TP53-dependence on the effect of doxorubicin and ... - SAGE Journals

3 downloads 0 Views 2MB Size Report
AZD8931, a novel EGFR/HER2/HER3 signalling inhi- bitor, with AZD5363 limits AKT inhibitor induced feed- back and enhances antitumour efficacy in HER2-.
Original Article

TP53-dependence on the effect of doxorubicin and Src inhibitor combination therapy

Tumor Biology August 2018: 1–11 Ó The Author(s) 2018 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/1010428318794217 journals.sagepub.com/home/tub

Yun Sun Lee1, Ji-Yoon Choi1, Jeeyun Lee2, Da Mi Shim1, Jaesoo Kim1, Woong-Yang Park3, Do-Hyun Nam4 and Sung Wook Seo1

Abstract The anticancer effects of Src kinase inhibitors are controversial. This study found an association between alterations in the TP53 gene and the synergy score for combination treatment with doxorubicin and an Src kinase inhibitor using human osteosarcoma cell lines (MG63 and U2OS) and human colon cancer cell line. Doxorubicin was found to activate signal transducer and activator of transcription 3 via Src kinase in cancer cells harboring alterations in TP53. A drug combination study using patient-derived cells confirmed that an Src kinase inhibitor synergizes with doxorubicin in cancer cells harboring alterations in TP53, while antagonizing its effect in cancer cells expressing wild-type TP53. Our findings suggest that genetic alterations in TP53 are a critical factor in determining the use of a combination treatment of doxorubicin and Src inhibitors. Keywords TP53, Src kinase, doxorubicin, drug resistant, saracatinib (AZD0530)

Date received: 12 April 2018; accepted: 18 July 2018

Introduction Doxorubicin (DOX) is a highly potent antitumor therapeutic agent widely used in the clinic. However, one of the main reasons for the problems encountered with its use in the clinic, and for its ultimate failure, is that cancer cells become resistant to DOX. Therefore, an understanding of the mechanisms underlying DOX resistance in various cancers is urgently needed.1,2 Src kinase, also known as Src proto-oncogene, plays a key role in the regulation of diverse cellular functions, including cell proliferation, differentiation, adhesion, and motility in both normal and cancer cells. Src kinase is overexpressed and abnormally activated in various human cancer cells and has been strongly implicated in tumorigenesis and metastatic progression. Current studies have demonstrated that increased Src activity correlates with disease progression and a poor clinical prognosis.3–6 Therefore, it would be expected that Src inhibitors could overcome the chemoresistance of common antitumor drugs.

In various clinical trials, Src inhibitors such as bosutinib, saracatinib, and dasatinib have been used either as single agents or in combination with chemotherapy. A clinical trial of bosutinib in chronic myeloid leukemia reported successful data. Two Phase 2 clinical trials using saracatinib failed to meet their primary endpoint in colon cancer and small-cell lung cancer patients. In

1

Department of Orthopaedic Surgery, Samsung Medical Center, School of Medicine, Sungkyunkwan University, Seoul, Korea 2 Department of Medicine, Division of Hematology and Oncology, Samsung Medical Center, School of Medicine, Sungkyunkwan University, Seoul, Korea 3 Samsung Genome Institute, Samsung Medical Center, Seoul, Korea 4 Department of Neurosurgery, Samsung Medical Center, School of Medicine, Sungkyunkwan University, Seoul, Korea Corresponding author: Sung Wook Seo, Department of Orthopaedic Surgery, Samsung Medical Center, School of Medicine, Sungkyunkwan University, 81 Irwon-Ro, Gangnam-gu, Seoul, 135-710, Korea. Email: [email protected]

Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

2 another clinical trial using dasatinib for the treatment of breast cancer, only one patient exhibited a clinical benefit,7 but unfortunately, the study failed to find a gene signature that defined tumor sensitivity to dasatinib. Combination therapy using Src kinase inhibitors and other agents have also reportedly shown poor outcomes. A combination of dasatinib and the receptor tyrosine kinase inhibitor erlotinib,8 a combination of bosutinib and the aromatase inhibitor letrozole,9 and a combination of dasatinib and docetaxel10 were all reported to be unsuccessful. During tumor progression, genetic heterogeneity occurs by the evolution of a variety of sub-clones. Within tumors, cellular heterogeneity is a major potential factor that can weaken the success of combination therapies.11,12 In this study, we aimed to find a key mutation that determines Src kinase sensitivity. This study characterizes the gene alterations that are associated with the synergistic effect of combination therapy with DOX and Src kinase. With our CancerSCAN V2 panel data, we have found that 380 cancer-related genes, including TP53, in these 14 cell lines.13 We have performed that cancers with TP53 mutations are sensitive to a combination therapy of DOX and an Src kinase inhibitor. In addition, we evaluated the TP53dependent mechanism for Src kinase that affects the use of Src kinase inhibitors. Finally, we demonstrate for the first time that TP53 is an important signature gene that defines tumors that are sensitive to combination therapy.

