Skip to main content
NCBI home page
British Journal of Pharmacology logoLink to British Journal of Pharmacology
. 2017 Apr 7;174(10):1116–1130. doi: 10.1111/bph.13759

Protein phosphatase 2A regulation of markers of extracellular matrix remodelling in hepatocellular carcinoma cells: functional consequences for tumour invasion

M P Ward 1, J P Spiers 1,✉
PMCID: PMC5406286  PMID: 28239848

Abstract

Background and Purpose

A hallmark of tumour invasion is breakdown of the extracellular matrix due to dysregulation of the matrix metalloproteinase (MMP) system. While our understanding of how this is regulated by kinase signalling pathways is well established, its counter‐regulation by protein phosphatases (PP) is poorly understood. Therefore, we investigated the effect of PP inhibition on markers of extracellular remodelling and how PP2A activity modulated MMP‐9 abundance and function of Hep3B cells.

Experimental Approach

Cells were exposed to okadaic acid (OA), tautomycetin and cyclosporin A, and the expression profile determined using PCR. Effects of OA and a protein inhibitor of PP2A, CIP2A, on MMP‐9 abundance, PP2A activity and cell migration were investigated using ELISA, promoter constructs, siRNA knockdown and transwell migration assays.

Key Results

OA increased expression and abundance of MMP‐9 and the tissue inhibitor of MMP, TIMP‐1, without affecting other MMPs, TIMPs and ADAMs. The effect on MMP‐9 was mimicked by CIP2A overexpression and knockdown of the PPP2CA catalytic, but not PPP2R1A scaffolding, subunit. Cyclosporin A and PPP1CA silencing did not alter MMP‐9 expression, while tautomycetin transiently increased it. Mutation of AP‐1, but not NF‐κB, binding sites inhibited OA‐mediated MMP‐9 transcriptional activity. OA and CIP2A decreased PP2A activity and increased cell migration.

Conclusion and Implications

OA increased MMP‐9 by decreasing PP2A activity and PP2Ac, through AP‐1 binding sites on the MMP‐9 promoter. The functional consequence of this and CIP2A overexpression was increased cell migration. Hence, PP2A inhibition induced a metastatic phenotype through alterations in MMP‐9 in Hep3B cells.

Abbreviations

ADAM

a disintegrin and metalloproteinase

AP‐1

activator protein 1

CIP2A

cancerous inhibitor protein of PP2A

ECM

extracellular matrix

HCC

hepatocellular carcinoma

OA

okadaic acid

PMA

phorbol‐12‐myristate‐13‐acetate

PP

protein phosphatase

SET

protein SET

TIMP

tissue inhibitor of matrix metalloproteinase

Tables of Links

TARGETS
Enzymes MMP‐2
ADAM‐10 MMP‐3
ADAM‐15 MMP‐9
ADAM‐17 MMP‐13
MMP‐1
LIGANDS
CSA, cyclosporin A TIMP‐1
IL‐1β TIMP‐2
IL‐8 TIMP‐3
OA, okadaic acid TIMP‐4
TGF‐β1
Open in a new tab

These Tables list key protein targets and ligands in this article that are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Southan et al., 2016), and are permanently archived in the Concise Guide to PHARMACOLOGY (Alexander et al., 2015).

Introduction

Hepatocellular carcinoma (HCC) is the third most common cause of death from cancer in the world (El‐Serag, 2011). Despite advances in its diagnosis and treatment, the prognosis is dismal. Aetiologically, more than 78% of patients who develop HCC have pre‐existing liver cirrhosis due to chronic inflammation as a consequence of hepatitis, alcohol abuse or non‐alcoholic fatty liver disease (Perz et al., 2006).

Within a tumour, inflammatory mediators, including cytokines, chemokines, growth and transcription factors and enzymes such as the MMPs and their associated tissue inhibitors (TIMP) cooperate to create a micro‐environment conducive to tumour growth and invasion (Kessenbrock et al., 2010). A recognized hallmark of invasion is the proteolytic breakdown of the extracellular matrix (ECM) through dysregulation of the matrix metalloproteinase system (Hanahan and Weinberg, 2011; Bonnans et al., 2014). In addition to MMP‐9 (El Samanoudy et al., 2014), MMP‐2, MMP‐13, TIMP‐1 and TIMP‐3 are also involved (Altadill et al., 2009; Gu et al., 2014). Interestingly, their effects are not limited to degradation of the ECM. For example, MMP‐9 modulates the innate immune response to tumours, facilitating both proliferation and regression of tumour cells (Demers et al., 2005; Leifler et al., 2013), while MMP‐14 and TIMP‐1 are implicated in development of chemoresistance in glioblastoma and breast cancer cell lines (Hekmat et al., 2013; Pratt and Annabi, 2014). Interestingly, MMP‐2 and MMP‐9 facilitate release of pro‐metastatic factors like VEGF‐A, TGF‐β, TNF‐α and IL‐8 (Wang et al., 2005; Kessenbrock et al., 2010). Other members of the superfamily of metalloproteinases, a disintegrin and metalloproteinase (ADAM), also ‘shed’ receptor ligands involved in cancer and inflammatory processes; ADAM‐17 catalyses the release of TNF‐α, while ADAM‐10 releases certain EGFR ligands (Sahin et al., 2004; Duffy et al., 2009).

A key molecular driver underpinning ECM degradation and release of pro‐metastatic factors is over‐activation of ERK, MEK and p38 MAP kinase‐dependent signalling pathways (Chin et al., 2007; He et al., 2008) or a reduction in activity of counter‐regulatory phosphatases (Stebbing et al., 2014). Within the latter family, serine/threonine protein phosphatases form a major subfamily, which includes phosphoprotein phosphatase type I (PP1) and type 2A (PP2A). These catalyse approximately 90% of all serine/threonine dephosphorylation events in cells (Haystead et al., 1989) and are regulated by endogenous protein inhibitors such as cancerous inhibitor of PP2A (CIP2A) and protein SET (SET) (Neviani et al., 2005; Junttila et al., 2007). In HCC, CIP2A and SET are overexpressed promoting cell proliferation, conferring drug resistance and are associated with poor clinical outcomes (He et al., 2012; Yu et al., 2013; Hung et al., 2016).

While our understanding of how kinases regulate cell motility, ECM degradation and subsequent migration are well established; less is known on the role of protein phosphatases. In particular, little is known on how protein phosphatases control the MMP system. Previous work from our group has shown knockdown of the catalytic subunit of PP2A by siRNA did not mimic the MMP‐9 response to okadaic acid (OA) in murine fibroblasts (Rietz et al., 2012). This may be explained by non‐selective inhibition of PP2A, as OA can also inhibit PP1, PP4, PP6 and PP2B or could be a consequence of indirect phosphorylation of Cdc25, histone H1, PKA, PKB, PKC and Iκ‐B kinases (Heilker et al., 1999; Swingle et al., 2007). Nevertheless, the role of other phosphoprotein phosphatases (PP2B and PP1) and CIP2A in regulation of the MMP system and other modulators of ECM remodelling remains to be established in HCC.

Therefore, the aims of the present study were to (1) investigate if pharmacological inhibition of PP1, PP2A and PP2B altered expression of the MMP system and other markers of ECM remodelling; (2) establish protein phosphatase subunit involvement in MMP‐9 regulation and the requirement for NF‐κB and/or AP‐1 binding motifs; (3) ascertain if overexpression of CIP2A modulated MMP‐9 abundance; and (4) examine if inhibition of PP2A altered cell invasion.

Methods

Cell culture

Human epithelial HCC (Hep3B), human HCC (Huh‐7) cells and human hepatic stellate (LX‐2) cells were cultured in DMEM supplemented with 10% fetal bovine serum, antibiotics (100 U·mL−1 penicillin and 100 mg·mL−1 streptomycin) and 4 mmol·L−1 glutamine, at 37°C in a humidified atmosphere containing 5% CO2. All experiments were performed under serum‐free conditions. siRNA knockdown and DNA transfection were performed under antibiotic‐free conditions.

Cell viability

Hep3B and LX‐2 cells seeded in 96‐well plates (5 × 103 cells per well) were exposed to OA (PP2A inhibition), tautomycetin (PP1 inhibition), cyclosporin A (1–100 nmol·L−1; PP2B inhibition), media, doxorubicin hydrochloride (5 μmol·L−1, positive control) or DMSO (0.1% v/v, solvent control) for 24 h. After 22 h, MTT (3‐(4,5‐dimethylthiazol‐2yl)‐2,5‐diphenyltetrazolium bromide (5 mg·mL−1)) was added, and the cells incubated for a further 2 h. At the end of the incubation period, the culture medium was removed, and the purple formazan deposits dissolved in DMSO. Colour development was quantified by spectrophotometry at a wavelength of 540 nm with a reference wavelength of 650 nm (BioTek, EL 808, Bedfordshire, UK).

