Glycosylation of a Fasciclin-Like Arabinogalactan-Protein (SOS5) Mediates Root Growth and Seed Mucilage Adherence via a Cell Wall Receptor-Like Kinase (FEI1/FEI2) Pathway in Arabidopsis

Abstract

Fundamental processes that underpin plant growth and development depend crucially on the action and assembly of the cell wall, a dynamic structure that changes in response to both developmental and environmental cues. While much is known about cell wall structure and biosynthesis, much less is known about the functions of the individual wall components, particularly with respect to their potential roles in cellular signaling. Loss-of-function mutants of two arabinogalactan-protein (AGP)-specific galactosyltransferases namely, GALT2 and GALT5, confer pleiotropic growth and development phenotypes indicating the important contributions of carbohydrate moieties towards AGP function. Notably, galt2galt5 double mutants displayed impaired root growth and root tip swelling in response to salt, likely as a result of decreased cellulose synthesis. These mutants phenocopy a salt-overly sensitive mutant called sos5, which lacks a fasciclin-like AGP (SOS5/FLA4) as well as a fei1fei2 double mutant, which lacks two cell wall-associated leucine-rich repeat receptor-like kinases. Additionally, galt2gal5 as well as sos5 and fei2 showed reduced seed mucilage adherence. Quintuple galt2galt5sos5fei1fei2 mutants were produced and provided evidence that these genes act in a single, linear genetic pathway. Further genetic and biochemical analysis of the quintuple mutant demonstrated involvement of these genes with the interplay between cellulose biosynthesis and two plant growth regulators, ethylene and ABA, in modulating root cell wall integrity.

Fig 1. Transcript profiling of GALT2, GALT5, SOS5, FEI1, and FEI2 genes in different organs/tissues and in response to salt treatment.

(A) qRT-PCR analysis of the five genes in different plant organs/tissues. Expression levels are the mean ± SE of three technical replicates relative to the Ubiquitin 10 (UBQ10) reference gene. The expression value of GALT2 in leaf was considered as 1. (B) Expression analysis of the five genes during the course of seed development. Total RNA was isolated from seeds 4, 7, and 11 DPA. Relative expression was normalized using GAPC as a reference gene. The expression value of GALT2 at 11 DPA was considered as 1. C. Induction of transcript abundance of GALT2, GALT5, SOS5, FEI1, and FEI2 in response to salt treatment. Total RNA was extracted from root tips of 250 WT and mutant seedlings grown in MS media and in MS media supplemented with 100 mM NaCl for 7d. Experiments in (B) and (C) are the mean ± SE of three technical replicates.

Fig 2. Quintuple mutants exhibit reduced root elongation in response to NaCl and sucrose.

(A) WT and mutant seedlings were grown on MS medium containing 1% sucrose for 5 d and then transferred to MS medium containing 100 mM NaCl. (B) WT and mutant seedlings were grown on MS medium containing 0% sucrose for 5 d and then transferred to MS medium containing 4.5% sucrose. Quantification of root elongation after transfer to non-permissive conditions was recorded after 7, 14, and 21 d. Data are means ± SE; n ≥ 25. Different letters indicate statistically significant differences (P < 0.05) between means.

Fig 3. Quintuple mutants display a conditional root anisotropic growth defect (i.e., root swelling).

(A) Root tips of WT and mutant seedlings four days after transfer from MS medium containing 1% sucrose to MS medium containing 100 mM NaCl as imaged by dark field microscopy. (B) Root tips of WT and mutant seedlings four days after transfer from MS medium containing 0% sucrose to MS medium containing 4.5% sucrose as imaged by dark field microscopy. (C) Graphical representation of root tip width in response to 100 mM NaCl. (D) Graphical representation of root tip width in response to 4.5% sucrose. The stripped bars indicate seedlings grown on control unsupplemented MS plates. Quantitation of root tip width was measured at the level of the youngest root hair using ImageJ software. Values are the means (n>15) ± SE. Different letters indicate statistically significant differences (P < 0.05) between means.

Fig 4. Hypersensitivity of the mutants towards isoxaben.

Root tips from WT and mutant seedlings were germinated and grown for five days on MS medium with 1% sucrose and transferred for 72 hours to MS medium supplemented with isoxaben at various concentrations. Root tips were imaged using dark field microscopy. Scale bar = 1 mm.

Fig 5. Ectopic lignin deposition in the mutants.

WT and mutant seedlings were grown on MS medium containing 0% sucrose for 5 d and then transferred to (A) MS medium containing 100 mM NaCl for 7 d or to (B) MS medium containing 4.5% sucrose for 7 d. Scale bar = 1 mm.

Fig 6. Quintuple mutants display a cellulose deficiency in response to salt.

(A) and (B) Incorporation of [¹⁴C]Glc into acid-soluble and acid-insoluble fractions from excised root tips from WT and mutant seedlings in response to 100 mM NaCl, (C) and (D) in response to 4.5% sucrose and (E) and (F) in response to AIB. Seedlings were grown on 0% sucrose for 5 d and then transferred to the three different treatment conditions for 7d. (A), (C), and (E) indicate incorporation of [¹⁴C]Glc into the acid soluble fraction, and (B), (D), and (F) indicate incorporation of [¹⁴C]Glc into the acid insoluble fraction. Values are means ± SD from two biological replicates, and the experiments were repeated at least two times with similar results. Different letters indicate statistically significant differences (P < 0.05) between means.

Fig 7. Role of ethylene and ABA on the conditional root swelling phenotype.

Images of WT and mutant seedlings grown on MS medium containing 0% sucrose for 4 d and transferred to MS medium containing 100 mM NaCl supplemented with AIB, ABA, AgNO₃, CoCl₂, and AVG for 7 d as indicated. Scale bar = 1 mm.

Fig 8. Aberrant adherent mucilage structure in the mutants as depicted by ruthenium red staining.

Ten milligrams of WT and mutant seeds were hydrated in water and occasionally shaken prior to ruthenium red staining. Scale bar = 0.2 mm.

Fig 9. Mutants display reduced cellulosic rays in seed coat mucilage.

(A) Calcofluor and Pontamine scarlet red staining of cellulosic rays of WT and mutant seeds was performed following gentle shaking in water. (B) Measurements of cellulosic ray length in WT and mutant seeds. Different letters indicate statistically significant differences (P < 0.05) between means. Error bars indicate SE of three replicates. Scale bars = 50 μm.

Fig 10. Impaired cellulosic rays in the mutants.

Crystalline cellulose content determined from WT and mutant seeds. Similar results were obtained in two biological repeats. Different letters indicate statistically significant differences (P < 0.05) between means.

Fig 11. Proposed model linking GALT2 and GALT5 with SOS5 and FEI1/FEI2 in regulating cellular signaling of root growth.

Signaling of normal root growth may involve GALT2/GALT5-dependent glycosylation of SOS5 and glycosylated SOS5 binding FEI1/FEI2, possibly inducing dimer formation and activation of the kinase domain as well as allowing for the binding of FEI1/FEI2 to ACC synthase. Such binding will inhibit the production of ACC, a potential signaling molecule and ethylene precursor, which directly or indirectly inhibits cellulose synthase/cellulose biosynthesis independent of ethylene. In contrast, when GALT2/GALT5-dependent SOS5 glycosylation is inhibited or SOS5 is mutated or FEI1/FEI2 is mutated, ACC synthase can no longer bind to FEI. Consequently, unbound ACC synthase produces ACC, which inhibits cellulose synthesis as well as leads to the production of ethylene.