Materials and methods Chemicals DOX, SU6656 were purchased from Sigma-Aldrich (St. Louis, MO, USA) and AZD0530 were from Selleck Chemicals (Houston, TX, USA).

Cell culture conditions A549, MCF7, MG63, and Huh7 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM), and U2OS, HepG2, and SaOS2 cells were cultured in Modified Eagle’s medium (MEM). Prostate cancer cell lines (LNCaP, PC3, and DU145), HCT116, and MDAMB468 cells were grown in RPMI1640 (HyClone Laboratories, Inc, Logan, UT, USA) with 10% fetal bovine serum (FBS) and 1% antibiotic-antimycotic solution. 143B cells were cultured in Eagle’s minimal essential medium (EMEM; Bio Whittaker, Lonza, Walkersville, MD) with 10% FBS and 1% antibioticantimycotic solution. All patient tissue samples were approved by the Institutional Review Board (IRB) in Samsung Medical center in accordance with the regulation of involved institutions and provided an informed

Tumor Biology consent from all patients. Patient tissues were washed with phosphate-buffered saline (PBS) and minced into small pieces of 2–4 mm using a scalped blade. The minced tissues were transferred into conical tube and dissociated at 37°C for 3 h with collagenase. Patientderived cells (PDCs) were cultured in RPMI1640 with 10% FBS and 1% antibiotic–antimycotic solution.

Determination of TP53 status The TP53 status of the cell lines was based on the International Agency for Research on Cancer (IARC) TP53 database as described in previous studies.13,14 The TP53 transcription status of the samples were determined by CancerSCAN, an internal development kit for sequencing of target gene that uses an Illumina HiSeq 2000 (Illumina Inc, San Diego, CA, USA). These were further confirmed by real-time polymerase chain reaction (RT-PCR) and Sanger sequencing, respectively, and the p53 mutation status of cell lines were shown in detail as supplementary table in the previous study.15 To amplify and sequence the p53 transcript, PCR was performed using the following primers: p53-specific primers, forward 5#-ctgggctccggggacacttt3# and reverse 5#-cgcacacctattgcaagc-3#.

Cell viability assay Cell viability assay was measured using a previously described procedure.15 Cell viability was determined by dissolve the formazan crystals using the MTT (3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay according to the manufacturer’s protocol (SigmaAldrich), and the absorbance was read at 570 nm.

Evaluation of the drug combination effect The combination effect of DOX/Src inhibitors was measured using the Bliss independence16 and Loewe additivity model,17,18 as described previously.15

Animal experiment HCT116–/– cells (5 3 106 cells) were injected subcutaneously into 6-week-old female BALB/c-nude mice (Orient Bio Inc, Seoul, Korea). At an approximate tumor volume of 50 mm3, the mice were injected intraperitoneally with DOX (2.5 mg/kg) every 3 days. The tumors were excised 2 weeks after DOX treatment and performed to immunohistochemistry analysis.

Immunohistochemistry analysis Tumor tissues taken from mice were fixed in 4% paraformaldehyde and then paraffin embedded, sections were deparaffinized in xylene. The samples were

Lee et al. subjected to heat-induced epitope retrieval (HIER) with citrate buffer (pH 6.0; Dako, Carpinteria, CA, USA) for 3 min at 121°C.19 Subsequently, endogenous peroxidase was blocked with 3% hydrogen peroxide in PBS for 10 min at room temperature. Sections were washed in PBS, and treated with normal goat for 20 min at room temperature to block non-specific binding,20 and incubated with an anti-phospho (Y416)-Src antibody (Novus Biologicals, Littleton, CO, USA) for 16 h at 4°C. The sections were incubated for 30 min at room temperature with a horseradish peroxidase (HRP)-conjugated secondary antibody against rabbit IgG. Finally, the sections were lightly counterstained with Mayer’s hematoxylin for 1 min.