Semi‐quantitative real‐time PCR

Cells were grown in 24‐well plates (4 × 104 cells per well) and total RNA isolated using TRI Reagenttm (Sigma Aldrich). Following treatment with DNase I (Sigma Aldrich), RNA was reverse transcribed using random hexamers and ReverseAid reverse transcriptase (Thermofisher, Dublin, Ireland). mRNA expression was analysed by semi‐quantitative PCR using target specific primers (Table 1) and Agilent Brilliant III SYBR green on a M×3000P qPCR system (Agilent Technology, Cork, Ireland). Gene expression was quantified using the comparative Ct method. For each primer pair, a no template control and no RT control was included. Authenticity of the primer sequences was verified by nucleotide search (Primer‐BLAST; NCBI).

Table 1.

Details of primers and their sequences used in mRNA expression profiling and for confirmation of siRNA knockdown

Primer NM ID Forward sequence Reverse sequence
MMP‐1 NM_001145938.1 AGAAAGAAGACAAAGGCAAGTTGA TTCCCAGTCACTTTCAGCCC
MMP‐2 NM_004530.5 CGCATCTGGGGCTTTAAACAT CCATTAGCGCCTCCATCGTA
MMP‐3 NM_002422.3 AAAGACAGGCACTTTTGGCG CTTCATATGCGGCATCCACG
MMP‐9 NM_004994.2 CATCCGGCACCTCTATGGTC CATCGTCCACCGGACTCAAA
MMP‐13 NM_002427.3 GGGTCTTCCAAAAGAAGTTAAG ATATCTCCAGACCTGGTTTC
TIMP‐1 NM_003254.2 CATCCGGTTCGTCTACACCC TCTGCAGTTTGCAGGGGATG
TIMP‐2 NM_003255.4 CAGCTTTGCTTTATCCGGGC ATGCTTAGCTGGCGTCACAT
TIMP‐3 NM_000362.4 GAAGAGAGTACCGGCATCGG CCATTCTCCCCCTGCCAAAT
TIMP‐4 NM_003256.3 CCTTGGTGCAGAGGGAAAGT GTCCAGAGGCACTCGTTAGG
TGFB1 NM_000660.5 CCCACAACGAAATCTATGAC TGTATTTCTGGTACAGCTCC
IL1B1 NM_000576.2 GCAACAAGTGGTGTTCTC CAGATTCTTTTCCTTGAGGC
IL‐8 NM_000584.3 CCACCGGAAGGAACCATCTC TTCCTTGGGGTCCAGACAGA
ADAM‐10 NM_001110.3 GACCACAGACTTCTCCGGAAT TGAAGGTGCTCCAACCCAAG
ADAM‐15 NM_001261466.1 ACCCCAGTGGCAGTTATGTG CTTGGAGTCCCTGCCTTTGA
ADAM‐17 NM_003183.4 TTGTGTGGTTTGGCCCTTCT CTGCTTTTGCACCACAGGTC
JUN NM_002228.3 GCATGAGGAAACGCATCGCTGCCTCCAAGT GCGACCAAGTCCTTCCCACTCGTGCACACT
FOS NM_005252.3 TGCCTCTCCTCATCAATGACCCTG ATAGGTCCATGTCTGGCACGG
NFκB1 NM_003998.3 AAGAGGAGGTTTCGCCACCG AAGGTATGGGCCATCTGCTGTT
PPP2CA NM_002715.2 GCCTCTGCGAGAAGGCTAAA GAATGGTGATGCGTTCACGG
PPP2R1A NM_014225.5 AAAGGGACGGAGCCAAGATG GGGTAGTGAAGGTTCCCAGC
PPP1CA NM_001244974.1 CAGGGTCCTGACACCCCATT AGGTAAAAGAGACGCCACGG
CIP2A NM_020890.2 TGGACCCACAAATCACCTCG ATTACCTCCAAGTGCCGCAA
GAPDH NM_002046.5 CTCTGCTCCTCCTGTTCGAC GCGCCCAATACGACCAAATC
Open in a new tab

Gelatin zymography

MMP‐9 and MMP‐2 activity were measured by gelatin zymography. In brief, conditioned media denatured in a non‐reducing Laemmli sample buffer (250 mmol·L−1 Tris–HCl, pH 6.8, 0.1% bromophenol blue, 0.1% phenol red, 50% glycerol, 5% Ficoll 400 and 5% lithium dodecyl sulphate; final concentrations) was loaded onto 8% SDS‐polyacrylamide gels co‐polymerized with gelatin (0.1% w/v). Following electrophoresis, gels were washed in 2.5% Triton X‐100 and incubated for 24 h at 37°C in development buffer (50 mmol·L−1 Tris–HCl, 10 mmol·L−1 CaCl2, 50 mmol·L−1 NaCl, pH 7.6). Gels were stained with 0.25% (w/v) Coomassie blue R‐250 in a glacial acetic acid : methanol : distilled H2O mixture (1:3:6). Following destaining, gels were imaged, and lytic activity was quantified by densitometry using a Fusion Fx imaging system. To test for non‐specific protease activity, identical gels were incubated in either the standard incubation buffer or one supplemented with EDTA (20 mmol·L−1) to inhibit gelatinase activity.

siRNA knockdown of protein phosphatases

Hep3B cells were seeded in 24‐well plates (4 × 104 cells per well) 24 h prior to transfection with siRNA targeting PPP2CA (10 nmol·L−1; catalytic subunit of PP2A; Table 2), PPP2R1A (10 nmol·L−1; scaffolding subunit of PP2A), PPP1CA (10 nmol·L−1; catalytic subunit α of PP1; Integrated DNA Technology, USA) or non‐target siRNA (Fisher, Dublin, Ireland) using HIPerFect, according to the manufacturer's instructions (Qiagen, Manchester, UK); a mock transfection was also included as a control. Knockdown of PPP2CA, PP2R1A and PPP1CA was confirmed by qPCR (Table 1), and knockdown of PP2Ac was confirmed by Western blot. The culture medium was collected 48 h post‐transfection, and MMP‐9 abundance and activity were determined using ELISA and zymography.

Table 2.

List of dicer siRNA used in this study along with their sequences

Duplex name Sequence
PPP2R1A Sense 5′‐rArGrGrCrGrGrArArCrUrUrCrGrArCrArGrUrA‐3′
Anti‐Sense 5′‐rArArArCrUrUrArArCrUrCrCrUrUrGrUrGrCrA′3
PPP2CA Sense 5′‐rGrGrArUrArGrCrArGrCrArArArCrArArUrCrArUrUrGrGrArG‐3′
Anti‐Sense 5′‐rCrUrCrCrArArUrGrArUrUrGrUrUrUrGrCrUrGrCrUrArUrCrCrUrUND3′
PPP1CA Sense 5′‐rCrArArGrArUrCrUrGrCrGrGrUrGrArCrArUrArUrUr‐3
Anti‐sense 5′‐rCrArArGrArGrArCrGrCrUrArCrArArCrArUrCrUrUr‐3
Open in a new tab

Overexpression of CIP2A

To assess the effect of overexpression of CIP2A on MMP‐9 activity and invasion potential, Hep3B cells were seeded (2 × 105 cells per well) in 6‐well plates and 24 h later transiently transfected with 1 μg of pcDNA3.1_CIP2aflag_WT plasmid (pCIP2A) or a pcDNA.3.1 (pDNA.3.1) control plasmid using PolyFect (Qiagen) in serum and antibiotic free DMEM. CIP2A overexpression was confirmed 24 and 48 h post‐transfection by qPCR and Western blot respectively. The effect of overexpression of CIP2A on MMP‐9 abundance and activity was determined in conditioned media 48 h post‐transfection using ELISA and gelatin zymography.

Plasmid stocks were generated in TOP10 Escherichia coli (transformation efficiency of 1.6 × 107 cfu·μg−1 DNA) and purified using a PureYieldtm Plasmid Midiprep System (Promega, Kilkenny, Ireland).

