- 1-PTHrP promotes malignancy of human oral cancer cell downstream of the EGFR signaling

Title PTHrP promotes malignancy of human oral cancer cell downstream of the EGFR signaling Author(s) Yamada, Tamaki; Tsuda, Masumi; Ohba, Yusuke; Kawaguchi, Hideaki; Totsuka, Yasunori; Shindoh, Masanobu Citation Biochemical and Biophysical Research Communications, 368(3): 575-581 Issue Date 2008-04-11 Doc URL http://hdl.handle.net/2115/54567 Type article (author version) File Information Biochem Biophys Res Commun_368(3)_575-581.pdf

Control siRNA (siControl, QIAGEN) was used as a control. After 48h or 72h, the cells were subjected to analyses after the determination of expression levels of PTHrP mRNA by RT-PCR.

Measurement of cell proliferation, motility, and invasiveness.
To measure the cell proliferation, 5 x 10 4 cells were plated onto 60 mm plates, and counted every other day for 8 days using a hematocytometer. To analyze the cell motility, confluent HSC-2 cells were wounded by scraping with a P200 pipette tip. Cell movements were observed by phase-contrast microscopy. The invasion assay was performed using 24-well BD BioCoat Matrigel Invasion Chambers (BD-Discovery Labware, Bedford, MA, USA). In the upper chamber, 2 x 10 4 HSC-2 cells were seeded in complete DMEM medium. After 22 h, the non-invading cells on the upper surface of the filters were removed by wiping with a cotton swab. Infiltrated cells were fixed with 3.7% formaldehyde in PBS, stained with 5% crystal violet, and quantified. In some experiments, the similar experiments were performed in the presence or absence of AG1478.

Human oral cancer cells express PTHrP
PTHrP has been detected in many tumors, particularly breast and prostate [2]. In this study, we first examined the levels of PTHrP by RT-PCR in five independent human OSCC cell lines: SAS, Ca9-22, HSC-2, HSC-3, and HSC-4. PTHrP mRNA was detected in all cell lines except SAS cells, but not in normal human gingival fibroblasts (HGF-1), and was highest in HSC-2 and HSC-4 cells (Fig. 1A). PTHrP protein levels were examined by immunoblotting ( Fig.   1B) and immunofluorescence (data not shown), and in parallel with mRNA levels. PTHrP expression in OSCC was substantially higher than in other types of highly bone-metastatic tumor cells including prostate cancers (PC-3 and LNCap), osteosarcoma (Saos-2), and synovial sarcoma (SYO-1) (Fig. 1C). Of note, since these cell lines have been reported to express abundant PTHrP, the expression level in OSCC appeared exceedingly high, indicating the potential role for PTHrP in the development of OSCC.

