Podoplanin promotes progression of malignant pleural mesothelioma by regulating motility and focus formation

Cell lines

The human mesothelioma cell lines MSTO-211H, H226, and H2452 were purchased from ATCC (Rockville, MD, USA). YMESO-14 cells were kindly donated by Dr. Y. Sekido (Aichi Cancer Research Center Institute, Nagoya, Japan) and EHMES-1 cells were kindly donated by Dr. H. Hamada (Hiroshima University, Hiroshima, Japan). NCI-H290 was provided by Dr. Adi F. Gazdar (University of Texas Southwestern Medical Center, Dallas, TX, USA). Cells were cultured in RPMI-1640 medium supplemented with 10% FBS (Life Technologies, Grand Island, NY, USA). All cell lines were tested and authenticated by the Japanese Cell Research Bank using short tandem repeat (STR) analysis and the GenePrint 10 System (Promega, Madison, WI, USA). Cells were regularly screened for Mycoplasma using a MycoAlert Mycoplasma Detection Kit (Lonza, Basel, Switzerland). ROCK inhibitors, Y-27632 and fasudil hydrochloride, were obtained from Wako Pure Chemical Industries (Osaka, Japan).

Western blotting

Lysates were prepared using Cell Lysis Buffer (Cell Signaling). The procedure for Western blotting was as previously described.11 The primary antibodies (Ab) used were anti- PDPN Ab (AngioBio Co.), anti-E-cadherin Ab (Cell Signaling), anti-N-cadherin Ab (Cell Signaling), anti-Vimentin Ab (Cell Signaling), anti-GAPDH Ab (Trevigen), and anti-β-actin Ab (Cell Signaling).

Cell viability assay

Cell viability was measured by the MTT dye reduction method. Tumor cells were plated onto 96-well plates at a density of 2 × 10³/100 μL per well in RPMI 1640 plus 10% FBS and cells were incubated for 24 h. Drugs were then added to each well, and incubation was continued for another 72 h. Cell growth was measured with MTT solution (2 mg/mL; Sigma, St. Louis, MO, USA), as described previously.12

Wound healing assay

Cells were plated onto 6-well plates at 500 000 cells per well in RPMI 1640 plus 10% FBS and allowed to form a confluent monolayer. A wound was introduced by running a P200 pipette tip evenly across the monolayer. After incubation for 36 and 48 h, cells were observed using a microscope.

Transwell assay

Transwell assays were performed using the modified Boyden chamber method,13 with an 8-μm pore filter separating the upper and lower transwell chambers (BD Biosciences, NJ, USA). Tumor cells (10⁴ cells/200 μL) were added to the upper chamber and incubated for 48 h. Cells that had not migrated were then removed from the upper surface of the filters with cotton swabs. Cells that had migrated to the lower surface of the filters were fixed, stained with H&E, and counted in six fields under a microscope at 200× magnification.

Transfection of the PDPN gene

Cells were seeded onto 6-well plates at a density of 1–2 × 10⁵ cells/well. Twenty-four hours later, cells were transfected with a PDPN expression vector2 using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's instructions. After treatment with neomycin (Sigma-Aldrich), cells were cultured in the presence of neomycin and clones expressing PDPN were isolated.

Small interfering RNA (siRNA) and short hairpin (sh) RNA for PDPN knockdown

shRNA was used to knock down PDPN. Lentiviruses were produced using 293T cells transfected with PCAG-HIV, CMV-VSV-G-RSV-Rev (RIKEN BioResource Center), and PDPN shRNA vectors (CS-H1-shRNA-EG; RIKEN BioResource Center).14 Transfection was performed using Lipofectamine 2000 reagent (Invitrogen, CA, USA) according to the manufacturer's instructions. The vector-containing medium was filtered through a 0.45-μm filter, and 8 μg/mL of Polybrene (SIGMA) was added for transduction of target tumor cells. siRNA was also used to knock down PDPN. Tumor cells were transfected with siRNAs against PDPN (Stealth siRNAs: HSS116395, HSS116397, HSS173792) or Stealth RNAi-negative control low GC Duplex #3 (Invitrogen) introduced into cells using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's instructions.

