Myopodin-Mediated Suppression of Prostate Cancer Cell Migration Involves Interaction with Zyxin

Introduction

Prostate cancer is one of the leading causes of death for men in the United States ( 1). The implementation of the serum prostate-specific antigen screening program has resulted in a substantial increase in the clinical detection rate of prostate cancer as otherwise unnoticed micro-adenocarcinomas of the prostate gland are now being found ( 2). Although many of these prostate cancers are very slow growing and are likely not to be clinically relevant for many patients, nearly 30,000 patients die of prostate cancer annually ( 1). Despite the tremendous advances in our knowledge about prostate cancer in recent years, the molecular mechanism by which a relatively indolent disease is converted into a highly aggressive and lethal one remains unclear.

Our previous studies suggested that myopodin, a newly characterized tumor suppressor gene, was deleted in a subset of prostate cancer cells ( 3). Further research has indicated that deletion of myopodin, either partial or complete, is associated with high rates of metastasis and prostate cancer clinical relapse. Deletion hotspot domains have been identified in myopodin. The absence of a myopodin deletion, on the other hand, is associated with low rates of invasion and clinical relapse, and this association seems to be independent of the Gleason grade of the tumor. Subsequent analysis suggested that suppression of myopodin expression exacerbates prostate cancer tumorigenesis and invasiveness ( 4). In the experiments reported herein, we used a yeast two-hybrid analysis system and found that myopodin interacts with zyxin both in vitro and in vivo. This interaction leads to slower migration of prostate cancer cells and reduced invasiveness.

Materials and Methods

Preparation of pACT2 library. To screen 1 × 10⁶ clones in yeast, the pre-made human prostate cDNA plasmid library pACT2 from MATCHMAKER was amplified according to the following protocol: titrations (10⁻³, 2 × 10⁻⁴, and 1 × 10⁻⁴) were done on three 100-mm ampicillin (100 μg/mL) agar plates before the amplification of the library. The plates were incubated at 30°C for 36 hours. The number of colonies in the three plates was 4.9 × 10⁸ cfu/mL. To amplify 1 × 10⁶ colonies, 10 μL of library mixed with 50 mL of LB were plated onto 300 plates and incubated at 30°C for 36 hours. The yeast cells were scraped and pooled in one 1-liter flask. A total volume of 1250 mL of cell solution was shaken at 200 rpm for 3 hours. DNA from an aliquot of library was extracted with a QIAGEN Plasmid Mega Prep kit (Qiagen, Valencia, CA) and stored at −20°C.

Constructions of fusion proteins. A mutagenic primer set (TGTGGCCTATAATCATATGCACTCGCCGTCTTACCCACTG and TTCATTTCAAGCAAAGTCGACACTCAGCTTCAGCTACAAG) was designed to create two restriction sites (NdeI and SalI) in the COOH terminus of myopodin, so that the PCR product could be ligated into a pGBK7T vector ( Fig. 1A ). An extended long PCR polymerase (Invitrogen, Carlsbad, CA) was used, and the PCR conditions were as follows: 94°C for 1 minute followed by 35 cycles of 94°C for 30 seconds, 68°C for 3 minutes, and a final 3-minute extension step at 68°C. The PCR product was restricted with NdeI and SalI, gel purified, and ligated into a similarly restricted pGBK7T vector. The fusion protein contained 187 amino acids from the myopodin COOH terminus. The construct was transformed into One Shot competent cells (Invitrogen). Plasmid DNA was extracted from selected transformed cells and digested with NdeI and SalI to detect the presence of the insert. The coding frame was confirmed by automated sequencing.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

