- Email: [email protected]
Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer
YGYNO-976836; No. of pages: 6; 4C: Gynecologic Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer Paweł Knapp a, Lubomir Bodnar b, Agnieszka Błachnio-Zabielska c,d, Magdalena Świderska d,⁎, Adrian Chabowski d a
Department of Gynecology and Gynecologic Oncology, Medical University of Bialystok, 24a Sklodowskiej-Curie Str., 15-276 Bialystok, Poland Department of Clinical Oncology, Military Institute of Medicine in Warsaw, 128 Szaserow Str., 04-141 Warsaw, Poland Department of Hygiene, Epidemiology and Metabolic Disorders, Medical University of Bialystok, 2c Mickiewicza Str., 15-222 Bialystok, Poland d Department of Physiology, Medical University of Bialystok, 2c Mickiewicza Str., 15-222 Bialystok, Poland b c
H I G H L I G H T S • • • •
Evaluation of selected sphingolipids in patients with advanced ovarian cancer Plasma C16-Cer,C18:1-Cer,C18-Cer concentration elevated in study group vs. control Increase concentration of 5 ceramide, S1P in ovarian tissue vs. control group Some sphingolipids can be used as potential biomarkers of ovarian cancer
a r t i c l e
i n f o
Article history: Received 11 May 2017 Received in revised form 27 July 2017 Accepted 30 July 2017 Available online xxxx Keywords: Ovarian cancer Carcinogenesis Sphingolipids Ceramide S1P Ovarian cancer biomarkers
a b s t r a c t Purpose. The role of lipids in carcinogenesis through induction of abnormal cell lines in the human body is currently undisputable. Based on the literature, bioactive sphingolipids play an essential role in the development and progression of cancer and are involved in the metastatic process. The aim of this study was to determine the concentration of selected sphingolipids in patients with advanced ovarian cancer (AOC, FIGO III/IV, high grade ovarian cancer). Methods. Seventy-four patients with ovarian cancer were enrolled. Plasma concentrations of C16-Cer, C18:1Cer and C18-Cer were assessed by LC/MS/MS. The content of tissue sphingolipids was measured using a UHPLC/ MS/MS. Results. Plasma concentration of 3 ceramides: C16-Cer, C18:1-Cer and C18-Cer was significantly elevated in women with advanced ovarian cancer compared to control group (P = 0.031; 0.022; 0.020; respectively). There were increases in concentration of 5 ceramides: C16-Cer, C18:1-Cer, C18-Cer, C24:1-Cer, C24-Cer (P = 0.025; 0.049; 0.032; 0.005; 0.013, respectively) and S1P (P = 0.004) in ovarian tissue of women with advanced ovarian cancer compared to healthy individuals. Importantly, significantly higher risk of ovarian cancer when the plasma concentration of C16-Cer N 311.88 ng/100 μl (AUC: 0.76, P = 0.0261); C18:1-Cer N 4.75 ng/100 μl (AUC: 0.77, P = 0.0160) and C18-Cer N 100.76 ng/100 μl (AUC:0.77, P = 0.0136) was noticed. Conclusions. Bioactive sphingolipids play an essential role in the development and progression of cancer and they also take part in the process of metastasizing. This study suggests that some sphingolipids can be used as potential biomarkers of advanced ovarian cancer and that they can play an important role in the pathogenesis of this disease. © 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction ⁎ Corresponding author at: Department of Physiology, Medical University of Bialystok, Poland, 2c Mickiewicza Str., 15-222 Bialystok, Poland. E-mail addresses: [email protected] (P. Knapp), [email protected] (L. Bodnar), [email protected] (A. Błachnio-Zabielska), [email protected] (M. Świderska), [email protected] (A. Chabowski).
Ovarian cancer is the fifth most common cancer in European women. Around the world, more than 200,000 women are estimated to develop ovarian cancer every year, and about 100,000 die of the disease [1].
http://dx.doi.org/10.1016/j.ygyno.2017.07.143 0090-8258/© 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
2
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Aggressive biology of ovarian cancer is strictly correlated with its histology. It is worth underlining that heterogeneity of observed ovarian cancer subtypes determining as far there are no screening tests for early diagnosis of this type of neoplasm and ovarian cancer is still routinely diagnosed in the advanced stages of that disease (manly FIGO stage III, IV) [2,3]. The role of lipids in carcinogenesis through induction of abnormal cell lines in the human body is currently undisputable. Based on the literature, bioactive sphingolipids play an essential role in the development and progression of cancer and are involved in the metastatic process [4,5]. Ceramides are some of the most important elements in the metabolism of sphingolipids. They can be either synthesized de novo from serine and palmitoyl-CoA in a reaction catalyzed by serine palmitoyltransferase (SPT) or produced through enzymatic hydrolysis of sphingomyelin by sphingomyelinase. Apart from ceramides, biologically active sphingolipids include: dihydroceramide, sphingosine-1phsophate (S1P), sphingosine and sphinganine. Based on recent research, ceramides have pro-apoptotic properties and constitute an important antineoplastic factor (lipid tumor suppressors). S1P acts antagonistically to ceramides, as it induces cell transformation, cancer cell proliferation, cell survival and blocks the anti-apoptotic mechanisms of ceramides. Moreover, an increase in the expression of sphingosine kinase (SPHK1), an enzyme that transforms sphingosine to S1P, causes accumulation of S1P in the cell, promoting carcinogenesis and tumor formation [6–8]. Despite the important role of sphingolipids in cancer biology, their metabolism in different malignant tumors is poorly investigated. Some studies showed marked differences in ceramide content between tumor and respective healthy tissue [9–11]. Interestingly, the level of this sphingolipid was either decreased or increased, suggesting that alterations in ceramide metabolism in cancer tissue may depend on a type of tumor [6,7,10,12]. However, these studies focused only on selected sphingolipid species, and to date there are no complex data on ceramide metabolism in cancer tissue. In addition, sphingolipid metabolism in ovarian cancer has been still poorly investigated. Therefore, measurement of the sphingolipids could lead to better understanding of the influence of ovarian cancer biology and possibly provide new biomarker(s) for non-invasive testing. 2. Material and methods The investigation conforms with the principles outlined in the Declaration of Helsinki and was approved by the Ethical Committee for Human Studies of the Medical University of Bialystok, Poland (recruitment between January 2015 till June 2016). All patients gave their informed consent prior to their inclusion in the study. The initial study included two groups of women: patients with suspicious of advanced ovarian cancer (AOC) (n = 165) and women with normal ovarian morphology confirmed by transvaginal sonography evaluation (control group, n = 149). The patients suspected to have an advanced ovarian cancer were examined in the Department of Gynecology and Gynecologic Oncology, Medical University of Bialystok and pre-treatment diagnosis were evaluated based on present adnexal masses with spread metastasis into pelvis and abdomen confirmed by CT scan, ascites, elevated level of Ca-125. 10 ml of peripheral blood (antecubital vein) was collected for EDTA probes from each patient 24 h before surgery procedures. The blood was centrifuged in 15 min, plasma subsequently separated and frozen at −80 °C temperature. Radical bulky surgery procedures according to gynecologic oncology guidelines (hysterectomy, bisalpingo-oophorectomy, total omentectomy, appendectomy, others) were performed in the group of patients clinically suspected of ovarian cancer. All applied procedures were standardized and repeatable. Neoplastic samples were collected by University Tissue Biobanking Team (one pathologist, one technician; according to Standard Operating Procedures for Biobanking - SOP) from an ovarian surface not deeper than 1 cm, just after removal of the uterus and ovaries. Dissected
tissues were immediately placed into liquid nitrogen and time from surgical closing last artery vasculating ovarian tumor to tissue collecting was not longer than 2 min. Normal ovarian tissues were obtained during operations because of fibroids or other non-oncology procedures. Hysterectomy was performed and ovarian samples were collected and handled as described above. Only women with histologically diagnosed advanced ovarian cancer serous type, the International Federation of Gynecology and Obstetrics (FIGO) Classification - stage III/IV, high-grade (2014 WHO Classification) were included in the study. We sequentially excluded from the group of 165 women: 48 borderline tumors, 43 of non-epithelial histology; and from the control group: 68 patients with some abnormalities in ovarian tissue diagnosed during histology evaluation. Final groups counted: advanced ovarian cancer: n = 74 patients, control group: n = 81 women. 2.1. The content of plasma sphingolipids The level of sphingolipids in plasma was analyzed by LC/MS/MS approach as previously described by Blachnio-Zabielska et al. with minor modification [13]. Briefly, to each plasma sample (100 μl) were added 50 μl of the internal standard solution (17C–sphingosine and 17C-S1P, and C17-Cer Avanti polar lipids) as well as 1.5 ml of an extraction mixture (isopropanol: water: ethyl acetate, 35:5:60; v:v:v). The following sphingolipids were quantified: SPH (sphingosine), S1P (sphingosine1-phosphate), SPA (sphinganine), ceramide C14:0-Cer (ceramides containing myristic acid), C16:0-Cer (ceramides containing palmitic acid), C18:1-Cer (ceramides containing oleic acid), C18:0-Cer (ceramides containing stearic acid), C20:0-Cer (ceramides containing arachidic acid), C22:0-Cer (ceramide containing behenic acid), C24:1-Cer (ceramides containing nervonic acid) and C24:0-Cer (ceramides containing lignoceric acid). The mixture was vortexed, sonicated and then centrifuged for 10 min at 4000 rpm (Sorvall Legend RT). The supernatant was transferred to a new tube and pellet was re-extracted. After centrifugation supernatants were combined and evaporated under nitrogen. The dried sample was reconstituted in 100 μl of LC Solvent A (2 mM ammonium formate, 0.15% formic acid in methanol) for LC/MS/MS analysis. Quantitative measurement was made using triple quadrupole mass spectrometer (Agilent 6460) in positive mode using multiple reaction monitoring (MRM). The chromatographic separation was performed on Agilent 1290 Infinity Ultra High Performance Liquid Chromatography (UHPLC). The analytical column was a reverse-phase Zorbax SB-C8 column 2.1 × 150 mm, 1.8 μm. Chromatographic separation was conducted in binary gradient using 2 mM ammonium formate, 0.15% formic acid in methanol as Solvent A and 1.5 mM ammonium formate, 0.1% formic acid in water as Solvent B at the flow rate of 0.4 ml/min. All sphingolipids were quantified against standard concentration curve. The flow was diverted to waste for the first and the last 4 min to prevent eluting impurities from entering the mass spectrometer. 