Immunofluorescent staining The effects of DOX on Src activity in HCT116+/+ and HCT116–/– cells were measured according to the previously described method with minor modification.21 Cells were cultured on glass coverslips (Fisher Scientific, Pittsburgh, PA, USA) and treated with 0.5 mM DOX for 16 h. Following wash with PBS, the cells were fixed in 3.7% paraformaldehyde and permeabilized with 0.5% Triton X-100 in PBS. Subsequently, coverslips were incubated for 1 h in 1% bovine serum albumin (BSA) and incubated with a rabbit anti-phospho-Src (Y416) antibody (Cell Signaling Technology, Danvers, MA, USA), followed by incubation with contained a fluorescent dye-conjugated secondary antibody (goat anti-rabbit Alexa 488, Invitrogen), for 2 h. The slides were mounted with Vectashield mounting medium containing DAPI (4’,6-Diamidino-2-Phenylindole; Vector Laboratories, Burlingame, CA, USA) and visualized using a Zeiss LSM 780 laser scanning confocal microscope (Carl Zeiss Inc, Go¨ttingen, Germany).

Plasmid, siRNA transfection pcDNA-p53 was cloned via the insertion of the p53 gene coding sequence (NM_000546) into pcDNA3.1. pcDNA-CA-Src (constitutive active form, Y530) was cloned using pcDNA3-MTS-CA-c-Src-FLAG that was provided by Dr Yoshimi Homma (Addgene plasmid #44654). The amplified coding sequences of p53 and CA-Src (Supplementary Table2) were digested with HindIII (underlined) and EcoRI (underlined in italic) and then ligated into HindIII and EcoRI digested pcDNA3.1, respectively. The sequence of plasmids was confirmed by Sanger sequencing. Plasmids were transfected using LipofectamineÒ2000 (Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. p53 and signal transducer and activator of transcription 3 (STAT3) siRNA (Supplementary Table2) were purchased from Bioneer and transfected

3 using LipofectamineÒRNAiMAX Technologies).

reagent

(Life

Immunoblot analysis The cells (1.5 3 105 cells) were plated in six-well plates. After drug treatment, the cells were washed with PBS and lysed in radioimmunoprecipitation assay (RIPA) buffer containing 150-mM NaCl, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 50-mM Tris-HCl (pH 8.0) supplemented with protease inhibitor cocktail (ThermoScientific, Piscataway, NJ, USA). Protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% gel and then transferred onto Polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Hercules, CA, USA). The following antibodies were used: anti-Src, anti-phospho(Y416)-Src, anti-STAT3, anti-phospho (Y705)-STAT3, anti-cleaved PARP (Cell Signaling Technology), anti-JNK, anti-phospho(T183, Y185)JNK, anti-phospho(S15)-p53 (Abcam, Cambridge, UK), anti-p53, anti-b-Actin, anti-rabbit IgG-HRP, and anti-mouse IgG-HRP (Santa Cruz Biotechnology, Santa Cruz, CA, USA).

Statistical analysis Statistical analysis of the data was signified as the mean 6 SD. Comparisons between two groups were performed using the non-parametric Mann–Whitney test or two tailed t-test using GraphPad Prism 5 (GraphPad Software, Inc, La Jolla, CA, USA).

Results Src kinase is a well-known survival factor that is found in various types of cancer cells. Our hypothesis was that a synergistic anticancer effect should occur when using a combination of DOX and an Src inhibitor. To address this hypothesis, we first analyzed the effect on cell viability in 14 different cancer cell lines treated with DOX and a Src inhibitor using the bliss synergy score (Figure 1(a)). Unexpectedly, we found that this combination showed a synergistic effect in some cancer cell lines (HCT116–/–, MDA-MB468, DU145, MG63, 143B, Huh7, SaOS2, PC3, LNCaP, and MCF7) but an antagonistic effect in other cell lines (HCT116+/+, A549, U2OS, and HepG2). In a previous study,15 we verified that 380 cancer-related genes (CancerSCAN V2) in a human colon cancer cell line (HCT116), a human hepatoma cell lines (HepG2 and Huh7), a human non-small-cell lung cancer cell line (A549), human osteosarcoma cell lines (MG63, SaOS2, 143B and U2OS), human prostate cancer cell lines (PC3, LNCaP and DU145) and human breast cancer cell line (MCF7 and MDA-MB468). The results for the

4

Figure 1. Continued

Tumor Biology

Lee et al.

