Chronobiology International The Journal of Biological and Medical Rhythm Research ISSN: 0742-0528 (Print) 1525-6073 (Online) Journal homepage: http://www.tandfonline.com/loi/icbi20 Sex-dependent correlation between survival and expression of genes related to the circadian oscillator in patients with colorectal cancer Kristina Hasakova, Marian Vician, Richard Reis, Michal Zeman & Iveta Herichova To cite this article: Kristina Hasakova, Marian Vician, Richard Reis, Michal Zeman & Iveta Herichova (2018): Sex-dependent correlation between survival and expression of genes related to the circadian oscillator in patients with colorectal cancer, Chronobiology International, DOI: 10.1080/07420528.2018.1488722 To link to this article: https://doi.org/10.1080/07420528.2018.1488722 Published online: 28 Jun 2018. Submit your article to this journal View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=icbi20 CHRONOBIOLOGY INTERNATIONAL https://doi.org/10.1080/07420528.2018.1488722 Sex-dependent correlation between survival and expression of genes related to the circadian oscillator in patients with colorectal cancer Kristina Hasakovaa, Marian Vicianb, Richard Reisc, Michal Zemana, and Iveta Herichova a a Department of Animal Physiology and Ethology, Faculty of Natural Sciences, Comenius University Bratislava, Bratislava, Slovak Republic; Fourth Surgery Department, University Hospital, Comenius University Bratislava, Bratislava, Slovak Republic; cFirst Surgery Department, University Hospital, Comenius University Bratislava, Bratislava, Slovak Republic b ABSTRACT ARTICLE HISTORY Recent evidence supports the important role of the circadian system in cancer progression in humans. The aim of the present study is to evaluate clock (cry1, cry2 and per2) and clockcontrolled (vascular endothelial growth factor-a, early growth response protein 1 and estrogen receptor β) gene expression in colorectal cancer and adjacent tissue and identify a possible link between survival of patients and expression of above mentioned genes. The study includes 64 patients of both sexes with previously diagnosed colorectal cancer. RNA was extracted from the tumor tissue and adjacent parts of the resected colon, and real-time PCR was used for detection of clock gene expression. Expression of cry2 and per2 was significantly downregulated in tumor tissue compared to adjacent tissues. After splitting of the cohort according to sex, we detected downregulated levels of cry2 and per2 in male patients, but not in females. Splitting of male and female sub-cohorts according to presence of metastases revealed significant donwregulation of cry2 expression in female patients without distant metastasis. Better survival rate was associated with low expression of cry2 in female patients. Moreover, we observed an increase in cry1 expression in female patients with distant metastases in tumor compared to adjacent tissue. Accordingly, women with high expression of cry1 in tumor tissue displayed worse survival, which was not observed in men. Taken together, expression of clock and clock-controlled genes in tumors of males and females clustered according to presence of distant metastases correlated with survival analysis. Studied clock-controlled genes also showed sex-dependent changes. Low expression of vegf-a in tumor correlated with better survival in men but not in women. High expression of estrogen receptor β mRNA was related to better survival in women but not in men. Low expression of vegf-a, egr1 and estrogen receptor β was associated with worse survival in women compared to men. Our data indicate sex-dependent associations between clock and clock-controlled gene expression in cancer tissue and patient’s survival prognosis. Received 16 February 2018 Revised 9 June 2018 Accepted 11 June 2018 Introduction An association between cancer progress and circadian system disruption is supported by both experimental and epidemiological evidence (Filipski and Lévi 2009; Schernhammer et al. 2001). The circadian system coordinates rhythms in physiology and behavior and allows anticipation of 24 h cycles in the environment. In mammals, this system consists of hierarchically organized oscillators localized in the suprachiasmatic nucleus (SCN) of the hypothalamus (master oscillator) and all other tissues (peripheral oscillators). At the molecular level, the circadian rhythms are generated by a transcriptional–translational feedback loop that results in oscillation of clock gene KEYWORDS Circadian; male; female; per2; cry1; cry2; vegf-a; egr1; estrogen expression. The key components of the molecular mechanism are transcriptional factors BMAL1 and CLOCK (or its functional homolog NPAS2) that, through binding to the regulatory region E-box, induce the transcription of period (per1-3) and cryptochrome (cry1-2) genes. The protein products of per and cry genes are allocated into the nucleus and repress their own transcription by inhibition of BMAL1-CLOCK heterodimer activity. It takes approximately 24 h to complete the negative feedback loop. The E-box element is also located in the promoter region of many clock-controlled genes (CCGs) so the complex BMAL1-CLOCK (NPAS2) can regulate transcription of other genes involved CONTACT Iveta Herichova herichova1@uniba.sk Department of Animal Physiology and Ethology, Faculty of Natural Sciences, Comenius University Bratislava, Ilkovicova 6, 842 15, Bratislava, Slovak Republic © 2018 Taylor & Francis Group, LLC 2 K. HASAKOVA ET AL. in cellular processes such as cell proliferation (Albrecht 2012). A cell cycle inhibitor, wee1, is a frequently studied CCG. The expression of wee1 displays a robust circadian rhythm in the liver and is positively regulated by the BMAL1-CLOCK heterodimer (Matsuo et al. 2003). In contrast, Sotak et al. (2013) did not observe a rhythm of wee1 in chemically induced colorectal tumors and surrounding healthy tissues in 52-week-old mice. However, when the colonic tissue from younger (10-weekold) healthy mice was analyzed, wee1 showed a significant daily rhythm. In the absence of cry1 and cry2, wee1 expression shows upregulated levels. Liver regeneration was disrupted in cry1/2 knockout mice due to increased levels of wee1 (Matsuo et al. 2003). Heterodimer BMAL1NPAS2 directly targets and