ATF3 and PRAP1 play important roles in cisplatin-induced damages in microvascular endothelial cells

Accepted Manuscript ATF3 and PRAP1 play important roles in cisplatin-induced damages in microvascular endothelial cells

Meifen Li, Guanghua Zhai, Xiuyu Gu, Kangyun Sun PII: DOI: Reference:

S0378-1119(18)30668-1 doi:10.1016/j.gene.2018.06.017 GENE 42950

To appear in:

Gene

Received date: Revised date: Accepted date:

30 November 2017 30 May 2018 6 June 2018

Please cite this article as: Meifen Li, Guanghua Zhai, Xiuyu Gu, Kangyun Sun , ATF3 and PRAP1 play important roles in cisplatin-induced damages in microvascular endothelial cells. Gene (2017), doi:10.1016/j.gene.2018.06.017

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Title page ATF3 and PRAP1 play important roles in cisplatin-induced damages in microvascular endothelial cells Running title: ATF3 and PRAP1 play roles in endothelial damage Authors: Meifen Li1*, Guanghua Zhai1*, Xiuyu Gu1 and Kangyun Sun2# *Contributed equally.

T

Affiliations:

IP

1 Department of Laboratory Medicine, the North District of Affiliated Suzhou Hospital, Nanjing

CR

Medical University, Suzhou 215008, China

2 Department of Cardiology, the North District of Affiliated Suzhou Hospital, Nanjing Medical

US

University, Suzhou 215008, China #corresponding author

AN

Kangyun Sun

Address: 242 Guangji Road, Suzhou 215008, Jiangsu, China

M

E-mail: [email protected]

AC

CE

PT

ED

Telephone number: +86-0512-62363516

1

ACCEPTED MANUSCRIPT Abstract Background: The early intervention is a rational approach to reduce the cardiovascular disease mortality in cancer patients. Here, we tried to identify potential biomarkers for the endothelial damage caused by cisplatin, a typical chemotherapy compound, and explore its underlying mechanisms. Methods: Microarray dataset GSE62523 were utilized to assess the gene differential expression

T

from human micro-vascular endothelial cells (HMEC-1) treated with cisplatin. Then, the

IP

potential key genes were further validated by qRT-PCR and the γH2AX level was evaluated to

CR

monitor the DNA damages caused by cisplatin.

Result: For the ‘acute-exposure’ settings that HMEC-1 were treated with 12.9 μM cisplatin for

US

6, 24 and 48 hours, ATF3, LRRTM2, VCAM1 and PAPPA were identified as potential key genes in endothelial damage, while for the ‘chronic-exposure’ settings that cells were exposed to

AN

0.52 μM cisplatin twice a week, SULF2, ACTA2 and PRAP1 were identified. In addition, further in vitro validation showed that knockdown of ATF3 attenuated the γH2AX level in cells

M

exposed to cisplatin for 6 or 24 hours and knockdown of PRAP1 increased the γH2AX level in cells exposed to cisplatin for 2 days. Notably, ATF3 has the ability to regulate the expression of

ED

HIST1H1D, FBXO6, APP, MDM2, STAT1 and TRAF1, while PRAP1 regulates YWHAB, MDM2, ISG15, LYN and CUL1 during cisplatin-induced DNA damage repair process.

PT

Conclusion: ATF3 and PRAP1 play important roles in cisplatin-induced DNA damage repair process. They may serve as potential early surrogate biomarkers of microvascular endothelial

CE

damage for cancer patients receiving chemotherapies.

AC

Keywords: ATF3; PRAP1; CDDP-specific; microvascular endothelial; biomarkers

2

ACCEPTED MANUSCRIPT Introduction Cisplatin (CDDP) is one of the most commonly used platinum based chemotherapy drug for various malignancies, such as ovary and lung carcinomas, lymphomas, sarcomas, and germ cell tumors (1). Even though cisplatin is well known in testicular germ cancer treatment and has been successfully used for tumor treatment, it leads to significant complications due to its

T

cardiotoxicity. Studies showed that cisplatin chemotherapy increased the risk of cardiovascular

IP

disease (CVD) for the testicular cancer survivors (2-4). A most recent study (5) reveals an

CR

elevated five-fold CVD mortality for chemotherapy patients within the first year compared to surgery alone patients. Additionally, heart failure is also observed in some breast cancer patients

US

receiving cisplatin treatment (6).

It is already known that cisplatin chemotherapy induces direct and indirect endothelial damage

AN

and impacts endothelial cell function, ultimately leading to the related CVD in cancer patients (7-10). Early intervention may reduce mortality and morbidity for CVD. However, up to now,

M

only a few studies focus on the cardiovascular disease biomarkers induced by chemotherapy, especially cisplatin specific biomarkers for cancer patients (11-13). Growth/differentiation factor

ED

15 (GDF-15) has been reported as a potentially useful biomarker related to endothelial damage during the combination of bleomycin-etoposide-cisplatin treatment of testicular cancer patients

PT

(13). The underlying mechanism for the specific cisplatin-induced endothelial damage is still

identified.

CE

largely unknown. Thus, biomarkers involved in cisplatin-induced endothelial damage need to be

AC

Activating Transcription Factor 3 (ATF3), encoded by an adaptive-response gene, plays an important role in cardiac remodeling processes (14), and protects against pressure overload heart failure (15), cardiac hypertrophy, and dysfunction (16). However, the function of ATF3 in cisplatin-induced CVD is still unknown. Very few studies focused on the proline-rich acidic protein 1 (PRAP1) gene, and the most recent study shows that it is involved in DNA damage repair pathway (17). We carried out this study to investigate the molecular markers involved in cisplatin-induced endothelial damage by analyzing gene expression profiles from human microvascular endothelial 3

ACCEPTED MANUSCRIPT cell line (HMEC-1) before and after cisplatin treatment for various time points and concentrations. The obtained differentially expressed genes (DEGs) were confirmed by

AC

CE

PT

ED

M

AN

US

CR

IP

T

quantitative Real Time PCR. Their functions in DNA damage repair were further evaluated.