RhoA-GTP binding assay

Direct activation of RhoA was measured using a G-LISA assay (Cytoskeleton Inc., CO, USA) according to the manufacturer's instructions. Briefly, the RhoA G-LISA kit used 96-well plates coated with the Rho-binding domain of the RhoA effector rhotekin. Rho-GDP was removed during washing steps and Rho-GTP was detected using a RhoA-specific antibody and chemiluminescence.

Orthotopic implantation

Tumor cells (1 × 10⁶/100 μL) were injected into the thoracic cavity of SCID mice as reported previously.15 After the indicated periods, the mice were euthanized and tumor development was evaluated.

Immunohistochemistry

Formalin-fixed paraffin-embedded tumor sections were subjected to antigen retrieval and endogenous peroxidase blocking, and sections were incubated with primary antibody (Ab), anti-Ki-67 Ab (Dako), or anti-Yes-associated protein 1 (YAP1) Ab (Cell Signaling) at 4°C overnight. After incubating overnight, slides were rinsed and incubated with a peroxidase-labeled polymer. The tissue sections were then rinsed and stained with 3,3′-diaminobenzidine (DAB) substrate-chromogen and then counterstained with Hematoxylin Gill I (EMD Millipore) and bluing reagent (EMD Millipore).

Focus formation assay

Tumor cells were plated onto 6-well plates at 500 000 cells per well in RPMI 1640 plus 10% FBS and allowed to form a confluent monolayer. The confluent cell cultures were incubated for 14 days. The cultured cells were then stained with crystal violet and the foci were identified under a microscope.

Statistical analysis

The statistical significance of difference between the in vitro and in vivo data were analyzed by one-way anova using GraphPad Prism Ver. 4.01 (GraphPad Software, Inc., San Diego, CA, USA). Survival was analyzed by the Kaplan–Meier method. Differences between treatment and control groups were compared with the log-rank test. Differences at P < 0.05 were deemed significant.

PDPN is highly expressed in pleural mesothelioma and promotes motility via RhoA/ROCK pathway activation

We first subjected tumors from 52 Japanese patients with MPM to immunostaining with the D2-40 antibody to determine if they expressed PDPN. Ninety percent of the tumors from Japanese patients with MPM tested positive for PDPN (Table S1, Fig. S1), so tumors from Japanese patients with MPM were found to express PDPN at high levels.

We then examined expression of PDPN in human MPM cell lines. High levels of expression were noted in three (H226, H2452, and EHMES-1) of six human MPM cell lines (Fig. 2a). In order to determine the role of PDPN in MPM cell lines, PDPN was knocked down with siRNA in 2 cell lines expressing high levels of PDPN (H226 and H2452). In H226 (Fig. 1b–d) and H2452 (Fig. S2) cells, knocking down PDPN did not alter cell viability but it did decrease cell motility. When PDPN was stably knocked down with shRNA in H226, motility was inhibited (Fig. 1e) but cell viability was not affected (Fig. S3c). The effects of shRNA were restored by transfection of shRNA-resistant PDPN mutants (Fig. S3a,b), so motility was definitely inhibited by knocking down PDPN.

In contrast, transfection of PDPN into MSTO-211H cells expressing low levels of PDPN (Fig. 2a) did not alter cell viability (data not shown) but it did enhance motility. Both a wound healing assay and a migration assay using a transwell system revealed enhanced motility as a result of overexpression of PDPN (Fig. 2b–d). These findings revealed that PDPN regulates the motility of MPM cells.