Confirmation of CIP2A overexpression by Western blot analysis

Protein samples (10–20 μg) were boiled in Laemmli buffer supplemented with 2‐β‐mercaptoethanol (1% v/v). The samples were subjected to SDS‐PAGE (10% gel) and semi‐dry transferred onto PVDF membrane (Amersham, Buckinghamshire, UK). Membranes were blocked for 2 h in TBS‐T (Tris–HCl 10 mmol·L−1, NaCl 100 mmol·L−1, pH 7.4 and Tween‐20 0.1% v/v) containing 5% w/v Marvel and probed with 1o antibody directed against CIP2A (Santa Cruz Biotechnology, Santa Cruz, CA, USA; 1:1000 dilution) for 24 h at 4°C. Following extensive washing, membranes were incubated with a secondary rabbit anti‐mouse antibody conjugated to HRP (1:2500 dilution). Signals were visualized by chemiluminescence detection (30% H2O2, 250 mmol·L−1 luminol in DMSO, 90 mmol·L−1 4‐iodophenylboronic acid in DMSO and 10 mmol·L−1 Tris–HCl) and the signal, recorded using a Fusion Fx imaging system. Membranes were stripped and re‐probed with a HRP‐conjugated β‐actin antibody as a loading control (1:7500, Santa Cruz Biotechnology).

PP2A phosphatase activity assay

PP2A activity was determined using a PP2A immunoprecipitation phosphatase assay kit (Millipore). Briefly, cells were lysed on ice in a buffer comprised of 50 mmol·L−1 Tris–HCl pH 7.4, 150 mmol·L−1 NaCl, 1 mmol·L−1 EDTA and 1% NP40 containing a protease inhibitor cocktail. About 50–100 μg of total protein was immunoprecipitated with anti‐PP2A‐C subunit antibody (4 μg; clone 1D6) along with protein A‐agarose. The mix was incubated for 2 h at 4°C, and the immunoprecipitate was extensively washed before determination of the phosphatase activity according to the manufacturer's instructions. Unmethylated PP2A phosphatase activity was calculated by dividing the free phosphate of the test cells by that of the untreated cells and expressed as a percentage. Each assay run contained a negative control (no lysate) as a quality control for phosphatase contamination.

MMP‐9 reporter assay

Hep3B cells were transfected with the full length pGL3‐MMP91285‐Luc plasmid (400 ng per well) or mutated counterparts (pGL3‐MMP9/NF‐κB(−600)‐Luc, pGL3‐MMP9/AP‐1(−79)‐Luc or pGL3‐MMP9/AP‐1(−533)‐Luc) in conjunction with pRL‐CMV (1 ng; internal standard) using PolyFect (Qiagen). Cells were subsequently exposed to OA (40 nmol·L−1) or media for 24 h, before being harvested and lysed. Firefly and Renilla luciferase activity was evaluated using a Dual‐Luciferase Reporter Assay System (Promega).

MMP‐9 ELISA

The effect of OA, silencing of PPP2CA by siRNA and overexpression of CIP2A on MMP‐9 abundance, was determined in conditioned media using ELISA following the vendor's protocol (Quantikine/DuoSet; R&D Systems, Abingdon, UK). Culture media was reserved 24 h post‐drug exposure and at 48 h post‐transfection. All samples were stored at −80°C until required.

Cell invasion assays

Transwell inserts (6.5 mm, 8 μm pore; Corning, Flintshire, UK) were pretreated with Matrigel (5 mg·mL−1; Sigma‐Aldrich, Ireland) for 24 h. Hep3B cells (2 × 104 cells per well) were loaded into the upper compartment of the transwell and exposed to OA (40 nmol·L−1) or media (control) under serum‐free conditions; 10% FBS was included in the lower chamber as chemoattractant. For CIP2A overexpression experiments, cells were transfected with pCIP2A for 48 h before being loaded into the upper compartment of the transwell. In all experiments, the filters were removed 24 h later; cells were fixed in 3% paraformaldehyde and stained with crystal violet. The matrigel membranes were imaged on a Cell Imaging System (EVOS® FL) for subsequent analysis. Each experiment was performed in triplicate.

Data and statistical analysis

The data and statistical analysis in this study comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). Data are expressed as mean ± SEM. All data were normalized to the appropriate control and expressed as either an absolute value, a percentage or ratio, with the exception of the MMP‐9 reporter data, which was expressed as the ratio of firefly luciferase to Renilla luciferase activity normalized to control (media) or pGL3‐MMP91285‐Luc promoter activity as appropriate. Cell invasion assays were evaluated using ImageJ software (Schneider et al., 2012). Data were analysed by one‐way anova or two‐way anova with post hoc analysis (Dunnett's or Bonferroni) as appropriate. Where applicable, data were analysed by unpaired Student's t‐test. A value of P < 0.05 was taken to indicate statistical significance.

Materials

OA, tautomycetin and cyclosporin A were purchased from Calbiochem (Carrigtwohill, Ireland). Mouse anti‐PP2A (C subunit, clone 1D6) and goat anti‐mouse HRP‐conjugated secondary antibodies were obtained from Millipore (Carrigtwohill, Ireland) and Sigma‐Aldrich (Arklow, Ireland) respectively. Mouse anti‐CIP2A and HRP‐conjugated β‐actin antibodies were acquired from Santa Cruz Biotechnology (Dublin, Ireland). All other chemicals were obtained from Sigma‐Aldrich (Ireland). LX‐2 hepatic stellate cells were a gift from Prof Scott L. Friedman, Mount Sinai School of Medicine, USA, while Hep3B and Huh‐7 cells were a gift from Dr Steven Grey, Trinity College Dublin, Ireland. siRNA was obtained from IDT (Integrated DNA Technologies, Leuven, Belgium), while DNA primers were purchased from Metabion (Steinkirchen, Germany). Details of the MMP‐9 wild‐type reporter construct (pGL3‐MMP91285‐Luc) and its NF‐κB (pGL3‐MMP9/NF‐κB(−600)‐Luc) and AP‐1 mutated counterparts (pGL3‐MMP9/AP‐1(−79)‐Luc and pGL3‐MMP9/AP‐1(−533)‐Luc) have been previously published by our group (Rietz et al., 2012). The CIP2A expression plasmid, pcDNA3.1_CIP2aflag_WT, was kindly provided by Prof J Westermarck, University of Turku, Finland.

Results

Effect of protein phosphatase inhibitors on cell viability and mRNA profile

Exposure to OA or tautomycetin for 24 h did not alter cell viability over the concentration range 0.1–100 nmol·L−1. However, cyclosporin A decreased cell viability at a concentration of 66 nmol·L−1 (45%) and higher (data not shown).

The effect of pharmacological inhibition of PP1 (tautomycetin), PP2A (OA) and PP2B (cyclosporin A) on a panel of markers representing extracellular matrix modulators, inflammation/growth factor releasers and transcription factors (Table 1) was screened by qPCR. With regard to the extracellular matrix modulators, OA (40 nmol·L−1) increased expression of MMP‐9 and TIMP‐1 mRNA in Hep3B cells (Table 3). In LX‐2 cells, MMP‐9 expression was increased, while that of TIMP‐1 decreased (Table 3). MMP‐9 and TIMP‐1 expressions were not altered by tautomycetin or cyclosporin A (10 nmol·L−1) in either cell line. In Hep3B and LX‐2 cells, expressions of MMP‐1, MMP‐2, MMP‐3, MMP‐13, TIMP‐2, TIMP‐3 and TIMP‐4 were unaltered by OA, tautomycetin and cyclosporin A at the concentrations studied. OA augmented IL‐8 mRNA expression compared with control (Table 3) in Hep3B and LX‐2 cell lines respectively; no effect on TGF‐β1, IL‐1β, ADAM‐10, ADAM‐15 or ADAM‐17 was noted. None of the inflammation/growth factor releasers were altered by tautomycetin or cyclosporin A, irrespective of cell line. Of the three transcription factors studied, only JUN expression was increased by OA (Table 3) in Hep3B and in LX‐2 cells. Tautomycetin and cyclosporin A did not modify mRNA expression of JUN, FOS or NF‐κB1. To validate the positive hits, the effects of the phosphatase inhibitors were also assessed in a second HCC cell line, Huh‐7 cells. MMP‐9 and TIMP‐1 mRNA expression was increased in these cells (Table 3) following exposure to OA. Similarly, OA increased IL‐8 and JUN mRNA expression ( Table 3). As in Hep3B and LX‐2 cells, tautomycetin and cyclosporin A had no effect on expression of mRNA for MMP‐9, TIMP‐1, IL‐8 or JUN, in Huh‐7 cells.

Table 3.

