In response to increased intracellular calcium, the formin INF2 polymerizes 20-30% of the total cellular actin pool within 30 sec, suggesting robust regulation. INF2 regulation requires an auto-inhibitory interaction between the N-terminal Diaphanous Inhibitory Domain (DID) and the C-terminal Diaphanous Auto-regulatory Domain (DAD). DID mutations are dominantly linked to two human diseases and constitutively activate INF2. However, DAD binding to actin monomers competes with DID binding, disrupting regulation. Here, we use a novel cell-free assay for detailed investigation of INF2 regulation. Contrary to our previous findings, INF2 inhibition does not require CAP proteins but does require actin ‘buffering’ by monomer-binding proteins such as profilin or thymosin. INF2 is activated by calcium-bound calmodulin (CALM) through CALM binding to the N-terminus. In addition, the N-terminus plays an important role in INF2 regulation beyond CALM binding. These findings support a role for actin monomer binding proteins in not only regulating overall actin dynamics but also in specific regulation of an actin polymerization factor. 

Plasmids and siRNA oligonucleotides

The GFP-F-tractin and mApple-F-tractin plasmids were a gift from A. Pasapera (NIH, Bethesda, MD). The ER-tagRFP (modified from the GFP-N1 backbone) was a gift from E. Snapp (Albert Einstein College of Medicine, New York, NY); and contains a prolactin signal sequence at the 5’ end of the fluorescent protein as well as a KDEL sequence at the 3’ end. The GFP-fusion constructs of human INF2-CAAX (UniProt Q27J81-1) and INF2-nonCAAX (UniProt Q27J81-2) constructs used in this study have been describe previously (Chakrabarti et al., 2018; A et al., 2019), and are in a modified eGFP-C1 vector (Clontech) containing a Strep-tag II (IBA Life Sciences) and an HRV3C cleavage site between the multiple cloning site and the start codon of INF2. INF2 mutations (R218Q; W11L14L18 to A; DNT; NM; LM; R9A, R9K10K15K17 to A; A13T; and E16A) were introduced into the GFP-INF2-CAAX construct using site-directed mutagenesis. For bacterial expression, human INF2 FH1-FH2-C (FFC, amino acids 469-1250) has been described previously (Ramabhadran et al., 2012). Human INF2 (amino acids 1–420) has been described previously (Sun et al., 2011). Mutants (W11L14L18 to A, ΔNT, and LM) were generated in the wild-type construct using site-directed mutagenesis. Human INF2-C term (amino acids 941-1249) was PCR amplified from GFP-INF2-CAAX and cloned into a pHis6-parallel1 plasmid (NovoPro Bioscience; V005547).**** His6-TEV-Hs CALM (residue 2-148) was obtained from addgene (159693) and has been described previously (Agamasu et al., 2019). The human thymosin b4 (Uniprot ID: P62328) open reading frame was made by gene synthesis and cloned into a pMW plasmid. The codon-optimized insert is 5’- ATGAGCGATAAACCGGATATGGCGGAAATTGAAAAATTTGATAAAAGCAAACTGAAAAAAACCGAAACCCAGGAAAAAAACCCGCTGCCGAGCAAAGAAACCATTGAACAGGAAAAACAGGCGGGCGAAAGCTGA-3’ (Integrated DNA Technologies). Oligonucleotides used for siRNA mediated gene silencing (Integrated DNA Technologies): CAP1 (exon 12): 5’-GTCAGTGCCAAATCTTCC-3’, CAP2 (3’ UTR): 5’-CTTTGAGAATCTAAGATG-3’.

Cell culture

Human osteosarcoma U2-OS cells (ATCC HTB-96) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Corning, 10-013-CV) supplemented with 10% newborn calf serum (R&D Systems; S11250). Human cervical cancer HeLa cells (ATCC CCL-2) were cultured in DMEM supplemented with 10% fetal bovine serum (Sigma-Aldrich; F4135). INF2 knockout (KO) U2-OS cells were generated in-house using CRISPR/Cas9 technology (Chakrabarti et al., 2018). In brief, a guide RNA sequence targeting INF2 (5’-TCCGTGGGGTCCGAATCCTG-3’) was designed and cloned into the LentiCRISPRv2 (addgene; 52961) vector. The resulting guide plasmid, together with the packaging plasmids psPAX2 (addgene; 12260) and pMD2.G (addgene; 12259), was transfected into HEK293 cells using Lipofectamine LTX (Invitrogen; 15338100). The transfection medium was replaced with fresh medium 24 h post-transfection. Lentiviral supernatant was collected 48 h after transfection and filtered through a 0.45 mm filter. U2OS cells were infected with the freshly prepared lentivirus and, after 48 h, the infection medium was replaced with fresh medium containing 2 mg/mL puromycin for selection. Surviving cells were subjected to puromycin selection, and single-cell clones were isolated by fluorescence-activated cell sorting (FACS) into 96-well plates. Clones were screened and validated for INF2 knockout by immunofluorescence and Western blotting. A U2-OS INF2 KO cell line stably expressing GFP-INF2-CAAX was made by transfecting the GFP-INF2-CAAX plasmid described above into INF2-KO U2-OS cells, then selecting cells with G418 (Sigma-Aldrich; A1720). G418-resistant cells were then enriched for GFP expression by multiple rounds of flow sorting (Dartmouth Flow Cytometry Core) to establish a cell population in which >90% of cells were GFP-positive. Cells were periodically flow sorted thereafter to maintain enrichment. All cell lines were maintained at 37 °C in a humidified incubator with 5% CO₂. Cells were assayed every 6 months for mycoplasma using the MycoStrip kit (InvivoGen; rep-mys-50).

