biology
Article PER2 Circadian Oscillation Sensitizes Esophageal Cancer Cells to Chemotherapy
Juan Alfonso Redondo 1,†, Romain Bibes 1,†, Alizée Vercauteren Drubbel 1, Benjamin Dassy 1, Xavier Bisteau 1 , Eleonore Maury 2 and Benjamin Beck 1,*
1 Institute of Interdisciplinary Research (IRIBHM), Faculty of Medicine, Erasme Campus of Université Libre de Bruxelles (ULB), 808 Route de Lennik, 1070 Brussels, Belgium; [email protected] (J.A.R.); [email protected] (R.B.); [email protected] (A.V.D.); [email protected] (B.D.); [email protected] (X.B.) 2 Endocrinology, Diabetes and Nutrition Unit, Institute of Experimental and Clinical Research (IREC), Faculty of Medicine, Bruxelles-Woluwe Campus of Université catholique de Louvain (UCLouvain), 55 Avenue Hippocrate, 1200 Woluwe-Saint-Lambert, Belgium; [email protected] * Correspondence: [email protected]; Tel.: + 32-2-555-4162 † These authors contributed equally to this work.
Simple Summary: There are growing evidences that the circadian rhythm modulates key cellular processes in physiological and pathological conditions. Here, we characterized the consequences of the daily oscillations of the clock-related gene PER2 in esophageal cancer cells and found that chemotherapy is more efficient when PER2 expression is low. These results suggest that chronother- apy might be used to potentiate the impact of current chemotherapy regimen.
Citation: Redondo, J.A.; Bibes, R.; Abstract: Esophageal squamous cell carcinoma (eSCC) accounts for more than 85% cases of esophageal Vercauteren Drubbel, A.; Dassy, B.; cancer worldwide and the 5-year survival rate associated with metastatic eSCC is poor. This low Bisteau, X.; Maury, E.; Beck, B. PER2 survival rate is the consequence of a complex mechanism of resistance to therapy and tumor relapse. Circadian Oscillation Sensitizes To effectively reduce the mortality rate of this disease, we need to better understand the molecular Esophageal Cancer Cells to mechanisms underlying the development of resistance to therapy and translate that knowledge Chemotherapy. Biology 2021, 10, 266. into novel approaches for cancer treatment. The circadian clock orchestrates several physiological https://doi.org/10.3390/ processes through the establishment and synchronization of circadian rhythms. Since cancer cells biology10040266 need to fuel rapid proliferation and increased metabolic demands, the escape from circadian rhythm
Academic Editor: Gianluigi is relevant in tumorigenesis. Although clock related genes may be globally repressed in human eSCC Ubaldo Mazzoccoli samples, PER2 expression still oscillates in some human eSCC cell lines. However, the consequences of this circadian rhythm are still unclear. In the present study, we confirm that PER2 oscillations Received: 1 February 2021 still occur in human cancer cells in vitro in spite of a deregulated circadian clock gene expression. Accepted: 24 March 2021 Profiling of eSCC cells by RNAseq reveals that when PER2 expression is low, several transcripts Published: 26 March 2021 related to apoptosis are upregulated. Consistently, treating eSCC cells with cisplatin when PER2 expression is low enhances DNA damage and leads to a higher apoptosis rate. Interestingly, this Publisher’s Note: MDPI stays neutral process is conserved in a mouse model of chemically-induced eSCC ex vivo. These results therefore with regard to jurisdictional claims in suggest that response to therapy might be enhanced in esophageal cancers using chronotherapy. published maps and institutional affil- iations. Keywords: circadian clock; esophagus cancer; squamous cell carcinoma; apoptosis; chemother- apy; chronotherapy
Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. 1. Introduction This article is an open access article In complex multicellular organisms, the circadian clock orchestrates several physiolog- distributed under the terms and ical processes and behaviors through the establishment and synchronization of circadian conditions of the Creative Commons Attribution (CC BY) license (https:// rhythms [1,2]. The circadian clock is a cell autonomous process. It is composed of mul- creativecommons.org/licenses/by/ tiple transcription regulators that participate in interconnected transcriptional feedback 4.0/). loops to generate their own oscillation. The clock is driven by positive regulators CLOCK
Biology 2021, 10, 266. https://doi.org/10.3390/biology10040266 https://www.mdpi.com/journal/biology Biology 2021, 10, 266 2 of 18
and BMAL1, which regulate expression of Cryptochromes (CRY1 and CRY2) and Period (PER1, PER2, and PER3) [3]. Circadian clock components directly or indirectly regulate the expression of hundreds of genes in different cell types. This leads to daily rhythms in many cellular processes including nutrient metabolism, redox regulation, autophagy, DNA damage repair, protein folding, and cellular secretion [4]. It has been shown that nearly half of the mouse genome oscillates with circadian rhythm, highlighting the pivotal role of the circadian clock in cell biology [5]. Since daily rhythm in numerous cellular processes is a crucial part of homeostasis, it is not surprising that perturbation of the clock is associated to pathological conditions such as neurodegeneration [6], metabolic disorders [7], and cancer [8,9]. Since cancer cells need to fuel rapid proliferation and increased metabolic demands, the escape from circadian rhythm is relevant in tumorigenesis. Studies have shown that cancer cells commonly display a disrupted circadian rhythm or a deregulated expression of the circadian clock genes [10]. In the same line of thought, loss of circadian clock gene expression has been correlated with poor prognosis in various cancer types such as breast cancer [11], head and neck cancer [12], and gastric cancer [13]. This loss of the molecular clock may even participate in resistance to apoptosis [14,15] or immune evasion [2]. Nonetheless, this does not mean that circadian clock related genes do not play a role in cancer cell physiology. Indeed, other pieces of evidence have revealed that dysfunctions in circadian rhythms affect tumorigenesis and clock genes regulate several cancer hallmarks [16]. It still remains unclear in which cancers circadian rhythms are altered and whether it may be exploited to improve current therapeutic approaches. Notably, very little is known about the circadian clock in esophageal cancers. Esophageal cancer is the eighth most frequent cancer in the world and is associated to a poor survival rate due its aggressive nature. Esophageal carcinoma occurs as either squamous cell carcinoma (SCC) or adenocarcinoma. Esophageal SCC (eSCC) accounts for more than 85% cases of esophageal cancer worldwide [17] and the 5-year survival rate associated with metastatic esophageal cancer is poor (5% in the United States) [18]. This low survival rate is the consequence of a complex mechanism of resistance to therapy and tumor relapse. To effectively reduce the mortality rate in these patients, we need to better understand the molecular mechanisms underlying the development of this deadly disease and translate that knowledge into novel approaches for its early diagnosis, treatment, and prevention. Very little is known about the circadian clock in esophageal cancers. Recent studies have suggested that clock related genes might be globally repressed in human eSCC samples as compared to normal tissue [19]. In spite of these alterations, PER2 oscillations still occur in some esophageal cancer cell lines [19]. Nonetheless, the consequences of these oscillations are still unclear. In the present study, we show that clock related gene expression is altered in esophageal SCC cells. By using a bioluminescent reporter under the control of the PER2 promoter in esophageal cells, we have observed the oscillation of PER2 expression in human cancer cells in vitro in spite of the deregulated circadian clock gene expression. Characterization of esophageal cancer cells by RNAseq indicated that when PER2 expression is low, several transcripts related to the apoptotic signaling pathway are upregulated. To determine whether this oscillation would modify the response to therapy, we have treated eSCC cells with cisplatin. Strikingly, treating esophageal cancer cells with cisplatin for 4 h when PER2 expression is high or low has different outcomes. Indeed, eSCC cells undergo apoptosis more frequently when PER2 expression is low. Interestingly, this process is conserved in a mouse model of chemically-induced eSCC. These results therefore suggest that the response to therapy might be enhanced in esophageal cancers using chronotherapy.
2. Materials and Methods 2.1. The Cancer Genome Atlas Analysis To obtain public Esophageal Carcinoma RNA sequencing data and the corresponding clinical information The Cancer Genome Atlas program (TCGA) was used. Illuminahiseq_ Biology 2021, 10, 266 3 of 18
rnaseqv2-RSEM_genes and Clinical_Pick_Tier1.Level_4 have been downloaded from The Genome Data Analysis Centers (GDACs) (https://gdac.broadinstitute.org/) (Accessed date: 15 December 2020). Based on histological type, 95 « esophagus squamous cell carcinoma » have been selected and 11 « Normal Solid Tissue ». Total raw counts were then loaded on degust 4.1.1 [20]. All analyses were performed using EdgeR, TMM normalization and “Min gene read count” set at 10. Normal Solid Tissue samples were used as reference to calculate the fold change of gene expressions. To have an overview of clock gene expressions in different clinical variables, boxplot have been realized using CPM value of gene of interest and “pathologic stage” or “pathology N stage”. To clarify the tendency per “pathologic stage” and increased the number of sample per stage, we removed the letter “a”, ”b”, and ”c” and grouped the pathologic stage in “i”, “ii”, or “iii”. Finally, we applied Kruskal–Wallis test to highlight significant differences between the clinical groups. When a significant value was return, the pairwise multiple tests Tukey–Kramer have been performed to identify which group was different from the other. All processing of the data and analyses have been done with R version 3.6.3.
2.2. Cell Culture Human eSCC lines KYSE-140 (ACC348), KYSE-180 (ACC379) and KYSE-410 (ACC381) were cultured in RPMI 1640 medium (Gibco/Thermo Fisher Scientific, Waltham, MA, USA, 21875-034) supplemented with 10% fetal bovine serum (Gibco, 10270-106) and 1% penicillin/streptomycin (Gibco, 15070-063). Human eSCC lines KYSE-270 (ACC380) and KYSE-450 (ACC387) were cultured in 49% RPMI 1640 and 49% Ham’s F12 (Gibco, 11765- 054) supplemented with 2% fetal bovine serum and 1% penicillin/streptomycin. All KYSE lines were obtained from DSMZ (German Collection of Microorganisms and Cell Cultures Gmbh, Braunschweig, Germany) and cultured as suggested by the company. Primary human esophagus epithelial cells (H-6046) were obtained from CellBiologics (Chicago, IL, USA) and cultured in complete epithelial cell medium (H6621) as suggested by the company. Primary culture of FACS sorted YFP+ murine eSCC cells, were cultured in RPMI 1640 supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. All cells were cultured under humidified atmosphere with 5% CO2.