Effect of protein phosphatase inhibitors on mRNA expression

Gene Okadaic acid (40 nmol·L−1) Tautomycetin (10 nmol·L−1) Cyclosporin A (10 nmol·L−1)
Hep3B LX‐2 Huh7 Hep3B LX‐2 Huh7 Hep3B LX‐2 Huh7
MMP‐1 NS NS NS NS NS NS
MMP‐2 NS NS NS NS NS NS
MMP‐3 NS NS NS NS NS NS
MMP‐9 1.90 ± 0.34* 1.28 ± 0.06* 1.29 ± 0.23* NS NS NS NS NS NS
MMP‐13 NS NS NS NS NS NS
TIMP‐1 1.74 ± 0.31* −0.71 ± 0.18* 1.47 ± 0.32* NS NS NS NS NS NS
TIMP‐2 NS NS NS NS NS NS
TIMP‐3 NS NS NS NS NS NS
TIMP‐4 NS NS NS NS NS NS
TGF‐β1 NS NS NS NS NS NS
COL1A NS NS NS NS NS NS
IL‐1β NS NS NS NS NS NS
IL‐8 1.57 ± 0.24* 1.85 ± 0.39* 2.72 ± 0.17* NS NS NS NS NS NS
ADAM‐10 NS NS NS NS NS NS
ADAM‐15 NS NS NS NS NS NS
ADAM‐17 NS NS NS NS NS NS
JUN 7.04 ± 0.45* 4.71 ± 1.07* 3.72 ± 0.59* NS NS NS NS NS NS
FOS NS NS NS NS NS NS
NFκB1 NS NS NS NS NS NS
Open in a new tab

Effect of pharmacological inhibition of protein phosphatases on change in gene expression. Hep3B, LX‐2 and Huh‐7 cells were exposed to OA (40 nmol·L−1), cyclosporin A (10 nmol·L−1) and tautomycetin (10 nmol·L−1) for 24 h before the mRNA expression profile was determined by semi‐quantitative RT‐PCR using SYBR green chemistry. Data are presented as fold change (mean ± SEM) and were analysed using Student's t‐test.

*

indicates P < 0.05 (n = 5). NS, no significant fold change following exposure to the inhibitor.

On the basis of these data, subsequent experiments focussed on regulation of MMP‐9.

Effect of protein phosphatase inhibitors on temporal expression of MMP‐9

To ensure that tautomycetin and cyclosporin A did not have a transient effect on MMP‐9 expression, the temporal response over 24 h was investigated. In Hep3B cells, OA (40 nmol·L−1) elicited a time‐dependent increase in expression of MMP‐9 mRNA that was detectable after 4 h and persisted for the remainder of the study period (P < 0.05; Figure 1). Although tautomycetin (10 nmol·L−1) caused a similar increase in MMP‐9 expression (~1.4‐fold) as OA, the response was transient, peaking at 4 h (P < 0.05; Figure 1), before quickly returning to baseline by 8 h and remaining stable thereafter. In contrast, cyclosporin A did not alter expression of MMP‐9 mRNA over the 24 h time course.

Figure 1.

Figure 1

Open in a new tab

Temporal effect of OA (40 nmol·L−1), cyclosporin A (CSA; 10 nmol·L−1) and tautomycetin (TMC; 10 nmol·L−1) on MMP‐9 mRNA expression determined by qPCR (comparative quantification) in Hep3B cells over 24 h. Data are normalized to baseline (mean ± SEM; n = 5). *P < 0.05, significantly different from T=0; two‐way anova with post hoc Dunnett's test.

Effect of protein phosphatase inhibitors on MMP‐9 abundance and MMP‐9 : TIMP‐1 ratio

OA (40 mmol·L−1) increased secreted MMP‐9 abundance in Hep3B cells (Figure 2A) and in LX‐2 cells (P < 0.05; Figure 2B). Tautomycetin and cyclosporin A (10 nmol·L−1) had no effect on secreted MMP‐9 abundance in either cell line studied. In Hep3B cells, OA increased TIMP‐1 abundance in the conditioned media (Figure 2C). Interestingly, TIMP‐1 levels decreased in conditioned media from LX‐2 cells (P < 0.05; Figure 2D). While OA had little effect on the MMP‐9 : TIMP‐1 ratio in Hep3B cells (Figure 2E), it was augmented in LX‐2 cells (Figure 2F). Tautomycetin and cyclosporin A did not alter TIMP‐1 abundance or the MMP‐9 : TIMP‐1 ratio in either cell line.

Figure 2.

Figure 2

Open in a new tab

Effect of protein phosphatase inhibitors on MMP‐9 (A, B) and TIMP‐1 abundance (C, D) and MMP‐9 : TIMP‐1 ratio (E, F) in Hep3B (A, C, E) and LX‐2 (B, D, F) cells. Cells were exposed to OA (40 nmol·L−1), tautomycetin (TMC; 10 nmol·L−1), cyclosporin A (CSA; 10 nmol·L−1) or media (UNTX) for 24 h, and abundance was determined in conditioned media using ELISA. Data are presented as pg·mL−1 or as a ratio (mean ± SEM; n = 5). P < 0.05, significantly different as indicated by the horizontal bars; one‐way anova with post hoc Dunnett's test.

Effect of protein phosphatase inhibitors on PP2A activity and protein expression

OA (40 mmol·L−1) decreased PP2A phosphatase activity by approximately 80% in Hep3B and Huh‐7 cells (P < 0.05; Figure 3A, B). Tautomycetin (10 nmol·L−1) and cyclosporin A (10 nmol·L−1) had no effect on PP2A phosphatase activity in either cell line at the concentration studied at 24 h. On investigating the OA response further, it was found that OA (40 mmol·L−1) did not alter PP2Ac abundance in either Hep3B or Huh‐7 cells (Figure 3C). The effects of longer exposure (48 and 72 h) were not investigated as OA (40 mmol·L−1) decreased cell viability in Hep3B cells at the later time points (P < 0.05; Figure 3D).

Figure 3.

Figure 3

Open in a new tab

Effect of exposure to OA (40 nmol·L−1), tautomycetin (TMC; 10 nmol·L−1), cyclosporin A (CSA; 10 nmol·L−1) and media (UNTX) for 24 h on PP2A activity in Hep3B cells (A) and Huh‐7 cells (B) and PP2Ac abundance in both cell lines (C). PP2Ac abundance was determined using Western blot, with β‐actin as the loading control. The effect of OA (40 nmol·L−1) on Hep3B cell viability was assessed over 72 h using an MTT assay (D). Data are mean ± SEM; n = 5. P < 0.05, significantly different as indicated by the horizontal bars; one‐way or two‐way anova with post hoc Dunnett's or Bonferroni test.

Role of binding sites for the transcription factors AP‐1 and NF‐κB on MMP‐9 promoter activity

In Hep3B‐transfected cells, mutation of the AP‐1(−79) but not AP‐1(−533) binding motif attenuated basal MMP‐9 promoter activity by ~55% compared with the p1285‐luc (wild type) promoter construct (P < 0.05; Figure 4A, C). Similarly, mutation of the NF‐κB(−600) binding site decreased basal promoter activity by ~40% (P < 0.05; Figure 4A). OA (40 nmol·L−1) increased p1285‐luc MMP‐9 promoter activity, compared with untreated cells (Figure 4B) at 24 h. Mutation of the AP‐1(−79) binding motif prevented OA‐mediated induction of MMP‐9 promoter activity, while mutation of the AP‐1(−533) binding motif reduced the response by 40% (P < 0.05; Figure 4B). When the NF‐κB(−600) binding site was mutated, the response to OA was similar to that in cells transfected with the p1285‐luc (wild type) promoter (Figure 4B).

Figure 4.

Figure 4

Open in a new tab

Effect of mutation of the AP‐1 and NF‐κB binding motifs on MMP‐9 promoter activity under basal conditions (A) and following exposure to OA for 24 h (40 nmol·L−1, B). A diagrammatic representation of the MMP‐9 promoter region detailing the location of the mutated binding motifs is presented (C). Hep3B cells were transfected with p1285‐luc, p1285‐luc AP‐1(−79), p1285‐luc AP‐1(−533) and p1285‐luc NF‐κB(−600) promoter constructs, and luciferase activity was determined in cell lysates. Basal promoter activity for each mutant was normalized to the p1285 construct, and their effect on the OA response was expressed as a change from the p1285‐luc construct under unstimulated conditions. Data are presented as mean ± SEM; n = 7–9. P < 0.05, significantly different as indicated by the horizontal bars; one‐way anova with post hoc Dunnett's or Bonferroni analysis, as appropriate.

Effect of PP2A and PP1 subunit knockdown on MMP‐9 activity

To ascertain PP2A and PP1 subunit involvement in MMP‐9 regulation, Hep3B cells were transfected with siRNA targeting the catalytic (PPP2CA) and scaffold (PPP2R1A) subunits of PP2A and the catalytic subunit of PP1 (PPP1CA). siRNA knockdown reduced basal mRNA expression of PPP2CA, PPP2R1A and PPP1CA by ~75% after 24 h compared with the mock‐transfected and non‐target transfected cells (Figure 5A). Knockdown of PPP2CA for 48 h decreased PP2Ac protein abundance by ~70% compared with mock and non‐template siRNA transfected cells (Figure 5B). Silencing of PPP2CA increased MMP‐9 abundance in conditioned media and increased detectable MMP‐9 activity, compared with media from mock‐transfected cells (Figure 5C, D). In contrast, knockdown of PPP2R1A and PPP1CA for 48 h did not alter detectable MMP‐9 activity or abundance (Figure 5C, D). Furthermore, knockdown of PPP2CA, PPP2R1A and PPP1CA had no effect on detectable MMP‐2 activity as determined by zymography. Transfection with the non‐target siRNA had no effect on protein phosphatase subunit expression or MMP‐9 activity/abundance.

