Mol Biol Rep (2014) 41:459–466 DOI 10.1007/s11033-013-2880-0
Carcinoembryonic antigen expression level as a predictive factor for response to 5-fluorouracil in colorectal cancer Ebrahim Eftekhar • Fakhraddin Naghibalhossaini
Received: 27 January 2013 / Accepted: 21 November 2013 / Published online: 29 November 2013 Ó Springer Science+Business Media Dordrecht 2013
Abstract Carcinoembryonic antigen (CEA) expression has been shown to protect cancer cell lines from apoptosis and anoikis. The aim of this study was to further elucidate the role of CEA expression on resistance to anticancer drugs in human colorectal cancer (CRC). We transfected CEA negative CRC cell line SW742 as well as CHO cells to overexpress CEA and their chemoresistance were assessed by MTT assay. In comparison to the parental cell lines, transfected cells had significantly increased resistance to 5-fluorouracil (5-FU). The results also showed a direct correlation between the amount of cellular CEA protein and 5-FU resistance in CEA expressing cells. We found no significant difference in sensitivity to cisplatin and methotrexate between CEA-transfected cells and their counter parental cells. We also compared the association between CEA expression and chemoresistance of 4 CRC cell lines which differed in the levels of CEA production. The CEA expression levels in monolayer cultures of these cell lines did not correlate with the 5-FU resistance. However, 5-FU treatment resulted in the selection of sub-populations of resistant cells that displayed increased CEA expression levels by increasing drug concentration. We analyzed the effect of 5-FU in a 3D multicellular culture generated from the two CRC cell lines, LS180 and HT29/219. Compared with monolayer culture, CEA production and 5-FU resistance in both cell lines were stimulated by 3D growth. In comparison to the 3D spheroids of parental CHO, we observed a significantly elevated 5-FU resistance in 3D culture of the CEAE. Eftekhar F. Naghibalhossaini (&) Department of Biochemistry, School of Medicine, Shiraz University of Medical Sciences, Zand Street, Shiraz, Iran e-mail: [email protected] F. Naghibalhossaini Autoimmune Research Center, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran
expressing CHO transfectants. Our findings suggest that the CEA level may be a suitable biomarker for predicting tumor response to 5-FU-based chemotherapy in CRC. Keywords CEA Chemotherapy Colorectal cancer Drug resistance 5-Fluorouracil
Introduction Colorectal cancer (CRC) is one of the most commonly occurring malignancies and one of the leading causes of cancer related death globally [1]. The primary treatment for CRC is surgery and chemotherapy is used complementary to decrease the risk of local recurrences. 5-Fluorouracil (5FU) in combination with other drugs has long been used in CRC chemotherapy and in a number of other common malignancies such as breast, esophageal, and gastric cancers [2]. However, tumors resistance in the course of treatment is an important cause of failure in clinical colon cancer therapy. In fact, response rates ranging 20–30 % have been reported in metastatic patients [3]. A goal of cancer therapy is to identify molecular markers to predict response and toxicity to drugs in order to facilitate the individualization of patient treatment. A large number of studies have attempted to define therapeutic and molecular determinants of interindividual differences in responses to the 5-FU based chemotherapy. The lack of response to 5-FU has been related to an increased expression of the target thymidylate synthase, and a decreased level of 5-FU-activating enzymes, including orotate phosphoribosyl-transferase and uridine kinase [4, 5]. Carcinoembryonic antigen (CEA), first described by Gold and Freeman in 1965, is a tumor associated antigen that is produced by a wide variety of human cancers
123
460
including colorectal, breast and lung tumors [6]. CEA is upregulated in approximately 90 % of advanced CRC and is associated with increased tumorigenicity [7, 8]. Elevated preoperative serum CEA levels have been correlated with high recurrence rates of CRC [9, 10]. In addition to its use as a tumor marker for colon carcinomas, elevated expression of CEA has been implicated in various biological aspects of neoplasia, including tumor cell adhesion, metastasis, the blocking of cellular differentiation and antiapoptosis functions [6]. Clinical observations have suggested that serum and tumor CEA concentration were associated with poor tumor response to chemoradiotherapy and poor outcome [11, 12]. Some previous studies have reported that CEA overexpression significantly protected tumor cells from undergoing anoikis [13–15], a form of apoptosis caused by detachment from cell matrix. A CEA-targeted ribozyme in human HT-29 colon cancer cells has been also shown to increase apoptosis in response to various apoptotic stimuli including UV irradiation, IFN-c treatment, and 5-FU treatment [16]. However, limited data are available regarding the correlation between cellular CEA expression levels and resistance to chemotherapy agents. Therefore, the goal of this study was to explore the relationship between CEA expression levels and resistance to anticancer drugs in human CRC cells.
Materials and methods
Mol Biol Rep (2014) 41:459–466
viability was determined by MTT dye reduction assay as described previously [17]. Briefly, 20 ll of the cellcounting solution (MTT solution) was added to each well to a final concentration of 0.5 mg/ml and incubated in a humidified 5 % CO2 atmosphere at 37 °C for 4 h. The crystals were dissolved in 150 ll of DMSO/well. The absorbance of the solution was read at 570 nm using a microtiter plate reader (Mikura, Englad). Cell viability was calculated according to the following formula: Cell viability (%) = OD570 (sample)/OD570 (control) 9 100. IC50 determination The IC50 was determined from the dose–response graph. Analysis of the dose–response curve was performed using the Software GraphPad PRISM Version 5.00 (GraphPad Software, San Diego, CA). Spheroid cultures Multicellular tumor spheroids were generated using a previously described method with minor modifications [18]. Briefly, a suspension of subconfluent monolayercultured tumor cells was plated over 1.5 % agarose in medium-coated 96-well plates at a density of *1,200 cells per well. After 5–7 days incubation in the incubator at 37 °C, three-dimensional multicellular spheroids had formed and were analyzed for 5-FU drug treatment. Spheroid experiments were done in three replicates and repeated at least two times.
