Skip to main content
NCBI home page
The Plant Cell logoLink to The Plant Cell
. 2020 Jun 2;32(8):2508–2524. doi: 10.1105/tpc.19.00700

FRA1 Kinesin Modulates the Lateral Stability of Cortical Microtubules through Cellulose Synthase–Microtubule Uncoupling Proteins

Anindya Ganguly 1,1, Chuanmei Zhu 1,2, Weizu Chen 1,3, Ram Dixit 1,4
PMCID: PMC7401024  PMID: 32487563

The FRA1 kinesin affects the positional stability of its tracks by regulating the levels of CMU proteins available to bind to cortical microtubules.

Abstract

Cell wall assembly requires harmonized deposition of cellulose and matrix polysaccharides. Cortical microtubules orient the deposition of cellulose by guiding the trajectory of cellulose synthase complexes. Vesicles containing matrix polysaccharides are thought to be transported by the FRAGILE FIBER1 (FRA1) kinesin to facilitate their secretion along cortical microtubules. The cortical microtubule cytoskeleton thus may provide a platform to coordinate the delivery of cellulose and matrix polysaccharides, but the underlying molecular mechanisms remain unknown. Here, we show that the tail region of the Arabidopsis (Arabidopsis thaliana) FRA1 kinesin physically interacts with cellulose synthase–microtubule uncoupling (CMU) proteins that are important for the microtubule-dependent guidance of cellulose synthase complexes. Interaction with CMUs did not affect microtubule binding or motility of the FRA1 kinesin but differentially affected the protein levels and microtubule localization of CMU1 and CMU2, thus regulating the lateral stability of cortical microtubules. Phosphorylation of the FRA1 tail region inhibited binding to CMUs and consequently reversed the extent of cortical microtubule decoration by CMU1 and CMU2. Genetic experiments demonstrated the significance of this interaction to the growth and reproduction of Arabidopsis plants. We propose that modulation of CMU protein levels and microtubule localization by FRA1 provides a mechanism that stabilizes the sites of deposition of both cellulose and matrix polysaccharides.

INTRODUCTION

Plants build a mechanically tough cell wall that confers cell shape by defining the direction of turgor-driven expansion. The cell wall is a composite material consisting primarily of cellulose microfibrils enmeshed within a complex network of hemicellulose and pectin (Lampugnani et al., 2018). Cells control the supply locations and delivery patterns of these components to regulate wall mechanics and thus spatially guide their growth. A key component of this control mechanism is the cortical microtubule cytoskeleton, which serves as a scaffolding structure that spatially organizes the cell wall synthesis machinery (Oda, 2015; Anderson, 2018). Disruption of the cortical microtubule array by genetic or pharmacological methods perturbs wall assembly and leads to aberrant plant growth and development (Hamada, 2014; Horio and Murata, 2014).

graphic file with name TPC_201900700R2_fx1.jpg

Open in a new tab

Cellulose microfibrils are synthesized at the cell surface by transmembrane cellulose synthase complexes (CSCs). CSCs contained in Golgi-derived membrane compartments are inserted into the plasma membrane adjacent to cortical microtubules (Crowell et al., 2009; Gutierrez et al., 2009). Catalytically active CSCs move steadily along cortical microtubules (Paredez et al., 2006) due to physical linkage provided by the CELLULOSE SYNTHASE INTERACTIVE protein (Bringmann et al., 2012; Li et al., 2012). The cortical microtubule-associated cellulose synthase–microtubule uncoupling proteins (CMUs) in turn maintain the positional stability of cortical microtubules in the face of forces generated by motile CSCs (Liu et al., 2016). Together, these mechanisms constitute a multistep process by which cortical microtubules orient cellulose deposition.

Hemicellulose and pectin are synthesized in the Golgi and secreted into the extracellular space for incorporation in the cell wall. Accumulating evidence indicates that deposition of at least some of these matrix polysaccharides is directed by cortical microtubules (Fukuda, 1997; McFarlane et al., 2008; Kong et al., 2015; Zhu et al., 2015; Takenaka et al., 2018). Recent work on the Arabidopsis (Arabidopsis thaliana) FRAGILE FIBER1 (FRA1) kinesin has uncovered a potential molecular mechanism for this process. Kinesins are ATP-dependent molecular motors that transport cellular cargo along microtubules in a directional manner. The FRA1 kinesin was found to move long distances along cortical microtubules (Zhu and Dixit, 2011; Kong et al., 2015; Zhu et al., 2015), and this property was shown to be essential for its function (Ganguly et al., 2017). Loss of the FRA1 kinesin leads to accumulation of vesicles in the vicinity of Golgi bodies and greatly reduced pectin secretion (Zhu et al., 2015). Based on these data, the FRA1 kinesin is thought to transport secretory vesicles containing matrix polysaccharides to exocytic sites positioned along cortical microtubules.

The position of cortical microtubules needs to be stable to maintain the pattern of deposition of both cellulose and matrix polysaccharides. When the demand for cell wall synthesis rises, CSCs increase in abundance and motility (Crowell et al., 2009; Zhu et al., 2015). Under these circumstances, the ability of cortical microtubules to withstand forces generated by motile CSCs would need to concomitantly increase. How this occurs remains unknown.

Here, we report that CMUs directly interact with the tail region of the FRA1 kinesin, and together they contribute to plant stature and fertility. We find that the steady state protein levels of CMUs are regulated by their interaction with FRA1, whereas the motility and function of the FRA1 kinesin is unaffected by CMUs. Finally, we show that phosphorylation of the FRA1 tail domain weakens its interaction with CMUs, providing a mechanism to tune the extent of cortical microtubule localization of CMUs and consequently the degree of lateral stability of cortical microtubules. Together, our data reveal a mechanism that functionally links the cellulose and matrix polysaccharide deposition machinery.

RESULTS

FRA1 Kinesin Physically Interacts with CMUs

A yeast two-hybrid screen conducted by Hybrigenics Services with a C-terminal fragment of the FRA1 kinesin (At5g47820) that includes the tail domain and part of the stalk region (amino acids 616 to 1035; Figure 1A) identified the Arabidopsis CMU1 (At4g10840) and CMU2 (At3g27960) as putative interacting partners. Based on amino acid sequence similarity to the human kinesin light-chain proteins, the Arabidopsis CMUs were previously also called kinesin light chain–related proteins (Bürstenbinder et al., 2013). We confirmed interaction between the FRA1 C-terminal fragment and full-length CMU1 and CMU2 using directed yeast two-hybrid assays (Figure 1B). In addition, we performed in vitro pull-down experiments and found that all three Arabidopsis CMU proteins interact with the FRA1 C-terminal fragment (Figures 1C and 1D; Supplemental Figure 1A). To determine whether this interaction occurs in vivo, we immunoprecipitated mCherry-tagged CMU1 and CMU2 expressed in wild-type Arabidopsis plants under their respective native promoters. We found that both mCherry-CMU1 and mCherry-CMU2 interact with full-length FRA1 in plants (Figures 1E and 1F). Similar experiments with mCherry-CMU3 were unsuccessful, probably because CMU3 expression is below the detection level in immunoblots (Supplemental Figure 1B). Together, these results identify CMUs as authentic interaction partners of the FRA1 kinesin.

Figure 1.

Figure 1.

Open in a new tab

FRA1 Interacts Directly with CMU1 and CMU2.

(A) Schematic representation of the FRA1 full-length protein and C-terminal region (FRA1-CT) used for pull-down and yeast two-hybrid assays in this study.

(B) Yeast two-hybrid assay using FRA1 C-terminal region fused with the activation domain (AD) as prey and CMU1 and CMU2 fused with the DNA binding domain (BD) as bait. Cells were grown on plates with or without 20 mM 3-amino-1,2,4-triazole (3AT).

(C) Pull-down of either MBP alone or MBP-tagged CMU1 and CMU2 proteins incubated with equal amounts of FRA1 C-terminal region (FRA1-CT). Immunoblotting was performed with anti-MBP and anti-FRA1 antibodies.

(D) Densitometry analysis of the FRA1 C-terminal region (FRA1-CT) bands in the pull-down lanes from (C), normalized to the amount of CMU protein pulled down. Center lines and error bars indicate mean ± sd (n = 3). Asterisks indicate significant difference between MBP-CMU1 and MBP-CMU2 lanes compared to MBP alone control: ***, P < 0.002 using ANOVA (Supplemental File). a.u., arbitrary units.

(E) Pull-down of mCherry-tagged CMU1 (mChr-CMU1) and CMU2 (mChr-CMU2) proteins expressed in Col-0 plants using their native promoters. Immunoblotting was performed with anti-mCherry and anti-FRA1 antibodies. The endogenous FRA1 protein frequently shows up as a doublet in immunoblots, as previously reported (Zhu et al., 2015; Ganguly et al., 2018). mChr, mCherry.