COOH terminus of myopodin binds with zyxin. A, diagram of construction strategy for pBD-myoC. A 187-amino-acid region of the myopodin COOH terminus was ligated in frame with DNA-binding domain of pGBK7T. B, growth of yeast harboring pBD-MyoC and pAD-zyxin in high-stringency nutrient selection agar (left) and positive α-galactosidase activity (right). C, coimmunoprecipitation of myopodin with antibodies against AD-zyxin in yeast (anti-hemagglutinin). Protein extracts from a yeast clone harboring pAD-zyxin and pBD-myoC were immunoprecipitated with antibodies as indicated (PreImS-preimmune rabbit serum, hemagglutinin-Ab anti-hemagglutinin antibodies, lysate-crude lysate without immunoprecipitation) and immunoblotted with anti-myopodin antibodies (MyoAb). D, coimmunoprecipitation of myopodin and zyxin in PC-3 cells. PC3 cell protein extracts were immunoprecipitated with anti-myopodin or control antibodies and blotted with zyxin antiserum. E, binding of zyxin with GST-myoC protein. Top, purified GST-myoC or other indicated control proteins were incubated with protein extracts from PC3 cells, washed, eluted, and immunoblotted with anti-zyxin antibodies. Bottom, purified GST-myoC or other indicated control proteins were incubated with protein from in vitro translation of T7 promoter driven AD-zyxin (ZIVT) or pGBKT7 (CIVT), washed, eluted, and immunoblotted with anti-zyxin antibodies. F, colocalization of myopodin and zyxin in prostate epithelial cells. Left, images of double immunostaining of myopodin (Rhodamine, top), zyxin (FITC, middle), and merge (bottom) in PC3 cells transformed with myopodin expression vector. Right, images of double immunofluorescence staining of myopodin (rhodamine, top), zyxin (FITC, middle), and merge (bottom) in normal prostate epithelial cells.

For construction of the glutathione S-transferase (GST)-MyoC fusion protein, a mutagenic primer set (TGTGGCCTATAATCGGATCCACTCGCCGTCTTACCCACT/TTCATTTCAAGCAAAGTCGACACTCAGCTTCAGCTACAAG) was designed to create a BamHI- and SalI-restricted site within the myopodin coding region that encodes a 170 (510 bp) amino acid region of the 3′ terminal of myopodin. A PCR was done using these primers under the following conditions: 94°C for 1 minute followed by 35 cycles of 94 °C for 30 seconds, 68°C for 3 minutes, and a final 3-min extension step at 68°C. The PCR product was subsequently gel purified and ligated into a pCR2.1 TA cloning vector. The DNA was transformed into Escherichia coli. DNA from the selected transformants was restricted with BamHI and SalI and ligated into a similarly restricted pGEX-5x-3 vector in frame. A series of deletions including 5′ or 3′ deletions of pGST-myoC were done using the primer sets listed in Table 1 (also see Fig. 2 for amino acid sequences in the constructs). The procedures for generating these mutants were similar to those described for pGST-myoC. The pGST-myoC and its mutants were transformed into E. coli BL21 cells for recombinant protein production.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Identification of myopodin sequence required for interaction with zyxin. A, mutants of NH₂ terminus deletion of GST-myoC and their interaction with zyxin. Top, diagram of GST-myoC deletion mutants. Bottom, immunoblot of zyxin that binds to GST-myoC mutants and Coomassie staining of GST-myoC mutant proteins. B, mutants of COOH terminus deletion of GST-myoC and their interaction with zyxin. Top, diagram of GST-myoC deletion mutants. Bottom, immunoblot of zyxin that binds to GST-myoC mutants and Coomassie staining of GST-myoC mutant proteins. C, peptide interference of myopodin-zyxin interaction. PC3 cell extracts incubated with GST-myoC in the presence of a peptide corresponding to the putative zyxin interaction site (ZI) or a peptide corresponding to an unrelated region (Im). Top, immunoblot of eluate. Bottom, immunoblot on unbound flow-through.

The construction of pCMV-myoC has been described previously ( 4). The following two sets of primers were designed for constructing a myopodin deletion mutant that lacks (amino acids 606-624) zyxin-interacting sequence (pCMV-Midel; Table 1). Two separate PCRs were done under the following conditions: 94°C for 1 minute followed by 35 cycles at 94°C for 30 seconds, 68°C for 3 minutes, and a final 3-minute extension step at 68°C. The PCR products were gel purified, mixed, and used as template in a PCR using the following primers: AATTAACCCTCACTAAAGGG/GTAATACGACTCACTATAGGGC. The PCR products were subsequently digested with BamHI and ApaI and ligated into the similarly digested pCMV-script vector. Selected clones were sequenced to confirm the mutation. For small interfering RNA (siRNA) analysis, 125 pmol siRNA specific for zyxin (CUGGACAUGGAGUUGGACCUGAGGCUU/GCCUCAGGUCCAACUCCAUGUCCAG) or nonspecific control (UAAUGUAUUGGAACGCAUAUU/UAUGCGUUCCAAUACAUUA) were transfected into Wild 2 cells using LipofectAMINE 2000 transfection kit (Invitrogen). Immunoblots, cell motility, and Matrigel transmigration assays were done 24 hours after transfection.

