PKCα Phosphorylates FACI to Switch Its Function in Clathrin-mediated Endocytosis to a Presumed Role in Macropinocytosis in Intestinal and Hepatic Cells

Abstract

Background & aims: We previously identified a fasting- and CREB-H-induced (FACI) protein and defined its adaptor function in clathrin-mediated endocytosis. Both CREB-H and FACI are specifically expressed in the liver and intestine. Here, we investigated the role of FACI in macropinocytosis and its activation by protein kinase Cα.

Methods: We employed a combination of biochemical, proteomic, cell biological, and microbiological assays to investigate the function of FACI in cultured cells and FACI^-/- mice.

Results: Phosphorylation of FACI at S9 and S37 by protein kinase Cα induces its detachment from clathrin-coated pits and relocation to the plasma membrane. FACI promotes phorbol ester-induced macropinocytosis in intestinal and hepatic cells. Interactome analysis reveals that FACI interacts with several actin remodeling proteins. FACI interacts with 14-3-3ζ to release SSH1 phosphatase from sequestration. Free SSH1 activates cofilin-1, which in turn enhances actin remodeling and macropinocytosis. Intestinal pathogens such as Salmonella typhimurium exploit FACI to facilitate their entry into host cells through macropinocytosis.

Conclusion: FACI modulates clathrin-mediated endocytosis and macropinocytosis in intestinal and hepatic cells. Protein kinase Cα phosphorylates FACI to switch its function in endocytosis to a presumed role in macropinocytosis in these cells. FACI facilitates enteric pathogen invasion by enhancing macropinocytosis in the intestine.

Keywords: Clathrin-mediated endocytosis; FACI; Macropinocytosis; PKCα; Phosphorylation; Salmonella.

Figure 1

PKCα phosphorylates FACI at S9 and S37. (A) Predicted 3D structure of FACI by AlphaFold2. FACI protein is composed of 2 parts: N-terminal intrinsic disorder region (IDR) and C-terminal helical region. The C-terminal helical region consists of motifs D (DxxxLI) and E (PI-binding motif). The N-terminal IDR contains motifs A (RxxpS), B (YxxL) and C (RxxpS). Motifs A and C are 2 predicted phosphorylatable motifs with unknown function. Model confidence: very high (blue, pLDDT >90); confident (cyan, 90 >pLDDT >70); low (yellow, 70 >pLDDT >50); very low (red, pLDDT <50). Regions below 50 pLDDT may be unstructured in isolation. pLDDT: predicted local distance difference test (pLDDT) score, a per-residue confidence score. (B) Multiple alignments of the N-terminal region of FACI homologs among 8 species. Conserved motifs A (RxxpS), B (YxxL) and C (RxxpS) were labeled. FACI homologs were from human (Homo sapiens, NP_001129957.1), mouse (Mus musculus, NP_081513.1), koala (Phascolarctos cinereus, XP_020824020.1), gray short-tailed opossum (Monodelphis domestica, XP_007502710.1), western painted turtle (Chrysemys_picta_bellii, XP_008177865.1), Reeves’ turtle (Mauremys reevesii, XP_039403575.1), graceful crag lizard (Hemicordylus capensis, XP_053133582.1), and central bearded dragon (Pogona_vitticeps, XP_020669464.1). (C) MS/MS spectrum of phosphopeptides SQpSFREPRPSYGR (motif A, left) and ALpSLRQGQEK (motif C, right). Phosphopeptides were obtained by digesting the mouse FACI (mFACI) protein. y-ions: red. b-ions: blue. ∗Denotes neutral loss of H3PO4 (−98 Da). (D) Coomassie blue staining of purified FACI-N and FACI-N-3A proteins. GST-FACI-N and GST-FACI-N-3A proteins were cut with PreScission protease to remove the GST-tag. Protein sequences of FACI-N and FACI-N-3A are in (E). (E) Protein sequences of FACI and its mutants. Motifs A to E were illustrated. (F) Sequences of synthetic peptides representing motif A (Pep-7S9S), non-phosphorylatable motif A (Pep-7A9A), motif C (Pep-37S) and non-phosphorylatable motif C (Pep-37A). (G–I) ADP-Glo kinase assay. Kinase activity of indicated kinases on FACI-N and FACI-N-3A (G), Pep-7S9S and Pep-7A9A (H), and Pep-37S and Pep-37A (I). Kinase activity was normalized with the mock group. (J) Schematic evolutionary tree of 11 PKC isozymes of humans. (K–M) ADP-Glo kinase assay. Kinase activity of indicated PKCs on FACI-N and FACI-N-3A (K), Pep-7S9S and Pep-7A9A (L), and Pep-37S and Pep-37A (M). Kinase activity was normalized to the mock group. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. (N) In-gel kinase assay. Recombinant PKCα was subjected to kinase reactions with the substrate FACI-N or FACI-N-3A. The reaction mixes were analyzed by SDS-PAGE and further probed with antibodies against phosphoserine in PKC substrates (left) or visualized by silver staining (right). Arrowhead demonstrates the recombinant FACI-N or FACI-N-3A. (O) ADP-Glo kinase assay. Kinase activity of indicated PKCs on FACI-C peptide. Kinase activity of PKCα on Pep-37S was used as positive control. Kinase activity was normalized to the mock group. All results are representative of 3 independent experiments. The mean values of 3 biological replicates (n = 3) were represented by the bars and their respective SDs were depicted as the error bars. (P) In vivo pull-down assay. AML12-CTR and AML12-Flag-FACI cells were treated with the indicated drugs, and the cell lysates were immunoprecipitated with anti-Flag antibody. The immunoprecipitates were analyzed by SDS-PAGE and probed with antibodies against phosphoserine in PKC substrates.