2.2. The content of tissue sphingolipids The content of sphingolipids was measured using a UHPLC/MS/MS (an ultra-high performance liquid chromatography coupled with triple quadrupole mass spectrometry) approach according to BlachnioZabielska et al. [13]. Briefly, the tissue samples (20 mg) were homogenized in a solution composed of 0.25 M sucrose, 25 mM KCl, 50 mM Tris and 0.5 mM EDTA, pH 7.4. Immediately afterwards, 50 μl of the internal standard solution (17C–sphingosine and 17C–S1P, and C17-Cer Avanti polar lipids) as well as 1.5 ml of an extraction mixture (isopropanol: water: ethyl acetate, 35:5:60; v:v:v) were added to each homogenate. The mixture was vortexed, sonicated and then centrifuged for 10 min at 4000 rpm (Sorvall Legend RT). The supernatant was transferred to a new tube and pellet was re-extracted. After centrifugation supernatants were combined and evaporated under nitrogen. The dried sample was reconstituted in 100 μl of LC Solvent A (2 mM
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
ammonium formate, 0.15% formic acid in methanol) for LC/MS/MS analysis. Sphingolipids were analyzed by means of an Agilent 6460 triple quadrupole mass spectrometer using positive ion electrospray ionization (ESI) source with multiple reaction monitoring (MRM) against standard curves constructed for each analyzed compound. The chromatographic separation was performed using an Agilent 1290 infinity ultra-performance liquid chromatography (UHPLC). The analytical column was a reverse-phase Zorbax SB-C8 column 2.1 × 150 mm, 1.8 μm. Chromatographic separation was conducted in binary gradient using 2 mM ammonium formate, 0.15% formic acid in methanol as solvent A and 1.5 mM ammonium formate, 0.1% formic acid in water as solvent B at the flow rate of 0.4 ml/min. 2.3. Statistical analysis The method was validated for the use of sphingolipid estimation in biological tissue and was also used for sphingolipids measurement in human plasma [14,15]. Inter and intra assay reproducibility of the method is around 1% for most of the sphingolipid species. The list of quantified sphingolipids with the limit of detection is provided in Table 1 (Supplementary Material). Descriptive statistics including mean concentration and standard error of the mean concentration were calculated for selected sphingolipids, henceforth called features. In order to detect statistically significant differences between examined groups, either fitting an analysis of variance model was conducted or non-parametric method (Wilcoxon rank-sum test) was applied. The choice of an appropriate method was made upon fulfilling the normality and the homogeneity of variances assumptions and in case of violation of at least one condition non-parametric approach was employed. The normality of features distribution was checked with the Shapiro-Wilk test and the homogeneity of variances with the Levene's test. Features that have been found significant, i.e. their distribution was statistically significantly different among experimental groups, were taken under further investigation to discover their prediction capability. To accomplish that, for each significant feature: ROC curves (receiver operating characteristic curve) were constructed and optimal threshold values were determined with the Youden method. Confidence intervals for sensitivity and specificity corresponding to a particular threshold were calculated with the use of the Wilson method and a test verifying the area under curve (AUC), that was significantly N 0.5 (random classification) was performed with the DeLong method - P-values and one-sided confidence intervals for AUC are reported. Calculations concerning ROC curves and corresponding tests were conducted with the functions provided by the pROC R package. Confidence intervals for sensitivity and specificity were constructed with the use of the binom.confint function, part of the binom R package. All calculations were carried out in R software environment. Finally, Spearman rank test was used with respect to correlate each examined sphindolipids derivativeness. Significance level alpha equal to 0.05 was applied for all statistical tests. 3. Results Patients' characteristic and clinicopathologic data are presented in Table 1 and 2. Patients from both groups were matched for several factors (age, body mass index, standard analyses, Ca-125, etc.) to make sure that two groups are comparable and there are no statistically significant differences between them. The values of mean concentration and 25% - 75% percentile of plasma sphingolipids and ovarian tissue sphingolipids in each study group are presented respectively in Table 3 and 4. There were no significant differences in mean values for any ceramide species between patients who were premenopausal versus postmenopausal. The levels did not correlate with tumor size (data not shown). We showed significant increase in concentration of 3 ceramides: C16-Cer and C18:1-Cer, C18-Cer (P = 0.031; 0.022; 0.020; respectively)
3
Table 1 Characteristic of the patients with serous advanced ovarian cancer (AOC) and control group included into the study. Parameters
Patients with AOC (n = 74)
Control group (n = 81)
P value
Age (median ± SD) BMI (median ± SD) Systolic BP (median ± SD) Diastolic BP (median ± SD) RBC (median ± SD) HGB (median ± SD) PLT (median ± SD) WBC (median ± SD) Ca125 (median ± SD) Glucose (median ± SD) Triglycerides (median ± SD) Total cholesterol (median ± SD) Total proteins (median ± SD)
54.2 ± 3.7 27.29 ± 4.24 123.47 ± 13.97 79.43 ± 10.93 4.07 ± 0.84 11.89 ± 1.80 580.53 ± 146.63 9.15 ± 2.48 822.40 ± 572.25 83.80 ± 11.16 1.53 ± 0.02 174.63 ± 20.12 5.96 ± 0.78
57.9 ± 1.9 27.26 ± 4.15 123.34 ± 14.13 80.13 ± 10.79 3.79 ± 0.75 12.72 ± 1.55 289.50 ± 96.18 7.81 ± 1.88 17.81 ± 8.08 82.47 ± 17.63 1.58 ± 0.3 173.56 ± 27.49 6.98 ± 0.68
0.005⁎ 0.893 0.833 0.854 0.157 0.056 0.0001⁎ 0.047⁎ 0.0001⁎ 0.481 0.513 0.735 0.0001⁎
SD – standard deviation. ⁎ P-value b 0.005.
in plasma of women with advanced ovarian cancer compared to control group (Table 3). Furthermore, we showed increase in concentration of 5 ceramides: C16-Cer, C18:1-Cer, C18-Cer, C24:1-Cer, C24-Cer (P = 0.025; 0.049; 0.032; 0.005; 0.013, respectively) and S1P (P = 0.004) in ovarian tissue of women with serous advanced ovarian cancer compared to control group. We also observed statistically higher level of SPH in healthy ovarian tissue then in ovarian cancer (P = 0.005) (Table 4). There was no correlation between plasma vs. ovarian cancer tissue sphingolipids in individual patients. Plasma C16, C18.1, C18 correlated with disease progression of ovarian cancer: FIGO staging (R = 0.26, P = 0.041, R = 0.26, P = 0.044, R = 0.31, P = 0.015 respectively), and with differentiate the grading (R = 0.26, P = 0.040, R = 0.28, P = 0.030, R = 0.29, P = 0.021 respectively) (Table 4). Moreover, plasma Sph, SPA and S1P concentration has shown a significant association with leukocyte (WBC) levels (R = −0.23, P = 0.001, R = −0.30, P = 0.018, R = −0.26, P = 0.039) (Table 5). We included all statistically significant sphingolipids in later ROC analyses but we created ROC curves only for sphingolipids significantly changed in plasma (which has potential for noninvasive diagnosis),
Table 2 Clinicopathologic characteristic of the patients with high-grade serous advanced ovarian cancer. Parameters
N
%
FIGO stage IIIA IIIB IIIC IVA IVB
2 18 49 1 4
2.7 24.3 66.2 1.4 5.4
Lymph nodes N0 N1 Not determined
2 71 1
2.7 95.9 1.4
Volume of ascites No ascites Ascites ≤ 500 ml Ascites N 500 ml
12 28 34
16.2 37.8 46.0
ECOG PS 0 1–2
69 5
93.2 6.8
Tumor residuals 0 1–10 mm N10 mm
50 19 5
67.5 25.7 6.8
ECOG PS – Eastern Cooperative Oncology Group performance status.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
4
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 3 Concentration of sphingolipids in plasma: ovarian cancer patients and healthy women (control group). Ovarian cancer patient (n = 74) Sphingolipids plasma concentration (ng/100 μl) mean (25–75% percentile) Total Cer 4236.3996 (2763.467–6872.3401) SPH 77.8381 (21.3466–132.4366) SPA 27.5618 (14.7352–42.4981) S1P 424.6286 (297.358–579.3469) C14-Cer 7.7032 (2.2389–12.1287) C16-Cer 403.6705 (224.8791–648.3847) C18:1-Cer 7.3511 (2.912–13.7634) C18-Cer 161.7371 (51.4581–270.3815) C20-Cer 117.1991 (46.4581–162.3947) C22-Cer 557.7762 (212.2437–795.3791) C24:1-Cer 1332.8207 (797.2736–1726.3849) C24-Cer 1648.1414 (1178.3746–2017.3884)
Healthy women (n = 81)
P-value
3772.7214 (1298.3961–6002.3411) 96.7350 (37.4936–167.4422) 26.8689 (13.4483–42.7346) 420.3498 (271.2312–690.3722) 8.0935 (3.3388–12.8849) 261.5972 (142.3771–559.9936) 3.8942 (0.8457–7.459) 92.4272 (33.4888–141.2882) 104.8481 (59.2777–177.6624) 514.9734 (201.2881–829.2836) 1174.8300 (523.3888–1579.2736) 1470.1474 (603.348–1992.3481)
0.231 0.102 0.966 0.8 0.955 0.031⁎ 0.022⁎ 0.020⁎ 0.057 0.464 0.076 0.921
⁎ Statistically significant value of b0.05 for Mann-Whitney test.
which set the threshold values and allowed predicting the likelihood of ovarian cancer with specific sensitivity and specificity. The area under the ROC curve for C16-Cer was 0.759, for C18:1-Cer was 0.768 and for C18-Cer was 0.771. All field values are satisfactory and indicate the usefulness of these biochemical markers as tools to predict the risk of ovarian cancer. We demonstrated a significantly higher risk of ovarian cancer when the plasma concentration of C16-Cer N 311.88 ng/100 μl (sensitivity: 0.77; specificity: 0.56, P = 0.0261); C18:1-Cer N 4.75 ng/100 μl (sensitivity: 0.9; specificity: 0.437, P = 0.0160) and C18-Cer N 100.76 ng/100 μl (sensitivity: 0.8; specificity: 0.57, P = 0.0136) (Supplementary Material Fig. 1, 2, 3). Diagnostic values of these sphingolipids in plasma are presented respectively in Table 6.