Yeast transformation and library screening. One single colony of the yeast strain AH109 (Clontech, Mountain View, CA) freshly formed after 3 days grown in an YPD plate at 30°C was selected and inoculated into 50 mL of YPDA and incubated at 30°C with shaking at 200 rpm for 16 hours. An additional 300 mL of YPD was added to the overnight culture, and the solution was allowed to shake for an additional 3 hours to enable the absorbance at 550 nm to reach 0.5. The cells were centrifuged at 2,000 × g for 5 minutes and resuspended in 60 mL of sterile water. The cells were recentrifuged to form pellets again, resuspended in 3 mL of 1.1× TE/LiAc, repelleted, and finally resuspended in 400 μL of 1× TE/LiAc. Fifty microliters of freshly prepared competent AH109 cells were mixed with plasmid DNA (0.25-0.50 μg) plus 5 μg Herring testes carrier DNA in 0.5 mL of polyethylene glycol/LiAc and incubated at 30°C for 30 minutes. Following this initial incubation with plasmid DNA, the cell solution was combined with 20 μL of DMSO and subjected to a 15-minute incubation at 42°C. The cells were pelleted, resuspended in 1 mL YPD medium, and shaken at 30°C for 40 minutes. The transformed cells were then pelleted, resuspended in 0.5 mL 0.9% NaCl, and plated to the appropriate SD agar plate. The transformants were first plated on low and medium stringency plates of SD-Leu/SD-Trp and SD-Leu/SD-Trp/SD-His, respectively. The grown colonies were subjected to the colony-lift filter α-galactosidase assay and allowed to grow further in the high-stringency plate (SD-Ade/SD-His/SD-Leu/Trp).

Validation of protein interactions in AH109. Plasmid DNA samples from positive clones were isolated from yeast, transformed into E. coli, and selected with ampicillin (100 μg/mL) to obtain genes interacting with the bait-domain fusion protein. The purified AD/library plasmid DNA was then cotransformed with MyoC-BD/bait plasmid into AH109 yeast cells and grown in a SD-Ade/SD-His/SD-Leu/Trp high-stringency medium. Colony-lift filter α-galactosidase activity was assayed on those cells grown in this medium. The positive clones were sequenced.

Immunoprecipitation. Protein extracts of PC3 cells transformed with pCMV-myopodin, or the in vitro translation product, was incubated with myopodin antibody (myoC; ref. 4) for 16 hours, then with protein G-Sepharose for 3 hours. The complex was washed five times with radioimmunoprecipitation assay buffer, and the bound proteins were eluted with SDS-PAGE sample buffer.

In vitro transcription and translation. DNA from a positive clone of the pACT2_1.1 prostate library were PCR amplified by DNA polymerase and RedMix plus (GeneChoice) with one of the primers incorporated in the T7 promoter sequence at its 5′ end (5′-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGGTACCCATACGATGTTCCAGATTACGCT). The gel-purified PCR product was used as the template in the in vitro transcription and translation reactions using TNT quick coupled transcription/translation systems (Promega, Madison, WI).

GST fusion pulldown. The cells were grown in 100 mL of Luria-Bertani (ampicillin 100 μg/mL) overnight and induced by isopropyl-l-thio-B-d-galactopyranoside (IPTG; final concentration of 1 mmol/L) for 3 hours. The cells were then pelleted, resuspended in 1× PBS, and sonicated for 2 minutes. The proteins were solubilized in 1% Triton X-100. The supernatant was collected after centrifugation at 15,000 × g for 5 minutes. The GST and GST-myoC fusion protein were purified through a glutathione-Sepharose 4B column (Amersham Biosciences, Piscataway, NJ). The PC3 cell protein extract, or the in vitro translation products, were preincubated with the column for 15 minutes at 4°C. The flow-through was collected after spinning at 3,000 × g for 1 minute. The flow-through cell lysates were then incubated with GST fusion protein packed glutathione-Sepharose 4B at 4°C for 2 hours. The column was spun at 3,000 × g at room temperature for 1 minute and further washed twice with PBS. The protein was eluted from the column with 40 μL of SDS-PAGE gel sample loading dye. SDS-PAGE protein gel analysis and Western blot analysis were subsequently conducted. For zyxin interaction peptide interference assay, a 19-amino- acid peptide (AMKQALPPRPVNAASPTNV) corresponding to myopodin codons 606 to 624 or a control (KMGKKKGKKPLNALDVMKHQ) was preincubated with GST-MyoC column for 1 hour before application cell extract.

