ORIGINAL ARTICLE

Telmisartan attenuates N-nitrosodiethylamine-inducedhepatocellular carcinoma in mice by modulatingthe NF-κB-TAK1-ERK1/2 axis in the context of PPARγ agonistic activity

Sameh Saber1 & Ahmed E. Khodir1 & Wafaa E. Soliman2,3& Mohamed M. Salama4 & Walied S. Abdo5

& Baraah Elsaeed4&

Karim Nader4 & Aya Abdelnasser4 & Nada Megahed4& Mohamed Basuony4 & Ahmed Shawky4 & Maryam Mahmoud4

&

Reham Medhat4 & Abdelrahman S. Eldin4

Received: 15 April 2019 /Accepted: 23 July 2019# Springer-Verlag GmbH Germany, part of Springer Nature 2019

AbstractHepatocellular carcinoma (HCC) is characterized by bad prognosis and is the second most common reason for cancer-linkedmortality. Treatment with sorafenib (SRF) alone increases patient survival by only a few months. A causal link has beendetermined between angiotensin II (Ang-II) and HCC. However, the mechanisms underlying the tumorigenic effects of Ang-IIremain to be elucidated. N-Nitrosodiethylamine was utilized to examine the effects of telmisartan (TEL) (15 mg/kg), SRF(30 mg/kg), and a combination of these two agents on HCC mice. Downregulation of NF-кBP65 mRNA expression andinhibition of the phosphorylation-induced activation of both ERK1/2 and NF-кB P65 were implicated in the anti-tumor effectsof TEL and SRF. Consequent regression of malignant changes and improvements in liver function associated with reduced levelsof AFP, TNF-α, and TGF-β1 were also confirmed. Anti-proliferative, anti-metastatic, and anti-angiogenic effects of treatmentwere indicated by reduced hepatic cyclin D1 mRNA expression, reduced MMP-2 levels, and reduced VEGF levels, respectively.TEL, but not SRF, demonstrated agonistic activity for PPARγ receptors, as evidenced by increased PPARγ DNA bindingactivity, upregulation of CD36, and HO-1 mRNA expression followed by increased liver antioxidant capacity. Both TEL andSRF inhibited TAK1 phosphorylation-induced activation, indicating that TAK1 might act as a central mediator in the interactionbetween ERK1/2 and NF-кB. TEL, by modulating the ERK1/2, TAK1, and NF-кB signaling axis in the context of PPARγagonistic activity, exerted anti-tumor effects and increased tumor sensitivity to SRF. Therefore, TEL is an encouraging agent forfurther clinical trials regarding the management of HCC.

Keywords Telmisartan . N-Nitrosodiethylamine . PPARγ . ERK1/2 . TAK1 . NF-кB

AbbreviationsCD36 Cluster of differentiation 36ERK Extracellular signal-regulated kinaseGST-P Placental glutathione S-transferaseHO-1 Heme oxygenase 1Iκ-Bα Nuclear factor kappa-B inhibitor alphaIKK IkB kinaseMAPK Mitogen-activated protein kinaseMEK Mitogen-activated protein kinaseMTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl

tetrazolium bromideNFκB Nuclear transcription factor kappa-BPPARγ Peroxisome proliferator-activated receptor gammaTAK1 Transforming growth factor beta-activated kinase 1

* Sameh SaberSampharm81@gmail.com; sameh.mohamed@deltauniv.edu.eg

1 Department of Pharmacology, Faculty of Pharmacy, Delta Universityfor Science and Technology, Costal International Road, GamasaCity, Mansoura, Egypt

2 Department of Microbiology and Biotechnology, Faculty ofPharmacy, Delta University for Science and Technology,Gamasa, Egypt

3 Department of Biomedical Science, Faculty of Clinical Pharmacy,King Faisal University, Al-Ahsa, Saudi Arabia

4 Department of Biochemistry, Faculty of Pharmacy, Delta Universityfor Science and Technology, Gamasa, Egypt

5 Department of Pathology, Faculty of Veterinary Medicine,Kafrelsheikh University, Kafrelsheikh, Egypt

Naunyn-Schmiedeberg's Archives of Pharmacologyhttps://doi.org/10.1007/s00210-019-01706-2

Introduction

Hepatocellular carcinoma (HCC) has bad prognosis and isconsidered the second most common reason for cancer-linked mortality (Dhanasekaran et al. 2016). Different riskfactors include viral hepatitis, ingesting foods contaminatedwith chemical carcinogens, and exposure to aflatoxins(Shlomai et al. 2014). During the course of HCC, varioussignaling pathways affecting cell proliferation and neovascu-larization are deregulated, including nuclear factor kappa-B(NF-кB) (Saber et al. 2018c) and extracellular signal-regulated kinase (ERK1/2) pathways (Ranjan et al. 2018). Inaddition, a link has been determined between liver inflamma-tion and HCC. NF-κB pathway activation results in gene tran-scription of many pro-inflammatory cytokines, such as tumornecrosis factor-alpha (TNF-α) (Saber et al. 2018a; Saber et al.2019), which is responsible for liver injury prior to the devel-opment of HCC. Moreover, several studies have indicated anessential role of transforming growth factor-beta (TGF-β) inthe pathogenesis of HCC (Zheng et al. 2013).

Many reports have suggested that renin-angiotensin system(RAS) components are major contributing factors to differenttypes of malignancy. Indeed, upregulation of angiotensin(Ang)-II type I receptor (AT1R) expression has been sug-gested to be an essential step in the pathogenesis of breasthyperplasia (De Paepe et al. 2002). The first evidence thatAT1R antagonists can inhibit breast tumor angiogenesis wasreported by Herr et al. (Herr et al. 2008), who usedcandesartan as an anti-angiogenic agent. Additionally,candesartan treatment effectively inhibits the development oflung metastatic nodules (Miyajima et al. 2002). Furthermore,it has been reported that Ang-II inhibition dramatically in-hibits the proliferation of prostate cancer cells in vitro(Ishiguro et al. 2007), decreases the possibility of developingesophageal cancer (Sjoberg et al. 2007), and reduces the inci-dence of keratinocyte cancers (Christian et al. 2008).

Telmisartan (TEL), an AT1R antagonist that has been ap-proved for the treatment of hypertension, shares structuralsimilarities with the peroxisome proliferator-activated recep-tor gamma (PPARγ) ligand pioglitazone (Koyama et al.2014). TEL displays anti-proliferative activity in prostate can-cer (Funao et al. 2008), kidney cancer (Funao et al. 2009),endometrial cancer (Ota et al. 2006), and colon cancer (Leeet al. 2014). However, the underlying mechanism remainsobscure.

The complex molecular pathogenesis and delayed diagno-sis of HCC affect the possibility for effective treatment.Sorafenib (SRF) chemotherapy is the main and initial therapyfor patients with HCC. However, incidence and mortality ofHCC patients treated with SRF are very similar to those ofuntreated patients (Brito et al. 2016). Therefore, a serious needpresents to explore and evaluate potential alternative strategiesto minimize resistance, improve prognosis, and guarantee

effective control of HCC. The effect of TEL on HCC hasnot yet been investigated. Hence, the current research wasdesigned to examine the anti-tumor activity of TEL and acombination of TEL and SRF on N-nitrosodiethylamine(NDEA)-induced HCC to get a deep vision into the potentialmechanisms underlying the action of TEL.

