Defective relocalization of ALS2/alsin missense mutants to Rac1-induced macropinosomes accounts for loss of their cellular function and leads to disturbed amphisome formation

ALS2 physically interacts with ALS2 by anti tag coimmunoprecipitation (View interaction) Mutations in the ALS2 gene account for several recessive motor neuron diseases (MNDs), including amyotrophic lateral sclerosis 2 [1], demonstrating that the ALS2 protein (ALS2 or alsin) is essential for motor neurons. ALS2 contains three predicted guanine nucleotide exchange factor (GEF) domains [2, 3], i.e., the regulator of chromosome condensation 1 (RCC1)-like domain (RLD), the Dbl-homology and pleckstrin-homology (DH/PH) domain, and the vacuolar protein sorting 9 (VPS9) domain [2, 3]. In addition, eight consecutive membrane occupation and recognition nexus (MORN) motifs are noted in the region between the DH/PH and VPS9 domains. We and others have reported that ALS2 acts as a GEF for Rab5 (Rab5GEF) [4-6]. Rab5 is a key regulator in endocytosis, endosome fusion, and endosome trafficking [7]. Indeed, it has been shown that ALS2 regulates endosome fusion and trafficking by activating Rab5 [4, 5]. Further, we have demonstrated that ALS2 is a novel Rac1 effector [8] rather than a Rac1GEF [5], and plays a role in macropinocytosis and the following endosome maturation [8]. Activated Rac1 interacts with ALS2 and induces the relocalization of ALS2 from cytoplasm to membranous compartments (e.g., membrane ruffle, macropinosome, and endosome) [8]. This Rac1-mediated relocalization of ALS2 is required for the ALS2-mediated Rab5 activation on the membranous compartments. As almost all mutations identified in the ALS2 gene result in the truncation of the C-terminal Rab5GEF domain of ALS2, loss of the ALS2-associated Rab5GEF activity may underlie the pathogenesis of ALS2-linked MNDs [1]. Recently, two missense mutations in the ALS2 gene were identified [9, 10]. Both mutations result in the mutant proteins (ALS2C157Y and ALS2G540E) carrying a single amino-acid substitution in the N-terminal RLD of ALS2 (ALS2-RLD) with a preservation of the C-terminal Rab5GEF domain [9, 10]. These missense mutations might lead to loss of the ALS2 function, because the pedigrees carrying these mutations are consistent with an autosomal recessive inheritance [9, 10]. However, the molecular mechanism by which the missense mutations cause loss of the ALS2 function is still unknown. To directly address this issue, we here investigated the molecular and cellular functions of these two pathogenic mutants. HeLa cells were cotransfected with pCMV10-2XHA-Rac1Q61L and either pCIneo-FLAG-ALS2WT, -ALS2C157Y, -ALS2G540E, -ALS2T510A, or -ALS2P1603A as previously described [8]. Twenty-four hours after transfection, the cells were fixed, permeabilized, and co-stained with anti-ALS2 (HPF1-680) and anti-Rac1 antibodies (Abs). Two hundred cells expressing FLAG-tagged ALS2 with HA-tagged Rac1Q61L were examined. The phenotypes of the cells were categorized into two groups; i.e., ‘cytoplasmic’ and ‘macropinosome (at least four vesicles/cell were over 2 μm in diameter)’, based on the localization of ALS2. Each value represents the mean and standard deviation of at least five independent experiments. Statistical significance was evaluated by ANOVA with Bonferroni's post hoc test. Other information concerning the materials and methods including antibodies, plasmid constructs, Western blot analysis and gel staining, cell culture and transfection, in vitro GST pull down assay, co-immunoprecipitation, in vitro GEF assay, immunocytochemistry, confocal microscopy, dextran-uptake assay, and protein-lipid overlay assay are available in the Supplementary data. ALS2 interacts with both Rac1 and Rab5 [4, 5, 8]. Thus, we first analyzed whether the pathogenic mutants (Fig. 1 A) interacted with Rac1 and Rab5. Consistent with the previous report [8], FLAG-ALS2WT was pulled down with GST-Rac1 and GST-Rab5, but not with GST, GST-Cdc42, and GST-RhoA (Fig. 1B). Notably, both mutants were also efficiently pulled down with either GST-Rac1 or GST-Rab5 but not with GST, GST-Cdc42, and GST-RhoA (Fig. 1B), as was ALS2WT. Next, we analyzed the Rab5GEF activities of the pathogenic mutants by in vitro GDP dissociation assay. Both mutants stimulated the GDP dissociation from Rab5A, although they showed slightly lower activities than ALS2WT (Fig. 2 A). Consistently, neither pathogenic missense mutations abolished the ALS2 self-association competence (Supplementary Fig. 1) [6]. These results indicate that the pathogenic ALS2 mutants retain the ability to activate Rab5 in vitro. Rac1-induced relocalization of ALS2 is necessary to proceed with the ALS2-mediated activation of Rab5 on membranous compartments. Thus, we assume that the mutants fail to be relocalized to membranous compartments irrespective of their Rac1 binding properties, which results in loss of the ALS2 