Endoplasmic reticulum stress controls PIN-LIKES abundance and thereby growth adaptation

Abstract

Extreme environmental conditions eventually limit plant growth [J. R. Dinneny, Annu. Rev. Cell Dev. Biol. 35, 1-19 (2019), N. Gigli-Bisceglia, C. Testerink, Curr. Opin. Plant Biol. 64, 102120 (2021)]. Here, we reveal a mechanism that enables multiple external cues to get integrated into auxin-dependent growth programs in Arabidopsis thaliana. Our forward genetics approach on dark-grown hypocotyls uncovered that an imbalance in membrane lipids enhances the protein abundance of PIN-LIKES (PILS) [E. Barbez et al., Nature 485, 119 (2012)] auxin transport facilitators at the endoplasmic reticulum (ER), which thereby limits nuclear auxin signaling and growth rates. We show that this subcellular response relates to ER stress signaling, which directly impacts PILS protein turnover in a tissue-dependent manner. This mechanism allows PILS proteins to integrate environmental input with phytohormone auxin signaling, contributing to stress-induced growth adaptation in plants.

Keywords: PIN-LIKES; auxin; endoplasmic reticulum; membrane lipids; protein turnover.

Fig. 1.

imp2 is defective in CHER1. (A) Representative images of 4-d-old dark-grown seedlings of Col-0 wild-type, PILS5-GFP^OX (p35S::PILS5-GFP), and imp2 (in the PILS5-GFP^OX background) grown on ½ MS. (Scale bars, 5 mm.) (B) Immunoblot of PILS5-GFP in 3-d-old dark-grown PILS5-GFP^OX and imp2; PILS5-GFP^OX seedlings. α-Actin antibody was used for normalization. (C) Representative images and quantifications of PILS5-GFP signal in 3-d-old dark-grown PILS5-GFP^OX and imp2; PILS5-GFP^OX seedlings. (Scale bars, 50 µm.) n = 16, Student’s t test (B: P < 0.0001). (D) Relative hypocotyl length of 4-d-old dark-grown PILS5-GFP^OX, cher1-4, and cher1-4; PILS5-GFP^OX seedlings compared to Col-0 wild-type. n = 22 to 38, one-way ANOVA followed by Tukey’s multiple comparison test (B: P < 0.0001). (E) Relative hypocotyl length of 4-d-old dark-grown PILS5-GFP^OX, imp2; PILS5-GFP^OX, and imp2; PILS5-GFP^OX complemented with pCHER1::CHER1-YFP seedlings compared to Col-0 wild-type. n = 22 to 38, one-way ANOVA followed by Tukey’s multiple comparison test (B: P < 0.0001). In all panels’ boxplots: Box limits represent the 25th percentile and the 75th percentile; the horizontal line represents the median. Whiskers display min. to max. values. Representative experiments are shown, and all experiments were repeated at least three times.

Fig. 2.

Imbalance in membrane lipids affects PILS abundance. (A) Representative images and quantifications of PILS5-GFP signal in 3-d-old dark-grown PILS5-GFP^OX and imp2; PILS5-GFP^OX seedlings. Seedlings were grown on dimethyl sulfoxide (DMSO; solvent control) or 0.5 µM FB1 containing ½ MS medium. (Scale bars, 50 µm.) n = 20, two-way ANOVA followed by Tukey’s multiple comparison test (PILS5-GFP^OX DMSO vs. FB1 and PILS5-GFP^OX DMSO vs. imp2; PILS5-GFP^OX FB1, b: P < 0.0001; PILS5-GFP^OX DMSO vs. imp2; PILS5-GFP^OX FB1, b: P < 0.001). (B) Relative hypocotyl length of 4-d-old dark-grown Col-0 wild-type, PILS5-GFP^OX and imp2; PILS5-GFP^OX seedlings. Seedlings were grown on DMSO or 0.5 µM FB1, and relative hypocotyl length was calculated. n = 22 to 60, one-way ANOVA followed by Tukey’s multiple comparison test (B: P < 0.0001). (C) Representative images and quantifications of pDR5::GFP signal in 4-d old dark-grown seedlings. Seedlings were grown on DMSO or 0.5 µM FB1 containing solid medium. (Scale bars, 50 µm.) n = 6 to 8, two-way ANOVA followed by Tukey’s multiple comparison test (B and C: P < 0.0001). (D and E) Kinetics of apical hook opening of (D) Col-0 wild-type, pils2 pils5, cher1-4 pils2 pils5, cher1-4, or (E) Col-0 wild-type and pils2 pils5 germinated on ½ MS media supplemented with solvent control DMSO or 0.5 µM FB1. n ≥ 12, statistical significance was evaluated by nonlinear regression and a subsequent extra sum of squares F test. End of maintenance phase (X₀) and speed of opening (K) were compared to Col-0 wild-type (D) or DMSO control (E) (B–D: P < 0.0001). In all panels’ boxplots: Box limits represent the 25th percentile and the 75th percentile; the horizontal line represents the median. Whiskers display min. to max. values. Representative experiments are shown, and all experiments were repeated at least three times.

