Fig 2: PTENe, but not other known PTEN isoforms, dephosphorylates ACTR2 and VASP by its protein phosphatase activity PTENe, but not other known PTEN isoforms, interacts with filopodia formation-related proteins. HEK293 cells were transfected with C-terminal HA and S-tagged PTEN, PTENa, PTENß, PTENe, or mock plasmid separately, followed by immunoprecipitation with S-protein beads and Western blotting with antibodies against CDC42, VASP, ACTR2, or FSCN1, and HA, respectively.Images of immunofluorescence staining for PTENe and endogenous interacting proteins (VASP, FSCN1, CDC42, and ACTR2). C-terminal GFP-tagged PTENe was introduced into Hela PTEN -/- cells. Then, these transfected cells were seeded on coverslips and sequentially stained with monoclonal antibodies (Santa Cruz, sc-8401(CDC42), sc-21743(FSCN1), sc-166103(ACTR2), and Cell Signaling Technology, 3132(VASP)) and DAPI separately. Then, all the coverslips were imaged by confocal microscopy. The scale bars represent 5 µm.Overexpression of PTENe, but not other PTEN variants, reduces the endogenous phosphorylation of ACTR2 (lower panel) and VASP (upper panel). HA and S-tagged canonical PTEN, PTENa, PTENß, PTENe, and PTENe Y210L were introduced into Hela PTEN -/- cells, followed by immunoblotting with antibodies against p-ACTR2 (Thr237 and Thr238), p-VASP(Ser157), p-VASP (Ser239), p-VASP (Thr278), VASP, ACTR2, GAPDH, and HA tag.Purified ACTR2 or VASP was phosphorylated in vitro by NIK or PKA and subsequently used in a phosphatase assay with wild-type His-PTENe, protein phosphatase-deficient His-PTENe Y210L or without purified PTENe protein as a control, followed by immunoblotting with antibodies against p-ACTR2 (Thr237 and Thr238, lower panel) and p-VASP (Ser157, upper panel). Purified proteins (PTENe, PTENe Y210L, PKA, NIK, ACTR2, and VASP) were detected by anti-His antibody (ZSGB-BIO, TA-02). Source data are available online for this figure.

Fig 3: PTENe interacts with and dephosphorylates filopodia formation-related proteins A In vivo S-tag-PTENe pull-down analysis. Whole-cell extracts from HEK293 cells transfected with S-tag-PTENe were immunoprecipitated with S-protein beads followed by mass spectrometric peptide sequencing. S-tagged-PTEN, S-tagged-PTENa, S-tagged-PTENß, or mock construct were introduced into HEK293 cells separately, and the transfected cells were treated in the same way as described above. Cell adhesion associated protein was found in the PTENe pull-down list (left panel). Molecular function analysis of PTENe interacting proteins compared with canonical PTEN, PTENa, PTENß, and MOCK groups (right panel, statistical significance was determined by Fisher’s exact test.). Mass spectrometry analysis data (as indicated in Figs 3C and E, and EV1A, and 6A) are available via ProteomeXchange with identifier PXD022245.BExogenous immunoprecipitation of PTENe with FLAG-tagged CDC42, ACTR2, VASP, and FSCN1. The plasmid expressing PTEN, PTENe, or mock carrying C-terminal S-tag and HA tag was co-transfected with FLAG-tagged CDC42, ACTR2, VASP, or FSCN1 plasmid in HEK293 cells separately. Cell lysates were incubated with S-protein agarose followed by Western blotting with antibodies against the FLAG tag and HA tag. G1–4 stand for immunoprecipitation group 1–4.CImmunoprecipitation of PTENe with endogenous CDC42, ACTR2, VASP, and FSCN1. The plasmid expressing PTEN, PTENe, or mock plasmid carrying C-terminal S-tag and HA tag was transfected in HEK293 cells separately. Cell lysates were incubated with S-protein agarose followed by Western blotting with antibodies against CDC42, VASP, ACTR2, FSCN1, and HA tag.DImages of immunofluorescence staining for PTENe and its interacting proteins (VASP and FSCN1). Hela PTEN -/- cells were co-transfected by a plasmid expressing C-terminal GFP-tagged PTENe with FLAG-tagged targets (FSCN1, VASP) separately, followed by staining with an anti-FLAG antibody and DAPI, and were imaged by confocal microscopy. The scale bars represent 5 µm.EThe cell plasma membrane localization of PTENe is required for its interaction with downstream targets. FLAG-tagged CDC42, VASP, ACTR2, or FSCN1 plasmid was co-transfected with a plasmid expressing PTENe with or without 69–81aa deletion carrying C-terminal S-tag and HA tag in HEK293 cells. Cell lysates were incubated with S-protein agarose before Western blotting with antibodies against the FLAG tag and HA tag.FPTENe protein phosphatase activity is required for its interaction with CDC42, FSCN1, VASP, and ACTR2. Wild-type PTENe and PTENe with protein phosphatase activity abolished mutation (Y210L, analogous to PTEN (Y138L)) were separately co-transfected with FLAG-tagged targets (CDC42, FSCN1, ACTR2 or VASP) in HEK293 cells. Cell lysates were incubated with S-protein agarose followed by Western blotting with antibodies against the FLAG tag and HA tag.GThe level of phospho-VASP and phospho-ACTR2 in PTENe overexpressed Hela cells. Wild-type PTENe and PTENe with protein phosphatase activity abolished mutation (Y210L, analogous to PTEN (Y138L)) carrying C-terminal S-tag and HA tag were introduced into Hela PTEN -/- cells, followed by immunoblotting with antibodies against p-ACTR2 (Thr237 and Thr238), p-VASP(Ser157), p-VASP (Ser239), p-VASP (Thr278), VASP, ACTR2, GAPDH, and HA tag.H, IPurified ACTR2 or VASP was phosphorylated in vitro by NIK or PKA and subsequently used in a phosphatase assay with His-PTENe or without purified PTENe protein as a control, followed by immunoblotting with antibodies against p-ACTR2 (Thr237 and Thr238) and p-VASP (Ser157). Purified proteins (PTENe, PKA, NIK, ACTR2, and VASP) were detected by anti-His antibody (ZSGB-BIO, TA-02). Source data are available online for this figure.