function as a Rab5GEF on membranous compartments. To verify this, we examined whether the Rac1-dependent subcellular relocalization of the mutants were affected. We expressed ALS2WT or its mutant proteins with constitutively activated-Rac1 mutant (Rac1Q61L) in HeLa cells, and observed the localization of ALS2 in Rac1Q61L expressing cells. Expression of Rac1Q61L induced macropinosomes (large phase bright organelle, over 2 μm in diameter, indicated by arrowheads in Fig. 3 A) arisen from peripheral and/or dorsal membrane ruffles as previously reported [11]. Although ALS2WT was widely distributed throughout the cytoplasm (Supplementary Fig. 2), it was redistributed to membrane ruffles and dextran-positive nascent macropinosomes in a Rac1Q61L dependent manner (Fig. 3A and B). Further, a significant proportion of ALS2 was colocalized with EEA1 (endosome marker) (Fig. 3C), suggesting that ALS2-localizing macropinosomes matured into early endosomes (EEs). Interestingly, unlike ALS2WT, both ALS2C157Y and ALS2G540E failed to be localized to both Rac1Q61L-induced macropinosomes and macropinosome-derived EEs, and yet sequestered in the cytoplasm (Fig. 3A–C). Quantitative analysis revealed that ALS2WT was localized to macropinosomes in over 40% of the Rac1Q61L/ALS2WT-expreeessing cells, whereas both ALS2C157Y and ALS2G540E were almost devoid of macropinosomal localization [ALS2WT, 42.00 ± 4.76%; ALS2C175Y, 0.28 ± 0.48%; ALS2G540E, 0.50 ± 0.54%] despite that a comparable number of macropinosomes was observed (Fig. 3A and B). Conversely, both mutants were widely distributed throughout the cytoplasm without localizing to small endosomes and membrane ruffles in over 80% of the cells expressing Rac1Q61L (Supplementary Fig. 3). By contrast, the ALS2 mutants carrying a single amino-acid substitution on a phosphorylation site [12] of the ALS2-RLD (ALS2T510A) or a critical site for Rab5GEF activity in the VPS9 domain (ALS2P1603A) [4] were localized to Rac1Q61L-induced macropinosomes and EEs (Fig. 3A–C), as was ALS2WT, although expression of ALS2P1603A slightly inhibited EEA1 recruitment [8]. These results suggest that the seemingly non-pathogenic amino-acid substitution in the RLD may not disturb the Rac1Q61L-dependent macropinosomal localization of ALS2. Collectively, our data demonstrate that the disease causing missense mutations in the ALS2-RLD specifically result in the mislocalization of ALS2 in cell. To clarify the mechanisms of mislocalization of the ALS2 mutants, we tested whether the ALS2 mutants could associate with phosphoinositide phosphates (PIPs). Since ALS2 has putative lipid association domains and motifs; i.e., RLD [8], DH/PH [2, 3] and MORN [2, 3] (Fig. 1A), we assume that macropinosomal localization of ALS2 is regulated by the lipid association. We performed PIP-Strip overlay assay using purified FLAG-ALS2WT and the mutants. FLAG-ALS2WT showed strong affinities to several PIPs including phosphatidylinositol-3-phosphate [PI(3)P], phosphatidylinositol-4-phosphate [PI(4)P], phosphatidylinositol-5-phosphate [PI(5)P], and phosphatidylinositol-3,5-phosphate [PI(3,5)P2] (Fig. 4 ). Notably, both FLAG-ALS2C157Y and FLAG-ALS2G540E exhibited lower affinities to PI(3)P and PI(4)P compared to FLAG-ALS2WT (Fig. 4). These results, combined with a recent finding indicating that PI(3)P is an important signaling lipid involved in macropinosomal maturation and traffficking after macropinocytosis [13], suggest that binding to specific PIPs, particularly PI(3)P, might regulate macropinosomal localization of ALS2, and that loss of such association in pathogenic ALS2 mutants may explain the preferential cytosolic distribution of these proteins. We next clarified the effect of ALS2 mislocalization in cells. To investigate whether the mislocalization of ALS2 causes a change in the activity of ALS2-mediated early endosomes (EEs) fusion [4], we expressed ALS2WT or the mutants in HeLa cells, and analyzed the morphology of EEA1-positive EEs. Consistent with the previous reports [4], expression of ALS2WT induced EEs enlargement (Fig. 5A, white arrowheads). Remarkably, both ALS2C157Y and ALS2G540E expression failed to induce EEs enlargement, as did the Rab5-GEF activity deficient mutant (ALS2P1603A) (Fig. 5A). These results indicate that mislocalization of ALS2 leads to loss of ALS2 competence to act as an activator for Rab5 in cells. Finally, we examined whether expression of ALS2 or its mutants regulate the autophagy-endolysosomal system. Recently, we have reported that ALS2 deficiency causes the delay of autophago-endolysosome mediated protein degradation [14]. Since a subpopulation of autophagosomes are fused with EEs and form hybrid-organelle called amphisomes before they fuse with lysosomes [15], we hypothesize that ALS2 plays a role in the autophagy-endolysosomal pathway by regulating the formation of amphisomes through its Rab5 GEF activity. To verify this, we examined