4

ACCEPTED MANUSCRIPT Methods Data Collection Microarray dataset GSE62523 were retrieved from Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo/). In this dataset, human micro-vascular endothelial cells

T

(HMEC-1) were treated with cisplatin for various time period and dosage settings. Untreated

IP

cells were used as control. Briefly, in the ‘acute’ - exposure setting, HMEC-1 cells were exposed to 12.9 μM CDDP for 6, 24 and 48 hours; in the ‘chronic’ setting, 0.52 μM CDDP was

CR

administrated twice a week, and cells were collected at day 2, 16 or 30 respectively. Total RNA from each treated sample or corresponding control was extracted and processed for cDNA

US

microarray analysis (the custom-made 18K cDNA microarray). Unprocessed data (.txt files)

AN

were available online and collected for further analysis.

Data Processing and Differentially Expressed Genes (DEGs) Analysis

M

Package limma in Bioconductor was applied to assess the differential expression of microarray

ED

signal. Methods ‘normexp’, ‘loess’ and ‘Aquantile’ were used for background correction, withinarray normalization and between-array normalization respectively. In addition, these two-color

PT

microarray experiments were converted into single channel experiments for DEGs analysis. Six statistical comparisons were performed. The first three comparisons were carried out in the

CE

‘acute’ setting; comparison 1 was HMEC-1 with CDDP treatment for 6 hours (t=6) and corresponding untreated control; comparison 2 was 24 hours (t=24) treatment and corresponding

AC

untreated control; comparison 3 was 48 hours (t=48) treatment and corresponding untreated control. The other three comparisons were made in the ‘chronic’ setting; comparison 4 was HMEC-1 with CDDP treatment for two days (d=2) and corresponding untreated control, comparison 5 was 16 days (d=16) treatment and corresponding untreated control; comparison 6 was 30 days (d=30) treatment and corresponding untreated control. Differentially expressed genes were obtained with the threshold of |log2(fold change)| > 1 and p value < 0.05. The DEGs then were compared in each setting. GO and KEGG Pathway Analysis 5

ACCEPTED MANUSCRIPT ClusterProfiler (18) in R packages was utilized to detect Gene Ontology (GO) categories and Kyoto Encyclopedia of Genes and Genomes (KEGG; http://www.genome.jp/kegg/) pathways with significant over-representation compared to whole genome. The significantly enriched biological processes were identified as adjusted p value less than threshold value 0.01. As for KEGG pathway, threshold of adjusted p value was set as 0.05.

IP

T

Construction of biological network

Protein-protein interaction (PPI) databases from HPRD (19), BIOGRID (20) and PIP (21)

CR

databases were retrieved online. Pair interactions in any of these three databases were included in our curated PPI database. Cyto scape (22) was utilized to construct interaction network.

US

Interacted gene pairs existed in our curated PPI database was imported as stored network. After functional enrichment analysis, the DEGs specified in significantly altered biological processes

AN

(Gene Ontology terms) and KEGG pathways were mapped to corresponding networks

M

respectively for interaction analysis.

ED

Tissue culture, Quantitative Real Time PCR, siRNA transfection, and Western Blot HMEC-1 cells were purchased from the American Type Culture Collection (ATCC, Manassas VA); maintained in MCDB-131 medium supplemented with 10% Fetal Bovine Serum (FBS)

PT

(Thermo Fisher Scientific, Shanghai, China), 10 ng/mL human epidermal growth factor (R&D

CE

Systems, Abingdon, UK), 10 mM L-glutamine (Invitrogen, Merelbeke, Belgium), 1 µg/mL hydrocortisone (Sigma-Aldrich, Amsterdam, the Netherlands) at 37°C in a humidified incubator

AC

containing 5% CO2.

Total RNA was extracted with TRIZOL reagent (Invitrogen, Carlsbad, CA) for different time points of CDDP treated or untreated cells. First strand cDNA was synthesized from 1 µg of total RNA using ImProm-IITM Reverse Transcriptase System for RT-PCR (Promega, Madison, Wisconsin). qRT-PCR was performed with SYBR Green Mix on the Applied Biosystems® 7500 FAST system (Applied Biosystems, Foster City, CA) following a standard protocol. The primers for each gene were shown in supplementary Table 1. All experiments were performed in triplicates. The expression levels of the target genes were calculated using mean cycle threshold 6

ACCEPTED MANUSCRIPT (CT) values normalized relative to mean CT of the housekeeping gene GAPDH. The p-value was

AC

CE

PT

ED

M

AN

US

CR

IP

T

set at 0.05.