4. Discussion Ovarian cancer is a disease associated with a high mortality [1]. This study shows that sphingolipid metabolism in the human ovarian carcinoma is augmented as compared to the healthy tissue. It is supported by the fact that the content of measured sphingolipids and ceramide species were increased in the ovarian cancer tissue. Accumulation of sphingolipids has been noted in various cancers such as tumors of the head and neck, lung cancer, sarcomas, breast cancer and endometrial cancer. There are studies showing that certain tumors, including colon cancer and brain cancer, are characterized by decreased concentration of ceramides compared to healthy tissues [18,19]. Such data clearly indicate that sphingolipid metabolism is closely associated with cell transformation depending on histological type of the tissue. Interestingly, in ovarian cancer ceramide levels are elevated as compared to healthy ovarian tissue. However, we have to point out that various agonists increase the level of endogenous, intracellular long-chain ceramides; for
example, rise in temperature is associated with enhanced cancer cell metabolism. Elevated concentration of ceramides and enzymes catalyzing the transformation to bioactive derivatives and accumulation of sphinganine and dihydroceramide coupled to activation of palmitoyltransferase suggests that ceramides are synthesized de novo [20]. It seems that, as in ovarian cancer, the synthesis of ceramides in endometrial cancer is performed de novo [4]. This is justified, as substantial increase in the concentrations of sphinganine, dihydroceramide and ceramide expression have been observed in other cancer types [21–24]. High concentration of total ceramide in ovarian cancer is puzzling in the context of its antiproliferative and pro-apoptotic functions. It is probably related to overactivation of ceramidase that produces ceramides. Park et al. described such a mechanism, demonstrating increased activation of acidic ceramidase in various cancer patients [25]. This reasoning is supported by our own research showing that the concentration of sphingosine is 5.5 times lower in ovarian cancer in comparison to healthy ovarian tissue. Possibly, compared to healthy ovarian tissue, the activation of ceramidase in ovarian cancer is significantly higher and leads to increased transformation of sphingosine to ceramide (115.69 vs. 654.10 ng/ml). We made similar observations in endometrial cancer, where we also noted high levels of ceramide and S1P [4]. Total ceramide concentration also clearly depends on the histological type of cancer and its clinical and the degree of histological progression (staging, grading). The observed high concentration of total ceramide in ovarian cancer (1.4 times higher in comparison to healthy ovarian tissue) was probably related to the advancement of cancer in studied patients (FIGO stage III/IV, G2/G3). The role of ceramides in modulating cell signaling is primarily related to their pro-apoptotic role and inhibition of tumor growth [26]. In this context, high concentrations of ceramides in ovarian cancer are not in line with the expectations; however, high levels of ceramides are probably associated with
Table 4 Concentration of sphingolipids in ovarian tissue: ovarian cancer patients and healthy women (control group). Ovarian cancer patient (n = 74) Ovarian tissue sphingolipids concentration (ng/ml) Mean (25–75% percentile) Total Cer 5299.1784 (1287.2736–8091.2736) SPH 115.6994 (52.3761–171.2871) SPA 31.5615 (11.2376–59.3481) S1P 1253.3658 (629.2761–1762.2399) C14-Cer 22.8915 (12.2761–37.2761) C16-Cer 1053.2861 (699.26661–1751.2881) C18:1-Cer 100.1249 (54.28871–161.2917) C18-Cer 273.0186 (179.2779–530.3801) C20-Cer 125.3704 (66.6892–209.0972) C22-Cer 225.8186 (142.3721–376.3821) C24:1-Cer 1885.1075 (1062.38271–3119.5622) C24-Cer 1613.5925 (777.2817–3001.2998)
Healthy women (n = 81)
P-value
4343.4923 (2376.8737–6981.2837) 654.1071 (201.2837–1082.8521) 30.1124 (12.34–78.2381) 811.9984 (267.3338–1491.2716) 15.2976 (8.8726–31.27771) 910.877 (509.2881–1487.2719) 20.9839 (9.2781–33.3433) 166.2519 (87.6509–290.6661) 91.7388 (55.8701–167.2771) 333.9555 (164.2871–612.3672) 1696.1389 (762.2281–3001.2971) 1108.2964 (661.2881–1982.3888)
0.061 0.215 0.894 0.004⁎ 0.626 0.025⁎ 0.049⁎ 0.032⁎ 0.859 0.063 0.005⁎ 0.013⁎
⁎ Statistically significant value of b0.05 for Mann-Whitney test.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
5
Table 5 Spearman rank correlation for analyzed sphingolipids in ovarian cancer patients' plasma.
Sph_plasma SPA_plasma S1P_plasma C14_plasma C16_plasma C18:1_plasma C18_plasma C20_plasma C22_plasma C24:1_plasma C24_plasma Total Cer_plasma
r p r p r p r p r p r p r p r p r p r p r p r p
FIGO
Grading
BMI
RBC
HGB
PLT
WBC
Ca125
Glucose
Triglycerides
Total cholesterol
Total proteins
−0,22 0,087 −0,05 0,697 0,06 0,658 −0,01 0,959 0,26 0,041 0,26 0,044 0,31 0,015 0,24 0,063 0,09 0,472 0,21 0,101 0,02 0,899 0,14 0,284
−0,23 0,074 0,00 0,993 0,01 0,947 0,00 0,970 0,26 0,040 0,28 0,030 0,29 0,021 0,22 0,084 0,08 0,538 0,20 0,123 0,01 0,917 0,14 0,279
−0,08 0,537 0,00 0,988 0,13 0,322 0,13 0,317 −0,07 0,602 −0,14 0,284 0,02 0,902 −002 0,883 0,03 0,799 −0,07 0,594 −0,01 0,934 −0,04 0,759
−0,16 0,217 0,05 0,710 −0,01 0,932 0,07 0,563 0,07 0,569 −0,03 0,802 0,08 0,557 0,11 0,401 0,12 0,369 0,08 0,556 −0,04 0,760 0,08 0,517
0,14 0,292 0,10 0,442 −0,19 0,141 −0,10 0,435 −0,29 0,021 −0,07 0,563 −0,23 0,068 −025 0,054 −0,20 0,118 −0,26 0,044 −0,13 0,331 −0,21 0,109
−0,17 0,182 −0,08 0,540 0,07 0,592 0,05 0,681 0,32 0,013 0,25 0,054 0,33 0,009 0,29 0,025 0,16 0,228 0,29 0,022 0,03 0,820 0,21 0,103
−0,42 0,001 −0,30 0,018 −0,26 0,039 −0,11 0,416 −0,04 0,764 −003 0,790 −0,06 0,622 −016 0,200 −0,23 0,072 −0,13 0,297 −0,18 0,169 −0,19 0,140
−0,23 0,068 −0,16 0,203 0,17 0,200 −0,09 0,503 0,25 0,053 0,28 0,026 0,30 0,018 0,22 0,086 0,02 0,859 0,16 0,203 −0,08 0,516 0,07 0,608
0,04 0,766 0,07 0,603 0,23 0,074 0,01 0,954 −0,21 0,107 −0,18 0,153 −0,12 0,350 −015 0,253 −0,16 0,209 −0,20 0,122 −0,07 0,602 −0,16 0,213
0,00 0,998 −0,03 0,805 0,00 0,988 −0,14 0,291 −0,13 0,306 −0,16 0,221 0,01 0,916 −008 0,532 −0,19 0,130 −0,10 0,450 −0,31 0,016 −0,20 0,127
−0,28 0,025 −0,19 0,143 −0,04 0,753 0,08 0,517 0,01 0,958 0,04 0,746 0,01 0,956 −002 0,847 −0,11 0,389 0,00 0,985 −0,10 0,429 −0,11 0,408
0,06 0,666 −0,08 0,544 −0,08 0,513 −0,02 0,893 −0,13 0,327 −0,09 0,509 −014 0,285 −023 0,067 −0,18 0,162 −0,25 0,051 −0,13 0,308 −0,21 0,094
r – Spearman rank (−1,+1); p - statistically significant P-value of b0.05. Significance of bold p b 0.05.
the risk of epithelial ovarian cancer. This could be explained by the aggressive biological nature of ovarian cancer whereby high levels of S1P produced by the tumor promote cell proliferation and cell transformation as well as survival of abnormal cells, neoangiogenesis and inhibition of pro-apoptotic processes [2,20]. Moreover, dysregulation of sphingolipid metabolism could also promote development of ovarian cancer. Ovarian cancer cells can produce an excessive amount of ceramides, resulting in increased conversion to S1P. According to such an assumption, increased risk of ovarian cancer could be caused by shifting of balance towards S1P through dysregulation of sphingolipid metabolism, resulting in overproduction of S1P and ceramides. Elevated concentrations of 11 out of 12 studied sphingolipid derivatives that we observed in this study in ovarian cancer patients compared to control subjects, clearly indicate disturbed S1P metabolism due to increase in the total bioactive sphingolipids. Alberg et al. came to similar conclusions, reporting that increase in the concentration of bioactive sphingolipids in lung cancer can also potentially be due to the dysregulation of S1P metabolism [27]. Numerous recent studies indicate that ceramides can play distinctive and opposing roles in neoplastic tumors possibly caused by differences in the lengths of lipid chains [17,28– 30]. Analysis of C18-Cer function demonstrated its role in tumor suppression, while C16-Cer was shown to induce cell proliferation. In advanced ovarian cancer production [4,10,16,17]of C16-Cer (1053.26 ng/ml) augments the aggressive nature of the tumor and its function is superior to C18-Cer and C18:1-Cer (5 and 10 times lower concentrations, respectively). This can be a result of the above-mentioned total dysregulation of S1P metabolism, which could be a dominant phenomenon seen in advanced ovarian cancer. Moreover, resistance to chemotherapy in ovarian cancer can also be caused by S1P dysregulation. It is underscored that S1P is not only a messenger in cell signaling, but it also binds to membrane receptors coupled to a
G protein (S1PRs), which is responsible for the majority of its biological actions [31–33]. It is possible that large amounts of S1P synthesized by the tumor stimulate ovarian cancer cell proliferation through binding to the above-mentioned receptor. Observed low serum level of S1P in ovarian cancer patients results from its consumption by biologically aggressive tumor cells that constantly proliferate. The above-mentioned resistance to chemotherapy is due not only to dysregulation of ceramide and S1P metabolism, but also to the presence of protein mediators and their signaling actions. According to Gatt et al., increased expression of ceramide transporting proteins (CERT) and their intracellular accumulation associated with S1P is a possible cause of ovarian cancer resistance to chemotherapy with platinum derivatives [34–36]. As the biology of ovarian cancer is very aggressive, the tumor is often diagnosed at a late stage (FIGO III/IV) characterized by a 5-year survival rate of 20–30%. Early detection of ovarian cancer using biomarkers could improve total survival of patients with this type of cancer. Interestingly, there is a positive statistically significant correlation between the cancer progression (FIGO and Grading) and the plasma C16, C18.1, C18 level (R = 26, P = 0.04; R = 0.26, P = 0.044; R = 0.31, P = 0.015 respectively). We can assume that ceramics can reflect the stages of ovarian cancer development. Besides that our research showed that patients with ovarian cancer are characterized by significantly higher serum levels of C16-Cer, C18:1-Cer and C18-Cer compared to healthy controls. Moreover, based on Youden analysis, we determined the optimal cut-off values for studied ceramides indicating significantly elevated risk of ovarian cancer. It was determined that C16-Cer, C18:1-Cer and C18-Cer meet the criteria for biomarkers (AUC = 0.759; =0.768; =0.771) and can be used in the diagnosis of patients at early stages of ovarian cancer. Sutphen et al. published similar results with regard to lisophospholipids [37]. These authors found significant differences in serum phospholipid concentrations between patients with ovarian cancer and healthy controls.
Table 6 Diagnostic values of plasma sphingolipids in patients with advanced ovarian cancer.
