A Functional Study of AUXILIN-LIKE1 and 2, Two Putative Clathrin Uncoating Factors in Arabidopsis

Abstract

Clathrin-mediated endocytosis (CME) is a cellular trafficking process in which cargoes and lipids are internalized from the plasma membrane into vesicles coated with clathrin and adaptor proteins. CME is essential for many developmental and physiological processes in plants, but its underlying mechanism is not well characterized compared with that in yeast and animal systems. Here, we searched for new factors involved in CME in Arabidopsis thaliana by performing tandem affinity purification of proteins that interact with clathrin light chain, a principal component of the clathrin coat. Among the confirmed interactors, we found two putative homologs of the clathrin-coat uncoating factor auxilin previously described in non-plant systems. Overexpression of AUXILIN-LIKE1 and AUXILIN-LIKE2 in Arabidopsis caused an arrest of seedling growth and development. This was concomitant with inhibited endocytosis due to blocking of clathrin recruitment after the initial step of adaptor protein binding to the plasma membrane. By contrast, auxilin-like1/2 loss-of-function lines did not present endocytosis-related developmental or cellular phenotypes under normal growth conditions. This work contributes to the ongoing characterization of the endocytotic machinery in plants and provides a robust tool for conditionally and specifically interfering with CME in Arabidopsis.

Figure 1.

Identification of CLC-Interacting Proteins. (A) TAP-MS-based isolation of proteins associated with clathrin. CLC1 in C- and N-terminal fusions with TAP tags was used as bait. Two repetitions with each construct were performed. (B) BiFC-based interactions between CLC1 and selected interactors identified with TAP-MS. Pairs of proteins of interest fused with halves of the GFP molecule were transiently expressed in wild tobacco leaves and reconstituted GFP fluorescence was imaged in epidermis on the abaxial leaf side. PIP5K1 did not interact with any of the proteins tested and was used as a negative control. (C) Schematic representation of all interactions detected with BiFC. Additional interactions are presented in Supplemental Figure 1.

Figure 2.

Subcellular Localizations of the Identified Clathrin Interactors. (A) and (B) PM (A) and cell plate (B) localization of CAP1, SH3P2, AUXILIN-LIKE1, and AUXILIN-LIKE2 fluorescent protein fusions in the seedling root epidermis of stable transformants. Background cytosol staining can be observed in all lines, while CAP1 and SH3P2 also show faint nuclear staining. 35S_pro:RFP-AUXILIN-LIKE1 was identical to 35S_pro:RFP-AUXILIN-LIKE2. Additional localization data, including colocalizations with PM markers, are presented in Supplemental Figure 2B. (C) VAEM colocalization of clathrin (CLC2_pro:CLC-GFP) and AUXILIN-LIKE1 (UBQ_pro:mCherry-AUXILIN-LIKE1) in the hypocotyl epidermis. Kymographs (middle and right) show examples of auxilin-like1 recruitment at the end of CCP lifetime, which was observed in a small subset of all endocytic events (arrowheads). The majority of clathrin foci were lacking AUXILIN-LIKE1, and AUXILIN-LIKE1 foci without clathrin could be observed. See Supplemental Movie 1.

Figure 3.

AUXILIN-LIKE1/2 Overexpression Causes an Arrest of Seed Germination and Seedling Growth and Alterations in Cell Morphology. Overexpression of auxilin-likes in XVE»AUXILIN-LIKE1/2 (A) or INTAM»GFP-auxilin-like1 (B) lines inhibits or strongly delays seed germination and inhibits seedling growth if the transgene is induced postgermination. The RAMs of AUXILIN-LIKE1/2-overexpressing seedlings appear semitransparent (C) due to enlargement of vacuoles and reduction of the cytoplasmic volume (D). V, vacuole; c, cytoplasm; n, nucleus.

Figure 4.

Inhibition of FM4-64 Endocytic Tracer Uptake by AUXILIN-LIKE1/2 Overexpression. (A) The uptake of the endocytotic tracer FM4-64 into RAM epidermal cells in XVE»AUXILIN-LIKE1/2 seedlings induced for 16 to 24 h is strongly reduced. At least 10 roots per genotype were imaged in three repeated experiments. The graphs shows quantification of intracellular-to-PM FM4-64 signal ratio from 40 cells coming from four roots in a representative experiment. (B) FM4-64 uptake at early stages (8 h) of INTAM»GFP-AUXILIN-LIKE1 induction. The degree of FM4-64 uptake correlated with GFP-AUXILIN-LIKE1 levels in individual root meristem epidermal cells. Graphs show examples of this correlation in four roots. Each data point represents an individual cell. Notice that at this stage of AUXILIN-LIKE1 overexpression, cell morphology remains normal, with abundant cytosol illuminated by GFP-AUXILIN-LIKE1 expression, and small, fragmented vacuoles seen by negative staining. a.u., arbitrary units.