Figure 5.

Figure 5

Open in a new tab

Effect of silencing the PP2A catalytic subunit (PPP2CA), PP2A regulatory subunit (PPP2R1) and PP1 catalytic subunit (PPP1CA) on (A) subunit mRNA expression under basal conditions, (B) PP2Ac protein abundance, (C) detectable MMP‐9 abundance and (D) MMP‐9 activity. Data are presented as a normalized ratio, a percentage of mock‐transfected cells or in pg·mL−1 (mean ± SEM; n = 5). P < 0.05 significantly different as indicated by the horizontal bars; one‐way anova with post hoc analysis (Bonferroni).

Effect of CIP2A overexpression on MMP‐9 activity, abundance and PP2A activity

To extend the physiological relevance of the OA and siRNA data, we investigated if overexpression of CIP2A, which is an endogenous inhibitor of PP2A, also regulated MMP‐9 in Hep3B cells.

Transfection of Hep3B cells with a pCIP2A plasmid augmented CIP2A mRNA expression and protein expression (Figure 6A, B) compared with mock‐transfected and empty vector (pDNA.3.1)‐transfected cells. CIP2A over‐expression increased total MMP‐9 abundance and increased detectable MMP‐9 activity compared with mock‐transfected and pDNA.3.1‐transfected cells (Figure 6C, D). Transfection of Hep3B cells with pDNA.3.1 did not alter CIP2A mRNA expression or total MMP‐9 abundance/activity compared with mock‐transfected cells (Figure 6A, C, D). However, overexpression of CIP2A attenuated PP2A activity by ~60% compared with mock‐transfected controls (Figure 6E).

Figure 6.

Figure 6

Open in a new tab

Effect of CIP2A overexpression on (A) CIP2A mRNA expression, (B) CIP2A protein abundance, (C) total MMP‐9 abundance, (D) MMP‐9 activity (representative zymogram) and (E) PP2A activity in Hep3B cells. Hep 3B cells were transfected with pcDNA3.1_CIP2aflag_WT plasmid and analysed for CIP2A expression, MMP‐9 abundance/activity and PP2A activity at 48 h. Data are presented as a normalized ratio/percentage of the mock‐transfected cells or as pg·mL−1 (mean ± SEM; n = 5–7).

Effect of OA and CIP2A overexpression on cell invasion

To determine if pharmacological or endogenous inhibition of PP2A is functionally important, a Matrigel transwell cell invasion assay was utilized. Exposure of Hep3B cells to OA (40 nmol·L−1) for 24 h augmented cell invasion through Matrigel‐coated transwell membranes compared with the untreated control cells (P < 0.05; Figure 7A, C). Similarly, overexpression of CIP2A for 48 h increased cell migration compared with mock‐transfected cells (P < 0.05; Figure 7B, D); transfection with the empty plasmid (pDNA.3.1) had no effect.

Figure 7.

Figure 7

Open in a new tab

Effect of OA (A, C) and CIP2A (B, D) on Hep3B cell invasion across a Matrigel‐coated transwell membrane. Cells were exposed to OA (40 nmol·L−1) or transfected with a CIP2A expressing plasmid and images captured on a Cell Imaging System (EVOS® FL). Media (UNTX) and empty vector (pDNA.3.1) were included as controls. Image analysis was preformed using ImageJ software. Data (mean ± SEM; n = 5) are presented normalized to either the untreated (UNTX) or pDNA.3.1 transfected groups. P < 0.05 significantly different as indicated by the horizontal bars; paired Student's t‐test or one‐way anova with post hoc analysis (Bonferroni), as appropriate. Representative images of Hep3B cell migration across the transwell inserts are inset (A, B).

Discussion

The present study provides evidence that OA increases expression of MMP‐9, IL‐8 and JUN mRNA in Hep3B, Huh‐7 and LX‐2 cells but differentially regulates TIMP‐1 expression, while tautomycetin and cyclosporin A had no effect at 24 h. Interestingly, in time course experiments, tautomycetin transiently increased MMP‐9 mRNA expression, while OA elicited a persistent response. With respect to OA, cell viability was decreased at 48 and 72 h. Furthermore, OA increased MMP‐9 and TIMP‐1 abundance in Hep3B and Huh‐7 cells, while increasing MMP‐9 but decreasing TIMP‐1 levels in LX‐2 cells. As such, the MMP‐9 : TIMP‐1 ratio was changed towards a degradative phenotype in LX‐2 cells. Expressions of other members of the MMP/TIMP/ADAM family and inflammatory/growth factors such as TGFβ1, IL‐1β and NFκ‐B1 were unaffected by OA, tautomycetin and cyclosporin A. OA exposure did not alter total PP2Ac protein expression in either Hep3B or Huh‐7 cell lines. Knockdown of PPP2CA increased MMP‐9 activity and abundance, while knockdown of the PP2A scaffolding subunit (PPP2R1A) or the catalytic subunit α of PP1 (PPP1CA) had no effect. At the promoter level, OA‐mediated induction of MMP‐9 required functional AP‐1 but not NF‐κB binding motifs. More importantly, overexpression of CIP2A, an endogenous inhibitor of PP2A, increased MMP‐9 abundance and activity and increased Hep3B invasiveness across a collagen matrix, as did OA.

Although the ECM provides structural support to cells, it also contributes to fundamental cell behaviour as diverse as cell proliferation, adhesion and migration, to cell differentiation and cell death; in part due to its ability to act as a reservoir for growth factors and cytokines (Meredith et al., 1993; Li et al., 2003; Lu et al., 2012). It is the pleiotropic nature of the ECM and its dynamic remodelling that makes it an important component of cancer. While numerous studies have investigated kinase dependent regulation of tumour remodelling and invasion (Dhillon et al., 2007; Whale et al., 2011), there is a paucity of data on the role of serine/threonine protein phosphatases in this regard. This is surprising given the importance of tumour dissemination as an indicator of clinical outcome in HCC and the emergence of protein phosphatases as novel ‘druggable’ targets (Perrotti and Neviani, 2008). On the basis of this, we screened key factors associated with ECM synthesis and degradation (Table 1) to ascertain if they are modulated by protein phosphatases in HCC. The work is based upon Hep3B cells, which are widely utilized as an in vitro model of HCC with validation experiments conducted in two further hepatic cell lines, Huh‐7 and LX‐2.

With regard to the ADAMs, we assessed the effects of pharmacological inhibition of PP2A, PP2B and PP1 on ADAM‐10, ADAM‐15 and ADAM‐17 expressions. These were specifically chosen as their expression is increased in HCC and is associated with tumour metastasis and proliferation. Our data clearly demonstrate that OA, tautomycetin and cyclosporin A do not alter expression of ADAM‐10, −15 or −17 in Hep3B or LX‐2 cells. While there is little to compare this to in the literature, a recent study reports OA to increase ADAM‐10 mRNA expression in neuronal cells from rat (Latta and Golding, 2012).

In addition to the above, inflammation also plays its role in shaping the ECM in the tumour micro‐environment. For example, in patients with HCC, serum levels of IL‐1β and IL‐8 are elevated and associated with poor prognosis (Ren et al., 2003). However, we report that IL‐1β expression is not affected by OA, cyclosporin A or tautomycetin in either Hep3B or LX‐2 cells. This is consistent with OA having little to no effect on IL‐1β expression in THP‐1 cells. Although this is at variance to other studies demonstrating OA to increase IL‐1β expression in rat astrocytes (Pshenichkin and Wise, 1997) and enhancement of phorbol‐12‐myristate‐13‐acetate (PMA)‐induced IL‐1β expression in THP‐1 monocytes (Shanley et al., 2001). Our data are also in disagreement with studies reporting cyclosporin A to decrease IL‐1β expression in transplanted islets of Langerhans and in a brain injury model in rats (Gurol et al., 2005; Qian et al., 2010). However, the concentration used in our study is well below the therapeutic levels used in the above studies and is therefore more likely to selectively inhibit PP2B.