7

ACCEPTED MANUSCRIPT Results Identification of Genes Using Differential expression analysis To identify biomarkers for the cisplatin-induced endothelial damage, we compared the gene expression profiles of HMEC-1 cells treated with cisplatin for various time period and dosage

T

settings. The differential expression analysis of up-regulated (increase in expression) or down-

IP

regulated (decrease in expression) between treatment and control groups will aid in our understanding of the molecular mechanism(s) involved. Thus, we first explored the CDDP

CR

treatment of 6 h, 24 h or 48 h respectively, of which were included in acute settings. Total of 246 genes were identified differentially expressed in HMEC-1 with CDDP treatment for 6 hours

US

(t=6) and corresponding untreated control with statistically significance, among which 108 were up regulated and 138 were down regulated. In comparison 2, 24 hours (t=24) treatment vs

AN

corresponding untreated control, 350 genes were classified as DEGs, with up-regulation of 121 genes and down-regulation of 229 genes; 509 genes were identified as DEGs in comparison 3

M

(48 hours (t=48) CDDP treatment and corresponding untreated control), among which 387 were up regulated and 122 were down regulated. We also analyzed the expression level change in

ED

chronical settings. Twenty-five genes were classified as DEGs in comparison (two days (d=2) CDDP treatment and corresponding untreated control) with 23 genes up-regulated and two genes

PT

down-regulated; 416 genes were identified as DEGs in comparison 5 (16 days (d=16) treatment vs corresponding untreated control), among which 216 genes were up regulated and 200 genes

CE

were down regulated; 140 genes were identified in comparison 6 (30 days (d=30) treatment vs corresponding untreated control) as DEGs, with 95 genes up regulated and 45 genes down

AC

regulated (Fig. 1).

Four common DEGs, ATF3 (Activating Transcription Factor 3), LRRTM2 (Leucine Rich Repeat Transmembrane Neuronal 2), VCAM1 (Vascular Cell Adhesion Molecule 1) and PAPPA (Pregnancy-Associated Plasma Protein A, Pappalysin 1), were identified in all the ‘acute’ settings (Table 1). Among them, ATF3 and LRRTM2 were up regulated in CDDP treatment groups compared to control groups, while VCAM1 and PAPPA were down regulated. Three common DEGs, SULF2 (Sulfatase 2), ACTA2 (actin, alpha 2, smooth muscle, aorta) and PRAP1

8

ACCEPTED MANUSCRIPT (Proline-Rich Acidic Protein 1), were identified in all the ‘chronic’ settings and they were all up regulated (Table 2). No down regulated common DEGs were identified in chronic settings. Identification of Hub Genes Using Construction of PPI network and Venn diagrams To identify the proteins and biological modules that involved in pathophysiological process of

T

microvascular endothelial cells damage, the PPI network was constructed using DEGs. The

IP

biological networks were visualized in Cytoscape (22). For comparison 1, hub genes of the network were HIST1H1D, FBXO6 and VCAM1; for comparison 2, hub genes were APP and

CR

MDM2; for comparison 3, hub genes were MDM2, STAT1, MMP2 and TRAF1; for comparison 5, hub genes were CUL1, YWHAB, CAND1 and LYN; for comparison 6, hub genes were

US

MDM2 and ISG15, no results in PPI analysis for comparison 4 (Table 3 and 4).

AN

PPI networks were also constructed for DEGs only identified in single comparison (Fig. 2). For DEGs only identified in comparison 1, hub genes of the network were FBXO6 and SMURF2

M

(SMAD specific E3 ubiquitin protein ligase 2); for DEGs identified only in comparison 2, hub gene of the network was APP; for DEGs identified only in comparison 3, hub genes of the

ED

network were LYN, STAT1 and MMP2; for DEGs identified only in comparison 5, hub genes of the network were CUL1, YWHAB, CAND1 and LYN; for DEGs identified only in comparison

PT

6, hub genes of the network were MDM2 and ISG15.

CE

Identification of Target Pathways Using GO and KEGG pathway analysis To further understand the function and biological pathways of CDDP-induced differential

AC

transcription profile alteration over the time course, we performed GO biological process and KEGG pathway analysis for DEGs in all comparisons or DEGs existing only in one comparison. Pathways enriched at threshold p<0.01. In ‘acute’ settings, four pathways were enriched for comparison 1 (cellular process, cell differentiation, movement of cell or subcellular component and cellular response to stimulus); four pathways were enriched for comparison 2 (cell differentiation, regulation of cell proliferation, cell adhesion and cell differentiation); five pathways were enriched for comparison 3 (apoptotic process, cell differentiation, immune response, cell death and cell differentiation). 9

ACCEPTED MANUSCRIPT KEGG pathways analysis of DEGs in ‘acute’ settings identified ‘TNF signaling pathway’ for comparison 1; ‘p53 signaling pathway’ for comparison 2; p53 signaling pathway and TNF signaling pathway for comparison 3 (Table 5 and 6). In chronic groups, four pathways were enriched for comparison 4 (response to external stimulus, positive regulation of biological process, apoptotic process and developmental process) in GO

T

pathway analysis; two pathways were identified for comparison 6 in GO pathway analysis (cell

IP

communication and immune system process). In KEGG pathways analysis, ‘p53 signaling

CR

pathway’ and ‘Salivary secretion’ were identified for comparison 4; ‘Influenza A’ was identified for comparison 6 (Table 7 and 8).

US

Validation and Functional analysis of these Hub Genes, suggesting ATF3 and PRAP1 play

AN

important roles in cisplatin-specific endothelial damage

Quantitative Real Time PCR was performed to detect the expression change of these overlapping

M

genes in HMEC-1 cells with or without cisplatin treatment. For comparison 1, 2 and 3 the common DEGs were ATF3, LRRTM2, VCAM1 and PAPPA. After 6, 24 and 48 hours of

ED

exposure to cisplatin, the mRNA expression level of ATF3 and LRRTM2 was significantly increased (3-4.5 and 2-5.8 fold, respectively), consistent with the cDNA microarray data in the

PT

acute exposure setting (Fig. 3a and 3b). The mRNA expression of VCAM1 and PAPPA were significantly decreased (2 fold, respectively) under the same settings (Fig. 3c and 3d). Similarly,

