Anti-RUVBL1 antibody (H00008607-M01) was purchased from Abnova (Taipei, Taiwan, R.O.C.). JLA20 anti-actin antibody (MABT219) was purchased from Millipore (Temecula, CA). Anti-vitamin D-binding protein antibody (ab65636) was purchased from Abcam (Cambridge, MA).

The human PDAC cell line S2-013, a derivative of SUIT-2, was obtained from Dr T. Iwamura (Miyazaki Medical College, Miyazaki, Japan) (25). All cells were grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco-BRL, Carlsbad, CA) supplemented with 10% heat-inactivated fetal calf serum (FCS) at 37°C in a humid atmosphere saturated with 5% CO₂.

A single mixture with four different short hairpin small interfering RNA (siRNA) oligonucleotides targeting RUVBL1 was purchased from Qiagen (FlexiTube GeneSolution GS8607; Valencia, CA) and a single mixture with four different scrambled negative control siRNA oligo-nucleotides was obtained from Santa Cruz Biotechnology (37007; Santa Cruz, CA). To examine the effect of the siRNAs on RUVBL1 expression, S2-013 and PANC-1 cells that expressed RUVBL1 were plated in 6-well plates. After 20 h, the cells were transfected with 80 pmols of each siRNA mixture in siRNA transfection reagent (Qiagen) following the manufacturer’s instructions. After incubation for 48 h, cells were processed for western blot analysis, transwell motility or Matrigel invasion assays.

Each cell pellet was resuspended in 20 mM HEPES (pH 7.4), 100 mM KCl, 2 mM MgCl₂, 0.5% Triton X-100, protease inhibitor cocktail tablets (Roche, Penzberg, Germany) and phosphatase inhibitor cocktail (Nacalai, Kyoto, Japan). The bicinchoninic acid (BCA) assay was used to determine protein concentration in each lysate; an aliquot of each lysate was then diluted with sample buffer (50 mM Tris, 2% SDS, 0.1% bromophenol blue and 10% glycerol) to a final concentration of 1–2 μg/μl and analyzed by SDS-PAGE and western blot analysis.

Coverslips were treated with 10 μg/ml fibronectin (Sigma-Aldrich, St. Louis, MO) for 1 h at room temperature. Cells were seeded on fibronectin-coated glass coverslips and incubated for 5 h; cells were then fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, covered with blocking solution (3% BSA/PBS), and then incubated with the primary antibody for 1 h. Alexa488- or Alexa594-conjugated secondary antibody (Molecular Probes, Carlsbad, CA) was used with or without rhodamine-conjugated phalloidin (Cytoskeleton, Denver, CO). Each specimen was visualized using a Zeiss LSM 510 META microscope (Carl Zeiss, Gottingen, Germany).

A plastic pipette tip was used to cut cross-shaped wounds through a confluent cell monolayer; cells were then allowed to polarize and migrate into a wounded area. After 4 h, cells were immunostained with a primary antibody and then incubated with a fluorophore-conjugated secondary antibody as described above. Each specimen was examined using a Zeiss LSM 510 META microscope (Carl Zeiss).

Cells (3.0×10⁴/chamber) were plated in the upper chamber of BD BioCoat Control Culture Inserts (24-well plates, 8-μm pore size; Becton-Dickinson, San Jose, CA). Serum-free culture medium was added to each upper chamber, and medium containing 5% FCS was added to each lower chamber. Cells were incubated on the membranes for 12 h. After this 12-h incubation, three independent visual fields in each lower chamber were examined via microscopic observation to count the number of cells that had moved from the top chamber to the lower chamber.

A two-chamber invasion assay was used to assess PDAC cell invasiveness (24-well plates, 8-μm pore size membrane coated with a layer of Matrigel extracellular matrix proteins; Becton-Dickinson). Cells (4.0×10⁴/chamber) suspended in serum-free medium were seeded into an upper chamber and allowed to invade towards a 5% FCS chemoattractant in a respective lower chamber. After 20-h incubation, three independent visual fields/lower chambers were examined via microscopic observation to count the number of cells that had moved to the bottom chamber.

