Article activity feed

  1. Evaluation Summary:

    Phosphoinositide phosphates (PIPs) are lipids that can convey distinct identities to different cellular membranes via different phosphorylation patterns. Here, Doumane and co-authors document the effects of the previously-characterized sac9 mutant, affecting a putative PIP-5-phosphatase in Arabidopsis, on PIP localization and endocytic trafficking. This work confirms that disrupting PI(4,5)P2 localization can affect endocytic trafficking in plants and will be of interest to the plant and cell biology research fields.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 and Reviewer #3 agreed to share their name with the authors.)

  2. Reviewer #1 (Public Review):

    The authors generate a fluorescently-tagged SAC9 that can complement sac9 mutants (Figure 1) and is localized to the cytosol and to small intracellular puncta (Figure 2) that colocalize with markers for the trans-Golgi network (TGN) (Figure 3). They also demonstrate that mutation of C459, a predicted catalytic residue in SAC9, causes SAC9(C459A) mislocalization and fails to complement the growth phenotype of sac9 mutants (Figures 1-3). Further characterization of sac9 subcellular phenotypes reveals that PI(4,5)P2 is mislocalized from the PM into intracellular structures and that PI(4)P and PS localization are also affected (Figure 4). They document endocytosis defects in sac9 mutants, including changes to FM4-46 tracer dye uptake, synergistic interactions with endocytosis inhibitors (Figure 6), and changes to the localization of a clathrin-mediated endocytosis marker (Figure 8). Finally, the authors present preliminary data that SAC9 might interact with SH3P2, a protein that may be involved in autophagy, endocytosis, and other intracellular trafficking processes (Figure 7).

    The authors present a model (Figure 8) in which SAC9 could associate with PI(4,5)P2 on endocytic vesicles through its interaction with SH3P2 to convert PI(4,5)P2 to PI(4)P, thus allowing endocytic vesicles to fuse with target membranes.

    The main strength of this work is the detailed documentation that loss of sac9 affects PI(4,5)P2, which provides the most direct evidence to-date that SAC9 affects PIPs.

    The main weaknesses are that the results are somewhat predictable given what is known about PI(4,5)P2 and endocytosis in other organisms, and that some of their other major claims of their model are not fully supported by the data presented here:

    1. The authors claim that "the phosphoinositide phosphatase activity of SAC9 is required for its function" but they have not documented any biochemical activity of SAC9. They say that these experiments were unsuccessful, but no supporting data are presented and instead, they rely on the CIT-SAC9(C459A) mutant, which they assume is catalytically inactive. It would support their claims that CIT-SAC9(C459A) is catalytically inactive to document similar changes to PI(4,5)P2 distribution in sac9 mutants carrying CIT-SAC9(C459A) and to document similar levels of CIT-SAC9(C459A) or CIT-SAC9 protein in their complementation lines.

    2. The authors claim that SAC9 labels "a subpopulation of endosomes close to the plasma membrane" but they have not documented this PM proximity and there are clearly SAC9-labelled puncta throughout the cytoplasm (e.g. Figure 2D).

    3. The interaction with SH3P2 is not well supported, relying on only Y2H data of truncated versions of both proteins. No attempts to verify this interaction using full-length clones or any independent method are documented. It is also unclear what role SH3P2 plays in plants, or even whether it is directly involved in endocytosis, since it has also been implicated in multivesicular body formation (Nagel et al 2017 PNAS), cell plate formation (Ahn et al 2017 Plant Cell), autophagy (Zhuang et al 2013 Plant Cell), which the authors do not assess in the sac9 mutants.

  3. Reviewer #2 (Public Review):

    Using a combination of genetic and imaging approaches Doumane et al. provide evidence that 1) SAC9 localizes to the trans-Golgi network/early endosomal system (TGN/EE), 2) SAC9 plays a role in maintaining the subcellular distribution of PI(4,5)P2 to the plasma membrane and 3) loss of SAC9 results in impaired endocytosis of the lipophilic dye, FM4-64 and the auxin efflux carrier, PIN2-GFP. Consistent with these endocytic defects, the sac9 loss-of-function mutant displayed impaired dynamics of the TPLATE endocytic adapter protein complex, as well as the localization of the SH3P2, which was identified here as an interactor of SAC9. Overall, this study provides new insights into our understanding of SAC9 function in phosphoinositide metabolism and trafficking. Intriguingly, the findings suggest that SAC9-mediated PI(4,5)P2 hydrolysis following (or concomitantly with) its internalization via endocytosis serves to spatially restrict PI(4,5)P2 to the plasma membrane.