Transfections

For plasmid transfections, cells were seeded at 5 × 10⁵ cells per 35-mm well approximately 18 hours before transfection. Transfections were performed in OPTI-MEM medium (Life Technologies; 31985062) using 2 ml Lipofectamine 2000 (Invitrogen; 11668) per well for 3.5 hours. Cells were then trypsinized and replated onto glass-bottom dishes (MatTek Corporation; P35G-1.5-14-C) at a density of ~3.3 × 10⁵ cells per well. Imaging was performed ~24 hours post-transfection in live-cell imaging medium (DMEM supplemented with 25 mM D-glucose, 4 mM L-glutamine, 25 mM HEPES, and 10% newborn calf serum). For all experiments, the following amounts of DNA were transfected per well (individually or combined for cotransfections): 400 ng for GFP-F-tractin, 400 ng for mApple-F-tractin, 400 ng for ER-tagRFP, and 200 ng for GFP-tagged INF2-CAAX constructs.

For siRNA transfections, 1 × 10⁵ cells were seeded per well in six-well plates. Each well received 2ml RNAiMAX (Invitrogen; 13778) and 63 pg of siRNA. Transfections were repeated after 48 hours, and cells were analyzed 96 hours after the original transfection.

For live-cell imaging experiments following siRNA treatment, mApple-F-tractin was transfected at 72 hours post-siRNA treatment as described above. 3.3 × 10⁵ cells were replated onto 35-mm glass-bottom dishes and imaged at 96 hours post-transfection.

Primary Antibodies

Anti-actin (mouse, mab1501R; Millipore) was used at 1:10,000. Anti-tubulin (mouse, T9026; Millipore) was used at 1:10,000. Anti-GFP (mouse, A11120; Thermo Fisher) was used at 1:1,000. Anti-calnexin (rabbit, Cell Signaling Technology; 2433) was used at 1:1,000. Anti-ATP synthase subunit b (mouse, A21351; Thermo Fisher Scientific) was used at 1:2,000. Anti-GM130 (rabbit, ab52649; Abcam) was used at 1:1,500. Anti-GAPDH (mouse, sc-365062; Santa Cruz Biotechnology) was used at 1:3,000. Anti-profilin1 (rabbit, P7624; Millipore) was used at 1:2,000. Anti-cofilin (rabbit, 5175; Cell Signaling Technology) was used at 1:1,000. Anti-FMNL1 (rabbit, HPA008129; Millipore) was used at 1:1,000. Anti-FMNL2 (rabbit, HPA005464; Millipore) was used at 1:1,500. Anti-FMNL3 (rabbit, NBP2-24724; Novus Biologicals) was used at 1:1,000. Anti-Arp2 (mouse, ab129018; Abcam) was used at 1:500. Anti-Arp3 (mouse, sc-48344; Santa Cruz Biotechnology) was used at 1:1,000. Anti-Myosin IIA (rabbit, 3403; Cell Signaling Technology) was used at 1:1,000. Anti-Myosin IIC (rabbit, 8189; Cell Signaling Technology) was used at 1:1,000. Anti-fascin (mouse, SC-21743; Santa Cruz Biotechnology) was used at 1:1,000. Anti-actinin 1 (mouse, 05-384; Millipore) was used at 1:1,000. Anti-actinin 4 (rabbit, 15145s; Cell Signaling Technology) was used at 1:1,000. Anti-calmodulin (rabbit, D1F7J; Cell Signaling Technology) was used at 1:1,000. Anti-INF2 was described previously (Ramabhadran et al., 2011), amino acid 469-1249, rabbit) and was used at 1:1,000. Polyclonal antibody against bacterially-expressed mouse mDia1 amino acids 1-549 (expression/purification described in (Li and Higgs, 2003) was raised in chicken (Cocalico Biologicals) and affinity purified by using the antigen coupled to Sulfolink (Pierce, Thermo Fisher Scientific, Rockford, IL). The rabbit polyclonal antibody against mouse mDia2 (also called DIAPH3) amino acids 1-520 was described in (Hager et al., 2012), and referred to as HNH3.1 in that publication. Rabbit polyclonal antibodies against CAP1 and CAP2 were described in (A et al., 2020). Rabbit polyclonal antibody against human INF2 FH1-FH2 region (amino acids 469-940) was previously described (Ramabhadran et al., 2011). Polyclonal antibody against mouse INF2 (Q0GNC1) N-terminal region (containing the DID) amino acids 1-424 was raised in rabbits (Cocalico Biologicals) and affinity purified by using the antigen coupled to Sulfolink (Pierce, Thermo Fisher Scientific, Rockford, IL).