Materials and methods

Evaluation of the in vitro anti-proliferative activityof TEL on HepG2 cells

A simple MTT assay was used to investigate the anti-proliferative activity of TEL on the HepG2 cell line(American Type Culture Collection (ATCC), HB-8065),which is an immortalized human liver carcinoma cell line.HepG2 cells were grown and maintained at 37 °C inDulbecco’s modified Eagle’s medium (DMEM) containing10% FBS, 2 mM glutamine, 100 IU/ml penicillin, and100 μg/ml streptomycin as the culture medium. Sub-culturing was performed in 96-well plates at a density of 2 ×104 cells/well in 100 μL of culture medium in a humidifiedatmosphere of 5% CO2. After 24 h, the medium was replacedwith DMEM (100 μL/well) containing 1% DMSO. Differentconcentrations (0.5, 1.0, 5.0, 10.0, 50.0, and 100.0 μmol/L) ofTEL pre-dissolved in DMSOwere supplemented to each well,and the cells were incubated for 48 h. Each concentration ofTEL was examined in triplicate.

Cytotoxic activity was evaluated as an indicator of anti-proliferative activity. After the medium was discarded and50 μL of MTT in DMEM was supplemented to each well,the cells were incubated for 3 h. Then, the medium wasdiscarded, and 50 μL of propanol was supplemented to eachwell. The microplates were then incubated for 30 min. Theabsorbance was detected at 540 nm. As described by Saberet al. (2018a), the percent growth inhibition and the percentcell viability were calculated to determine anti-proliferativeactivity by approximately estimating the half-maximal cyto-toxic concentration (CTC50; μmol/L).

Animals

One hundred and twenty-one male Swiss albino mice of CD-1strain (20 ± 2 g) aging 4–6 weeks were purchased from theanimal facility at Delta University, Egypt. The mice were fedstandard diet and given access to water ad libitum. Standardconditions of 21 °C with 45–55% humidity and 12 h:12 hlight:dark cycles were maintained. The mice were acclima-tized for 1 week before induction of the experiment. All pro-cedures followed the ARRIVE guidelines and the EUDirective 2010/63/EU for animal experiments and compliedwith standards equivalent to the UKCCCR guidelines for the

Naunyn-Schmiedeberg's Arch Pharmacol

welfare of animals in experimental neoplasia (Br J Cancer 58:109–113, 1988). In addition, all procedures were approved bythe FPDU/IACUC, approval no: FPDU1/2018/2.

Experimental design

The mice were randomly assigned to five groups. Group 1(CTRL) mice administered intraperitoneal (i.p.) injections of5 ml/kg normal saline solution weekly (n = 20). Group 2(NDEA) mice administered i.p. injections of NDEA (Sigma-Aldrich, St. Louis, MO, USA) as previously described bySaber et al. (2018c) (n = 41). Group 3 (SRF) mice adminis-tered i.p. injections of NDEA + SRF (30 mg/kg) (n = 19)(Bayer, Berlin, Germany). Group 4 (TEL) mice administeredi.p. injections of NDEA + TEL (15 mg/kg) (n = 21)(Boehringer Ingelheim, Berlin, Germany). Group 5 (SRF +TEL) mice administered i.p. injections of NDEA + SRF(30 mg/kg) + TEL (15 mg/kg) (n = 20). NDEA was dilutedto 1% in normal saline solution, while the other drugs weresuspended in distilled water immediately before use. The drugdoses were administered once/day by oral tube on the 45thday and continued for up to the 120th day (Table 1).

Rationale for drug dosing

The dose of SRF was selected dependent on a previous study(Saber et al. 2018b). One log concentration below the calcu-lated CTC50 is approximately 5 μmol/L. Notably, plasma con-centrations of 5 μmol/LTEL can be achieved by conventionaloral dosing (Stangier et al. 2000). Therefore, the conventionalclinical dose of 80 mg of TEL was used to calculate the equiv-alent mouse dose using approximate conversion factors de-scribed by Sharma and McNeill (2009).

Determination of alpha-fetoprotein (AFP), alanineaminotransferase (ALT), aspartate aminotransferase(AST), alkaline phosphatase (ALP), catalase, and totalantioxidant capacity

The serum levels of AFP were measured using a QuantikineELISA kit (R&D Systems, Minneapolis, MN). Commercialkits provided by Biodiagnostic (Giza, Egypt) were used tomeasure serum levels of ALT, AST and ALP, hepatic levelsof catalase, and total antioxidant capacity as instructed.

Determination of TNF-α, TGF-β, matrixmetalloproteinase-2 (MMP-2), and vascularendothelial growth factor (VEGF) levels

The serum levels of TNF-α, TGF-β, and MMP-2 and thehepatic levels of VEGF were measured by ELISA using aQuantikine ELISA kit (R&D Systems) as instructed.

Determination of total NF-кB p65and phosphorylated p65 (NF-кBp-p65) and totalERK1/2 and phosphorylated ERK1/2 (p-ERK1/2) levels

A 1× cell extraction buffer PTR was prepared as per manu-facturer’s instructions. Liver tissues were homogenized in achilled 1× cell extraction buffer PTR, then incubated on ice for20 min followed by 20 min of centrifugation at 18000×g and4 °C. Pellets were discarded and supernatants were stored at −80 °C for subsequent quantitative measurement of ERK1/2(pT202/Y204) and total ERK1/2 protein using a simple stepAbcam ELISA kit (ab176660) (Cambridge, MA, USA).Ultimately, the generated signal is proportional to the amountof bound analyte (P-ERK1/2 or total ERK1/2) and the inten-sity is measured at 450 nm. Similarly, a simple step Abcam

Table 1 Experimental designExp. groups Days (1:45) Days (46:120)

CTRL group I.P. injection of saline solution once a week I.P. injection of saline solution once a week

NDEA group I.P. injection of NDEA once a week I.P. injection of NDEA once a week

SRF group I.P. injection of NDEA once a week I.P. injection of NDEA once a week

+

SRF (30 mg/kg/day p.o.)

TEL group I.P. injection of NDEA once a week I.P. injection of NDEA once a week

+

TEL (15 mg/kg/day p.o.)

TEL+SRF group I.P. injection of NDEA once a week I.P. injection of NDEA once a week

+

TEL (15 mg/kg/day p.o.)

+

SRF (30 mg/kg/day p.o.)

NDEA, N-nitrosodiethylamine; I.P., intraperitoneal; p.o., per oral; SRF, sorafenib; TEL, telmisartan

Naunyn-Schmiedeberg's Arch Pharmacol

ELISA kit (ab176663) was used for the determination ofNF-ĸB p65 (pS536) + total NF-ĸB p65 utilizing the sameextraction buffer that is provided in the kit in the way asdescribed by the supplier. Phosphorylated NF-ĸB p65 andphosphorylated ERK1/2 were normalized to total NFĸB p65and total ERK1/2, respectively, and normalized by the controlvalue.

Western blot analysis of the expressionof TGF-β-activated kinase 1 (TAK1) in liver tissue

Western blot analysis of the expression of TGF-β-activatedkinase 1 (TAK1) in liver tissue was performed according tomethods previously described by Saber et al. (2018a) usingRIPA lysis buffer (Bio Basic, Inc., Markham, Ontario,Canada), 10% SDS-PAGE (Bio-Rad Laboratories, Inc.,USA), PVDF membranes (Bio-Rad Laboratories, Inc.),monoclonal antibodies specific for TAK1, p-TAK1 and glyc-eraldehyde 3-phosphate dehydrogenase (GAPDH), and rabbitanti-mouse secondary antibody conjugated to horseradish per-oxidase (Thermo Fisher Scientific, Inc., USA). As a final step,image analysis was performed to determine the band intensityof the p-TAK1 and TAK-1 proteins against that of the controlsample after normalization by GAPDH.