Figure 2

Sequence alignment of motif A in FACI orthologs from 223 species. The conserved S9 residue was indicated with an asterisk. Up to May 2024, FACI orthologs from 228 species were deposited in the NCBI gene database. 223 of them were used for sequence alignment. A consensus sequence logo of FACI motif A generated from 223 FACI homologs was also provided.

Figure 3

Sequence alignment of motif C in FACI orthologs from 223 species. The conserved S37 residue was indicated with an asterisk. Up to May 2024, FACI orthologs from 228 species were deposited in the NCBI gene database. 223 of them were used for sequence alignment. A consensus sequence logo of FACI motif C generated from 223 FACI homologs was also provided.

Figure 4

Putative phosphorylation sites of FACI. The consensus sequence recognized by 8 kinases. S/T phosphorylation sites and flanking ±6 amino acids were shown. Sequences of kinase substrates were acquired from PhosphoSitePlus PTM Database (CST). Consensus sequence logos of FACI motifs A and C generated from 223 FACI homologs were also indicated.

Figure 5

Functional impact of PMA-induced PKC activation on FACI. (A, B) Spinning disc confocal images. AML12 cells expressing mRuby2-FACI were transfected with plasmids encoding mEmerald-CLC (A) or mEGFP-PKCα (B). Cells were imaged before and after PMA treatment. Representative images were shown. The basal PM layers (A) and the cytoplasmic layers (B) were indicated. CLC, clathrin light chain. Scale bar, 10 μm. (C, D) Spinning disc confocal images. AML12 cells expressing mRuby2-FACI were transfected with plasmids encoding mEmerald-CLC (C) or mEGFP-PKCα (D). AML12 cells were pretreated with GO6983 for 30 minutes, followed by GO6983 + PMA treatment for another 30 minutes. Images before and after drug treatment were acquired. Representative images were shown. The basal PM layers (C) and the cytoplasmic layers (D) were indicated. Scale bar, 10 μm. (E) Quantitative analysis of distribution patterns of mRuby2-FACI in the PM. AML12 cells expressing mRuby2-FACI and mEmerald-CLC were imaged before and after drug treatment. The distribution patterns of mRuby2-FACI in the PM are categorized into 2 types: cells with mRuby2-FACI localized to CCPs, and cells with mRuby2-FACI evenly distributed in the PM. The percentage count of both cell types was quantified. For each group, more than 100 cells were quantified. (F) Quantification of mRuby2-FACI in the PM. AML12 cells expressing mRuby2-FACI were transfected with mEmerald-Farnesyl (PM marker). Cells underwent the specified treatments and were subsequently imaged. The PM fluorescence of mRuby2-FACI was quantified by calculating the fluorescent ratio between the PM and the entire cell. One-way ANOVA with the Tukey post hoc tests was used for statistical analysis. ∗∗∗P < .001; ns: not significant. (G) Immunoprecipitation. AML12 cells with stable expression of FACI-Flag (AML12-FACI) and control AML12 stable cells (AML12-CTR) were used for immunoprecipitation. Cells were mock treated or treated with PMA for 30 min. Cell lysates were collected, immunoprecipitated with anti-Flag antibodies and analyzed as specified. (H, I) Subcellular fractionation. The PM fractions of AML12-FACI cells with the indicated treatments were isolated. Total cell lysates and PM fractions were analyzed by SDS-PAGE and probed with anti-Flag and antibodies against the indicated organelle marker (H). E-cadherin, GAPDH, GM-130, Calnexin, and Lamin A/C serve as markers for PM, cytoplasm, Golgi apparatus, endoplasmic reticulum (ER), and nucleus, respectively. Statistical analysis of subcellular fractionation was performed (I). The relative amounts of FACI protein in total cell lysates and PM fractions were quantified. One-way ANOVA with the Tukey post hoc tests was used for statistical analysis. ∗∗∗P < .001; ns: not significant. All results are representative of 3 independent experiments.