represses expression of transcription factor c-myc, which was found to be deregulated in mPer2 mutant mice (mPer2m/m) (Fu et al. 2002). Expression of c-myc is arrhythmic in the colorectal tumors and surrounding tissues (Sotak et al. 2013). Deregulated expression of several genes involved in cell cycle regulation, including cyclin D1, cyclin A, Mdm-2 and Gadd45, has been described in mPer2m/m mice (Fu et al. 2002). Transcriptional factors involved in cell cycle control can influence the basic circadian molecular feedback loop. A frequently mutated cell cycle component, p53, has the capacity to decrease expression of per2 through binding to its response element that overlaps with the E-box in the per2 promoter (Miki et al. 2013). A direct association between mutated per2 and cancer growth was found in mPer2m/m mice (Fu et al. 2002). γ-irradiation induced tumor development and decreased apoptosis occurrence in thymocytes in mPer2m/m mice. Moreover, expression of the core clock genes was inducible by γ-irradiation in the wild type mice but not in mPer2m/m mice. All these findings support the assumption that per2 possesses tumor suppressor activity (Fu et al. 2002). The negative correlation between per2 expression and tumor staging and grading and age in human colorectal cancer tissue has also been observed (Wang et al. 2011; Zeman et al. 2008). Our previous study revealed a correlation between cell cycle regulatory factors hus1, gadd45α, rb1, cdkn2α, and mre11α and per2 expression in human colorectal cancer tissue depending on TNM (tumor, node, metastasis) stage (Storcelova et al. 2013). The circadian rhythm in expression of per1, per2, dbp and reverb is reduced in colorectal cancer tissue compared to normal colonic tissue in mice. Furthermore, a daily rhythm in bmal1 expression is completely abolished in this tissue (Sotak et al. 2013). The increased activity of the cyclin D and βcatenin signaling pathway is the possible mechanism that explains how the downregulation or mutation in per2 may contribute to increased cell proliferation and worse cancer prognosis (Wood et al. 2008). The role of cry1 and cry2 in carcinogenesis is not well understood yet. Despite the fact that the cry1−/- cry2−/- double mutant does not show any signs of increased radiation sensitivity, there are experimental data which indicate association of cry genes with cell proliferation and DNA repair. When the cry1−/- cry2−/- mutation is combined with mutation of p53, animals are more resistant to early onset of cancer and have better survival compared to p53 mutant animals (Gauger and Sancar 2005). DNA repair factor XPA is also controlled by circadian system and the rhythm of XPA is abolished in the liver of double mutant cry1−/cry2−/- mice (Kang et al. 2010). The link between the circadian system deregulation and the development of cancer has been established by epidemiological analysis, meta-analysis and animal studies. Schernhammer et al. (2001) demonstrated a higher incidence of breast cancer in nurses working on rotating night shift for up to 29 years. Long-term exposure to night shifts was also positively correlated with incidence of colorectal cancer (Schernhammer et al. 2003). These findings contributed to shift work that involves a circadian disruption being classified as “probably carcinogenic in humans” in 2007 (Straif et al. 2007). Additional research is needed for better understanding and clarification of the mechanism of circadian disruption-induced carcinogenesis in vulnerable individuals (Haus and Smolensky 2013). In our study we also focused on several clockcontrolled genes that are involved in regulation of cell proliferation or apoptosis. Vascular endothelial growth factor (VEGF) plays an important role 3 CHRONOBIOLOGY INTERNATIONAL in tumor induced vascular proliferation and angiogenesis. Increased level of vegf was reported in many malignant tumors and the expression of vegf is associated with worse prognosis (George et al. 2001; Koutras et al. 2015). The mRNA level of vegf shows circadian oscillation in tumor tissue and per2 and cry1 inhibit the hypoxia induced vegf promoter activity (Koyanagi et al. 2003). Early growth response protein 1 (egr1) is a transcription factor that is in the mouse liver expressed in circadian manner and is directly regulated by CLOCK-BMAL1 heterodimer through E-box element (Tao et al. 2015). Depending on the type of tissue and external stimuli, egr1 is considered to be either tumor supressor (Krones-Herzig et al. 2005) or tumor promoter (Gitenay and Baron 2009). The estrogen receptor beta (ERβ) mRNA exhibit circadian oscillation in mouse lung and this rhythm is abolished in BMAL1 knockout mice. Promoter of ERβ gene contains conserved E-box element and is under direct circadian control (Cai et al. 2008). Recently it was demonstrated that men and women respond differently to chronotherapy of colorectal cancer (Giacchetti et al. 2012). This finding implicates sex-dependent differences in circadian oscillator functioning in tumor tissue. Therefore, the aim of our study was to analyze the expression of clock (cry1, cry2 and per2) and clock-controlled (vegf-a, egr1 and estrogen receptor β) genes in tumor and adjacent tissues and associate gene expression with sex, occurrence of distant metastases and survival of patients with colorectal cancinoma. Materials and methods The study included 64 patients of both sexes with previously diagnosed colorectal cancer (38 males and 26 females, average age 69 years). All patients were exposed to a standard hospital practice with lights on from 6:00 a.m. to 9:00 p.m. (The First Surgery Department, University Hospital Bratislava). The protocol was explained and informed consent was obtained from all participants included in the study. The experimental protocol was approved by the Ethics Committee. Histopathological examinations were performed by a hospital pathologist. Tissue samples taken during surgery were collected from the tumor and the proximal (≥ 10 cm above the tumor) and distal (≥ 2 cm under the tumor) parts of the resected colon. The surgery was conducted between 10:00 a.m. and 1:00 p.m. Tissue samples were placed into liquid nitrogen and stored at −80° C until further processing. Description of the cohort (tumor