C16-Cer C18:1-Cer C18-Cer
Threshold value (ng/100 μl)
Sensitivity
95%CI for sensitivity
Specificity
95%CI for specificity
AUC
95%CI for AUC
SE
P-value
311.88 4.74 100.76
0.77 0.9 0.8
0.5654–0.9431 0.5959–0.9845 0.4997–0.9472
0.56 0.437 0.57
0.4971–0.7762 0.3916–0.6922 0.4278–0.7795
0.759 0.768 0.771
0.51–0.8 0.53–0.81 0.53–0.8
0.071 0.07 0.06
0.0261 0.0160 0.0136
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
6
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
We obtained similar results and documented the usefulness of sphingolipids as biomarkers for monitoring of early and asymptomatic ovarian cancer (data not published). Clearly, further research is needed in order to assess the dynamics of sphingolipid concentrations for determining the scope of surgical resection, monitoring of chemotherapy outcomes and possible tumor recurrence. Moreover, studies assessing the usefulness of sphingolipids as biomarkers in assessing patients with a familial risk of breast and ovarian cancer who are carriers of the BRCA gene mutation are planned. In this publication, we show that 3 ceramides (C16-Cer, C18:1-Cer, C18-Cer) can be used as potential biomarkers of ovarian cancer and they can play an important role in this disease. It seems that this is one of the few publications dealing with the role of ceramides and S1P in tissue metabolism of ovarian cancer. In the literature, there is still a shortage of studies explaining the role of sphingolipids as transducers in cell signaling in the metabolism and development of ovarian cancer. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ygyno.2017.07.143. Conflict of interest None. Acknowledgements
[14]
[15]
[16]
[17] [18]
[19] [20] [21]
[22]
[23]
[24]
Medical University of Bialystok Grants supported this study: 14333601L; 143-29595L. [25]
Author contributions [26]
Conceived and designed the experiments: PK LB ABZ AC. Performed the experiments: PK LB ABZ AC. Analyzed the data: PK LB MS AC. Contributed reagents/materials/analysis tools: PK LB ABZ AC. References [1] P.M. Webb, S.J. Jordan, Epidemiology of epithelial ovarian cancer, Best Pract Res Clin Obstet Gynaecol (2016). [2] T.S. Latha, K. Panati, D.S. Gowd, M.C. Reddy, D. Lomada, Ovarian cancer biology and immunotherapy, Int. Rev. Immunol. 33 (2014) 428–440. [3] I. Romero, R.C. Bast Jr., Minireview: human ovarian cancer: biology, current management, and paths to personalizing therapy, Endocrinology 153 (2012) 1593–1602. [4] P. Knapp, M. Baranowski, M. Knapp, P. Zabielski, A. Blachnio-Zabielska, J. Gorski, Altered sphingolipid metabolism in human endometrila cancer, Prostaglandins and Onther Lipid Mediators 92 (2010) 62–66. [5] T.S. Tirodkar, C. Voelkel-Johnson, Sphingolipids in apoptosis, Exp. Oncol. 34 (2012) 231–242. [6] B. Carroll, J.C. Donaldson, L. Obeid, Sphingolipids in the DNA damage response, Adv Biol Regul 58 (2015) 38–52. [7] K. Kitatani, M. Taniguchi, T. Okazaki, Role of sphingolipids and metabolizing enzymes in hematological malignancies, Mol. Cell 38 (2015) 482–495. [8] T. Levade, N. Andrieu-Abadie, O. Micheau, P. Legembre, B. Segui, Sphingolipids modulate the epithelial-mesenchymal transition in cancer, Cell Death Dis. 1 (2015) 15001. [9] C.S. Bonder, S.M. Pitson, Sphingolipids–who's controlling who in disease? Immunol. Cell Biol. 93 (2015) 767–768. [10] R. Ghidoni, A. Caretti, P. Signorelli, Role of sphingolipids in the pathobiology of lung inflammation, Mediat. Inflamm. 2015 (2015) 487508. [11] K. Marycz, A. Smieszek, M. Jelen, K. Chrzastek, J. Grzesiak, J. Meissner, The effect of the bioactive sphingolipids S1P and C1P on multipotent stromal cells—new opportunities in regenerative medicine, Cell. Mol. Biol. Lett. 20 (2015) 510–533. [12] M. Garcia-Barros, N. Coant, J.P. Truman, A.J. Snider, Y.A. Hannun, Sphingolipids in colon cancer, Biochim. Biophys. Acta 1841 (2014) 773–782. [13] A.U. Blachnio-Zabielska, X.M. Persson, C. Koutsari, P. Zabielski, M.D. Jensen, A liquid chromatography/tandem mass spectrometry method for measuring the in vivo
[27]
[28] [29]
[30]
[31]
[32] [33]
[34] [35] [36]
[37]
incorporation of plasma free fatty acids into intramyocellular ceramides in humans, Rapid Commun. Mass Spectrom. 26 (2012) 1134–1140. M. Knapp, M. Zendzian-Piotrowska, A. Blachnio-Zabielska, P. Zabielski, K. Kurek, J. Gorski, Myocardial infarction differentially alters sphingolipid levels in plasma, erythrocytes and platelets of the rat, Basic Res. Cardiol. 107 (2012) 294. M. Baranowski, A. Blachnio-Zabielska, M. Charmas, J.W. Helge, F. Dela, M. Ksiazek, et al., Exercise increases sphingoid base-1-phosphate levels in human blood and skeletal muscle in a time- and intensity-dependent manner, Eur. J. Appl. Physiol. (2014). M. Nagahashi, J. Tsuchida, K. Moro, M. Hasegawa, K. Tatsuda, I.A. Woelfel, K. Takabe, T. Wakai, High levels of sphingolipids in human breast cancer, J. Surg. Res. 204 (2016) 435–444. O. Sukocheva, C. Wadham, Role of sphingolipids in oestrogen signalling in breast cancer cells: an update, J. Endocrinol. 220 (2014) R25–35. V. del Solar, D.Y. Lizardo, N. Li, J.J. Hurst, C.J. Brais, G.E. Atilla-Gokcumen, Differential regulation of specific sphingolipids in colon cancer cells during staurosporine-induced apoptosis, Chem. Biol. 22 (2015) 1662–1670. S. Sonnino, A. Prinetti, The role of sphingolipids in neuronal plasticity of the brain, J. Neurochem. 137 (2016) 485–488. R. Kolesnick, Ceramide: a novel second messenger, Trends Cell Biol. 2 (1992) 232–236. T. Goldkorn, S. Filosto, S. Chung, Lung injury and lung cancer caused by cigarette smoke-induced oxidative stress: molecular mechanisms and therapeutic opportunities involving the ceramide-generating machinery and epidermal growth factor receptor, Antioxid. Redox Signal. 21 (2014) 2149–2174. Y.Y. Ma, X.Z. Mou, Y.H. Ding, H. Zou, D.S. Huang, Delivery systems of ceramide in targeted cancer therapy: ceramide alone or in combination with other anti-tumor agents, Expert Opin. Drug Deliv. 13 (2016) 1397–1406. I.G. Rivera, M. Ordonez, N. Presa, P. Gangoiti, A. Gomez-Larrauri, M. Trueba, T. Fox, M. Kester, A. Gomez-Munoz, Ceramide 1-phosphate regulates cell migration and invasion of human pancreatic cancer cells, Biochem. Pharmacol. 102 (2016) 107–119. K. Tanaka, K. Tamiya-Koizumi, M. Yamada, T. Murate, R. Kannagi, M. Kyogashima, Individual profiles of free ceramide species and the constituent ceramide species of sphingomyelin and neutral glycosphingolipid and their alteration according to the sequential changes of environmental oxygen content in human colorectal cancer Caco-2 cells, Glycoconj. J. 31 (2014) 209–219. S.J.H. Park, Acid ceramidase and human disease, Biochim. Biophys. Acta 1758 (2006) 2133–2138. R. Kolesnick, The therapeutic potential of modulating the ceramide/sphingomyelin pathway, J. Clin. Invest. 110 (2002) 3–8. A.J. Alberg, K. Armeson, J.S. Pierce, J. Bielawski, A. Bielawska, K. Visvanathan, E.G. Hill, B. Ogretmen, Plasma sphingolipids and lung cancer: a population-based, nested case-control study, Cancer Epidemiol. Biomark. Prev. 22 (2013) 1374–1382. H. Furuya, Y. Shimizu, T. Kawamori, Sphingolipids in cancer, Cancer Metastasis Rev. 30 (2011) 567–576. P. Giussani, C. Tringali, L. Riboni, P. Viani, B. Venerando, Sphingolipids: key regulators of apoptosis and pivotal players in cancer drug resistance, Int. J. Mol. Sci. 15 (2014) 4356–4392. J. Zou, X. Ma, G. Zhang, L. Shen, L. Zhou, Y. Yu, F. Zhu, Z. Chen, Evaluation of the change in sphingolipids in the human multiple myeloma cell line U266 and gastric cancer cell line MGC-803 treated with arsenic trioxide, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1004 (2015) 98–107. S. Li, Y. Zhou, X. Zheng, X. Wu, Y. Liang, S. Wang, Y. Zhang, Sphk1 promotes breast epithelial cell proliferation via NF-kappaB-p65-mediated cyclin D1 expression, Oncotarget (2016). J. Long, Y. Xie, J. Yin, W. Lu, S. Fang, SphK1 promotes tumor cell migration and invasion in colorectal cancer, Tumour Biol. 37 (2016) 6831–6836. S. Tsukamoto, M. Kumazoe, Y. Huang, C. Lesnick, N.E. Kay, T.D. Shanafelt, H. Tachibana, SphK1 inhibitor potentiates the anti-cancer effect of EGCG on leukaemia cells, Br. J. Haematol. (2016). R. Kolesnick, D. Altieri, Z. Fuks, A CERTain role for ceramide in taxane-induced cell death, Cancer Cell 11 (2007) 473–475. S. Chiantia, E. London, Sphingolipids and membrane domains: recent advances, Handb. Exp. Pharmacol. (2013) 33–55. L.R. Cordeiro Pedrosa, O. van Tellingen, T. Soullie, A.L. Seynhaeve, A.M. Eggermont, T.L. Ten Hagen, M. Verheij, G.A. Koning, Plasma membrane targeting by short chain sphingolipids inserted in liposomes improves anti-tumor activity of mitoxantrone in an orthotopic breast carcinoma xenograft model, Eur. J. Pharm. Biopharm. 94 (2015) 207–219. R. Sutphen, Y. Xu, G.D. Wilbanks, J. Fiorica, E.C. Grendys Jr., J.P. LaPolla, H. Arango, M.S. Hoffman, M. Martino, K. Wakeley, D. Griffin, R.W. Blanco, A.B. Cantor, Y.J. Xiao, J.P. Krischer, Lysophospholipids are potential biomarkers of ovarian cancer, Cancer Epidemiol. Biomark. Prev. 13 (2004) 1185–1191.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer Paweł Knapp a, Lubomir Bodnar b, Agnieszka Błachnio-Zabielska c,d, Magdalena Świderska d,⁎, Adrian Chabowski d a
Department of Gynecology and Gynecologic Oncology, Medical University of Bialystok, 24a Sklodowskiej-Curie Str., 15-276 Bialystok, Poland Department of Clinical Oncology, Military Institute of Medicine in Warsaw, 128 Szaserow Str., 04-141 Warsaw, Poland Department of Hygiene, Epidemiology and Metabolic Disorders, Medical University of Bialystok, 2c Mickiewicza Str., 15-222 Bialystok, Poland d Department of Physiology, Medical University of Bialystok, 2c Mickiewicza Str., 15-222 Bialystok, Poland b c
H I G H L I G H T S • • • •
Evaluation of selected sphingolipids in patients with advanced ovarian cancer Plasma C16-Cer,C18:1-Cer,C18-Cer concentration elevated in study group vs. control Increase concentration of 5 ceramide, S1P in ovarian tissue vs. control group Some sphingolipids can be used as potential biomarkers of ovarian cancer