Figure 6

Requirement of S7 and S37 for PMA-induced FACI dissociation from CCPs. (A) Protein sequences of FACI and FACI-3A mutants. Motifs A to E were illustrated. (B) Protein steady-states of FACI-Flag and FACI-3A-Flag. AML12-FACI and AML12-FACI-3A cells were treated with vehicle or PMA for 30 minutes. The cell lysates were analyzed by SDS-PAGE and probed by the indicated antibodies. (C, D) Spinning disc confocal images. AML12 cells expressing either mRuby2-FACI (C) or mRuby2-FACI-3A (D) were transfected with mEmerald-CLC and imaged before and after PMA treatment. Over 100 cells were imaged and analyzed. Representative images were shown. The basal PM layers were imaged. Scale bar, 10 μm. (E) Quantitative analysis of mRuby2-FACI-3A and mRuby2-FACI distribution in the PM. AML12 cells expressing mRuby2-FACI-3A and mEmerald-CLC were imaged before and after drug treatment. The number of cells with mRuby2-FACI-3A localized in CCPs and cells with mRuby2-FACI-3A evenly distributed in the PM were quantified. The distribution of mRuby2-FACI in the PM was also analyzed using the same method. For each group, more than 100 cells were quantified. (F–G) Spinning disc confocal images. AML12 cells expressing either mRuby2-FACI (F) or mRuby2-FACI-3A (G) were transfected with mEGFP-PKCα and imaged before and after PMA treatment. Representative images were shown. The cytoplasmic layers were imaged. Scale bar, 10 μm. (H) Quantification of mRuby2-FACI-3A and mRuby2-FACI amounts in the PM. AML12 cells expressing mRuby2-FACI-3A were transfected with mEmerald-Farnesyl (PM marker) and imaged before and after PMA treatment. The fluorescence intensity of mRuby2-FACI-3A in the PM was quantified. The amounts of mRuby2-FACI in the PM were also analyzed using the same method. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. (I, J) Fluorescent LDL uptake assay. AML12-CTR, AML12-FACI, and AML12-FACI-3A cells were treated with mock (I) or PMA (J) for 30 minutes, followed by a pHrodo-Red-LDL uptake assay. The amounts of endocytosed pHrodo-Red-LDL were measured by flow cytometry. Data are represented by medium fluorescence intensity (MFI) ± SD. ∗∗∗P < .001; ns: not significant. A 2-tailed unpaired Student’s t-test was used. All results are representative of 3 independent experiments. (K, L) AML12-FACI and AML12-FACI-3A cells were treated with mock (K) or PMA (L) for 30 minutes, followed by a pHrodo-Red-LDL uptake assay. The amounts of endocytosed pHrodo-Red-LDL were measured by flow cytometry. The measured results were normalized based on their protein expression levels. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test.