CE

after 2, 16 and 30 days of exposure to cisplatin (same as the settings for comparisons 4, 5 and 6), the mRNA expression level of SULF2, ACTA2 and PRAP1 was significantly increased,

AC

consistent with the microarray results for the ‘chronic’ settings (Fig. 3e, 3f and 3g). To explore the possible biological implication of these genes in CDDP-induced DNA double strand damage (DSB) repair process, the overall γH2AX protein level (which is an indicator of unrepaired DSBs) was measured by western blot. As shown in Fig. 4, the γH2AX protein level was increased after the CDDP treatment in all settings. The knockdown of ATF3 by siRNA attenuated the γH2AX level in cells exposed to CDDP for 6 or 24 hours (Fig. 5a). The γH2AX level did not significantly change in LRRTM2 knockdown cells (Fig. 5b). In chronic settings, increased γH2AX protein level was only observed in PRAP1 knockdown cells 2 days after 10

ACCEPTED MANUSCRIPT CDDP treatment (Fig. 5c). No significant change of γH2AX protein level was detected for SULF2 or ACTA2 knockdown cells (Fig. 5d and 5e). To further evaluate the biological function of ATF3 and PRAP1, q-RT PCR was performed to quantify the expression level of other genes. As shown in Fig. 6, knockdown of ATF3 reversed the CDDP-induced alterations of HIST1H1D, FBXO6, APP, MDM2 and STAT1; while for

T

TRAF1, the expression level was significantly increased in ATF3 knockdown cells after CDDP

IP

treatment compared to untreated cells or mock knockdown cells. Knockdown of PRAP1 in

CR

HMEC-1 cells significantly increased the gene expression level of YWHAB, MDM2 and ISG15 (Fig. 7). Further decreased expression of LYN and CUL1 was observed in the PRAP1

AC

CE

PT

ED

M

AN

US

knockdown cells; no difference was observed for CAND1 (Fig. 7).

11

ACCEPTED MANUSCRIPT Discussion Despite a wealth of evidence outlining the increased cardiovascular diseases incidences associated with chemotherapy for cancer patients (23), the exact molecular mechanism underlying is still lacking. The early intervention targeting the related pathways is a rational approach to reduce the CVD risk. Several studies have suggested von Willebrand factor (vWF),

T

hsCRP, microalbuminuria and GDF-15 as biomarkers for endothelial damage in testicular cancer

IP

patients (2, 13, 24, 25). In current study, we carried out an unbiased analysis of the gene

CR

expression profile changes in CDDP treated human microvascular endothelial cell line (HMEC1) at various time points, and identified potential biomarkers for cisplatin-induced endothelial

US

damage to explore the underlying mechanisms.

The expression levels of ATF3 and PRAP1 were significantly changed after CDDP treatment in

AN

our cDNA microarray library analysis and the alterations of these two genes were consistent with our qRT-PCR results. GO and KEGG pathway analysis suggested several pathways for cluster

M

the DEGs from ‘acute’ or ‘chronic’ settings, including ‘TNF signaling’ and ‘p53 signaling’.

ED

ATF3, a member of the activation transcription factor/cAMP responsive element-binding (CREB) protein family of transcription factors, expressed ubiquitously and the level maintained

PT

low without stress stimuli (26). The cellular ATF3 mRNA level is increased rapidly upon exposure to various stress conditions, such as genotoxic agents, hyponutrition, serum factors,

CE

hypoxia, cytokines (19, 27-29). Our qRT-PCR results also detected this significant elevation after CPPD treatment for 6 h, 24 h and 48 h. Although studies found some pathways, such as the

AC

JNK, IL-6 and p53, involved in the activation of ATF3 under certain conditions, the direct impact of inducing ATF3 remains elusive (26). Kawauchi et al (30) found that the over expression of ATF3 prevents HUVEC endothelial cells from TNF-induced apoptosis. Similarly, increased cisplatin-induced apoptosis is observed in ATF3 knock-down cells (31). On the contrary, Tanaka et al (20) reported that ATF3 knockdown cell has higher viability after MMS treatment compared to controls. Several studies highlight ATF3 function as the downstream effector in p53-mediated cell death in stress response (21, 22, 32). Our results showed that the CDDP-induced γH2AX level was actually reduced in ATF3 siRNA knockdown HMEC-1 cells in the ‘acute’ 6 h and 24 h treatment settings, while no detectable change in LRRTM2 12

ACCEPTED MANUSCRIPT knockdown cells, suggesting the overall DNA damage induced by cisplatin was actually reduced in ATF3 knockdown cells compared to control or LRRTM2 knockdown groups. Thus the ATF3 is a pro-apoptotic gene in HMEC-1 cells. We further explored the function of ATF3 in regulation of other genes during chemotherapy, and the results showed that ATF3 activates some target genes but represses others during stress responses (20). Herein, we checked the effects of ATF3 down regulation on gene expression alteration for HIST1H1D, FBXO6, APP, MDM2, STAT1

T

and TRAF1. The expression of HIST1H1D and APP was reduced in cells under CDDP exposure

IP

and this reduction was partially reversed by inhibition of ATF3, while FBXO6, MDM2 and

CR

STAT1 were increased after CDDP treatment but significantly down regulated in ATF3 knockdown background. No difference was observed for the expression of TRAF1 gene treated

US

with or without CDDP for 48 h, however when ATF3 was knocked down, the TRAF1 expression was increased significantly, suggesting the TNF signaling pathway involved in the acute DNA

AN

repair process for CDDP damages. Our findings are consistent with the result of previous study that inhibition of ATF3 reduces γH2AX level and TRAF1 over expression inhibits antigen-

M

induced apoptosis (33).