Western blotting

For analysis of whole-cell extract preparations, confluent cells in 35-mm dishes were lysed in ~300 µl of 1× DB buffer (50 mM Tris-HCl, pH 6.8; 2 mM EDTA; 20% glycerol; 0.8% SDS; 0.02% bromophenol blue; 1 M NaCl; 4 M urea), followed by heating at 95 °C for 5 min. Genomic DNA was sheared by passing the lysate through a 27G needle. Proteins were resolved by standard SDS–PAGE and transferred onto PVDF membranes (Millipore). Membranes were blocked in TBS-T (20 mM Tris-HCl, pH 7.6; 136 mM NaCl; 0.1% Tween-20) containing 3% bovine serum albumin for 1 h at room temperature, then incubated overnight at 4 °C with primary antibodies. After washing with TBS-T, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (goat anti-mouse, #1721011, Bio-Rad; goat anti-rabbit, #1706515, Bio-Rad; rabbit anti-chicken/turkey, #613120, Molecular Probes) for 1 h at room temperature. Detection was performed using ECL Prime Western Blotting Detection Reagent (45-002-40, Cytiva Amersham), and chemiluminescent signals were captured with an ECL Chemocam imaging system (SYNGENE G:BOX Chemi XRQ). For fluorescent Western blotting, membranes were incubated for 1 h at room temperature with IRDye-conjugated secondary antibodies: IRDye 680 goat anti-mouse (926-68070, LI-COR) or IRDye 800CW goat anti-rabbit (926-32211, LI-COR). Signals were detected using the LI-COR Odyssey CLx imaging system.

Protein expression and purification

GFP-INF2-nonCAAX (plasmid described above) was expressed and purified following a modified version of a previously published protocol (A et al., 2020). Briefly, the construct was transfected into 2 L of Expi293 cells and protein was expressed for 48 h. Cells were harvested by centrifugation at 300 × g for 15 min and resuspended in 100 ml of EB-S buffer (100 mM HEPES, pH 7.4, 500 mM NaCl, 5 mM EDTA, 1 mM DTT, 1% Triton X-100, supplemented with 2 mg/ml leupeptin, 10 mg/ml aprotinin, 2 mg/ml pepstatin A, 1 mg/ml calpeptin, 1 mg/ml calpain inhibitor I, 1 mM benzamidine, and 1:1,000 dilution of universal nuclease [Thermo Fisher; 88702]). Cells were lysed using a microfluidizer. Cell debris was removed by ultracentrifugation at 185,000 × g for 1 h (Ti45 rotor, Beckman). The supernatant was incubated with 20 mg/ml avidin (Sigma-Aldrich; 189725) to block endogenous biotin before loading onto a 5 ml Strep-Tactin Superflow column (IBA; 2-1206-025) pre-equilibrated with EB-S buffer. The column was washed extensively with WB buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM DTT) and eluted with Strep elution buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM DTT, and 2.5 mM desthiobiotin). Eluted protein was concentrated using a 100 kDa MWCO Amicon Ultra-15 centrifugal filter (Millipore) and further purified on a Superdex 200 Increase 10/300 GL column (GE Biosciences) equilibrated in 1x K50MEHD buffer (50 mM KCl, 1 mM MgCl₂, 1 mM EGTA, 10 mM HEPES, pH 7.4, and 1 mM DTT). Protein was aliquoted and snap-frozen in liquid nitrogen.