Chemicals Plasmids and transfections All chemicals were purchased from Sigma (Gillingham, UK) unless otherwise stated. Reagents were prepared and stored according to the manufacturers’ instructions. Cell culture and drug treatment The Chinese hamster ovary (CHO) cell line and the human colon carcinoma cell lines SW742, LS180, HT29/219, and Caco2 were obtained from the National Cell Bank of Iran (NCBI, Pasteur Institute, Tehran). Cells were cultured in either RPMI 1640 or in DMEM supplemented with 10 % fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, and 100 lg/ml streptomycin (all from Gibco) in a humidified 5 % CO2 atmosphere at 37 °C. To determine IC50, a triplicate aliquot of cells were seeded in 96-well microtiter plates (3 9 103 cells/well in 100 ll medium). Twenty-four hours after seeding, 5-FU, methotrexate, and cisplatin solutions were added at desired concentrations. The medium of control cells was replenished with medium without chemicals. The cells were incubated at 37 °C for 72 h in a humidified incubator. Cell
123
The plasmid pcDNA3.1 (?) (Invitrogen, USA) containing human CEA cDNA was constructed by our laboratory as previously described [19]. The CHO cells were seeded into 3 9 10 cm petri dishes and stably transfected at 80 % confluence with pcDNA3.1 (?) containing full length CEA cDNA by calcium phosphate co-precipitation. The human colon carcinoma SW742 cells were transfected by electroporation as follows. Briefly, 106 cells in 400 ll of medium were mixed with 20 lg of pcDNA3.1 (?)/CEA. Electroporation was performed by using 1200 V and 25 lF. Stable CHO and SW742 transfectants were selected by incubating cells in Geneticin (G418 sulfate) (400 lg/ml) for 15 days. G418-resistant colonies were pooled and enriched for stable transfectants. Cell lysate preparation and immunoblotting Monolayer cultures of cells were removed from plastic surfaces by scrapping in whole-cell lysis buffer (50 mM Tris–HCl, pH 7.5, 0.5 M NaCl, and 1 % NP-40)
Mol Biol Rep (2014) 41:459–466
supplemented with antiprotease cocktail (Roche Applied Science, Germany). Lysates were stored frozen at -80 °C until use. 30–50 lg of total protein was subjected to electrophoresis using 7.5 % SDS-PAGE. The gels were transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore, MA, USA) and analyzed for CEA as described [20]. The blots were probed with anti-actin antibody as an internal control of protein loading. CEA assay Cell lysates prepared as described above and CEA content of cells was determined using a commercially available CEA ELISA kit (CanAg Diagnostics AB, Gothenburg, Sweden), according to the procedure provided by the manufacturer. The total cellular protein was determined according to Bradford method [21] using crystalline BSA as a standard and the CEA content was normalized to nanograms per milligram cellular protein.
Results Expression of CEA in CRC and CHO cell lines We examined the role of CEA in mediating the chemoresistance of cells to anticancer drugs by transfecting an expression vector (PCDNA3.1) containing the human CEA cDNA into CHO and SW742 human colon carcinoma cells. These cell lines were chosen as targets because CHO cell line does not contain CEA gene and SW742 human carcinoma cells express no detectable CEA protein. The transfection experiments were repeated three times. Fifteen to eighteen geneticin-resistant colonies of CHO and SW742 cells were pooled and their lysates were examined for CEA expression. A significant increase of CEA expression was detected in crude cell extracts by ELISA assay. By immunoblot analysis, a band corresponding to the recombinant CEA was only detected in the transfected cells, with an apparent size of 180 kD (Fig. 1). Maximum CEA expression levels measured in crude cell extracts in
Fig. 1 Western blot analysis of CEA expression in human CRC cell line SW742 and in CHO-cells transfected with human CEA gene The right panel indicates blot of two different pools of CHO–CEA transfectants with different levels of CEA expression
461
CHO and SW742 transfectants were 70 and 13.02 ng of CEA protein per mg total protein content at sub-confluent stage, respectively. Under the same conditions, in the control parental CHO and SW742 cells, no detectable expression of CEA was observed (Fig. 1). The growth rate of the transfected cells did not differ from the growth rate of the wild-type cells (data not shown). Sensitivity of CEA-transfected cell lines to anticancer drugs Stably transfected SW742 and CHO cells (3,000 cells/well in 96-well plates) were exposed to graded concentrations of 5-FU on day 1 after seeding. After 72 h, cell proliferation was assessed by MTT dye reduction assay and relevant IC50 values (50 % reduction in cell viability) determined. Increasing the 5-FU concentrations caused progressively greater cytotoxicity and growth inhibition. The IC50 value in the wild type CHO and SW742 were 0.41 and 21.9 lM, respectively (Table 1). Both transfected cell lines were more resistant to the cytotoxic action of 5-FU when compared to the wild type cells, having a 1.8 to 5-fold increase in the IC50 value. For SW742 transfectants, the IC50 value for 5-FU treatment in low and high CEA expresser was 40.2 ± 8.6 and 44.4 ± 11.75 lM, respectively and the IC50 for 5-FU treatment in low and high CEA expressing CHO cells was 1.06 ± 0.36 and 2.15 ± 0.56 lM, respectively (Table 1). We next examined the specificity of CEA effect on drug resistance by evaluating the sensitivity of CEA-transfected CHO and SW742 cells to cisplatin and methotrexate, two other commonly used anticancer drugs with different mechanisms of action. Cells (3,000 cells/well) were seeded in 96-well plates. 24 h later, cisplatin and methotrexate were added at the concentrations 0.2–12.5 and 0.02–2.5 lM, respectively, for 72 h. As shown in Table 1, there was no significant difference in sensitivity to cisplatin and methotrexate between wild type parental and their CEA-transfected cells. Sensitivity of cell lines to chemotherapy by 5-FU Since our DNA transfection studies clearly demonstrated that CEA production confers resistance to 5-FU, we compared the effect of 5-FU on a panel of 4 CRC cell lines (SW742, HT29/219, LS180 and Caco-2) with a wide range of CEA production. Our previous work has determined that LS180 has very high level of CEA expression and the other two well-differentiated cell lines, Caco-2 and HT29/219 express relatively moderate levels of CEA, but SW742 cells express no detectable level of CEA protein [20]. Three thousand cells were seeded in a 96-well plate. Twenty-four hours later cells were treated with graded