(F) Densitometry analysis of the full-length FRA1 bands in the pull-down lanes from (E), normalized to the amount of CMU protein pulled down. Center lines and error bars indicate mean ± sd (n = 3). Asterisks indicate significant difference between mChr-CMU1 and mChr-CMU2 lanes compared to Col-0 control: **, P < 0.02; ***, P < 0.002 using ANOVA (Supplemental File). a.u., arbitrary units; mChr, mCherry.

Loss-of-Function cmu Mutants Enhance the Growth and Fertility Defects of the fra1-5 Mutant

Previous work had identified T-DNA–induced loss-of-function Arabidopsis mutants of CMU1 and CMU2, named cmu1 and cmu2 (Liu et al., 2016). Since T-DNA insertion alleles are unavailable for CMU3 (At1g27500), we used artificial microRNA technology to knock down CMU3 expression in the cmu1 cmu2 mutant background (Supplemental Figure 1C).

We found that cmu1 seedlings have skewed roots, whereas cmu2 seedlings are indistinguishable from the wild type (Figures 2A and 2B). The cmu1 cmu2 double mutant and cmu1 cmu2 cmu3 triple mutant have skewed roots similar to the cmu1 single mutant (Figures 2A and 2B). In adult plants, none of the cmu mutants showed significant growth defects compared to wild-type plants (Figures 2C and 2D; Supplemental Figures 1D and 1E).

Figure 2.

Figure 2.

Open in a new tab

Phenotypes of cmu and fra1-5 Mutant Combinations.

(A) and (B) Root images (A) and quantification of the root-skewing angle (B) of 5-d-old, light-grown seedlings. Center lines and error bars indicate mean ± sd from 20 seedlings for each genotype. ns, not significant.

(C) and (D) Whole-plant appearance (C) and quantification of stem heights (D) of plants 42 d after germination. Center lines and error bars indicate mean ± sd (n > 12 plants). Bar = 5 cm.

(E) Images of primary inflorescence stems, isolated siliques, and flowers. Bar = 1 cm.

(F) Scatter plot of silique length. Center lines and error bars indicate mean ± sd (n > 20 mature siliques).

Asterisks denote significant difference between the indicated genotypes as determined by ANOVA (Supplemental File),: ****, P < 0.0001; ***, P < 0.002. ns, not significant.

To determine whether the physical interaction between FRA1 and CMU proteins is functionally important, we crossed the cmu mutants with the previously described fra1-5 knockout mutant (Zhu et al., 2015). We found that the root-skewing phenotype of the cmu1 single mutant was partially suppressed in the fra1-5 cmu1 double mutant (Figures 2A and 2B). However, the dwarf phenotype of the fra1-5 cmu1 double mutant was indistinguishable from the fra1-5 single mutant (Supplemental Figures 1D and 1E). By contrast, the fra1-5 cmu2 double mutant was significantly shorter than the fra1-5 mutant (Supplemental Figures 1D and 1E). Stem height was further reduced in the fra1-5 cmu1 cmu2 triple mutant (Figures 2C and 2D), indicating that CMU1 and CMU2 contribute additively to stem height in the fra1-5 background. By contrast, knockdown of CMU3 did not further decrease stem height in the fra1-5 cmu1 cmu2 cmu3 quadruple mutant (Figures 2C and 2D).

While fertility of the cmu1 cmu2 double mutant is similar to that of wild-type plants, we found that these mutations greatly enhanced the fertility defect of the fra1-5 mutant. As previously reported, the fra1-5 mutant has ∼10% undeveloped siliques compared to ∼1% undeveloped siliques in the wild-type plants (Kong et al., 2015). Introduction of the cmu1 cmu2 double mutation into the fra1-5 line resulted in ∼30% undeveloped siliques in the fra1-5 cmu1 cmu2 triple mutant (Figure 2E). In addition, fully developed siliques of the fra1-5 cmu1 cmu2 triple mutant were about half the length of the fra1-5 siliques (Figure 2F). To determine whether paternal or maternal tissues contribute to reduced fertility of these mutants, we performed reciprocal pollination experiments and measured the extent of pollen tube growth in the transmitting tract of the pistil. When wild-type pistils were used, we found that pollen grains from the cmu1 cmu2 double mutant grew to a similar length as wild-type pollen (Figures 3A and 3B). By contrast, pollen grains from the fra1-5 single mutant and the fra1-5 cmu1 cmu2 triple mutant grew significantly less over the same time period (Figures 3A and 3B). In the reciprocal experiments, wild-type pollen grew to a lesser extent on fra1-5 pistils than on wild-type and cmu1 cmu2 pistils, and this defect was exacerbated on fra1-5 cmu1 cmu2 pistils (Figures 3C and 3D).

Figure 3.

Figure 3.

Open in a new tab

Pistil and Pollen Defects in cmu and fra1-5 Mutant Combinations.

(A) Images of aniline blue–stained, wild-type pistils after 2 h of pollination with pollen from the indicated genotypes. Blue arrowheads point to pollen tubes growing in the transmitting tract. Red bars indicate the distance of pollen tube growth. Bar = 120 µm.

(B) Scatter plot of pollen tube lengths from experiments shown in (A). Center lines and error bars indicate mean ± sd (n > 12 sets).

(C) Images of aniline blue–stained pistils of the indicated genotype after 2 h of pollination with wild-type pollen. Blue arrowheads point to pollen tubes growing in the transmitting tract. Red bars indicate the distance of pollen tube growth. Pistils of the fra1-5 mutant reproducibly show a band of callose beneath the stigma (yellow arrowhead). Bar = 120 µm.

(D) Scatter plot of pollen tube lengths from experiments shown in (C). Center lines and error bars indicate mean ± sd (n > 12 sets).

(E) Micrographs of pollen grains stained with 1 µg/mL propidium iodide for 15 min. The dye is excluded from the cytoplasm of viable pollen. Bar = 20 µm.

(F) Scatter plot of the percentage of viable pollen grains. Bars indicate mean ± sd (n > 12 sets with at least 15 pollen in each set).

Asterisks denote significant difference between the indicated genotypes as determined by ANOVA (Supplemental File): ****, P < 0.0001; ***, P < 0.002; **, P < 0.02.

During these experiments, we observed that pollen grains from the fra1-5 cmu1 cmu2 triple mutant often failed to germinate. To determine whether this was due to reduced pollen viability, we stained pollen grains with propidium iodide. We found that while pollen grains from the fra1-5 and cmu1 cmu2 mutants showed similar viability as the wild type, pollen grains from the fra1-5 cmu1 cmu2 triple mutant showed an ∼30% decrease in viability (Figures 3E and 3F). Together, our data indicate that the CMU1 and CMU2 proteins work together with the FRA1 kinesin to achieve normal pollen and pistil function.

Motility of the FRA1 Kinesin Remains Unaffected in the cmu1 cmu2 Mutant

In mammals, kinesin light chains regulate the processive motility of kinesin-1 motors (Verhey et al., 1998; Cai et al., 2007; Wong and Rice, 2010). Since CMUs resemble kinesin light chains (Bürstenbinder et al., 2013), we investigated whether the motility of FRA1 is altered in the cmu1 cmu2 double mutant. For this purpose, we generated a single GFP-tagged version of full-length FRA1 driven by its own promoter and introduced it into the fra1-5 and fra1-5 cmu1 cmu2 mutants. The ProFRA1:FRA1-GFP construct fully complemented the dwarf phenotypes of both of these mutants, indicating that it is functional (Figures 4A and 4B). Live imaging of hypocotyl epidermal cells revealed that FRA1-GFP particles move processively in both fra1-5 and fra1-5 cmu1 cmu2 mutants (Figure 4C). We generated kymographs to quantify motility of the FRA1-GFP particles and found that their velocity, run length, and motile density were indistinguishable in the fra1-5 and fra1-5 cmu1 cmu2 backgrounds (Figures 4D and 4E). Therefore, we conclude that CMU1 and CMU2 do not regulate motility of the FRA1 kinesin.

Figure 4.

Figure 4.

Open in a new tab

CMU Proteins Do Not Regulate the Motility of FRA1.

(A) and (B) Whole-plant appearance (A) and quantification of stem heights (B) of plants 42 d after germination. Center lines and error bars indicate mean ± sd (n = 12 plants each). Bar = 5 cm.

(C) Single frames and time projections of 120 frames of FRA1-GFP. Directional movement of FRA1-GFP molecules appear as linear tracks in the time projection. The bright oval structures in these images are chloroplasts imaged because of chlorophyll autofluorescence. Bar = 5 µm.