the colocalization of EEA1 (endosome marker) and microtubule-associated protein 1A/1B-light chain 3 (LC3) (autophagosome marker) in vesicular compartments. Expression of ALS2WT induced enlarged EEA1-positive vesicles that were colocalized with LC3 (Fig. 5 B, white arrowheads), indicating an enhanced formation of amphisomes. By contrast, expression of ALS2P1603A did not induce enlarged amphisomes (Fig. 5B). Notably, expression of neither ALS2C157Y nor ALS2G540E enhanced the formation and/or enlargement of amphisomes (Fig. 5B). These results indicate that like Rab5-GEF activity-defective ALS2, relocalization-defective ALS2 mutants lead to the decreased formation of amphisomes in cells. In this study, we demonstrated that the pathogenic missense mutations in the ALS2-RLD impaired Rac1-induced relocalization of ALS2, probably by decreasing the affinities to specific lipid molecules in membranes. Further, these mutants lost the competence to enhance the formation of amphisomes in cells. Based on these findings, mislocalization of ALS2 leads to loss of the ALS2 function as a Rab5 activator on macropinosomes/endosomes, resulting disturbance of the autophagosome and endosome maturation, might account for the ALS2-linked MNDs. Previously, we have shown that loss of the ALS2-associated Rab5GEF activity underlies the pathogenesis of ALS2-linked MNDs [1, 4]. Further, the ALS2-associated Rab5GEF activity is involved in macropinocytosis and the following early endocytic pathway [4, 5, 8]. Although the physiological role of macropinocytosis is still unclear, macropinocytosis may link cargo molecules for the incorporation into macropinosomes to the degradative pathway [16]. Since ALS2 activates Rab5 on macropinosomes and endosomes, it may enhance endosome trafficking leading to the degradation of certain cargo molecules that are endocytosed via macropinocytosis. Indeed, we have shown that mouse embryonic fibroblasts derived from Als2-knockout mice displays slower degradation rate of endocytosed epidermal growth factor [17]. Further, loss of ALS2 in a mutant Cu/Zn-superoxide dismutase (SOD1H46R)-expressing ALS mouse model results in the delay of autophago-lysosome mediated protein degradation and accelerates neurodegeneration [14]. Moreover, we here demonstrated that ALS2-mediated Rab5 activation was involved in the amphisome formation, which may link the endocytosed molecules to autophago-lysosome mediated protein degradation. Collectively, ALS2 is implicated in macropinocytosis, endosome trafficking, and the following degradative pathways. Thus, loss of ALS2 might promote the accumulation of misfolded proteins in particular groups of motor neurons and neurodegeneration. Future studies on the molecular basis of ALS2 may uncover the roles for macropinocytosis in selective cargo sorting and degradation, which bring new insights into the molecular pathogenesis for ALS2-linked MNDs and other neurodegenerative diseases. This work was funded by Grant-in-Aid for (JSPS) (to A.O., R.K., and S.H.), and partly by the Japan Science and Technology Agency (to J.E.I.) and the Ministry of Health, Labor and Welfare (to J.E.I.). A.O. is supported by Japan Amyotrophic Lateral Sclerosis Association. S.H. receives support for a Research Aid from the Tokai University General Research Organization. A.O., R.K., K.S.-U., and S.H. are also supported by a Tokai University School of Medicine Research Aid. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2011.01.045. Supplementary Fig. 1. Oligomerization of the ALS2 mutants. Cell lysates from COS-7 cells expressing FLAG- and HA-ALS2 were immunoprecipitated by using anti-FLAG M2 Ab conjugated beads. Cell lysates and immunoprecipitated samples were immunoblotted and probed with either anti-FLAG M2 or anti-HA Ab. Supplementary Fig. 2. Subcellular localization of the ALS2 mutants. Subcellular localization of ALS2WT or its mutants expressed in HeLa cells was analyzed by immunocytochemistry using anti-ALS2 Ab (HPF1-680). Under standard cultural conditions, all of the ALS2 proteins used in these experiments (shown in Fig. 1A) were distributed throughout the cytoplasm and small portions of them were present to membrane ruffles. Supplementary Fig. 3. Quantification of the cells showing cytoplasmic localization of ALS2 in Rac1Q61L expressing HeLa cells. Two hundred cells expressing Rac1Q61L with ALS2WT, ALS2C157Y, ALS2G540E, or ALS2T510A were analyzed, and the cells showing “cytoplasmic ALS2 localization” were counted (see Supplementary materials and methods). The values are shown as mean ± S.D. (each genotype; n = 6–7). Statistical significance is evaluated by ANOVA with Bonferroni’s post hoc test. There are significant differences in the percentage of the cells showing cytoplasmic localization of ALS2 between indicated genotypes (∗∗∗P < 0.001). Supplementary data. Materials and methods. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.