INF2-FH1-FH2-C (FFC) constructs (wild-type and the chimera with mDia1-DAD) and INF2 1–420 constructs (wild-type, W11L14L18>A mutant, ΔNT, and LM), and the INF2 C-terminal region (CT) were expressed in E. coli as GST fusion proteins, as described previously (Gaillard et al., 2011). Bacterial lysates were applied to glutathione-Sepharose resin (GE Biosciences; 17513201), and bound proteins were cleaved on-resin with either tobacco etch virus (TEV) protease for INF2-FFC or thrombin for INF2 1–420 and INF2-CT. Proteins were further purified by gel filtration using a Superdex 200 16/600 column (GE Biosciences) equilibrated with 1x K50MEHD buffer.

Profilin I was expressed in E. coli and purified as described previously (Harris et al., 2004). In brief, bacterial pellet expressing profilin I was resuspended in extraction buffer (50 mM Tris-HCl, pH 8.0, 500 mM NaCl, 5 mM EDTA,1mM DTT, and 1pill/50ml Complete protease inhibitors cocktail [Roche; 04693116001]) and lysed by sonication. The lysate was cleared by ultracentrifugation at 185,000 × g for 1 h at 4˚C (Ti45 rotor, Beckman), and the supernatant was loaded to a poly-L-proline (Sigma P-3886) affinity column coupled to CNBr-activated Sepharose (Amersham Biosciences). The column was washed sequentially with buffer 1 (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1 mM DTT), buffer 2 (buffer 1 containing 3 M urea), and buffer 3 (buffer 1 containing 8 M urea). The buffer 3 eluate was dialyzed overnight at 4 °C against dialysis buffer (DB; 2 mM Tris-HCl, pH 8.0, 0.2 mM EGTA, 1 mM DTT, 0.01% NaN₃), followed by an additional 2 h dialysis in fresh DB. Thymosin β4 was also expressed in E. coli in the same manner, resuspended in buffer containing 10 mM HEPES, pH 7.4, and 1 mM EDTA, and lysed by sonication. Lysates were cleared by ultracentrifugation at 185,000 × g for 1 h at 4˚C (Ti45 rotor, Beckman), and the supernatant was subjected to stepwise ammonium sulfate precipitation (40%, 60%, and 80% saturation). The 80% supernatant was dialyzed into water (three 24 hr dialyses in 20 L), then lyophilized in a speed vac. Final protein preparations were aliquoted and snap-frozen in liquid nitrogen.

For INF2 constructs, concentration of the final fraction was determined by Bradford assay (BioRad 5000006) and confirmed by Coomassie-stained SDS-PAGE, using known amounts of actin as standards. For profilin and thymosin, concentration and purity of the final fraction was determined by amino acid analysis (Molecular Structure Facility, UC Davis). 

Preparation of INF2-N terminal FITC labeled peptide

N-terminally FITC-Ahx–labeled INF2 N-terminal peptide (FITC-MSVKEGAQRKWAALKEKLGPQDSDPTEAN) was synthesized by BIOMATIK. A total of 5 mg peptide was dissolved in Milli-Q water to 1 mM, and the pH was adjusted to 7.0 using NaOH.

Fluorescent labeling of INF2-C term

INF2-Cterm was dialyzed overnight against 1 L of DB buffer (2 mM NaPO4, pH 7.0; 100 mM NaCl; 0.5 mM EDTA; 0.25 mM DTT). A 37.5 mM solution of the protein was then mixed with 375 µM fluorescein succinimide (FITC) (Thermo Fisher; 46409) in 50 mM NaPO4, pH 6.0, 100 mM NaCl at 25 °C. The labeling reaction was terminated by adding Tris-HCl (pH 8.0) to a final concentration of 100 mM and DTT to 20 mM. The mixture was subsequently gel-filtered using a Superose 12 column (GE Biosciences) equilibrated in polymerization buffer (1× K50MEH). Final protein concentration was measured by Bradford assay, and FITC concentration was determined using an extinction coefficient of 75,000 M⁻¹ cm⁻¹ at 490 nm. The calculated FITC:INF2-Cterm labeling ratio was 1.02.

mDia1 DAD and INF2-mDia1-DAD chimera

The DAD sequence from mouse mDia1 (DETGVMDSLLEALQSGAAFRRKRG) was used in these experiments. For fluorescence anisotropy binding studies, the peptide was synthesized as described in (Li & Higgs, 2005), and labeled with TAMRA-maleimide following the same procedure as described in that publication. The chimeric INF2-mDia1-DAD constructs were made in both human INF2-FFC and full-length INF2-CAAX by replacing amino acids 968-986 (EEVCVIDALLADIRKGFQL) with the mDia1 sequence above. To generate INF2-mDia1-DAD, bases 2026-3062 (containing amino acids 675-1020) of human INF2 were synthesized (Integrated DNA Technologies) with mDia1-DAD replacing the INF2-DAD. The construct includes an AflII site at the 5’ end and an SbfI site at the 3’ end, that are unique within the human INF2 sequence. The gene fragment was cloned into either GFP-INF2-CAAX or INF2-FH1-FH2-C using the AflII and SbfI restriction sites.