123
462
Mol Biol Rep (2014) 41:459–466
Table 1 Sensitivity of CHO/CEA and SW742/CEA transfectants to anticancer drugs Cell lines
CEA protein content (ng/mg total protein)b
IC50 (5-FU) (lM)a Mean ± SD
IC50 (cisplatin) (lM)a Mean ± SD
IC50 (methotrexate) (lM)a Mean ± SD
CHO
0
0.41 ± 0.22
1.46 ± 0.23
0.014 ± 0.01
CHO/CEA-1
70
2.15 ± 0.56
1.12 ± 0.32
0.014 ± 0.00
CHO/CEA-2
26.96
1.06 ± 0.36
ND
ND
SW742
0
21.9 ± 8.04
1.28 ± 0.08
0.07 ± 0.01
SW742/CEA-1
13.02
44.4 ± 11.75
0.94 ± 0.19
0.08 ± 0.01
SW742/CEA-2
9.6
40.2 ± 8.6
ND
ND
ND not determined a
IC50 value for each transfectant is presented from three independent assays
b
Data indicate the CEA expression in parental cell lines and in two independent pools of their CEA transfectants (CEA-1and CEA-2)
Table 2 CEA expression levels and growth inhibition of colon cancer cell lines by anti-cancer drug 5-FU Cell line
CEA expression level (ng/mg protein)
IC50 (lM)a Mean ± SD
LS180
174.3 ± 16.6
9.64 ± 4.33
20.2 ± 0.3
8.67 ± 3.17
5.7 ± 1.7
15.7 ± 1.01
HT29/219 Caco2 SW742
Undetectable
21.9 ± 8.04
a
IC50 value is presented as a mean of two independent assays, each done in triplicate
HT29/219 cancer cell lines (Table 2). Similar results on cell death were obtained by trypan blue staining (data not shown). Antigenic heterogeneity within cell lines and variations in 5-FU chemosensitivity
Fig. 2 Sensitivity of CRC cell lines to growth inhibition by anticancer agent 5-FU Cell viability was determined using MTT assay after treatment of cells with various concentrations of anticancer agents at 37 °C for 72 h. Data are means of three experiments, each done in triplicate. a The linear scale of X-axes; b The log scale of X-axes
concentrations of 5-FU. MTT assay results indicated that all the cells undergo growth inhibition in a dose-dependent manner (Fig. 2). However, there was no direct correlation between 5-FU resistance and cellular CEA expression levels in these cell lines. The IC50 of 5-FU was 21.9 ± 8.04 lM for SW742, 9.64 ± 4.33 lM for LS180, 15.7 ± 1.01 lM for Caco-2, and 8.67 ± 3.17 lM for
123
Human CRC cell lines are composed of heterogeneous population of cells that express different amounts of CEA antigen [22]. Considering the heterogeneity of cells within each cell line in terms of CEA expression, we investigated the relationship between 5-FU resistance and CEA expression levels of cellular subpopulations in two CRC cell lines. To elucidate antigen expression profile of cells after selective resistance to chemotherapy, HT29/219 and LS180 cells were treated with 2 different concentrations of 5-FU and CEA levels were measured in CRC cells that survived chemotherapy. For this purpose 2 9 105 cells were seeded in T25 flasks. Twenty-four hours later, 5-FU was added at the concentrations indicated for 72 h and protein extracts of surviving cells were examined for CEA expression by ELISA assay (Table 3). The results indicated that 5-FU resisting cells display higher CEA contents than control cells (unexposed to 5-FU) and cells survived exposure to high concentrations of 5-FU display higher CEA expression than cells surviving exposure to low 5-FU
Mol Biol Rep (2014) 41:459–466
463
Table 3 CEA protein expression in 5-FU resistant subpopulations of CRC cells Cell lines
5-FU conc (lM) 0
LS180
HT29/219
ng CEA/mg proteina,c
ng CEA/mg proteinb,c
Table 4 Expression of CEA in 2D and 3D culture of CRC cells and growth inhibition by 5-FU Cell lines
174.3 ± 16.6
20
187.6 ± 18.3
40 0
223.8 ± 19.8 20.2 ± 0.3
12
34.6 ± 2.3
24
37.5 ± 10.2
192 ± 15.9
32.5 ± 3.1
CEA expression (ng/mg protein)a
IC50 (lM)b
2D culture
2D culture
3D culture
LS180
174.3 ± 16.6
207.8 ± 20.3
9.6 ± 4.3
12.1 ± 5.3
HT29/ 219
20.2 ± 0.3
36.3 ± 2.3
8.6 ± 3.1
25.2 ± 1.4
0.4 ± 0.22
1.77 ± 0.82
2.15 ± 0.56
5.7 ± 1.76
CHO
0.00
0.00
70
68
a
CHO/ CEA1
b
a, b
CEA expression levels after growing in medium containing 5-FU drug
CEA expression levels after growing the 5-FU pre-treated cells for two passages in 5-FU-free medium
3D culture
Mean ± SD, mean values were calculated from two independent experiments, each done in triplicate
c
Mean ± SD, mean values were calculated from two independent experiments, each done in triplicate
concentration. To verify whether the 5-FU-mediated increase of CEA was a stable characteristic, we re-cultivated the 5-FU pre-treated cells in 5-FU-free medium. The CEA expression did not change after growing cells for two passages in 5-FU-free medium (Table 3). These results suggest that within each cell line the subpopulations of cells that have retained high-level of CEA expression resist chemotherapy and survive longer. Multicellular spheroid culture induce CEA expression and 5-FU resistance in CRC cells Numerous studies have reported that the formation of multicellular aggregates of tumor cells (3D culture) increase their resistance to chemotherapy agents [23–25]. To investigate the effect of spheroid culture on the chemosensitivity of cells to 5-FU, HT29/219, SW742, LS180, and Caco-2 cells as well as CHO and CHO/CEA transfectants were cultured under 3D culture condition as described in ‘‘Materials and methods‘‘ section. SW742 and Caco-2 cells were unable to form spheroids, but remained
as free-floating single cells and died after a few days. However, on day 5–7 after seeding, LS180, HT29/219, and CHO/CEA transfectants produced large, approximately 500 lm aggregates (Fig. 3) and 5-FU added as described above. In comparison to 2D monolayer culture, the IC50 value of multicellular spheroids of LS180 and HT29/219 increased by 1.3 times (IC50 = 12.1 ± 5.3 lM) and 3 times (IC50 = 25.2 ± 1.4 lM), respectively (Table 4). Interestingly, the multicellular spheroids of CHO/CEA transfectant was also more resistant to 5-FU (IC50 = 5.7 ± 1.76 lM) than multicellular aggregates of untransfected CHO cells (IC50 = 1.77 ± 0.82 lM). We then measured the CEA expression in multicelluar aggregates of CRC cells. As shown in Table 4, CEA expression was upregulated in LS180 and HT29/219 spheroids by 20 and 80 %, respectively, compared with monolayer culture of these cells (P \ 0.05).