(D) Representative kymographs showing the movement of FRA1-GFP. Diagonal lines represent motile events.

(E) Motile parameters of processive FRA1-GFP puncta. Values are mean ± sd.

FRA1 Differentially Regulates CMU1 and CMU2 Cortical Microtubule Localization

To determine whether FRA1 affects CMU proteins, we generated constructs that express mCherry-tagged CMU1 and CMU2 under the control of their respective native promoters. We found that mCherry-CMU1 and mCherry-CMU2 fusion proteins colocalize with cortical microtubules (Supplemental Figure 2A), as reported previously (Liu et al., 2016). Importantly, both fusion proteins are functional because they fully complement the growth defects of cmu1 and fra1-5 cmu2 mutants, respectively (Supplemental Figures 2B to 2E).

To examine the effect of FRA1 on these proteins, we introduced the mCherry-CMU1 and mCherry-CMU2 markers into the fra1-5 mutant. We found that the mCherry-CMU1 signal along cortical microtubules was significantly reduced in the fra1-5 mutant compared to the wild type (Figures 5A and 5B). Conversely, the mCherry-CMU2 signal along cortical microtubules was significantly enhanced in the fra1-5 mutant compared to the wild type (Figures 5C and 5D). Complementation of the fra1-5 mutant by FRA1-GFP restored the cortical microtubule signal of mCherry-CMU1 and mCherry-CMU2 to wild-type levels (Figures 5A and 5C).

Figure 5.

Figure 5.

Open in a new tab

FRA1 Differentially Regulates CMU1 and CMU2 Localization on Cortical Microtubules.

All data are from 4-d-old seedlings. The immunoblots are representative of at least three independent experiments. Coomassie-stained gels are shown below each immunoblot as a loading control.

(A) to (D) Fluorescence micrographs ([A] and [C]) and quantification ([B] and [D]) of mCherry-CMU1 or mCherry-CMU2 signal on cortical microtubules in hypocotyl epidermal cells of the indicated genotypes. Center lines and error bars indicate mean ± sd (three to four microtubules were quantified per cell, total eight cells from three seedlings). Bar in (A) = 10 µm; bar in (C) = 5 µm. a.u., arbitrary units.

(E) and (F) Immunoblotting (E) and quantification of mCherry-tagged CMU1 and CMU2 protein levels (F) in the indicated genotypes. Total protein extracts were probed with anti-mCherry and anti-FRA1 antibodies. Center lines and error bars indicate mean ± sd (n = 3). a.u., arbitrary units

(G) and (H) Immunoblots of mCherry-tagged CMU1 (G) or CMU2 (H) proteins expressed in Col-0, fra1-5, and fra1-5 plants complemented with untagged FRA1 or FRA1-GFP and probed with anti-mCherry antibody.

Asterisks denote significant difference between the indicated genotypes as determined by ANOVA (Supplemental File): ****, P < 0.0001; ***, P < 0.002.

To determine whether the changes in signal intensity of mCherry-CMU1 and mCherry-CMU2 are due to alteration of protein levels, we performed immunoblot analysis with an anti-mCherry antibody. These experiments revealed that the amount of mCherry-CMU1 protein was approximately twofold lower and the amount of mCherry-CMU2 protein was approximately twofold higher in the fra1-5 mutant than in wild-type plants (Figures 5E and 5F). Consistent with our microscopy data, mCherry-CMU1 and mCherry-CMU2 protein levels returned to wild-type levels upon introduction of a functional copy of FRA1 in the fra1-5 mutant (Figures 5G and 5H).

We used drug treatments to study whether FRA1 affects the turnover of CMU proteins. Treatment of Columbia-0 (Col-0) seedlings expressing mCherry-CMU1 with cycloheximide revealed that the CMU1 protein levels significantly decrease over 24 h (Supplemental Figure 3A). Cotreatment of these seedlings with cycloheximide and the 26S proteasome inhibitor MG132 prevented mCherry-CMU1 protein loss (Supplemental Figure 3A). In addition, treatment of fra1-5 seedlings expressing mCherry-CMU1 with MG132 restored the mCherry-CMU1 protein to wild-type levels (Supplemental Figure 3B). Together, these data indicate that CMUs are degraded by the proteasome system and the FRA1 kinesin regulates this process to influence the steady state level of CMU proteins available to bind to cortical microtubules.

Lateral Stability of Cortical Microtubules Is Reduced in cmu and fra1-5 Mutants

CMU proteins prevent the lateral displacement of cortical microtubules by active CSCs (Liu et al., 2016). To determine whether CMU1 and CMU2 differ in their ability to stabilize cortical microtubule positions, we conducted live imaging of cortical microtubules in the cmu1 and cmu2 single mutants. In the cmu1 mutant, entire cortical microtubules could be seen drifting in the cell cortex as if they were detached from the plasma membrane (Figure 6A). In addition, the growing plus ends of cortical microtubules often abruptly changed orientation in the cmu1 mutant (Figure 6B). By contrast, typically only small portions of cortical microtubules showed lateral displacement in the cmu2 mutant (Figure 6C). To measure the extent of lateral displacement of cortical microtubules, we used kymograph analysis as described previously (Liu et al., 2016). Cortical microtubules that stably maintain their position appear as vertical lines in kymographs, whereas cortical microtubules that drift laterally appear as oblique lines in kymographs (Figure 6D). We found that the extent and frequency of lateral microtubule displacements were significantly higher in the cmu1 mutant than in the wild type, whereas they were indistinguishable between cmu2 and the wild type (Figures 6E and 6F).

Figure 6.

Figure 6.

Open in a new tab

Lateral Displacement of Cortical Microtubules in cmu and fra1-5 Mutants.

(A) Example of an entire cortical microtubule detaching from the cell cortex in the cmu1 mutant.

(B) Example of the plus end of a cortical microtubule detaching from the cell cortex and abruptly changing orientation in the cmu1 mutant.

(C) Example of a cortical microtubule being laterally displaced in the cmu2 mutant.

(D) Kymographs showing lateral displacement of cortical microtubules in hypocotyl epidermal cells. Positionally fixed microtubules appear as straight lines, whereas laterally drifting microtubules appear as slanted lines (orange arrowheads).

(E) and (F) Plots of the angle of lateral displacement of cortical microtubules (E) and the frequency of displaced microtubules (F) determined from kymographs. Center lines and error bars indicate mean ± sd. For (E), n > 200 microtubules from 15 to 20 cells from at least four independent seedlings. For (F), n = 12 to 14 cells from at least four independent seedlings. Asterisks denote a significant difference from Col-0 as determined by ANOVA (Supplemental File): ****, P < 0.0001; ***, P < 0.002; **, P < 0.02.

Numbers indicate time in seconds. Yellow dotted line traces the microtubule of interest. Bar in (A), (B), and (C) = 3 µm.

Since cortical microtubule localization of CMU proteins is altered in the fra1-5 mutant, we wondered whether the lateral stability of cortical microtubules was compromised in fra1-5 cells. Live imaging revealed that cortical microtubules in the fra1-5 mutant showed lateral deflections similar to that observed in the cmu1 single mutant and cmu1 cmu2 double mutant (Figures 6D to 6F; Supplemental Figure 4A). This defect was not enhanced in the fra1-5 cmu1 cmu2 triple mutant compared to the fra1-5 and cmu1 cmu2 mutants (Figures 6D to 6F; Supplemental Figure 4B). Together, these data indicate that FRA1 itself does not contribute to the positional stability of cortical microtubules and that this phenotype is likely the indirect consequence of lower CMU1 protein levels in the fra1-5 mutant.

FRA1 C-Terminal Region Is Phosphorylated

Mass spectrometry data in the Arabidopsis protein phosphorylation site database identified Thr-687 and Thr-694 as being phosphorylated in the C-terminal region of FRA1 kinesin (Durek et al., 2010). The NetPhos3.1 and KinasePhos2.0 phosphorylation site prediction tools (Blom et al., 2004; Wong et al., 2007) indicated that several different kinases can potentially phosphorylate FRA1. However, casein kinase-1 family members were predicted to phosphorylate both Thr-687 and Thr-694, making them promising candidates for this task.