a r t i c l e
i n f o
Article history: Received 11 May 2017 Received in revised form 27 July 2017 Accepted 30 July 2017 Available online xxxx Keywords: Ovarian cancer Carcinogenesis Sphingolipids Ceramide S1P Ovarian cancer biomarkers
a b s t r a c t Purpose. The role of lipids in carcinogenesis through induction of abnormal cell lines in the human body is currently undisputable. Based on the literature, bioactive sphingolipids play an essential role in the development and progression of cancer and are involved in the metastatic process. The aim of this study was to determine the concentration of selected sphingolipids in patients with advanced ovarian cancer (AOC, FIGO III/IV, high grade ovarian cancer). Methods. Seventy-four patients with ovarian cancer were enrolled. Plasma concentrations of C16-Cer, C18:1Cer and C18-Cer were assessed by LC/MS/MS. The content of tissue sphingolipids was measured using a UHPLC/ MS/MS. Results. Plasma concentration of 3 ceramides: C16-Cer, C18:1-Cer and C18-Cer was significantly elevated in women with advanced ovarian cancer compared to control group (P = 0.031; 0.022; 0.020; respectively). There were increases in concentration of 5 ceramides: C16-Cer, C18:1-Cer, C18-Cer, C24:1-Cer, C24-Cer (P = 0.025; 0.049; 0.032; 0.005; 0.013, respectively) and S1P (P = 0.004) in ovarian tissue of women with advanced ovarian cancer compared to healthy individuals. Importantly, significantly higher risk of ovarian cancer when the plasma concentration of C16-Cer N 311.88 ng/100 μl (AUC: 0.76, P = 0.0261); C18:1-Cer N 4.75 ng/100 μl (AUC: 0.77, P = 0.0160) and C18-Cer N 100.76 ng/100 μl (AUC:0.77, P = 0.0136) was noticed. Conclusions. Bioactive sphingolipids play an essential role in the development and progression of cancer and they also take part in the process of metastasizing. This study suggests that some sphingolipids can be used as potential biomarkers of advanced ovarian cancer and that they can play an important role in the pathogenesis of this disease. © 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction ⁎ Corresponding author at: Department of Physiology, Medical University of Bialystok, Poland, 2c Mickiewicza Str., 15-222 Bialystok, Poland. E-mail addresses: [email protected] (P. Knapp), [email protected] (L. Bodnar), [email protected] (A. Błachnio-Zabielska), [email protected] (M. Świderska), [email protected] (A. Chabowski).
Ovarian cancer is the fifth most common cancer in European women. Around the world, more than 200,000 women are estimated to develop ovarian cancer every year, and about 100,000 die of the disease [1].
http://dx.doi.org/10.1016/j.ygyno.2017.07.143 0090-8258/© 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
2
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Aggressive biology of ovarian cancer is strictly correlated with its histology. It is worth underlining that heterogeneity of observed ovarian cancer subtypes determining as far there are no screening tests for early diagnosis of this type of neoplasm and ovarian cancer is still routinely diagnosed in the advanced stages of that disease (manly FIGO stage III, IV) [2,3]. The role of lipids in carcinogenesis through induction of abnormal cell lines in the human body is currently undisputable. Based on the literature, bioactive sphingolipids play an essential role in the development and progression of cancer and are involved in the metastatic process [4,5]. Ceramides are some of the most important elements in the metabolism of sphingolipids. They can be either synthesized de novo from serine and palmitoyl-CoA in a reaction catalyzed by serine palmitoyltransferase (SPT) or produced through enzymatic hydrolysis of sphingomyelin by sphingomyelinase. Apart from ceramides, biologically active sphingolipids include: dihydroceramide, sphingosine-1phsophate (S1P), sphingosine and sphinganine. Based on recent research, ceramides have pro-apoptotic properties and constitute an important antineoplastic factor (lipid tumor suppressors). S1P acts antagonistically to ceramides, as it induces cell transformation, cancer cell proliferation, cell survival and blocks the anti-apoptotic mechanisms of ceramides. Moreover, an increase in the expression of sphingosine kinase (SPHK1), an enzyme that transforms sphingosine to S1P, causes accumulation of S1P in the cell, promoting carcinogenesis and tumor formation [6–8]. Despite the important role of sphingolipids in cancer biology, their metabolism in different malignant tumors is poorly investigated. Some studies showed marked differences in ceramide content between tumor and respective healthy tissue [9–11]. Interestingly, the level of this sphingolipid was either decreased or increased, suggesting that alterations in ceramide metabolism in cancer tissue may depend on a type of tumor [6,7,10,12]. However, these studies focused only on selected sphingolipid species, and to date there are no complex data on ceramide metabolism in cancer tissue. In addition, sphingolipid metabolism in ovarian cancer has been still poorly investigated. Therefore, measurement of the sphingolipids could lead to better understanding of the influence of ovarian cancer biology and possibly provide new biomarker(s) for non-invasive testing. 2. Material and methods The investigation conforms with the principles outlined in the Declaration of Helsinki and was approved by the Ethical Committee for Human Studies of the Medical University of Bialystok, Poland (recruitment between January 2015 till June 2016). All patients gave their informed consent prior to their inclusion in the study. The initial study included two groups of women: patients with suspicious of advanced ovarian cancer (AOC) (n = 165) and women with normal ovarian morphology confirmed by transvaginal sonography evaluation (control group, n = 149). The patients suspected to have an advanced ovarian cancer were examined in the Department of Gynecology and Gynecologic Oncology, Medical University of Bialystok and pre-treatment diagnosis were evaluated based on present adnexal masses with spread metastasis into pelvis and abdomen confirmed by CT scan, ascites, elevated level of Ca-125. 10 ml of peripheral blood (antecubital vein) was collected for EDTA probes from each patient 24 h before surgery procedures. The blood was centrifuged in 15 min, plasma subsequently separated and frozen at −80 °C temperature. Radical bulky surgery procedures according to gynecologic oncology guidelines (hysterectomy, bisalpingo-oophorectomy, total omentectomy, appendectomy, others) were performed in the group of patients clinically suspected of ovarian cancer. All applied procedures were standardized and repeatable. Neoplastic samples were collected by University Tissue Biobanking Team (one pathologist, one technician; according to Standard Operating Procedures for Biobanking - SOP) from an ovarian surface not deeper than 1 cm, just after removal of the uterus and ovaries. Dissected
tissues were immediately placed into liquid nitrogen and time from surgical closing last artery vasculating ovarian tumor to tissue collecting was not longer than 2 min. Normal ovarian tissues were obtained during operations because of fibroids or other non-oncology procedures. Hysterectomy was performed and ovarian samples were collected and handled as described above. Only women with histologically diagnosed advanced ovarian cancer serous type, the International Federation of Gynecology and Obstetrics (FIGO) Classification - stage III/IV, high-grade (2014 WHO Classification) were included in the study. We sequentially excluded from the group of 165 women: 48 borderline tumors, 43 of non-epithelial histology; and from the control group: 68 patients with some abnormalities in ovarian tissue diagnosed during histology evaluation. Final groups counted: advanced ovarian cancer: n = 74 patients, control group: n = 81 women. 2.1. The content of plasma sphingolipids The level of sphingolipids in plasma was analyzed by LC/MS/MS approach as previously described by Blachnio-Zabielska et al. with minor modification [13]. Briefly, to each plasma sample (100 μl) were added 50 μl of the internal standard solution (17C–sphingosine and 17C-S1P, and C17-Cer Avanti polar lipids) as well as 1.5 ml of an extraction mixture (isopropanol: water: ethyl acetate, 35:5:60; v:v:v). The following sphingolipids were quantified: SPH (sphingosine), S1P (sphingosine1-phosphate), SPA (sphinganine), ceramide C14:0-Cer (ceramides containing myristic acid), C16:0-Cer (ceramides containing palmitic acid), C18:1-Cer (ceramides containing oleic acid), C18:0-Cer (ceramides containing stearic acid), C20:0-Cer (ceramides containing arachidic acid), C22:0-Cer (ceramide containing behenic acid), C24:1-Cer (ceramides containing nervonic acid) and C24:0-Cer (ceramides containing lignoceric acid). The mixture was vortexed, sonicated and then centrifuged for 10 min at 4000 rpm (Sorvall Legend RT). The supernatant was transferred to a new tube and pellet was re-extracted. After centrifugation supernatants were combined and evaporated under nitrogen. The dried sample was reconstituted in 100 μl of LC Solvent A (2 mM ammonium formate, 0.15% formic acid in methanol) for LC/MS/MS analysis. Quantitative measurement was made using triple quadrupole mass spectrometer (Agilent 6460) in positive mode using multiple reaction monitoring (MRM). The chromatographic separation was performed on Agilent 1290 Infinity Ultra High Performance Liquid Chromatography (UHPLC). The analytical column was a reverse-phase Zorbax SB-C8 column 2.1 × 150 mm, 1.8 μm. Chromatographic separation was conducted in binary gradient using 2 mM ammonium formate, 0.15% formic acid in methanol as Solvent A and 1.5 mM ammonium formate, 0.1% formic acid in water as Solvent B at the flow rate of 0.4 ml/min. All sphingolipids were quantified against standard concentration curve. The flow was diverted to waste for the first and the last 4 min to prevent eluting impurities from entering the mass spectrometer. 2.2. The content of tissue sphingolipids The content of sphingolipids was measured using a UHPLC/MS/MS (an ultra-high performance liquid chromatography coupled with triple quadrupole mass spectrometry) approach according to BlachnioZabielska et al. [13]. Briefly, the tissue samples (20 mg) were homogenized in a solution composed of 0.25 M sucrose, 25 mM KCl, 50 mM Tris and 0.5 mM EDTA, pH 7.4. Immediately afterwards, 50 μl of the internal standard solution (17C–sphingosine and 17C–S1P, and C17-Cer Avanti polar lipids) as well as 1.5 ml of an extraction mixture (isopropanol: water: ethyl acetate, 35:5:60; v:v:v) were added to each homogenate. The mixture was vortexed, sonicated and then centrifuged for 10 min at 4000 rpm (Sorvall Legend RT). The supernatant was transferred to a new tube and pellet was re-extracted. After centrifugation supernatants were combined and evaporated under nitrogen. The dried sample was reconstituted in 100 μl of LC Solvent A (2 mM