location, tumor type, nodus, metastasis classification, grading, clinical stage, age and sex) is provided in Table 1. Gene expression analysis Total RNA was isolated from the tissue samples using RNAzol reagent (MRC, USA) according to the manufacturer’s instructions. Complementary DNA was synthesized with the ImProm-II Reverse Transcription System (Promega, USA) according to the manufacturer’s instructions. Relative quantification of gene expression was performed with the Table 1. Patient’s characteristics including sex, age, tumor location, grading stage, clinical stage and TNM classification. All patients Sex Male Female Age Mean ± SD (years) Tumor location Proximal colon Transverse colon Distal colon Rectosigmoid colon Rectum Grading stage G1–1.5 G2–2.5 G3 Clinical stage I IIA, IIB IIIA, IIIB IVA, IVB TNM classification Primary tumor invasion T1–T2 T3 T4 Regional lymph node N0 N1 N2 Distant metastasis M0 M1 n 64 % 38 26 59 41 69 ± 12 18 7 11 10 18 28 11 17 16 28 11 49 4 17 77 6 4 29 15 16 6 45 24 25 4 48 12 6 75 19 35 13 16 55 20 25 48 16 75 25 4 K. HASAKOVA ET AL. QuantiTect SYBR Green PCR Kit (QIAGEN, Germany) and the StepOne TM Real-Time PCR System (Applied Biosystems, USA). The primer pairs used for the amplification of per2 were those used in our previous study (Zeman et al. 2008). Primers for the detection of cry1, cry2, vegf-a, egr1, ERβ and U6 were: cry1 (NM_004075.4) sense 5'-CCGTCTGTTTGTGATTCGTG-3', antisense 5'-AAGTTAGAGGCGGTTGTCCA-3'; cry2 (NM_ 001127457.2) sense 5'-GGAGGCTGGTGTGGAA GTAG-3', antisense 5'-CGTAGGTCTCGTCGTGG TTC-3'; vegf-a (NM_001171623.1) sense 5'AGAAGGAGGAGGGCAGAATC-3', antisense 5'CATCAGGGGCACACAGGAT-3'; egr1 (NM_0019 64.2) sense 5'- CAGCACCTTCAACCCTCAG-3', antisense 5'-GTCTCCACCAGCACCTTCTC-3'; ERβ (NM_001437.2) sense 5'-TGAGGGGAA ATGCGTAGAAGG-3', antisense 5'- CGTTCAGC AAGTGAGCCAG-3' and U6 (NR_004394.1) sense 5'-GCTTCGGCAGCACATATACTAA-3', antisense 5'-AAAATATGGAACGCTTCACGA-3'. Real-time PCR conditions were: activation of hot start polymerase at 95ºC for 15 min followed by 35–45 cycles at 94ºC for 15 s, 49ºC–53°C for 30 s (49º for cry1, cry2, per2; 52°C for U6; 53°C; vegf-a, egr1, ERβ), and 72ºC for 30 s. The specificity of PCR products were validated by melting curve analysis. Expression of U6 was used for normalization of clock and clock-controlled gene expression. Statistical analysis Differences in gene expression in tumor versus proximal and distal tissue were evaluated by oneway analysis of variance (ANOVA) and the Tukey post hoc test. Differences between male and female patients were calculated using the unpaired t-test. Differences between mRNA expression in tumor and adjacent tissue (averaged expression in the proximal and distal tissue) were performed by paired t-test. The Kaplan–Meier survival curve and log-rank test were used to evaluate 5 year survival of patients in relation to clock gene expression. Starting point for log-rank test was a day of surgery. Three living patients were excluded from this analysis since time from surgery did not meet 5 years period criterion. P-value < 0.05 was considered significantly different. Results In the whole cohort of patients expression of per2 and cry2 was significantly decreased in tumor tissue compared to adjacent proximal and distal tissues (ANOVA, P < 0.05, Figure 1). Expression of cry1 did not show this difference (Figure 1). Analysis focused on sex-dependent expression of clock genes revealed that expression of per2 was significantly reduced in tumor tissue compared with adjacent proximal and distal tissue in male patients (ANOVA, P < 0.05, Figure 2), but not in female patients. Moreover, expression of per2 was significantly lowered in female proximal and distal tissue compared with males (unpaired t-test, P < 0.05, Figure 2). Significant differences in cry1 expression were not observed between the two sexes and between the tumor and adjacent tissues (Figure 2) unless cohort was clustered more (Figure 4B). Expression of cry2 was significantly downregulated in tumor tissue in comparison to Figure 1. Expression of clock genes per2, cry1 and cry2 in tumor tissue and adjacent proximal and distal tissues. Values are presented as arithmetic means ± SEM (n = 64). Columns marked with different letters are significantly different. (ANOVA, Tukey post hoc test P < 0.05). CHRONOBIOLOGY INTERNATIONAL 5 Figure 2. Sex-dependent expression of clock genes per2, cry1 and cry2 in tumor and proximal and distal tissue in male (n = 38) and female patients (n = 26). Values are presented as arithmetic means ± SEM. White columns indicate proximal tissue, dashed columns indicate cancer tissue and plain gray columns are attributed to distal tissue. Letters indicate comparison between tumor and adjacent tissues within one sex and refer to significant difference by ANOVA (P < 0.05). An asterisk indicates comparison of the same tissue between two sexes and refers to significant difference by unpaired t-test. *P < 0.05 **P < 0.01. proximal and distal tissues in males (ANOVA, P < 0.05). This difference was not observed in female patients (Figure 2) unless cohort was clustered more (Figure 4B). The 5 year survival of patients was evaluated after splitting the cohort into low and high expression subgroups according to the median of clock and clock-controlled gene expression in the tumor tissue. The Kaplan–Meier analysis of a patient’s survival revealed significant sex-dependent differences. Low cry1 expression in tumor tissue was beneficial for female patients, who showed significantly better survival rates (Figure 3A) in comparison to female patients with high cry1 expression. Female patients with distant metastases exhibited higher expression of cry1 in tumor tissue when compared with adjacent tissues (Figure 4B). This difference was not observed in female patients without distant metastasis (Figure 4A,B). This observation is in a good agreement with female survival rate based on cry1 mRNA expression in tumor. No differences were observed in the survival of male patients with low and high expression of cry1 in tumor (Figure 3A). Female patients showed worse survival associated with high cry1 expression compared to male patients (Figure 1S A). Female patients with low cry2 expression in tumor tissue showed better survival