT, aluminum-activated malate transporter; AVP, Arabidopsis vacuolar H⁺-pyrophosphatase; CAX, cation proton exchanger; CLC, chloride channel; CNGC, cyclic nucleotide gated channel; GORK, guard cell outwardly rectifying K⁺ channel; GLR, Glu receptor-like Ca²⁺ channel; GORK, gated outwardly rectifying K⁺ channel; HAK, high-affinity K⁺ transporter; HKT, high-affinity K⁺/Na⁺ transporter; KAT, guard cell inwardly rectifying K⁺ channel; NHX, Na⁺/H⁺ antiporter; TPK, tonoplast K⁺ channel; OST1, open stomatal1; PIP, plasma membrane intrinsic protein; PP2C, Protein Phosphatase 2C; RCAR, regulatory component of ABA receptor; SLAC, slow anion channel; SnRK2, SNF1-related protein kinase2; SUC, Suc transporter; TIP, tonoplast intrinsic protein; TPC, two-pore channel; TPK, two-pore channel, a guard cell membrane transporter; VHA, vacuolar H⁺-ATPase.

Figure 3.

Phylogenetic trees of key ABA signaling proteins in species of plants and algae. A to E, The regulatory component of ABA receptor 11 (RCAR11; A), ABA insensitive1 (ABI1; B), open stomatal 1 (OST1; C), slow anion channel1 (SLAC1; D), and vacuolar H⁺-pyrophosphatase1 (AVP1; E) were estimated. The maximum-likelihood method was used to construct the trees and evolutionary distances were computed in the software MEGA 6. Bootstrap values are shown next to each branch of the trees. Clades are indicated by different colors: angiosperm (blue), gymnosperm (pink), fern (red), lycophyte (green), moss (brown), liverwort (purple), and algae (orange). Af, Azolla filiculoides; Amt, Amborella trichopoda; At, Arabidopsis thaliana; Bd, Brachypodium distachyon; Br, Brassica rapa; Cm, Cyanidioschyzon merolae; Cr, Chlamydomonas reinhardtii; Eg, Eucalyptus grandis; Es, Ectocarpus siliculosus; Gm, Glycine max; Gr, Gossypium raimondii; Hv, Hordeum vulgare; Kf, Klebsormidium flaccidum; Md, Malus domestica; Mp, Marchantia polymorpha; Mt, Medicago truncatula; Os, Oryza sativa; Oss, Ostreococcus sp.; Pa, Picea abies; Ph, Phyllostachys heterocycla; Pl, Pinus lambertiana; Pot, Populus trichocarpa; Pp, Physcomitrella patens; Pt, Pinus taeda; Py, Porphyra yezoenessi; Sb, Sorghum bicolor; Sc, Salvinia cucullata; Sf, Sphagnum fallax; Sl, Solanum lycopersicum; Sm, Selaginella moellendorffii; Sp, Spirodela polyrhiza; Ta, Triticum aestivum; Th, Theobroma cacao; Vc, Volvox carteri; Vv, Vitis vinifera; Zm, Zea mays.

Figure 4.

Relative expression of ABA signaling genes in epidermal peels of P. proliferum. Forty genes involved in the ABA signaling pathway were classified into five groups by function. Relative expression levels were calculated and normalized. Data are mean ± se. Significant up- or down-regulation is indicated with an asterisk at P < 0.05.

Figure 5.

Conserved ABA signaling pathway in epidermis of H. vulgare and P. proliferum. The signaling network is formed by four main functional categories: ABA biosynthesis (pink), ABA transportation (green), signal transduction (purple), and the downstream ion channels and transporters (light blue). Heatmap of transcriptome of H. vulgare epidermis in drought stress (left and middle) and qPCR of P. proliferum epidermis in ABA treatment (right) are shown. ABCC, ATP-binding cassette C transporter; ABCG, ATP-binding cassette G; ABF, ABA responsive elements-binding factor; ABI4, ABA insensitive4; AHA, Arabidopsis plasma membrane H⁺-ATPase; AKT1, Ser/Thr kinase1; ALMT, aluminum-activated malate transporter; AREB, AREB-like protein; AVP, Arabidopsis vacuolar H⁺-pyrophosphatase; CAS, calcium sensing receptor; CHLH, protoporphyrin IX magnesium chelatase, subunit H; CLC-C, chloride channel C; CNGC, cyclic nucleotide gated channel; CYP707A, cytochromeP450 family 707 superfamily A; GORK, gated outwardly rectifying K⁺ channel; GTG, GPCR-type g protein1; KAT, guard cell inwardly rectifying K⁺ channel; MAPK, mitogen activated kinase-like protein; MYB, MYB domain protein; NADPBRF, NAD(P)-binding Rossmann-fold superfamily protein; NCED3, 9-cis-epoxycarotenoid dioxygenase6; OST1, open stomata1; PLDa1, phospholipase Dα1; PP2C, Protein Phosphatase 2C; RBOH, respiratory burst oxidase homolog protein; RCAR, regulatory component of ABA receptor; SLAC, slow anion channel; STP1, sugar transporter1; VHA, vacuolar H⁺-ATPase; ZEP, zeaxanthin epoxidase.

Figure 6.

Stomatal responsiveness to ABA and CaCl₂ in ferns. A to D, Stomatal aperture of P. proliferum (A) and N. exaltata (B) in the control, 50 μm ABA, and 10 mm CaCl₂ treatments at 0, 60, and 120 min. B, ABA dose-dependent (0.1, 1, 10, 50, and 200 μm ABA) stomatal closure in P. proliferum (C) and N. exaltata (D). Data are mean ± se (n = 30–40 stomata from five biological replicates). Asterisks indicate significant difference at P < 0.01 level.

Figure 7.

ABA-induced stomatal closure in ferns. A and B, Stomatal aperture of P. proliferum (A) and N. exaltata (B) in the control (0–20 min) and 50 μm ABA (20–120 min) treatment. Blank (control) stomatal aperture measurements were also conducted for 120 min in both species. Data are mean ± se (n = 30–40 stomata from five biological replicates).

Figure 8.

ABA-induced stomatal guard cell volume changes in ferns. A, Representative confocal images of open, half-open, and closed stomata in P. proliferum. Green and red fluorescence indicate the ROS and endomembrane stain, respectively. Bars = 20 μm. B and C, Stomatal aperture (B) and estimated vacuole volume (C) of open, half-open, and closed stomata in P. proliferum and N. exaltata. Data are mean ± se (n = 6–10 stomata from three biological replicates). D, Correlation between stomatal aperture and estimated vacuole volume. Data are plotted from all the measured stomata in confocal imaging.