Figure 1. Continued

5

6

Tumor Biology

Figure 1. The paradoxical effect of Src inhibition: (a) additive growth inhibition by SU6656 during DOX treatment in 14 cancer cell lines. Cells were treated with 0.5-mM DOX or the combination of 0.5-mM DOX and 25-mM SU6656. For the combination of DOX and SU6656, cells were pre-treated with SU6656 for 1 h prior to DOX treatment. After 2 days, cell viability was measured using the MTT assay. Additive growth inhibition was evaluated by subtracting the viable value with the combination treatment from the viable value with the DOX treatment. Data represent the mean 6 S.D. (b) pSrc (Y416) levels in TCGA samples (TCGA RPPA score) from the cBioPortal open source. The significance of the statistical comparisons was based on the two-tail t-test. (c) Increased expression of pSrc in the DOX treated HCT116–/– tumor model. Phosphorylation of Src was analyzed using immunohistochemistry as described in ‘‘Materials and methods’’ section. (d) The TP53 status depends on the differential activation of Src. HCT116+/+ and HCT116–/– cells were treated with 0.5 mM of DOX for 16 h. Indirect immunofluorescence was performed using an anti-pSrc antibody. (e) HCT116+/+, HCT116–/–, U2OS, and MG63 cells were treated with 0.5-mM DOX. After treatment for the indicated time, cell lysates were prepared and Western blotting was performed to assess the levels of pSrc, Src, pp53, and p53. (f) Synergism and antagonism of SU6656 in cell lines. The box and whisker plot represents the distribution of excess over bliss score for the combination of DOX and SU6656. Cells were treated with each drug or with the combination of DOX and SU6656, for 2 days (0.5mM DOX, 25-mM SU6656). Excess over bliss for TP53 wild-type cell lines (n = 5; HepG2, HCT116+/+, U2OS, MCF7, and A549 cells) and TP53-altered cell lines (n = 7; DU145, MG63, SaOS2, MDA-MB468, HCT116–/–, 143B, and Huh7 cells) were compared by the Mann–Whitney test. Significant differences are marked by asterisks (*p\0.05). (g) Synergy scores in colon cancer (HCT116+/ +, HCT116–/–) and osteosarcoma (U2OS, MG63) cell lines for the combination of DOX and AZD0530. Cells were treated with increasing doses of DOX (0, 0.25, 0.5, 1, 2, and 4 mM) or AZD0530 (0, 0.25, 0.5, 1, 2, and 4 mM), respectively or the combination of DOX and AZD0530. Cell viability was measured with an MTT assay after 2 days of treatment. The synergy scores were calculated by the Loewe dose-additivity model. (h) The synergistic effect of AZD0530 with DOX in p53 silenced U2OS cells. (i) The antagonistic effect of AZD0530 with DOX in MG63 cells ectopically expressing p53. The upper and lower panels represent the Loewe excess and isobologram, respectively.

mutational status of p53 in the cancer cell lines were also showed in the supplementary table of a previous study.15 We found that an alteration in TP53 was significantly associated with a high synergy score. In other words, the data in Figure 1(a) show that the eight cancer cells with an alteration in TP53 had a synergistic effect of DOX plus the Src inhibitor, whereas four of the six cancer cells with wild-type TP53 showed an antagonistic effect. We also found the relationship between an alteration in TP53 and Src activity by using

the open-source data generated by The Cancer Genome Atlas (TCGA) research network. Among 33 data sets that contained both the phosphoprotein level (measured by Reverse Phase Protein Array, RPPA) and genomic data, five of these data sets demonstrated a significant increase in phosphorylated Src in the TP53altered group (p \ 0.001; Figure 1(b)). Based on these results, we evaluated the activation of Src by DOX in TP53-altered cells. As shown in Figure 1(c), the expression levels of Src increased markedly following DOX