2.3. 4-NQO Mediated Induction of eSCC in Transgenic Mice 4-NQO (Sigma-Aldrich/Merck KGaA, Darmstadt, Germany, N8141) was dissolved in propylene glycol to a final concentration of 5 mg/mL and stored at −20 ◦C until use. When needed, this stock solution was further diluted 1/50 in distilled, sterile water to a final 4-NQO concentration of 100 mg/L. The expression of the genetic construction K5CreER:p53fl/fl:RosaYFP was activated through intraperitoneal injection of 2.5 mg of tamoxifen (BOC Science, Shirley, NY, USA, B0084-358326) in sterile sunflower seed oil in 8 weeks-old mice housed under SOPF condi- tions. One week after injection, the drinking water of the mice was replaced by the aqueous 4-NQO solution, in opaque bottles, for 8 weeks (changing the solution twice a week). After the treatment, distilled, sterile drinking water was restored, and the animals were weighed weekly and carefully inspected daily.
2.4. Tissue Digestion and FACS Isolation 4-NQO-induced esophagus SCC were dissected, minced and digested in 2 mg/mL of collagenase I (A. G. Scientific, San Diego, CA, USA, C-2823) for 1 h. Collagenase I activity was blocked by the addition of 5 mM EDTA and incubation for 15 min. Trypsin (0.125%) (Capricorn Scientific, Ebsdorfergrund, Germany, TRY-2B10) was then added for 15 min and then the cells were rinsed in PBS supplemented with 2% FBS. All incubations were done on a rocking plate at 37 ◦C. Immunostaining was performed on single cell suspension using PE-conjugated anti- CD45 (1:500, BioLegend, San Diego, CA, USA), PE-conjugated anti CD31 (1:500, BioLegend), PE-conjugated anti-CD140a (1:500, BioLegend) and Hoechst 33258 (1:10,000, Molecular Biology 2021, 10, 266 4 of 18
Probes/Thermo Fisher Scientific, Waltham, MA, USA, H3569), during 45 min at 4 ◦C on a rocking plate. Living epithelial cells were selected by forward scatter, side scatter, doublets discrimination and by Hoechst dye exclusion. YFP+/Lin- cells were selected based on the expression of YFP and the exclusion of CD45, CD31, CD140a (Lin-). Control esophageal cells were selected based on the expression of APC-Cy7 EpCam (1:250, BioLegend) and the exclusion of CD45, CD31, CD140a (Lin-). Fluorescence-activated cell sorting analysis was performed using FACSAria III and FACSDiva software (BD Biosciences, San Jose, CA, USA).
2.5. Circadian Cycle Synchronization Human KYSE-410 and murine eSCC cell lines were synchronized in the circadian cycle by addition of 1 µM dexamethasone (Sigma-Aldrich, D1756) for 1 h. Cells were then collected or treated 24 h post synchronization, at the top of PER2 expression, and 36 h post-treatment, at the lowest PER2 expression level. Mouse esophageal SCC cells were subjected to an additional cycle being also collected or treated 48 (high PER2 expression) and 60 (low PER2 expression) hours post-synchronization.
2.6. Bioluminescence Measurement and Data Analysis Esophageal cancer cells (KYSE-410 and 4-NQO) harbored a rapidly-degradable form of luciferase, dLuc, driven by the PER2 gene promoter through lentiviral infection (PER2- dLuc) [21] purchased from DNA/RNA Delivery Core, Northwestern University, us- ing Polybrene Infection Reagent (Millipore, Burlington, MA, USA TR1003G). The cells were maintained in a culture medium containing RPMI-1640 supplemented with 10% heat-inactivated FBS, 2 mM L-Glutamine, antibiotic antimycotic (all products from Life Technologies—Thermo Fisher Scientific, Waltham, MA, USA) and 2 µg/mL blasticidin (Sigma-Aldrich, 15205) to select for stable PER2-dLuc integration. Selected cells were then trypsinized (Trypsin-EDTA, Capricorn Scientific, TRY-1B10) and seeded in 3.5 cm dishes (1 × 105 cells/cm2 for luciferase assay the following day). Cells were synchronized with 1 µM dexamethasone for 1 h. Cells were washed twice with PBS and the medium was replaced with 1.2 mL of phenol-red free medium supplemented with 352.5 µg/mL of sodium bicarbonate, 10 mM HEPES, 2 mM L-Glutamine, 5% FBS, 25 units/mL penicillin, and 25 µg/mL streptomycin (all products from Life Technologies), and 0.1 mM luciferin potassium salt (Biosynth AG, Staad, Switzerland, L-8240). Sealed cultures (VWR, cover glass 631-0177; Sigma-Aldrich Dow corning high-vacuum silicone grease, Z273554-EA) were placed in a Kronos Dio luminometer (Atto, Tokyo, Japan, AB-2550) at 37 ◦C. Biolu- minescence was recorded continuously for 5 days; data were analyzed with the Kronos Dio software. Data were acquired and graphed using a custom-made script in RStudio. To measure the PER2-dLuc luminescence period from the time-series datasets, the “meta2d” function from R’ “MetaCycle” package (https://CRAN.R-project.org/package=MetaCycle) (Accessed date: 1 February 2019) [22] was used.
2.7. Cisplatin Treatment
Cisplatin EC50 value for the human KYSE-410 cell line was determined by incubating the cells in complete RPMI 1640 with 2.5 to 50 µM of cisplatin (Selleckchem, Munich, Germany, S1166) for 48 h at 37 ◦C. Cell viability was measured by MTS assay (Promega, Madison, WI, USA, G3581) 48 h after treatment following the instructions from the manufacturer. Both cell lines were treated by the addition of 50 µM cisplatin to complete RPMI 1640 medium. Cells were incubated in the presence of cisplatin or the vehicle (NaCl) for 4 h at 37 ◦C.
2.8. Annexin V FACS Analysis 48 h after chemotherapy, the culture medium was removed and the cells were trypsinized (trypsin-EDTA). The removed medium was used to stop the trypsinization en- suring that the whole cell population was recovered. Cells were pelleted by centrifugation at 600× g for 5 min and the supernatant was discarded. The pellets were resuspended in 1X HBSS buffer (Gibco, 14025-050) containing 1:100 of Annexin V APC (Invitrogen—Thermo Biology 2021, 10, 266 5 of 18
Fisher Scientific, Waltham, MA, USA, 17-8007-74) and 1:10,000 of Hoechst 33258 dye. After incubation at room temperature on a rocking platform for 30 min in the dark, the cells were further diluted in 1x HBSS and analyzed. FACS analysis was performed using a BD LSRFortessa X-20 and FACSDiva software (BD Biosciences).
2.9. Immunofluorescence Microscopy eSCC cells were directly fixed with 10% neutral buffered formalin (VWR, 11699455) for 5 min at room temperature. After 3 washes of 5 min with PBS, all fixed cells were blocked with PBST-BSA [PBS containing 0.2% Triton X-100, 2% BSA (Sigma-Aldrich, A7906)] supple- mented with 10% normal horse serum (Capricorn Scientific, HOS-1B). Primary antibodies (GFP from chicken, GenScript, A01694-40, 1:500; Krt 14 from chicken, Biolegend, 906004, 1:4000; p63 from rabbit, Abcam, ab124762, 1:1000) were diluted in PBST-BSA and incubated overnight at 4 ◦C. Cells were washed 3 times with PBS for 5 min and incubated with secondary antibody (Jackson Immunoresearch, West Grove, PA, USA, 1:500) and 1:1,000 of Hoechst dye for 1 h in the dark at room temperature. After washing, samples were mounted with Immu-Mount (Thermo Fisher Scientific, 9990402). Images were acquired using an Epifluorescence Zeiss Axioimager Z1 microscope equipped with Zeiss AxioCam MRc5 and a 20x objective (Plan-Neofluar, 20×/0.5; Zeiss, Jena, Germany). Acquired CZI images were then processed using ImageJ.
2.10. Immunoblot Cells were lysed in Laemmli 1x sample buffer. Lysates were centrifuged for 10 min at 13,000 RPM at RT. 10-30 µg of protein extract were separated on polyacrylamide gels and transferred onto polyvinylidene difluoride membranes (PVDF, Millipore, Burlington, MA, USA, IPVH0010) using a wet system. The membranes were blocked in tris-buffered saline (TBS) with 0.1% Tween20 and 5% skim milk (VWR, 84615). Blots were probed with the appropriate primary antibodies overnight at 4 ◦C, followed by secondary goat anti-mouse (Cell Signaling Technologies, 7076) or anti-rabbit antibodies (Cell Signaling Technologies, 7074) conjugated to horseradish peroxidase and developed using enhanced chemilumines- cence (PerkinElmer, Waltham, MA, USA, NEL104001EA). Images were obtained on a Vilber Lourmat Fusion Solo S. Primary antibodies used: Anti-PER2 (Abcam, Cambridge, United Kingdom, ab180655) and anti-γH2AX (p Ser139) (Cell Signaling Technologies, Danvers, MA, USA, 9718).