The proline-rich acidic protein 1 (PRAP1) is a p53-responsive gene induced by genotoxic stress;

ED

and its expression level is significantly elevated after DNA damage (34). Here, we demonstrated that in HMEC-1 cells after exposed to CDDP, the expression of PRAP1 increased at least 3 fold.

PT

While knocking down of PRAP1 in CDDP treated cells leads to increased γH2AX level compared to control group, indicating more DNA damage accumulated over the time. Similar

CE

results were observed in 5-FU treated PRAP1 knockdown HCT-116 cells, which lead to reduced cell survival and 2.5-fold increased DNA damage (34). Together, our results suggested that loss

AC

of PRAP1 impaired cell DNA damage repair capacity and ultimately caused cell death. The regulation of other genes by PRAP1 was further explored by qRT-PCR. As showed in Fig. 4b, the ‘chronic’ CDDP exposure leads to down-regulation of CUL1 and LYN, and the expression of these two genes was further reduced after PRAP1 knockdown. The expression level of YWHAB, CAND1, MDM2 and ISG15 was elevated after CDDP treatment, except for CAND1, all other three genes expression was further increased significantly by PRAP1 knockdown. All these genes are involved in the p53 pathway (35), thus, the correlation between the chronic DNA

13

ACCEPTED MANUSCRIPT damage-caused low-grade endothelial damage in long-term cancer survivors and the activation of this pathway need to be further investigated. Conclusions In conclusion, several genes were significantly differentially expressed in HMEC-1 cells exposed to cisplatin: ATF3, LRRTM2, VCAM1 and PAPPA were identified in all the ‘acute-

T

exposure’ settings; SULF2, ACTA2 and PRAP1 were identified in all the ‘chronic-exposure’

IP

settings. Among them, ATF3 and PRAP1 were found involved in DNA damage repair process.

CR

Clusters of genes involved in ‘TNF signaling’ and ‘p53 signaling’ were affected. ATF3 and

AC

CE

PT

ED

M

AN

US

PRAP1 may be potential early surrogate biomarkers of endothelial damage.

14

ACCEPTED MANUSCRIPT Authors' contributions: ML, GZ and KS contributed to the conception and design; XG and KS contributed to administrative support and provision of study materials; ML performed data collection and assembly; GZ contributed to analysis and interpretation of data; ML, GZ, XG and KS contributed to manuscript writing; ML, GZ, XG and KS contributed to the final approval of manuscript.

T

Acknowledgements

IP

Funding: This research did not receive any specific grant from funding agencies in the public,

CR

commercial, or not-for-profit sectors.

AC

CE

PT

ED

M

AN

US

Conflicts of interest: none.

15

ACCEPTED MANUSCRIPT Reference

AC

CE

PT

ED

M

AN

US

CR

IP

T

1. Einhorn LH. Treatment of testicular cancer: a new and improved model. J Clin Oncol 1990;8:1777-81. 2. Meinardi MT, Gietema JA, van der Graaf WT, et al. Cardiovascular morbidity in long-term survivors of metastatic testicular cancer. J Clin Oncol 2000;18:1725-32. 3. Haugnes HS, Wethal T, Aass N, et al. Cardiovascular risk factors and morbidity in long-term survivors of testicular cancer: a 20-year follow-up study. J Clin Oncol 2010;28:4649-57. 4. Haugnes HS, Aass N, Fossa SD, et al. Predicted cardiovascular mortality and reported cardiovascular morbidity in testicular cancer survivors. J Cancer Surviv 2008;2:128-37. 5. Fung C, Fossa SD, Milano MT, et al. Cardiovascular Disease Mortality After Chemotherapy or Surgery for Testicular Nonseminoma: A Population-Based Study. J Clin Oncol 2015;33:3105-15. 6. Nieto Y, Cagnoni PJ, Bearman SI, et al. Cardiac toxicity following high-dose cyclophosphamide, cisplatin, and BCNU (STAMP-I) for breast cancer. Biol Blood Marrow Transplant 2000;6:198-203. 7. Lechner D, Kollars M, Gleiss A, et al. Chemotherapy-induced thrombin generation via procoagulant endothelial microparticles is independent of tissue factor activity. J Thromb Haemost 2007;5:2445-52. 8. Nuver J, De Haas EC, Van Zweeden M, et al. Vascular damage in testicular cancer patients: a study on endothelial activation by bleomycin and cisplatin in vitro. Oncol Rep 2010;23:247-53. 9. Yu M, Han J, Cui P, et al. Cisplatin up-regulates ICAM-1 expression in endothelial cell via a NFkappaB dependent pathway. Cancer Sci 2008;99:391-7. 10. Montiel M, Urso L, de la Blanca EP, et al. Cisplatin reduces endothelial cell migration via regulation of type 2-matrix metalloproteinase activity. Cell Physiol Biochem 2009;23:441-8. 11. Bissett D, Kunkeler L, Zwanenburg L, et al. Long-term sequelae of treatment for testicular germ cell tumours. Br J Cancer 1990;62:655-9. 12. Berger CC, Bokemeyer C, Schneider M, et al. Secondary Raynaud's phenomenon and other late vascular complications following chemotherapy for testicular cancer. Eur J Cancer 1995;31A:2229-38. 13. Altena R, Fehrmann RS, Boer H, et al. Growth differentiation factor 15 (GDF-15) plasma levels increase during bleomycin- and cisplatin-based treatment of testicular cancer patients and relate to endothelial damage. PLoS One 2015;10:e0115372. 14. Koren L, Alishekevitz D, Elhanani O, et al. ATF3-dependent cross-talk between cardiomyocytes and macrophages promotes cardiac maladaptive remodeling. International Journal of Cardiology 2015;198:232-40. 15. Lin H, Li HF, Chen HH, et al. Activating transcription factor 3 protects against pressure-overload heart failure via the autophagy molecule Beclin-1 pathway. Mol Pharmacol 2014;85:682-91. 16. Zhou H, Shen DF, Bian ZY, et al. Activating transcription factor 3 deficiency promotes cardiac hypertrophy, dysfunction, and fibrosis induced by pressure overload. PLoS One 2011;6:e26744. 17. Huang BH, Zhuo JL, Leung CHW, et al. PRAP1 is a novel executor of p53-dependent mechanisms in cell survival after DNA damage. Cell Death & Disease 2012;3. 18. Kohli S, Kohli M. Acute chemotherapy-induced cardiovascular changes in patients with testicular cancer: are there implications for blood pressure management in patients receiving chemotherapy? J Clin Oncol 2006;24:2399-400. 19. Fu M, Zhu X, Zhang J, et al. Egr-1 target genes in human endothelial cells identified by microarray analysis. Gene 2003;315:33-41. 20. Tanaka Y, Nakamura A, Morioka MS, et al. Systems analysis of ATF3 in stress response and cancer reveals opposing effects on pro-apoptotic genes in p53 pathway. PLoS One 2011;6:e26848. 21. Zhang C, Gao C, Kawauchi J, et al. Transcriptional activation of the human stress-inducible transcriptional repressor ATF3 gene promoter by p53. Biochem Biophys Res Commun 2002;297:130210. 22. Amundson SA, Bittner M, Chen Y, et al. Fluorescent cDNA microarray hybridization reveals complexity and heterogeneity of cellular genotoxic stress responses. Oncogene 1999;18:3666-72. 16