Lee et al. treatment in the TP53-altered HCT116 (–/–) tumor model. To test whether DOX could induce Src activity depending on the p53 status, the phosphorylation of Src was evaluated at the protein level. TP53-altered HCT116 (–/–) and TP53 wild-type HCT116 (+/+) cells were treated with DOX for 16 h, and the expression of pSrc was quantified by immunostaining. pSrc levels were found to be increased following treatment with 0.5-mM DOX in both cells (Figure 1(d)). DOX-induced Src activation was further evaluated in TP53 wild-type cells (HCT116+/+, U2OS) and TP53-altered cells (HCT116–/–), as well as TP53-null cells (MG63) by Western blotting. A higher level of Src phosphorylation was detected in TP53-altered cells and TP53-null cells, compared with wild-type TP53 cells (Figure 1(e)). Bliss analysis also showed the synergy of DOX/ SU6656 treatment in the TP53-altered cells, but there was antagonism in the wild-type TP53 cells (Figure 1(f)). In addition, another Src inhibitor AZD0530 also showed a significant synergy score dependent on TP53 status (Figure 1(g)).

TP53 gain and loss of function study When p53 was silenced in TP53 wild-type cells (U2OS), the synergy score was increased compared with the control cells (Figure 1(h)). An isobologram showed that AZD0530 possessed an antagonistic effect at a low dose of DOX. However, when p53 was silenced, AZD0530 had a synergic effect at all doses of DOX tested (Figure 1(h) lower panel). In contrast, when p53 was ectopically expressed in TP53-altered MG63 cells, the synergy score was decreased compared with the control (Figure 1(i)). Therefore, alterations in TP53 are a key mechanism for the synergistic effect of both agents.

A gain and loss of function study for Src kinase identifies a novel Src kinase/TP53 pathway To understand the role of Src kinase, we introduced a constitutively active Src mutant (c-Src Y530 mutant) into HCT116 cells. The activation of JNK and STAT3, which respectively mediate cell death and survival, were then evaluated using Western blot. The activated Src mutant induced the activation of both JNK and STAT3, independently of p53 expression. Phosphorylation of JNK was increased in wild-type TP53 cells compared with TP53-altered cells. However, STAT3 activation occurred in a converse manner (Figure 2(a)). Surprisingly, we found activated Src can also induce p53 phosphorylation, which promotes apoptosis with DOX treatment (Figure 2(b)). Using an Src inhibitor, SU6656, we further confirmed that inhibition of Src led to the inactivation of p53 and STAT3. Interestingly, DOX induced the activation of STAT3 in TP53-altered MG63 cells, but there was an inactivation

7 of STAT3 in U2OS cells expressing wild-type TP53. P53 is known to negatively modulate STAT3 phosphorylation and DNA binding activity in cancer cells.22 It appears that although STAT3 can be activated in a TP53 status-independent manner, STAT3 activation was repressed by the predominant activity of p53 in U2OS cells expressing wild-type TP53 (Figure 2(a) and (d)). As shown in Figure 2(c), the protein levels of pJNK increased in response to DOX exposure. To confirm activation of JNK which mediated cell death result, apoptosis-related proteins by DOX were observed by immunoblotting. The levels of active form of apoptotic marker, cleaved poly [ADP-ribose] polymerase (PARP) increased in response to treatment of DOX alone in U2OS cells, and the combination of DOX and SU6656 decreased the active form of PARP (Figure 2(e)), which could not be confirmed in p53altered MG63 cells. Next, we asked whether STAT3 was related to cell viability by DOX, depending on the state of TP53. Cells were transfected with an siRNAtargeting STAT3 to decrease its levels and were then treated with DOX at concentrations ranging from 0.25 to 4 mM for 24 h (Figure 2(f)). An MTT assay showed the effect of STAT3 expression levels on the growth of cells with an alteration in TP53 (HCT116–/– and MG63 cells). Our data suggest that DOX induces cell death in wild-type TP53 cells but increases resistance via STAT3 in cells with an alteration in TP53. Based on these data, we propose a model for the TP53-dependent biphasic role of Src (Figure 2(g))

Further testing of TP53-associated synergism with patient-derived cancer cells (PDCs) We found that Src inhibitors enhanced the anticancer effect of DOX in cancer cells with an alteration in TP53– but repressed the effect of DOX in cancer cells expressing wild-type TP53. Based on this result, we established a new treatment strategy based on TP53 status, and conducted a synergy test with PDCs to confirm the possibility of clinical application. First of all, CancerSCAN or Sanger sequencing was performed on patient-derived tissues to determine their TP53 status. TP53 wild-type and TP53-altered PDCs were defined as the control and experimental groups, respectively. There were a total of 26 PDCs that satisfied the conditions described in section ‘‘Materials and methods,’’ with 16 in the control group and 10 in the experimental group. To test the effect of combination treatment with DOX/AZD0530, the PDCs were treated with a combination of different doses of DOX (0, 0.25, 1, 2, and 4 mM) and AZD0530 (0, 0.25, 0.5, 1, 2, and 4 mM). Following the combination treatment, cell viability was measured to compare growth inhibition between control and experimental groups

