CESA TRAFFICKING INHIBITOR inhibits cellulose deposition and interferes with the trafficking of cellulose synthase complexes and their associated proteins KORRIGAN1 and POM2/CELLULOSE SYNTHASE INTERACTIVE PROTEIN1

Abstract

Cellulose synthase complexes (CSCs) at the plasma membrane (PM) are aligned with cortical microtubules (MTs) and direct the biosynthesis of cellulose. The mechanism of the interaction between CSCs and MTs, and the cellular determinants that control the delivery of CSCs at the PM, are not yet well understood. We identified a unique small molecule, CESA TRAFFICKING INHIBITOR (CESTRIN), which reduces cellulose content and alters the anisotropic growth of Arabidopsis (Arabidopsis thaliana) hypocotyls. We monitored the distribution and mobility of fluorescently labeled cellulose synthases (CESAs) in live Arabidopsis cells under chemical exposure to characterize their subcellular effects. CESTRIN reduces the velocity of PM CSCs and causes their accumulation in the cell cortex. The CSC-associated proteins KORRIGAN1 (KOR1) and POM2/CELLULOSE SYNTHASE INTERACTIVE PROTEIN1 (CSI1) were differentially affected by CESTRIN treatment, indicating different forms of association with the PM CSCs. KOR1 accumulated in bodies similar to CESA; however, POM2/CSI1 dissociated into the cytoplasm. In addition, MT stability was altered without direct inhibition of MT polymerization, suggesting a feedback mechanism caused by cellulose interference. The selectivity of CESTRIN was assessed using a variety of subcellular markers for which no morphological effect was observed. The association of CESAs with vesicles decorated by the trans-Golgi network-localized protein SYNTAXIN OF PLANTS61 (SYP61) was increased under CESTRIN treatment, implicating SYP61 compartments in CESA trafficking. The properties of CESTRIN compared with known CESA inhibitors afford unique avenues to study and understand the mechanism under which PM-associated CSCs are maintained and interact with MTs and to dissect their trafficking routes in etiolated hypocotyls.

Figure 1.

CESTRIN reduces GFP-CESA3 velocity (particle movement rate) and induces its accumulation in endomembrane compartments. Arabidopsis seedlings expressing GFP-CESA3 were grown in the dark for 3 d and imaged by spinning-disk confocal microscopy. A, Seedlings expressing GFP-CESA3 were treated with DMSO (control). A single optical section and, as an indication of motility, an average of 60 frames are shown. B, Upon a 2-h 15 μm CESTRIN treatment, GFP-CESA3 particles no longer follow linear trajectories and are accumulated in punctae exhibiting increased fluorescence intensity. A single optical section and an average of 58 frames are shown. Bars in A and B = 10 μm. C, Histogram showing the distribution of GFP-CESA3 velocities at the PM focal plane under DMSO (white bars; n = 349) and CESTRIN (black bars; n = 105) treatment.

Figure 2.

CESTRIN does not broadly disrupt trafficking to the PM or secretion in Arabidopsis hypocotyls. Three-day-old etiolated Arabidopsis seedling hypocotyls were observed in the confocal microscope for either DMSO or 2-h CESTRIN treatment. Seedlings expressing CLC2-GFP (A), sec-GFP, a secretion marker (B), and THE1-GFP, a PM receptor-like kinase (C), did not show localization or morphological defects after CESTRIN treatment. Bars = 10 µm.

Figure 3.

CESTRIN inhibits anisotropic growth in Arabidopsis. A and B, Concentration-dependent growth inhibition of 5-d-old Arabidopsis etiolated hypocotyls under CESTRIN treatment. The half-maximal inhibitory concentration is calculated to be 4.85 µm using an exponential trend line (r² = 0.9416, n = 48). C and D, Propidium iodide staining of hypocotyl cells in 5-d-old Arabidopsis seedlings treated with CESTRIN shows decreased elongation and increased radial swelling. Bars = 50 µm.

Figure 4.

CESTRIN alters the localization and velocity of GFP-KOR1 and POM2/CSI1-3xYpet. Arabidopsis seedlings expressing GFP-KOR1 and POM2/CSI1-3xYpet were grown in the dark for 3 d and imaged by spinning-disk confocal microscopy. A, Seedlings expressing GFP-KOR1 were treated with DMSO (control). A single optical section and an average of 50 frames are shown. B, Upon a 2-h 15 μm CESTRIN treatment, GFP-KOR1 particles are accumulated in punctae exhibiting increased fluorescence intensity. A single optical section and an average of 55 frames are shown. Bars in A and B = 5 μm. C, Histogram showing the frequency of GFP-KOR1 velocities at the PM focal plane under DMSO (white bars; n = 406) or 15 μm CESTRIN (black bars; n = 247) treatment. D, Seedlings expressing POM2/CSI1-3xYpet were treated with DMSO (control). A single optical section and an average of 58 frames are shown. E, Upon a 1.5-h 15 μm CESTRIN treatment, POM2/CSI1-3xYpet particles show an altered distribution pattern, as shown by a 60-frame average. Upon a 2-h 15 μm CESTRIN treatment, POM2/CSI1-3xYpet was localized to the cytoplasm; an average of 60 frames is shown. Bars in D and E = 5 μm. F, Histogram showing the frequency of POM2/CSI1-3xYpet velocities at the PM focal plane under DMSO (white bars; n = 420) or 15 μm CESTRIN (black bars; n = 106) treatment.

Figure 5.

CESTRIN alters MT organization concurrent with GFP-CESA3 mislocalization. Three-day-old etiolated Arabidopsis seedlings expressing both RFP-TUA5 and GFP-CESA3 were imaged. A, DMSO-treated seedlings. A trace from six frames shows colocalization of GFP-CESA3 particles and MTs. B, A 2-h 15 μm CESTRIN treatment causes loss of MT organization and redistribution of GFP-CESA3 particles. The trace was composed of 58 frames. Bars = 10 μm.

Figure 6.

CESTRIN causes an increase in colocalization between CSCs and SYP61 in Arabidopsis hypocotyls. Colocalization of CFP-SYP61 (magenta) and GFP-CESA3 (cyan) in etiolated Arabidopsis hypocotyls (A) increases under 2 h of 15 µm CESTRIN treatment when compared with the DMSO control (B). A 15% increase in the colocalization of GFP-CESA3 and CFP-SYP61 particles was observed. Analysis was performed on three time series per treatment, acquired by sequential line scanning. Bars = 10 µm.

Figure 7.

The isoxaben-resistant mutant ixr1-1 did not display cross resistance to CESTRIN. A, Hypocotyl growth of wild-type Col-0, the isoxaben-resistant CESA3 mutant ixr1-1, and the CESA6 mutant prc1-1 in 8 µm CESTRIN compared with DMSO-supplemented medium (n = 16 seedlings per treatment). B, Wild-type Col-0, ixr1-1, and prc1-1 5-d-old seedlings were grown on medium containing 4 nm isoxaben or DMSO (n = 16 seedlings per treatment). The hypocotyl growth of ixr1-1 was reduced under CESTRIN treatment by 62% (P = 8.6 × 10⁻⁵), while it was not reduced under isoxaben treatment (P = 0.2) when compared with wild-type Col-0. When comparing Col-0 and ixr1-1, the two genotypes showed significantly different responses to the two chemicals (two-way ANOVA, P = 4.7 × 10⁻¹²). Significantly different responses of prc1-1 compared with Col-0 were observed for both chemical treatments (two-way ANOVA, P = 0.008).