Cell treatments

For ionomycin or histamine treatment, 5 x 10⁵ cells were plated onto 35-mm glass-bottom dishes (MatTek). The following day, regular growth medium was replaced by 1 ml of live-cell medium (DMEM with 25 mM D-glucose, 4 mM L-glutamine, and 25 mM HEPES, supplemented with 10% newborn calf serum) (Gibco; 31053028). Another 1 ml of live-cell medium containing 8 mM ionomycin (Sigma; I0634; from a 1 mM stock in DMSO) or 100 mM histaimine (Sigma; H7125; from a 100 mM stock in DMSO) was added swiftly in a circular motion onto the outside of the plate (so as not to activate CIA by mechanical stimulation, (Shao et al., 2015). For ionomycin treatment during live-cell imaging, ionomycin stimulation was applied at the fifth frame (15-second intervals; four imaging fields selected) and imaging was continued for an additional 5 minutes. For histamine stimulation, the procedure was the same, except images were acquired at 10-second intervals for an additional 3 minutes and 20 seconds after drug addition. For fixed-cell imaging, medium was removed and cells were fixed in pre-warmed 4% paraformaldehyde (PFA; 15170; Electron Microscopy Sciences) and 0.25% glutaraldehyde (16020; Electron Microscopy Sciences) in PBS for 15 minutes at room temperature. After fixation, cells were washed three times with PBS and permeabilized with 0.1% Triton X-100 in PBS for 1 minute. Cells were then washed again with PBS and blocked with 10% calf serum in PBS for 20 minutes. Actin filaments were stained with 1mM TRITC–phalloidin (Sigma-Aldrich, P1951) and DNA was stained with 0.1 mg/L (w/v) 4,6-diamidino-2-phenylindole (DAPI; Calbiochem, 268298). Cells were washed three times with PBS and stored in PBS at 4 °C until imaging.

Microscopy

Both fixed and live cell samples were imaged using the Dragonfly 302 spinning disk confocal system (Andor Technology) mounted on a Nikon Ti-E base, equipped with an iXon Ultra 888 EMCCD camera and a Tokai Hit stage-top incubator maintained at 37°C. A solid-state 405 smart diode 100-mW laser, solid-state 488 OPSL smart laser 50-mW laser, and a solid-state 560 OPSL smart laser 50-mW laser were used for excitation. Images were acquired with a 100× 1.4 NA CFI Plan Apo objective (Nikon) using Fusion software (Andor Technology).

CIA intensity measurement in live and fixed cells

For both live-cell and fixed-cell imaging, exposure times and laser intensities were kept constant across all cell lines and treatment conditions. A 25 mm² circular region was selected in the perinuclear cytoplasmic area of each cell, specifically within a medial confocal plane to avoid signal contributions from basal stress fibers or cortical actin. Mean fluorescence intensity (arbitrary units) in this region was measured using ImageJ. For live-cell imaging, Ftractin values at each time point after drug treatment (F) were normalized to the baseline fluorescence (F₀; average of the first four frames) and plotted over time as F/F₀ using Microsoft Excel. For fixed-cell imaging, TRITC–phalloidin fluorescence intensity values were normalized to the TRITC fluorescence measured from the corresponding nuclear area in each cell and plotted as dot plot using GraphPad Prism. Statistical significance between two groups was determined using unpaired Student’s t-tests, comparing the full datasets as indicated in the figure legends.

Actin preparation

Rabbit skeletal muscle actin (RSKA) was purified from acetone powder as previously described**** (Spudich and Watt, 1971). In brief, RSKA acetone powder was resuspended in G-buffer (2 mM Tris-HCl, pH 8.0, 0.5 mM DTT, 0.2 mM ATP, 0.1 mM CaCl₂, and 0.01% sodium azide) and incubated at 4 °C for 30 min with end-to-end mixing. The suspension was then ultracentrifuged at 185500 × g (40,000rpm in Beckman Ti45 rotor), and the supernatant containing actin was collected. The supernatant was filtered through a paper towel to remove the debris, mixed with final 50 mM KCl, 2 mM MgCl₂ and incubated at 4 °C for 1 h with stirring. Subsquently, final 0.8 M KCl was added and the mixture was incubated for an additional 30 min. Polymerized actin was pelleted by ultracentrifugation at 185500 × g (40,000rpm in Beckman Ti45 rotor), and resuspended in G-buffer. The actin solution was then dialyzed against fresh G-buffer for 2 days, with daily buffer change. Finally, monomeric actin was obtained by ultracentrifugation at 329700 × g (60,000rpm in Beckman Ti70.1 rotor) and labeled with either pyrenyliodoacetamide (Molecular Probes; P29) or TAMRA NHS ester (Molecular Probes; C1171). Both unlabeled and labeled actin were gel-filtered using a Superdex 75 column and stored in dialysis in G-buffer at 4 °C. Dialysis buffer changed weekly.