rates in comparison with female patients with high cry2 expression (Figure 3B). This is in accordance with our finding that female patients without distant metastasis demonstrate lower cry2 expression in tumor tissue compared to adjacent tissue that was not observed in those with distant metastases (Figure 4B). A difference was not observed when female’s survival curves were related to per2 expression (Figure 3C). Male patients did not show differences in survival rates based on cry2 and per2 expression in tumor tissue (Figure 3B and C). Female patients showed worse survival associated with high cry2 expression compared to male patients (Figure 1S B). Low expression of vegf-a in tumor was related to better survival in men (Figure 3D). Survival of women did not show dependency on vegf-a expression (Figure 3D). vegf-a expression was significantly increased in tumor compared to adjacent tissue in men and higher variability was observed in patients distant metastases Figure 4A). In female patients, an increase in variability in sub-cohort with distant metastases caused that a difference between vegf-a expression in tumor and adjacent tissue did not reach level of significance (Figure 4B). Male patients with low expression of vegf-a showed better survival in comparison with female patients with low expression of vegf-a (Figure 1S D). When sub-cohort of male patients was split according egr1 expression a pronounced trend (P = 0.057) indicating association between low egr1 expression in tumor and better survival was 6 K. HASAKOVA ET AL. Figure 3. The Kaplan–Meier survival curves for patients with colorectal cancer sorted according to sex and cry1 (A), cry2 (B), per2 (C), vegf-a (D), egr1 (E) and estrogen receptor β (ERβ, F) mRNA expression. Solid line indicates low expression and dotted line indicates high expression. P - level of significance between high and low expression. Figure 4. Clock (cry1, cry2, per2) and clock-controlled gene (vegf-a, egr1, estrogen receptor β) expression in cancer tissue relativized to adjacent tissue in male (A) and female (B) patients with colorectal carcinoma. Values of mRNA expression in the adjacent tissue were calculated as an average of particular gene expression in proximal and distal tissue. Values above solid black line indicate increased gene expression in tumor compared to adjacent tissue. Values below solid black line indicate decreased gene expression in tumor compared to adjacent tissue. White columns are related to expression in patients without distant metastases (M0), gray columns show mRNA expression in patients with distant metastases (M). Paired t-test was used to compare expression in tumor and adjacent tissue. *P < 0.05 **P < 0.01, ***P < 0.001. CHRONOBIOLOGY INTERNATIONAL observed. On the contrary, low expression of egr1 in tumor was associated with trend to worse survival in female patients (Figure 3E). Decreased survival of females with low expression of egr1 is in agreement with our observation that egr1 levels in tumor tissue of female patients with distant metastases are significantly lower compared to adjacent tissue (Figure 4B). Survival of female patients with low egr1 expression in tumor was significantly lower compared to male patients (Figure 1S E). Low expression of ERβ mRNA in tumor was associated with worse survival in female patients. This observation was not confirmed in men (Figure 3F). ERβ expression showed significantly lower expression in tumor in female patients with metastases in comparison with those without nodus and metastases involvement (data not shown, unpaired t-test, P < 0.05). There was a significant decrease in ERβ mRNA expression in tumor compared to adjacent tissue in male patients, however, this decrease was not dependent on TNM stage as it was observed in female patients (Figure 4A,B). Low expression of ERβ mRNA in tumor was associated with worse survival in female compared to male patients (Figure 1S F). Discussion According to our data, there are differences in cry2 and per2 gene expression in tumor tissue compared to adjacent non-cancerous tissues in the complete cohort. Splitting of cohort according to sex revealed that decrease in cry2 and per2 expression was generated more by the male patients since decrease did not reach level of significance in female patients. When male and female subcohorts were split more, according distant metastases occurrence, a decrease in cry2 expression reached threshold of significance in female patients without distant metastases. Moreover, a trend to increased cry1 expression in tumor compared to adjacent tissues was significant in female patients with metastases. A survival analysis revealed correlation between high expressions of cry1, cry2 in tumor in female patients with worse 5 year survival that was not observed in male patients. Accordingly, female 7 patients showed worse survival associated with high cry1 and cry2 expression compared to male patients. In our study, the expression of cry1 did not exhibit any difference between tumor and noncancerous tissue in patients with colorectal cancer (CRC) without clustering of cohort. Other studies report either no difference between tumor and matched adjacent tissue or downregulation of cry1 expression in tumor tissue at the protein and mRNA level (Mazzoccoli et al. 2011, 2016). The authors of the previous study also described downregulated levels of cry1 mRNA in females, in patients between 62 and 74 years old, and in tumors located in the transverse colon. On the other hand, expression of protein CRY1 was found to be upregulated in the majority of CRC cell lines in comparison to the normal colon epithelium cell line and in primary CRC tissues compared to non-cancerous adjacent tissues. Higher expression of cry1 also correlated with lymph node metastasis and the TNM stage (Yu et al. 2013). We observed an increase of cry1 expression in tumor tissue in comparison to adjacent tissue in the female part of the cohort with distant metastasis. Our finding that expression of cry2 was lower in tumor tissue than in adjacent distal tissue in the whole cohort has been previously observed by Mazzoccoli et al. (2016). They found reduced levels of both protein and mRNA coding cry2 in CRC tissue samples in comparison to