EGF induces PTHrP expression via ERK1/2 and p38 MAPK in OSCC
OSCC arising from the EGF-enriched oral environment express abundant EGFR [6], suggesting that EGF-EGFR signaling is associated with the malignant biology of this cancer.
We hypothesized that activation of EGFR signaling was correlated to enhanced PTHrP expression and subsequent malignant conversions in OSCC. The summary of the properties of EGFR in the OSCC cell lines used here was as follows: OSCCs expressed abundant EGFR ( Supplementary Fig. 1A); EGFR was predominantly localized to the plasma membrane in the quiescent state ( Supplementary Fig. 1B); The phosphorylation on Tyr1173, the direct binding site for SHC, SHP1, and PLC-, was readily detectable in OSCCs even without EGF stimulation, but not in A431 epidermoid carcinoma cells ( Supplementary Fig. 1C); none of the cell lines possessed any mutations in the catalytic domain and C-terminal docking sites (data not shown); EGF stimulation significantly increased the EGFR phosphorylation ( Supplementary Fig. 1C).
We investigated the role of EGFR activation in PTHrP expression, and found that EGF stimulation induced PTHrP expression in HSC-2, HSC-3, HSC-4, and Ca9-22 cells by RT-PCR and Quantitative real-time RT-PCR ( Fig. 2A and Supplementary Fig. 2A-B). PTHrP mRNA was approximately 1.7-fold higher following EGF stimulation of HSC-2 cells ( Fig. 2A). The effect of EGF on PTHrP expression seems to be limited probably due to high basal levels of Next, we investigated the downstream pathway contributed to the expression of PTHrP.
Immunoblotting analyses using phosphorylation site-specific antibodies revealed EGF-dependent phosphorylation of ERK1/2, p38 MAPK, and SAPK/JNK in HSC-2 cells (Fig. 2C). Among these MAPKs, ERK and p38 MAPK appeared to mediate EGF-dependent PTHrP expression, since the ERK inhibitor U0126 and p38 MAPK inhibitor SB203580, but not the JNK inhibitor SP600125, reduced EGF-induced PTHrP expression (Fig. 4D). These data indicate that EGFR and the ERK and p38 MAPK cascades are primary regulators of PTHrP expression in OSCC.
To further clarify the relationship between the EGFR-MAPK signaling and PTHrP expression, we examined the effects of AG1478 and U0126 on EGF-stimulated MAPK cascades and subsequent PTHrP expression. These reagents suppressed target molecules in a dose dependent manner as expected ( Supplementary Fig. 3A-C). Interestingly, U0126 treatment in combination with AG1478 produced an additive effect on the inhibition of ERK1/2 phosphorylation ( Supplementary Fig. 4D). Concomitantly with this ERK1/2 suppression, combined treatment of HSC-2 cells with AG1478 and U0126 most effectively suppress the PTHrP expression (Fig. 2E). Taken together, the EGFR-MAPK cascades are profoundly involved in the PTHrP expression in OSCC.

PTHrP regulates malignant phenotypes in OSCC
To examine whether PTHrP accelerates the development of malignant phenotypes of OSCC, we thus utilized RNAi to knockdown endogenous PTHrP and investigated its effects on cell proliferation, migration, and invasion. PTHrP-targeted siRNA (siPTHrP) transfection significantly decreased PTHrP gene expression, by 70%, 80%, and 60% in HSC-2, HSC-4, and HSC-3 cells, respectively, when compared with a scrambled control siRNA (siControl), whereas To gain further insight into the mechanisms by which PTHrP regulates OSCC malignancy, we examined the effects of PTHrP silencing on MAPK and phosphatidylinositol 3kinase (PI3K)/Akt signaling, which are involved in cell proliferation, survival, and motility. The phosphorylations of ERK1/2 and Akt at Ser473 were reduced in PTHrP-silenced HSC-2 cells compared to control cells, whereas phosphorylation of Akt at Thr308, in addition to the total expression levels, was not affected (Fig. 3F or data not shown).
Next, we investigated the expression level of PTH/PTHrP receptor PTH1R, and noticed that three OSCCs expressed PTH1R at a similar level of Saos-2, which is reported to express abundant PTH1R [8] (Fig. 3G). This observation raises the possibility that PTHrP functions in a paracrine/autocrine manner via PTH1R in OSCCs. In fact, exogenous PTHrP stimulation induced substantial phosphorylations of ERK1/2 and Akt in HSC-2 cells (Fig. 3H). In addition, PTHrP enhanced cell motility, and completely rescued the suppression by siPTHrP (Fig. 3I). On the other hand, Matrigel invasiveness was slightly increased in the presence of exogenous PTHrP ( Fig. 3J). Taken together, in OSCCs, PTHrP functions via cell surface PTH1R at least in part.