2.11. mRNA Extraction and RT-qPCR RNA extraction was performed using the MicroElute Total RNA kit (Omega Biotek, Norcross, GA, USA, R6831-02) according to the manufacturer’s recommendations. Purified RNA was used to synthesize the first strand cDNA using Superscript II (Invitrogen) and random hexamers. RT-qPCR analyses were performed with 4 ng of cDNA as template, using a SYBR Green Mix (Applied Biosystems, Foster City, CA, USA, 100029284) and a QuantStudio 3 (Applied Biosystems) real-time PCR System. Primers: Ms_Actb Fw(5->3) CACTGTCGAGTCGCGTCC Rv(5->3) TCATCCATGGCGAACTGGTG Ms_Per2 Fw(5->3) CCACTATGTGACAGCGGAGG Rv(5->3) CTCTGGAATAAGCGCTTCGC Ms_Arntl Fw(5->3) GAGCGGATTGGTCGGAAAGTA Rv(5->3) TCTTCCAAAATCCAATGAAGGC Relative quantitative RNA was normalized using the housekeeping gene β-actin. Primers were designed using NCBI Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/ primer-blast/) (Accessed date: 1 December 2020). Analysis of the results was performed using QuantStudio Design and Analysis Software v1.4 (Applied Biosystems) and relative quantification was performed using the DDCt method using β-actin as reference. Biology 2021, 10, 266 6 of 18
2.12. RNAseq and Analysis of Bulk Samples RNA quality was checked using a Fragment Analyzer 5200 (Agilent technologies, Santa Clara, CA, USA). Indexed cDNA libraries were obtained using the Ovation Solo RNA-Seq Library Preparation Kit (Tecan, Männedorf, Switzerland) following manufac- turer’s recommendations. The multiplexed libraries were loaded on a NovaSeq 6000 (Illumina, San Diego, CA, USA) using a S2 flow cell and sequences were produced using a 200 Cycle Kit. Paired-end reads were mapped against the mouse reference genome GRCm38 using STAR software (version 2.5.3a) to generate read alignments for each sample. Annotations Mus_musculus.GRCm38.90.gtf were obtained from ftp.Ensembl.org (accessed on 15 December 2020). For human KYSE lines, paired-end reads were mapped against the human reference genome GRCh38 using STAR software (version 2.5.3a) to generate read alignments for each sample. Annotations Homo_sapiens.GRCh38.90.gtf were obtained from ftp.Ensembl.org (accessed on 15 December 2020). After transcripts assembling, gene level counts were obtained using HTSeq-0.9.1 [23]. Total raw counts were loaded on degust 4.1.1 [20]. All analyses were performed using EdgeR, TMM normalization and “Min gene read count” set at 10. Control esophagus samples were used as reference to calculate the fold change of gene expressions. Volcano plots were generated using the package “EnhancedVolcano” [24] from Bioconductor in R version 3.6.3.
2.13. Gene Set Enrichment Analysis (GSEA) Analysis GSEA analysis was performed using preranked gene set enrichment analysis from the fgsea package [25] in R version 3.6.3, with “nperm = 1000” and “maxSize = 500”. The values of LFC were used as the ranking metric. For this purpose, thresholds were enlarged to include more genes in the analysis and therefore increase the strength of the analysis. The human C5 collection which contains gene sets annotated by GO terms have been downloaded (http://bioinf.wehi.edu.au/software/MSigDB/[ 3]) (Accessed date: 10 January 2021) and all gene sets have been tested. For GSEA analysis, ranked fold change values correspond to 36 h over 24 h post-synchronization (abs(LFC) > 1, p-value < 0.05).
3. Results 3.1. Expression of Clock-Related Genes Is Altered in Human eSCC Samples First, to determine putative alteration of the clock related genes in eSCC, we have analyzed publicly available data from the Cancer Genome Atlas (TCGA). Comparison between the 95 available eSCC samples and the 11 samples of normal esophagus unveiled a significant downregulation of CLOCK as well as members of the Period (PER) and Cryptochrome (CRY) families (Figure1A). Other transcripts were either not modified (ARNTL, NR1D2 and RORA) or upregulated (NR1D1 and TIMELESS). Then, we aimed at characterizing the impact of cancer progression on the pattern of clock related genes. To this end, we sorted TCGA eSCC samples based on their pathological classification (i, ii, iii, and iv). To summarize, stage i describes small tumors without local invasion (n = 7), stage ii (n = 55) and iii (n = 27) describe invasive cancers potentially invading lymph nodes, and stage iv describes cancers with distant metastases (n = 4). We have investigated the expression of PER1, PER2, PER3, CRY1, CRY2, ARNTL (BMAL1), TIMELESS, NR1D1 (REVERBα), NR1D2 (REVERBβ), CLOCK, and RORA (RORα) in samples of eSCC at different stages (Figure1B). We did not observe a significant downregulation of clock related genes in advances stages as compared to small, localized tumors. A detailed representation of the expression of these genes depending on pathological classification is provided in the Supplementary Figure S1. In the same line, we did not find a significant downregulation of clock related genes depending on the criterion of lymph node invasion (pathological N stage) (Supplementary Figure S2). Biology 2021, 10, 266 7 of 18 6 A 500 − 1.45e TCGA samples 400 13 5 − −
normal esophagus 300 N.S 0.0009 0.002 29 − 4.92e 6.93e eSCC N.S 6
(CPM) 200 − 1.60e 0.0081
100 1.86e N.S mRNA expression 0
PER1 PER2 PER3 ARNTL CRY1 CRY2 RORA CLOCK NR1D1 NR1D2 TIMELESS
B CRY1 CRY2 PER1 p=0.047 p=0.217 p=0.812 0.031 80 40 300
eSCC stages 60 30 Stagei (n=7) 200 Stage ii (n=55) 40 20 Stage iii (n=27) 100 Stage iv (n=4) 20 10 i ii iii iv i ii iii iv i ii iii iv mRNA expression (CPM) Pathological stage PER2 PER3 ARNTL TIMELESS 200 p=0.971 200 p=0.911 40 p=0.183 p=0.162 200 30
100 100 20 100
10
0 0 i ii iii iv i ii iii iv i ii iii iv i ii iii iv mRNA expression (CPM) Pathological stage NR1D1 NR1D2 CLOCK RORA p=0.821 p=0.342 p=0.956 p=0.566
200 100 150 200 150 150 100 60 100 100 50 50 50 20 0 i ii iii iv i ii iii iv i ii iii iv i ii iii iv mRNA expression (CPM) Pathological stage
Figure 1. Expression of clock-related genes in human esophageal squamous cell carcinoma (eSCC) samples. (A) Expression of clock related genes measured by RNA sequencing in biopsies from control esophagus (n = 11) and eSCC (n = 95). Data are represented as boxplots. False discovery rate (FDR) is indicated for each gene. Benjamini–Hochberg method on the p-values was used to control the FDR. (B) Expression of clock related genes measured by RNA sequencing in biopsies from eSCC samples depending on their pathological stage. Data are represented as boxplots. Statistics were calculated using a Kruskal–Wallis test followed by a Tukey–Kramer test. p-values < 0.05 are considered as significant. Biology 2021, 10, 266 8 of 18
3.2. PER2 Expression Oscillates in Human Esophageal SCC To determine the pattern of clock related genes more specifically in esophageal SCC epithelial cells, we have used RNA sequencing to profile a commercially available non- immortalized human esophagus epithelial cell line as well as 5 different esophageal SCC cell lines from the KYSE series: KYSE-140, −180, −270, −10 and −450 [20] (Figure2A). Using a threshold of false discovery rate (FDR < 0.01), we could find several deregulated genes in cancer cells (Figure2B–F). Interestingly, instead of a global downregulation of all clock related genes as previously reported in human SCC samples from different organs [19], we found both upregulated and downregulated transcripts in esophageal cancer cell lines. This pattern also differed from the one found in the eSCC samples from the TCGA. Moreover, the 5 eSCC lines are different from each other. For instance, we observed a significant upregulation of PER2 and PER3 in 4 lines out of 5. However, other mRNAs such as ARNTL (coding for BMAL1 protein) is upregulated in some lines (KYSE- 270 and −450), downregulated in another (KYSE-410 and 180) or not modified (KYSE-140) (Figure2B–F). To determine whether esophageal cancer cells have somehow conserved a circadian rhythm in spite of these modifications, we then focused our analysis on the KYSE-410 eSCC line. KYSE-410 cells are established from the poorly differentiated invasive esophageal squamous cell carcinoma resected from the cervical esophagus of a 51-year-old Japanese man prior treatment [26]. Co-immunostaining showed that these cells are characterized by p63 and Krt14 expression as expected from eSCC cells (Figure2G). We transduced a PER2: Luc construct in KYSE-410 cells using lentiviral particles (Figure2H). In this model, a 1-h dexamethasone treatment is sufficient to synchronize the cells and trigger oscillations of the luciferase activity that can be monitored for several days (Figure2I). These results are consistent with those recently published with another cell line from the KYSE series [19]. Our results thus suggest that in spite of an altered expression of the clock related genes, the circadian oscillation of PER2 expression is still active in human eSCC cells.