surrounding tissue. Lower expression of cry2 was also detected in tumors localized in the transverse colon in comparison with tumors localized in other parts of colon (Mazzoccoli et al. 2016). Silencing of cry2 in a human breast cancer cell line increased accumulation of mutagen-induced DNA damage compared with the cry2 positive cells (Hoffman et al. 2010a). Breast cancer patients have a higher degree of cry2 promoter methylation than control subjects and the difference is evident mainly in postmenopausal women (Hoffman et al. 2010b). These results indicate tumor suppressor activity of cry2. On the contrary, ectopic expression of both cry1 and cry2 genes in several CRC cell lines was associated with reduced apoptosis, increased proliferation and different responses to chemotherapy 8 K. HASAKOVA ET AL. (Mazzoccoli et al. 2016). Sex-dependent differences in cry2 expression were not previously reported. Decreased expression of mRNA and protein levels of per2 were found in several kinds of human cancers, (e.g. colorectal, breast, and gastric cancers) (Chen et al. 2005; Mazzoccoli et al. 2011; Zhao et al. 2014). These findings are in agreement with our data, which show decreased levels of per2 mRNA expression in tumor tissue. Downregulation of per2 induces proliferation of a murine breast cancer-derived cell line in vitro and tumor growth in vivo at specific times of the day (Yang et al. 2009). Higher expression of per2 in the colorectal carcinoma is associated with significantly better outcomes than lower expression (Oshima et al. 2011). This survival assumption was also confirmed by Cadenas et al. (2014) who demonstrated an association between higher expression of clock genes per2, per3 and cry2 and better metastasis free survival in breast cancer patients. Upregulation of per2 leads to increased levels of p53 and bax expression and greater apoptosis in the mouse mammary and Lewis lung carcinoma cell line (Hua et al. 2006). Our study revealed that only male patients show a difference between tumor and adjacent tissues, while no changes were observed in female tissue samples. Moreover, expression of per2 was significantly reduced in adjacent tissue of females in comparison to males. Colorectal cancer is a frequently diagnosed type of cancer in both males and females (Torre et al. 2015), however, the sex-dependent differences in the epidemiology of CRC are obvious. In general, women reach the same level of incidence and mortality of CRC 4–8 years later than men (Brenner et al. 2007). The reason for this sexual dimorphism is not fully understood, although the effects of sex steroid hormones have been suggested. Studies focused on postmenopausal women using hormone therapy identified lower incidence and risk of CRC in patients with hormonal supplementation in comparison with the control (Grodstein et al. 1999; Chlebowski et al. 2004). In contrast, another study demonstrated a promotion of CRC carcinogenesis by male hormones instead of protection by female hormones (Amos-Landgraf et al. 2014). Male steroid hormones, especially testosterone, can indirectly promote colorectal tumor development and growth in the ApcPirc/+b rat model, and depletion of male hormones protected ApcPirc/+b rats from CRC development. Ovariectomized female rats did not exhibit an increased incidence of colorectal tumors in this study (Amos-Landgraf et al. 2014). The sex steroid hormone status differs between men and women even after menopause (He et al. 2007). It is known that the level of sex steroid hormones varies during the course of a person’s life. Unlike estrogens, testosterone levels are higher in men in comparison with women even in older age. Decreased levels of estrogens in women after menopause can be lower than plasma estrogen concentration in men (Greendale et al. 1997; He et al. 2007). This decrease in estrogen levels can be followed by changed expression of clock genes and/or cell cycle control in a sex-dependent manner. This hypothesis has been supported by experimental evidence to some extent already. So far it is known that estrogens and testosterone influence tumor cell proliferation and apoptosis differently (Gupta and Singh 2008). Estrogen receptor α (ERα) reduces the pro-inflammatory activity and tumor-promoting properties of macrophages (Clocchiatti et al. 2016). ERβ is the dominant receptor in the colon and exerts cancerprotective effects as well. Estrogen mediates the colorectal cancer protective effects through ERβ in wild-type mice but not in ERβ knock-out mice (Weige et al. 2009). Moreover deficiency of ERβ is related to adenoma multiplicity, especially in the small intestine of female mice with Apc Min/+, and these tumors are larger than tumors collected from control Apc Min/+ mice (Giroux et al. 2008). Expression of ERβ is significantly reduced in colon cancer tissue compared with normal mucosa in female patients with an average age of 67. Moreover, the ERβ1 isoform exhibited decreased levels in female tumor tissue compared with tumors from male patients (Campbell-Thompson et al. 2001). Compared with high ERβ expression, low or lack of ERβ in tumor was associated with higher stage and worse patient’s survival (Rudolph et al. 2012). Tumor-promoting effects of androgens have been reported previously (Amos-Landgraf et al. CHRONOBIOLOGY INTERNATIONAL 2014). Studies have shown that androgen receptors may promote pro-inflammatory cytokine release (Clocchiatti et al. 2016). There is also evidence supporting the protective role of testosterone in carcinogenesis that can be mediated via membrane androgen receptors and its non-genomic proapoptotic effects in neoplastic cells (Gu et al. 2011). Increased CRC development and higher risk of colorectal neoplasia is seen in men on long-term androgen deprivation therapy (Gillessen et al. 2010). Therefore, more studies are needed to describe the role of testosterone in colorectal neoplasia. Gupta and Singh (2008) reported that cancer cells obtained from the tumor-bearing mice with Dalton’s lymphoma exhibited dimorphism in growth kinetics with higher growth in cells from females compared to cells obtained from males. Cells derived from male mice showed higher levels of apoptosis in tumor cells in comparison with cells derived from female mice. Expression