Interestingly, in the present study, inhibition of PP2A by OA increased IL‐8 expression in Hep3B, Huh‐7 and LX‐2 cells. This supports and extends previous studies in human lung epithelial and human promyelocytic leukaemia cells (Sonoda et al., 1997; Cornell et al., 2009). However, pharmacological inhibition of PP2B and PP1 had no effect, despite previous studies showing cyclosporin A to attenuate PMA‐induced IL‐8 expression through functional NF‐κB and AP‐1 sites on the IL‐8 promoter (Sonoda et al., 1997); Wakabayashi et al. (2004) in human glioblastoma cells.

With regard to the matrix metalloproteinase system, pharmacological inhibition of PP2A (OA) increased TIMP‐1 expression in Hep3B and Huh‐7 cells while decreasing it in LX‐2 hepatic stellate cells. This is likely to reflect cell line differences as OA increases TIMP‐1 mRNA expression in HT‐1080 fibrosarcoma cells (Westermarck et al., 1997), while PPP2CA siRNA knockdown decreases it in human HA22T HCC cells (Dung et al., 2013). In the current study, we did not find any evidence that pharmacological inhibition of PP2B (cyclosporin A) and PP1 (tautomycetin) altered TIMP‐1 expression in the cell lines used. While there is no published literature on tautomycetin regulating TIMP‐1, there are conflicting data with regard to cyclosporin A. For example, cyclosporin A decreased TIMP‐1 mRNA but not protein expression in gingival fibroblasts (Hyland et al., 2003) but increased it in MRC‐5 fibroblasts and HK‐2 renal tubular epithelial cells (Esposito et al., 2000). However, these studies are limited in that the concentrations of cyclosporin A used are well above that associated with selective inhibition of PP2B (65 nmol·L−1)(Groblewski et al., 1994). Furthermore, we are the first to report that pharmacological inhibition of PP2A, PP2B and PP1 did not affect TIMP‐2, ‐3 or ‐4 expression in our cell lines, although silencing of CIP2A increased TIMP‐4 mRNA expression in MCF‐7 cells (Niemela et al., 2012).

When we investigated the effect of OA, cyclosporin A and tautomycetin on MMP expression, we found that expression of MMP‐1, ‐2, ‐3 and ‐13 in Hep3B and LX‐2 hepatic stellate cells was unaltered. In contrast, OA increases MMP‐1 promoter activity in murine and human fibroblasts (Westermarck et al., 1994; Westermarck et al., 1997), MMP‐2 expression and mRNA stability in U937 cells exposed to amsacrine (Liu et al., 2014), MMP‐3 in HT1080 cells (Westermarck et al., 1994) and MMP‐13 in murine fibroblasts (Westermarck et al., 2000). The reported literature is less consistent with regard to cyclosporin A as it decreases MMP‐1 in human gingival fibroblasts but increases it in HUVEC cells (Hyland et al., 2003; Ha and Mun, 2012). In the present study, tautomycetin did not alter expression of MMP‐1, ‐2, ‐3 or ‐13 in either Hep3B or LX‐2 hepatic stellate cells. While this has not been assessed before, activation of PP1 decreases MMP‐1 expression in fibroblasts (Westermarck et al., 2001).

With regard to MMP‐9, the story is quite different. As previously reported by our group (Rietz et al., 2012), OA increases MMP‐9 expression and activity in murine fibroblasts. We have extended previous work to show that tautomycetin and cyclosporin A do not affect MMP‐9 expression or abundance in either cell line. Although, when the temporal response was investigated, tautomycetin elicited a rapid but transient increase in MMP‐9 expression, indicating that PP1 may be involved; OA produced a sustained response. The transient effect of tautomycetin may be an ‘off target’ response, as knockdown of the α catalytic subunit of PP1 did not alter MMP‐9 activity or abundance. However, we cannot rule out the involvement of other PP1 catalytic subunits (PPP1CB and PPP1CC (Cohen, 2002)). In contrast, we have shown PPP2CA knockdown to increase MMP‐9 activity and abundance in Hep3B cells supporting data from HA22T cells (Dung et al., 2013) but is at variance to our earlier study in murine fibroblasts (Rietz et al., 2012). The latter may simply reflect species differences or regulation via other PP2A subunits. While this was not investigated in our earlier paper, we show that knockdown of the PPP2R1A scaffolding subunit of PP2A is not involved in regulating MMP‐9 in Hep3B cells.

The MMP‐9 promoter contains several transcription factor binding sites including AP‐1 sites and one NF‐κB site (Westermarck and Kahari, 1999). Mutation of the proximal AP‐1(−79) site or NF‐κB(−600) attenuated basal MMP‐9 promoter activity indicating the importance of these motifs in maintaining constitutive activity of the MMP‐9 promoter. Indeed, mutation of the proximal AP‐1(−79) site in HIG‐82 cells diminished basal MMP‐9 promoter activity (Ray et al., 2005). Moreover, mutation of the proximal AP‐1(−79) and distal AP‐1(−533) binding site indicating that they are required for activation of the p1285‐luc promoter by OA in Hep3B cells, while the NF‐κB motif is not. Previous studies have shown OA to enhance expression of JunB and trans‐activation of AP‐1 complexes containing c‐Jun and JunB to regulate MMPs (Westermarck et al., 1994; Rietz et al., 2012).

While SET inhibits PP2A activity through interactions with the catalytic subunit (Irie et al., 2012), the precise interaction between CIP2A and PP2A remains to be determined as no crystal structure yet exist. Data from immunoprecipitation experiments indicates that CIP2A interacts with the scaffold (PP2Aa; R65) subunit of PP2A (Junttila et al., 2007). Although overexpression of SET decreases MMP‐9 mRNA expression in neck squamous cell carcinoma (Sobral et al., 2014), we found that overexpression of CIP2A increases MMP‐9 activity and abundance in Hep3B cells. This was associated with decreased PP2A activity and increased invasiveness of Hep3B cells across a Matrigel membrane; an effect mimicked by OA and silencing of the PP2A catalytic subunit. This is consistent with overexpression of CIP2A promoting cell invasion in lung adenocarcinoma (Sung et al., 2013), and silencing CIP2A decreasing MMP‐9 mRNA in osteosarcoma cells (Zhai et al., 2014). Importantly, the increase in cell invasiveness in the present study is likely due to MMP‐9 rather than MMP‐1‐, MMP‐2‐, MMP‐3‐ and MMP‐13‐mediated degradation of collagen type IV in the Matrigel basement membrane, as their expression was unaltered. Clinically, our data may in part explain why increased expression of CIP2A in many cancers including HCC is associated with increased tumour aggressiveness and poor prognosis (He et al., 2012; Yu et al., 2014).

In conclusion, an OA‐mediated reduction in PP2A activity increased MMP‐9 activity and expression in Hep3B cells. This was mediated through the PP2A catalytic (PPP2CA) subunit independent of the PPP2R1A scaffolding subunit and required functional AP‐1 binding motifs on the MMP‐9 promoter. The pharmacological effects of OA were mimicked by both siRNA knockdown of PPP2CA but not PPP1CA and by overexpression of CIP2A, an endogenous inhibitor of PP2A. This study directly links the loss of PP2A activity with an increase in cell invasion and ECM remodelling, with MMP‐9 being the predominant driving force. Our study indicates that other members of the MMP and ADAM family were not affected by pharmacological inhibition of PP2A, PP1 or PP2B. Interestingly, reduced phosphatase activity did not alter TGF‐β or IL‐1β, both of which are key in driving the inflammatory aspect of ECM remodelling. Other factors that influence tumour ECM remodelling such as the pro‐inflammatory chemokine IL‐8 and TIMP‐1 may too have a role to play in aiding MMP‐9‐mediated remodelling in Hep3B cells. The functional consequences of PP2A inhibition resulting in an increase in MMP‐9 highlight the role of PP2A in promoting HCC invasion through extracellular remodelling.

Author contributions

M.P.W. and J.P.S. designed the study and interpreted the data. M.P.W. performed the experiments and analysed the data. M.P.W. and J.P.S. wrote and revised the paper for intellectual content. All authors approved the paper for submission and agreed to be accountable for all aspects of the work.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organisations engaged with supporting research.

Acknowledgements

The authors acknowledge the generosity of Professor Jukka Westermarck, University of Turku, Finland, for the kind gift of the CIP2A plasmid and Professor Scott L. Friedman, Mount Sinai School of Medicine, New York, for donation of the LX‐2 hepatic cell line.

Ward, M. P. , and Spiers, J. P. (2017) Protein phosphatase 2A regulation of markers of extracellular matrix remodelling in hepatocellular carcinoma cells: functional consequences for tumour invasion. British Journal of Pharmacology, 174: 1116–1130. doi: 10.1111/bph.13759.