Low speed pellet (LSP) preparation

Cells were plated at 4 × 10⁶ cells per 100 mm dish, with two dishes prepared for each cell line. The following day, cells were washed twice with 5 ml PBS and collected using a cell scraper in 2 ml of 2× extraction buffer (1× KMEH: 50 mM KCl, 1 mM MgCl2, 1 mM EGTA, 10 mM HEPES, pH 7.4) supplemented with 1 mM DTT and protease inhibitors (leupeptin 2 mg/ml, aprotinin 10 mg/ml, pepstatin A 2 mg/ml, benzamidine 2 mM, calpain inhibitor I (ALLN) 1 mg/ml, calpeptin 1 mg/ml, cathepsin B inhibitor II 0.05 mg/ml). Cells were lysed using a metal Dounce homogenizer (DWK Life Sciences). The total cell lysate was centrifuged at 600 × g (1800 rpm in Beckman Allegra 6R swinging bucket centrifuge) for 5 min at 4°C. The supernatant was removed, and the pellet was washed with 2 ml of 2× extraction buffer, followed by another centrifugation at 600 × g for 5 min at 4°C. The pellet was then resuspended in 0.5 ml of 2× extraction buffer. Total protein concentration in the LSP fraction was determined by Bradford assay (Bio-Rad). For LSP preparation from transiently transfected cells, 2 × 10⁶ cells were plated on a 100 mm dish. The following day, transfections were performed in OPTI-MEM medium (Life Technologies; 31985062) using 7 mg DNA and 10 ml Lipofectamine 2000 (Invitrogen; 11668) per dish for 3.5 hours. LSP was then prepared as described above, with the final resuspension in 200 ml of 2× extraction buffer. For LSP preparation from CAP1 and CAP2 knockdown cells, 1.5 × 10⁶ cells were used. Knockdown was performed as described, and cells were harvested for LSP preparation 48 hours post-transfection. LSP protein concentration determined by BCA assay (ThermoFisher 23250).

Free calcium concentration calculations

Free calcium was concentrated for our standard actin polymerization buffer (1× KMEH: 50 mM KCl, 1 mM MgCl2, 1 mM EGTA, 10 mM HEPES, pH 7.4) using Ca-EGTA Calculator v1.3 (https://somapp.ucdmc.ucdavis.edu/pharmacology/bers/maxchelator/CaEGTA-TS.htm). As an example, total CaCl₂ added to conduct the free calcium concentration curve in Figure 3C were: 0.113 mM (10 nM free), 0.273 mM (25 nM), 0.463 mM (50 nM free), 0.683 mM (100 nM), 0.893 mM (200 nM free), 0.993 mM (300 nM), 1.048 mM (400 nM free), 1.087 mM (500 nM), and 1.174 mM (1 mM free).

Pyrene actin polymerization assay and fluorescence anisotropy

Actin monomers in G-buffer (6.7 mM actin, 5% pyrene-labeled) were converted to the Mg²⁺-bound form by adding 0.1 volumes of 10 mM EGTA, 1 mM MgCl₂ for 2 minutes at 23 °C immediately prior to polymerization. Formin protein or 10 mg of LSP was premixed with other proteins (profilin, thymosin, capping protein) in 1.43× polymerization buffer (71.4 mM KCl, 1.43 mM MgCl₂, 1.43 mM EGTA, 14.3 mM HEPES pH 7.4, 1.9 mM DTT, 1.9 mM Tris-HCl, 0.19 mM ATP, 0.09 mM CaCl₂, and 0.009% w/v NaN₃). Pyrene fluorescence (excitation 365 nm, emission 410 nm) was measured using a 96-well fluorescence plate reader (Infinite M1000; Tecan), starting within 1 minute of mixing actin with the formin/LSP premix in the plate. The slope of the fluorescence curve at half-maximum intensity was calculated to represent INF2 activity using KaleidaGraph 4.5 (Synergy Software).

Anisotropy readings were conducted with FITC- or TAMRA-labeled peptides and the indicated binding proteins, in 50 mM KCl, 1 mM MgCl2, 1 mM EGTA, 10 mM Hepes pH 7.4, 0.01% Thesit (Sigma-Aldrich).