of apoptosis regulation proteins p53 and caspaseactivated DNase showed a sex-dependent pattern as well, with higher expression in tumor cells obtained from male mice. Taken together, although more experimental evidence is needed, it is quite possible that sex hormones and distribution of their receptors may influence tumor progression. In our study we analyzed expression of ERβ in tumor and adjacent tissues and revealed strong dependence of ERβ on TNM staging in female patients that was not observed in male patients. Generally, expression of ERβ was strongly suppressed in tumor compared to adjacent tissues in both sexes. However, in female patients without nodus and metastasis involvement this decrease was only present as a nonsignificant trend and became a very pronounced with distance metastases occurrence. Accordingly, low ERβ expression in tumor was associated with lower survival rate in female patients. Female patients with low ERβ expression in tumor showed decreased survival rate compared to male patients with low ERβ expression in tumor. It was proven that sex steroid hormones can also affect a circadian system. The hormone 17βestradiol (E2) was previously shown to influence expression of cry1 and cry2 in a tissue-dependent 9 manner. E2 significantly increased levels of cry1 mRNA in the cerebral cortex but decreased cry2 mRNA expression in the SCN of the female mice (Nakamura et al. 2001). Positive correlation of fold change expression in a colorectal tumor versus normal mucosa in per1 and ERβ expression was observed in patients with grade G3. Lower per1 expression in female tumor tissue compared to males was detected previously (Mostafaie et al. 2009). Expression of per2 is inducible by E2 in estrogen receptor-positive breast cancer cell lines and binding of PER2 enhances ERα degradation, while suppression of per2 leads to ERα stabilization. Induction of per2 by E2 may occur through conserved estrogen response elements (ERE) localized in per2 promoter (Gery et al. 2007). E2 was also shown to affect the circadian rhythm of per2 and per1 expression in the liver and kidney in ovariectomized rats. Administration of E2 caused an increase in amplitude of per1 expression in the liver and kidney and an increase in per2 expression in the liver (Nakamura et al. 2005). Androgen receptors also possess the capacity to influence clock gene expression. After castration, mice exhibited upregulated levels of per2 expression in the prostate that were subsequently lowered by administering testosterone. Expression of clock and bmal1 showed the opposite pattern than per2 in this experimental model, and expression of clock and bmal1 was inducted by testosterone after castration (Kawamura et al. 2014). The male sex hormones can also influence the function of the SCN. Rats that underwent gonadectomy had significantly altered neuronal circuits in the SCN and changed response of per1 and per2 expression to photic stimulation compared with intact animals (Karatsoreos et al. 2011). The molecular mechanism of interactions between the sex steroid hormones, the circadian system and cell cycle control is not clear at this moment; however, epidemiological evidence supports the existence of clinically relevant interactions. This assumption is supported by a study demonstrating that chronotherapy does not appear to hold the same benefit for males and females (Giacchetti et al. 2012). Male patients benefit from chronotherapy more than females, and their survival improves more by chronotherapy than by conventional therapy. Females 10 K. HASAKOVA ET AL. exhibited different drug schedule tolerance and experienced severe toxicity more often than males while undergoing chronotherapy (Lévi et al. 2007). Yet it is possible that the optimal chronomodulated schedule differs between males and females. Our data support the existence of sex-dependent differences in clock gene expression and survival. Early growth response protein 1 is involved in the cell cycle control, proliferation and apoptosis. Previously, protein and mRNA of egr1 was found increased in colorectal cancer tissue compared to paired normal mucosa. Positive expression of egr1 was significantly associated with age, lymph node and distant metastasis, tumor stage and poor survival (Myung et al. 2014). Tumor promoting role in colorectal cancer also supports a study which revealed that egr1 negatively regulated the apoptosis in HCT15 colon carcinoma cells (Mahalingam et al. 2010). In our study we observed a trend to increased egr1 expression in tumor compared to adjacent tissue in men and in women without distant metastases. According to our data, egr1 expression was decreased in tumor tissue of female patients with distant metastasis. Better survival rate in female patients was associated with high egr1 expression. On the other hand, male patients exhibited strong trend to better 5 year survival with low egr1 expression. Obviously egr1 plays a role in tumor progression regulation but the mechanism behind is not completely elucidated yet. Expression of vegf is significantly up-regulated in colonic adenoma and carcinoma compared to normal colon, which suggest that angiogenesis is stimulated from early stages of CRC development (Wong et al. 1999). Negative or weak expression of vegf in tumor tissue was also correlated with better survival of patients with advanced colorectal cancer patients (Bendardaf et al. 2017). These observations are in agreement with our data, since we detected strong up-regulation of vegf-a expression in tumor tissue compared to adjacent tissues in all subgroups of patients. However, only male patients demonstrated significantly better survival rate associated with low vegf-a expression in tumor. Accordingly, low expression of vegf-a in men was associated with better survival in comparison to women. To summarize, our study demonstrated sexdependent changes in clock gene expression in tumor of patients with colorectal cancer as well as in 5 year survival. Expression of studied clockcontrolled genes also exerted sex-dependent associations in survival analysis. Expression of clock and clock-controlled genes in tumors of males and females clustered by the presence of distant metastases are in good agreement with our results from survival analysis. These findings might be potentially of interest in respect to implementation of personalized medicine and chronopharmacology into practice. Declaration of Interest The authors declare that they have no conflict of interest. Funding The study was supported by VEGA 1/0499/15, APVV-140318 and APVV-16-0209. ORCID Iveta Herichova http://orcid.org/0000-0002-0475-0461 References Albrecht U. 2012. Timing to perfection : The biology of central and peripheral circadian clocks. Neuron. 