References

  1. Alexander SPH, Fabbro D, Kelly E, Marrion N, Peters JA, Benson HE et al. (2015). The Concise Guide to PHARMACOLOGY 2015/16: Enzymes. Br J Pharmacol 172: 6024–6109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Altadill A, Rodriguez M, Gonzalez LO, Junquera S, Corte MD, Gonzalez‐Dieguez ML et al. (2009). Liver expression of matrix metalloproteases and their inhibitors in hepatocellular carcinoma. Dig Liver Dis 41: 740–748. [DOI] [PubMed] [Google Scholar]
  3. Bonnans C, Chou J, Werb Z (2014). Remodelling the extracellular matrix in development and disease. Nat Rev Mol Cell Biol 15: 786–801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chin R, Earnest‐Silveira L, Koeberlein B, Franz S, Zentgraf H, Dong X et al. (2007). Modulation of MAPK pathways and cell cycle by replicating hepatitis B virus: factors contributing to hepatocarcinogenesis. J Hepatol 47: 325–337. [DOI] [PubMed] [Google Scholar]
  5. Cohen PT (2002). Protein phosphatase 1 – targeted in many directions. J Cell Sci 115: 241–256. [DOI] [PubMed] [Google Scholar]
  6. Cornell TT, Hinkovska‐Galcheva V, Sun L, Cai Q, Hershenson MB, Vanway S et al. (2009). Ceramide‐dependent PP2A regulation of TNFalpha‐induced IL‐8 production in respiratory epithelial cells. Am J Physiol Lung Cell Mol Physiol 296: L849–L856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Curtis MJ, Bond RA, Spina D, Ahluwalia A, Alexander SP, Giembycz MA et al. (2015). Experimental design and analysis and their reporting: new guidance for publication in BJP. Br J Pharmacol 172: 3461–3471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Demers M, Couillard J, Belanger S, St‐Pierre Y (2005). New roles for matrix metalloproteinases in metastasis. Crit Rev Immunol 25: 493–523. [DOI] [PubMed] [Google Scholar]
  9. Dhillon AS, Hagan S, Rath O, Kolch W (2007). MAP kinase signalling pathways in cancer. Oncogene 26: 3279–3290. [DOI] [PubMed] [Google Scholar]
  10. Duffy MJ, McKiernan E, O'Donovan N, McGowan PM (2009). Role of ADAMs in cancer formation and progression. Clin Cancer Res 15: 1140–1144. [DOI] [PubMed] [Google Scholar]
  11. Dung TD, Feng CC, Kuo WW, Pai P, Chung LC, Chang SH et al. (2013). Suppression of plasminogen activators and the MMP‐2/−9 pathway by a Zanthoxylum avicennae extract to inhibit the HA22T human hepatocellular carcinoma cell migration and invasion effects in vitro and in vivo via phosphatase 2A activation. Biosci Biotechnol Biochem 77: 1814–1821. [DOI] [PubMed] [Google Scholar]
  12. El Samanoudy A, Monir R, Badawy A, Ibrahim L, Farag K, El Baz S et al. (2014). Matrix metalloproteinase‐9 gene polymorphism in hepatocellular carcinoma patients with hepatitis B and C viruses. Genet Mol Res 13: 8025–8034. [DOI] [PubMed] [Google Scholar]
  13. El‐Serag HB (2011). Hepatocellular carcinoma. N Engl J Med 365: 1118–1127. [DOI] [PubMed] [Google Scholar]
  14. Esposito C, Fornoni A, Cornacchia F, Bellotti N, Fasoli G, Foschi A et al. (2000). Cyclosporine induces different responses in human epithelial, endothelial and fibroblast cell cultures. Kidney Int 58: 123–130. [DOI] [PubMed] [Google Scholar]
  15. Groblewski GE, Wagner AC, Williams JA (1994). Cyclosporin A inhibits Ca2+/calmodulin‐dependent protein phosphatase and secretion in pancreatic acinar cells. J Biol Chem 269: 15111–15117. [PubMed] [Google Scholar]
  16. Gu X, Fu M, Ding Y, Ni H, Zhang W, Zhu Y et al. (2014). TIMP‐3 expression associates with malignant behaviors and predicts favorable survival in HCC. PLoS One 9: e106161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gurol AO, Kursun AO, Suzergoz F, Kucuksezer UC, Kiran B, Kaya S et al. (2005). Peritransplant and long‐term secretion of interleukin‐1beta in cyclosporine treated syngeneic rats allografted with islets of langerhans. Transplant Proc 37: 2375–2378. [DOI] [PubMed] [Google Scholar]
  18. Ha E, Mun KC (2012). Effects of cyclosporine on metalloproteinase in endothelial cells. Transplant Proc 44: 991–992. [DOI] [PubMed] [Google Scholar]
  19. Hanahan D, Weinberg RA (2011). Hallmarks of cancer: the next generation. Cell 144: 646–674. [DOI] [PubMed] [Google Scholar]
  20. Haystead TA, Sim AT, Carling D, Honnor RC, Tsukitani Y, Cohen P et al. (1989). Effects of the tumour promoter okadaic acid on intracellular protein phosphorylation and metabolism. Nature 337: 78–81. [DOI] [PubMed] [Google Scholar]
  21. He X, Zhu Z, Johnson C, Stoops J, Eaker AE, Bowen W et al. (2008). PIK3IP1, a negative regulator of PI3K, suppresses the development of hepatocellular carcinoma. Cancer Res 68: 5591–5598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. He H, Wu G, Li W, Cao Y, Liu Y (2012). CIP2A is highly expressed in hepatocellular carcinoma and predicts poor prognosis. Diagn Mol Pathol 21: 143–149. [DOI] [PubMed] [Google Scholar]
  23. Heilker R, Freuler F, Pulfer R, Di Padova F, Eder J (1999). All three IkappaB isoforms and most Rel family members are stably associated with the IkappaB kinase 1/2 complex. Eur J Biochem 259: 253–261. [DOI] [PubMed] [Google Scholar]
  24. Hekmat O, Munk S, Fogh L, Yadav R, Francavilla C, Horn H et al. (2013). TIMP‐1 increases expression and phosphorylation of proteins associated with drug resistance in breast cancer cells. J Proteome Res 12: 4136–4151. [DOI] [PubMed] [Google Scholar]
  25. Hung MH, Chen YL, Chu PY, Shih CT, Yu HC, Tai WT et al. (2016). Upregulation of the oncoprotein SET determines poor clinical outcomes in hepatocellular carcinoma and shows therapeutic potential. Oncogene 35: 4891–4902. [DOI] [PubMed] [Google Scholar]
  26. Hyland PL, Traynor PS, Myrillas TT, Marley JJ, Linden GJ, Winter P et al. (2003). The effects of cyclosporin on the collagenolytic activity of gingival fibroblasts. J Periodontol 74: 437–445. [DOI] [PubMed] [Google Scholar]
  27. Irie A, Harada K, Araki N, Nishimura Y (2012). Phosphorylation of SET protein at Ser171 by protein kinase D2 diminishes its inhibitory effect on protein phosphatase 2A. PLoS One 7: e51242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Junttila MR, Puustinen P, Niemela M, Ahola R, Arnold H, Bottzauw T et al. (2007). CIP2A inhibits PP2A in human malignancies. Cell 130: 51–62. [DOI] [PubMed] [Google Scholar]
  29. Kessenbrock K, Plaks V, Werb Z (2010). Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 141: 52–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Latta EJ, Golding JP (2012). Regulation of PP2A activity by Mid1 controls cranial neural crest speed and gangliogenesis. Mech Dev 128: 560–576. [DOI] [PubMed] [Google Scholar]
  31. Leifler KS, Svensson S, Abrahamsson A, Bendrik C, Robertson J, Gauldie J et al. (2013). Inflammation induced by MMP‐9 enhances tumor regression of experimental breast cancer. J Immunol 190: 4420–4430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Li J, Zhang YP, Kirsner RS (2003). Angiogenesis in wound repair: angiogenic growth factors and the extracellular matrix. Microsc Res Tech 60: 107–114. [DOI] [PubMed] [Google Scholar]
  33. Liu WH, Chen YJ, Chien JH, Chang LS (2014). Amsacrine suppresses matrix metalloproteinase‐2 (MMP‐2)/MMP‐9 expression in human leukemia cells. J Cell Physiol 229: 588–598. [DOI] [PubMed] [Google Scholar]
  34. Lu PF, Weaver VM, Werb Z (2012). The extracellular matrix: a dynamic niche in cancer progression. J Cell Biol 196: 395–406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Meredith JE Jr, Fazeli B, Schwartz MA (1993). The extracellular matrix as a cell survival factor. Mol Biol Cell 4: 953–961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Neviani P, Santhanam R, Trotta R, Notari M, Blaser BW, Liu S et al. (2005). The tumor suppressor PP2A is functionally inactivated in blast crisis CML through the inhibitory activity of the BCR/ABL‐regulated SET protein. Cancer Cell 8: 355–368. [DOI] [PubMed] [Google Scholar]