Microscopy-based assays using LSP

For TAMRA-actin imaging, LSP was prepared as described above from U2-OS INF2 KO cells or U2-OS INf2 KO stably expressing GFP-INF2-CAAX. Then, 10 mg of LSP was premixed with other proteins (profilin, capping protein) in 1.43× polymerization buffer (71.4 mM KCl, 1.43 mM MgCl₂, 1.43 mM EGTA, 14.3 mM HEPES pH 7.4, 1.9 mM DTT, 1.9 mM Tris-HCl, 0.19 mM ATP, 0.09 mM CaCl₂, and 0.009% w/v NaN₃). Actin monomers in G-buffer (6.7 mM actin, 10% TAMRA-labeled) were converted to the Mg²⁺-bound form by adding 1 mM EGTA and 0.1 mM MgCl₂ (from a 10× stock) for 2 minutes at 23 °C immediately prior to polymerization. Actin was then mixed with LSP and other components, incubated at 23 °C for 10 minutes, and subsequently placed on ice. The mixture was then combined with an equal volume of 1× extraction buffer containing 0.2 mg/L (w/v) DAPI (Calbiochem, 268298). A 2 μl aliquot was mounted onto a glass slide and covered with a coverslip (72222-01; Electron Microscopy Sciences) for imaging. For TRITC-phalloidin imaging, 10 mg of LSP was incubated with 1 mM TRITC–phalloidin and 0.1 mg/L (w/v) DAPI in 100 ml of 1× extraction buffer. Samples were imaged after a 10-minute incubation at room temperature.

Calculation of ratios of actin and actin binding proteins in U2OS cells

We used publically-available data from the DepMap portal (Broad Institute, 25Q2 dataset) to determine overall transcript levels for actin, capping protein, thymosin, and profilin. Values used are in TPM (transcripts per million). For actin, thymosin, and profilin, we used the sum of expression from all genes. Actin transcript levels (seven genes): ACTA1, 0.11; ACTA2, 2.40; ACTB, 3843.33; ACTBL2, 3.56; ACTC1, 0.10; ACTG1, 1425.99; ACTG2, 0.24; total actin, 5275.73. Thymosin transcript levels (five genes): TMSB4X, 3236.75; TMSB4Y, 0.00; TMSB10, 2973.20; TMSB15A, 108.11; TMSB15B, 14.45; TMSB15C, 3.99; total thymosin, 6336.50. Profilin transcript levels (four genes): PFN1, 1243.73; PFN2, 163.93; PFN3, 0.00; PFN4, 1.11; total profilin, 1408.77. For the capping protein heterodimer, we used the average of CAPZA (sum of the three CAPZA genes) and CAPZB (single gene). Capping protein transcript levels (four genes): CAPZA1, 138.76; CAPZA2, 51.42; CAPZA3, 0.00; CAPZB, 168.44; total capping protein, 179.31. From these values, the ratios of actin:capping protein:thymosin:profilin are 1:0.034:1.20:0.27.

Actin pelleting assay

Performed as previously described (Kage et al., 2022) with modifications. Briefly, U2-OS cells were seeded 6 x 10⁵ cells per 35 mm well the day prior to the experiment. The following day, cells were treated with one of three conditions: live cell medium alone, 4 mM ionomycin in live-cell medium for the indicated time or 100 mM histamine in U2-OS culture medium for the indicated time. Following treatment, the medium was removed quickly, and cells were extracted with 2ml of extraction buffer (1xNa50MEH [500 mM NaCl, 20 mM MgCl2, 5 mM EGTA, 100 mM HEPES], 1 mM DTT, 0.4 mM phalloidin, 0.4 mM LatA, and 1% Triton X-100). From each extract, 750 ml was collected as the input sample and mixed with 250 ml of 4x sample buffer (500 mM Tris, pH 6.8, 4 mM EDTA, 40% glycerol, 8% SDS, 40 mM DTT). An additional 1 ml of lysate was transferred to a TLA100 ultracentrifuge tube and centrifuged at 80,000 rpm in a TLA120 rotor (Beckman) for 22 minutes at 4 °C. After centrifugation, 750 ml of the supernatant (containing monomeric actin) was carefully removed and mixed with 250 ml of 4x sample buffer. The pellet (containing filamentous actin) was washed once with 1 ml of 1× Na50MEH containing 1% Triton X-100 (by carefully adding then immediately removing), then resuspended in 1.33 ml of 1x sample buffer. All samples were heated at 95 °C for 5 minutes. Ten microliters of each sample was loaded onto SDS-PAGE gels alongside standard dilutions of purified actin and tubulin protein. Proteins were analyzed by Western blot using an Odyssey CLx imager (LI-COR), using the following antibodies: anti-actin (mouse, mab1501R; Millipore; 1: 10,000) and anti-tubulin (mouse, T9026; Millipore; 1: 10,000). Intensities of actin and tubulin bands determined using LI-COR software Image Studio Lite Ver. 5.2.