3.3. The Level of PER2 Expression Is Associated to Transcription Modifications in Esophageal Cancer Cells To determine the impact of PER2 oscillations on esophageal cancer cells, we have synchronized the KYSE-410 cells using dexamethasone and profiled the cells when PER2 expression is high or low (24 and 36 h after synchronization respectively) (Figure3A). We found hundreds of genes significantly deregulated between these 2 time points (Log2 Fold of change > 0.65 and p-val <0.05), 482 transcripts were upregulated at 36 h as compared to 24 h, and 442 were downregulated (Figure3B). Of note, PER1 was the most significant downregulated transcript at this timepoint. Gene set enrichment analysis (GSEA) suggested that multiples processes were impacted depending on PER2 expression (Figure3C,D). Notably, we found that “regulation of nucleotide metabolic process” and “regulation of cyclic nucleotide metabolic process” were among the most significant negatively enriched gene sets, thus indicating that transcripts related to these biological processes were downregulated when PER2 expression is low. Inversely, “apoptotic signaling pathway” gene set related transcripts were enriched when PER2 expression is low (Figure3C,D). These data suggest that depending on the circadian clock and PER2 expression, eSCC cells may be more or less sensitive to pro-apoptotic signals. These results are in line with a previous work that showed that CRY1/2 and PER1 influence the extrinsic TNFα-dependent pathway and intrinsic apoptotic pathways in cancer cells [15]. Biology 2021, 10, 266 9 of 18
4 KYSE-140 A CT eso cells (Hu) B primary eso epith 2 N.S N.S N.S N.S
eSCC cells (Hu) N.S 0 RNA seq 5
KYSE-140 − N.S N.S N.S KYSE-180 N.S 1.67e Log2 FC over CT mRNA expression KYSE-270 −2 N.S KYSE-410 KYSE-450 PER1PER2PER3 ARNTLCRY1CRY2 RORA CLOCK NR1D1NR1D2 TIMELESS C 4 KYSE-180 D 4 KYSE-270
2 2 3 6 11 27 4 − − − − − N.S N.S 1.01e 1.37e 4.58e 2.35e 0 4.49e 0 5 9 5 4 11 59 12 − − − − − − − N.S N.S N.S N.S N.S N.S N.S N.S 1.17e Log2 FC over CT Log2 FC over CT 4.61e 6.11e 1.26e mRNA expression mRNA expression 4.58e 3.56e −2 −2 1.07e
PER1PER2PER3 ARNTLCRY1CRY2 RORA PER1PER2PER3 ARNTLCRY1CRY2 RORA CLOCK NR1D1NR1D2 CLOCK NR1D1NR1D2 TIMELESS TIMELESS E 4 KYSE-410 F 4 KYSE-450
2 2 86 5 3 7 − − 4 − − − N.S N.S N.S N.S 6.6e 1.64e 2.81e 5.8e 0 1.96e 0 11 46 14 10 16 25 181 191 150 − − − − − − N.S 0.01 − − − N.S N.S Log2 FC over CT Log2 FC over CT mRNA expression mRNA expression 9.96e 4.18e 3.36e 2.77e 1.43e 1.92e 1.13e 8.47e −2 −2 1.41e
PER1PER2PER3 ARNTLCRY1CRY2 RORA PER1PER2PER3 ARNTLCRY1CRY2 RORA CLOCK NR1D1NR1D2 CLOCK NR1D1NR1D2 TIMELESS TIMELESS G merge p63 krt14 KYSE-410 (eSCC)
KYSE-410 H I 1
Dexa luciferase monitoring Per2 Luc 0.5 0 24 48 72 96 120 Time (hours)
0 0 24 48 72 96 120
Luminescence (A.Ux1000) Time, hours
Figure 2. PER2 expression oscillates in human eSCC cells in spite of alterations in the pattern of clock-related gene expression. (A) Experimental design. (B) Histogram depicting the expression of clock related genes measured by RNA sequencing in the KYSE-140 human eSCC cell line. Data are represented as the mean of a biological duplicate. FDR is indicated for each gene. Benjamini–Hochberg method on the p-values was used to control the false discovery rate (FDR). (C) Same as in (B) in the KYSE-180 human eSCC cell line. (D) Same as in (B) in the KYSE-270 human eSCC cell line. (E) Same as in (B) in the KYSE-410 human eSCC cell line. (F) Same as in (B) in the KYSE-450 human eSCC cell line. (G) Co-immunostaining of p63 and Krt14 in the KYSE-410 human eSCC cell line. (H) Experimental design to monitor the activity of the PER2 promotor in KYSE-410 cells. (I) Real time monitoring of luciferase activity in the KYSE-410 human eSCC cell line. FDR>5% are considered non-significant (N.S). Biology 2021, 10, 266 10 of 18
Biology 2021, 10, x FOR PEER REVIEW 10 of 18
AB30 RNA sequencing PER1
20 -Log10(P) Dexa
10 NR1D1
02436 0 Time (hours) −6 0 6 Log2 FC 36h vs 24h
C Apoptotic signaling Regulation of nucleotide pathway metabolic process 0.8 NES 0 0.6 +1.72 −0.2 0.4 −0.4 0.2 −0.6 NES 0 −1.67
Enrichment score Enrichment score −0.8 0 250 500 750 0 250 500 750 Rank Rank D KYSE-410 36h vs 24h - Gene Ontology NES p-val GO_APOPTOTIC_SIGNALING_PATHWAY +1.72 0.020 GO_REGULATION_OF_NUCLEOTIDE_METABOLIC_PROCESS −1.67 0.013 GO_REGULATION_OF_CYCLIC_NUCLEOTIDE_METABOLIC_PROC. −1.67 0.013 GO_HYDROLASE_ACTIVITY_ACTING_ON_ESTER_BONDS −1.77 0.006 −2.01 0.002 GO_HORMONE_ACTIVITY FigureFigure 3. 3.PER2 PER2 expressionexpression oscillationsoscillations areare associated associated to to modifications modifications of of the the transcriptome. transcriptome.( (AA)) Experimental Experimental design. design. ((BB)) VolcanoVolcano plotsplots representing representing results results of of RNA-seq RNA-seq as as the the statistical statistical significance significance versus versus the the magnitude magnitude of fold of fold of change of change (FC) (FC) in KYSE-410 in KYSE- compared410 compared to control to control esophageal esophageal cells. cells. Data Data are are filtered filtered based based on on FC FC (abs(LFC) (abs(LFC) > 1)> 1) and andp-value p-value (p (
1, p-value < 0.05). (D) Table summarizing the results of the GSEA. Pathways with a p-value < 0.05 are (abs(LFC) > 1, p-value < 0.05). (D) Table summarizing the results of the GSEA. Pathways with a p-value < 0.05 are listed and listed and sorted based on normalized enrichment score. sorted based on normalized enrichment score. 3.4. Efficiency of Cisplatin-Induced Apoptosis Depends on the Level of PER2 Expression 3.4. Efficiency of Cisplatin-Induced Apoptosis Depends on the Level of PER2 Expression To determine whether the oscillations of PER2 expression might indeed regulate To determine whether the oscillations of PER2 expression might indeed regulate apoptosis in eSCC cells, we have treated cells with cisplatin. First, we have tested several apoptosis in eSCC cells, we have treated cells with cisplatin. First, we have tested several concentrations of cisplatin on KYSE-410 cells to assess their sensitivity to the drug. We concentrations of cisplatin on KYSE-410 cells to assess their sensitivity to the drug. We have have treated KYSE-410 cells for 48 h with different concentrations of cisplatin (from 2.5 to treated KYSE-410 cells for 48 h with different concentrations of cisplatin (from 2.5 to 50 µM) and50 µM) measured and measured cell survival cell usingsurvival MTS using assay MTS (Figure assay4A). (Figure This experiment4A). This experiment indicated thatindi- cated that the EC50 for cisplatin in KYSE-410 cells is around 20 µM (Figure 4B). Since we the EC50 for cisplatin in KYSE-410 cells is around 20 µM (Figure4B). Since we had to treat eSCChad to cells treat for eSCC a short cells period for a short of time period (4 h), of we time have (4 chosenh), we have to use chosen a concentration to use a concentra- that was sufficienttion that was to kill sufficient the majority to kill of the KYSE-410 majority cells of KYSE-410 (50 µM). Wecells treated (50 µM). KYSE-410 We treated cells KYSE- with cisplatin410 cells forwith 4 hcisplatin at either for 24 4 ( PER2h at eitherhigh 24 expression) (PER2 high or expression) 36 h (PER2 lowor 36 expression) h (PER2 low after ex- dexamethasonepression) after dexamethasone synchronization synchronization (Figure4C). We validated(Figure 4C). by immunoblottingWe validated by thatimmunob- PER2 waslotting indeed that downregulatedPER2 was indeed at 36downregulated h comparedto at 2436 hh aftercompared synchronization to 24 h after (Figure synchroniza-4D). tion To(Figure determine 4D). whether a difference in the level of PER2 expression would have conse- quencesTo ondetermine the global whether response a difference to therapy, in we the then level investigated of PER2 expression whether cisplatin-induced would have con- DNAsequences damage on the would global be response modified to depending therapy, we on then the level investigated of PER2 whether expression. cisplatin-in- To this end,duced we DNA harvested damage KYSE-410 would be cells modified after a 4-hdepending cisplatin on treatment the level at of either PER2 24 expression. or 36 h post- To synchronizationthis end, we harvested and measured KYSE-410γH2AX cells after expression a 4-h cisplatin by Western treatment blot (Figureat either4E,F). 24 or This 36 h experimentpost-synchronization shows that and cisplatin measured induces γH2AX more DNAexpression damage by when Western PER2 blot expression (Figure is4E,F). low. WeThis then experiment measured shows apoptosis that cisplatin using annexin induces V more staining DNA by damage flow cytometry when PER2 48 hexpression after the endis low. of theWe treatmentthen measured (Figure apoptosis4G). Interestingly, using annexin the percentageV staining by of flow eSCC cytometry cells positive 48 h after for annexinthe end Vof appeared the treatment significantly (Figurehigher 4G). Interest when cellsingly, were the treatedpercentage 36 h afterof eSCC synchronization, cells positive atfor the annexin time when V appeared PER2 expression significantly is low higher (Figure when4H). cells Hence, were these treated results 36 h suggest after synchroni- that the circadianzation, at clockthe time influences when PER2 chemotherapy-induced expression is low (Figure cell death 4H). in Hence, esophageal these cancerresults cells.suggest that the circadian clock influences chemotherapy-induced cell death in esophageal cancer cells.