S2-013 cells were seeded onto fibronectin and incubated for 5 h. Cells were lysed in lysis buffer [20 mM HEPES (pH 7.4), 100 mM KCl, 5 mM MgCl₂, 0.5% Triton X-100, protease inhibitor cocktail tablets (Roche), and phosphatase inhibitor cocktail (Nacalai)]. Lysates were immunoprecipitated with Dynabeads Protein G (Dynal, Oslo, Norway) and with anti-RUVBL1 antibody or normal mouse IgG (isotype control) for 2 h at 4°C. Beads were pelleted on a magnetic rack (Dynal). Co-immunoprecipitated proteins were separated on a 4 to 20% gradient SDS-PAGE gel and then silver stained. Bands precipitated by the anti-RUVBL1 antibody were excised from the gel; a nano-LC MS/MS system, which consisted of an Ultimate HPLC system (Agilent 1100; Agilent Technologies, Santa Clara, CA) and a QSTAR XL mass spectrometer (Applied Biosystems/MDS SCIEX, Concord, ON, Canada) equipped with a nano-ESI source was used characterize the excised proteins (Genomine, Inc., Pohang, Korea). MASCOT v1.9.0 (Matrix Science, Boston, MA) was used to perform database searches with findings from the MS/MS spectra (Genomine, Inc.).

Actin exists in equilibrium between monomeric subunits and polymeric filaments. To accurately quantify these actin forms, a commercially available actin polymerization assay (BK003; Cytoskeleton) was used to measure actin polymerization under defined conditions; specifically, polymerization-dependent increases in fluorescence of pyrene-conjugated actin were measured. Each of four concentrations (10, 30, 100 or 300 μg/ml) of recombinant human RUVBL1 protein (TP301170; Origene, Rockville, MD) were added to a separate actin polymerization assay. Briefly, the actin polymerization assays were based on the enhanced fluorescence of pyrene-conjugated actin that occurs during polymerization. The enhanced fluorescence was measured by pyrene monomer G-actin formed polymer pyrene F-actin in a fluorometer at excitation wavelength 365 nm and emission wavelength 407 nm. To quantify changes in polymerization rate, Boltzmann sigmoidal equations (GraphPad Prism version 6.0 software; GraphPad Software, Inc., La Jolla, CA) were used to fit curves to the fluorescence data. Half-maximal saturated polymerization values [T^1/2max (s)] were calculated from raw data.

GraphPad Prism version 6.0 software was used for all statistical analyses. Statistical significance was determined using a two-tailed Student’s t-test and standard deviations. For all analyses, p<0.05 was considered significant.

To investigate whether RUVBL1 regulates cell growth, motility and invasion, RUVBL1 expression was suppressed by a single mixture with four different siRNA oligonucleotides against RUVBL1 in moderately differentiated PDAC cells (line S2-013) and the poorly differentiated PDAC cell line PANC-1 that endogenously expressed high levels of RUVBL1. Based on western blot data, RUVBL1 expression was markedly lower in RUVBL1-RNAi cells than in control cells 72 h after transfection of the respective siRNAs (Fig. 1A). RUVBL2 expression was not changed in RUVBL1-RNAi cells, compared to control cells (data not shown). These results indicated that RUVBL1-siRNAs specifically suppressed endogenous expression of RUVBL1 in S2-013 and PANC-1 cells. RNAi-mediated suppression of RUVBL1 did not affect cell growth in an in vitro MTT assay of S2-013 and PANC-1 (data not shown). Transwell motility and Matrigel invasion assays were used to examine the effect of RUVBL1 on cell motility and invasiveness. In Transwell motility assays, motility of S2-013 and PANC-1 cells was significantly lower in RUVBL1-RNAi cells than in control cells (Fig. 1B). In two-chamber invasion assays, invasiveness of RUVBL1-RNAi cells of S2-013 and PANC-1 was significantly lower than that of control cells (Fig. 1C). These results indicated that RUVBL1 promoted the motility and invasiveness of PDAC cells.

Endogenous RUVBL1 localizes with transcription factors mainly to the nucleus, and regulates its target genes including p53 in colon cancer cells (26). We used immunocytochemistry to determine the subcellular localization of RUVBL1 in S2-013 cells. Notably, when S2-013 cells that were initially in suspension attach to an immobilized fibronectin substrate, nascent membrane protrusions (de novo formation of actin patches at the cell periphery) form, and as these protrusions mature, they promote cell motility and invasion (22). Therefore, we analyzed the subcellular distribution of RUVBL1 in PDAC cells cultured on fibronectin. Spreading of S2-013 cells on fibronectin promoted accumulation of cytoplasmic RUVBL1 in membrane protrusions (Fig. 2A). These results indicated that the function of cytoplasmic RUVBL1 may be different from that of nuclear RUVBL1 in PDAC cells. Z stack panels substantiated this result in S2-013 cells cultured on fibronectin (Fig. 2B). Additionally, an immunostaining wound-healing assay was used to analyze localization of RUVBL1 in polarized migrating S2-013 cells; results from this assay showed that RUVBL1 was recruited to the leading edges, in which peripheral actin structures were abundant, during wound healing (Fig. 1C). Additionally, an immunostaining wound-healing assay was used to analyze localization of RUVBL1 in polarized migrating S2-013 cells (Fig. 2C); results from this assay showed that cytoplasmic RUVBL1 was recruited to the leading edges of S2-013 cells during wound healing. Based on these results, we reasoned that localization of RUVBL1 in membrane protrusions may be important to cell motility and invasiveness.