Fig. 3.

ER stress–inducing conditions stabilize PILS protein levels. (A and B) Representative images and quantifications of PILS5-GFP^OX, PILS3-GFP^OX, and PILS6-GFP^OX signal in 3-d-old dark-grown seedlings. Seedlings were grown on ½ MS and treated with or without 75 mM NaCl (A), DMSO (solvent control), or 0.5 µg/mL TM (B) in liquid ½ MS for 4 h. (Scale bars, 50 µm.) n = 8 to 14, Student’s t test between ctrl. and treatment (**P < 0.01, ***P < 0.001, ****P < 0.0001). (C and D) Representative images and quantifications of p35S::DER1-mScarlet and p35S::GFP-HDEL signal in 3-d-old dark-grown seedlings. Seedlings were grown on ½ MS and treated with or without 75 mM NaCl (C) or DMSO or 0.5 µg/mL TM (D) in liquid ½ MS for 4 h. (Scale bars, 50 µm.) n = 8 to 14, Student’s t test between ctrl. and treatment (**P < 0.01, ***P < 0.001, ****P < 0.0001). Representative experiments are shown, and all experiments were repeated at least three times.

Fig. 4.

ER stress defines PILS-dependent growth. (A and B) Representative images and quantifications of pPILS3::PILS3-GFP in the apical hook region (in pils3-1 background) signal in 3-d-old dark-grown seedlings. Seedlings were grown on solid ½ MS and treated with or without 75 mM NaCl (A) or with solvent control DMSO or 5 µg/mL TM (B) in liquid ½ MS for 1 to 4 h. Representative images for untreated and 2 h time point are shown (see additional images in SI Appendix, Fig. S4 C and D). (Scale bars, 50 µm.) n = 10, one-way ANOVA followed by Tukey’s multiple comparison test for each treatment against control or DMSO (*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001). (C) Representative images and relative hypocotyl length of 5-d-old dark-grown seedlings germinated on ½ MS media supplemented with DMSO or 0.15 µg/mL TM. (Scale bars, 10 mm, n = 35 to 50.) One-way ANOVA followed by Tukey’s multiple comparison test (B: P < 0.0001). (D) Relative hypocotyl length of 3-d-old dark-grown seedlings transferred for 2 additional days on ½ MS media supplemented with or without 100 mM NaCl. n = 60, one-way ANOVA followed by Tukey’s multiple comparison test (B: PILS3 OX vs. Col.-0 P < 0.001, PILS3 OX vs. PILS5 OX P < 0.05, PILS3 OX vs. pils2 pils5 P < 0.0001; C: PILS5 OX vs. Col-0 and vs. pils2 pils5 P < 0.0001; D: pils2 pils5 vs. Col-0 P < 0.05). (E) Relative hypocotyl length of 3-d-old dark-grown seedlings transferred for 2 additional days on ½ MS media supplemented with DMSO or 0.5 µg/mL TM. n = 90 to 150, pooled data of three biological replicates are shown. One-way ANOVA followed by Tukey’s multiple comparison test (B: PILS3^OX vs. Col.-0 P < 0.001, PILS5^OX vs. Col-0 P < 0.0001; C: pils2 pils5 vs. Col-0 P < 0.05, pils2 pils5 vs. PILS3 and 5^OX P < 0.0001). In all panels’ boxplots: Box limits represent the 25th percentile and the 75th percentile; the horizontal line represents the median. Whiskers display min. to max. values. Representative experiments are shown, and all experiments were repeated at least three times.