    The strengths of this manuscript include the quantitation of confocal microscopy images and subsequent determination of statistical significance, as well as the use of multiple marker lines to localize specific phospholipids and organelles in wild-type and sac9 backgrounds. Co-localization of wild-type SAC9 with internalized FM4-64 and clathrin light chain (CLC2) as well as its association with BFA bodies shown in Figure 3 support the authors' conclusion that SAC9 is associated with the TGN/EE. The use of multiple phosphoinositide marker lines to demonstrate that intracellular pools of PI(4,5)P2 and PI(4)P, as well as PS, but not other phospholipids, is well-done (Figures 4A-4B). Furthermore, the authors clearly demonstrate that PI(4,5)P2 and PI(4)P do not localize to the same intracellular puncta (Figures 4D-4F). The authors effectively explain the broader significance of their work and describe the limitations (i.e. the authors recognize the caveats that the biochemical activity of SAC9 in vitro and function of SH3P2 in endocytosis remain to be defined).

    One weakness of the manuscript is the authors' claim that mCIT-SAC9 and the catalytically inactive mCIT-SAC9C459A localize to a spatially restricted TGN/EE subpopulation in the cell cortex (Figures 2 and 3). The qualitative and quantitative analysis, however, appear to have been only conducted on a single plane of focus. Also, it is not clear from the methods whether the fluorescent 'cortical' signal described by the authors excludes all fluorescent signal from the interior of the cell. The authors should present quantitative imaging of SAC9 fusion protein distribution at multiple regions (Z planes of focus) of the cell to support their claim that SAC9 distribution is spatially restricted.

    As stated above, the results of the BFA and FM4-64 colocalization experiments in Figures 3A and 3C are consistent with the authors' conclusion that the wild-type SAC9 is associated with the TGN/EE. However, there is a disconnect between the representative images of the colocalization analyses of mCIT-SAC9 and mCIT-SAC9C459A with various endomembrane marker proteins and the quantitation shown in Figure 3E-3H which undermine the confidence in the authors' conclusion that the intracellular compartments labeled by SAC9 are TGN/EE. On one level, colocalization has been described using markers which are not quantitated (e.g. RabF2a in panel 3H) or which are quantitated but not depicted (e.g. VTI12, 3E-3F, line 137). Furthermore, the images in panel H showing the merged co-localizations of RabD1 and Got1p with mCIT-SAC9C459A show a number of puncta with signal from both fluorophores, which is not evident in the quantitative data shown in Figure 3G. I understand that the colocalization studies are challenging due to the high signal of mCIT-SAC9 in the cytosol. Nevertheless, given the high cytosolic signal it would be nice to see imaging and quantitation of a control cytosolic protein (e.g. mCIT alone) for comparison in the mCIT-SAC9 colocalization studies. In addition, the authors should present representative images of mCIT-SAC9 and mCIT-SAC9C459A colocalization analysis that more convincingly reflect the quantitative colocalization analyses.

    It is certainly plausible that the endocytic defects including reduction in levels of PM- associated SH3P2 and TPLATE in the sac9 mutant cells are directly related to the loss of SAC9 and/or altered PI(4,5)P2 distribution (as implied in lines 286-288). However, the authors cannot rule out that loss of SAC9 and/or altered metabolism of PI(4,5)P2 may indirectly affect SH3P2 localization and/or TPLATE recruitment/dynamics at the plasma membrane due to alterations in the organization/function of the TGN/EE and/or general inhibition of post-Golgi trafficking. The authors have the necessary tools and expertise including the TGN/EE markers and assays to address these questions (e.g. analysis of PIN2-GFP recycling +/- cycloheximide or quantitating secretion of marker sec-GFP in sac9 vs wild-type).

  4. Reviewer #3 (Public Review):

    Doumane and colleagues describe in this paper how the phosphatase enzyme SAC9 is involved in controlling the homeostasis of the PI(4,5)P2 lipid levels at the plasma membrane and during the formation of the endocytic vesicles in root meristematic epidermal cells. They show how SAC9 is localized mainly in the cytosol and in structures close to the plasma membrane. Furthermore, their data indicate that SAC9 phosphatase activity (its catalytic cysteine 459) is necessary for an adequate clathrin-mediated endocytosis, and to keep a proper PI(4,5)P2 and PI4P lipids balance. Additionally, the authors show SAC9-SH3P2 interaction via Y2H, based on which they hypothesize an interaction of these proteins at the plasma membrane and their potential cooperation to regulate clathrin-mediated endocytosis and membrane phosphoinositide homeostasis.
    Overall, this is potentially a very interesting and high-quality work that try to unravel part of the endocytosis mechanism and to clarify the different role of the lipids that form the vesicles during the process.

    However, there are some open questions and important points for the authors to consider:
    a) Describe with images and numbers the focal planes used in the imaging,
    b) Clarify the exact cells and focal planes used to quantify the cortical early endosomes when cells in different focal planes are shown in the images (see as example cells 1 and 2 of Fig 2G),
    c) Re-analyze endocytosis events considering the intensity of signal internalized vs PM-localized in at least 3/4 cell sides in focus, and
    d) Describe the methodology followed to select the analyzed particles.