Phorbolester-activated Munc13-1 and ubMunc13-2 exert opposing effects on dense-core vesicle secretion

  1. Sébastien Houy
  2. Joana S Martins
  3. Noa Lipstein
  4. Jakob Balslev Sørensen

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    Houy and co-workers investigated the function of Munc13-1 and ubMunc13-2 in chromaffin cells and the interaction with phorbol esters (PMA). They combined calcium uncaging, capacitance measurements, amperometry, and activity-dependent movements of the EGFP-labeled Munc13 proteins. This study reveals that phorbolesters have a stimulatory effect via ubMunc13-2 but an inhibitory effect via Munc13-1. These opposing effects of the two Munc13 paralogs are surprising considering the closely related domain architectures.

    (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. The reviewers remained anonymous to the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Munc13 proteins are priming factors for SNARE-dependent exocytosis, which are activated by diacylglycerol (DAG)-binding to their C1-domain. Several Munc13 paralogs exist, but their differential roles are not well understood. We studied the interdependence of phorbolesters (DAG mimics) with Munc13-1 and ubMunc13-2 in mouse adrenal chromaffin cells. Although expression of either Munc13-1 or ubMunc13-2 stimulated secretion, phorbolester was only stimulatory for secretion when ubMunc13-2 expression dominated, but inhibitory when Munc13-1 dominated. Accordingly, phorbolester stimulated secretion in wildtype cells, or cells overexpressing ubMunc13-2, but inhibited secretion in Munc13-2/ Unc13b knockout (KO) cells or in cells overexpressing Munc13-1. Phorbolester was more stimulatory in the Munc13-1/ Unc13a KO than in WT littermates, showing that endogenous Munc13-1 limits the effects of phorbolester. Imaging showed that ubMunc13-2 traffics to the plasma membrane with a time-course matching Ca 2+ -dependent secretion, and trafficking is independent of Synaptotagmin-7 (Syt7). However, in the absence of Syt7, phorbolester became inhibitory for both Munc13-1 and ubMunc13-2-driven secretion, indicating that stimulatory phorbolester x Munc13-2 interaction depends on functional pairing with Syt7. Overall, DAG/phorbolester, ubMunc13-2 and Syt7 form a stimulatory triad for dense-core vesicle priming.

Article activity feed

  1. Version published to 10.7554/elife.79433 on eLife
  2. eLife

    Evaluation Summary:

    Houy and co-workers investigated the function of Munc13-1 and ubMunc13-2 in chromaffin cells and the interaction with phorbol esters (PMA). They combined calcium uncaging, capacitance measurements, amperometry, and activity-dependent movements of the EGFP-labeled Munc13 proteins. This study reveals that phorbolesters have a stimulatory effect via ubMunc13-2 but an inhibitory effect via Munc13-1. These opposing effects of the two Munc13 paralogs are surprising considering the closely related domain architectures.

    (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. The reviewers remained anonymous to the authors.)

  3. eLife

    Reviewer #1 (Public Review):

    Houy and co-workers investigated the function of Munc13-1 and ubMunc13-2 in chromaffin cells and the interaction with phorbolesters (PMA). They combined calcium uncaging, capacitance measurements, amperometry, and activity dependent movements of the EGFP-labeled Munc13 proteins. The most exciting finding is that phorbolesters have a stimulatory effect via ubMunc13-2 but an inhibitory effect via Munc13-1.

    The effects of phorbolester on release were studied in mutants lacking either Munc13-1 or ubMunc13-2 and in cells over-expressing either Munc13-1 or ubMunc13-2. Both approaches consistently show that phorbolesters facilitate release only in the presence of ubMunc13-2 but inhibit release when Munc13-1 dominates. The data also indicate another interesting difference between both isoforms. Only ubMunc13-2 trafficked to the plasma membrane with a time course matching secretion after flash-evoked calcium increase. To investigate if Munc13-1 exerts its inhibitory function in the presence of phorbolester only by displacing the apparently more potent ubMunc13-2 or really has itself an inhibitory function, Munc13-1-EGFP was expression in cells lacking ubMunc13-2. The data indicated that for the initial phase of release (burst) the inhibitory function of Munc13-1 in the presence of phorbolester might be due to a displacement of ubMunc13-2, but for the sustained release Munc13-1 has an inhibitory effect in the presence of phorbolester. Finally, the interaction with Syt7 was investigated. The data indicate that only ubMunc13-2 can interact productively with Syt7. In the absence of Syt7, both ubMunc13-2 and Munc13-1 have a negative effect when interacting with phorbolester.