Fig 4: CD2AP/CIN85 promotes actin-branched network formation around the base of primary cilia.A and B, RPE1 cells transfected with EHBP1L1 #1 siRNA or cotransfected with CD2AP and CIN85 siRNAs were stained with antibodies against Arp2 (A) and PCNT (A and B) or with Alexa-568 phalloidin (B). Dotted circles indicate a 1 µm radius around the PCNT-labeled basal body. Magnified images of the circled areas are shown in the insets. Scale bar represent (main images) 2 µm; (insets) 0.5 µm. D and E, fluorescence intensities of Arp2 and Alexa-568 phalloidin within a 1 µm radius around the basal body were measured and plotted against siRNA-treated cells (control, CD2AP/CIN85, or EHBP1L1 #1 and #2). n = 50 for each experiment. Error bars represent S.D. p values were calculated using Student’s t test. Statistical significance was set at ***p < 0.001. C, cell lysates treated with siRNAs (control, CD2AP/CIN85, or EHBP1L1 #1 and #2) were immunoblotted with EHBP1L1, CD2AP, CIN85, and Arp2 antibodies. GAPDH was used as the loading control. PCNT, pericentrin; RPE, retinal pigment epithelium.

Fig 5: ARP2 levels are enriched in the plaque microenvironment of AsymAD cases.A. Staining against ARP2 (green), IBA1 (red) and DAPI, to identify amyloid-plaques and cell nuclei, respectively in control, AsymAD, and AD cases. Colocalization between ARP2 and IBA1 pixels are shown in white B-D. Quantification of overall mean ARP2 intensity within 50 µm of the core plaque (B) Mean ARP2 intensity per IBA1 cell (C) and Pearson correlation of ARP2 and IBA intensity (D) E. ARP2 and IBA1 immunostaining in areas free of amyloid-plaques F. Quantification of mean ARP2 intensity per IBA1 cell. Data is shown as mean ± SEM. In A, B and C, 50–60 plaques were analyzed per condition, n=6 cases per condition. Significance was determined by Mann-Whitney test. In F, a total of 90–170 microglia cells were analyzed per condition, n=6 cases per condition. Data is shown as ±SEM, One-way ANOVA, following Tukey’s multiple comparisons test (n.s.; 0.2037 and n.s.; 0.1875).