To determine whether FRA1 is a substrate of these kinases, we selected casein kinase1-like3 (CKL3; At4g28880) and casein kinase1-like 6 (CKL6; At4g28540) because they have been shown to be active kinases in vitro (Ben-Nissan et al., 2008; Tan et al., 2013). We were particularly interested in CKL6 because it localizes to cortical microtubules (Ben-Nissan et al., 2008). We performed in vitro pull-down experiments and found that full-length, maltose-binding protein (MBP)-tagged CKL3 and CKL6 both interact directly with the FRA1 C-terminal domain (Figure 7A). However, only CKL6 was able to phosphorylate the FRA1 C-terminal domain in vitro (Figure 7B). Mutating the Thr-687 and Thr-694 residues to Ala reduced phosphorylation of the FRA1 C-terminal domain by CKL6 by approximately threefold (Figure 7C). Control experiments showed that both CKL3 and CKL6 phosphorylate tubulin dimers in vitro (Supplemental Figure 5), demonstrating that the lack of phosphorylation of the FRA1 C-terminal domain by CKL3 was not due to a lack of kinase activity.

Figure 7.

Figure 7.

Open in a new tab

C-Terminal Tail Region of FRA1 Is Phosphorylated by CKL6.

(A) Pull-down of either MBP alone or MBP-tagged CKL3 and CKL6 proteins incubated with equal amounts of C-terminal region of FRA1 (FRA1-CT) protein. Immunoblotting was performed with anti-MBP and anti-FRA1 antibodies.

(B) In vitro phosphorylation experiment with wild type FRA1 C-terminal region (FRA1-CT), and a mutant C-terminal region containing T687A and T694A mutations (FRA1-AA) incubated with either CKL3 or CKL6. Phosphorylation was detected by an anti–phospho-Ser/Thr antibody. Coomassie-stained gel is shown below as a loading control.

(C) Densitometry analysis of the phosphorylated FRA1 bands from the experiments in (B). a.u., arbitrary units; FRA1-CT, C-terminal region of wild-type FRA1; FRA1-AA, C-terminal region of FRA1 containing T687A and T694A mutations.

(D) and (E) Whole-plant appearance (D) and quantification of stem heights (E) of plants of the indicated genotypes 42 d after germination. Center lines and error bars indicate mean ± sd (n = 12 plants each). Asterisks indicate significant difference as determined by ANOVA (Supplemental File): ***, P < 0.002; **, P < 0.02; *, P < 0.05. Bar = 5 cm. FRA1-AA-TOM, full-length FRA1 containing T687A and T694A mutations tagged with tdTomato; FRA1-DD-TOM, full-length FRA1 containing T687D and T694D mutations tagged with tdTomato; FRA1-TOM, full-length, wild-type FRA1 tagged with tdTomato.

(F) Immunoblot of total protein extracts probed with anti-FRA1 antibody. Coomassie-stained gel is shown below as a loading control.

(G) Representative kymographs showing the movement of tdTomato-labeled wild-type FRA1, FRA1-AA, and FRA1-DD mutants. Diagonal lines represent motile events.

(H) Motile parameters of processive FRA1-tdTomato puncta. Values are mean ± sd.

To study whether phosphorylation of FRA1 at these residues contributes to its function, we introduced untagged and tdTomato-tagged wild-type FRA1 or phosphodefective (FRA1-AA) or phosphomimetic (FRA1-DD) versions expressed under the control of the native FRA1 promoter in the fra1-5 mutant. In agreement with previous work by Ganguly et al. (2018), wild-type FRA1-tdTomato fully rescued the dwarf phenotype of the fra1-5 mutant (Figures 7D and 7E). By contrast, the FRA1-AA and FRA1-DD proteins restored stem height to ∼90% of the wild type, but did not completely rescue the fra1-5 mutant (Figures 7D and 7E). Importantly, the steady state protein levels and motile properties of the FRA1 phosphorylation mutants were similar to those of the wild-type FRA1 (Figures 7F to 7H). Therefore, phosphorylation of Thr-687 and Thr-694 does not significantly affect the abundance and motility of the FRA1 kinesin.

Phosphorylation of the FRA1 C-Terminal Region Regulates the Extent of CMU Localization on Cortical Microtubules

Next, we investigated whether phosphorylation of Thr-687 and Thr-694 regulates the interaction of FRA1 with CMUs. In vitro pull-down experiments showed that the phosphodefective mutations of the FRA1 C-terminal region enhanced binding to CMU1 and CMU2, whereas the phosphomimetic mutations decreased binding to CMU1 and CMU2 (Figures 8A and 8B). To determine whether phosphorylation of FRA1 affects the microtubule localization of CMUs in vivo, we introduced untagged FRA1-AA and FRA1-DD proteins in the fra1-5 mutant expressing either mCherry-CMU1 or mCherry-CMU2. We found that the phosphodefective FRA1 mutant restored the levels of mCherry-CMU1 and mCherry-CMU2 on cortical microtubules to wild-type levels (Figures 8C to 8F). By contrast, the phosphomimetic FRA1 mutant was only partially able to perform this function (Figures 8C to 8F).

Figure 8.

Figure 8.

Open in a new tab

Phosphorylation of FRA1 Regulates CMU1 and CMU2 Localization on Cortical Microtubules.

(A) Pull-down of MBP-CMU1 and MBP-CMU2 incubated with equal amounts of either the wild-type FRA1 C-terminal region (FRA1-CT), C-terminal region with T687A and T694A mutations (FRA1-AA), or C-terminal region with T687D and T694D mutations (FRA1-DD). Immunoblotting was performed with anti-MBP and anti-FRA1 antibodies.

(B) Densitometry analysis of the FRA1 bands in the pull-down lanes from the experiments shown in (A). a.u., arbitrary units; FRA1-CT, FRA1 C-terminal region.

(C) to (F) Fluorescence micrographs ([C] and [E]) and quantification ([D] and [F]) of mCherry-CMU1 or mCherry-CMU2 signal on cortical microtubules in hypocotyl epidermal cells of 4-d-old seedlings of the indicated genotypes. Center lines and error bars indicate mean ± sd (three to four microtubules were quantified per cell, total eight cells from three seedlings). Bar in (B) = 10 µm; bar in (D) = 5 µm. a.u., arbitrary units.

Asterisks denote significant difference between the indicated genotypes as determined by ANOVA (Supplemental File): ****, P < 0.0001; ***, P < 0.002.

DISCUSSION

The cortical microtubule cytoskeleton spatially directs the deposition of cell wall material. Individual cortical microtubules need to be laterally stable to prevent unwanted changes in the pattern of cellulose and matrix polysaccharides in the cell wall, which could lead to aberrant cell expansion. This requirement becomes particularly important during periods of rapid growth when cells boost the abundance and activity of both CSCs and FRA1 kinesin to meet the increased demand for cellulose and matrix polysaccharides, respectively (Crowell et al., 2009; Zhu et al., 2015; Ganguly et al., 2018). An increase in the number of motile CSCs in the plasma membrane could be deleterious because they would predictably deflect greater numbers of cortical microtubules and thus potentially disrupt array organization. Therefore, increased cellulose synthesis needs to be accompanied by increased positional stability of cortical microtubules (Liu et al., 2016). Here, we show that the FRA1 kinesin physically interacts with CMU proteins and leads to an increase in CMU1 protein levels. Since CMU1 appears to be primarily responsible for the positional stability of cortical microtubules, an increase in the amount of FRA1 in growing cells would simultaneously enhance the resistance of cortical microtubules to lateral displacement by CSCs (Figure 9A). The transport activity of FRA1 combined with its ability to bolster the lateral stability of cortical microtubules through CMUs provides a mechanism to stabilize the sites of deposition of both cellulose and matrix polysaccharides.

Figure 9.

Figure 9.

Open in a new tab

Model of How Regulated Interaction of FRA1 with CMU1 and CMU2 Contributes to the Lateral Stability of Cortical Microtubules.

(A) In seedlings, interaction with FRA1 leads to an increase in the levels and thus cortical microtubule localization of CMU1, whereas it leads to a decrease in the levels and cortical microtubule localization of CMU2. An increase in FRA1 levels would thus lead to more CMU1 binding and greater lateral stability of cortical microtubules. CMUs may bind to FRA1 in the cytoplasm and/or to FRA1 that is attached to a microtubule. Whether CMU1 and CMU2 form a ternary complex with FRA1 remains to be determined. Since CMUs do not show directed movement on cortical microtubules, they are unlikely to be bound to motile FRA1 molecules (denoted by the filled black arrows). Green arrows represent positive regulation and red bars represent negative regulation.

(B) In adult plants, the tail region of FRA1 is proposed to be phosphorylated by kinases such as CKL6, which inhibits the interaction of FRA1 with both CMU1 and CMU2. This reverses the effect of FRA1 on CMU1 and CMU2 protein levels, leading to greater cortical microtubule localization of CMU2 than CMU1. Hence, phosphorylation provides a mechanism to control the relative amount of CMU1 and CMU2 on cortical microtubules.