8

Tumor Biology

Figure 2. p53-dependent biphasic effect of Src: (a) and (b) activation of JNK, p53, and STAT3 following CA-Src overexpression. Cells were transfected with a CA-Src expressing vector and were treated with 0.5 mM of DOX for 16 h or left untreated. (c) and (d) inactivation of JNK, p53, and STAT3 by SU6656. Cells were pre-treated with an Src inhibitor (SU6656) at a concentration of 25 mM for 1 h followed by treatment with 0.5-mM DOX for 16 h. A Western blot was performed to examine the activation of JNK, p53, and STAT3. (e) A Western blot was performed to examine the activation of the apoptotic indicators, cleaved PARP. b-Actin was used as an internal control. TP53 wild-type U2OS and p53-null MG63 cells were treated the combination of DOX (0.5 mM) and Src inhibitor (SU6656 25 mM) for 24 h. (f) Cell viability was analyzed using the MTT assay. STAT3 siRNA-transfected cancer cell lines were treated with DOX (0.25, 0.5, 1, 2, and 4 mM) for 24 h. Significant differences are marked by asterisks (**p\0.01, ***p\0.001). (g) Schematic diagram of the signal transduction pathways involved in DOX-induced cell death or survival.

Lee et al.

Figure 3. Synergistic effect of TP53. The significance of the difference between the TP53 wild-type and altered groups was compared using the Mann–Whitney test. Significant differences are marked by asterisks (***p\0.001).

using the Loewe synergy scores (Supplementary Figure 1, Supplementary Table1). The average synergy scores of the control group and experimental groups were 1.138 6 0.620 (DOX/ AZD0530) and 2.882 6 0.736 (DOX/AZD0530), respectively. In contrast to the control group, the antagonistic effect of the combination was observed in experimental group. The difference between two groups was evaluated using a non-parametric method and was found to be highly significant (***p \ 0.001; Figure 3). Especially, gastric cancer (#31, #34, #36) showed significantly higher synergy scores (Supplementary Table 1). Based on the PDCs study, we conclude that the combination of DOX/AZD0530 has a synergistic effect in cancers with alterations in TP53. In addition, TP53 status is a critical factor in predicting the response to DOX/AZD0530 combination therapy.

Discussion The mechanism of resistance to DOX has been widely studied and is thought to occur through an inherent resistance pathway.23 Several cell-signaling molecules have been found to be altered in DOX-resistant cells, such as p-glycoprotein, MAPK, and cyclin. Naturally DOX-resistant K562 cells were found to have a reduced resistance to DOX when a p-glycoprotein inhibitor was administered.24 MAPK inhibition had a similar effect because MAPK is known to be downstream of p-glycoprotein.25 Cyclin proteins stimulate cell growth, so DOX antagonizes the effect of cyclin due to its growth arrest function.26 However, signal transduction-related proteins are incredibly complicated, and it is difficult to justify a single mechanism of resistance.27