74 (2):246–60. Amos-Landgraf JM, Heijmans J, Wielenga MC, Dunkin E, Krentz KJ, Clipson L, Ederveen AG, Groothuis PG, Mosselman S, Muncan V, et al. 2014. Sex disparity in colonic adenomagenesis involves promotion by male hormones, not protection by female hormones. Proc Natl Acad Sci USA. 111(46):16514–165199. doi:10.1073/ pnas.1323064111. Bendardaf R, El-Serafi A, Syrjänen K, Collan Y, Pyrhönen S. 2017. The effect of vascular endothelial growth factor-1 expression on survival of advanced colorectal cancer patients. Libyan J Med. 12(1):1290741. doi:10.1080/ 19932820.2017.1290741. Brenner H, Hoffmeister M, Arndt V, Haug U. 2007. Gender differences in colorectal cancer : Implications for age at initiation of screening. Br J Cancer. 96(5):828–31. Cadenas C, van de Sandt L, Edlund K, Lohr M, Hellwig B, Marchan R, Schmidt M, Rahnenfuhrer J, Oster H, Hengstler JG. 2014. Loss of circadian clock gene expression is associated with tumor progression in breast cancer. Cell Cycle. 13(20):3282–91. CHRONOBIOLOGY INTERNATIONAL Cai W, Rambaud J, Teboul M, Masse I, Benoit G, Gustafsson JA, Delaunay F, Laudet V, Pongratz I. 2008. Expression levels of estrogen receptor beta are modulated by components of the molecular clock. Mol Cell Biol. 28(2):784–93. Campbell-Thompson M, Lynch IJ, Bhardwaj B. 2001. Expression of estrogen receptor (ER) subtypes and ERbeta isoforms in colon cancer. Cancer Res. 61(2):632– 40. Chen ST, Choo KB, Hou MF, Yeh KT, Kuo SJ, Chang JG. 2005. Deregulated expression of the PER1, PER2 and PER3 genes in breast cancers. Carcinogenesis. 26(7):1241–46. Chlebowski RT, Wactawski-Wende J, Rittenbaugh C, Hubbel FA, Ascensao J, Rodabough RJ, Rosenberg CA, Taylor VM, Harris R, Chen C, et al. 2004. Estrogen plus progestin and colorectal cancer in postmenopausal women. N Engl J Med. 350(10):991–1004. doi:10.1056/ NEJMoa032071. Clocchiatti A, Cora E, Zhang Y, Dotto GP. 2016. Sexual dimorphism in cancer. Nat Rev Cancer. 16(5):330–39. Filipski E, Lévi F. 2009. Circadian disruption in experimental cancer processes. Integr Cancer Ther. 8(4):298–302. Fu L, Pelicano H, Liu J, Huang P, Lee C. 2002. The circadian gene period2 plays an important role in tumor suppression and DNA damage response in vivo. Cell. 111(1):41–50. Gauger MA, Sancar A. 2005. Cryptochrome, circadian cycle, cell cycle checkpoints, and cancer. Cancer Res. 65 (15):6828–34. George ML, Tutton MG, Janssen F, Arnaout A, Abulafi AM, Eccles SA, Swift RI. 2001. VEGF-A, VEGF-C, and VEGF-D in colorectal cancer progression. Neoplasia. 3(5):420–27. Gery S, Virk RK, Chumakov K, Yu A, Koeffler HP. 2007. The clock gene Per2 links the circadian system to the estrogen receptor. Oncogene. 26(57):7916–20. Giacchetti S, Dugué PA, Innominato PF, Bjarnason GA, Focan C, Garufi C, Tumolo S, Coudert B, Iacobelli S, Smaaland R, et al. 2012. Sex moderates circadian chemotherapy effects on survival of patients with metastatic colorectal cancer : a meta-analysis. Ann Oncol. 23 (12):3110–16. doi:10.1093/annonc/mds148. Gillessen S, Templeton A, Marra G, Kuo Y, Valtorta E, Shahinian VB. 2010. Risk of colorectal cancer in men on long-term androgen deprivation therapy for prostate cancer. J Natl Cancer Inst. 102(23):1760–70. Giroux V, Lemay F, Bernatchez G, Robitaille Y, Carrier JC. 2008. Estrogen receptor beta deficiency enhances small intestinal tumorigenesis in ApcMin/+ mice. Int J Cancer. 123(2):303–11. Gitenay D, Baron VT. 2009. Is EGR1 a potential target for prostate cancer therapy? Future Oncol. 5(7):993–1003. Greendale GA, Edelstein S, Barrett-Connor E. 1997. Endogenous sex steroids and bone mineral density in older women and men: The Rancho Bernardo Study. J Bone Miner Res. 12(11):1833–43. Grodstein F, Newcomb PA, Stampfer MJ. 1999. Postmenopausal hormone therapy and the risk of colorectal cancer : A review and meta-analysis. Am J Med. 106 (5):574–82. 11 Gu S, Papadopoulou N, Nasir O, Föller M, Alevizopoulos K, Lang F, Stournaras C. 2011. Activation of membrane androgen receptors in colon cancer inhibits the prosurvival signals Akt/Bad in vitro and in vivo and blocks migration via vinculin/actin signaling. Mol Med. 17(1–2):48–58. Gupta V, Singh SM. 2008. Gender dimorphism of tumor growth: Role of gonadal hormones in differential regulation of apoptosis of a murine T cell lymphoma. J Biomed Sci. 15(2):147–62. Haus EL, Smolensky MH. 2013. Shift work and cancer risk : Potential mechanistic roles of circadian disruption, light at night, and sleep deprivation. Sleep Med Rev. 17(4):273–84. He H, Yang F, Liu X, Zeng X, Hu Q, Zhu Q, Tu B. 2007. Sex hormone ratio changes in men and postmenopausal women with coronary artery disease. Menopause. 14(3 Pt 1):385–90. Hoffman AE, Zheng T, Ba Y, Stevens RG, Yi CH, Leaderer D, Zhu Y. 2010a. Phenotypic effects of the circadian gene Cryptochrome 2 on cancer-related pathways. BMC Cancer. 10:110. Hoffman AE, Zheng T, Ch-H Y, Stevens RG, Ba Y, Zhang Y, Leaderer D, Holford T, Hansen J, Zhu Y. 2010b. The core circadian gene Cryptochrome 2 influences breast cancer risk, possibly by mediating hormone signaling. Cancer Prev Res. 3(4):539–48. Hua H, Wang Y, Wan C, Liu Y, Zhu B, Yang C, Wang X, Wang Z, Cornelissen-Guillaume G, Halberg F. 2006. Circadian gene mPer2 overexpression induces cancer cell apoptosis. Cancer Sci. 97(7):589–96. Kang TH, Lindsey-Boltz LA, Reardon JT, Sancar A. 2010. Circadian control of XPA and excision repair of cisplatinDNA damage by cryptochrome and HERC2 ubiquitin ligase. Proc Natl Acad Sci USA. 107:4890−4895. Karatsoreos IN, Butler MP, Lesauter J, Silver R. 2011. Androgens modulate structure and function of the suprachiasmatic nucleus brain clock. Neuroendocrinology. 152 (52):1970–78. Kawamura M, Tasaki H, Misawa I, Chu G, Yamauchi N, Hattori MA. 2014. Contribution of testosterone to the clock system in rat prostate mesenchyme cells. Andrology. 