Quantitative real-time PCR analysis of the expressionof NF-кB p65 mRNA, cyclin D1 mRNA, hemeoxygenase 1 (HO-1) mRNA and clusterof differentiation 36 (CD36) mRNA in liver tissue

Total RNAwas obtained from hepatic tissues using a Qiagenkit (Venlo, Netherlands) according to the manufacturer’s in-structions. Complementary DNAwas synthesized using 1 μgof total RNA and Moloney Murine Leukemia Virus ReverseTranscriptase (M-MLV RT) (Promega, Madison, WI, USA).The purity (A260/A280 ratio) and the concentration of RNAwere obtained using spectrophotometry (GeneQuant 1300,Uppsala, Sweden), and RT-qPCR was performed in real timeusing a StepOnePlus Real-Time PCR System (AppliedBiosystems, Foster City, CA, USA) and SYBR Green PCRmaster mix (Applied Biosystems, Foster City, CA, USA). Thefollowing PCR primer pair sequences were used: NF-кB p65,F: 5′-CTTCCTCAGCCATGGTACCTCT-3′ and R: 5′-CAAGTCTTCATCAGCATCAAACTG-3′; cyclin D1, F: 5′-CCAACAACTTCCTCTCCTGCT-3 ′ and R : 5 ′ -GACTCCAGAAGGGCTTCAATC-3′); HO-1, F: 5 ′-AGCCCCACCAAGTTCAAACA-3 ′ a nd R : 5 ′ - CATCACCTGCAGCTCCTCAA-3 ′; GAPDH, F: 5 ′-AATGGGAAGCTTGTCATCAA- 3 ′ a n d R : 5 ′ - TACTTGGCAGGTTTCTCCAG-3′; and CD36, F: 5′-TCCTCTGACATTTGCAGGTCTATC-3′ and R: 5′-AAAGGCATTGGCTGGAAGAA-3′. The thermal cycling condi-tions involved an initial denaturation step at 95 °C for

10 min followed by 40 cycles of 94 °C for 30 s (denaturation),63 °C for 30 s (annealing), and 72 °C for 30 s (extension). Afinal extension was carried out at 72 °C for 5 min. Relativeexpressions were determined as previously described (Saberet al. 2018c).

Determination of PPARγ DNA binding activity

An ELISA kit provided by Abcamwas used for the estimationof PPARγ activity according to the manufacturer ’sinstructions.

Histological and immunohistochemical examination

Liver tissues from neutral buffered formalin were fixed inparaffin. Sections (4–5 μm thick) from the paraffin blockswere stained with hematoxylin and eosin (H&E) for histolog-ical analysis. HCC was classified as previously reported(Bosman et al. 2010). Immunostaining of placental glutathi-one S-transferase (GST-P) was performed according to theprevious description by (Abdo et al. 2013; Abdo et al. 2015)using a polyclonal rabbit anti-mouse GST-P antibody(Medical & Biological Laboratories Co., Ltd., Nagoya,Japan; 1:500). The GST-P expression was presented as thenumbers of foci/cm2 and areas mm2/cm2.

Statistical analysis

The data are presented as the mean ± standard deviation (SD).Statistical analysis was done by the GraphPad software ver-sion 6 (CA, USA). One-way analysis of variance (ANOVA)followed by Tukey’s multiple comparisons test was used toanalyze differences between groups. The Kaplan-Meier meth-od was used for construction of survival curves, and theMantel-Cox method was used to assess the significance ofdifferences. A p value of < 0.05 was considered significant.

Results

In vitro cytotoxicity of TEL towards HepG2 cells

As shown in Table 2, TEL exhibited anti-proliferative effectson HepG2 cells, and the CTC50 value was estimated to beapproximately 50 μmol/L.

Histopathological and immunohistochemicalexamination

Tissue sections from the normal CTRL group displayed nor-mal histology (Fig. 1a). On the other hand, liver sections fromthe NDEA-treated mice (Fig. 1b, c) displayed loss of hepaticlobular architecture, intralobular inflammatory cell infiltration

Naunyn-Schmiedeberg's Arch Pharmacol

and fibrous tissue in addition to malignant giant hepatocyteswith atypical nuclei, darkly stained nuclei due to increasedDNA content, increased nucleo-cytoplasmic ratios, dysplasia,and pleomorphism. Administration of TEL (Fig. 1e) improvedliver histology; the sections from TEL-treated mice showedmodest reversion of malignancy and were comparable with

liver tissue sections from mice treated with SRF (Fig. 1d).Sections from mice treated with the combination of TEL andSRF (Fig. 1f) displayed marked histological improvement andregression of malignancy. GST-P immunolabeling was per-formed to assess the difference between groups on quantita-tive scale of both foci areas and numbers (Fig. 2f). Liversections from control animals were negative to GST-P exceptpositive few bile duct lining epithelia (Fig. 2a). The DEN-treated animals showed marked increase in the GST-P fociarea and number (Fig. 2b). In addition, sections from TheSRF- (Fig. 2c) and TEL-treated animals (Fig. 2d) revealeddecrease in the number and area of altered foci. Meanwhile,the combination of TEL and SRF revealed marked decrease inGST-P-positive hepatic cells (Fig. 2e).

Effect on serum AFP level, relative liver weight,and the number of surface nodules/liver

AFP serum levels were significantly higher in NDEA-treatedmodel mice than in CTRLmice, but animals treated with SRF,

Fig. 1 Histological examination of liver tissues. Shown are representativelight micrographs from a normal CTRL samples showing normal hepatichistology with radial arrangement of hepatocyte rays around a centralvein (arrows) that are separated by sinusoids (H&E, 200x) and b, cNDEA-treated mouse samples showing HCC with loss of lobular archi-tecture, hepatocellular atypia, pleomorphism (arrows), and intralobular

inflammatory cell infiltration (arrowhead) (H&E, 200× (b) and 400×(c)). Shown are micrographs from d SRF-treated mouse samples (H&E,200×) and e TEL-treated mouse samples (H&E, 200×) showing moderateregression of HCC and from f TEL+SRF-treated mouse samples showingmarked regression of HCC (H&E, 200×). Also shown are morphometricpictures of livers from different treatment groups (g)

Table 2 The anti-proliferative activity of TEL on HepG2

TEL (μmol/L) OD540 ± SD % growth inhibition % cell viability

0 0.648 ± 0.09 0 100

0.5 0.642 ± 0.09 0.93 99.07

1 0.618 ± 0.07 4.63 95.37

5 0.501 ± 0.06 22.69 77.31

10 0.483 ± 0.05 65.8 74.53

50 0.303 ± 0.05 53.3 46.7

100 0.213 ± 0.05 67.13 32.87

CTC50 is approximately 50 μmol/L

CTC, cytotoxic concentration; TEL, telmisartan; OD, optical density

Naunyn-Schmiedeberg's Arch Pharmacol

TEL, or the combination of SRF and TEL demonstrated signif-icantly lower AFP levels than NDEA-only mice. AFP serumlevels in the combination therapy group were significantly dif-ferent than those in the TEL treatment group. In addition, treat-ment of mice with SRF, TEL, or the combination of SRF andTEL significantly attenuatedNDEA-induced increases in relativeliver weight. Furthermore, mice treated with SRF, TEL, or thecombination of SRF and TEL exhibited significantly fewer sur-face nodules/liver than NDEA-only mice (Table 3).