Biology 2021, 10, 266 11 of 18 Biology 2021, 10, x FOR PEER REVIEW 11 of 18
ABKYSE-410 (Hu) C DKYSE-410 100 Cisplatin 50uM 24h 36h Analyse Cisplatin Per2
50 Dexa 0 24 48 Time (hours) HSP90 02436
Survival, % of CTRL 0 110100Time (hours) Cisplatin (uM) EF Cisplatin 50uM 24h 36h NaCl Cisplatin NaCl Cisplatin
b-actin Dexa
36 +4 +48 Time 024+4 +48 (hours) gH2AX gH2AX Annexin V
GHNo Cisp (CT) +Cisp / 24h +Cisp / 36h 5 p=0,0073
4 p=0.8066
3 0.4% 6.8% 11.4% 2
Cell count 1
2 3 4 5 2 3 4 5 2 3 4 5 % of AnnexinV+0 cells 10 10 10 10 10 10 10 10 10 10 10 10 l t lat C la p is Annexin V (in Hoechst negative cells) +NaClC h+ 24h 4 36h +Na 2 36h +Cisp Figure 4. A low PER2 expression is associatedassociated toto higherhigher sensitivitysensitivity toto cisplatin-inducedcisplatin-induced apoptosis. (A) Experimental design. ( (B)) KYSE-410 survival measured using MTS assa assayy 48 h after cisplatin treatment. The red sigmoid curve is is a a guide guide for for the the eye. eye. Data are represented asas meanmean +/+/−− SEMSEM (n (n == 5). 5). (C (C) )Experimental Experimental design design to to treat treat KYSE KYSE-410-410 when when PER2 PER2 expression expression is high or low (24 or 36 h after synchronization respectively). ( (DD)) Western Western blot showing the expression of PER2 24 or 36 h after synchronization. (E) Schematic representation of the cisplatin treatment timeline and the time-points selected to collect synchronization. (E) Schematic representation of the cisplatin treatment timeline and the time-points selected to collect the samples for DNA damage (γH2AX) evaluation and apoptosis (annexin V) measurement. (F) Western blot showing the the samples for DNA damage (γH2AX) evaluation and apoptosis (annexin V) measurement. (F) Western blot showing the expression of γH2AX after a 4-h cisplatin treatment either 24 or 36 h post-synchronization. Cells were either treated with γ expressionvehicle alone of (NaCl)H2AX or after with a 50 4-h µM cisplatin cisplatin. treatment (G) Representative either 24 or 36 histogram h post-synchronization. showing the proportion Cells were of eitherHoechst-/Annexin treated with vehicleV KYSE-410 alone cells (NaCl) measured or with by 50 µflowM cisplatin. cytometry (G 48) Representative h after the end histogram of the cisplatin showing treatment. the proportion Cells were of Hoechst-/Annexin either not treated V(NaCl KYSE-410 = vehicle) cells or measured treated for by 4 flow h with cytometry 50 µM cisplatin 48 h after 24 the or 36 end h ofafter the synchronization. cisplatin treatment. (H) CellsHistogram were eithersummarizing not treated the (NaClproportion = vehicle) of Hoechst-/Annexin or treated for 4 V h KYSE-410 with 50 µM cells cisplatin measured 24 or 48h 36 after h after the synchronization. end of the cisplati (Hn )treatment. Histogram Data summarizing are repre- thesented proportion as mean of+/ Hoechst-/Annexin− SEM (n = 8). p-values V KYSE-410 were calculated cells measured using a 48htwo-way after theANOVA end of followed the cisplatin by Sidak’s treatment. post Datahoc test. are representedFixed effect: asTime/ meanp = +/ 0.0110;− SEM Cisplatin/ (n = 8). pp-values < 0.0001; were Time calculated x Cisplatin/ usingp = a 0.0527. two-way ANOVA followed by Sidak’s post hoc test. Fixed effect: Time/p = 0.0110; Cisplatin/p < 0.0001; Time x Cisplatin/p = 0.0527. 3.5. Characterization of the Pattern of Expression of Clock Related Genes in Mouse Esophageal SCC3.5. CharacterizationPrimary Culture of the Pattern of Expression of Clock Related Genes in Mouse Esophageal SCCIn Primary order Cultureto determine whether the pattern of clock related genes in biopsies of human eSCCIn would order be to affected determine by whethertheir contamination the pattern with of clock non-epithelial related genes cells, in we biopsies have inves- of hu- tigatedman eSCC clock would genes bein a affected mouse model by their of contaminationeSCC. To this end, with we non-epithelial used a model cells,of chemically- we have inducedinvestigated esophageal clock genes cancer in based a mouse on the model additi ofon eSCC. of 4-NQO To this (4-Nitroquinoline end, we used a1-oxide) model ofin thechemically-induced drinking water. We esophageal treated K5CreER:p53f cancer basedl/fl:RosaYFP on the addition (K5:p53:YFP) of 4-NQO (4-Nitroquinolinemice with tamox- ifen1-oxide) to remove in the drinkingp53 and turn water. on We the treated YFP fluo K5CreER:p53fl/fl:RosaYFPrescent reporter in esophageal (K5:p53:YFP) cells (Figure mice 5A).with We tamoxifen treated to8 weeks-old remove p53 K5:p53:YFP and turn onmice the with YFP 4-NQO fluorescent over reporter8 weeks inand esophageal then fol- lowedcells (Figure them over5A). time Wetreated until they 8 weeks-old developed K5:p53:YFP tumors (Figure mice 5B). with We 4-NQO could dissect over 8 individ- weeks ualand tumors then followed from K5:p53:YFP them over esophagus time until and they dige developedsted these tumorsto isolate (Figure tumor5 epithelialB). We could cells bydissect flow individualcytometry. tumorsWe verified from the K5:p53:YFP histology esophagusof the primary and tumors digested prior these to tosequencing. isolate tu- Thesemor epithelial tumors were cells bycharacterized flow cytometry. by classical We verified eSCC themarkers histology p63 and of the Krt14 primary (Figure tumors 5C). FACSprior toanalysis sequencing. confirmed These the tumors expression were characterized of YFP by the by tumor classical cells, eSCC allowing markers us p63 to andsort pureKrt14 YFP+ (Figure tumor5C). epithelial FACS analysis cells and confirmed get rid of the the expression contamination of YFP with by thelineage tumor negative cells, cellsallowing (CD45+ us to leukocytes, sort pure YFP+ CD31+ tumor endothelial epithelial cells cells and and CD140a+ get rid of fibroblasts) the contamination (Figure with5D). Welineage profiled negative YFP+ cells eSCC (CD45+ epithelial leukocytes, cells by CD31+RNAseq endothelial and compared cells them and CD140a+ to FACS fibrob-sorted
Biology 2021, 10, 266 12 of 18
Biology 2021, 10, x FOR PEER REVIEW 12 of 18
lasts) (Figure5D) . We profiled YFP+ eSCC epithelial cells by RNAseq and compared them EpCam+to FACS control sorted esophageal EpCam+ epithelial control esophageal cells (Figure epithelial 5D). We cells found (Figure 3,4925D). transcripts We found sig- 3,492 nificantlytranscripts deregulated significantly in cancer deregulated cells (Log2 in Fold cancer of change cells (Log2 >1 and Fold FDR of < change0.01). We >1 focused and FDR < our analysis0.01). We on focused clock related our analysis genes onand clock found related no difference genes and except found for no Per3 difference, which except was for upregulatedPER3, which (LFC was = 1.21) upregulated in cancer (LFCcells =compared 1.21) in cancer to normal cells comparedcells and Rora to normal, which cells ap- and pearedRora downregulated, which appeared (LFC= downregulated −2.07) in cancer (LFC= cells− compared2.07) in cancer to control cells comparedesophageal to cells control (Figureesophageal 5E). cells (Figure5E).
FigureFigure 5. Per2 5. PER2expression expression oscillates oscillates in primary in primary culture culture of mou ofse mouse eSCC eSCCcells and cells influences and influences sensitivity sensitivity to chemotherapy. to chemotherapy. (A) Genetic(A) Genetic strat- egy andstrategy experimental and experimental design. (B design.) Representative (B) Representative picture of picture 4-NQO-induce of 4-NQO-inducedd eSCC in eSCCK5:p53:YFP in K5:p53:YFP mouse esophagus. mouse esophagus. (C) Co- immunostaining of p63 and Krt14 in normal esophagus and 4-NQO-induced eSCC. (D) FACS plot showing the expression of (C) Co-immunostaining of p63 and Krt14 in normal esophagus and 4-NQO-induced eSCC. (D) FACS plot showing the EpCam and lineage negative markers (CD45, CD31, and CD140a) in control esophageal cells, and YFP and lineage negative expression of EpCam and lineage negative markers (CD45, CD31, and CD140a) in control esophageal cells, and YFP and markers in 4-NQO-induced eSCC. (E) Expression of clock-related genes measured by RNA sequencing in FACS sorted Epcam+ normallineage esophagus negative epithelial markers cells in 4-NQO-induced (n = 2) and YFP+ eSCC. 4-NQO-induced (E) Expression eSCC of clock-relatedepithelial cells genes (n = 3). measured FDR is indicated by RNA sequencing for each gene. in Benjamini–HochbergFACS sorted Epcam+ method normal on the esophagus p-values epithelial was used cells to control (n = 2) the and false YFP+ discov 4-NQO-inducedery rate (FDR). eSCC FDR>5% epithelial are cells considered (n = 3). FDRnon- significantis indicated (N.S). for“Epith” each gene.means Benjamini–Hochberg epithelial cells. method on the p-values was used to control the false discovery rate (FDR). FDR>5% are considered non-significant (N.S). “Epith” means epithelial cells. 3.6. A Low Per2 Expression is Associated to a Higher Sensitivity to Cisplatin-Mediated Apoptosis3.6. A in Low Mouse PER2 Esophageal Expression SCC Is Associated Cells to a Higher Sensitivity to Cisplatin-Mediated Apoptosis in Mouse Esophageal SCC Cells We grew FACS sorted YFP+ tumor epithelial cells in vitro (Figure 6A) and transduced the PER2:LucWe grew construct FACS sortedin these YFP+ cells tumor using epithelial lentiviral cells particles.in vitro Similar(Figure to6A) human and transduced eSCC cells,the a 1-h PER2:Luc dexamethasone construct intreatment these cells was using suff lentiviralicient to particles.synchronize Similar the cells to human and trigger eSCC cells, a 1-h dexamethasone treatment was sufficient to synchronize the cells and trigger luciferase luciferase daily oscillations that can be monitored on real time for several days (Figure daily oscillations that can be monitored on real time for several days (Figure6B). These 6B). These results therefore demonstrate that Per2 expression oscillates in primary cul- results therefore demonstrate that PER2 expression oscillates in primary cultures of mouse tures of mouse esophageal cancer epithelial cells. We then measured Per2 and Arntl ex- esophageal cancer epithelial cells. We then measured PER2 and Arntl expression by qPCR pression by qPCR in mouse eSCC epithelial cells after synchronization (Figure 6C). This experiment confirmed the variations of Per2 expression measured using bioluminescence
Biology 2021, 10, 266 13 of 18
Biology 2021, 10, x FOR PEER REVIEW 13 of 18
in mouse eSCC epithelial cells after synchronization (Figure6C). This experiment confirmed the variations of PER2 expression measured using bioluminescence assay, although there is assay,a although 4-h delay therebetween is a bioluminescence 4-h delay between and mRNAbioluminescence expression. and Nonetheless, mRNA expressio we didn. not see Nonetheless,a typical we antiphase did not expressionsee a typical of antiphaseArntl in mouse expression eSCC of cells, Arntl suggesting in mouse that eSCC the cells, biological suggestingcircadian that clockthe biological would be circadian altered clock in these would cells be and altered that onlyin these PER2 cells expression and that only may still Per2 expressionoscillate in may this model.still oscillate in this model. To determineTo determine whether whether Per2 oscillations PER2 oscillations had an had impact an impact on response on response to therapy to therapy as we as we foundfound out in out human in human eSCC eSCCcells, we cells, treated we treated mouse mouse tumor tumor epithelial epithelial cells with cells cisplatin with cisplatin for for 4 h when4 h whenPer2 expression PER2 expression is high is (24 high and (24 48 and h) or 48 when h) or whenit is low it is(36 low and (36 60 and h) (Figure 60 h) (Figure 6D). 6D). We thenWe meas thenured measured cell death cell death by counting by counting the proportion the proportion of annexin of annexin V positive V positive cells 48 cells h 48 h after theafter end the of end the oftreatment the treatment (Figure (Figure 6E). As6E). we As observed we observed in human in human eSCC eSCC cells, cells,the level the level of cisplatinof cisplatin-induced-induced apoptosis apoptosis was significantly was significantly higher higher when when Per2 PER2expression expression is low is (at low 36 (at 36 and 60and h after 60 h synchronization) after synchronization) (Figure (Figure 6F). Hence,6F). Hence, it seems it seems that in that both in bothmouse mouse and human and human eSCC eSCCcells, Per2 cells, oscillations PER2 oscillations lead to lead variable to variable sensitivity sensitivity to chemotherapy to chemotherapy (Figure (Figure 6G). 6G).