    Although it remains unclear how these findings relate to release of synaptic vesicles, the study provides important mechanistic differences between the two isoforms of this central priming protein. The manuscript is clearly written and the conclusions are justified.

  4. eLife

    Reviewer #2 (Public Review):

    In this manuscript, Houy et al. studied the interactions of phorbolesters with Munc13-1 and ubMunc13-2 in dense core vesicle secretion in mouse adrenal chromaffin cells. Using calcium uncaging and capacitance measurement of secretion in chromaffin cells, they have identified that phorbolesters exposure enhances secretion when Munc13-2 is dominant but inhibits secretion when Munc13-1 is dominantly expressed. Phorbolesters positively regulate RRP in the presence of ubMunc13-2. To strengthen the conclusion, they have done the experiments in Munc13-1 or Munc13-2 knockout and with overexpression of Munc13-1 or ubMunc13-2. Using live-cell imaging, they further showed that calcium-dependent ubMunc13-2 translocation to the plasma membrane is independent of Syt7, but Syt7 seems to be involved in the augmentation of secretion of phorbolesters mediated by ubMunc13-2. The study's strengths are to use rigorous methods to establish an exciting phenomenon, i.e. Munc13-1 and ubMunc13-2 differentially mediate phorbol ester impact on dense core vesicle release. The weakness is the mechanism of such a divergent function mediated by highly domain conserved proteins, i.e. Munc13-1 vs. unMunc13-2 is still not known. Overall, these experiments are well done, and the conclusion appears to be justified.

    1. The conserved domain structures of Munc13-1 and ubMun13-2 are remarkably similar. The result from this study is somewhat surprising but indeed interesting. Mechanistically, I still feel a little bit struggling by claiming that " plasma membrane targeting of Munc13-1 is inhibitory for chromaffin cell DV secretion" (line 498). After all, overexpression of Munc13-1 facilitates the release in wild-type cells (Figure 4-S1). Should we expect this overexpressed Munc13-1 not to increase the Ca2+-dependent membrane targeting? Moreover, the author also acknowledged that in the Rosenmund et al. 2002 paper, they had shown that phorbol ester potentiated glutamate release (actually by more than two folds). I wonder if any mechanistic studies can be conducted. For example, whether the H567K Munc13-1 (Rhee Cell 2002), which is phorbol ester binding deficient Munc13-1, abolishes the inhibitory effect of phorbol ester in chromaffin cells?
    2. For imaging analysis, it is unclear how the membrane portion was determined. How do the authors determine the inside intensities? How to choose the confocal images to quantify the integrated density shown in figure 1M?
    3. From the result, it appears that PMA itself can translocate Muncs to the plasma membrane (Figure 4 &7), which might be more potent than calcium-mediated membrane targeting Figure 6 vs. 7. The expression level of protein expression mediated by the Smiliki viruses is very difficult to control. For data shown in figure 4A, are those fluorescence aggregates inside the cell? Moreover, can the author show time-lapse images of the translocation of Muncs to membrane-mediated by PMA?
    4. The choice of statistical analyses should be reconsidered. For example, they used non-parametric Mann-Whitney tests for most of the data but did not use the student t-test. In figure 2, they used the Kruskal-Wallis test, which is a one-way ANOVA but they have genotype differences and also the effect of PMAs, two independent variables. I suggest the authors consult with a statistician for the analysis. I found Ho et al. "Moving beyond P values: data analysis with estimation graphics" Nature Methods 16, 565-566 (2019) to be useful.

  5. eLife

    Reviewer #3 (Public Review):