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN

US

CR

IP

T

23. van den Belt-Dusebout AW, Nuver J, de Wit R, et al. Long-term risk of cardiovascular disease in 5year survivors of testicular cancer. J Clin Oncol 2006;24:467-75. 24. Nuver J, Smit AJ, van der Meer J, et al. Acute chemotherapy-induced cardiovascular changes in patients with testicular cancer. J Clin Oncol 2005;23:9130-7. 25. Nuver J, Smit AJ, Sleijfer DT, et al. Microalbuminuria, decreased fibrinolysis, and inflammation as early signs of atherosclerosis in long-term survivors of disseminated testicular cancer. Eur J Cancer 2004;40:701-6. 26. Hai T, Wolfgang CD, Marsee DK, et al. ATF3 and stress responses. Gene Expr 1999;7:321-35. 27. Nawa T, Nawa MT, Adachi MT, et al. Expression of transcriptional repressor ATF3/LRF1 in human atherosclerosis: colocalization and possible involvement in cell death of vascular endothelial cells. Atherosclerosis 2002;161:281-91. 28. Koike M, Ninomiya Y, Koike A. Characterization of ATF3 induction after ionizing radiation in human skin cells. J Radiat Res 2005;46:379-85. 29. Kool J, Hamdi M, Cornelissen-Steijger P, et al. Induction of ATF3 by ionizing radiation is mediated via a signaling pathway that includes ATM, Nibrin1, stress-induced MAPkinases and ATF-2. Oncogene 2003;22:4235-42. 30. Kawauchi J, Zhang C, Nobori K, et al. Transcriptional repressor activating transcription factor 3 protects human umbilical vein endothelial cells from tumor necrosis factor-alpha-induced apoptosis through down-regulation of p53 transcription. J Biol Chem 2002;277:39025-34. 31. Hamdi M, Popeijus HE, Carlotti F, et al. ATF3 and Fra1 have opposite functions in JNK- and ERKdependent DNA damage responses. DNA Repair (Amst) 2008;7:487-96. 32. Fan F, Jin S, Amundson SA, et al. ATF3 induction following DNA damage is regulated by distinct signaling pathways and over-expression of ATF3 protein suppresses cells growth. Oncogene 2002;21:7488-96. 33. Speiser DE, Lee SY, Wong B, et al. A regulatory role for TRAF1 in antigen-induced apoptosis of T cells. J Exp Med 1997;185:1777-83. 34. Huang BH, Zhuo JL, Leung CH, et al. PRAP1 is a novel executor of p53-dependent mechanisms in cell survival after DNA damage. Cell Death Dis 2012;3:e442. 35. Soucy TA, Dick LR, Smith PG, et al. The NEDD8 Conjugation Pathway and Its Relevance in Cancer Biology and Therapy. Genes Cancer 2010;1:708-16.

17

ACCEPTED MANUSCRIPT Figure legends Figure 1. Overlap analysis to identify the common genes. For comparison 1, 2 and 3, the common DEGs were ATF3, LRRTM2 (up regulated), VCAM1 and PAPPA (down regulated); for comparison 4, 5 and 6, the common DEGs were SULF2, ACTA2, PRAP1 (up regulated).

T

Figure 2. Network of DEGs in comparison 1, 2, 3, 5, 6. a) Network of DEGs in comparison 1.

IP

Hub genes: HIST1H1D, FBXO6, VCAM1. b) Network of DEGs in comparison 2. Hub genes: APP, MDM2. c) Network of DEGs in comparison 3. Hub genes: MDM2, STAT1, MMP2,

CR

TRAF1. d) Network of DEGs in comparison 5. Hub genes: CUL1, YWHAB, CAND1, LYN. e).