Discussion The usual treatment of CRC is surgery, which in many cases is followed by chemotherapy using 5-FU and its derivatives to reduce tumor recurrence. However, human
Fig. 3 Micrograph of LS80 and HT29/219 spheroid
123
464
colon carcinomas are heterogeneous at the molecular and cellular levels and a considerable heterogeneity in the response to chemotherapy are observed in clinical studies [3]. The identification of molecular variables that are involved in tumor resistance to chemotherapy is of major interest in predicting treatment response and avoiding adverse events. CEA is a widely used tumor marker for monitoring of the course of CRC progression and a popular molecular target for cancer immunotherapy. Clinical studies suggest that CEA overexpression is associated with poor clinical outcome and reduced survival of colon cancer patients [26, 27]. Soeth et al. [16] reported that a CEA-targeted ribozyme that silenced CEA expression in human HT-29 colon cancer cells increases apoptosis in response to various apoptotic stimuli including 5-FU treatment. However, no previous studies have directly examined the effect of CEA overexpression on resistance of CRC cells to anticancer drugs. Therefore, the aim of this study was to determine the correlation between CEA expression levels and response to treatment with anticancer drugs in CRC cells. For this purpose the full length CEA cDNA, cloned in the eukaryotic expression vector pcDNA3.1, was transfected into human colon carcinoma SW742 and into CHO cell line. Transfection experiments were done in three replicates and repeated at least two times. We obtained several pools of CEA-transfectants that expressed different levels of the transfected cDNA. Several of the moderate- and high-expressing pools of clones were examined for drug resistance. CEA transfected cells were more resistant to 5-FU than untransfected parental cells (Table 1). Moreover, there was a significant correlation between the CEA expression levels in transfected cells and their resistance to 5-FU cytotoxity. When we examined the cytotoxic effects of methotrexate and cisplatin, the sensitivity of SW742and CHO–CEA transfectants was not significantly different from that of parental cells. The results suggest that the selective resistance to 5-FU cytotoxicity was a direct effect of the elevated CEA expression in the transfected cells rather than a general effect due to selection process. Results presented here confirm and extend the findings of Soeth et al. [16] who demonstrated that decreasing CEA expression in HT-29 cells by an inducible hammerhead ribozyme to CEA, increases apoptosis in response to5-FU. When we compared the effect of 5-FU on a panel of four CRC cell lines, the correlation between CEA expression and sensitivity to 5-FU did not bear a simple relationship (Table 2). Among the cell lines examined, the maximal IC50 value for 5-FU induced cytotoxicity was observed with SW742 cells, in which no CEA protein was detected. Obviously, other differences between human colon carcinoma cells other than CEA expression levels contribute to the drug resistance in neoplastic cells. These include
123
Mol Biol Rep (2014) 41:459–466
mechanisms involved in drug influx and efflux, processing of drug-induced damage, drugs inactivation, and induction of cell apoptosis etc. The genotypes and activity of several enzymes including thymidylate synthase, thymidine phosphorylase, and MTHFR enzyme have been shown previously to correlate with cancer response to 5-FU in vitro and in vivo [4, 28, 29]. On the basis of the observed heterogeneity within CRC cell lines in terms of the amount of CEA expression [22], we anticipated that the high CEA-expressing fraction of CRC cell lines would be more resistant and selected by 5-FU treatment. To test this hypothesis, we cultured two CRC cell lines HT29/219 and LS180 with two different concentrations of 5-FU and the cellular CEA contents were analyzed before and after 5-FU treatment. In both cell lines, surviving cells displayed an elevated CEA expression in comparison to untreated cells and the fractions survived at high drug concentration expressed higher levels of CEA protein than cells survived at low drug concentration (Table 3). Some authors previously suggested that treatment of neoplastic cell lines with 5-FU may induce CEA up-regulation transiently [30, 31]. When we re-cultivated the 5-FU-pretreated cells for two passages in 5-FU-free medium, the CEA expression did not decline. Moreover, previous investigations showed that 5-FU was not able to induce CEA expression in normal or in CEA-negative tumor cells [31, 32]. These observations support the hypothesis that the mechanism of 5-FU-mediated increase of cellular CEA levels might be due to the selection process rather than drug-induced CEA expression. Sorbye and Dahl also reported a transient elevation of serum CEA during FOLFOX therapy in CRC patients [33]. But, they did not measure CEA expression in tumor tissues. There are many factors affecting serum CEA concentrations in cancer patients with CRCs [34]. Indeed, the plasma CEA values in response to chemotherapy may show an initial delay or even a transient surge due to cytolysis in patients responding to chemotherapy before demonstrating the expected pattern of change [35]. Human tumors, although often treated as a single entity, comprise a population of cells with heterogeneous expression of several surface molecules including CEA antigen [22, 28, 36]. Therefore, 5-FU treatment may result in the selection of high CEA-expressing tumor cells that would tolerate high drug concentrations. Additional experimental and clinical studies are needed to determine the correlation between levels of CEA expression and dose of 5-FU drug required for treatment of CEA expressing tumors. Three dimensional cultures of cancer cells are better representation of solid tumors when compared to monolayer cultures. In agreement with previous studies [37, 38], we
Mol Biol Rep (2014) 41:459–466