CMUs are microtubule-associated proteins (MAPs) by virtue of their microtubule localization in vivo and direct microtubule binding in vitro (Liu et al., 2016). Accruing evidence suggests that certain MAPs serve as positive or negative regulators of kinesins. For example, tau and MAP2 proteins both inhibit kinesin-1 motility (Heins et al., 1991; Seitz et al., 2002; Vershinin et al., 2007; Dixit et al., 2008). By contrast, MAP7 recruits kinesin-1 to microtubules and activates motility (Barlan et al., 2013; Tymanskyj et al., 2018; Chaudhary et al., 2019; Hooikaas et al., 2019). Similarly, the MAP65 family of microtubule-bundling proteins recruit kinesin-4 to antiparallel microtubule overlaps in both plants and animals (Bieling et al., 2010; de Keijzer et al., 2017). However, we found that CMUs do not influence the amount, microtubule localization or motility of the FRA1 kinesin during interphase. Instead, we found that FRA1 regulates microtubule localization of CMUs, providing an example of a motor protein that modifies the behavior of its track. This activity appears to require direct binding of CMUs to FRA1 because inhibitory phosphomimetic mutations in the tail domain of FRA1 significantly attenuated FRA1’s impact on the microtubule localization of CMUs. Since FRA1 is a dimeric motor protein (Zhu and Dixit, 2011; Zhu et al., 2015), it has two binding sites for CMU proteins. Whether CMU1 and CMU2 can together bind to FRA1 and whether they compete for this interaction remains to be determined.

Since the steady state protein levels of CMU1 and CMU2 change in the fra1-5 mutant, one potential mechanism by which FRA1 regulates the extent of cortical microtubule localization of CMUs is through protein degradation. Inhibition of the 26S proteasome restores CMU1 in the fra1-5 mutant to wild-type levels, indicating that FRA1 regulates the degradation of CMUs by the proteasome system. In addition to regulating the turnover of CMU proteins, FRA1 might also affect their microtubule binding activity. One possibility is that FRA1 directly regulates CMUs microtubule binding through an allosteric effect. Such a mechanism combined with competition between CMU1 and CMU2 for binding to FRA1 might explain the differential effect of FRA1 on the microtubule localization of CMU1 and CMU2. Alternatively, FRA1 might indirectly affect the microtubule localization of CMU1 and CMU2 by influencing their interaction with other regulatory proteins. A plant-specific IQ67 domain (IQD)–containing protein called IQD1 has been shown to bind to CMUs and regulate their microtubule localization (Bürstenbinder et al., 2013). Therefore, it will be interesting to determine whether FRA1 affects the binding of CMUs and IQD1.

We found that roots of the cmu1 mutant grow in a skewed manner, in agreement with the previously reported cell file twisting phenotype in hypocotyls of the cmu1 mutant (Liu et al., 2016). By contrast, the cmu2 seedlings did not show any measurable growth defects, at least under normal growth conditions. The severe loss of lateral stability of cortical microtubules in the cmu1 mutant compared to cmu2 provides a plausible explanation for the aberrant growth of the cmu1 mutant. The partial suppression of the root-skewing phenotype in the fra1-5 cmu1 double mutant could be due to an increase in the amount of CMU2 on cortical microtubules in the absence of FRA1. None of the cmu mutant combinations showed significant growth defects in adult plants. However, when introduced into the fra1-5 background, the cmu lesions exacerbated the growth and fertility defects of the fra1-5 mutant. Thus, fra1-5 provides a sensitized genetic background to study the contribution of cortical microtubule lateral stability conferred by CMUs to the growth and fecundity of plants. It is also possible that cell wall and potentially other cellular defects caused by the fra1-5 and cmu mutations work additively to produce the more severe phenotypes observed in fra1-5 cmu mutant combinations.

Our genetic experiments revealed that both CMU1 and CMU2 contribute to stem growth. We found that CMU3 does not contribute to this phenotype, probably because the CMU3 protein is undetectable in plants. Reciprocal pollination experiments demonstrated that reduced fertility of the fra1-5 and fra1-5 cmu1 cmu2 mutants was due to defective pistil and pollen functions. Successful pollen–pistil interactions involve pollen adhesion, hydration, and germination on the stigma surface to produce a pollen tube that then penetrates the stigma and elongates into the transmitting tract of the style on its way to the ovary (Kandasamy et al., 1994). These steps require secretion of specific degradative enzymes, pollen tube growth-promoting molecules, and guidance factors into the cell wall of pistil tissues (Hiscock et al., 1994; Cheung et al., 1995; Lavithis and Bhalla, 1995; Hiscock et al., 2002; Okuda et al., 2009; Samuel et al., 2009). The reduced receptivity of the fra1-5 and fra1-5 cmu1 cmu2 pistils might be due to compromised secretion of these factors. It is also possible that cell walls of the transmitting tract of fra1-5 and fra1-5 cmu1 cmu2 pistils have altered composition and/or structure that might impede pollen tube growth. On the pollen side, our results suggest that FRA1 contributes to pollen tube elongation, perhaps by facilitating secretion of wall material at the growing tip. Furthermore, pollen viability was significantly reduced in the fra1-5 cmu1 cmu2 mutant, indicating that proper cell wall production is important for pollen development.

Arabidopsis protein phosphorylation data sets indicated that the tail region of FRA1 is phosphorylated at residues Thr-687 and Thr-694. Here, we show that CKL6 physically binds to the FRA1 tail region and can mediate this phosphorylation in vitro. Since mutating Thr-687 and Thr-694 to Ala did not eliminate phosphorylation of the FRA1 tail region by CKL6, there are likely additional residues that serve as substrates for CKL6. In addition, we cannot exclude the possibility that other kinases phosphorylate FRA1 at these or other tail domain locations in vivo. Notably, phosphorylation of the FRA1 tail region did not affect its abundance or motility. Therefore, these phosphorylation events are not working to regulate motor activity. Rather, they regulate the binding of CMUs to FRA1, although we cannot exclude the possibility that the AA and DD mutations affect this interaction by changing the structure of the C-terminal region of FRA1. Since interaction with FRA1 differentially affects the levels and hence the cortical microtubule localization of CMU1 and CMU2, phosphorylation of the FRA1 tail region provides a mechanism to alter the balance of CMU1 and CMU2 along cortical microtubules. We found that while CMU1 is the major contributor to seedling growth, CMU2 is important for inflorescence stem growth in adult plants. The switch between CMU1 and CMU2 during seedling to adult plant transition could be achieved by phosphorylation of the FRA1 tail region (Figure 9B). Regulated interaction between FRA1 and CMUs may also be important for the development and function of plant reproductive tissues. Taken together, our data provide insights into how mechanisms for the oriented deposition of cellulose and matrix polysaccharides functionally intersect at the cortical microtubule interface.

METHODS

Plant Material and Growth

Arabidopsis (Arabidopsis thaliana), Col-0 accession, was used throughout. The cmu1 and cmu2 mutants were isolated from T-DNA insertion lines SAIL_335_B08 (cmu1) and SALK_148296C (cmu2), obtained from Arabidopsis Biological Resource Center (Alonso et al., 2003). Homozygous mutants were identified by PCR-based genotyping with primers listed in the Supplemental Table. The fra1-5 mutant was described by Zhu et al. (2015). The cmu1 cmu2 double mutant was generated by crossing the cmu1 and cmu2 single mutants. The homozygous cmu1 cmu2 double mutant was crossed with the fra1-5 mutant, and the resulting segregating progeny were screened by PCR genotyping to isolate fra1-5 cmu1, fra1-5 cmu2, and fra1-5 cmu1 cmu2 lines. For generating the cmu1 cmu2 cmu3 triple mutant, a CMU3 artificial microRNA construct (see the Supplemental Table for primers) was introduced into the cmu1 cmu2 double mutant using Agrobacterium-mediated floral dip transformation, and independent homozygous cmu1 cmu2 cmu3 lines were isolated by antibiotic selection and PCR genotyping. A cmu1 cmu2 cmu3 line showing significant reduction in CMU3 expression was then crossed with the fra1-5 mutant to generate the fra1-5 cmu1 cmu2 cmu3 quadruple mutant.

For growth on plates, seeds were sterilized with 5% (v/v) sodium hypochlorite for 10 min, rinsed with sterile water four times, and planted on 0.5× Murashige and Skoog medium (Caisson Laboratories). Seeds were stratified at 4°C for 2 d and then germinated at 20°C with 16 h of 120- to 140-µmol light with Philips F96T8/TL841/HO/PLUS 86-W bulbs. Live imaging used 4-d-old, light-grown seedlings unless otherwise stated. Root length and root-skewing angle were measured using the Fiji ImageJ package (Schindelin et al., 2012) on 5-d-old, light-grown seedlings. For growth in soil, seeds were grown under continuous light at 120- to 140-µmol intensity with GE Ecolux with Starcoat F40T8SPX41 40-W 4100K bulbs, 70% humidity, and 21°C after stratification at 4°C for 2 d. Stem height was measured with a ruler.