To investigate the mechanism by which RUVBL1 promoted cell motility and invasiveness, immunoprecipitation (IP) experiments were performed with lysates from fibronectin-stimulated S2-013 cells; a specific anti-RUVBL1 antibody was used to detect multiprotein complexes that contained RUVBL1. Control and anti-RUVBL1 immunoprecipitates were subject to SDS-PAGE; the separated proteins were silver stained. A 40-kDa band was evident in the anti-RUVBL1 sample that was very weak in the isotype control sample (Fig. 3A). The band was excised, and LC-MS/MS was used to identify the constituent protein after in-gel trypsin digestion; the protein was actin. The peptide sequence coverage was 58% (Fig. 3B). Immunoblot analysis showed that a faint band of actin was detected in control-immunoprecipitates from fibronectin-stimulated S2-013 cells (Fig. 3C), suggesting the presence of low levels of non-specific actin in all IP samples. Strong actin band was detected in the anti-RUVBL1-immunoprecipitates (Fig. 3C), indicating that actin was enriched in RUVBL1-IP materials compared to control IgG-IPs.

Immunocytochemical signal from RUVBL1 and fluorescent signal from peripheral F-actin structures (labeled by phalloidin) were colocalized in cell protrusions of fibronectin-stimulated S2-013 cells (arrowheads in Fig. 3D). Thus, we hypothesized that the RUVBL1-actin complexes localized in cell protrusions could function in promotion of the motility and invasiveness.

To investigate whether RUVBL1 influenced the structural organization of F-actin, we assessed whether RUVBL1 had an effect on the apparent rate of actin polymerization. Actin polymerization was monitored using an in vitro pyrene-labeled G-actin polymerization assay. The kinetics of actin polymerization in the presence of 30, 100 or 300 μg/ml RUVBL1 was measured by increase in pyrene fluorescence (Fig. 4A). Half-maximal saturated polymerization values [T^1/2max (s)] calculated from Boltzmann sigmoidal curve fits are summarized in Fig. 4B. It is likely that T^1/2max values for actin reactions in the presence of 30, 100 or 300 μg/ml RUVBL1 were faster than that in vehicle control. A significant difference in synergistic polymerization rates between vehicle control and added RUVBL1 became evident at a time point of 60 min (Fig. 4C). In contrast, RUVBL1 had no discernible effect on the kinetics of in vitro pyrene-labeled F-actin depolymerisation assays that involved measurement of F-actin fluorescence (Fig. 4D). These results indicated that binding of RUVBL1 to F-actin enhanced elongation of existing actin filaments.

Immunocytochemistry and an antibody against vitamin D-binding protein [DBP (27)], a protein that specifically binds G-actin, was used to indirectly assess the presence and localization of G-actin in fibronectin-stimulated S2-013 cells; the cells had transfected with either scrambled control-siRNA or RUVBL1-siRNA. In control cells, G-actin accumulated in cell protrusions of motile cells (arrows in Fig. 5A); notably, RUVBL1 and peripheral F-actin filaments were colocalized in these cell protrusions (Fig. 3D), but RUVBL1 did not bind to G-actin in these control protrusions (Fig. 5A). In contrast, siRNA-mediated suppression of RUVBL1 inhibited the accumulation of G-actin near the cell membranes (Fig. 5B). A significant difference in rates of G-actin concentration in cell protrusion between scrambled control and knockdown of RUVBL is shown in Fig. 5C. Suppression of RUVBL1 decreased G-actin concentration in the protrusion of S2-013. These results indicated that RUVBL1 played a role in induction of G-actin concentration in cell protrusions.

We analyzed peripheral F-actin structures in membrane ruffles of S2-013 cells transfected with scrambled control or RUVBL1 siRNA; all cells were cultured on fibronectin. Knockdown of RUVBL1 inhibited the increase in peripheral F-actin structures compared to control cells (Fig. 6A). To determine whether RUVBL1 that increased F-actin structures in cell protrusions has a role in inducing membrane protrusions, immunofluorescence was carried out; all cells were treated with either scrambled control or RUVBL1 RNAi and cultured on fibronectin. siRNA-mediated suppression of RUVBL1 significantly inhibited the formation of fibronectin-mediated membrane protrusions in which peripheral actin structures were abundant, compared to scrambled control (Fig. 6B). These results indicated that RUVBL1 induced peripheral F-actin polymerization in cell protrusions that promoted formation of cell protrusions, and that these RUVBL1-mediated actin rearrangements promote the motility and invasiveness of PDAC cells.