    In this manuscript Houy and coworkers report new experiments regarding the role of phorbolester-activated Munc13 paralogs, Munc13-1 and ubMunc13-2, on the secretion response of mouse chromaffin cells. They report that expression of either paralog enhanced secretion. Using single knock outs (Figs. 1, 2) or with the expression of either paralog (Figs 3, 4) they found that treatment with the phorbolester PMA was stimulatory when ubMunc13-2 was the predominating paralog, but inhibitory when Munc13-1 dominated. The opposing PMA effects in the presence of either Munc13-1 or ubMunc13-2 were interpreted in the context of a potential competition of both proteins in essential priming reactions (Fig. 5). In simultaneous fluorescence recordings of EGFP tagged Munc13 variants they studied the Ca2+- and PMA-dependent translocation of Munc13 to the plasma membrane (PM). They found that only Munc13-2 (Fig. 3) but not Munc13-1 (Figs. 4, 5) is translocated to the PM in response an intracellular Ca2+-elevation. In this context, they also report that Ca2+ -dependent recruitment of ubMunc13-2 is independent of Synaptotagmin-7 (Fig. 6) and that in the absence of Synaptotagmin-7, ubMunc13-2-dependent secretion is inhibited by PMA (Fig. 7). Based on these results the authors argue that ubMunc13-2, Synaptotagmin-7 and DAG/phorbolester form a stimulatory entity to facilitate dense core vesicle fusion.
    Although the manuscript presents interesting observations, some conclusions appear to be compromised by methodological and conceptual concerns.

    Major criticism
    1. In order to track Munc13 translocation the authors have chosen EGFP-tagged variants which overlap in the emission with the standard FuraII/Furaptra emission. Consequently, the authors omitted Ca2+-imaging in these experiments and thereby lost crucial information regarding the development of [Ca]I before and after the uncaging flash. These parameters are of central importance for the Ca2+-dependent priming and exocytosis timing, respectively. This is particularly worrisome, because in several experiments with Munc13 expression hardly any RRP component is apparent in the displayed capacitance traces, which may indicate insufficient Ca2+-dependent vesicle priming (Fig. 4). Under proper calcium control, both Ashery et al 2000 (Fig. 2) and Betz et al 2001 (Fig. 6) reported that Munc13-1 overexpression in wt chromaffin cells causes at least a 300% increase in the size of the EB compared to wt cells. Performing the same experiment, but without calcium imaging, the authors in Fig4-Sup1 show hardly any increase in the size of the EB (violet trace Fig4-Sup1) but a rather strong increase in the sustained phase of exocytosis, a phenotype that could be a result of low intracellular pre-flash calcium levels leading to insufficient vesicle priming. I do not understand why the authors have not chosen any other red-shifted protein tag to prevent such uncertainties. Furthermore, the display of the capacitance traces in several figures does not allow the appreciation of changes in the EB size or its components (e.g. RRP).
    2. The authors speculate about the possibility, that PMA treatment PMA-treatment of Unc13b KO cells may lead to spontaneous release, depleting the cells of secretory vesicles. To test this, they determined the integrated CgA-fluorescence over the entire cell (Fig. 1M, N) rather than analyzing submembrane CgA-fluorescence. With the latter strategy, they will be able to focus on a potential subcellular depletion of release-ready vesicles.
    3. After showing a detailed analysis of the exocytotic burst components and their kinetics in Fig. 1 and 2 the authors argue on page 9 Line 275 'Since the measurements above indicated that the main effect of PMA is on secretion amplitude, not kinetics (see also (Nagy et al., 2006)), we only distinguished between burst secretion (first 1s secretion after Ca2+ uncaging, corresponding approximately to RRP and SRP fusion) and sustained secretion (last 4 s of secretion), as well as total secretion (the sum of burst and sustained release). '
    I have some concerns with this argumentation because the expression of Munc13 paralogs apparently leads to changes in the burst components and/or it kinetics (e.g. Fig. 4B compare to Fig. 1 or 2). In fact, these differences cannot be directly appreciated, because experiments like in Fig. 3 and 4 lack the littermate wt control without and with PMA.
    Moreover, Munc13 expression leads to a disproportionate increase in the sustained phase of release, which is not present with PMA.
    I would recommend at least to include detailed analyses of the exocytotic burst components and their kinetics to address these uncertainties.

    4. As central hypothesis, the authors propose that they have identified a unique stimulatory triad of ubMunc13-2, Syt7 and DAG/phorbolesters, which is needed for dense core vesicle priming and fusion. For example, in contrast to the behavior of wt cells (e.g. Fig 1A) phorbolester treatment becomes inhibitory in cells lacking Syt7 and expressing ubMunc13-2 (Fig. 7). Nonetheless, previously published data by Sorensen's group, obtained under similar preflash [Ca]I conditions (Tawfik et al., 2021; Fig 6-figure supplement 2 E-H), clearly show that PMA strongly potentiates exocytosis even in the absence of Syt7. Therefore, these previous findings by Tawfik et al. clearly counter the central hypothesis of the manuscript. The authors should clarify these disparate results.

  6. Version published to 10.1101/2022.04.27.489716v1 on bioRxiv