found that 3D growth of CRC cells was associated with an increased 5-FU resistance (LS180: IC50 = 12.1 ± 5.3 lM, 1.3-fold increase; HT29/219: IC50 = 25.2 ± 1.4 lM, threefold increase). As shown in Table 4, CEA levels are increased in 3D culture of CRC cell lines when compared to monolayer cultures of these cells. Compared with monolayer culture, the 3D culture of LS180 and HT29/219 cells resulted in a 20 and 80 % increase in CEA expression levels, respectively. This observation is similar to another investigation which found that the CEA expression is stimulated by 3D growth of poorly differentiated MIP-101 human colorectal carcinoma cells that do not produce CEA in monolayer cultures [39]. It has been reported that it is more difficult for drug to penetrate into the center cell mass of 3D spheroids. Some prior studies have suggested that 5-FU specifically targets proliferating cells, and thus would not kill the quiescent cells in the spheroids. Whereas in 2D monolayer cultures, cells proliferate at a faster rate and thus 5-FU inhibits cellular growth more effectively [40]. In our study, the 3D spheroids of CHO/CEA transfectants were significantly more resistant to 5-FU when compared to untransfected parental CHO cells cultured under 3D condition. Although above mentioned factors are likely to play some roles in increasing chemo-resistance of multicellular spheroids of cells, other factors, such as up-regulated CEA expression and cell–cell interactions may also influence the survival of aggregated CRC cells to cytotoxic drugs. The actual mechanisms underlying CEA-mediated 5-FU resistance is unknown. It has been reported that CEA can protect tumor cells from undergoing anoikis, i.e., apoptosis induced by loss of cell contact with the extracellular matrix, by directly binding to TRAIL-R2 (DR5) [13, 15]. Previous analysis of 273 genes in HT29 cells by microarray experiment also revealed that CEA down-regulate expression of various groups of cancer-related genes, in particular apoptotic genes [16]. Definition of the biological characteristics of resistant tumor cells is important for developing treatment strategies and selecting patients who are most likely to benefit from the treatment. CEA is an important biomarker in the diagnosis and monitoring of relapse in CRC. We believe that the results of our study offer useful information regarding the potential use of CEA expression levels as a marker for tumor response to fluorouracil-based chemotherapy in CRC. Acknowledgments This study was a part of the dissertation of Ebrahim Eftekhar, submitted to Shiraz University of Medical Sciences in partial fulfillment of the requirements for the Ph. D in biochemistry. This work was supported by a grant from the Vice Chancellor for Research, Shiraz University of Medical Sciences.
465
References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61(2):69–90 2. Malet-Martino M, Jolimaitre P, Martino R (2002) The prodrugs of 5-fluorouracil. Curr Med Chem Anticancer Agents 2(2):267–310 3. Isacoff WH, Borud K (1997) Chemotherapy for the treatment of patients with metastatic colorectal cancer: an overview. World J Surg 21(7):748–762 4. Furuta T (2009) Pharmacogenomics in chemotherapy for GI tract cancer. J Gastroenterol 44(10):1016–1025 5. Peters GJ, Backus HH, Freemantle S, van Triest B, CodacciPisanelli G, van der Wilt CL, Smid K, Lunec J, Calvert AH, Marsh S, McLeod HL, Bloemena E, Meijer S, Jansen G, van Groeningen CJ, Pinedo HM (2002) Induction of thymidylate synthase as a 5-fluorouracil resistance mechanism. Biochim Biophys Acta 1587(2–3):194–205 6. Chan CH, Stanners CP (2007) Recent advances in the tumour biology of the GPI-anchored carcinoembryonic antigen family members CEACAM5 and CEACAM6. Curr Oncol 14(2):70–73 7. Goldstein MJ, Mitchell EP (2005) Carcinoembryonic antigen in the staging and follow-up of patients with colorectal cancer. Cancer Invest 23(4):338–351 8. Ilantzis C, DeMarte L, Screaton RA, Stanners CP (2002) Deregulated expression of the human tumor marker CEA and CEA family member CEACAM6 disrupts tissue architecture and blocks colonocyte differentiation. Neoplasia 4(2):151–163 9. Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, Somerfield MR, Hayes DF, Bast RC Jr (2006) ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 24(33):5313–5327 10. Park IJ, Kim HC, Yu CS, Yoo JH, Kim JC (2005) Cutoff values of preoperative s-CEA levels for predicting survivals after curative resection of colorectal cancer. J Korean Med Sci 20(4):624–627 11. Lin JZ, Zeng ZF, Wu XJ, Wan DS, Chen G, Li LR, Lu ZH, Ding PR, Pan ZZ (2010) Phase II study of pre-operative radiotherapy with capecitabine and oxaliplatin for rectal cancer and carcinoembryonic antigen as a predictor of pathological tumour response. J Int Med Res 38(2):645–654 12. Li M, Li JY, Zhao AL, He JS, Zhou LX, Li YA, Gu J (2009) Comparison of carcinoembryonic antigen prognostic value in serum and tumour tissue of patients with colorectal cancer. Colorectal Dis 11(3):276–281 13. Ordonez C, Screaton RA, Ilantzis C, Stanners CP (2000) Human carcinoembryonic antigen functions as a general inhibitor of anoikis. Cancer Res 60(13):3419–3424 14. Chan CH, Camacho-Leal P, Stanners CP (2007) Colorectal hyperplasia and dysplasia due to human carcinoembryonic antigen (CEA) family member expression in transgenic mice. PLoS ONE 2(12):e1353 15. Samara RN, Laguinge LM, Jessup JM (2007) Carcinoembryonic antigen inhibits anoikis in colorectal carcinoma cells by interfering with TRAIL-R2 (DR5) signaling. Cancer Res 67(10):4774–4782 16. Soeth E, Wirth T, List HJ, Kumbhani S, Petersen A, Neumaier M, Czubayko F, Juhl H (2001) Controlled ribozyme targeting demonstrates an antiapoptotic effect of carcinoembryonic antigen in HT29 colon cancer cells. Clin Cancer Res 7(7):2022–2030 17. Li HL, Wang TS, Li XY, Li N, Huang DZ, Chen Q, Ba Y (2007) Overexpression of cyclin L2 induces apoptosis and cell-cycle arrest in human lung cancer cells. Chin Med J (Engl) 120(10):905–909
123