A eSCC primary culture B C 4NQO (eSCC) Per2 10 0.08 Arntl
e
)
c
0
n
)
t
0
e
0
c
C
1
s d 0.07
(
x
e
P
5 /
(
n
F
1
i
U
Y
.
m
u A 0.06
L 0 0 24 48 72 96 120 12 24 36 48 Time (hours) Time (hours) D E +Cisplatin 50uM Cisplatin 50uM 24h 36h 17.4% 29.3%
s
i
t
s
a
n
y
x
l
u
e
a
o
D
n
c
A
l
l 48h 60h
e 15.2% 22.7%
C 0 24 36 48 60 +48
Time (hours) 2 3 4 5 2 3 4 5 10 10 10 10 10 10 10 10 Annexin V (in Hoechst neg cells) F G p=0.0266 p=0.0041 s 30 s 25 High resistance
l l
l l
e p=0.2252 e p=0.855 PER2 to chemo c c 20
+ +
n
V 20 V
o
i
n n 15
n n
i i
s
i i
t t
x x
s
a a
l l
e e
e
p p
r
n n 10
s s
i i
p
n 10 n
x
C C
A A
e
f f 5
o o
2
R
% 0 % 0 l p l l p l E s sp s sp Low sensitivity C i C i C i C i P a C a C a C a C N + N + N + N + Time PER2 to chemo + h + h + h + h h 4 h 6 h 8 h 0 4 2 6 3 8 4 0 6 2 3 4 6
FigureFigure 6. A low 6. APer2 low expression PER2 expression is associated is associated to a higher to a highersensitivity sensitivity to cisplatin to cisplatin in mouse inmouse eSCC eSCCepithelial epithelial cells. ( cells.A) Endogenous(A) Endogenous YFP fluorescenceYFP fluorescence in primary in culture primary of culture4-NQO of-induced 4-NQO-induced eSCC epithelial eSCC epithelial cells. (B) cells.Real time (B) Real monitoring time monitoring of luciferase of luciferase activity activityin pri- mary inculture primary of 4 culture-NQO-induced of 4-NQO-induced eSCC epithelial eSCC cells. epithelial (C) Per2 cells. and (C )ArntlPER2 expressionand Arntl expressionon the 4-NQO on theinduced 4-NQO eSCC induced epithelial eSCC cells measuredepithelial by cells qPCR measured for 48 byh after qPCR dexamethasone for 48 h after- dexamethasone-inducedinduced synchronization. synchronization.(D) Experimental ( Ddesign) Experimental to treat 4-NQO design-in- to ducedtreat eSCC 4-NQO-induced epithelial cells when eSCC Per2 epithelial expression cells when is high PER2 or low expression (24 or 36 is h high after or dexamet low (24hasone or 36 h-induced after dexamethasone-induced synchronization). (E) Representativesynchronization). histogram (E showing) Representative the proportion histogram of Hoechst showing-/Annexin the proportion V 4-NQO of Hoechst-/Annexin-induced eSCC epithelial V 4-NQO-induced cells measured eSCC by flow cytometryepithelial cells48 h after measured the end by of flow the cytometry cisplatin treatment. 48 h after theCells end were of the either cisplatin not treated treatment. (NaCl Cells = vehicle) were either or treated not treated for 4 h (NaClwith 50 μM cisplatin 24, 36, 48 or 60 h after synchronization. (F) Histogram summarizing the proportion of Hoechst-/Annexin V+ 4- = vehicle) or treated for 4 h with 50 µM cisplatin 24, 36, 48 or 60 h after synchronization. (F) Histogram summarizing the NQO-induced eSCC epithelial cells measured 48 h after the end of the cisplatin treatment. Data are represented as mean +/− proportion of Hoechst-/Annexin V+ 4-NQO-induced eSCC epithelial cells measured 48 h after the end of the cisplatin SEM. p-values were calculated using a two-way ANOVA followed by Sidak’s post hoc test. First cycle (n = 8). Fixed effect: Time/ptreatment. = 0.3788; Cisplatin Data are represented/p < 0.0001; Time as mean x Cisplatin +/− SEM./p =p 0.0067-values. Second were calculated cycle (n = using 5). Fixed a two-way effect: Time ANOVA/p = followed0.0231; Cisplatin by Sidak’s/p < 0.0001;post Time hoc x test. Cisplatin First cycle/p = 0.0 (n231= 8).. (G Fixed) Scheme effect: depicting Time/p our= 0.3788; main Cisplatin/observationsp < in 0.0001; esophageal Time xcancer Cisplatin/ cells.p = 0.0067. Second cycle (n = 5). Fixed effect: Time/p = 0.0231; Cisplatin/p < 0.0001; Time x Cisplatin/p = 0.0231. (G) Scheme depicting our main observations in esophageal4. Discussion cancer cells. Many studies have investigated the role of the circadian clock in cancer cells over the last 2 decades [16]. It is now clear that the disruption of the circadian rhythm increases the
Biology 2021, 10, 266 14 of 18
4. Discussion Many studies have investigated the role of the circadian clock in cancer cells over the last 2 decades [16]. It is now clear that the disruption of the circadian rhythm increases the risk of developing cancer [8,9]. CLOCK and BMAL1 clock master genes may harbor tumor-suppressive functions both in humans and rodents [27]. However, depending on the model and the tissue, CLOCK and BMAL1 may be oncogenic or tumor suppressive [28–31]. Mouse models with an invalidation of the clock master genes PER1 and/or PER2 display an increased risk of spontaneous and radiation-induced cancer development [32–34]. In spite of the downregulation of clock related genes in cancer, the persistence of a circadian clock or at least of PER2 oscillations have been reported in several cancer models. Indeed, the use of PER2:Luc had been previously used to demonstrate this rhythmic expression of PER2 in low malignancy breast cancer cells [35], and more recently in cervical and esophageal cancer cell lines [19]. Our results unveil that human eSCC cell lines display various patterns of expression of clock related genes. This deregulation is not associated to a downregulation of the PER family members as observed in biopsies from human eSCC. In addition, our results also clearly evidence an oscillation of PER2 expression in esophageal squamous cell carcinoma (eSCC) cells from two different species. The difference between the pattern found in biopsies and tumor epithelial cells could be the consequence of the presence of non-tumoral cells in biopsies such as immune cells or cancer associated fibroblasts, but the status of clock-related genes in stromal cells should be thoroughly investigated to test this hypothesis. This difference might also reflect various stages of cancer progression. However, our data show that the tumor pathological stage is not associated to a significant modification of the clock-related genes expression. Since the number of eSCC samples is quite limited in the TCGA database, especially for metastatic cancers, further experiments will be required to determine how cancer progression impacts the expression of clock related genes. Clock related genes are associated to multiple functions in cancer cells, such as cell- cycle, cell signaling, cell death, DNA damage response, or metabolism [16]. It has already been reported that clock disruption results in the accumulation of DNA damage. CRY2 expression is stimulated by DNA-damage and can regulate an intra-S checkpoint function in a ATR/CHK1 dependent manner [36]. PER1 interacts with ATM/CHK2 following radiation-induced DNA damage [37]. In the same line, PER2 downregulation would promote the resistance to radiation-induced apoptosis by delaying CHK2 expression [38]. In our model, we could show by RNA sequencing analysis that a low PER2 expression is associated to a downregulation of transcripts involved in “metabolism of nucleic acids”, and to an increase in “apoptotic signaling pathway” related genes. This pattern suggests that response to therapy may be different depending on the circadian rhythm, mostly because cells might accumulate more DNA damage and undergo apoptosis more frequently when PER2 expression is low. Consistently, mice homozygous for the mPer2m/m mutation show an increased sensitivity to radiation and a higher risk to develop lymphoma [32], while PER1 is involved in the regulation of proapoptotic signals [15]. Cisplatin is one of the most common chemotherapeutic drugs. It is widely used to treat many solid tumors, including lung cancer, head and neck cancer and esophageal cancer [39]. Despite its effectiveness, many patients have intrinsic resistance to cisplatin or develop resistance to cisplatin over time. Platinum resistance thus remains a major issue in the therapy for many types of cancer. The mechanisms of resistance are complex and have not been fully elucidated. Among these mechanisms, enhanced DNA repair is considered as the most important cause of cisplatin resistance [39]. Indeed, cisplatin causes DNA intra- and interstrand cross-links by forming cisplatin-DNA adducts. These cisplatin-DNA adducts lead to damages that are repaired by nucleotide excision repair [40]. Interestingly, it has been reported that cisplatin-induced DNA damage is repaired by two circadian programs [41]. Our results show that cisplatin-induced DNA damage is enhanced when PER2 expression is low in human eSCC cells. It will be important to investigate the molecular mechanisms that link PER2 expression, DNA damage and apoptosis in Biology 2021, 10, 266 15 of 18