Effect on ALT, AST, ALP, catalase levels, and totalantioxidant capacity

As depicted in Fig. 3a, serum ALT, AST, and ALP levels weresignificantly higher in NDEA-treated model mice than inCTRL mice. However, animals treated with SRF, TEL, orthe combination of SRF and TEL exhibited significantly low-er levels of these markers than NDEA-only mice. In addition,

significantly lower levels of ALT and ALP were observed inthe combination therapy group than in the monotherapygroups. Significantly lower catalase levels (Fig. 3b) and totalantioxidant capacity (Fig. 3c) were measured in the NDEA-treated model mice than in the CTRL mice. Mice receivingTEL or combination therapy demonstrated significantlyhigher levels than the NDEA-only group, while the combina-tion therapy group demonstrated significantly higher levelsthan the SRF-treated group. However, a trend towards in-creased levels of both catalase (p = 0.08) and total antioxidantcapacity (p = 0.07) was observed in SRF-treated mice com-pared with NDEA-only mice.

Effect on TNF-α, TGF-β, MMP-2, and VEGF levels

Serum levels of TNF-α (Fig. 4a), TGF-β (Fig. 4b), andMMP-2 (Fig. 4c) and hepatic levels of VEGF (Fig. 4d) were signif-icantly higher in NDEA-treated model mice than in CTRL

Fig. 2 Immunohistochemicaldetection of GST-P in liver tis-sues. Shown are representativelight micrographs from a normalCTRL samples showing negativeGST-P immunolabeling and bNDEA-treated mouse samplesshowing marked increase in GST-P foci number and area. Alsoshown are micrographs from cSRF-treated mouse samplesshowing decrease in GST-P focinumber and area d TEL-treatedmouse samples showing decreasein GST-P foci number and areathat was comparable with that ofSRF-treated mice and from eTEL+SRF-treated mouse samplesshowing marked decrease inGST-P-positive hepatic cells. fShown are the effects of SRF(30 mg/kg), TEL (15 mg/kg) anda combination of both on area %and number of GST-P foci. Thedata are expressed as the mean ±SD (n = 6). Statistical analysiswas performed using ordinaryone-way ANOVA followed byTukey’s post-test. +p < 0.05 vs.CTRL, ++++p < 0.0001 vs. CTRL,***p < 0.001 vs. NDEA,****p < 0.0001 vs. NDEA,@@p < 0.01 vs. NDEA+SRF,@@@p < 0.001 vs. NDEA+SRF,###p < 0.001 vs. NDEA+TEL,####p < 0.0001 vs. NDEA+TEL.scale bar, 50 μm

Naunyn-Schmiedeberg's Arch Pharmacol

mice. Mice treated with SRF, TEL, or the combination of SRFand TEL exhibited significantly lower TGF-β, TNF-α, MMP-2, and VEGF levels than NDEA-only mice. In addition, sig-nificant differences were observed in the levels of these pro-teins between the combination therapy group and the mono-therapy groups.

Effect on total NF-кB p65, NF-кBp-p65, TAK1, totalERK1/2, and p-ERK1/2 levels in liver tissue

Drug treatment significantly inhibited NDEA-induced in-creases in the ratios of p-p65 to p65 (Fig. 5a), p-TAK1 toTAK1 (Fig. 5c), and p-ERK1/2 to ERK1/2 (Fig. 5b), exceptin the TEL-treated group, which showed a trend towards asignificant reduction (p = 0.06) in the p-ERK1/2:ERK1/2 ratiocompared with the NDEA-only group. In addition, the com-bination treatment group exhibited significantly lower p-p65:p65, p-TAK1:TAK1, and p-ERK1/2:ERK1/2 ratios thanthe monotherapy groups.

Effect on the expression of NF-кB p65 mRNA, cyclinD1mRNA, HO-1mRNA, and CD36mRNA in liver tissue

Mice treated with NDEA had significantly higher mRNAexpression levels of NF-ĸB p65 (Fig. 6a) and cyclin D1(Fig. 6b) than CTRL mice. However, SRF and TEL treat-ment significantly attenuated the NDEA-induced upregu-lation of NF-ĸB p65 and cyclin D1 mRNA expression. Onthe other hand, mice treated with NDEA had significantlylower mRNA expression levels of HO-1 (Fig. 6c) andCD36 (Fig. 6d) mRNA expression than CTRL mice. TELtreatment significantly attenuated the NDEA-induceddownregulation of HO-1 and CD36 mRNA expression,while SRF-treated mice exhibited a trend towards signifi-cantly higher HO-1 mRNA expression (p = 0.06) thanNDEA-only mice. Moreover, the SRF-treated group didnot exhibit a significant difference in CD36 expressioncompared with the NDEA-treated model group and dem-onstrated a trend towards a significant decrease (p = 0.06)compared with the CTRL group.

Fig. 3 Effects of SRF (30 mg/kg), TEL (15mg/kg), and a combination ofboth on a serum alanine aminotransferase (ALT), aspartateaminotransferase (AST), and alkaline phosphatase (ALP) levels; b cata-lase levels; and c total antioxidant capacity inmice treatedwith NDEA for16 weeks. The data are expressed as the mean ± SD (n = 4). Statistical

analysis was performed using ordinary one-way ANOVA followed byTukey’s post-test. +p < 0.05 vs. CTRL, ++p < 0.01 vs. CTRL,++++p < 0.0001 vs. CTRL, **p < 0.01 vs. NDEA, ***p < 0.001 vs.NDEA, ****p < 0.0001 vs. NDEA, @p < 0.05 vs. NDEA+SRF,#p < 0.05 vs. NDEA+TEL

Table 3 Effect of SRF (30mg/kg), TEL (15mg/kg), and their combination on relative liver weight, serum level of alpha-fetoprotein (AFP), and surfacenodules/liver in NDEA-intoxicated mice for 16 weeks

Groups Liver weight (g) Body weight (BW) (g) Relative liver weight (g/g BW) AFP (ng/ml) Surface nodules/liver

CTRL 1.01 ± 0.14 35.16 ± 2.04 2.33 ± 0.26 2.12 ± 0.35 0

NDEA 1.651 ± 0.17 27.5 ± 1.87 5.24 ± 0.22++++ 8.64 ± 1.2++++ 4 ± 1.26++++

NDEA+SRF 1.26 ± 0.15 30.16 ± 1.94 3.80 ± 0.74++++,*** 4.14 ± 0.9+,**** 0.83 ± 0.75****

NDEA+TEL 1.3 ± 0.19 28.3 ± 3.50 3.45 ± 0.46++,**** 4.76 ± 0.92++,*** 1.16 ± 0.752****

NDEA+TEL+SRF

1.14 ± 0.12 34.16 ± 1.94 3.24 ± 0.42+,**** 2.78 ± 0.81****,# 0.16 + 0.4****

Values are expressed as means ± SD (n = 6), + p < 0.05 vs. CTRL, ++ p < 0.01 vs. CTRL, ++++ p < 0.0001 vs. CTRL, *p < 0.05 vs. NDEA, **p < 0.01 vs.NDEA, ***p < 0.001 vs. NDEA, ****p < 0.0001 vs. NDEA, # p < 0.05 vs. NDEA+TEL