US

Network of DEGs in comparison 6. Hub genes: MDM2, ISG15

Figure 3. Q-PCR validation of overlapping genes obtained from "acute" groups (comparison 1, 2

AN

and 3) in HMEC-1 cells with cisplatin treatment for 0, 6, 24, 48 h. a) The expression level of ATF3. b) The expression level of LRRTM2. c) The expression level of VCAM1. d) The

M

expression level of PAPPA. **p<0.01. Q-PCR validation of overlapping genes obtained from "chronic" groups (comparison 4, 5 and 6) in HMEC-1 cells with cisplatin treatment for 0, 6, 24,

level of PRAP1. **p<0.01.

ED

48 h. e) The expression level of SULF2. f) The expression level of ACTA2. g) The expression

PT

Figure 4. The γH2AX level in cisplatin treated cells at different times.

CE

Figure 5. The effect of ATF3, LRRTM2, PRAP1, SULF2 or ACTA2 absence on Cisplatininduced γH2AX level alterations. a) Cisplatin-induced γH2AX level alterations in ATF3 knock-

AC

down cells. b) Cisplatin-induced γH2AX level alterations in LRRTM2 knock-down cells. c) Cisplatin-induced γH2AX level alterations in PRAP1 knock-down cells. d) Cisplatin-induced γH2AX level alterations in SULF2 knock-down cells. e) Cisplatin-induced γH2AX level alterations in ACTA2 knock-down cells. * p<0.05, **p<0.01. Figure 6. Cisplatin-specific induced gene expression level change in ATF3 knock-down cells. a) The expression level of HIST1H1D. b) The expression level of APP. c) The expression level of STAT1. d) The expression level of FBXO6. e) The expression level of MDM2. f) The expression level of TRAF1. * p<0.05, **p<0.01. 18

ACCEPTED MANUSCRIPT Figure 7. Cisplatin-specific induced gene expression level change in PRAP1 knock-down cells. a) The expression level of CUL1. b) The expression level of YWHAB. c) The expression level of MDM2. d) The expression level of CAND1. e) The expression level of LYN. f) The

AC

CE

PT

ED

M

AN

US

CR

IP

T

expression level of ISG1. * p<0.05, **p<0.01.

19

ACCEPTED MANUSCRIPT Table 1. Expression of Common DEGs in HMEC-1 after Exposed to 12.9 μM Cisplatin (CDDP) for 6, 24 and 48 Hours Respectively

P Value

Adjusted

C ATF3

1.80

LogFC

P Value 1.27E-

1.33E-01

1.98

03 2.45

2 VCAM 1 PAPPA

6.41E-

9.53E-02

1.48

Value

P Value

7.13E

1.04E-02

1.74E

LogFC

1.42

2.71E-

2.15E-01

-1.28

1.16E -02

3.78E-

7.38E-02

-1.48

1.29E

04

-06

P

Adjusted

Value

P Value

3.27E-

1.15E-01

03 3.74E-01

-02

03 -1.20

Adjusted

-05

04 -1.74

P

3.06E-01

4.45E-04

US

LRRTM

Comparison3

T

LogF

Comparison2

IP

Symbol

Comparison1

CR

Gene

2.31

2.35E-

2.71E-02

04 -1.53

2.47E-

9.69E-02

03 -1.27

1.79E-

7.11E-03

05

AN

Footnotes: ATF3, Activating Transcription Factor 3; LRRTM2, Leucine Rich Repeat Transmembrane Neuronal 2; VCAM1, Vascular Cell Adhesion Molecule 1; PAPPA, Pregnancy-Associated Plasma

AC

CE

PT

ED

M

Protein A, Pappalysin 1.

20

ACCEPTED MANUSCRIPT Table 2. Expression of Common DEGs in HMEC-1 after Exposed to 0.52 μM CDDP for 2, 16 and 30 Days Respectively

LogFC

SULF2

1.40

P

Adjusted

Value

P Value

7.83E-

7.76E-01

LogFC

P Value

1.43

2.24

2.48E-

1.51

LogFC

1.53E-

8.79E-01

1.37E-01

1.93

1.86E-

5.30E-01

1.38

4.55E-

03

02

Adjusted

2.02

2.72E-

4.21E-01

03 3.13E-01

02

2.55E-

P Value

P Value

03

02 PRAP1

Adjusted P Value

03 ACTA2

Comparison6

T

Symbol

Comparison5

1.60

IP

Comparison4

4.51E-01

CR

Gene

1.63

3.24E-

9.82E-01

02 1.02E-

7.14E-01

02

US

Footnotes: SULF2, Sulfatase 2; ACTA2, actin, alpha 2, smooth muscle, aorta; PRAP1, Proline-Rich

AC

CE

PT

ED

M

AN

Acidic Protein 1.

21

ACCEPTED MANUSCRIPT Table 3. Expression of Hub Genes in HMEC-1 after Exposed to 12.9 μM Cisplatin (CDDP) for 6, 24 and 48 Hours Respectively

Symbol

LogFC

Comparison2

P

Adjusted

Value

P Value

HIST1H -1.28

3.36E-

7.52E-01

1D

02

FBXO6

1.75

1.06E-

LogFC

Comparison3

P

Adjusted

Value

P Value

4.55E-

2.05E-06

4.56E-01

-1.03

02 2.47

2.68E-

Adjusted

Value

P Value

3.41E-

3.39E-02

2.74E-02

US

MDM2

P

CR

02 APP

LogFC

T

Comparison1

IP

Gene

04

AN

STAT1

M

MMP2

04 -1.64

2.65E-

7.48E-04

07 1.05

6.19E-

1.59E-01

03

ED

TRAF1

1.09

Footnotes: HIST1H1D, histone cluster 1; FBXO6, F-box only protein 6; VCAM1, Vascular Cell

PT

Adhesion Molecule 1; APP, amyloid beta precursor protein; MDM2, MDM2 proto-oncogene, E3 ubiquitin protein ligase; STAT1, signal transducer and activator of transcription 1; MMP2, matrix

AC

CE

metallopeptidase 2; TRAF1, TNF receptor associated factor 1.