  37. Niemela M, Kauko O, Sihto H, Mpindi JP, Nicorici D, Pernila P et al. (2012). CIP2A signature reveals the MYC dependency of CIP2A‐regulated phenotypes and its clinical association with breast cancer subtypes. Oncogene 31: 4266–4278. [DOI] [PubMed] [Google Scholar]
  38. Perrotti D, Neviani P (2008). Protein phosphatase 2A (PP2A), a drugable tumor suppressor in Ph1(+) leukemias. Cancer Metastasis Rev 27: 159–168. [DOI] [PubMed] [Google Scholar]
  39. Perz JF, Armstrong GL, Farrington LA, Hutin YJ, Bell BP (2006). The contributions of hepatitis B virus and hepatitis C virus infections to cirrhosis and primary liver cancer worldwide. J Hepatol 45: 529–538. [DOI] [PubMed] [Google Scholar]
  40. Pratt J, Annabi B (2014). Induction of autophagy biomarker BNIP3 requires a JAK2/STAT3 and MT1‐MMP signaling interplay in concanavalin‐A‐activated U87 glioblastoma cells. Cell Signal 26: 917–924. [DOI] [PubMed] [Google Scholar]
  41. Pshenichkin SP, Wise BC (1997). Okadaic acid stimulates nerve growth factor production via an induction of interleukin‐1 in primary cultures of cortical astroglial cells. Neurochem Int 30: 507–514. [DOI] [PubMed] [Google Scholar]
  42. Qian NS, Xu XP, Chen Y, Dou KF, Wang YY, Yang YL (2010). Protective effect of cyclosporin A on brain injury in rats with acute necrotic pancreatitis. Life Sci 87: 64–68. [DOI] [PubMed] [Google Scholar]
  43. Ray A, Bal BS, Ray BK (2005). Transcriptional induction of matrix metalloproteinase‐9 in the chondrocyte and synoviocyte cells is regulated via a novel mechanism: evidence for functional cooperation between serum amyloid A‐activating factor‐1 and AP‐1. J Immunol 175: 4039–4048. [DOI] [PubMed] [Google Scholar]
  44. Ren Y, Poon RT, Tsui HT, Chen WH, Li Z, Lau C et al. (2003). Interleukin‐8 serum levels in patients with hepatocellular carcinoma: correlations with clinicopathological features and prognosis. Clin Cancer Res 9: 5996–6001. [PubMed] [Google Scholar]
  45. Rietz A, Volkov Y, Davies A, Hennessy M, Spiers JP (2012). Okadaic acid induces matrix metalloproteinase‐9 expression in fibroblasts: crosstalk between protein phosphatase inhibition and beta‐adrenoceptor signalling. Br J Pharmacol 165: 274–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sahin U, Weskamp G, Kelly K, Zhou HM, Higashiyama S, Peschon J et al. (2004). Distinct roles for ADAM10 and ADAM17 in ectodomain shedding of six EGFR ligands. J Cell Biol 164: 769–779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Schneider CA, Rasband WS, Eliceiri KW (2012). NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9: 671–675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Shanley TP, Vasi N, Denenberg A, Wong HR (2001). The serine/threonine phosphatase, PP2A: endogenous regulator of inflammatory cell signaling. J Immunol 166: 966–972. [DOI] [PubMed] [Google Scholar]
  49. Sobral LM, Sousa LO, Coletta RD, Cabral H, Greene LJ, Tajara EH et al. (2014). Stable SET knockdown in head and neck squamous cell carcinoma promotes cell invasion and the mesenchymal‐like phenotype in vitro, as well as necrosis, cisplatin sensitivity and lymph node metastasis in xenograft tumor models. Mol Cancer 13: 32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sonoda Y, Kasahara T, Yamaguchi Y, Kuno K, Matsushima K, Mukaida N (1997). Stimulation of interleukin‐8 production by okadaic acid and vanadate in a human promyelocyte cell line, an HL‐60 subline. Possible role of mitogen‐activated protein kinase on the okadaic acid‐induced NF‐kappaB activation. J Biol Chem 272: 15366–15372. [DOI] [PubMed] [Google Scholar]
  51. Southan C, Sharman JL, Benson HE, Faccenda E, Pawson AJ, Alexander SP et al. (2016). The IUPHAR/BPS guide to PHARMACOLOGY in 2016: towards curated quantitative interactions between 1300 protein targets and 6000 ligands. Nucleic Acids Res 44: D1054–D1068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Stebbing J, Lit LC, Zhang H, Darrington RS, Melaiu O, Rudraraju B et al. (2014). The regulatory roles of phosphatases in cancer. Oncogene 33: 939–953. [DOI] [PubMed] [Google Scholar]
  53. Sung WW, Wang YC, Lin PL, Cheng YW, Chen CY, Wu TC et al. (2013). IL‐10 promotes tumor aggressiveness via upregulation of CIP2A transcription in lung adenocarcinoma. Clin Cancer Res 19: 4092–4103. [DOI] [PubMed] [Google Scholar]
  54. Swingle M, Ni L, Honkanen RE (2007). Small‐molecule inhibitors of ser/thr protein phosphatases: specificity, use and common forms of abuse. Methods Mol Biol 365: 23–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wakabayashi K, Kambe F, Cao X, Murakami R, Mitsuyama H, Nagaya T et al. (2004). Inhibitory effects of cyclosporin A on calcium mobilization‐dependent interleukin‐8 expression and invasive potential of human glioblastoma U251MG cells. Oncogene 23: 6924–6932. [DOI] [PubMed] [Google Scholar]
  56. Wang L, Clutter S, Benincosa J, Fortney J, Gibson LF (2005). Activation of transforming growth factor‐beta1/p38/Smad3 signaling in stromal cells requires reactive oxygen species‐mediated MMP‐2 activity during bone marrow damage. Stem Cells 23: 1122–1134. [DOI] [PubMed] [Google Scholar]
  57. Westermarck J, Kahari VM (1999). Regulation of matrix metalloproteinase expression in tumor invasion. FASEB J 13: 781–792. [PubMed] [Google Scholar]
  58. Westermarck J, Lohi J, Keski‐Oja J, Kahari VM (1994). Okadaic acid‐elicited transcriptional activation of collagenase gene expression in HT‐1080 fibrosarcoma cells is mediated by JunB. Cell Growth Differ 5: 1205–1213. [PubMed] [Google Scholar]
  59. Westermarck J, Seth A, Kahari VM (1997). Differential regulation of interstitial collagenase (MMP‐1) gene expression by ETS transcription factors. Oncogene 14: 2651–2660. [DOI] [PubMed] [Google Scholar]
  60. Westermarck J, Li SP, Jaakkola P, Kallunki T, Grenman R, Kahari VM (2000). Activation of fibroblast collagenase‐1 expression by tumor cells of squamous cell carcinomas is mediated by p38 mitogen‐activated protein kinase and c‐Jun NH2‐terminal kinase‐2. Cancer Res 60: 7156–7162. [PubMed] [Google Scholar]
  61. Westermarck J, Li SP, Kallunki T, Han J, Kahari VM (2001). p38 mitogen‐activated protein kinase‐dependent activation of protein phosphatases 1 and 2A inhibits MEK1 and MEK2 activity and collagenase 1 (MMP‐1) gene expression. Mol Cell Biol 21: 2373–2383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Whale A, Hashim FN, Fram S, Jones GE, Wells CM (2011). Signalling to cancer cell invasion through PAK family kinases. Front Biosci (Landmark Ed) 16: 849–864. [DOI] [PubMed] [Google Scholar]
  63. Yu HC, Chen HJ, Chang YL, Liu CY, Shiau CW, Cheng AL et al. (2013). Inhibition of CIP2A determines erlotinib‐induced apoptosis in hepatocellular carcinoma. Biochem Pharmacol 85: 356–366. [DOI] [PubMed] [Google Scholar]
  64. Yu HC, Hung MH, Chen YL, Chu PY, Wang CY, Chao TT et al. (2014). Erlotinib derivative inhibits hepatocellular carcinoma by targeting CIP2A to reactivate protein phosphatase 2A. Cell Death Dis 5: e1359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Zhai M, Cong L, Han Y, Tu G (2014). CIP2A is overexpressed in osteosarcoma and regulates cell proliferation and invasion. Tumour Biol 35: 1123–1128. [DOI] [PubMed] [Google Scholar]

Articles from British Journal of Pharmacology are provided here courtesy of The British Pharmacological Society

RESOURCES