Immunofluorescence staining. PC3 cells that were transfected with pCMV-myopodin were cultured on chamber slides for 24 hours. The slides were washed with PBS for three times. The cells were fixed with 4% paraformaldehyde for 1 hour at room temperature. After washing the slides with PBS twice, the cells were blocked with 10% donkey serum with 0.4% Triton X-100. The cells were then incubated with rabbit antisera against myopodin (myoC) and goat antisera against zyxin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at room temperature for 1 hour. The slides were washed with PBS twice. Secondary antibodies from Donkey against goat (fluorescence conjugated) and against rabbit (rhodamine conjugated) were added and incubated at room temperature for 1 hour. The slides were then washed with PBS twice before addition of Hoescht 33342. After additional washes with PBS, slides were mounted with immunomounting buffer. Immunofluorescence staining was examined under Olympus inverted system fluorescence microscope IX. Immunofluorescence staining on normal prostate was done on fresh frozen section of a prostate gland obtained from a 35-year-old organ donor.

Matrigel transmigration analysis. Cells from each clone were suspended in F12K medium were added to the upper chamber at a concentration of 1 × 10⁵ cells per insert, whereas the lower chamber contained F12K medium with 10% fetal bovine serum. After 24 hours of culture, the upper surface of the inserts was wiped with cotton swabs, and the inserts were stained with H&E. Each experiment was done twice with each sample in triplicate. The cells that migrated through the Matrigel and the filter pores to the lower surface were counted under a light microscope.

Wound-healing assays and random migration analysis. For the wound-healing assay, Du145 and PC-3 cells were seeded into six-well culture plates. A plastic pipette tip was drawn across the center of the well to produce a clean 1-mm-wide crevice after the cells had reached confluence. After culturing for 24 hours at 37°C in F12K (PC-3) or MEM (Du145) medium containing 10% fetal bovine serum (FBS), images were taken, and recovered areas were measured. For random migration analysis, cells were plated to 5% confluence, and medium without FBS was applied. Several areas were selected for migration analysis. Images of these areas were collected every 6 hours, with the final image being collected 48 hours after FBS application. The distances traveled by cells with and without myopodin (or its mutants) transformation during this period were tabulated, averaged, and compared.

Results

Demonstration of myopodin-zyxin binding activity. Deletion analysis indicated that the tumor suppressor and metastasis suppressor activity of myopodin is located within the COOH terminus region. To identify the target protein that mediates myopodin-induced suppression of cancer metastasis, we did a yeast two-hybrid analysis using COOH terminus of myopodin fused with DNA-binding domain to screen a human prostate pACT2-cDNA library. Twelve clones were isolated after three rounds of nutrient selection. DNA samples from these clones were subsequently extracted and transformed into E. coli. After eliminating several recombinant plasmid DNA duplicates, seven of these clones remained and were sequenced using an automated sequencer.

One of the sequenced clones was found to contain zyxin cDNA. As shown in Fig. 1A, the DNA from this clone was cotransfected with pBD-MyoC. The cotransfectants were able to grow in the high-stringency nutrient selective agar lacking leucine, tryptophan, histidine, and adenine ( Fig. 1B). These colonies were positive for α-galactosidase activity, confirming that there was an interaction of two human proteins that brought the DNA-binding domain and the activation domain of GAL4 together.