466 18. Friedrich J, Seidel C, Ebner R, Kunz-Schughart LA (2009) Spheroid-based drug screen: considerations and practical approach. Nat Protoc 4(3):309–324 19. Naghibalhossaini F, Pakdel A, Ghaderi AA, Saberi Firoozi M (2005) Effective production of carcinoembryonic antigen by conversion of the membrane-bound into a recombinant secretory protein by site-specific mutagenesis. Pathol Oncol Res 11(4):211–217 20. Pakdel A, Naghibalhossaini F, Mokarram P, Jaberipour M, Hosseini A (2012) Regulation of carcinoembryonic antigen release from colorectal cancer cells. Mol Biol Rep 39(4):3695–3704 21. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 22. Han JS, Nair PP (1995) Flow cytometric identification of cell surface markers on cultured human colonic cell lines using monoclonal antibodies. Cancer 76(2):195–200 23. Green SK, Francia G, Isidoro C, Kerbel RS (2004) Antiadhesive antibodies targeting E-cadherin sensitize multicellular tumor spheroids to chemotherapy in vitro. Mol Cancer Ther 3(2):149–159 24. St Croix B, Kerbel RS (1997) Cell adhesion and drug resistance in cancer. Curr Opin Oncol 9(6):549–556 25. Smalley KS, Lioni M, Herlyn M (2006) Life isn’t flat: taking cancer biology to the next dimension. In Vitro Cell Dev Biol Animal 42(8–9):242–247 26. Ballesta AM, Molina R, Filella X, Jo J, Gimenez N (1995) Carcinoembryonic antigen in staging and follow-up of patients with solid tumors. Tumour Biol 16(1):32–41 27. Jantscheff P, Terracciano L, Lowy A, Glatz-Krieger K, Grunert F, Micheel B, Brummer J, Laffer U, Metzger U, Herrmann R, Rochlitz C (2003) Expression of CEACAM6 in resectable colorectal cancer: a factor of independent prognostic significance. J Clin Oncol 21(19):3638–3646 28. Bandres E, Zarate R, Ramirez N, Abajo A, Bitarte N, GariiaFoncillas J (2007) Pharmacogenomics in colorectal cancer: the first step for individualized-therapy. World J Gastroenterol 13(44):5888–5901 29. Longley DB, Johnston PG (2005) Molecular mechanisms of drug resistance. J Pathol 205(2):275–292 30. Prete SP, Turriziani M, Massara MC, De Rossi A, Correale P, De Vecchis L, Torino F, Bonmassar L, Aquino A (2008) Combined effects of 5-fluorouracil, folinic acid and oxaliplatin on the
123
Mol Biol Rep (2014) 41:459–466
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
expression of carcinoembryonic antigen in human colon cancer cells: pharmacological basis to develop an active antitumor immunochemotherapy. J Exp Clin Cancer Res 270:5 Aquino A, Prete SP, Guadagni F, Greiner JW, Giuliani A, Orlando L, Masci G, De Santis S, Bonmassar E, Graziani G (2000) Effect of 5-fluorouracil on carcinoembryonic antigen expression and shedding at clonal level in colon cancer cells. Anticancer Res 20(5B):3475–3484 Cappelletti D, Cardillo A, Bonanno E, Prete SP, Cucchiara G, Turriziani M, Greiner JW, Cottarelli A, Breda E, Aquino A, Bonmassar E, De Vecchis L (2000) Drug-induced modulation of carcinoembryonic antigen (CEA) expression in neoplastic cells from a patient with rectal cancer. J Exp Clin Cancer Res 19(4):467–469 Sørbye H, Dahl O (2004) Transient CEA increase at start of oxaliplatin combination therapy for metastatic colorectal cancer. Acta Oncol 43(5):495–834 Duffy MJ (2001) Carcinoembryonic antigen as a marker for colorectal cancer: is it clinically useful? Clin Chem 47(4):624–630 Bidart JM, Thuillier F, Augereau C, Chalas J, Daver A, Jacob N, Labrousse F, Voitot H (1999) Kinetics of serum tumor marker concentrations and usefulness in clinical monitoring. Clin Chem 45(10):1695–1707 Leteurtre E, Gouyer V, Rousseau K, Moreau O, Barbat A, Swallow D, Huet G, Lesuffleur T (2004) Differential mucin expression in colon carcinoma HT-29 clones with variable resistance to 5-fluorouracil and methotrexate. Biol Cell 96(2):145–151 Graham CH, Kobayashi H, Stankiewicz KS, Man S, Kapitain SJ, Kerbel RS (1994) Rapid acquisition of multicellular drug resistance after a single exposure of mammary tumor cells to antitumor alkylating agents. J Natl Cancer Inst 86(13):975–982 Sutherland RM (1988) Cell and environment interactions in tumor microregions: the multicell spheroid model. Science 240(4849):177–184 Jessup JM, Brown D, Fitzgerald W, Ford RD, Nachman A, Goodwin TJ, Spaulding G (1997) Induction of carcinoembryonic antigen expression in a three-dimensional culture system. In Vitro Cell Dev Biol Animal 33(5):352–357 Tung YC, Hsiao AY, Allen SG, Torisawa YS, Ho M, Takayama S (2011) High-throughput 3D spheroid culture and drug testing using a 384 hanging drop array. Analyst 136(3):473–478
Carcinoembryonic antigen expression level as a predictive factor for response to 5-fluorouracil in colorectal cancer Ebrahim Eftekhar • Fakhraddin Naghibalhossaini
Received: 27 January 2013 / Accepted: 21 November 2013 / Published online: 29 November 2013 Ó Springer Science+Business Media Dordrecht 2013
Abstract Carcinoembryonic antigen (CEA) expression has been shown to protect cancer cell lines from apoptosis and anoikis. The aim of this study was to further elucidate the role of CEA expression on resistance to anticancer drugs in human colorectal cancer (CRC). We transfected CEA negative CRC cell line SW742 as well as CHO cells to overexpress CEA and their chemoresistance were assessed by MTT assay. In comparison to the parental cell lines, transfected cells had significantly increased resistance to 5-fluorouracil (5-FU). The results also showed a direct correlation between the amount of cellular CEA protein and 5-FU resistance in CEA expressing cells. We found no significant difference in sensitivity to cisplatin and methotrexate between CEA-transfected cells and their counter parental cells. We also compared the association between CEA expression and chemoresistance of 4 CRC cell lines which differed in the levels of CEA production. The CEA expression levels in monolayer cultures of these cell lines did not correlate with the 5-FU resistance. However, 5-FU treatment resulted in the selection of sub-populations of resistant cells that displayed increased CEA expression levels by increasing drug concentration. We analyzed the effect of 5-FU in a 3D multicellular culture generated from the two CRC cell lines, LS180 and HT29/219. Compared with monolayer culture, CEA production and 5-FU resistance in both cell lines were stimulated by 3D growth. In comparison to the 3D spheroids of parental CHO, we observed a significantly elevated 5-FU resistance in 3D culture of the CEAE. Eftekhar F. Naghibalhossaini (&) Department of Biochemistry, School of Medicine, Shiraz University of Medical Sciences, Zand Street, Shiraz, Iran e-mail: [email protected] F. Naghibalhossaini Autoimmune Research Center, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran
expressing CHO transfectants. Our findings suggest that the CEA level may be a suitable biomarker for predicting tumor response to 5-FU-based chemotherapy in CRC. Keywords CEA Chemotherapy Colorectal cancer Drug resistance 5-Fluorouracil