Naunyn-Schmiedeberg's Arch Pharmacol

Fig. 4 Effects of SRF (30mg/kg),TEL (15 mg/kg), and acombination of both on serum atumor necrosis factor-alpha(TNF-α) levels, b transforminggrowth factor-beta (TGF-β)levels, c matrixmetalloproteinase-2 (MMP-2)levels, and d vascular endothelialgrowth factor (VEGF) levels inmice treated with NDEA for16 weeks. The data are expressedas the mean ± SD (n = 4).Statistical analysis was performedusing ordinary one-way ANOVAfollowed by Tukey’s post-test.+p < 0.05 vs. CTRL, ++p < 0.01vs. CTRL, +++p < 0.001 vs.CTRL, ++++p < 0.0001 vs. CTRL,*p < 0.05 vs. NDEA, **p < 0.01vs. NDEA, ****p < 0.0001 vs.NDEA, @p < 0.05 vs. NDEA+SRF, @@p < 0.01 vs. NDEA+SRF, #p < 0.05 vs. NDEA+TEL,##p < 0.01 vs. NDEA+TEL

Fig. 5 Effects of SRF (30 mg/kg), TEL (15mg/kg), and a combination ofboth on hepatic a NF-ĸB p-p65:p65 ratios (measured by ELISA), b p-ERK1/2:ERK1/2 ratios (measured by ELISA), and c p-TAK1:TAK1ratios (measured by western blot analysis and normalized to GAPDH)in mice treated with NDEA for 16 weeks. The data are presented as themean ± SD (n = 4) of the ratios as the fold changes from the normal CTRL

value. Statistical analysis was performed using ordinary one-wayANOVA followed by Tukey’s post-test. +p < 0.05 vs. CTRL, ++p < 0.01vs. CTRL, +++p < 0.001 vs. CTRL, ++++p < 0.0001 vs. CTRL, **p < 0.01vs. NDEA, ***p < 0.001 vs. NDEA, ****p < 0.0001 vs. NDEA,@p < 0.05 vs. NDEA+SRF, @@p < 0.01 vs. NDEA+SRF, ##p < 0.01 vs.NDEA+TEL

Naunyn-Schmiedeberg's Arch Pharmacol

Effect of telmisartan (TEL) on PPARγ DNA bindingactivity

As depicted in Fig. 7, compared with the normal mice(CTRL), the NDEA-treated mice did not exhibit a significantchange in PPARγ DNA binding activity. However, comparedwith either the normal or the NDEA-treated group, the PPARγDNA binding activity was increased significantly in theNDEA+TEL- and NDEA+TEL+SRF-treated mice. Thus,the TEL regimen in our model induces higher levels of phos-phorylated PPARγ (active form) that is able to bind to perox-isome proliferator responsive elements. As a target gene,CD36 expression is transcriptionally regulated by PPARγ innormal mice. We observed that NDEA induced a significantchange in the expression level of CD36 compared with eitherthe CTRL group. However, the relative mRNA expression ofCD36 was increased significantly in in the NDEA+TEL- andNDEA+TEL+SRF-treated mice, consistent with increasedPPARγ DNA binding activity.

Survival and mortality assessment

As depicted in Fig. 8, Kaplan-Meier analysis showed thatNDEA-treated model mice had higher mortality rates than

Fig. 6 Effects of SRF (30mg/kg),TEL (15 mg/kg), and thecombination of both on hepaticexpression of a NF-ĸBp65mRNA, b cyclin D1 mRNA, cHO-1 mRNA, and d CD36mRNA in mice treated withNDEA for 16 weeks. The data arepresented as the mean ± SD (n =4). Statistical analysis was per-formed using ordinary one-wayANOVA followed by Tukey’spost-test. +p < 0.05 vs. CTRL,++p < 0.01 vs. CTRL,+++p < 0.001 vs. CTRL,++++p < 0.0001 vs. CTRL,***p < 0.001 vs. NDEA,****p < 0.0001 vs. NDEA,@p < 0.05 vs. NDEA+SRF,@@p < 0.01 vs. NDEA+SRF,@@@p < 0.001 vs. NDEA+SRF,@@@@p < 0.0001 vs. NDEA+SRF, #p < 0.05 vs. NDEA+TEL

Fig. 7 Effects of SRF (30 mg/kg), TEL (15 mg/kg), and the combinationof both on PPARγ DNA binding activity in mice treated with NDEA for16 weeks. The data are presented as the mean ± SD (n = 4). Statisticalanalysis was performed using ordinary one-way ANOVA followed byTukey’s post-test. +++p < 0.001 vs. CTRL, **p < 0.01 vs. NDEA,@p < 0.05 vs. NDEA+SRF

Naunyn-Schmiedeberg's Arch Pharmacol

mice in other groups. Administration of SRF, TEL, or TEL+SRF significantly increased the survival probability; (p =0.046 vs. NDEA, Fig. 8a), (p = 0.029 vs. NDEA, Fig. 8b),and (p = 0.015 vs. NDEA, Fig. 8c), respectively.

Discussion

Even though HCC is one of the foremost reasons for cancer-related mortality, the currently available therapeutic interven-tions are yet limited. Therefore, the discovery of new thera-peutic strategies is essentially required for improving the prog-nosis and survival of patients with HCC. RAS inhibition is anexciting target for preventing the development of HCC. It iswell documented that Ang-II can promote tumor growth(Saber et al. 2017; Saber 2018; Saber et al. 2018d).Furthermore, inhibition of RAS signaling using angiotensinreceptor blockers (ARBs) has been reported to repress varioustypes of cancer, including ovarian, prostate, brain, breast, co-lon, pancreatic, and lung cancer (Rosenthal and Gavras 2009).Therefore, we were interested in examining the effect of TELon experimental HCC. Here, we provide the first evidence thatTEL can inhibit HCC progression in an NDEA-induced HCCmouse model, which has been considered a good representa-tion of human HCC with poor prognosis (Lee et al. 2004).

HCC in NDEA-treated animals was confirmed based onthe presence of surface nodules, the appearance of malignantenlarged hepatocytes as revealed by histological analysis anda significant elevation in serum AFP level, which is consid-ered a marker of some malignancies, particularly HCC; onlytrace amounts of AFP are normally detected in serum.Concomitant drug treatment increased lifespan, decreasedthe number of tumor nodules, significantly reduced serumAFP levels, improved liver function, and decreased relativeliver weight in NDEA-treated mice. These effects were asso-ciated with HCC regression, indicating an inhibition of

malignant transformation and a reduction in tumor productionrates. In addition, our results indicate that GST-P is a potentialmarker for hepatocarcinogenesis as confirmed byimmunolabeling of liver tissue sections.

In the present study, NDEA-treated mice exhibited elevatedVEGF levels; VEGF upregulation is known to be an essentialfactor in neovascularization, which is an important process forthe development and growth of solid tumors. Our results are con-sistent with those of other reports demonstrating VEGF upregula-tion during HCC (Schoenleber et al. 2009). It has been reportedthat Ang-II-mediated induction of VEGF mRNA and protein ex-pression is significantly inhibited by an AT1R antagonist via themitogen-activated protein kinase (MAPK) (Anandanadesan et al.2008) and NF-κB signaling pathways (Saber et al. 2018c). Inaddition, NF-κB and ERK signaling have been implicated in var-ious cellular processes, such as immunomodulation, differentia-tion, and proliferation (Abdel-Ghany et al. 2015; Saber et al.2018a), and VEGF has been suggested to stimulate signaling cas-cades through ERK phosphorylation to lead to enhanced prolifer-ation (Takahashi et al. 1999).