Figure 7

Functional impact of PMA-induced PKC activation on FACI in Caco2 cells. (A) Spinning disc confocal images. Caco2 cells expressing mRuby2-FACI (left) or mRuby2-FACI-3A (right) were transfected with plasmids encoding mEmerald-CLC. Cells were imaged before and after PMA treatment. Over 50 cells were imaged and analyzed. Representative images were shown. The basal PM layers were indicated. Scale bar, 10 μm. (B) Spinning disc confocal images. Caco2 cells stably expressing mRuby2-FACI (left) or mRuby2-FACI-3A (right) were imaged before and after PMA treatment. Representative images were shown. The cytoplasmic layers were indicated. Scale bar, 10 μm. (C) Quantitative analysis of the distribution of mRuby2-FACI and mRuby2-FACI-3A in the PM. Caco2 cells expressing either mRuby2-FACI or mRuby2-FACI-3A were imaged before and after drug treatment. The number of cells with mRuby2-FACI localized in CCPs and cells with mRuby2-FACI evenly distributed in the PM were quantified. The distribution of mRuby2-FACI-3A in the PM was analyzed using the same method. For each group, more than 75 cells were quantified. (D) Quantification of protein amounts in the PM. Caco2 cells expressing either mRuby2-FACI or mRuby2-FACI-3A were transfected with mEmerald-Farnesyl (PM marker) and imaged before and after PMA treatment. The fluorescence intensity of mRuby2-FACI or mRuby2-FACI-3A in the PM was quantified. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. All results are representative of 3 independent experiments.

Figure 8

Requirement of motif E for FACI interaction with PI(4,5)P₂ in the PM and PMA-induced FACI relocation to the PM. (A) Amino acid sequences of FACI, FACI-motif-ABCD, and FACI-motif-E proteins. Motif A to E regions were indicated above the sequences. (B) Spinning disc confocal images of AML12 cells expressing mEmerald-FACI-motifs-ABCD and mRuby2-PKCα before and after PMA treatment. Over 50 cells were imaged, and representative images were presented. Scale bar, 10 μm. (C) Left. Spinning disc confocal images of AML12 cells expressing mEmerald-FACI-motif-E and mRuby2-PKCα before and after PMA treatment. Representative images were presented. Scale bar, 10 μm. Right. Quantification of protein amounts in the PM. AML12 cells expressing mRuby2-FACI-motif-E were transfected with mEmerald-Farnesyl (PM marker) and imaged before and after PMA treatment. The PM fluorescence of mRuby2-FACI-motif-E was quantified as the fluorescent ratio of the PM against the whole cell. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. (D) 3D structure of human FACI (hFACI) motif E was predicted by AlphaFold2 and illustrated by ribbons (upper left) and hydrophobicity surface model (upper right). Protein sequence of hFACI motif E was shown with its amphipathic α-helix and hydrophobic region highlighted (lower). Yellow: hydrophobic amino acids. Blue: positively charged amino acids. Red: negatively charged amino acids. Green (in ribbon model) and gray (in hydrophobicity surface model): other amino acids. (E) The amphipathic α-helix (83–100 amino acids of hFACI) predicted by the Heliquest was visualized. (F) Theoretical net charges of hFACI motif E (83–115 amino acids) at different pH values. Net charges at physiological condition (pH = 7.4) were shown. (G) Fat blot assays were performed with indicated PI strips and a synthetic FACI peptide (75–115 amino acids of hFACI). Results of 2 independent experiments were displayed. PM-distributed PIs were highlighted in red. PtdIns: phosphatidylinositol. (H) Semi-quantitative analysis of fat blot assay in (G). Results were normalized to PI(3,5)P₂. (I) AML12 cells expressing mEmerald-FACI-motif-E and the indicated plasmids were treated with PMA. For each group, over 50 cells were imaged and analyzed. Representative images were shown. Scale bar, 10 μm. (J) Schematic diagram depicting the rapamycin-inducible heterodimerization system, which was used for acutely depriving PI(4,5)P₂ from the PM. After rapamycin treatment, the cytoplasmic FKBP-fused 5-phosphatase (5-ptase) binds to the PM-anchored FRB. FKBP-5-ptase at the PM causes a rapid degradation of PI(4,5)P₂. (K, L) AML12 cells expressing mEmerald-FACI and PM-FRB-T2A-FKBP-5-ptase were treated with rapamycin to acutely deprive PI(4,5)P₂ from PM. Representative images (K) and pixel fluorescent intensity (L) showing distinct mEmerald-FACI distribution pre- and post-rapamycin treatment. ∗: nucleus. Scale bar, 10 μm. All results are representative of 3 independent experiments.