RT-PCR Analysis

Total RNA was extracted from basal internodes of 4-week-old plants using an Omega Plant RNA Mini Kit (Omega Bio-Tek). Approximately 500 ng of DNase-treated RNA was used for cDNA synthesis using qScript cDNA supermix (Quanta BioSciences). Primers for RT-PCR were designed using sequence downstream of the T-DNA insertion sites and are listed in the Supplemental Table.

Generation of Transgene Constructs and Transgenic Arabidopsis Plants

Primers used for construct assembly are listed in the Supplemental Table. The ProCMU1:mCherry-CMU1 and ProCMU2:mCherry-CMU2 constructs were generated using ∼2 kb of CMU1 or CMU2 promoter, mCherry cDNA, and full-length genomic DNA of CMU1 or CMU2 genes, respectively. The constructs included a 36-bp linker sequence between mCherry and CMU1 or CMU2 (Supplemental Table). The assembled constructs were ligated into pCAMBIA3300 vector and introduced into wild-type Arabidopsis plants expressing the GFP-fused β-tubulin 6 isoform (GFP-TUB6) microtubule marker (gift from Takashi Hashimoto, Nara Institute of Science and Technology, Japan) via Agrobacterium-mediated floral dip transformation. Transgenic plants were selected using 10 µg/mL phosphinothricin, and homozygous T4 lines expressing a single copy of the transgene, as determined by segregation analysis, were used for imaging. Two independent mCherry-CMU1 and mCherry-CMU2 lines were crossed with the cmu1, cmu2, and fra1-5 mutants to generate the cmu1/mCherry-CMU1, cmu2/mCherry-CMU2, fra1-5/mCherry-CMU1, and fra1-5/mCherry-CMU2 lines, which were used for complementation analysis and live imaging. The fra1-5 plants expressing mCherry-CMU1 or mCherry-CMU2 were transformed with either untagged FRA1 or FRA1-enhanced GFP (EGFP; see below) to examine the effect of FRA1 expression on CMU protein levels and microtubule localization. All transgene insertions and mutant backgrounds were confirmed by PCR-based screening.

For complementation of the fra1-5 cmu2 mutant, the mCherry-CMU2 construct was transformed into this background, and three independent lines homozygous for the transgene and both mutations were subsequently analyzed. All FRA1 constructs were cloned into the pCAMBIA1300 vector with a modified multiple cloning site consisting of 5′-SalI-XbaI-BamHI-AvrII-XmaI-AvrII-NruI-BglII-MluI-AscI-SpeI-NcoI-SacI-XbaI-3′. The ProFRA1:FRA1-tdTomato construct has been described by Zhu et al. (2015). The ProFRA1:FRA1-EGFP construct was made by replacing the 3X-GFP in pFRA1:FRA1-3GFP (Zhu et al., 2015) by a single EGFP gene, and the ProFRA1:FRA1 (untagged) version was made by deleting the EGFP gene from this construct. The tdTomato and untagged versions of the FRA1 T687A-T694A and T687D-T694D mutations were constructed by the megaprimer method using primers listed in the Supplemental Table. The FRA1-EGFP and FRA1 constructs were transformed into the fra1-5 mutant to generate the fra1-5/FRA1-EGFP and fra1-5/FRA1 lines containing a single transgene insertion as determined by segregation analysis. The fra1-5 cmu1 cmu2/FRA1-EGFP line was generated by crossing the fra1-5 cmu1 cmu2 line with a fra1-5/FRA1-EGFP line followed by antibiotic selection of the FRA1-EGFP transgene and PCR-based genotyping to select the fra1-5 cmu1 cmu2 background.

To visualize cortical microtubules in the cmu1 and cmu2 mutants, the GFP-TUB6 microtubule marker was introduced into these lines via Agrobacterium-mediated floral dip transformation. At least four independent T1 lines were used for live imaging for each genotype.

Live Imaging and Image Analysis

Arabidopsis seedlings were imaged using variable-angle epifluorescence microscopy (Konopka and Bednarek, 2008). Live imaging of GFP-TUB6– and mRuby-TUB6–labeled microtubules was conducted on 4-d-old seedlings mounted in 0.5× Murashige and Skoog media between two layers of double-sided adhesive tape. Unless otherwise stated, epidermal cells in the apical or subapical region of the hypocotyl were imaged. GFP and mRuby were excited using 3-mW 488-nm and 5-mW 561-nm diode-pumped solid-state lasers (Melles Griot), respectively. To image FRA1-EGFP and FRA1-tdTomato, specimens were excited using 3-mW 488-nm and 5-mW 561-nm diode-pumped solid-state lasers (Melles Griot), respectively. Images were collected using 100× (NA 1.45) objective and a back-illuminated electron-multiplying charge-coupled device camera (ImageEM; Hamamatsu) at 1 s for FRA1 imaging and 3 s for microtubule imaging.

To analyze the signal intensity of mCherry-CMU1 and mCherry-CMU2 on cortical microtubules, all images were acquired with identical settings within a course of 3 to 4 h. For these measurements, single microtubules were chosen based on their signal intensity being similar to newly nucleated single microtubules. At least 25 independent microtubules from multiple cells were analyzed in each case, and the mean mCherry signal per unit length of cortical microtubule was measured using the Fiji ImageJ package (Schindelin et al., 2012). Background signal intensity was measured in multiple regions inside the cell with no visible microtubules. The mean background signal was subtracted from the mCherry-CMU signal intensity. For microtubule lateral displacement measurements, kymographs from at least 15 cells were generated by selecting regions in the middle of the cell along their long axis. The extent of lateral displacement of all microtubules in each kymograph was quantified by the angular deviation of a microtubule from its straight (vertical) path.

Yeast Two-Hybrid Assay

The initial yeast two-hybrid screen that identified CMU1 and CMU2 as potential interaction partners of FRA1 was conducted by Hybrigenics Services. Directed yeast two-hybrid experiments were performed according to the ProQuest Two-Hybrid System with Gateway Technology manual (Invitrogen). Full-length CMU1 and CMU2 along with a FRA1 C-terminal fragment (amino acids 616 to 1035) were first cloned into the entry vector pENTR 11 with enzymes Kpn1 and Not1 for CMU1 and CMU2 and Kpn1 and EcoR1 for FRA1 (for primers, see Supplemental Table). From the entry clones, CMU1, CMU2, and FRA1-tail domain were then cloned into pDEST32 and pDEST22 vectors, respectively, using LR Clonase II enzyme (Invitrogen). These constructs together with all the controls were transformed into the MaV203 competent yeast cells (Saccharomyces cerevisiae) and selected on SD-Leu-Trp plates. For each set, at least four independent colonies were used for yeast two-hybrid assay on SD-Leu-Trp-His plates with and without 3-amino-1,2,4-triazole (Sigma-Aldrich). Colonies were imaged every 24 h after plating for 4 to 5 d.

In Vitro Pull-Down Experiments

CMU1, CMU2, and CMU3 were cloned into pMAL:c5x vector to obtain N-terminal MBP-tagged fusion proteins. The FRA1 C-terminal tail fragment (amino acids 616 to 1035) was cloned into a pTEV vector to generate a C-terminal 6×His fusion construct. Plasmids were transformed into BL21-DE3-RIPL–competent cells (Agilent) and induced with 0.5 mM isopropyl β-d-1-thiogalactopyranoside for 4 h. Cells were lysed with Pierce B-PER Bacterial Protein Extraction Reagents (1 mL of B-Per, 2 μL of DNase1, 3 μL of lysozyme, 1× protease inhibitor, and 1 mM phenylmethylsulfonyl fluoride; Thermo Fisher Scientific). For the pull-down experiments, 40 μL of MBP-Trap agarose beads (Chromotek) was first incubated with either MBP, MBP-CMU1, MBP-CMU2, or MBP-CMU3 (∼5 µg each) at 4°C on a shaker (speed, 10 rpm) for 4 to 6 h to allow binding. Next, purified FRA1 tail fragment was added (∼5 µg) into each tube and incubated for an additional 10 to 12 h as before. Subsequently, the beads were washed at least five times with 1× PBS + 0.1% (v/v) Tween 20. Bound proteins were isolated from the beads by adding 1× SDS loading dye, separated by SDS-PAGE, and transferred to 0.45-µm polyvinylidene fluoride membrane (Thermo Fisher Scientific). The blots were probed with monoclonal anti-MBP antibody DSHB-MBP-2A1 (1:5000) from Developmental Studies Hybridoma Bank, and a polyclonal anti-FRA1 antibody (1:2000) as described previously (Zhu et al., 2015). For secondary antibodies, anti-rabbit IgG horseradish peroxidase (HRP, 1:10,000, catalog no. 711-035-152; Jackson Immuno Research Laboratories) and anti-mouse IgG HRP (1:10,000, catalog no. 715-035-151; Jackson Immuno Research Laboratories) were used. Detection was conducted using SuperSignal West Dura chemiluminescence substrate (Thermo Fisher Scientific). CKL3 and CKL6 were cloned into the pMAL:c5x vectors to obtain N-terminal MBP-tagged fusion proteins. Pull-down experiments with 6×His-tagged FRA1 tail fragment were performed using the above-mentioned protocol.