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
ammonium formate, 0.15% formic acid in methanol) for LC/MS/MS analysis. Sphingolipids were analyzed by means of an Agilent 6460 triple quadrupole mass spectrometer using positive ion electrospray ionization (ESI) source with multiple reaction monitoring (MRM) against standard curves constructed for each analyzed compound. The chromatographic separation was performed using an Agilent 1290 infinity ultra-performance liquid chromatography (UHPLC). The analytical column was a reverse-phase Zorbax SB-C8 column 2.1 × 150 mm, 1.8 μm. Chromatographic separation was conducted in binary gradient using 2 mM ammonium formate, 0.15% formic acid in methanol as solvent A and 1.5 mM ammonium formate, 0.1% formic acid in water as solvent B at the flow rate of 0.4 ml/min. 2.3. Statistical analysis The method was validated for the use of sphingolipid estimation in biological tissue and was also used for sphingolipids measurement in human plasma [14,15]. Inter and intra assay reproducibility of the method is around 1% for most of the sphingolipid species. The list of quantified sphingolipids with the limit of detection is provided in Table 1 (Supplementary Material). Descriptive statistics including mean concentration and standard error of the mean concentration were calculated for selected sphingolipids, henceforth called features. In order to detect statistically significant differences between examined groups, either fitting an analysis of variance model was conducted or non-parametric method (Wilcoxon rank-sum test) was applied. The choice of an appropriate method was made upon fulfilling the normality and the homogeneity of variances assumptions and in case of violation of at least one condition non-parametric approach was employed. The normality of features distribution was checked with the Shapiro-Wilk test and the homogeneity of variances with the Levene's test. Features that have been found significant, i.e. their distribution was statistically significantly different among experimental groups, were taken under further investigation to discover their prediction capability. To accomplish that, for each significant feature: ROC curves (receiver operating characteristic curve) were constructed and optimal threshold values were determined with the Youden method. Confidence intervals for sensitivity and specificity corresponding to a particular threshold were calculated with the use of the Wilson method and a test verifying the area under curve (AUC), that was significantly N 0.5 (random classification) was performed with the DeLong method - P-values and one-sided confidence intervals for AUC are reported. Calculations concerning ROC curves and corresponding tests were conducted with the functions provided by the pROC R package. Confidence intervals for sensitivity and specificity were constructed with the use of the binom.confint function, part of the binom R package. All calculations were carried out in R software environment. Finally, Spearman rank test was used with respect to correlate each examined sphindolipids derivativeness. Significance level alpha equal to 0.05 was applied for all statistical tests. 3. Results Patients' characteristic and clinicopathologic data are presented in Table 1 and 2. Patients from both groups were matched for several factors (age, body mass index, standard analyses, Ca-125, etc.) to make sure that two groups are comparable and there are no statistically significant differences between them. The values of mean concentration and 25% - 75% percentile of plasma sphingolipids and ovarian tissue sphingolipids in each study group are presented respectively in Table 3 and 4. There were no significant differences in mean values for any ceramide species between patients who were premenopausal versus postmenopausal. The levels did not correlate with tumor size (data not shown). We showed significant increase in concentration of 3 ceramides: C16-Cer and C18:1-Cer, C18-Cer (P = 0.031; 0.022; 0.020; respectively)
3
Table 1 Characteristic of the patients with serous advanced ovarian cancer (AOC) and control group included into the study. Parameters
Patients with AOC (n = 74)
Control group (n = 81)
P value
Age (median ± SD) BMI (median ± SD) Systolic BP (median ± SD) Diastolic BP (median ± SD) RBC (median ± SD) HGB (median ± SD) PLT (median ± SD) WBC (median ± SD) Ca125 (median ± SD) Glucose (median ± SD) Triglycerides (median ± SD) Total cholesterol (median ± SD) Total proteins (median ± SD)
54.2 ± 3.7 27.29 ± 4.24 123.47 ± 13.97 79.43 ± 10.93 4.07 ± 0.84 11.89 ± 1.80 580.53 ± 146.63 9.15 ± 2.48 822.40 ± 572.25 83.80 ± 11.16 1.53 ± 0.02 174.63 ± 20.12 5.96 ± 0.78
57.9 ± 1.9 27.26 ± 4.15 123.34 ± 14.13 80.13 ± 10.79 3.79 ± 0.75 12.72 ± 1.55 289.50 ± 96.18 7.81 ± 1.88 17.81 ± 8.08 82.47 ± 17.63 1.58 ± 0.3 173.56 ± 27.49 6.98 ± 0.68
0.005⁎ 0.893 0.833 0.854 0.157 0.056 0.0001⁎ 0.047⁎ 0.0001⁎ 0.481 0.513 0.735 0.0001⁎
SD – standard deviation. ⁎ P-value b 0.005.
in plasma of women with advanced ovarian cancer compared to control group (Table 3). Furthermore, we showed increase in concentration of 5 ceramides: C16-Cer, C18:1-Cer, C18-Cer, C24:1-Cer, C24-Cer (P = 0.025; 0.049; 0.032; 0.005; 0.013, respectively) and S1P (P = 0.004) in ovarian tissue of women with serous advanced ovarian cancer compared to control group. We also observed statistically higher level of SPH in healthy ovarian tissue then in ovarian cancer (P = 0.005) (Table 4). There was no correlation between plasma vs. ovarian cancer tissue sphingolipids in individual patients. Plasma C16, C18.1, C18 correlated with disease progression of ovarian cancer: FIGO staging (R = 0.26, P = 0.041, R = 0.26, P = 0.044, R = 0.31, P = 0.015 respectively), and with differentiate the grading (R = 0.26, P = 0.040, R = 0.28, P = 0.030, R = 0.29, P = 0.021 respectively) (Table 4). Moreover, plasma Sph, SPA and S1P concentration has shown a significant association with leukocyte (WBC) levels (R = −0.23, P = 0.001, R = −0.30, P = 0.018, R = −0.26, P = 0.039) (Table 5). We included all statistically significant sphingolipids in later ROC analyses but we created ROC curves only for sphingolipids significantly changed in plasma (which has potential for noninvasive diagnosis),
Table 2 Clinicopathologic characteristic of the patients with high-grade serous advanced ovarian cancer. Parameters
N
%
FIGO stage IIIA IIIB IIIC IVA IVB
2 18 49 1 4
2.7 24.3 66.2 1.4 5.4
Lymph nodes N0 N1 Not determined
2 71 1
2.7 95.9 1.4
Volume of ascites No ascites Ascites ≤ 500 ml Ascites N 500 ml
12 28 34
16.2 37.8 46.0
ECOG PS 0 1–2
69 5
93.2 6.8
Tumor residuals 0 1–10 mm N10 mm
50 19 5
67.5 25.7 6.8
ECOG PS – Eastern Cooperative Oncology Group performance status.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
4
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 3 Concentration of sphingolipids in plasma: ovarian cancer patients and healthy women (control group). Ovarian cancer patient (n = 74) Sphingolipids plasma concentration (ng/100 μl) mean (25–75% percentile) Total Cer 4236.3996 (2763.467–6872.3401) SPH 77.8381 (21.3466–132.4366) SPA 27.5618 (14.7352–42.4981) S1P 424.6286 (297.358–579.3469) C14-Cer 7.7032 (2.2389–12.1287) C16-Cer 403.6705 (224.8791–648.3847) C18:1-Cer 7.3511 (2.912–13.7634) C18-Cer 161.7371 (51.4581–270.3815) C20-Cer 117.1991 (46.4581–162.3947) C22-Cer 557.7762 (212.2437–795.3791) C24:1-Cer 1332.8207 (797.2736–1726.3849) C24-Cer 1648.1414 (1178.3746–2017.3884)
Healthy women (n = 81)
P-value
3772.7214 (1298.3961–6002.3411) 96.7350 (37.4936–167.4422) 26.8689 (13.4483–42.7346) 420.3498 (271.2312–690.3722) 8.0935 (3.3388–12.8849) 261.5972 (142.3771–559.9936) 3.8942 (0.8457–7.459) 92.4272 (33.4888–141.2882) 104.8481 (59.2777–177.6624) 514.9734 (201.2881–829.2836) 1174.8300 (523.3888–1579.2736) 1470.1474 (603.348–1992.3481)
0.231 0.102 0.966 0.8 0.955 0.031⁎ 0.022⁎ 0.020⁎ 0.057 0.464 0.076 0.921
⁎ Statistically significant value of b0.05 for Mann-Whitney test.
which set the threshold values and allowed predicting the likelihood of ovarian cancer with specific sensitivity and specificity. The area under the ROC curve for C16-Cer was 0.759, for C18:1-Cer was 0.768 and for C18-Cer was 0.771. All field values are satisfactory and indicate the usefulness of these biochemical markers as tools to predict the risk of ovarian cancer. We demonstrated a significantly higher risk of ovarian cancer when the plasma concentration of C16-Cer N 311.88 ng/100 μl (sensitivity: 0.77; specificity: 0.56, P = 0.0261); C18:1-Cer N 4.75 ng/100 μl (sensitivity: 0.9; specificity: 0.437, P = 0.0160) and C18-Cer N 100.76 ng/100 μl (sensitivity: 0.8; specificity: 0.57, P = 0.0136) (Supplementary Material Fig. 1, 2, 3). Diagnostic values of these sphingolipids in plasma are presented respectively in Table 6.