Figure 9

Further bioinformatic analysis of motif E. (A) Sequence alignment of motif E between hFACI and Zebrafish Sb:cb1058 (zSb:cb1058, Uniprot No: A0A0R4IHZ3). Sb:cb1058 is the ancestral gene of FACI that exists in fishes. The amphipathic α-helix (predicted by HeliQuest) and hydrophobic regions were indicated. (B) 3D structure of zSb:cb1058 motif E was predicted by AlphaFold2 and illustrated by ribbons (upper left) and hydrophobicity surface model (upper right). Sequence information of zSb:cb1058 motif E was also provided (lower). Yellow: hydrophobic amino acids. Blue: positively charged amino acids. Red: negatively charged amino acids. Green (ribbons model) and gray (hydrophobicity surface model): other amino acids. (C) The amphipathic α-helix within zSb:cb1058 motif E was predicted with Heliquest. (D) Theoretical net charges of zSb:cb1058 motif E (190-222 amino acids) at different pH values. Net charges at physiological conditions (pH = 7.4) were shown.

Figure 10

FACI promotes PMA-induced intestinal and hepatic macropinocytosis. (A) RT-qPCR analysis of FACI mRNA expression in WT and FACI-overexpressing Caco2 and AML12 cell lines. ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. (B) Immunofluorescence staining. Caco2-FACI and AML12-FACI cells with unattached edges (ie, cell edges not in contact with other cells) were imaged after PMA treatment. Flag-FACI and F-actin were stained by anti-Flag antibodies and phalloidin-iFluor 647, respectively. After PMA treatment, FACI and F-actin were present in membrane ruffle regions. X-Z images for the chosen y-axis (dash line) were displayed. Scale bar, 10 μm. (C) An ImageJ macro was developed for the batch quantification of macropinocytosis. Caco2-FACI cells were incubated with TD-70 and PMA. Cells were fixed, stained with phalloidin-iFluor 647, and imaged. The acquired 647 nm channel (phalloidin) images were used to identify cell area and the 561 nm channel (TD-70) images were used to determine macropinosome area. As indicated, the macro successfully assisted in the identification of endocytosed TD-70. Scale bar, 10 μm. (D) In vivo macropinocytosis assay. Caco2 and AML12 cells were incubated with TD-70, with or without PMA. Cells were fixed at the indicated time points and imaged using a spinning disc confocal microscope. For each time point, more than 30 microscopic fields were imaged. The macropinocytosis index was calculated using the methods described in Figure 10C. RAW264.7 macrophages were also included in the experiments as a reference. Statistical significance was analyzed between the mock-treated and PMA-treated groups. ∗∗∗P < .001. (E, F) Time-series images. Caco2-mRuby2-FACI (E) and AML12-mRuby2-FACI (F) were incubated with FITC-dextran 70 kD (FD-70) and PMA for 15 minutes. Macropinocytosis events were dynamically monitored and represented. Scale bar, 10 μm. Refer to Supplementary Videos 4–5 for more details. BF, bright field. (G, H) In vitro macropinocytosis assay. Caco2-CTR, Caco2-FACI, Caco2-FACI-3A, and Caco2-motif-ABCD cells were treated with TD-70 and indicated drugs. Details of the drug treatment can be found in the Methods section. Representative images (G) and quantitative analysis (H) of TD-70 uptake were presented. Scale bar, 10 μm. One-way ANOVA with the Tukey post hoc tests was used for statistical analysis. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001; ns: not significant. (I) In vitro macropinocytosis assay. Primary enterocytes from WT and FACI^-/- mice (7 to 10 days old; n = 4–5 for each group) were isolated and treated with TD-70 and indicated drugs. Ingested TD-70 was measured by flow cytometry. Data are represented by medium fluorescence intensity (MFI) ± SD. ∗∗P < .01; ∗∗∗P < .001 by 1-way ANOVA with the Tukey post hoc tests.