Podoplanin is known to bind to the ERM protein family and activate Rho.5 Thus, we examined whether or not PDPN promotes the motility of MPM cells via the Rho/ROCK/Rac pathway. Transfection of PDPN into MSTO-211H cells resulted in increased RhoA-GTP binding (Fig. 3a). Conversely, knocking down PDPN with specific siRNA in H226 cells resulted in decreased RhoA-GTP binding (Fig. 3b). We also explored the effects of compounds that inhibit ROCK downstream of Rho. Y-27632 is widely used as a ROCK inhibitor, and fasudil hydrochloride has been clinically approved for treatment of delayed cerebral vasospasms following a subarachnoid hemorrhage since it inhibits ROCK.16 Neither Y-27632 nor fasudil hydrochloride altered the viability of MSTO-211H/PDPN cells (Fig 3c, Fig. S4a), but the two ROCK inhibitors did inhibit motility in a dose-dependent manner (Fig. 3d, Fig. S4b). These findings indicate that PDPN activates the RhoA/ROCK pathway, thus promoting the motility of MPM cells.

PDPN promotes the progression of mesothelioma in the orthotopic implantation model

The effects of PDPN on tumor progression were examined in a model of orthotopic intrathoracic implantation in SCID mice. In H226 cells, tumor progression (the intrathoracic tumor burden) was inhibited by the knockdown of PDPN with shRNA (Fig. 4a). In contrast, tumor progression was promoted by transfection of PDPN into MSTO-211H cells, and mice had a significantly reduced survival time (Fig. 4b,c). Similarly, tumors produced by MSTO-211H cells transfected with PDPN had an increased number of Ki-67-positive proliferating cells. In contrast, tumors produced by H226 cells when PDPN was knocked down with shRNA had a reduced number of Ki-67-positive proliferating cells (Fig. 4d, Fig. S5). In another cell line expressing low levels of PDPN (H290), transfection of the PDPN gene resulted in enhanced tumor progression in a model of orthotopic implantation and an increased number of Ki-67-positive proliferating cells (Fig. S6). However, PDPN expression did not affect the engraftment rate or the number of tumors produced by MPM cells. These findings revealed that PDPN sustains the growth of MPM cells in vivo and that it promotes tumor progression in the thoracic cavity.

PDPN promotes focus formation in vitro and induces YAP1 activation associated with a low level of E-cadherin expression in vivo

Promotion of MPM cell motility by PDPN may not be the only factor responsible for tumor enlargement in vivo. Therefore, we focused on contact inhibition as another mechanism. Loss of contact inhibition is a strong indicator of cell transformation17 and facilitates tumor progression. We performed a focus formation assay to examine the effect of PDPN on contact inhibition in MPM cells. PDPN blocked contact inhibition and promoted the formation of foci in MSTO-211H (Fig. 5a) and H290 (Fig. S7) cells. In contrast, knockdown of PDPN enhanced contact inhibition in H226 cells (Fig. 5b) resulting in a remarkable decrease in the number of foci.

YAP1 is reported to block contact inhibition and promote tumor progression.18 In order to determine the mechanisms by which PDPN blocks contact inhibition, YAP1 expression was examined in tumors obtained from an orthotopic implantation model. YAP1 is a transcription factor that facilitates the transcription of various genes upon nuclear translocation.18 In tumors produced by H226 cells expressing high levels of PDPN, YAP1 was detected in the nuclei of 50% or more tumor cells, indicating that YAP1 was activated. In tumors produced by H226 cells upon PDPN knockdown with shRNA, YAP1 was not detected in the nuclei of most tumor cells, indicating that YAP1 was inactive (Fig. 6a). In tumors produced by MSTO-211H or H290 cells that express low levels of PDPN, YAP1 was not detected in the nuclei of most tumor cells. In tumors produced by MSTO-211H or H290 cells transfected with PDPN, YAP1 was detected in the nuclei of 60% or more tumor cells (Fig. 6b, Fig. S8). Moreover, PDPN knockdown in H226 cells resulted in increased E-cadherin expression, whereas transfection of PDPN into MSTO-211H cells resulted in decreased E-cadherin expression (Fig. 6c,d). These findings suggest that PDPN blocks contact inhibition via decreased expression of E-cadherin and YAP1 activation.