9 A role for Src in the mechanism of resistance to DOX has not been previously described; therefore, our study is the first to find that Src kinase is associated with DOX resistance in cancer cells. Src has been implicated in the transcriptional regulation of other transcription factors, which are important components of many signaling pathways. For example, Src has been shown to be associated with myogenin, MEF-2, transcriptional enhancer factor (TEF), NF-kappaB, AP-1, JNK, STAT, p53, and E2F1, which are known to be involved in many physiological processes.28 STAT3, a transcription factor with vital roles in tumors, can also be activated by Src.29 Here, we showed that DOX activates Src kinase and its downstream STAT3, which is closely linked to cell proliferation and survival pathways. Therefore, we think that DOX-induced Src kinase activity is a crucial event in DOX resistance. Indeed, inhibition of Src kinase successfully synergized the cytotoxic effect of DOX in a group of cancers that had alterations in TP53. However, in cancer cells expressing wild-type TP53, Src kinase inhibition did not show effective synergism with DOX. Numerous other studies have shown diverse outcomes with respect to the anticancer effect of Src kinase inhibitors, but the reason for this has not yet been elucidated. The mechanism we investigated here focused on the effect of the Src/p53/STAT3 pathway and its regulation of DOX resistance. This study has identified for the first time a significant subgroup of cancers (i.e. a TP53 mutant subgroup) that were sensitive to the Src kinase inhibitor. We also found that Src kinase activates p53, which explains why Src kinase inhibitors did not effectively synergize with the cytotoxic effect of DOX in TP53 wild-type cancers. Other studies have shown that Src kinase can induce the JNK/p53 pathway, with the JNK activation establishing a positive feedback loop with p53.30 From the data shown in Figures 2(c) and (d), we noticed that Src kinase inhibition decreased p53-induced apoptosis by inhibiting the phosphorylation of p53, which was unexpected. The p53 pathway is likely to be the predominant signaling pathway in TP53 wild-type cancers; therefore, the Src kinase–induced survival signal was not able to inhibit the cytotoxic effect of DOX. Instead, we hypothesize that this pathway supports the anticancer effect by activating the p53-induced apoptosis pathway. In this PDCs study, gastric cancer showed significantly higher synergy scores, and analysis with TCGA indicated that there are 47% mutations in TP53. We confirmed that Src inhibitors might be safe for use in cancer patients harboring an alteration in TP53, and tumors with high TP53 mutations are expected to be helpful in therapy. In contrast, it is obvious that Src inhibitors should be avoided in cancer patients

10

Tumor Biology

harboring wild-type TP53 because the chemotherapeutic effect of DOX is significantly reduced by the Src inhibitor. This study has a limitation in not evaluating side effects, such as cytotoxicity, in human body. Because normal cells have an intact TP53 gene in most patients, combination therapy may not critically affect normal tissues that are vulnerable to chemotherapeutic agents. Cancers have complex genetic mutations, and single tumors have genetic heterogeneity. Therefore, our simple therapeutic strategy based on TP53 status cannot be applied to all clinical cases. Nevertheless, this study suggests that TP53 status can predict if an Src inhibitor therapy would be effective in cancer, and we expect that this study will trigger more clinical and preclinical studies examining the crosstalk between TP53 status and chemotherapy. Acknowledgements Yun Sun Lee and Ji-Yoon Choi contributed equally to this study.

Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is partially supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education and Science Technology (MEST; grant no.: 2017092544).

ORCID iD Sung Wook Seo

https://orcid.org/0000-0001-9048-0367

References 1. Kawaguchi K, Igarashi K, Kiyuna T, et al. Individualized doxorubicin sensitivity testing of undifferentiated soft tissue sarcoma (USTS) in a patient-derived orthotopic xenograft (PDOX) model demonstrates large differences between patients. Cell Cycle 2018; 17: 627–633. 2. Davis T, van Niekerk G, Peres J, et al. Doxorubicin resistance in breast cancer: a novel role for the human protein AHNAK. Biochem Pharmacol 2018; 148: 174–183. 3. Liu W, Kovacevic Z, Peng Z, et al. The molecular effect of metastasis suppressors on Src signaling and tumorigenesis: new therapeutic targets. Oncotarget 2015; 6: 35522–35541. 4. Martin GS. The hunting of the Src. Nat Rev Mol Cell Biol 2001; 2: 467–475.