We further detected a significant increase in serum TGF-βlevels in NDEA-treated mice. The vital TGF-β pathway isknown to play an essential role in promoting the epithelial-mesenchymal transition (EMT), a fundamental mechanism fortumor metastasis. In addition, TGF-β1 modulates the expres-sion of several genes associated with tumor development. Thecurrent study showed that TEL administration significantlyreduced TGF-β levels in a manner corresponding to that ofSRF administration. Previous reports have shown that admin-istration of single ARBs can inhibit TGF-β expression in dif-ferent models (Wu et al. 2016).

The capability of TGF-β to augment metastasis is believedto be facilitated by stimulation of MMP-2 secretion (Lin et al.2000). MMPs are capable of degrading the extracellular ma-trix (ECM) to boost migration and tumor metastasis. In thecurrent work, NDEA-treated model animals revealed

Fig. 8 Effects of SRF (30 mg/kg), TEL (15 mg/kg), and the combination of both on survival rate. Shown are the Kaplan-Meier survival curves for a SRFvs. NDEA, b TEL vs. NDEA, and c TEL+SRF vs. NDEA. Statistical analysis was performed using the log-rank test (Mantel-Cox method)

Naunyn-Schmiedeberg's Arch Pharmacol

significantly higher MMP-2 levels than CTRL animals.Conversely, TEL- and SRF- treated mice exhibited signifi-cantly lower MMP-2 levels than NDEA-only mice, which isprobably attributable to TGF-β inhibition. Furthermore, dif-ferent studies have indicated the ability of ARBs to inhibitMMPs (Yang et al. 2009).

Dai et al. (2013) reported that TGF-β is involved in upreg-ulating cyclin D1 in breast tissue carcinoma. During the cellcycle, cyclin D1 plays a key role in regulating the cell transi-tion through the G1 phase. Cyclin D1 upregulation has beendetected in various types of malignancies, such as HCC. Ourdata showed that the relative expression of cyclin D1 wassignificantly higher in NDEA-treated model animals than inCTRL mice, consistent with the findings of Shirakami et al.(2012). Administration of TEL and SRF exerted anti-proliferative effects, as indicated by the significantly lowercyclin D1 mRNA expression in TEL- and SRF-treated micethan in NDEA-only mice. Ang-II has also been shown toupregulate the mRNA expression of cyclin D1 (Guillemotet al. 2001).

TNF-α stimulates a downstream NF-κB signaling pathwaythat is involved in regulating inflammation and cell prolifera-tion and contributes to the development and progression ofHCC. In the present work, we revealed significantly higher

TNF-α levels in NDEA-treated model mice than in controlmice, suggesting that inflammatory pathways were stimulat-ed. Our data are consistent with those of earlier reports indi-cating that NDEA treatment increases TNF-α levels (Songet al. 2013). Administration of TEL and SRF significantlyrepressed the inflammation pathway, which was indicated bythe significant decline in TNF-α levels in drug-treated micecompared with NDEA-only mice. AT1R blockers have beenfound to downregulate TNF-α both experimentally and clin-ically (Ali et al. 2015). Moreover, TNF-α activates the ERKpathway, leading to upregulation of TGF-β1 mRNA and pro-tein expression (Sullivan et al. 2005), and ERK signaling in-fluences TNF-α expression (Sabio and Davis 2014) (Fig. 9).

The pro-inflammatory activity of Ang-II strongly dependson the recruitment of an IkB kinase (IKK) complex signalingcascade that leads to phosphorylation of p65 subunits (Weiet al. 2008). Moreover, Ang-II leads to phosphorylation-initiated degradation of Iκ-Bα and increases NF-κB p65transactivation, nuclear translocation, and DNA binding ca-pacity (Liu et al. 2006). Previous reports have revealed thatSRF decreases the expression levels of NF-κB p65 subunitsand inhibits tumor growth through inactivation of NF-κB(Park et al. 2017). In the cytosol, NF-κBp65 is inactivatedby complexing with Iκ-Bα and is stimulated upon

Fig. 9 Proposed mechanism of action of telmisartan (TEL) as a modula-tor of NF-Κb-TAK1-ERK1/2 axis in the context of PPARγ agonisticactivity. AT1R, angiotensin II type 1 receptor; CD36, cluster of differen-tiation 36; ERK, extracellular signal-regulated kinase; HO-1, heme oxy-genase 1; Iκ-Bα, nuclear factor kappa-B inhibitor alpha; IKK, IkB kinase;MEK, mitogen-activated protein kinase kinase; NFκB, nuclear transcrip-tion factor kappa-B; PKC, PPARγ, peroxisome proliferator-activated

receptor gamma; ROS, reactive oxygen species; SRF, sorafenib; TAB1,TGF-beta-activated kinase 1-binding protein 1; TAK1, transforminggrowth factor beta-activated kinase 1; TGF-βR, transforming growthfactor beta receptor; TNFR, tumor necrosis factor alpha receptor;TRAF6, tumor necrosis factor receptor-associated factor 6; VEGFR, vas-cular endothelial growth factor receptor

Naunyn-Schmiedeberg's Arch Pharmacol

inflammation. Upon activation, Iκ-Bα is phosphorylated byIKKs and degraded through proteasomal degradation. Theseactions permit the transactivation and p65 subunits transloca-tion to the nucleus with consequent increases in the productionof various pro-inflammatory mediators and tumor-promotinggenes. Phosphorylation of p65 occurs in the cytosol and hasbeen revealed to promote both p65 transcription activity andDNA binding capacity and to reduce the capability of p65 tocomplex with Iκ-Bα. The current study provides evidence forthe involvement of the NF-кB pathway in the observed anti-tumor effects of TEL and SRF, as indicated by lower hepaticp-p65/total p65 ratios and p65 mRNA expression in drug-treated mice than in NDEA-only mice, representing inactiva-tion of the NF-кB pathway (Fig. 9).

It has been demonstrated that PPARγ activation potentlyinhibits NF-кB-dependent transcriptional activation andstrongly suppresses transcription of NF-кB-dependent genes(Remels et al. 2009). TEL is considered a PPARγ agonist andcan influence the expression of PPARγ target genes involvedin the metabolism of carbohydrates and lipids, such as CD36(Fujita et al. 2007). Our study demonstrated that TEL, but notSRF, upregulated the mRNA expression of CD36 and HO-1(Fig. 9). Notably, crosstalk between HO-1 and PPARs hasbeen identified (Ndisang 2014). Emerging evidence indicatesthat many of the effects modulated by the HO system aremediated through PPARs through a synergistic relationship.Consistent with these findings, in the current work, PPARγactivation by TEL increased the expression of HO-1 mRNA,upregulated the catalase enzyme and increased total antioxi-dant capacity. In addition, it has been reported that upregula-tion of HO-1 expression is associated with increases in cata-lase levels (Turkseven et al. 2005).

In the present study, TEL inhibited the phosphorylation-induced activation of TAK1. TAK1 has been suggested to beinvolved in regulating the activation of IKK. TAK1 is a mem-ber of the MAP3K family that stimulates the activation ofNF-κB as well as the MAPK signaling pathways. Therefore,TAK1 acts as a central mediator in the crosstalk betweenNF-κB and MAPK (Fig. 9).