Introduction Colorectal cancer (CRC) is one of the most commonly occurring malignancies and one of the leading causes of cancer related death globally [1]. The primary treatment for CRC is surgery and chemotherapy is used complementary to decrease the risk of local recurrences. 5-Fluorouracil (5FU) in combination with other drugs has long been used in CRC chemotherapy and in a number of other common malignancies such as breast, esophageal, and gastric cancers [2]. However, tumors resistance in the course of treatment is an important cause of failure in clinical colon cancer therapy. In fact, response rates ranging 20–30 % have been reported in metastatic patients [3]. A goal of cancer therapy is to identify molecular markers to predict response and toxicity to drugs in order to facilitate the individualization of patient treatment. A large number of studies have attempted to define therapeutic and molecular determinants of interindividual differences in responses to the 5-FU based chemotherapy. The lack of response to 5-FU has been related to an increased expression of the target thymidylate synthase, and a decreased level of 5-FU-activating enzymes, including orotate phosphoribosyl-transferase and uridine kinase [4, 5]. Carcinoembryonic antigen (CEA), first described by Gold and Freeman in 1965, is a tumor associated antigen that is produced by a wide variety of human cancers
123
460
including colorectal, breast and lung tumors [6]. CEA is upregulated in approximately 90 % of advanced CRC and is associated with increased tumorigenicity [7, 8]. Elevated preoperative serum CEA levels have been correlated with high recurrence rates of CRC [9, 10]. In addition to its use as a tumor marker for colon carcinomas, elevated expression of CEA has been implicated in various biological aspects of neoplasia, including tumor cell adhesion, metastasis, the blocking of cellular differentiation and antiapoptosis functions [6]. Clinical observations have suggested that serum and tumor CEA concentration were associated with poor tumor response to chemoradiotherapy and poor outcome [11, 12]. Some previous studies have reported that CEA overexpression significantly protected tumor cells from undergoing anoikis [13–15], a form of apoptosis caused by detachment from cell matrix. A CEA-targeted ribozyme in human HT-29 colon cancer cells has been also shown to increase apoptosis in response to various apoptotic stimuli including UV irradiation, IFN-c treatment, and 5-FU treatment [16]. However, limited data are available regarding the correlation between cellular CEA expression levels and resistance to chemotherapy agents. Therefore, the goal of this study was to explore the relationship between CEA expression levels and resistance to anticancer drugs in human CRC cells.
Materials and methods
Mol Biol Rep (2014) 41:459–466
viability was determined by MTT dye reduction assay as described previously [17]. Briefly, 20 ll of the cellcounting solution (MTT solution) was added to each well to a final concentration of 0.5 mg/ml and incubated in a humidified 5 % CO2 atmosphere at 37 °C for 4 h. The crystals were dissolved in 150 ll of DMSO/well. The absorbance of the solution was read at 570 nm using a microtiter plate reader (Mikura, Englad). Cell viability was calculated according to the following formula: Cell viability (%) = OD570 (sample)/OD570 (control) 9 100. IC50 determination The IC50 was determined from the dose–response graph. Analysis of the dose–response curve was performed using the Software GraphPad PRISM Version 5.00 (GraphPad Software, San Diego, CA). Spheroid cultures Multicellular tumor spheroids were generated using a previously described method with minor modifications [18]. Briefly, a suspension of subconfluent monolayercultured tumor cells was plated over 1.5 % agarose in medium-coated 96-well plates at a density of *1,200 cells per well. After 5–7 days incubation in the incubator at 37 °C, three-dimensional multicellular spheroids had formed and were analyzed for 5-FU drug treatment. Spheroid experiments were done in three replicates and repeated at least two times.
Chemicals Plasmids and transfections All chemicals were purchased from Sigma (Gillingham, UK) unless otherwise stated. Reagents were prepared and stored according to the manufacturers’ instructions. Cell culture and drug treatment The Chinese hamster ovary (CHO) cell line and the human colon carcinoma cell lines SW742, LS180, HT29/219, and Caco2 were obtained from the National Cell Bank of Iran (NCBI, Pasteur Institute, Tehran). Cells were cultured in either RPMI 1640 or in DMEM supplemented with 10 % fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, and 100 lg/ml streptomycin (all from Gibco) in a humidified 5 % CO2 atmosphere at 37 °C. To determine IC50, a triplicate aliquot of cells were seeded in 96-well microtiter plates (3 9 103 cells/well in 100 ll medium). Twenty-four hours after seeding, 5-FU, methotrexate, and cisplatin solutions were added at desired concentrations. The medium of control cells was replenished with medium without chemicals. The cells were incubated at 37 °C for 72 h in a humidified incubator. Cell
123
The plasmid pcDNA3.1 (?) (Invitrogen, USA) containing human CEA cDNA was constructed by our laboratory as previously described [19]. The CHO cells were seeded into 3 9 10 cm petri dishes and stably transfected at 80 % confluence with pcDNA3.1 (?) containing full length CEA cDNA by calcium phosphate co-precipitation. The human colon carcinoma SW742 cells were transfected by electroporation as follows. Briefly, 106 cells in 400 ll of medium were mixed with 20 lg of pcDNA3.1 (?)/CEA. Electroporation was performed by using 1200 V and 25 lF. Stable CHO and SW742 transfectants were selected by incubating cells in Geneticin (G418 sulfate) (400 lg/ml) for 15 days. G418-resistant colonies were pooled and enriched for stable transfectants. Cell lysate preparation and immunoblotting Monolayer cultures of cells were removed from plastic surfaces by scrapping in whole-cell lysis buffer (50 mM Tris–HCl, pH 7.5, 0.5 M NaCl, and 1 % NP-40)
Mol Biol Rep (2014) 41:459–466
supplemented with antiprotease cocktail (Roche Applied Science, Germany). Lysates were stored frozen at -80 °C until use. 30–50 lg of total protein was subjected to electrophoresis using 7.5 % SDS-PAGE. The gels were transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore, MA, USA) and analyzed for CEA as described [20]. The blots were probed with anti-actin antibody as an internal control of protein loading. CEA assay Cell lysates prepared as described above and CEA content of cells was determined using a commercially available CEA ELISA kit (CanAg Diagnostics AB, Gothenburg, Sweden), according to the procedure provided by the manufacturer. The total cellular protein was determined according to Bradford method [21] using crystalline BSA as a standard and the CEA content was normalized to nanograms per milligram cellular protein.