Synergetic effects of AG1478 and PTHrP knockdown on cell proliferation and invasion
Currently, the EGFR signaling is one of the most frequently targeted pathways for caner therapy; however, the therapy are well tolerated and have reproducible and unique toxicities, which often require dose interruptions or reductions. When we examined the effect of EGFR and MAPK inhibitors on cell proliferation, motility, and invasion of HSC-2 cells, essentially similar results were obtained as those of siPTHrP, except that p38 MAPK inhibitor enhanced cell proliferation ( Supplementary Fig. 5, 6). Therefore, to evaluate the possibility of combination of EGFR-TKI and PTHrP suppression for therapeutics, we finally investigated whether the combination exhibit synergistic effects on OSCC malignancy. Treatment of AG1478 in PTHrPdepleted HSC-2 cells most completely inhibited cell proliferation of HSC-2 cells (Fig. 4A).
Furthermore, the combination produced an additive effect on Matrigel invasiveness, reducing by 75% (Fig. 4B). These results may offer the dual targeting to EGFR and PTHrP as a potent novel therapeutics against OSCCs.

Discussion
Enhanced cell migration, invasion, and uncontrolled cell growth are crucial components for tumor progression, and, indeed, OSCC have a high potential for invasiveness, frequently targeting the mandibular bone. In this study, we used five OSCC cell lines to explore the role of  Fig. 1C), suggesting the autocrine production of EGFR ligands to constitutively activate EGFR and to dictate high levels of PTHrP (Fig. 1C). Alternatively, EGFR overexpression may cause transactivation of EGFR in a ligand-independent manner.
We showed that EGF activated ERK, p38 MAPK, and JNK in OSCC. Among them, ERK and p38 MAPK were involved in PTHrP expression. Both kinases are known to activate a distinct set of transcriptional factors and subsequent gene induction to contribute tumor progression [11,12]. Several transcription factors have been implicated in PTHrP expression [13,14]. Although high levels of Ets1 were detected in OSCC (unpublished result), we found no specific involvement of Ets1 in PTHrP expression by chromatin immunoprecipitation assay (data not shown). Thus, further detailed analyses will be required to clarify the transcription factors responsible for PTHrP expression in OSCC.
We found that phosphorylation of ERK and Akt was decreased in PTHrP-silenced HSC-2 cells (Fig. 3F). Furthermore, exogenous PTHrP stimulation induced ERK and Akt phosphorylation. PTHrP may functions, at least partly, in a paracrine/autocrine manner via PTH1R in OSCC (Fig. 3G-J). Recently, it is reported that PTHrP translocates into the nucleus and functions in an intracrine manner, which is associated with altered cell-cycle progression and protection from apoptosis [15][16][17]. We noted nuclear localization of PTHrP in addition to a cytoplasmic signal; however, PTHrP may not participate in apoptotic pathways in OSCC, since we found no cleaved Caspase-3, -9, or PARP in cells deficient in PTHrP (data not shown). The other intracrine functions of PTHrP seem to be connected to cell adhesiveness and motility through integrin expression and actin reorganization [3,18,19]. In this study, we showed that stimulation by recombinant PTHrP significantly promoted cell motility, whereas the effect on invasiveness was marginal (Fig. 3I, J). It is possible that PTHrP regulates the OSCC motility and invasiveness by different manners: the autocrine and intracrine pathways, respectively. Thus, PTHrP may coordinately enhance cell proliferation and invasiveness of OSCC through autocrine, paracrine, and intracrine signaling (Fig. 4C).
Targeting PTHrP in preclinical settings has been attempted using a neutralizing antibody, antisense oligonucleotides, and RNAi, which resulted in prolonged survival of hypercalcemic mice, decreased cell proliferation of renal carcinoma, and increased apoptosis in human medulloblastomas, respectively [20][21][22]. We also demonstrated the inhibitory effects of siPTHrP on cell proliferation and motility of OSCC. Moreover, PTHrP knockdown with AG1478 -13 -treatment exhibited more effective suppression on cell proliferation and invasiveness, even with the lower dose of AG1478 (Fig. 4A, B). Our results suggest the fascinating possibility that the combined application of PTHrP on dose-reduction of EGFR-TKI may contribute to avoid the EGFR-TKI-derived adverse aspects, and to maintain the anti-malignant potency. We therefore strongly propose targeting both EGFR and PTHrP in anticipation of their synergistic effects, which could lead to the conquest of oral cancer.