Figure 11

Further exploration of FACI's function in macropinocytosis. (A, B) In vivo macropinocytosis assay. WT and FACI^-/- mice (2 months old; n = 4 for each group) were anesthetized. Ligated ileal loops were generated and filled with 1.5 mg/mL TD-70 and 200 nM PMA for 15 minutes. Ileal loop tissues were sectioned, stained with phalloidin-iFluor 647 and DAPI, and imaged by confocal microscopy. Panel (A) presents representative images of the ileum sections from both WT and FACI^-/- mice. Scale bar, 50 μm. Panel (B) shows pixel fluorescent intensities corresponding to the solid straight lines marked by yellow rectangles in Panel (A). (C) In vivo macropinocytosis assay. WT mice (7 to 10 days old) were intraperitoneally injected with PBS (for mock group) or EIPA (10 mg/kg) for 3 days and then subjected to in vivo macropinocytosis assay. Flow cytometry was used to quantify the TD-70 uptake levels. Data are represented by medium fluorescence intensity (MFI). ∗∗∗P < .001 by 2-tailed unpaired Student’s t-test. n = 4–5 for each group. (D–E) In vivo macropinocytosis assay. WT and FACI^-/- mice (7 to 10 days old) were subjected to in vivo macropinocytosis assay as described in Figure 11A. (D) Ileal loop tissues were sectioned, stained with phalloidin-iFluor 647 and DAPI, and imaged. Scale bar, 50 μm. For each group, sections from 4 mice were imaged and analyzed. Representative images were shown. Experiments were conducted 2 times. (E) Ileal loop tissues were digested, and intestine epithelial cells were isolated. TD-70 engulfment of intestine epithelial cells was measured by flow cytometry. Data were represented by medium fluorescence intensity (MFI) ± SD. ∗∗P < .01 by 2-tailed unpaired Student’s t-test. Results are representative of 3 independent experiments. (F) In vitro macropinocytosis assay in AML12 cells. AML12-CTR, and AML12-FACI cells were treated with TD-70 and the indicated drugs. The ingested TD-70 was measured by flow cytometry. The experiments were performed 3 times. Data are represented by medium fluorescence intensity (MFI). ∗∗∗P < .001 by 1-way ANOVA with the Tukey post hoc tests. (G) Transcriptomic data showing FACI expression levels in human immune cells. FACI was not expressed in monocytes and dendritic cells. Data were retrieved from Human Protein Atlas. Upper panel: RNA Monaco immune cell gene data; lower panel: RNA HPA immune cell gene data. (H) RT-qPCR analysis. FACI mRNA expression was compared in the intestinal epithelium, liver, BMDMs, and BMDCs of mice. The mRNA levels of FACI were normalized to β-actin as an internal control.

Figure 12

FACI facilitates Salmonella invasion by enhancing intestinal macropinocytosis. (A) Spinning disc confocal images. Caco2-mRuby2-FACI cells were infected with S. typhimurium SL7207-GFP. Over 50 cells were imaged and analyzed. Representative images were shown. mRuby2-FACI protein was enriched on Salmonella-induced apical ruffles (upper) and peripheral ruffles (lower). X-Z images for the chosen y-axis (dash line) were displayed. Scale bar, 10 μm. (B) Caco2-mRuby2-FACI cells infected with SL7207-GFP were imaged. Representative images were shown. Salmonella-containing macropinosomes were enlarged. Scale bar, 10 μm. (C) In vitro Salmonella invasion assay. Caco2-CTR and Caco2-FACI cells were treated with 200 nM PMA and infected with SL7207-GFP. Invaded SL7207-GFP were quantified by CFU counting. ∗P < .05 by 2-tailed unpaired Student’s t-test. (D–G) In vivo Salmonella infection. Three-month-old mice (n = 6–8 for each group) were inoculated with S. typhimurium SL7207 by oral gavage. The mice were sacrificed 72 hours after inoculation, and samples were harvested for analysis. Representative gross images (D) and histology images (E) of ceca from uninfected WT, infected WT, and FACI^-/- mice were presented. Quantification of invaded SL7207 in ceca of WT and FACI^-/- mice was presented in (F). ∗P < .05 by 2-tailed Mann-Whitney U test. The mRNA expression of the indicated inflammatory markers in ceca of WT and FACI^-/- mice was presented in (G). ∗P < .05. ∗∗P < .01 by 2-tailed unpaired Student’s t-test. All results are representative of 3 independent experiments.