To validate the observed protein interaction, immunoprecipitation of AD-zyxin using antibodies against the tag sequence (anti-hemagglutinin) was done. This was followed by immunoblotting with antibodies against the myopodin COOH terminus (anti-myoC). As shown in Fig. 1C, a clear band of BD-myoC was identified in this coimmunoprecipitation. Our previous studies indicated that the protein expression of myopodin was significantly reduced in PC3 cells due to hemizygous deletion. To investigate whether myopodin interacts with zyxin in human cells, coimmunoprecipitations of zyxin and myopodin were done on PC3 cells transformed with myopodin expression vector pCMV-myopodin using anti-myopodin antibodies. The results confirmed that a complex association between myopodin and zyxin occurs in PC-3 cells ( Fig. 1D). We subsequently ligated the COOH terminus of myopodin with GST and expressed GST-myoC in bacteria. We tested whether myoC binds with zyxin in cell-free system. The results were consistent with our hypothesis that zyxin and myopodin bind directly ( Fig. 1E), independent of any other accessory proteins.

Interaction between myopodin and zyxin requires colocalization of these two proteins in the same cell compartment. To investigate whether myopodin and zyxin colocalize in prostate epithelial cells, we did double immunofluorescence staining with antibodies against myopodin (with rhodamine-labeled secondary antibodies) and antibodies against zyxin (with FITC-labeled secondary antibodies) on PC3 cells transformed with myopodin. The results suggested a dominant cytoplasmic localization of both zyxin and myopodin in PC3 cells. Overlapping of zyxin and myopodin localizations was extensive ( Fig. 1F). We also did a double immunostaining on normal prostate epithelial cells. Similar zyxin/myopodin colocalization was seen, although the subcellular distribution of either protein seemed to include both nucleus and cytoplasm.

To identify the motifs required for myopodin interaction with zyxin, a series of myoC NH₂ terminus ( Fig. 2A) and COOH terminus deletion mutants were constructed and ligated into GST in frame to form fusion GST-myoC deletion mutant proteins. The IPTG-induced GST-myoC mutant proteins were purified using glutathione column and incubated with protein extracts from PC3 cells. As shown in Fig. 2A and B, binding of zyxin and myoC was found to occur in the myoC wild-type construct and certain mutants. By overlapping the mutants with binding activity, a sequence that encompass 19 amino acids from codon 606 to 624 were identified as the critical region for zyxin-binding activity. Several GST-myoC/zyxin binding assays were done in the presence or absence of a chemically synthesized peptide with a sequence that corresponded to this 19-amino-acid sequence. As shown in Fig. 2C, inclusion of the peptide containing the putative zyxin-interacting sequence severely impeded the binding between zyxin and GST-myoC. In contrast, a peptide corresponding to another region of myopodin did not inhibit such activity.

Cell motility. Zyxin is known for its regulatory role in cell motility ( 5– 8). We examined cell motility by performing wound-healing assays and random migration analysis on PC3 cells that had been transformed with wild-type or mutant myopodin with deletion of the zyxin-interacting sequence ( Fig. 3A ). As shown in Fig. 3B, expression of myopodin retarded migration in the wound-healing assay by an average of 40% (P < 0.001) and inhibited random migration by 74% (P < 0.001). In contrast, the migration retarding effect was virtually abolished in wound-healing assays (97% of controls, P = 0.63) and in random migration assays (82% of controls, P = 0.43) when mutant myopodin that lacked the zyxin interaction site was expressed. Consistent with these findings, siRNA specific for zyxin virtually abolished the myopodin-mediated cell motility inhibition effect ( Fig. 3B and C).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Zyxin interaction is required for myopodin-mediated motility retardation. A, expression of wild-type and mutant myopodin (Midel). Top, diagram of wild-type myopodin and deletion mutant Midel. Black strips, sequences homologous to synaptopodin. Bottom left, immunoblot of myopodin on protein extracts from pCMV-myopodin (Wild 1 and 2), pCMV-Midel (Midel 3 and 1), or pCMVscript (P2 and P3) vector-transformed PC3 cells. β-Actin is the control. Bottom right, immunoblot of zyxin on protein extracts from Wild 2 cell lines treated with scramble or zyxin-specific siRNA. β-Actin is the control. B, wound-healing assay of wild-type PC3 cells transformed with pCMVscript vector (P2 and P3), with pCMV-myopodin (Wild 1 and 2), or myopodin deletion mutant (Midel 3 and Midel 1). Wild 2 was treated with either scramble siRNA (control) or siRNA specific for zyxin. C, random migration assay of wild-type PC3 cells transformed with pCMVscript vector (P2 and P3), or with pCMV-myopodin (Wild 1 and 2), or myopodin deletion mutant (Midel 3 and 1). Wild 2 was treated with either scramble siRNA (control) or siRNA specific for zyxin. D, Matrigel transmigration analysis of wild-type PC3 cells transformed with pCMVscript vector (P2 and P3), with pCMV-myopodin (Wild 1 and 2), or myopodin deletion mutant (Midel 3 and 1). Wild 2 was treated with either scramble siRNA (control) or siRNA specific for zyxin. Columns, average; bars, SE.