In conclusion, in addition to agonistic activity on PPARγ,inactivation of the NF-κB and ERK1/2 pathways is implicatedin the anti-tumor effect of TEL. Moreover, TAK1 seems toplay a crucial role in modulating the interplay between NF-κBand ERK1/2. As a result of these specific effects, TEL dem-onstrated anti-inflammatory, anti-proliferative, anti-metasta-tic, and anti-angiogenic activity. In addition, TEL prolongedsurvival, restored hepatic lobular architecture, increased anti-oxidant capacity, improved liver function, and augmented theanti-tumor effect of SRF. Also, we conclude that GST-P is apotential marker for hepatocarcinogenesis. Finally, TEL is apromising agent for further clinical trials on HCC manage-ment as a single agent or as a co-adjuvant with other chemo-therapies to minimize costs and resistance.

Author’s contribution SS conceived, designed, conducted the experi-ment, and analyzed data; AK, WS, and MS conceived, designed, andconducted the experiment; WA performed histological and immunohis-tochemical analysis; BE, KN, AA, NM, MB, AS, MM, RM, and AEconducted the experiment. All authors equally contributed in writing,reading, and approving the manuscript.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict ofinterest.

References

Abdel-Ghany R, Rabia I, El-Ahwany E, Saber S, Gamal R, Nagy F,Mahmoud O, Hamad RS, Barakat W (2015) Blockade of PGE2,PGD2 receptors confers protection against prepatent schistosomiasismansoni in mice. J Egypt Soc Parasitol 45:511–520

Abdo W, Hirata A, Sakai H, El-Sawak A, Nikami H, Yanai T (2013)Combined effects of organochlorine pesticides heptachlor and hexa-chlorobenzene on the promotion stage of hepatocarcinogenesis inrats. Food Chem Toxicol 55:578–585

Abdo W, Hirata A, Shukry M, Kamal T, Abdel-Sattar E, Mahrous E,Yanai T (2015) Calligonum comosum extract inhibitsdiethylnitrosamine-induced hepatocarcinogenesis in rats. OncolLett 10:716–722

Ali Z, Sudiro T, Mulia D (2015) Losartan effect on lowering TNF-[alpha]serum levels in asymptomatic hyperuricemia hypertension patients.J Hypertens 33:e33

Anandanadesan R, Gong Q, Chipitsyna G, Witkiewicz A, Yeo CJ, ArafatHA (2008) Angiotensin II induces vascular endothelial growth fac-tor in pancreatic cancer cells through an angiotensin II type 1 recep-tor and ERK1/2 signaling. J Gastrointest Surg 12:57–66

Bosman FT, Carneiro F, Hruban RH, Theise ND (2010) WHO classifi-cation of tumours of the digestive system. World HealthOrganization

Brito AF, Abrantes AM, Tralhao JG, Botelho MF (2016) Targeting he-patocellular carcinoma: what did we discover so far? Oncol Rev 10:302

Christian JB, Lapane KL, Hume AL, Eaton CB, Weinstock MA, Trial V(2008) Association of ACE inhibitors and angiotensin receptorblockers with keratinocyte cancer prevention in the randomizedVATTC trial. J Natl Cancer Inst 100:1223–1232

Dai M, Al-Odaini AA, Fils-Aime N, Villatoro MA, Guo J, Arakelian A,Rabbani SA, Ali S, Lebrun JJ (2013) Cyclin D1 cooperates with p21to regulate TGFbeta-mediated breast cancer cell migration and tu-mor local invasion. Breast Cancer Res 15:R49

De Paepe B, Verstraeten VM, De Potter CR, Bullock GR (2002)Increased angiotensin II type-2 receptor density in hyperplasia,DCIS and invasive carcinoma of the breast is paralleled with in-creased iNOS expression. Histochem Cell Biol 117:13–19

Dhanasekaran R, Bandoh S, Roberts LR (2016) Molecular pathogenesisof hepatocellular carcinoma and impact of therapeutic advances.F1000Res 5

Fujita K, Yoneda M, Wada K, Mawatari H, Takahashi H, Kirikoshi H,Inamori M, Nozaki Y, Maeyama S, Saito S (2007) Telmisartan, anangiotensin II type 1 receptor blocker, controls progress of nonalco-holic steatohepatitis in rats. Dig Dis Sci 52:3455–3464

Funao K, Matsuyama M, Kawahito Y, Sano H, Chargui J, Touraine J-L,Nakatani T, Yoshimura R (2008) Telmisartan is a potent target forprevention and treatment in human prostate cancer. Oncol Rep 20:295–300

Naunyn-Schmiedeberg's Arch Pharmacol

Funao K, Matsuyama M, Kawahito Y, Sano H, Chargui J, Touraine J-L,Nakatani T, Yoshimura R (2009) Telmisartan as a peroxisomeproliferator-activated receptor-γ ligand is a new target in the treat-ment of human renal cell carcinoma. Mol Med Rep 2:193–198

Guillemot L, LevyA, RaymondjeanM, Rothhut B (2001) Angiotensin II-induced transcriptional activation of the cyclin D1 gene is mediatedby Egr-1 in CHO-AT(1A) cells. J Biol Chem 276:39394–39403

Herr D, Rodewald M, Fraser HM, Hack G, Konrad R, Kreienberg R,Wulff C (2008) Potential role of renin-angiotensin-system for tumorangiogenesis in receptor negative breast cancer. Gynecol Oncol 109:418–425

Ishiguro H, Ishiguro Y, Kubota Y, Uemura H (2007) Regulation of pros-tate cancer cell growth and PSA expression by angiotensin II recep-tor blocker with peroxisome proliferator-activated receptor gammaligand like action. Prostate 67:924–932

Koyama N, Nishida Y, Ishii T, Yoshida T, Furukawa Y, Narahara H(2014) Telmisartan induces growth inhibition, DNA double-strandbreaks and apoptosis in human endometrial cancer cells. PLoS One9:e93050

Lee J-S, Chu I-S, Mikaelyan A, Calvisi DF, Heo J, Reddy JK,Thorgeirsson SS (2004) Application of comparative functional ge-nomics to identify best-fit mouse models to study human cancer. NatGenet 36:1306–1311

Lee LD, Mafura B, Lauscher JC, Seeliger H, Kreis ME, Gröne J (2014)Antiproliferative and apoptotic effects of telmisartan in human coloncancer cells. Oncol Lett 8:2681–2686

Lin SW, Lee MT, Ke FC, Lee PP, Huang CJ, Ip MM, Chen L, Hwang JJ(2000) TGFbeta1 stimulates the secretion of matrix metalloprotein-ase 2 (MMP2) and the invasive behavior in human ovarian cancercells, which is suppressed by MMP inhibitor BB3103. Clin ExpMetastasis 18:493–499

Liu H-q, Wei X-b, Sun R, Cai Y-w, Lou H-y, Wang J-W, Chen AF, ZhangX-M (2006) Angiotensin II stimulates intercellular adhesionmolecule-1 via an AT 1 receptor/nuclear factor-κB pathway in brainmicrovascular endothelial cells. Life Sci 78:1293–1298

Miyajima A, Kosaka T, Asano T, Asano T, Seta K, Kawai T, HayakawaM (2002) Angiotensin II type I antagonist prevents pulmonary me-tastasis of murine renal cancer by inhibiting tumor angiogenesis.Cancer Res 62:4176–4179