esophageal cancer cells. Since we found that PER1 was the most significant downregulated genes when PER2 expression is low, the process may be related ATM/CHK2 as reported in a human colon cancer cell line [37]. Since the level of PER2 expression correlates with apoptosis related genes, it will be important to determine whether PER2 oscillatory expression sensitizes esophageal cancer cells to other anti-cancer therapies and whether PER1 is involved in this process. The goal of chronotherapy is to deliver the drug at a time of the day that is optimal for killing cancer cells while avoiding resistance and/or toxicity [42,43]. Since cisplatin half-life in the blood stream is short (4h), administering this chemotherapy when the cells are the most sensitive to it can be relevant for its efficiency. In melanoma, the efficiency of cisplatin in PER1/2−/− animals is higher than in control specimens [44]. In the same line, our results clearly indicate that cisplatin treatment of esophageal cancer cells is more efficient when PER2 expression is low. Previous studies already suggested that circadian rhythm may affect the response to therapy in esophageal cancer cells. Indeed, apoptotic regulators DEC1 and DEC2 repress CLOCK/BMAL-induced transactivation of the PER1 promoter and modulate the response to cisplatin in several types of cancers [45–47], including human eSCC cell lines [48,49]. Nonetheless, a treatment based on physiological oscillations of PER2 has never been tested before in eSCC. Our results therefore suggest that chronotherapy could be used to improve the efficiency of current chemotherapy in esophageal cancers. Several attempts have been made to develop chronotherapy regimens for combinations of drugs that included platinum salts [50,51]. Nevertheless, the gains have been either minor or have not been replicated in further studies [52]. Hence, cisplatin chronotherapy is not practiced in clinical oncology. Further characterization of the process by which circadian clock modulates chromatin stability and genome maintenance will be required. Indeed, although there are growing evidences of connections between circadian rhythm and genome maintenance [53], the underlying mechanisms are still largely unknown. To determine whether chronotherapy might be used to improve treatment of esophageal cancers, the characterization of PER2 oscillations in eSCC cells in vivo will be required. Then the efficiency of cisplatin treatment based on the time of the day and/or PER2 expression should be tested in an in vivo setting such as xenografts of multiple eSCC lines and/or patient derived xenograft (PDX). Since cancer progression is associated to reduced BMAL1 and PER2 oscillations, it will be important to determine whether more malignant esophageal cancer epithelial cells are characterized by additional modifications of clock related genes and how the circadian clock is impacted. In conclusion, our results reveal that in spite of alterations of the expression of clock related genes, PER2 oscillations could be used to improve response to chemotherapy in esophageal cancer.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/biology10040266/s1, Figure S1: Expression of clock-related genes in human eSCC samples depending on tumor stages. Figure S2: Expression of clock-related genes in human eSCC samples regarding N tumor pathological stage. Author Contributions: J.A.R., R.B. and B.B. designed the experiments and performed data analysis. J.A.R. and R.B. performed all the experiments. B.D. and A.V.D. performed analysis of the RNAseq data, X.B. provided technical help and E.M. performed the bioluminescence measurement to monitor PER2 oscillations. B.B. wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: B.B. is an investigator of Walloon Excellence in Lifesciences & BIOtechnology (WELBIO) and FNRS at the ULB (https://www.benjaminbecklab.com accessed date: 15 December 2020). J.A.R. is supported by a fellowship from the WELBIO. B.D. is supported by the FRIA (Fonds pour la formation à la recherche dans l’industrie et dans l’agriculture). A.V.D. is supported by the Fondation contre le cancer (FCC/ULB 2018-067). X.B. is supported by a MSCA-IF-2018 (843107) and by a Innoviris BB2B-Attract (RBC/BFB1) and E.M. is supported by INNOVIRIS ATTRACT (reference 2016-BB2B-4, to E.M.). This work has been supported by the WELBIO (FRFS-WELBIO-CR-2017S-01), the FNRS (F4538.16 MIS). Biology 2021, 10, 266 16 of 18
Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of the Faculty of Medicine from the ULB (protocol code 668N (2018-2023)). Data Availability Statement: The results shown here are in whole or part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga (accessed date: 15 December 2020). Acknowledgments: We thank the genomic facility and the FACS core facility from the Université libre de Bruxelles (ULB) for their precious help. Sequencing was performed at the Brussels Interuniversity Genomics High Throughput core (www.brightcore.be accessed date: 15 December 2020). We also thank the animal house facility of the ULB. Conflicts of Interest: The authors declare no conflict of interest.
References 1. Delaunay, F.; Laudet, V. Circadian Clock and Microarrays: Mammalian Genome Gets Rhythm. Trends Genet. 2002, 18, 595–597. [CrossRef] 2. Panda, S.; Antoch, M.P.; Miller, B.H.; Su, A.I.; Schook, A.B.; Straume, M.; Schultz, P.G.; Kay, S.A.; Takahashi, J.S.; Hogenesch, J.B. Coordinated Transcription of Key Pathways in the Mouse by the Circadian Clock. Cell 2002, 109, 307–320. [CrossRef] 3. Takahashi, J.S. Transcriptional Architecture of the Mammalian Circadian Clock. Nat. Rev. Genet. 2017, 18, 164–179. [CrossRef] [PubMed] 4. Kim, P.; Oster, H.; Lehnert, H.; Schmid, S.M.; Salamat, N.; Barclay, J.L.; Maronde, E.; Inder, W.; Rawashdeh, O. Coupling the Circadian Clock to Homeostasis: The Role of Period in Timing Physiology. Endocr. Rev. 2019, 40, 66–95. [CrossRef] 5. Zhang, R.; Lahens, N.F.; Ballance, H.I.; Hughes, M.E.; Hogenesch, J.B. A Circadian Gene Expression Atlas in Mammals: Implications for Biology and Medicine. Proc. Natl. Acad. Sci. USA 2014, 111, 16219–16224. [CrossRef][PubMed] 6. Hastings, M.H.; Goedert, M. Circadian Clocks and Neurodegenerative Diseases: Time to Aggregate? Curr. Opin. Neurobiol. 2013, 23, 880–887. [CrossRef] 7. Marcheva, B.; Ramsey, K.M.; Peek, C.B.; Affinati, A.; Maury, E.; Bass, J. Circadian Clocks and Metabolism. In Circadian Clocks. Handbook of Experimental Pharmacology; Kramer, A., Merrow, M., Eds.; Springer: Berlin/Heidelberg, Germany, 2013; Volume 217, pp. 127–155. ISBN 978-3-642-25949-4. 8. Stevens, R.G.; Blask, D.E.; Brainard, G.C.; Hansen, J.; Lockley, S.W.; Provencio, I.; Rea, M.S.; Reinlib, L. Meeting Report: The Role of Environmental Lighting and Circadian Disruption in Cancer and Other Diseases. Environ. Health Perspect. 2007, 115, 1357–1362. [CrossRef][PubMed] 9. Straif, K.; Baan, R.; Grosse, Y.; Secretan, B.; Ghissassi, F.E.; Bouvard, V.; Altieri, A.; Benbrahim-Tallaa, L.; Cogliano, V. Carcino- genicity of Shift-Work, Painting, and Fire-Fighting. Lancet Oncol. 2007, 8, 1065–1066. [CrossRef] 10. Davis, K.; Roden, L.C.; Leaner, V.D.; van der Watt, P.J. The Tumour Suppressing Role of the Circadian Clock: TUMOUR SUPPRESSING ROLE OF THE CIRCADIAN CLOCK. IUBMB Life 2019, 71, 771–780. [CrossRef] 11. Cadenas, C.; van de Sandt, L.; Edlund, K.; Lohr, M.; Hellwig, B.; Marchan, R.; Schmidt, M.; Rahnenführer, J.; Oster, H.; Hengstler, J.G. Loss of Circadian Clock Gene Expression Is Associated with Tumor Progression in Breast Cancer. Cell Cycle 2014, 13, 3282–3291. [CrossRef] 12. Hsu, C.-M.; Lin, S.-F.; Lu, C.-T.; Lin, P.-M.; Yang, M.-Y. Altered Expression of Circadian Clock Genes in Head and Neck Squamous Cell Carcinoma. Tumor Biol. 2012, 33, 149–155. [CrossRef] 13. Zhao, H.; Zeng, Z.-L.; Yang, J.; Jin, Y.; Qiu, M.-Z.; Hu, X.-Y.; Han, J.; Liu, K.-Y.; Liao, J.-W.; Xu, R.-H.; et al. Prognostic Relevance of Period1 (PER1) and Period2 (PER2) Expression in Human Gastric Cancer. Int. J. Clin. Exp. Pathol. 2014, 7, 619–630. 