For the actin pelleting assay of LSP, 150 mg of LSP was resuspended in 200 ml of extraction buffer containing 1 unit benzonase (Millipore; 70664) along with other additives as indicated (2 mM phalloidin, and 2 mM LatA) and incubated overnight at 4˚C. Samples were then centrifuged at 80,000 rpm in a TLA 100 rotor for 22 minutes at 4 °C. After centrifugation, 150 ml of the supernatant (containing monomeric actin) was carefully removed and mixed with 50 ml of 4x sample buffer. The pellet (containing filamentous actin) was washed once with 200 ml of 1× Na50MEH containing 1% Triton X-100, then resuspended in 267 ml of 1x sample buffer.

Calculation of theoretical rate maximal cellular rate of actin addition to INF2-bound barbed ends took into account the on-rate of an actin monomer onto an INF2-bound barbed end in the presence of profilin (~6.5 mM^-1sec^-1,**** Gurel et al., 2015), 192 million actin molecules/cell in U2-OS (the mean of three independent measurements of total actin content conducted in our laboratory (Hatch et al., 2016; A et al., 2020; Kage et al., 2022), a cytoplasmic volume of 3.14 pL (A et al., 2020), and an estimate that 50% of the cellular actin is monomeric at the time of cell stimulation. Based on these values, U2-OS cells possess 91 mM total actin, of which 45.5 mM is monomeric. Each INF2-bound barbed end should, therefore, allow 296 actins per sec to add to the barbed end. Our estimate of 35 actins/sec/INF2 is lower than this value.

Actin quantification in LSP

For the quantification of actin in the LSP fraction (Supplementary Figure 2C), U2-OS INF2 KO cells stably expressing GFP-INF2-CAAX were seeded at 4 x 10⁶ cells per 100 mm dish, with two dishes prepared. LSP was prepared according to the method described above. Briefly, cells were extracted with 2.2 ml of extraction buffer. A 0.2 ml aliquot of the total cell extract was reserved, and the remaining 2 ml was processed for LSP preparation. The resulting LSP was resuspended in 0.5 ml of extraction buffer. Protein concentrations of both the total cell extract and the LSP were determined using the Bradford assay, and the indicated amount of proteins were loaded onto SDS-PAGE gel alongside actin and GAPDH protein standards. Proteins were analyzed by standard Western blot using using the following antibodies: anti-actin (mouse, mab1501R; Millipore; 1: 10,000) and anti-GAPDH (mouse, sc-365062 (G-9); Santa Cruz; 1: 3,000).

Computational Modeling

A structural model of full-length human INF2-CAAX (Uniprot Q27J81-1) was made using AlphaFold3. AlphaFold returned five models, of which the one with the highest confidence score is depicted in Supplementary Figure 8, though the orientations of DID, DAD, and N-terminus were similar in the other four models produced. A second structural model of a human INF2 variant lacking the DAD (https://alphafold.ebi.ac.uk/entry/A0A087X118) was also used. This model was further subjected to molecular dynamics simulations using Amber 22. The model was prepared for simulation using the ff19SB force field (Tian et al., 2020) and solvated in an octagonal box of OPC3 waters (Izadi and Onufriev, 2016) with a minimum of 8 Å between the protein and the edge of the box in all dimensions. Hydrogen mass repartitioning (Hopkins et al., 2015) was applied to allow for a 4-fs simulation time step. The model was energy minimized over 5000 steps, then heated at constant volume from 200 to 310 K over 11,000 4-fs steps. Finally, 100-ns of constant-pressure simulation were performed. Simulations used the Andersen thermostat (DOI: 10.1063/1.439486) with a frequency of 1000 steps.

Data processing and statistical analyses

For the cell imaging data, brightness and contrast were adjusted uniformly using ImageJ software. Figures were compiled and finalized using PowerPoint 2016. Data processing was performed in ImageJ and Microsoft Excel. Statistical analyses, including p-value calculations, were conducted using GraphPad Prism version 6.01. Comparisons between two groups were made using unpaired Student’s t-tests. Statistical significance was defined as p ≤ 0.05 and is indicated by an asterisk in figure panels. Additional significance levels are represented as follows: ** for p ≤ 0.01, *** for p ≤ 0.001, and **** for p ≤ 0.0001.