Ndisang JF (2014) Cross-talk between heme oxygenase and peroxisomeproliferator-activated receptors in the regulation of physiologicalfunctions. Front Biosci (Landmark Ed) 19:916–935

Ota K, Ito K, Suzuki T, Saito S, Tamura M, Hayashi S-i, Okamura K,Sasano H, Yaegashi N (2006) Peroxisome proliferator-activated re-ceptor γ and growth inhibition by its ligands in uterine endometrialcarcinoma. Clin Cancer Res 12:4200–4208

Park GB, Ko HS, Kim D (2017) Sorafenib controls the epithelial-mesenchymal transition of ovarian cancer cells via EGF and theCD44-HA signaling pathway in a cell type-dependent manner.Mol Med Rep 16:1826–1836

Ranjan A, Iyer SV, Ward C, Link T, Diaz FJ, Dhar A, Tawfik OW,Weinman SA, Azuma Y, Izumi T (2018) MTBP inhibits theErk1/2-Elk-1 signaling in hepatocellular carcinoma. Oncotarget 9:21429

Remels A, Langen R, Gosker HR, Russell A, Spaapen F, Voncken J,Schrauwen P, Schols AM (2009) PPARγ inhibits NF-κB-dependent transcriptional activation in skeletal muscle. A J PhysiolEndocrinol Metab 297:E174–E183

Rosenthal T, Gavras I (2009) Angiotensin inhibition and malignancies: areview. J Hum Hypertens 23:623–635

Saber S (2018) Angiotensin II: a key mediator in the development of liverfibrosis and cancer. Bull Nat Res Centre 42:18

Saber S, Goda R, El-Tanbouly GS, Ezzat D (2018a) Lisinopril inhibitsnuclear transcription factor kappa B and augments sensitivity tosilymarin in experimental liver fibrosis. Int Immunopharmacol 64:340–349

Saber S, Khalil RM, Abdo WS, Nassif D, El-Ahwany E (2019)Olmesartan ameliorates chemically-induced ulcerative colitis in ratsvia modulating NFκB and Nrf-2/HO-1 signaling crosstalk. ToxicolAppl Pharmacol 364:120–132

Saber S, Mahmoud A, Helal N, El-Ahwany E, Abdelghany R (2018b)Liver protective effects of renin-angiotensin system inhibition haveno survival benefits in hepatocellular carcinoma induced by repeti-tive administration of diethylnitrosamine in mice. Open AccessMaced J Med Sci 6:955–1179

Saber S,MahmoudAA, Helal NS, El-Ahwany E, Abdelghany RH (2017)Losartan, an angiotensin-II type 1 receptor blocker, attenuates CCl4-induced liver fibrosis with a positive impact on survival in mice.World J Pharm Pharm Sci 5:121–126

Saber S, Mahmoud AAA, Goda R, Helal NS, El-ahwany E, AbdelghanyRH (2018c) Perindopril, fosinopril and losartan inhibited the pro-gression of diethylnitrosamine-induced hepatocellular carcinoma inmice via the inactivation of nuclear transcription factor kappa-B.Toxicol Lett 295:32–40

Saber S, Mahmoud AAA, Helal NS, El-Ahwany E, Abdelghany RH(2018d) Renin–angiotensin system inhibition ameliorates CCl4-induced liver fibrosis in mice through the inactivation of nucleartranscription factor kappa B. Can J Physiol Pharmacol 96:569–576

Sabio G, Davis RJ (2014) TNF and MAP kinase signalling pathways.Seminars in immunology. Elsevier, pp 237–245

Schoenleber SJ, Kurtz DM, Talwalkar JA, Roberts LR, Gores GJ (2009)Prognostic role of vascular endothelial growth factor in hepatocel-lular carcinoma: systematic review and meta-analysis. Br J Cancer100:1385–1392

Sharma V, McNeill JH (2009) To scale or not to scale: the principles ofdose extrapolation. Br J Pharmacol 157:907–921

Shirakami Y, Gottesman ME, Blaner WS (2012) Diethylnitrosamine-induced hepatocarcinogenesis is suppressed in lecithin:retinolacyltransferase-deficient mice primarily through retinoid actions im-mediately after carcinogen administration. Carcinogenesis 33:268–274

Shlomai A, de Jong YP, Rice CM (2014) Virus associated malignancies:the role of viral hepatitis in hepatocellular carcinoma. Semin CancerBiol 26:78–88

Sjoberg T, Garcia Rodriguez LA, Lindblad M (2007) Angiotensin-converting enzyme inhibitors and risk of esophageal and gastriccancer: a nested case-control study. Clin Gastroenterol Hepatol 5:1160–1166 e1161

SongY, Jin SJ, Cui LH, Ji XJ, Yang FG (2013) Immunomodulatory effectof Stichopus japonicus acid mucopolysaccharide on experimentalhepatocellular carcinoma in rats. Molecules 18:7179–7193

Stangier J, Su C, Roth W (2000) Pharmacokinetics of orally and intrave-nously administered telmisartan in healthy young and elderly vol-unteers and in hypertensive patients. J Int Med Res 28:149–167

Sullivan DE, Ferris M, Pociask D, Brody AR (2005) Tumor necrosisfactor-α induces transforming growth factor-β1 expression in lungfibroblasts through the extracellular signal–regulated kinase path-way. Am J Respir Cell Mol Biol 32:342–349

Takahashi T, Ueno H, Shibuya M (1999) VEGF activates protein kinaseC-dependent, but Ras-independent Raf-MEK-MAP kinase pathwayfor DNA synthesis in primary endothelial cells. Oncogene 18:2221–2230

Turkseven S, Kruger A, Mingone CJ, Kaminski P, Inaba M, Rodella LF,Ikehara S, Wolin MS, Abraham NG (2005) Antioxidant mechanismof heme oxygenase-1 involves an increase in superoxide dismutaseand catalase in experimental diabetes. Am J Phys Heart Circ Phys289:H701–H707

Wei Y, Sowers JR, Clark SE, Li W, Ferrario CM, Stump CS (2008)Angiotensin II-induced skeletal muscle insulin resistance mediatedbyNF-κB activation via NADPH oxidase. Am J Physiol EndocrinolMetab 294:E345–E351

Naunyn-Schmiedeberg's Arch Pharmacol

Wu M, Peng Z, Zu C, Ma J, Lu S, Zhong J, Zhang S (2016) Losartanattenuates myocardial endothelial-to-mesenchymal transition inspontaneous hypertensive rats via inhibiting TGF-beta/Smad signal-ing. PLoS One 11:e0155730

Yang D, Ma S, Li D, Tang B, Yang Y (2009) Angiotensin II receptorblockade improves matrix metalloproteinases/tissue inhibitor of ma-trix metalloproteinase-1 balance and restores fibronectin expressionin rat infarcted myocardium. Biochem Biophys Res Commun 388:606–611

Zheng X, Gai X, Han S, Moser CD, Hu C, Shire AM, Floyd RA, RobertsLR (2013) The human sulfatase 2 inhibitor 2,4-disulfonylphenyl-tert-butylnitrone (OKN-007) has an antitumor effect in hepatocellu-lar carcinoma mediated via suppression of TGFB1/SMAD2 andhedgehog/GLI1 signaling. Genes Chromosom Cancer 52:225–236

Publisher’s note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

Naunyn-Schmiedeberg's Arch Pharmacol