In Vivo Pull-Down Experiments

Native promoter-driven CMU1 and CMU2 constructs were introduced into the pCAMBIA3300 vector using primers listed in Supplemental Table. The constructs were transformed into Col-0 plants, and homozygous transgenic plants were selected. For the pull-down assay, 7-d-old seedlings were homogenized in a mortar using lysis buffer (20 mM Hepes, pH 7.5, 40 mM KCl, 1 mM EDTA, 0.1% [v/v] Triton X-100, 1× protease inhibitor cocktail, 1 mM phenylmethylsulfonyl fluoride, and 10% [v/v] glycerol). The homogenate was then centrifuged at 4°C twice at 15,000g for 10 min and incubated with 20 μL of anti-mCherry Trap agarose beads (Chromotek) for 12 to 16 h at 4°C on a rocking shaker. Beads were then washed at least five times with 1× PBS + 0.1% (v/v) Tween 20 solution. Bound proteins were isolated from the beads by adding 1× SDS loading dye, separated by SDS-PAGE, and detected by immunoblotting using anti-mCherry (Living Colors mCherry Monoclonal Antibody, catalog no. 632543; Clontech Laboratories) and FRA1 antibody.

Pollen Viability Tests

Pollen from six fully open flowers of 4- to 5-week-old plants was incubated in 50 μL of pollen germination medium (10% [w/v] Suc, 0.01 mM boric acid, 5 mM CaCl2, 5 mM KCl, and 1 mM MgSO4, pH 7.5) on a slide for 10 min and then stained with 1 µg/mL propidium iodide for 15 min. Stained pollen was imaged using a UV light-fluorescence microscope with red fluorescent protein filter and 20× lens.

Pollination and Aniline Blue Staining

Pistils from stage 12 flowers were exposed by removing sepals, petals, and stamens and kept upright in 1% (w/v) agar plates. Any pistils with pollen already on them were discarded. Virgin pistils were pollinated using anthers from fully open flowers. After 2 h at room temperature, pistils were fixed using acetic acid:ethanol (3:1; v/v) at 60°C for 5 min. The fixative solution was replaced with 1 N NaOH and incubated at 60°C for 15 min to soften the tissue. After rinsing with water, the samples were stained with 0.1% (w/v) decolorized aniline blue, which stains callose. Stained pistils were imaged under a UV light-fluorescence microscope with a 4′,6-diamidino-2-phenylindole filter. Pollen tube length was measured from the stigma tip to where the majority of pollen tubes ended within the style using the Fiji ImageJ package (Schindelin et al., 2012). For each pistil, the pollen tube lengths were measured three times, and the average was used.

In Vitro Kinase Assay

The FRA1 tail fragment (amino acids 616 to 1035) was cloned into the pMAL:c5x vector (MBP tag) with Sal1 and EcoR1 enzymes, and CKL3 and CKL6 were cloned in the pTEV vector (6×His tag) with Nde1 and BamH1 sites. For the kinase assay, MBP, MBP-FRA1-tail, and MBP-FRA1(AA)-tail were used as potential substrates for CKL3-6×His and CKL6-6×His. For each assay, 2 µg of substrate and 500 ng of kinase were incubated in 1× kinase buffer (25 mM Tris-HCl, pH 7.5, 2 mM DTT, 50 µM ATP, and 12 mM MgCl2) supplemented with 50 µM ATP and 1× phosphatase inhibitor (phosphatase inhibitor complex set III, catalog no. 524627; EMD Biosciences) in a 50-μL reaction. Reactions were performed at 30°C for 45 min and then terminated by adding 4× SDS-PAGE loading buffer and heated at 75°C for 6 to 8 min. Bands were separated by SDS-PAGE and detected via immunoblotting using anti–phospho-Ser/Thr antibody (rabbit polyclonal anti-phospho-Ser-Thr antibody, catalog no. PP2551ECM; EMD Biosciences) at 1:2500 dilution and anti-rabbit goat secondary antibody with HRP at 1:10,000 dilution.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: FRA1 (At5g47820); CMU1 (At4g10840); CMU2 (At3g27960); CMU3 (At1g27500); CKL3 (At4g28880); CKL6 (At4g28540).

Supplemental Data

DIVE Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

AUTHOR CONTRIBUTIONS

A.G., C.Z., and R.D. designed the research and analyzed the data. A.G., C.Z., and W.C. performed the research. A.G. and R.D. wrote the article. All authors read and approved the final version of the article.

Acknowledgments

We thank Graham Burkart for the mRuby-TUB6 microtubule marker line used in this study, Katharina Bürstenbinder for sharing unpublished data on FRA1–CMU interaction, and Yanbing Wang for help with pollination experiments. This work was supported by National Science Foundation (grant 1453726 to R.D.).