Figure 13

FACI might play a role in actin remodeling as revealed by interactome analysis. (A) Schematic diagram of the AP-MS workflow. AP-MS assays were conducted in Caco2 and AML12 cell lines stably expressing V5-FACI (Caco2-V5-FACI and AML12-V5-FACI). Mock-transfected Caco2 and AML12 stable cell lines (Caco2-V5-CTR and AML12-V5-CTR) were used as control. V5-FACI served as bait. (B) Immunoblotting. Constitutive expression of V5-FACI was verified in Caco2-V5-FACI and AML12-V5-FACI cells. (C, D) Immunoprecipitation. Total proteins from Caco2-V5-FACI (C), AML12-V5-FACI (D) and their control cells were collected and immunoprecipitated with anti-V5 antibodies. Immunoprecipitates were separated by SDS-PAGE and then analyzed through anti-V5 immunoblotting and silver staining. FACI was indicated by arrowheads. (E) Volcano plots illustrating 3 independent AP-MS assays in Caco2 (Caco2-IP-A, -B and -C) and one AP-MS assay in AML12 (AML12-IP-A). Y-axis: log₂ fold change (FC) of protein iBAQ intensity (FACI-IP group vs Control group). X-axis: protein iBAQ values of FACI-IP group. “Potential FACI-interactors” were highlighted in red. The criteria for selecting “Potential FACI-interactors” were described in the Methods section. iBAQ: intensity-based absolute quantification. Proteins identified in the FACI-IP group only were boxed. (F, G) STRING protein–protein interaction networks of FACI-interactors in Caco2 (F) and AML12 (G) cells. The criteria for selecting FACI-interactors were described in the Methods section. Actin skeleton-related clusters (dash line) and clathrin-related clusters (solid line) were boxed. STRING settings: physical interactions; interaction sources = “text-mining, experiments, and databases”; cutoff score ≥ 0.5; disconnected nodes hidden.

Figure 14

Further analysis of the FACI interactome. (A, B) Bubble plots depicting GO enrichment of FACI-interactors in Caco2 (A) and AML12 (B) cells. Actin skeleton-related GO terms were highlighted. Y-axis: GO terms. X-axis: fold enrichment. Bubble colors: -log₁₀ (adjust P value). Bubble sizes: gene counts. (C) A list of 19 FACI-interactors identified more than 4 times in 6 rounds of AP-MS assays. Actin-binding or actin-regulatory proteins were highlighted. Protein identification times in AML12 and Caco2 cells were listed. (D) STRING PPI networks of 19 FACI-interacting proteins in Figure 14C. STRING settings: physical interactions; interaction sources = “text-mining, experiments, and databases”; cutoff score ≥ 0.5.