22

ACCEPTED MANUSCRIPT Table 4. Expression of Hub Genes in HMEC-1 after Exposed to 0.52 μM CDDP for 2, 16 and 30 Days Respectively Gene

Comparison 4

Symbol

Log FC

Comparison 5

P

Adjusted

Value

P Value

CUL1

Log FC

-1.11

Comparison6

P

Adjusted

Value

P Value

1.03E-

2.47E-01

LogFC

P

Adjusted

Value

P Value

3.35E-

9.94E-01

1.23

4.47E02

CAND1

2.01

3.60E04

1.16

1.58E-

2.91E-01

US

LYN

5.59E-02

CR

B

4.49E-01

IP

YWHA

T

02

02

AN

MDM2

M

ISG15

1.52

02 1.01

5.78E-

5.67E-01

03

(Note: There’re no PPI analysis results for comparison 4)

ED

Footnotes: CUL1, cullin 1; YWHAB, tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, beta; CAND1, cullin-associated and neddylation-dissociated 1; LYN, LYN proto-oncogene, Src

AC

CE

ubiquitin-like modifier.

PT

family tyrosine kinase; MDM2, MDM2 proto-oncogene, E3 ubiquitin protein ligase; ISG15, ISG15

23

ACCEPTED MANUSCRIPT Table 5. Selected GO Pathways for the Analysis of DEGs in Acute Groups GO Comparison 1 ID

Comparison 2 Term

ID

Overall GO0009987 cellular process

Comparison 3 Term

ID

GO0030154 cell

Term

GO0006915 apoptotic

differentiation

process

T

GO0030154 cell differentiation GO0042127 regulation of GO0030154 cell

IP

cell

differentiation

proliferation

or

CR

GO0006928 movement of cell GO0007155 cell adhesion subcellular

US

component

GO0051716 cellular response GO0030154 cell

differentiation

CE

PT

ED

M

AN

to stimulus

AC

Special

24

GO0006955 immune response GO0008219 cell death GO0030154 cell differentiation

ACCEPTED MANUSCRIPT Table 6. Selected KEGG Pathways for the Analysis of DEGs in Acute Groups KEGG Comparison 1 ID

Comparison 2 Term

ID

Comparison 3 Term

ID

Overall KEGG04668 TNF signaling KEGG04115 p53 signaling KEGG04115 pathway

pathway

Term p53 signaling pathway

AC

CE

PT

ED

M

AN

US

CR

IP

T

KEGG04668 TNF signaling

25

pathway

ACCEPTED MANUSCRIPT Table 7. Selected GO Pathways for the Analysis of DEGs in Chronic Groups GO Comparison 5

Overall

Comparison 6

ID

Term

ID

GO0009605

response

to

Term

external GO0007154

cell

Positive

regulation

of

IP

GO0048518

communication

T

stimulus

biological process GO0006915

GO0002376

developmental process

GO0007154

US

GO0032502

apoptotic process

CR

Special

AC

CE

PT

ED

M

AN

(Note: There’re no GO analysis results for comparison 4)

26

immune

system

process cell communication

ACCEPTED MANUSCRIPT Table 8. Selected KEGG Pathways for the Analysis of DEGs in Chronic Groups GO Comparison 4

Comparison 6 Term

ID

Term

Overall

hsa04115

p53 signaling pathway

hsa05164

Influenza A

Special

hsa04970

Salivary secretion

hsa05164

Influenza A

T

ID

AC

CE

PT

ED

M

AN

US

CR

IP

(Note: There’re no KEGG analysis results for comparison 5)

27

ACCEPTED MANUSCRIPT List of abbreviations

T

IP

CR

US

AN

M

ED

CE AC

CAND1 LYN SMURF2 vWF CREB

Gene Ontology Kyoto Encyclopedia of Genes and Genomes differentially expressed genes Cisplatin cardiovascular disease Activating Transcription Factor 3 proline-rich acidic protein 1 human microvascular endothelial cell Gene Expression Omnibus Protein-protein interaction American Type Culture Collection cycle threshold Leucine Rich Repeat Transmembrane Neuronal 2 Vascular Cell Adhesion Molecule 1 Pregnancy-Associated Plasma Protein A Sulfatase 2 histone cluster 1 F-box only protein 6 amyloid beta precursor protein signal transducer and activator of transcription 1 matrix metallopeptidase 2 TNF receptor associated factor 1 cullin 1 tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, beta cullin-associated and neddylation-dissociated 1 LYN proto-oncogene, Src family tyrosine kinase SMAD specific E3 ubiquitin protein ligase 2 von Willebrand factor cAMP responsive element-binding

PT

GO KEGG DEGs CDDP CVD ATF3 PRAP1 HMEC-1 GEO PPI ATCC CT LRRTM2 VCAM1 PAPPA SULF2 HIST1H1D FBXO6 APP STAT1 MMP2 TRAF1 CUL1 YWHAB

28

ACCEPTED MANUSCRIPT Highlights 1. ATF3 and PRAP1 play important roles in cisplatin-induced DNA damage repair process. 2. ATF3 regulates the expression of HIST1H1D, FBXO6, APP, MDM2, STAT1 and TRAF1.

AC

CE

PT

ED

M

AN

US

CR

IP

T

3. PRAP1 regulates YWHAB, MDM2, ISG15, LYN and CUL1.

29

Figure 1

Figure 2a

Figure 2b

Figure 2c

Figure 2d

Figure 2e

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7