Invasiveness. To investigate whether zyxin-myopodin interaction is related to prostate cancer invasiveness, we did Matrigel transmigration analysis on these myopodin-transformed cells. The invasiveness of cells is expressed as a ratio (invasion index) of number of invasion cells (penetrated through Matrigel) over the number of migration cells (traveled through membrane). As shown in Fig. 3D, expression of wild-type myopodin resulted in a 4-fold suppression of invasiveness, whereas expression of the myopodin mutants lacking the zyxin interaction site had a negligible effect on the migration index and invasiveness. Knock down of zyxin using siRNA specific for this gene resulted in partial reversal of myopodin-mediated inhibition of invasiveness ( Fig. 3A and D).

Discussion

Myopodin was originally identified as a gene located in the 4q25 common minimal deletion region of aggressive prostate cancer samples ( 3). It shares limited homology with the synaptopodin protein that is expressed in neurons and podocytes. Expression of myopodin has been shown to suppress tumor growth and invasiveness in several prostate cancer cell lines in vitro ( 4). In mice, PC3-xenografted tumors transformed with wild-type myopodin were found to have a smaller tumor volume, a lower metastasis rate, and a lower mortality. Furthermore, loss of myopodin expression was found to be associated with development of aggressive urothelial carcinomas ( 9).

It is of great interest to elucidate the mechanism by which myopodin mediates metastasis suppression. Employing the yeast two-hybrid system in the present study, we identified and characterized the interaction between myopodin and zyxin. We found that deletion of the sequence required for myopodin-zyxin interaction virtually abolished myopodin-mediated suppression of cell motility and invasiveness. These findings are consistent with the hypothesis that interaction of myopodin and zyxin induces retardation of migration and cell motility. Furthermore, because migration and cell motility are the basis of metastasis and invasion, our observations suggest that zyxin is a component of myopodin-mediated suppression of prostate cancer metastasis. This hypothesis is further supported by a recent report that zyxin exhibited tumor suppressor activity in Ewing sarcoma ( 8).

The physiologic implications of myopodin-zyxin interaction are not entirely clear. Zyxin is known for its role in regulating cell motility and reorganization of cytoskeletal elements. Consistent with previous reports indicating that myopodin can suppress metastasis in vitro and in vivo, the current study showed that myopodin inhibits cell motility and migration. The interaction of myopodin and zyxin is likely essential for maintaining mature epithelial cell association with the prostate gland, presumably by cell immobilization. Because myopodin is expressed in both the basal cells and the acinar cells of the prostate gland ( 4), it is unlikely that myopodin is a differentiation-related protein. Based on its growth suppression and immobilization activity, we speculate that myopodin is a balancing protein that normally prevents overgrowth and random migration of cells from the prostate gland. Thus, myopodin may have very different roles in prostate gland than in skeletal muscle, where it is directly involved in muscle differentiation ( 10).

Zyxin involvement in the myopodin functional pathway may have important clinical implications. First, it would be of interest to know whether zyxin itself is a component in cancer metastasis. Future research examining the level of zyxin expression in metastasized cancer and analysis of structural, genomic, or epigenomic alterations of zyxin in cancer cells may provide key information about the role of zyxin in prostate cancer development and progression. Additionally, because deletion of myopodin or complete inactivation of myopodin expression has been associated with a high rate of prostate cancer metastasis, and because zyxin seems to mediate myopodin-induced inhibition of cell motility and migration and thus suppression of metastasis, it may be possible to devise a therapeutic intervention based upon the zyxin-myopodin interaction to treat or even prevent prostate cancer metastasis.