Figure 15

FACI impaired the binding of SSH1 with 14-3-3ζ to activate cofilin-1. (A) Immunoprecipitation. AML12 cells were transfected with plasmids of HA-14-3-3ζ, HA-14-3-3ε, and Flag-FACI. Cells were treated with PMA for 30 minutes, and immunoprecipitation was performed as indicated. (B, C) Confocal immunofluorescence images of Caco2 (B) and AML12 (C) cells expressing HA-14-3-3ζ and mRuby2-FACI after PMA treatment. Over 20 cells were imaged and analyzed. Representative images were shown. HA-14-3-3ζ was probed with anti-HA. Scale bar, 10 μm. (D) Consensus sequence logos of the 14-3-3 binding motif. Protein sequences (n = 322) for generating the 14-3-3 binding motif were acquired from a previous study. Motifs A (R-S-x-pS-[LF]-R) and C (R-x-x-pS-[LF]-R) in FACI display similarity to the consensus binding motif of 14-3-3 (R-S-x-pS/T-x-P). (E) Immunoprecipitation. HEK293T cells were transfected with plasmids expressing HA-14-3-3ζ and Flag-FACI. After being treated with PMA or mock treated for 30 minutes, cell lysates were collected and immunoprecipitated with anti-Flag antibodies. Immunoprecipitates were analyzed by immunoblotting with the indicated antibodies. (F) Plasmids expressing HA-14-3-3ζ and either Flag-FACI or Flag-FACI-3A were transfected into HEK293T cells. Cells were treated with PMA for 30 minutes, and immunoprecipitation was performed with anti-Flag antibodies. Results are representative of 3 independent experiments. (G) Immunoprecipitation. Caco2-FACI and Caco2-CTR cells were treated with PMA or mock treatment for 30 minutes. Cell lysates were collected and immunoprecipitated with anti-Flag antibodies. Immunoprecipitates were analyzed by immunoblot with anti-Flag and anti-14-3-3ζ antibodies. Representative immunoblot results (left) and relative quantification of immunoprecipitated 14-3-3ζ proteins (right) were shown. ∗P < .05 by 2-tailed unpaired Student’s t-test. (H) Immunoprecipitation. Caco2-FACI, Caco2-FACI-3A, and Caco2-CTR cells were treated with PMA for 30 minutes. Immunoprecipitation was performed with anti-Flag and analyzed by immunoblot with anti-Flag and anti-14-3-3ζ antibodies. (I, J) Immunoprecipitation. Stable Caco2 cell lines with doxycycline-inducible V5-FACI expression were treated with mock and Dox for 24 hours. Cells were then treated with PMA for 30 minutes before harvest. Cell lysates were immunoprecipitated with anti-SHH1 antibodies or IgG. Immunoprecipitated proteins were probed by indicated antibodies. Experiments were performed 3 times. Representative immunoblot results (I) and relative quantification of immunoprecipitated 14-3-3ζ proteins (J) were shown. ∗∗P < .01 by 2-tailed unpaired Student’s t-test. (K, L) Stable Caco2 cell lines with Dox-inducible V5-FACI expression were treated with mock and Dox for 24 hours. Cells were then treated with PMA for 30 minutes and analyzed by immunoblotting with the indicated antibodies. Representative immunoblot images (K) were presented, and the ratio of phospho- to total proteins (L) was quantified. ∗P < .05; ∗∗P < .01 by 2-tailed unpaired Student’s t-test. (M, N) Enterocytes were isolated from the small intestines of WT and FACI^-/- mice (7 days old; male, n = 3 for each group) and lysed. The cell lysates were subjected to SDS-PAGE analysis and probed with indicated antibodies. Representative immunoblot images (M) were presented, and the ratio of phospho- to total proteins (N) was quantified. ∗∗P < .01 by 2-tailed unpaired Student’s t-test. All results are representative of 3 independent experiments.

Figure 16

(A) Proposed model depicting the physiological effects of FACI phosphorylation by PKCα. In the non-phosphorylated states, FACI is situated at CCPs of the PM, via binding with the AP2 complex through its motif D. PKCα phosphorylates FACI protein at S9 within motif A and S37 within motif C. Phosphorylated FACI loses its interaction with the AP2 complex, dissociates from CCPs, and subsequently binds with 14-3-3ζ. PKC activation also results in the translocation of intracellular FACI to the PM. FACI contains a poly-basic and hydrophobic motif E, which mediates FACI binding with PI(4,5)P₂ in the PM. FACI promotes PMA-induced macropinocytosis by two mechanisms. First, FACI impairs SSH1 binding with 14-4-3ζ. Increased free SSH1 activates cofilin-1, which severs actin filaments and provides more barbed ends for actin polymerization underneath the membrane ruffles. Second, FACI interacts with some actin-binding or actin-regulatory proteins and recruits them to the ruffle area, which facilitates actin remodeling and membrane ruffling required for macropinocytosis. (B) Summary of physiological functions of 5 motifs of FACI. The 3D structure of human FACI protein was predicted with AlphaFold2. Five FACI motifs are highlighted. Motif E contains a positively charged amphipathic α-helix and a hydrophobic disordered region, accounting for the binding of FACI to PIs in the inner PM. Motif E also contributed to PM translocation of FACI following PMA treatment. Motif D (DxxxLI) directly binds with the AP2 complex, localizing FACI to CCPs. Motifs A and C are 2 phosphorylatable motifs with similar sequence patterns (R-x-x-pS-[LF]-R). Their phosphorylation by conventional PKCs dissociates FACI from CCPs. Phosphorylated motifs A and C also enable FACI to bind with 14-4-3ζ. This interaction modulates the SSH1-cofilin-1 signaling, which in part mediates the promoting effects of FACI on macropinocytosis.