References

  1. Alonso J.M., et al. (2003). Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301: 653–657. [DOI] [PubMed] [Google Scholar]
  2. Anderson C.T.(2018). Finding order in a bustling construction zone: Quantitative imaging and analysis of cell wall assembly in plants. Curr. Opin. Plant Biol. 46: 62–67. [DOI] [PubMed] [Google Scholar]
  3. Barlan K., Lu W., Gelfand V.I.(2013). The microtubule-binding protein ensconsin is an essential cofactor of kinesin-1. Curr. Biol. 23: 317–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ben-Nissan G., Cui W., Kim D.J., Yang Y., Yoo B.C., Lee J.Y.(2008). Arabidopsis casein kinase 1-like 6 contains a microtubule-binding domain and affects the organization of cortical microtubules. Plant Physiol. 148: 1897–1907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bieling P., Telley I.A., Surrey T.(2010). A minimal midzone protein module controls formation and length of antiparallel microtubule overlaps. Cell 142: 420–432. [DOI] [PubMed] [Google Scholar]
  6. Blom N., Sicheritz-Pontén T., Gupta R., Gammeltoft S., Brunak S.(2004). Prediction of post-translational glycosylation and phosphorylation of proteins from the amino acid sequence. Proteomics 4: 1633–1649. [DOI] [PubMed] [Google Scholar]
  7. Bringmann M., Li E., Sampathkumar A., Kocabek T., Hauser M.T., Persson S.(2012). POM-POM2/cellulose synthase interacting1 is essential for the functional association of cellulose synthase and microtubules in Arabidopsis. Plant Cell 24: 163–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bürstenbinder K., Savchenko T., Müller J., Adamson A.W., Stamm G., Kwong R., Zipp B.J., Dinesh D.C., Abel S.(2013). Arabidopsis calmodulin-binding protein IQ67-domain 1 localizes to microtubules and interacts with kinesin light chain-related protein-1. J. Biol. Chem. 288: 1871–1882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cai D., Hoppe A.D., Swanson J.A., Verhey K.J.(2007). Kinesin-1 structural organization and conformational changes revealed by FRET stoichiometry in live cells. J. Cell Biol. 176: 51–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chaudhary A.R., Lu H., Krementsova E.B., Bookwalter C.S., Trybus K.M., Hendricks A.G.(2019). MAP7 regulates organelle transport by recruiting kinesin-1 to microtubules. J. Biol. Chem. 294: 10160–10171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cheung A.Y., Wang H., Wu H.M.(1995). A floral transmitting tissue-specific glycoprotein attracts pollen tubes and stimulates their growth. Cell 82: 383–393. [DOI] [PubMed] [Google Scholar]
  12. Crowell E.F., Bischoff V., Desprez T., Rolland A., Stierhof Y.D., Schumacher K., Gonneau M., Höfte H., Vernhettes S.(2009). Pausing of Golgi bodies on microtubules regulates secretion of cellulose synthase complexes in Arabidopsis. Plant Cell 21: 1141–1154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. de Keijzer J., Kieft H., Ketelaar T., Goshima G., Janson M.E.(2017). Shortening of microtubule overlap regions defines membrane delivery sites during plant cytokinesis. Curr. Biol. 27: 514–520. [DOI] [PubMed] [Google Scholar]
  14. Dixit R., Ross J.L., Goldman Y.E., Holzbaur E.L.(2008). Differential regulation of dynein and kinesin motor proteins by tau. Science 319: 1086–1089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Durek P., Schmidt R., Heazlewood J.L., Jones A., MacLean D., Nagel A., Kersten B., Schulze W.X.(2010). PhosPhAt: The Arabidopsis thaliana phosphorylation site database. An update. Nucleic Acids Res. 38: D828–D834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fukuda H.(1997). Tracheary element differentiation. Plant Cell 9: 1147–1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ganguly A., DeMott L., Dixit R.(2017). The Arabidopsis kinesin-4, FRA1, requires a high level of processive motility to function correctly. J. Cell Sci. 130: 1232–1238. [DOI] [PubMed] [Google Scholar]
  18. Ganguly A., DeMott L., Zhu C., McClosky D.D., Anderson C.T., Dixit R.(2018). Importin-β directly regulates the motor activity and turnover of a kinesin-4. Dev. Cell 44: 642–651.e5. [DOI] [PubMed] [Google Scholar]
  19. Gutierrez R., Lindeboom J.J., Paredez A.R., Emons A.M., Ehrhardt D.W.(2009). Arabidopsis cortical microtubules position cellulose synthase delivery to the plasma membrane and interact with cellulose synthase trafficking compartments. Nat. Cell Biol. 11: 797–806. [DOI] [PubMed] [Google Scholar]
  20. Hamada T.(2014). Microtubule organization and microtubule-associated proteins in plant cells. Int. Rev. Cell Mol. Biol. 312: 1–52. [DOI] [PubMed] [Google Scholar]
  21. Heins S., Song Y.H., Wille H., Mandelkow E., Mandelkow E.M.(1991). Effect of MAP2, MAP2c, and tau on kinesin-dependent microtubule motility. J. Cell Sci. Suppl. 14: 121–124. [DOI] [PubMed] [Google Scholar]
  22. Hiscock S., Bown D., Gurr S., Dickinson H.(2002). Serine esterases are required for pollen tube penetration of the stigma in Brassica. Sex. Plant Reprod. 15: 65–74. [Google Scholar]
  23. Hiscock S., Dewey F., Coleman J., Dickinson H.(1994). Identification and localization of an active cutinase in the pollen of Brassica napus L. Planta 193: 377–384. [Google Scholar]
  24. Hooikaas P.J., et al. (2019). MAP7 family proteins regulate kinesin-1 recruitment and activation. J. Cell Biol. 218: 1298–1318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Horio T., Murata T.(2014). The role of dynamic instability in microtubule organization. Front. Plant Sci. 5: 511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kandasamy M., Nasrallah J., Nasrallah M.(1994). Pollen-pistil interactions and developmental regulation of pollen tube growth in Arabidopsis. Development 120: 3405–3418. [Google Scholar]
  27. Kong Z., Ioki M., Braybrook S., Li S., Ye Z.H., Julie Lee Y.R., Hotta T., Chang A., Tian J., Wang G., Liu B.(2015). Kinesin-4 functions in vesicular transport on cortical microtubules and regulates cell wall mechanics during cell elongation in plants. Mol. Plant 8: 1011–1023. [DOI] [PubMed] [Google Scholar]
  28. Konopka C.A., Bednarek S.Y.(2008). Variable-angle epifluorescence microscopy: A new way to look at protein dynamics in the plant cell cortex. Plant J. 53: 186–196. [DOI] [PubMed] [Google Scholar]
  29. Lampugnani E.R., Khan G.A., Somssich M., Persson S.(2018). Building a plant cell wall at a glance. J. Cell Sci. 131: jcs207373. [DOI] [PubMed] [Google Scholar]
  30. Lavithis M., Bhalla P.(1995). Esterases in pollen and stigma of Brassica. Sex. Plant Reprod. 8: 289–298. [Google Scholar]
  31. Li S., Lei L., Somerville C.R., Gu Y.(2012). Cellulose synthase interactive protein 1 (CSI1) links microtubules and cellulose synthase complexes. Proc. Natl. Acad. Sci. USA 109: 185–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liu Z., Schneider R., Kesten C., Zhang Y., Somssich M., Zhang Y., Fernie A.R., Persson S.(2016). Cellulose-microtubule uncoupling proteins prevent lateral displacement of microtubules during cellulose synthesis in Arabidopsis. Dev. Cell 38: 305–315. [DOI] [PubMed] [Google Scholar]
  33. McFarlane H.E., Young R.E., Wasteneys G.O., Samuels A.L.(2008). Cortical microtubules mark the mucilage secretion domain of the plasma membrane in Arabidopsis seed coat cells. Planta 227: 1363–1375. [DOI] [PubMed] [Google Scholar]
  34. Oda Y.(2015). Cortical microtubule rearrangements and cell wall patterning. Front. Plant Sci. 6: 236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Okuda S., et al. (2009). Defensin-like polypeptide LUREs are pollen tube attractants secreted from synergid cells. Nature 458: 357–361. [DOI] [PubMed] [Google Scholar]
  36. Paredez A.R., Somerville C.R., Ehrhardt D.W.(2006). Visualization of cellulose synthase demonstrates functional association with microtubules. Science 312: 1491–1495. [DOI] [PubMed] [Google Scholar]
  37. Samuel M.A., Chong Y.T., Haasen K.E., Aldea-Brydges M.G., Stone S.L., Goring D.R.(2009). Cellular pathways regulating responses to compatible and self-incompatible pollen in Brassica and Arabidopsis stigmas intersect at Exo70A1, a putative component of the exocyst complex. Plant Cell 21: 2655–2671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schindelin J., et al. (2012). Fiji: An open-source platform for biological-image analysis. Nat. Methods 9: 676–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Seitz A., Kojima H., Oiwa K., Mandelkow E.M., Song Y.H., Mandelkow E.(2002). Single-molecule investigation of the interference between kinesin, tau and MAP2c. EMBO J. 21: 4896–4905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Takenaka Y., Watanabe Y., Schuetz M., Unda F., Hill J.L. Jr., Phookaew P., Yoneda A., Mansfield S.D., Samuels L., Ohtani M., Demura T.(2018). Patterned deposition of xylan and lignin is independent from that of the secondary wall cellulose of Arabidopsis xylem vessels. Plant Cell 30: 2663–2676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Tan S.T., Dai C., Liu H.T., Xue H.W.(2013). Arabidopsis casein kinase1 proteins CK1.3 and CK1.4 phosphorylate cryptochrome2 to regulate blue light signaling. Plant Cell 25: 2618–2632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tymanskyj S.R., Yang B.H., Verhey K.J., Ma L.(2018). MAP7 regulates axon morphogenesis by recruiting kinesin-1 to microtubules and modulating organelle transport. eLife 7: e36374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Verhey K.J., Lizotte D.L., Abramson T., Barenboim L., Schnapp B.J., Rapoport T.A.(1998). Light chain-dependent regulation of kinesin’s interaction with microtubules. J. Cell Biol. 143: 1053–1066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Vershinin M., Carter B.C., Razafsky D.S., King S.J., Gross S.P.(2007). Multiple-motor based transport and its regulation by Tau. Proc. Natl. Acad. Sci. USA 104: 87–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wong Y.H., Lee T.Y., Liang H.K., Huang C.M., Wang T.Y., Yang Y.H., Chu C.H., Huang H.D., Ko M.T., Hwang J.K.(2007). KinasePhos 2.0: A web server for identifying protein kinase-specific phosphorylation sites based on sequences and coupling patterns. Nucleic Acids Res. 35: W588-94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wong Y.L., Rice S.E.(2010). Kinesin’s light chains inhibit the head- and microtubule-binding activity of its tail. Proc. Natl. Acad. Sci. USA 107: 11781–11786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zhu C., Dixit R.(2011). Single molecule analysis of the Arabidopsis FRA1 kinesin shows that it is a functional motor protein with unusually high processivity. Mol. Plant 4: 879–885. [DOI] [PubMed] [Google Scholar]
  48. Zhu C., Ganguly A., Baskin T.I., McClosky D.D., Anderson C.T., Foster C., Meunier K.A., Okamoto R., Berg H., Dixit R.(2015). The fragile Fiber1 kinesin contributes to cortical microtubule-mediated trafficking of cell wall components. Plant Physiol. 167: 780–792. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Plant Cell are provided here courtesy of Oxford University Press

RESOURCES