Figure 5.

Inhibition of Protein Cargo Endocytosis by AUXILIN-LIKE1/2 Overexpression. (A) Immunolocalization of PIN1 in the RAM stele after 90 min of BFA (25 µM) treatment. The BFA body number is markedly reduced, indicative of inhibition of PIN1 endocytosis. The graph shows collated results from three repetitions of the experiment. Each data point represents one root. (B) and (C) Following AUXILIN-LIKE2 overexpression, PIN2-GFP does not localize to BFA-induced bodies in RAM epidermis after 90 min of BFA (25 µM) treatment (B). Control experiment with CLC2-GFP used as an endosome marker (C) shows that BFA bodies can be easily observed in auxilin-like overexpressing seedlings despite the enlarged vacuoles. (D) Long-term PIN2 turnover monitoring in XVE»AUXILIN-LIKE2 with photoconvertible PIN2-Dendra fusion. PIN2-Dendra has been converted from green into red form in whole root tips at 6 h of AUXILIN-LIKE2 induction. At the 24-h time point, significant amounts of the red form have been retained at the PMs of auxilin-like overexpressing seedlings, while the protein has been internalized and degraded in the mock-treated controls. Panels at 6 h show merged green and red channels, while panels at 24 h show green and red channels separately. (E) Immunolocalization of KNOLLE in the RAM of XVE»AUXILIN-LIKE lines showing diffusion of KNOLLE out of the completed cell plates and into the old PM in late cytokinetic cells. The graph shows percentage of cells with diffused KNOLLE signals; error bars indicate sd between 3 experiments. Experiments in (A) to (C) and (E) were performed after 16 to 24 h of XVE»AUXILIN-LIKE1/2 induction.

Figure 6.

Loss of Clathrin and Dynamin from the PMs of XVE»AUXILIN-LIKE1 Line. (A) and (B) Following AUXILIN-LIKE1 overexpression, CLC2-GFP signals were absent from the PMs of RAM epidermis cells (A) and normal CLC2-GFP endocytic foci could not be observed in hypocotyl epidermis by VAEM (B). The endosomal clathrin populations were retained in both tissues. (C) to (E) DRP1C-GFP PM signals were markedly reduced in RAM epidermis (C), and the density of DRP1C-positive PM foci was reduced in the hypocotyl epidermis (D). The remaining DRP1C-positive PM foci had exceptionally long lifetimes in the majority of observed cells (E). Graphs in (A) and (C) show average ± sd percentage of roots with strong PM signals from the fluorescent markers. Graph in (D) shows collated measurements from three repetitions of the experiment. All experiments in this figure were performed after 16 to 24 h of XVE»AUXILIN-LIKE1 induction.

Figure 7.

Increased PM Binding of Endocytic Adaptor Protein Complexes in XVE»AUXILIN-LIKE1 Line. Overexpression of AUXILIN-LIKE1 caused increased PM binding of TPLATE, a subunit of the TPLATE complex, as well as AP2A1, a subunit of the AP-2 complex. CLSM images from RAM epidermis ([A] and [B]) and VAEM images from hypocotyl epidermis ([C] and [D]) are shown. Graphs in (A) and (B) show PM signal intensity measurements from a representative experiment. Kymographs ([E] and [F]) show reduced lifetimes of TPLATE and AP2A1 foci in XVE»AUXILIN-LIKE1 line, likely indicating that endocytic pits aborted after initiation. All experiments in this figure were performed after 16 to 24 h of XVE»AUXILIN-LIKE1 induction.

Figure 8.

Excess Membrane Accumulation at the PM as a Result of Inhibited Endocytosis. Overexpression of AUXILIN-LIKE2 caused excess deposition of membrane material at the PMs of root epidermis, as visualized with FM4-64 staining ([A] and [B]) or with PIN2-GFP (C). (A) and (C) are maximum intensity projections of z-stacks through outer region of the epidermis, while (B) is a longitudinal section of epidermal cells showing membrane aggregation mainly in the outer domain. FM4-64-stained deposits in Col-0 in (A) are likely remnants of apoptosed lateral root cap cells.

Figure 9.

auxilin-like1/2 Loss-of-Function Mutants. (A) Isolation of auxilin-like1/2 CRISPR mutants. Top: Overview of the AUXILIN-LIKE1/2 genomic locus showing CRISPR target sites in the second exons of AUXILIN-LIKE1 and AUXILIN-LIKE2. Bottom: Details of the CRISPR target sequence and the isolated mutant alleles. (B) auxilin-like1/2 double homozygous mutants, as well as single mutants, developed normally at both the seedling and adult stages. (C) FM4-64 uptake into auxilin-like1/2^c1 seedlings was comparable to that of the controls, indicating functional endocytosis.