4. Discussion Ovarian cancer is a disease associated with a high mortality [1]. This study shows that sphingolipid metabolism in the human ovarian carcinoma is augmented as compared to the healthy tissue. It is supported by the fact that the content of measured sphingolipids and ceramide species were increased in the ovarian cancer tissue. Accumulation of sphingolipids has been noted in various cancers such as tumors of the head and neck, lung cancer, sarcomas, breast cancer and endometrial cancer. There are studies showing that certain tumors, including colon cancer and brain cancer, are characterized by decreased concentration of ceramides compared to healthy tissues [18,19]. Such data clearly indicate that sphingolipid metabolism is closely associated with cell transformation depending on histological type of the tissue. Interestingly, in ovarian cancer ceramide levels are elevated as compared to healthy ovarian tissue. However, we have to point out that various agonists increase the level of endogenous, intracellular long-chain ceramides; for
example, rise in temperature is associated with enhanced cancer cell metabolism. Elevated concentration of ceramides and enzymes catalyzing the transformation to bioactive derivatives and accumulation of sphinganine and dihydroceramide coupled to activation of palmitoyltransferase suggests that ceramides are synthesized de novo [20]. It seems that, as in ovarian cancer, the synthesis of ceramides in endometrial cancer is performed de novo [4]. This is justified, as substantial increase in the concentrations of sphinganine, dihydroceramide and ceramide expression have been observed in other cancer types [21–24]. High concentration of total ceramide in ovarian cancer is puzzling in the context of its antiproliferative and pro-apoptotic functions. It is probably related to overactivation of ceramidase that produces ceramides. Park et al. described such a mechanism, demonstrating increased activation of acidic ceramidase in various cancer patients [25]. This reasoning is supported by our own research showing that the concentration of sphingosine is 5.5 times lower in ovarian cancer in comparison to healthy ovarian tissue. Possibly, compared to healthy ovarian tissue, the activation of ceramidase in ovarian cancer is significantly higher and leads to increased transformation of sphingosine to ceramide (115.69 vs. 654.10 ng/ml). We made similar observations in endometrial cancer, where we also noted high levels of ceramide and S1P [4]. Total ceramide concentration also clearly depends on the histological type of cancer and its clinical and the degree of histological progression (staging, grading). The observed high concentration of total ceramide in ovarian cancer (1.4 times higher in comparison to healthy ovarian tissue) was probably related to the advancement of cancer in studied patients (FIGO stage III/IV, G2/G3). The role of ceramides in modulating cell signaling is primarily related to their pro-apoptotic role and inhibition of tumor growth [26]. In this context, high concentrations of ceramides in ovarian cancer are not in line with the expectations; however, high levels of ceramides are probably associated with
Table 4 Concentration of sphingolipids in ovarian tissue: ovarian cancer patients and healthy women (control group). Ovarian cancer patient (n = 74) Ovarian tissue sphingolipids concentration (ng/ml) Mean (25–75% percentile) Total Cer 5299.1784 (1287.2736–8091.2736) SPH 115.6994 (52.3761–171.2871) SPA 31.5615 (11.2376–59.3481) S1P 1253.3658 (629.2761–1762.2399) C14-Cer 22.8915 (12.2761–37.2761) C16-Cer 1053.2861 (699.26661–1751.2881) C18:1-Cer 100.1249 (54.28871–161.2917) C18-Cer 273.0186 (179.2779–530.3801) C20-Cer 125.3704 (66.6892–209.0972) C22-Cer 225.8186 (142.3721–376.3821) C24:1-Cer 1885.1075 (1062.38271–3119.5622) C24-Cer 1613.5925 (777.2817–3001.2998)
Healthy women (n = 81)
P-value
4343.4923 (2376.8737–6981.2837) 654.1071 (201.2837–1082.8521) 30.1124 (12.34–78.2381) 811.9984 (267.3338–1491.2716) 15.2976 (8.8726–31.27771) 910.877 (509.2881–1487.2719) 20.9839 (9.2781–33.3433) 166.2519 (87.6509–290.6661) 91.7388 (55.8701–167.2771) 333.9555 (164.2871–612.3672) 1696.1389 (762.2281–3001.2971) 1108.2964 (661.2881–1982.3888)
0.061 0.215 0.894 0.004⁎ 0.626 0.025⁎ 0.049⁎ 0.032⁎ 0.859 0.063 0.005⁎ 0.013⁎
⁎ Statistically significant value of b0.05 for Mann-Whitney test.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
5
Table 5 Spearman rank correlation for analyzed sphingolipids in ovarian cancer patients' plasma.
Sph_plasma SPA_plasma S1P_plasma C14_plasma C16_plasma C18:1_plasma C18_plasma C20_plasma C22_plasma C24:1_plasma C24_plasma Total Cer_plasma
r p r p r p r p r p r p r p r p r p r p r p r p
FIGO
Grading
BMI
RBC
HGB
PLT
WBC
Ca125
Glucose
Triglycerides
Total cholesterol
Total proteins
−0,22 0,087 −0,05 0,697 0,06 0,658 −0,01 0,959 0,26 0,041 0,26 0,044 0,31 0,015 0,24 0,063 0,09 0,472 0,21 0,101 0,02 0,899 0,14 0,284
−0,23 0,074 0,00 0,993 0,01 0,947 0,00 0,970 0,26 0,040 0,28 0,030 0,29 0,021 0,22 0,084 0,08 0,538 0,20 0,123 0,01 0,917 0,14 0,279
−0,08 0,537 0,00 0,988 0,13 0,322 0,13 0,317 −0,07 0,602 −0,14 0,284 0,02 0,902 −002 0,883 0,03 0,799 −0,07 0,594 −0,01 0,934 −0,04 0,759
−0,16 0,217 0,05 0,710 −0,01 0,932 0,07 0,563 0,07 0,569 −0,03 0,802 0,08 0,557 0,11 0,401 0,12 0,369 0,08 0,556 −0,04 0,760 0,08 0,517
0,14 0,292 0,10 0,442 −0,19 0,141 −0,10 0,435 −0,29 0,021 −0,07 0,563 −0,23 0,068 −025 0,054 −0,20 0,118 −0,26 0,044 −0,13 0,331 −0,21 0,109
−0,17 0,182 −0,08 0,540 0,07 0,592 0,05 0,681 0,32 0,013 0,25 0,054 0,33 0,009 0,29 0,025 0,16 0,228 0,29 0,022 0,03 0,820 0,21 0,103
−0,42 0,001 −0,30 0,018 −0,26 0,039 −0,11 0,416 −0,04 0,764 −003 0,790 −0,06 0,622 −016 0,200 −0,23 0,072 −0,13 0,297 −0,18 0,169 −0,19 0,140
−0,23 0,068 −0,16 0,203 0,17 0,200 −0,09 0,503 0,25 0,053 0,28 0,026 0,30 0,018 0,22 0,086 0,02 0,859 0,16 0,203 −0,08 0,516 0,07 0,608
0,04 0,766 0,07 0,603 0,23 0,074 0,01 0,954 −0,21 0,107 −0,18 0,153 −0,12 0,350 −015 0,253 −0,16 0,209 −0,20 0,122 −0,07 0,602 −0,16 0,213
0,00 0,998 −0,03 0,805 0,00 0,988 −0,14 0,291 −0,13 0,306 −0,16 0,221 0,01 0,916 −008 0,532 −0,19 0,130 −0,10 0,450 −0,31 0,016 −0,20 0,127
−0,28 0,025 −0,19 0,143 −0,04 0,753 0,08 0,517 0,01 0,958 0,04 0,746 0,01 0,956 −002 0,847 −0,11 0,389 0,00 0,985 −0,10 0,429 −0,11 0,408
0,06 0,666 −0,08 0,544 −0,08 0,513 −0,02 0,893 −0,13 0,327 −0,09 0,509 −014 0,285 −023 0,067 −0,18 0,162 −0,25 0,051 −0,13 0,308 −0,21 0,094
r – Spearman rank (−1,+1); p - statistically significant P-value of b0.05. Significance of bold p b 0.05.
the risk of epithelial ovarian cancer. This could be explained by the aggressive biological nature of ovarian cancer whereby high levels of S1P produced by the tumor promote cell proliferation and cell transformation as well as survival of abnormal cells, neoangiogenesis and inhibition of pro-apoptotic processes [2,20]. Moreover, dysregulation of sphingolipid metabolism could also promote development of ovarian cancer. Ovarian cancer cells can produce an excessive amount of ceramides, resulting in increased conversion to S1P. According to such an assumption, increased risk of ovarian cancer could be caused by shifting of balance towards S1P through dysregulation of sphingolipid metabolism, resulting in overproduction of S1P and ceramides. Elevated concentrations of 11 out of 12 studied sphingolipid derivatives that we observed in this study in ovarian cancer patients compared to control subjects, clearly indicate disturbed S1P metabolism due to increase in the total bioactive sphingolipids. Alberg et al. came to similar conclusions, reporting that increase in the concentration of bioactive sphingolipids in lung cancer can also potentially be due to the dysregulation of S1P metabolism [27]. Numerous recent studies indicate that ceramides can play distinctive and opposing roles in neoplastic tumors possibly caused by differences in the lengths of lipid chains [17,28– 30]. Analysis of C18-Cer function demonstrated its role in tumor suppression, while C16-Cer was shown to induce cell proliferation. In advanced ovarian cancer production [4,10,16,17]of C16-Cer (1053.26 ng/ml) augments the aggressive nature of the tumor and its function is superior to C18-Cer and C18:1-Cer (5 and 10 times lower concentrations, respectively). This can be a result of the above-mentioned total dysregulation of S1P metabolism, which could be a dominant phenomenon seen in advanced ovarian cancer. Moreover, resistance to chemotherapy in ovarian cancer can also be caused by S1P dysregulation. It is underscored that S1P is not only a messenger in cell signaling, but it also binds to membrane receptors coupled to a
G protein (S1PRs), which is responsible for the majority of its biological actions [31–33]. It is possible that large amounts of S1P synthesized by the tumor stimulate ovarian cancer cell proliferation through binding to the above-mentioned receptor. Observed low serum level of S1P in ovarian cancer patients results from its consumption by biologically aggressive tumor cells that constantly proliferate. The above-mentioned resistance to chemotherapy is due not only to dysregulation of ceramide and S1P metabolism, but also to the presence of protein mediators and their signaling actions. According to Gatt et al., increased expression of ceramide transporting proteins (CERT) and their intracellular accumulation associated with S1P is a possible cause of ovarian cancer resistance to chemotherapy with platinum derivatives [34–36]. As the biology of ovarian cancer is very aggressive, the tumor is often diagnosed at a late stage (FIGO III/IV) characterized by a 5-year survival rate of 20–30%. Early detection of ovarian cancer using biomarkers could improve total survival of patients with this type of cancer. Interestingly, there is a positive statistically significant correlation between the cancer progression (FIGO and Grading) and the plasma C16, C18.1, C18 level (R = 26, P = 0.04; R = 0.26, P = 0.044; R = 0.31, P = 0.015 respectively). We can assume that ceramics can reflect the stages of ovarian cancer development. Besides that our research showed that patients with ovarian cancer are characterized by significantly higher serum levels of C16-Cer, C18:1-Cer and C18-Cer compared to healthy controls. Moreover, based on Youden analysis, we determined the optimal cut-off values for studied ceramides indicating significantly elevated risk of ovarian cancer. It was determined that C16-Cer, C18:1-Cer and C18-Cer meet the criteria for biomarkers (AUC = 0.759; =0.768; =0.771) and can be used in the diagnosis of patients at early stages of ovarian cancer. Sutphen et al. published similar results with regard to lisophospholipids [37]. These authors found significant differences in serum phospholipid concentrations between patients with ovarian cancer and healthy controls.
Table 6 Diagnostic values of plasma sphingolipids in patients with advanced ovarian cancer.