14. Matsunaga, N.; Kohno, Y.; Kakimoto, K.; Hayashi, A.; Koyanagi, S.; Ohdo, S. Influence of CLOCK on Cytotoxicity Induced by Diethylnitrosamine in Mouse Primary Hepatocytes. Toxicology 2011, 280, 144–151. [CrossRef] 15. Ozturk, N.; Lee, J.H.; Gaddameedhi, S.; Sancar, A. Loss of Cryptochrome Reduces Cancer Risk in P53 Mutant Mice. Proc. Natl. Acad. Sci. USA 2009, 106, 2841–2846. [CrossRef][PubMed] 16. Shafi, A.A.; Knudsen, K.E. Cancer and the Circadian Clock. Cancer Res. 2019, 79, 3806–3814. [CrossRef][PubMed] 17. Arnold, M.; Laversanne, M.; Brown, L.M.; Devesa, S.S.; Bray, F. Predicting the Future Burden of Esophageal Cancer by Histological Subtype: International Trends in Incidence up to 2030. Am. J. Gastroenterol. 2017, 112, 1247–1255. [CrossRef] 18. Siegel, R.L.; Miller, K.D.; Fuchs, H.E.; Jemal, A. Cancer Statistics, 2021. CA Cancer J. Clin. 2021, 71, 7–33. [CrossRef][PubMed] 19. Van der Watt, P.J.; Roden, L.C.; Davis, K.T.; Parker, M.I.; Leaner, V.D. Circadian Oscillations Persist in Cervical and Esophageal Cancer Cells Displaying Decreased Expression of Tumor-Suppressing Circadian Clock Genes. Mol. Cancer Res. 2020, 18, 1340–1353. [CrossRef][PubMed] 20. Powell, D. Drpowell/Degust 4.1.1; Zenodo: Geneva, Switzerland, 2019. [CrossRef] 21. Liu, A.C.; Tran, H.G.; Zhang, E.E.; Priest, A.A.; Welsh, D.K.; Kay, S.A. Redundant Function of REV-ERBα and β and Non-Essential Role for Bmal1 Cycling in Transcriptional Regulation of Intracellular Circadian Rhythms. PLoS Genet. 2008, 4, e1000023. [CrossRef] 22. Wu, G.; Anafi, R.C.; Hughes, M.E.; Kornacker, K.; Hogenesch, J.B. MetaCycle: An Integrated R Package to Evaluate Periodicity in Large Scale Data. Bioinformatics 2016, 32, 3351–3353. [CrossRef][PubMed] Biology 2021, 10, 266 17 of 18
23. Anders, S.; Pyl, P.T.; Huber, W. HTSeq—A Python Framework to Work with High-Throughput Sequencing Data. Bioinformatics 2015, 31, 166–169. [CrossRef][PubMed] 24. Blighe, K. EnhancedVolcano. Bioconductor. 2018. Available online: https://bioconductor.org/packages/release/bioc/html/ EnhancedVolcano.html (accessed on 15 December 2020). 25. Korotkevich, G.; Sukhov, V.; Budin, N.; Shpak, B.; Artyomov, M.N.; Sergushichev, A. Fast Gene Set Enrichment Analysis. BioRxiv 2016, e060012. [CrossRef] 26. Shimada, Y.; Imamura, M.; Wagata, T.; Yamaguchi, N.; Tobe, T. Characterization of 21 Newly Established Esophageal Cancer Cell Lines. Cancer 1992, 69, 277–284. [CrossRef] 27. Fu, L.; Kettner, N.M. The circadian clock in cancer development and therapy. In Progress in Molecular Biology and Translational Science; Elsevier: Amsterdam, The Netherlands, 2013; Volume 119, pp. 221–282. ISBN 978-0-12-396971-2. 28. Kettner, N.M.; Voicu, H.; Finegold, M.J.; Coarfa, C.; Sreekumar, A.; Putluri, N.; Katchy, C.A.; Lee, C.; Moore, D.D.; Fu, L. Circadian Homeostasis of Liver Metabolism Suppresses Hepatocarcinogenesis. Cancer Cell 2016, 30, 909–924. [CrossRef] 29. Papagiannakopoulos, T.; Bauer, M.R.; Davidson, S.M.; Heimann, M.; Subbaraj, L.; Bhutkar, A.; Bartlebaugh, J.; Vander-Heiden, M.G.; Jacks, T. Circadian Rhythm Disruption Promotes Lung Tumorigenesis. Cell Metab. 2016, 24, 324–331. [CrossRef] 30. Puram, S.V.; Tirosh, I.; Parikh, A.S.; Patel, A.P.; Yizhak, K.; Gillespie, S.; Rodman, C.; Luo, C.L.; Mroz, E.A.; Emerick, K.S.; et al. Single-Cell Transcriptomic Analysis of Primary and Metastatic Tumor Ecosystems in Head and Neck Cancer. Cell 2017, 171, 1611–1624.e24. [CrossRef] 31. Tang, Q.; Cheng, B.; Xie, M.; Chen, Y.; Zhao, J.; Zhou, X.; Chen, L. Circadian Clock Gene Bmal1 Inhibits Tumorigenesis and Increases Paclitaxel Sensitivity in Tongue Squamous Cell Carcinoma. Cancer Res. 2017, 77, 532–544. [CrossRef] 32. Fu, L.; Pelicano, H.; Liu, J.; Huang, P.; Lee, C.C. The Circadian Gene Period2 Plays an Important Role in Tumor Suppression and DNA Damage Response In Vivo. Cell 2002, 111, 41–50. [CrossRef] 33. Lee, S.; Donehower, L.A.; Herron, A.J.; Moore, D.D.; Fu, L. Disrupting Circadian Homeostasis of Sympathetic Signaling Promotes Tumor Development in Mice. PLoS ONE 2010, 5, e10995. [CrossRef] 34. Matsuo, T. Control Mechanism of the Circadian Clock for Timing of Cell Division In Vivo. Science 2003, 302, 255–259. [CrossRef] 35. Lellupitiyage Don, S.S.; Lin, H.-H.; Furtado, J.J.; Qraitem, M.; Taylor, S.R.; Farkas, M.E. Circadian Oscillations Persist in Low Malignancy Breast Cancer Cells. Cell Cycle 2019, 18, 2447–2453. [CrossRef] 36. Barnes, J.W. Requirement of Mammalian Timeless for Circadian Rhythmicity. Science 2003, 302, 439–442. [CrossRef] 37. Gery, S.; Komatsu, N.; Baldjyan, L.; Yu, A.; Koo, D.; Koeffler, H.P. The Circadian Gene PER1 Plays an Important Role in Cell Growth and DNA Damage Control in Human Cancer Cells. Mol. Cell 2006, 22, 375–382. [CrossRef] 38. Yang, X.; Wood, P.A.; Hrushesky, W.J.M. Mammalian TIMELESS Is Required for ATM-Dependent CHK2 Activation and G2/M Checkpoint Control. J. Biol. Chem. 2010, 285, 3030–3034. [CrossRef] 39. Kelland, L. The Resurgence of Platinum-Based Cancer Chemotherapy. Nat. Rev. Cancer 2007, 7, 573–584. [CrossRef] 40. Reardon, J.T.; Vaisman, A.; Chaney, S.G.; Sancar, A. Efficient Nucleotide Excision Repair of Cisplatin, Oxaliplatin, and Bis-Aceto- Ammine-Dichloro-Cyclohexylamine-Platinum(IV) (JM216) Platinum Intrastrand DNA Diadducts. Cancer Res. 1999, 59, 3968–3971. [PubMed] 41. Yang, Y.; Adebali, O.; Wu, G.; Selby, C.P.; Chiou, Y.-Y.; Rashid, N.; Hu, J.; Hogenesch, J.B.; Sancar, A. Cisplatin-DNA Adduct Repair of Transcribed Genes Is Controlled by Two Circadian Programs in Mouse Tissues. Proc. Natl. Acad. Sci. USA 2018, 115, E4777–E4785. [CrossRef] 42. Smolensky, M.H.; Peppas, N.A. Chronobiology, Drug Delivery, and Chronotherapeutics. Adv. Drug Deliv. Rev. 2007, 59, 828–851. [CrossRef] 43. Cederroth, C.R.; Albrecht, U.; Bass, J.; Brown, S.A.; Dyhrfjeld-Johnsen, J.; Gachon, F.; Green, C.B.; Hastings, M.H.; Helfrich-Förster, C.; Hogenesch, J.B.; et al. Medicine in the Fourth Dimension. Cell Metab. 2019, 30, 238–250. [CrossRef] 44. Dakup, P.P.; Porter, K.I.; Little, A.A.; Gajula, R.P.; Zhang, H.; Skornyakov, E.; Kemp, M.G.; Van Dongen, H.P.A.; Gaddameedhi, S. The Circadian Clock Regulates Cisplatin-Induced Toxicity and Tumor Regression in Melanoma Mouse and Human Models. Oncotarget 2018, 9, 14524–14538. [CrossRef] 45. Liu, Y.; Sato, F.; Kawamoto, T.; Fujimoto, K.; Morohashi, S.; Akasaka, H.; Kondo, J.; Wu, Y.; Noshiro, M.; Kato, Y.; et al. Anti- Apoptotic Effect of the Basic Helix-Loop-Helix (BHLH) Transcription Factor DEC2 in Human Breast Cancer Cells. Genes Cells 2010, 15, 315–325. [CrossRef] 46. Wu, Y.; Sato, F.; Bhawal, U.K.; Kawamoto, T.; Fujimoto, K.; Noshiro, M.; Morohashi, S.; Kato, Y.; Kijima, H. Basic Helix-Loop-Helix Transcription Factors DEC1 and DEC2 Regulate the Paclitaxel-Induced Apoptotic Pathway of MCF-7 Human Breast Cancer Cells. Int. J. Mol. Med. 2011, 27, 491–495. [CrossRef][PubMed] 47. Wu, Y.; Sato, F.; Bhawal, U.K.; Kawamoto, T.; Fujimoto, K.; Noshiro, M.; Seino, H.; Morohashi, S.; Kato, Y.; Kijima, H. BHLH Transcription Factor DEC2 Regulates Pro-Apoptotic Factor Bim in Human Oral Cancer HSC-3 Cells. Biomed. Res. 2012, 33, 75–82. [CrossRef] 48. Sato, H.; Wu, Y.; Kato, Y.; Liu, Q.; Hirai, H.; Yoshizawa, T.; Morohashi, S.; Watanabe, J.; Kijima, H. DEC2 Expression Antagonizes Cisplatin-induced Apoptosis in Human Esophageal Squamous Cell Carcinoma. Mol. Med. Rep. 2017, 16, 43–48. [CrossRef] 49. Seino, H.; Wu, Y.; Morohashi, S.; Kawamoto, T.; Fujimoto, K.; Kato, Y.; Takai, Y.; Kijima, H. Basic Helix-Loop-Helix Transcription Factor DEC1 Regulates the Cisplatin-Induced Apoptotic Pathway of Human Esophageal Cancer Cells. Biomed. Res. 2015, 36, 89–96. [CrossRef] [PubMed] Biology 2021, 10, 266 18 of 18
50. Levi, F.; Schibler, U. Circadian Rhythms: Mechanisms and Therapeutic Implications. Annu. Rev. Pharmacol. Toxicol. 2007, 47, 593–628. [CrossRef] 51. Lévi, F.; Okyar, A.; Dulong, S.; Innominato, P.F.; Clairambault, J. Circadian Timing in Cancer Treatments. Annu. Rev. Pharmacol. Toxicol. 2010, 50, 377–421. [CrossRef][PubMed] 52. Sancar, A.; Lindsey-Boltz, L.A.; Gaddameedhi, S.; Selby, C.P.; Ye, R.; Chiou, Y.-Y.; Kemp, M.G.; Hu, J.; Lee, J.H.; Ozturk, N. Circadian Clock, Cancer, and Chemotherapy. Biochemistry 2015, 54, 110–123. [CrossRef] 53. Kolinjivadi, A.M.; Chong, S.T.; Ngeow, J. Molecular Connections between Circadian Rhythm and Genome Maintenance Pathways. Endocr. Relat. Cancer 2021, 28, 55–66. [CrossRef]