5. Elsberger B, Stewart B, Tatarov O, et al. Is Src a viable target for treating solid tumours? Curr Cancer Drug Targets 2010; 10: 683–694. 6. Irby RB and Yeatman TJ. Role of Src expression and activation in human cancer. Oncogene 2000; 19: 5636. 7. Kantarjian HM, Cortes JE, Kim D-W, et al. Bosutinib safety and management of toxicity in leukemia patients with resistance or intolerance to imatinib and other tyrosine kinase inhibitors. Blood 2014; 123: 1309–1318. 8. Pusztai L, Moulder S, Altan M, et al. Gene signatureguided dasatinib therapy in metastatic breast cancer. Clin Cancer Res 2014; 20: 5265–5271. 9. Gold KA, Lee JJ, Harun N, et al. A phase I/II study combining erlotinib and dasatinib for non-small cell lung cancer. Oncologist 2014; 19: 1040–1041. 10. Moy B, Neven P, Lebrun F, et al. Bosutinib in combination with the aromatase inhibitor letrozole: a phase II trial in postmenopausal women evaluating first-line endocrine therapy in locally advanced or metastatic hormone receptor-positive/HER2-negative breast cancer. Oncologist 2014; 19: 348–349. 11. Araujo JC, Trudel GC, Saad F, et al. Docetaxel and dasatinib or placebo in men with metastatic castrationresistant prostate cancer (READY): a randomised, double-blind phase 3 trial. Lancet Oncol 2013; 14: 1307–1316. 12. Holohan C, Van Schaeybroeck S, Longley DB, et al. Cancer drug resistance: an evolving paradigm. Nat Rev Cancer 2013; 13: 714. 13. Nowell PC:. The clonal evolution of tumor cell populations. Science 1976; 194: 23–28. 14. Petitjean A, Mathe E, Kato S, et al. Impact of mutant p53 functional properties on TP53 mutation patterns and tumor phenotype: lessons from recent developments in the IARC TP53 database. Hum Mutat 2007; 28: 622–629. 15. Lee YS, Choi YJ, Lee J, et al. TP53 alteration determines the combinational cytotoxic effect of doxorubicin and an antioxidant NAC. Tumor Biol 2017; 39: 1010428317700159. 16. Borisy AA, Elliott PJ, Hurst NW, et al. Systematic discovery of multicomponent therapeutics. Proc Natl Acad Sci U S A 2003; 100: 7977–7982. 17. Yeh PJ, Hegreness MJ, Aiden AP, et al. Drug interactions and the evolution of antibiotic resistance. Nat Rev Microbiol 2009; 7: 460. 18. Crafter C, Vincent JP, Tang E, et al. Combining AZD8931, a novel EGFR/HER2/HER3 signalling inhibitor, with AZD5363 limits AKT inhibitor induced feedback and enhances antitumour efficacy in HER2amplified breast cancer models. Int J Oncol 2015; 47: 446–454. 19. Zhang S, Huang W-C, Zhang L, et al. SRC family kinases as novel therapeutic targets to treat breast cancer brain metastases. Cancer Res 2013; 73: 5764–5774. 20. Zhou X, Tan M, Hawthorne VS, et al. Activation of the Akt/mammalian target of rapamycin/4E-BP1 pathway by ErbB2 overexpression predicts tumor progression in breast cancers. Clin Cancer Res 2004; 10: 6779–6788. 21. Cordero JB, Ridgway RA, Valeri N, et al. C-Src drives intestinal regeneration and transformation. EMBO J 2014; 33: 1474–1491.

Lee et al. 22. Lin J, Tang H, Jin X, et al. P53 regulates STAT3 phosphorylation and DNA binding activity in human prostate cancer cells expressing constitutively active STAT3. Oncogene 2002; 21: 3082–3088. 23. Volm M and Efferth T. Prediction of cancer drug resistance and implications for personalized medicine. Front Oncol 2015; 5: 282. 24. Dash TK and Konkimalla VB. Selection of P-glycoprotein inhibitor and formulation of combinational nanoformulation containing selected agent curcumin and DOX for reversal of resistance in K562 cells. Pharm Res 2017; 34: 1741–1750. 25. Chen Y, Zhao Y, Wang C, et al. Inhibition of p38 MAPK diminishes doxorubicin-induced drug resistance associated with P-glycoprotein in human leukemia K562 cells. Med Sci Monit 2012; 18: BR383–BR388.

11 26. Yao CX, Shi JC, Ma CX, et al. EGF protects cells against DOX-induced growth arrest through activating cyclin D1 expression. J Cell Biochem 2015; 116: 1755–1765. 27. Smith L, Watson MB, O’Kane SL, et al. The analysis of doxorubicin resistance in human breast cancer cells using antibody microarrays. Mol Cancer Ther 2006; 5: 2115–2120. 28. Li S and Shang Y. Regulation of Src family coactivators by post-translational modifications. Cell Signal 2007; 19: 1101–1112. 29. Sievers E, Trautmann M, Kindler D, et al. Src inhibition represents a potential therapeutic strategy in liposarcoma. Int J Cancer 2015; 137: 2578–2588. 30. Shi Y, Nikulenkov F, Zawacka-Pankau J, et al. ROSdependent activation of JNK converts p53 into an efficient inhibitor of oncogenes leading to robust apoptosis. Cell Death Differ 2014; 21: 612–623.