2(2):225–33. Koutras A, Kotoula V, Fountzilas G. 2015. Prognostic and predictive role of vascular endothelial growth factor polymorphisms in breast cancer. Pharmacogenomics. 16(1):79–94. Koyanagi S, Kuramoto Y, Nakagawa H, Aramaki H, Ohdo S, Soeda S, Shimeno H. 2003. A molecular mechanism regulating circadian expression of vascular endothelial growth factor in tumor cells. Cancer Res. 63(21):7277– 83. Krones-Herzig A, Mittal S, Yule K, Liang H, English C, Urcis R, Soni T, Adamson ED, Mercola D. 2005. Early growth response 1 acts as a tumor suppressor in vivo and in vitro via regulation of p53. Cancer Res. 65(12):5133–43. Lévi F, Focan C, Karaboué A, de la Valette V, Focan-Henrard D, Baron B, Kreutz F, Giacchetti S. 2007. Implications of circadian clocks for the rhythmic delivery of cancer therapeutics. Adv Drug Deliv Rev. 59(9–10):1015–35. 12 K. HASAKOVA ET AL. Mahalingam D, Natoni A, Keane M, Samali A, Szegezdi E. 2010. Early growth response-1 is a regulator of DR5induced apoptosis in colon cancer cells. Br J Cancer. 102 (4):754–64. Matsuo T, Yamaguchi S, Mitsui S, Emi A, Shimoda F, Okamura H. 2003. Control mechanism of the circadian clock for timing of cell division in vivo. Science. 302 (5643):255–59. Mazzoccoli G, Colangelo T, Panza A, Rubino R, De Cata A, Tiberio C, Valvano MR, Pazienza V, Merla G, Augello B, et al. 2016. Deregulated expression of cryptochrome genes in human colorectal cancer. Mol Cancer. 15:6. Mazzoccoli G, Panza A, Valvano MR, Palumbo O, Carella M, Pazienza V, Biscaglia G, Tavano F, Di Sebastiano P, Andriulli A, et al. 2011. Clock gene expression levels and relationship with clinical and pathological features in colorectal cancer patients. Chronobiol Int. 28(10):841–51. doi:10.3109/07420528.2011.615182. Miki T, Matsumoto T, Zhao Z, Lee CC. 2013. p53 regulates Period2 expression and the circadian clock. Nat Commun. 4:2444. Mostafaie N, Kállay E, Sauerzapf E, Bonner E, Kriwanek S, Cross HS, Huber KR, Krugluger W. 2009. Correlated downregulation of estrogen receptor beta and the circadian clock gene Per1 in human colorectal cancer. Mol Carcinog. 48(7):642–47. Myung DS, Park YL, Kim N, Chung CY, Park HC, Kim JS, Cho SB, Lee WS, Lee JH, Joo YE. 2014. Expression of early growth response-1 in colorectal cancer and its relation to tumor cell proliferation and apoptosis. Oncol Rep. 31(2):788–94. Nakamura TJ, Moriya T, Inoue S, Shimazoe T, Watanabe S, Ebihara S, Shinohara K. 2005. Estrogen differentially regulates expression of Per1 and Per2 genes between central and peripheral clocks and between reproductive and nonreproductive tissues in female rats. J Neurosci Res. 82(5):622–30. Nakamura TJ, Shinohara K, Funabashi T, Kimura F. 2001. Effect of estrogen on the expression of Cry1 and Cry2 mRNAs in the suprachiasmatic nucleus of female rats. Neurosci Res. 41(3):251–55. Oshima T, Takenoshita S, Akaike M, Kunisaki C, Fujii S, Nozaki A, Numata K, Shiozawa M, Rino Y, Tanaka K, et al. 2011. Expression of circadian genes correlates with liver metastasis and outcomes in colorectal cancer. Oncol Rep. 25(5):1439–46. doi:10.3892/or.2011.1207. Rudolph A, Toth C, Hoffmeister M, Roth W, Herpel E, Jansen L, Marx A, Brenner H, Chang-Claude J. 2012. Expression of oestrogen receptor β and prognosis of colorectal cancer. Br J Cancer. 107(5):831–39. Schernhammer ES, Laden F, Speizer FE, Willett WC, Hunter DJ, Fuchs CS, Colditz GA. 2003. Night-shift work and risk of colorectal cancer in the nurses’ health study. J Natl Cancer Inst. 95(11):825–28. Schernhammer ES, Laden F, Speizer FE, Willett WC, Hunter DJ, Kawachi I, Colditz GA. 2001. Rotating night shifts and risk of breast cancer in women participating in the nurses’ health study. J Natl Cancer Inst. 93(20):1563–68. Sotak M, Polidarová L, Ergang P, Sumova A, Pacha J. 2013. An association between clock genes and clock-controlled cell cycle genes in murine colorectal tumors. Int J Cancer. 132(5):1032–41. Storcelova M, Vician M, Reis R, Zeman M, Herichova I. 2013. Expression of cell cycle regulatory factors hus1, gadd45a, rb1, cdkn2a and mre11a correlates with expression of clock gene per2 in human colorectal carcinoma tissue. Mol Biol Rep. 40(11):6351–61. Straif K, Baan R, Grosse Y, Secretan B, El Ghissassi F, Bouvard V, Altieri A, Benbrahim-Tallaa L, Cogliano V, WHO International Agency For Research on Cancer Monograph Working Group. 2007. Carcinogenicity of shift-work, painting, and fire-fighting. Lancet Oncol. 8 (12):1065–66. doi:10.1016/S1470-2045(07)70373-X. Tao W, Wu J, Zhang Q, Lai SS, Jiang S, Jiang C, Xu Y, Xue B, Du J, Li CJ. 2015. EGR1 regulates hepatic clock gene amplitude by activating Per1 transcription. Sci Rep. 16 (5):15212. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. 2015. Global cancer statistics, 2012. CA Cancer J Clin. 65(2):87–108. Wang Y, Hua L, Lu C, Chen Z. 2011. Expression of circadian clock gene human Period2 (hPer2) in human colorectal carcinoma. World J Surg Oncol. 9:166. Weige CC, Allred KF, Allred CD. 2009. Estradiol alters cell growth in nonmalignant colonocytes and reduces the formation of preneoplastic lesions in the colon. Cancer Res. 69(23):9118–24. Wong MP, Cheung N, Yuen ST, Leung SY, Chung LP. 1999. Vascular endothelial growth factor is up-regulated in the early pre-malignant stage of colorectal tumour progression. Int J Cancer. 81(6):845–50. Wood PA, Yang X, Taber A, Oh EY, Ayers SE, Al-Assaad Z, Carnevale K, Berger FG, Peña MM, Hrushesky WJ. 2008. Period 2 mutation accelerates ApcMin/+ tumorigenesis. Mol Cancer Res. 6(11):1786–93. Yang X, Wood PA, Ansell C, Hrushesky WJ. 2009. Circadian time-dependent tumor suppressor function of period genes. Integr Cancer Ther. 8(4):309–16. Yu H, Meng X, Wu J, Pan C, Ying X, Zhou Y, Liu R, Huang W. 2013. Cryptochrome 1 overexpression correlates with tumor progression and poor prognosis in patients with colorectal cancer. PLoS One. 8(4):e61679. Zeman M, Vician M, Monosikova J, Reis R, Herichova I. 2008. Deregulated expression of the per2 gene in human colorectal carcinoma. Mol Med Rep. 1(4):599–603. Zhao H, Zeng ZL, Yang J, Jin Y, Qiu MZ, Hu XY, Han J, Liu KY, Liao JW, Xu RH, et al. 2014. Prognostic relevance of Period1 (Per1) and Period2 (Per2) expression in human gastric cancer. Int J Clin Exp Pathol. 7 (2):619–30.