C16-Cer C18:1-Cer C18-Cer
Threshold value (ng/100 μl)
Sensitivity
95%CI for sensitivity
Specificity
95%CI for specificity
AUC
95%CI for AUC
SE
P-value
311.88 4.74 100.76
0.77 0.9 0.8
0.5654–0.9431 0.5959–0.9845 0.4997–0.9472
0.56 0.437 0.57
0.4971–0.7762 0.3916–0.6922 0.4278–0.7795
0.759 0.768 0.771
0.51–0.8 0.53–0.81 0.53–0.8
0.071 0.07 0.06
0.0261 0.0160 0.0136
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143
6
P. Knapp et al. / Gynecologic Oncology xxx (2017) xxx–xxx
We obtained similar results and documented the usefulness of sphingolipids as biomarkers for monitoring of early and asymptomatic ovarian cancer (data not published). Clearly, further research is needed in order to assess the dynamics of sphingolipid concentrations for determining the scope of surgical resection, monitoring of chemotherapy outcomes and possible tumor recurrence. Moreover, studies assessing the usefulness of sphingolipids as biomarkers in assessing patients with a familial risk of breast and ovarian cancer who are carriers of the BRCA gene mutation are planned. In this publication, we show that 3 ceramides (C16-Cer, C18:1-Cer, C18-Cer) can be used as potential biomarkers of ovarian cancer and they can play an important role in this disease. It seems that this is one of the few publications dealing with the role of ceramides and S1P in tissue metabolism of ovarian cancer. In the literature, there is still a shortage of studies explaining the role of sphingolipids as transducers in cell signaling in the metabolism and development of ovarian cancer. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ygyno.2017.07.143. Conflict of interest None. Acknowledgements
[14]
[15]
[16]
[17] [18]
[19] [20] [21]
[22]
[23]
[24]
Medical University of Bialystok Grants supported this study: 14333601L; 143-29595L. [25]
Author contributions [26]
Conceived and designed the experiments: PK LB ABZ AC. Performed the experiments: PK LB ABZ AC. Analyzed the data: PK LB MS AC. Contributed reagents/materials/analysis tools: PK LB ABZ AC. References [1] P.M. Webb, S.J. Jordan, Epidemiology of epithelial ovarian cancer, Best Pract Res Clin Obstet Gynaecol (2016). [2] T.S. Latha, K. Panati, D.S. Gowd, M.C. Reddy, D. Lomada, Ovarian cancer biology and immunotherapy, Int. Rev. Immunol. 33 (2014) 428–440. [3] I. Romero, R.C. Bast Jr., Minireview: human ovarian cancer: biology, current management, and paths to personalizing therapy, Endocrinology 153 (2012) 1593–1602. [4] P. Knapp, M. Baranowski, M. Knapp, P. Zabielski, A. Blachnio-Zabielska, J. Gorski, Altered sphingolipid metabolism in human endometrila cancer, Prostaglandins and Onther Lipid Mediators 92 (2010) 62–66. [5] T.S. Tirodkar, C. Voelkel-Johnson, Sphingolipids in apoptosis, Exp. Oncol. 34 (2012) 231–242. [6] B. Carroll, J.C. Donaldson, L. Obeid, Sphingolipids in the DNA damage response, Adv Biol Regul 58 (2015) 38–52. [7] K. Kitatani, M. Taniguchi, T. Okazaki, Role of sphingolipids and metabolizing enzymes in hematological malignancies, Mol. Cell 38 (2015) 482–495. [8] T. Levade, N. Andrieu-Abadie, O. Micheau, P. Legembre, B. Segui, Sphingolipids modulate the epithelial-mesenchymal transition in cancer, Cell Death Dis. 1 (2015) 15001. [9] C.S. Bonder, S.M. Pitson, Sphingolipids–who's controlling who in disease? Immunol. Cell Biol. 93 (2015) 767–768. [10] R. Ghidoni, A. Caretti, P. Signorelli, Role of sphingolipids in the pathobiology of lung inflammation, Mediat. Inflamm. 2015 (2015) 487508. [11] K. Marycz, A. Smieszek, M. Jelen, K. Chrzastek, J. Grzesiak, J. Meissner, The effect of the bioactive sphingolipids S1P and C1P on multipotent stromal cells—new opportunities in regenerative medicine, Cell. Mol. Biol. Lett. 20 (2015) 510–533. [12] M. Garcia-Barros, N. Coant, J.P. Truman, A.J. Snider, Y.A. Hannun, Sphingolipids in colon cancer, Biochim. Biophys. Acta 1841 (2014) 773–782. [13] A.U. Blachnio-Zabielska, X.M. Persson, C. Koutsari, P. Zabielski, M.D. Jensen, A liquid chromatography/tandem mass spectrometry method for measuring the in vivo
[27]
[28] [29]
[30]
[31]
[32] [33]
[34] [35] [36]
[37]
incorporation of plasma free fatty acids into intramyocellular ceramides in humans, Rapid Commun. Mass Spectrom. 26 (2012) 1134–1140. M. Knapp, M. Zendzian-Piotrowska, A. Blachnio-Zabielska, P. Zabielski, K. Kurek, J. Gorski, Myocardial infarction differentially alters sphingolipid levels in plasma, erythrocytes and platelets of the rat, Basic Res. Cardiol. 107 (2012) 294. M. Baranowski, A. Blachnio-Zabielska, M. Charmas, J.W. Helge, F. Dela, M. Ksiazek, et al., Exercise increases sphingoid base-1-phosphate levels in human blood and skeletal muscle in a time- and intensity-dependent manner, Eur. J. Appl. Physiol. (2014). M. Nagahashi, J. Tsuchida, K. Moro, M. Hasegawa, K. Tatsuda, I.A. Woelfel, K. Takabe, T. Wakai, High levels of sphingolipids in human breast cancer, J. Surg. Res. 204 (2016) 435–444. O. Sukocheva, C. Wadham, Role of sphingolipids in oestrogen signalling in breast cancer cells: an update, J. Endocrinol. 220 (2014) R25–35. V. del Solar, D.Y. Lizardo, N. Li, J.J. Hurst, C.J. Brais, G.E. Atilla-Gokcumen, Differential regulation of specific sphingolipids in colon cancer cells during staurosporine-induced apoptosis, Chem. Biol. 22 (2015) 1662–1670. S. Sonnino, A. Prinetti, The role of sphingolipids in neuronal plasticity of the brain, J. Neurochem. 137 (2016) 485–488. R. Kolesnick, Ceramide: a novel second messenger, Trends Cell Biol. 2 (1992) 232–236. T. Goldkorn, S. Filosto, S. Chung, Lung injury and lung cancer caused by cigarette smoke-induced oxidative stress: molecular mechanisms and therapeutic opportunities involving the ceramide-generating machinery and epidermal growth factor receptor, Antioxid. Redox Signal. 21 (2014) 2149–2174. Y.Y. Ma, X.Z. Mou, Y.H. Ding, H. Zou, D.S. Huang, Delivery systems of ceramide in targeted cancer therapy: ceramide alone or in combination with other anti-tumor agents, Expert Opin. Drug Deliv. 13 (2016) 1397–1406. I.G. Rivera, M. Ordonez, N. Presa, P. Gangoiti, A. Gomez-Larrauri, M. Trueba, T. Fox, M. Kester, A. Gomez-Munoz, Ceramide 1-phosphate regulates cell migration and invasion of human pancreatic cancer cells, Biochem. Pharmacol. 102 (2016) 107–119. K. Tanaka, K. Tamiya-Koizumi, M. Yamada, T. Murate, R. Kannagi, M. Kyogashima, Individual profiles of free ceramide species and the constituent ceramide species of sphingomyelin and neutral glycosphingolipid and their alteration according to the sequential changes of environmental oxygen content in human colorectal cancer Caco-2 cells, Glycoconj. J. 31 (2014) 209–219. S.J.H. Park, Acid ceramidase and human disease, Biochim. Biophys. Acta 1758 (2006) 2133–2138. R. Kolesnick, The therapeutic potential of modulating the ceramide/sphingomyelin pathway, J. Clin. Invest. 110 (2002) 3–8. A.J. Alberg, K. Armeson, J.S. Pierce, J. Bielawski, A. Bielawska, K. Visvanathan, E.G. Hill, B. Ogretmen, Plasma sphingolipids and lung cancer: a population-based, nested case-control study, Cancer Epidemiol. Biomark. Prev. 22 (2013) 1374–1382. H. Furuya, Y. Shimizu, T. Kawamori, Sphingolipids in cancer, Cancer Metastasis Rev. 30 (2011) 567–576. P. Giussani, C. Tringali, L. Riboni, P. Viani, B. Venerando, Sphingolipids: key regulators of apoptosis and pivotal players in cancer drug resistance, Int. J. Mol. Sci. 15 (2014) 4356–4392. J. Zou, X. Ma, G. Zhang, L. Shen, L. Zhou, Y. Yu, F. Zhu, Z. Chen, Evaluation of the change in sphingolipids in the human multiple myeloma cell line U266 and gastric cancer cell line MGC-803 treated with arsenic trioxide, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1004 (2015) 98–107. S. Li, Y. Zhou, X. Zheng, X. Wu, Y. Liang, S. Wang, Y. Zhang, Sphk1 promotes breast epithelial cell proliferation via NF-kappaB-p65-mediated cyclin D1 expression, Oncotarget (2016). J. Long, Y. Xie, J. Yin, W. Lu, S. Fang, SphK1 promotes tumor cell migration and invasion in colorectal cancer, Tumour Biol. 37 (2016) 6831–6836. S. Tsukamoto, M. Kumazoe, Y. Huang, C. Lesnick, N.E. Kay, T.D. Shanafelt, H. Tachibana, SphK1 inhibitor potentiates the anti-cancer effect of EGCG on leukaemia cells, Br. J. Haematol. (2016). R. Kolesnick, D. Altieri, Z. Fuks, A CERTain role for ceramide in taxane-induced cell death, Cancer Cell 11 (2007) 473–475. S. Chiantia, E. London, Sphingolipids and membrane domains: recent advances, Handb. Exp. Pharmacol. (2013) 33–55. L.R. Cordeiro Pedrosa, O. van Tellingen, T. Soullie, A.L. Seynhaeve, A.M. Eggermont, T.L. Ten Hagen, M. Verheij, G.A. Koning, Plasma membrane targeting by short chain sphingolipids inserted in liposomes improves anti-tumor activity of mitoxantrone in an orthotopic breast carcinoma xenograft model, Eur. J. Pharm. Biopharm. 94 (2015) 207–219. R. Sutphen, Y. Xu, G.D. Wilbanks, J. Fiorica, E.C. Grendys Jr., J.P. LaPolla, H. Arango, M.S. Hoffman, M. Martino, K. Wakeley, D. Griffin, R.W. Blanco, A.B. Cantor, Y.J. Xiao, J.P. Krischer, Lysophospholipids are potential biomarkers of ovarian cancer, Cancer Epidemiol. Biomark. Prev. 13 (2004) 1185–1191.
Please cite this article as: P. Knapp, et al., Plasma and ovarian tissue sphingolipids profiling in patients with advanced ovarian cancer, Gynecol Oncol (2017), http://dx.doi.org/10.1016/j.ygyno.2017.07.143