Results Expression of CEA in CRC and CHO cell lines We examined the role of CEA in mediating the chemoresistance of cells to anticancer drugs by transfecting an expression vector (PCDNA3.1) containing the human CEA cDNA into CHO and SW742 human colon carcinoma cells. These cell lines were chosen as targets because CHO cell line does not contain CEA gene and SW742 human carcinoma cells express no detectable CEA protein. The transfection experiments were repeated three times. Fifteen to eighteen geneticin-resistant colonies of CHO and SW742 cells were pooled and their lysates were examined for CEA expression. A significant increase of CEA expression was detected in crude cell extracts by ELISA assay. By immunoblot analysis, a band corresponding to the recombinant CEA was only detected in the transfected cells, with an apparent size of 180 kD (Fig. 1). Maximum CEA expression levels measured in crude cell extracts in
Fig. 1 Western blot analysis of CEA expression in human CRC cell line SW742 and in CHO-cells transfected with human CEA gene The right panel indicates blot of two different pools of CHO–CEA transfectants with different levels of CEA expression
461
CHO and SW742 transfectants were 70 and 13.02 ng of CEA protein per mg total protein content at sub-confluent stage, respectively. Under the same conditions, in the control parental CHO and SW742 cells, no detectable expression of CEA was observed (Fig. 1). The growth rate of the transfected cells did not differ from the growth rate of the wild-type cells (data not shown). Sensitivity of CEA-transfected cell lines to anticancer drugs Stably transfected SW742 and CHO cells (3,000 cells/well in 96-well plates) were exposed to graded concentrations of 5-FU on day 1 after seeding. After 72 h, cell proliferation was assessed by MTT dye reduction assay and relevant IC50 values (50 % reduction in cell viability) determined. Increasing the 5-FU concentrations caused progressively greater cytotoxicity and growth inhibition. The IC50 value in the wild type CHO and SW742 were 0.41 and 21.9 lM, respectively (Table 1). Both transfected cell lines were more resistant to the cytotoxic action of 5-FU when compared to the wild type cells, having a 1.8 to 5-fold increase in the IC50 value. For SW742 transfectants, the IC50 value for 5-FU treatment in low and high CEA expresser was 40.2 ± 8.6 and 44.4 ± 11.75 lM, respectively and the IC50 for 5-FU treatment in low and high CEA expressing CHO cells was 1.06 ± 0.36 and 2.15 ± 0.56 lM, respectively (Table 1). We next examined the specificity of CEA effect on drug resistance by evaluating the sensitivity of CEA-transfected CHO and SW742 cells to cisplatin and methotrexate, two other commonly used anticancer drugs with different mechanisms of action. Cells (3,000 cells/well) were seeded in 96-well plates. 24 h later, cisplatin and methotrexate were added at the concentrations 0.2–12.5 and 0.02–2.5 lM, respectively, for 72 h. As shown in Table 1, there was no significant difference in sensitivity to cisplatin and methotrexate between wild type parental and their CEA-transfected cells. Sensitivity of cell lines to chemotherapy by 5-FU Since our DNA transfection studies clearly demonstrated that CEA production confers resistance to 5-FU, we compared the effect of 5-FU on a panel of 4 CRC cell lines (SW742, HT29/219, LS180 and Caco-2) with a wide range of CEA production. Our previous work has determined that LS180 has very high level of CEA expression and the other two well-differentiated cell lines, Caco-2 and HT29/219 express relatively moderate levels of CEA, but SW742 cells express no detectable level of CEA protein [20]. Three thousand cells were seeded in a 96-well plate. Twenty-four hours later cells were treated with graded
123
462
Mol Biol Rep (2014) 41:459–466
Table 1 Sensitivity of CHO/CEA and SW742/CEA transfectants to anticancer drugs Cell lines
CEA protein content (ng/mg total protein)b
IC50 (5-FU) (lM)a Mean ± SD
IC50 (cisplatin) (lM)a Mean ± SD
IC50 (methotrexate) (lM)a Mean ± SD
CHO
0
0.41 ± 0.22
1.46 ± 0.23
0.014 ± 0.01
CHO/CEA-1
70
2.15 ± 0.56
1.12 ± 0.32
0.014 ± 0.00
CHO/CEA-2
26.96
1.06 ± 0.36
ND
ND
SW742
0
21.9 ± 8.04
1.28 ± 0.08
0.07 ± 0.01
SW742/CEA-1
13.02
44.4 ± 11.75
0.94 ± 0.19
0.08 ± 0.01
SW742/CEA-2
9.6
40.2 ± 8.6
ND
ND
ND not determined a
IC50 value for each transfectant is presented from three independent assays
b
Data indicate the CEA expression in parental cell lines and in two independent pools of their CEA transfectants (CEA-1and CEA-2)
Table 2 CEA expression levels and growth inhibition of colon cancer cell lines by anti-cancer drug 5-FU Cell line
CEA expression level (ng/mg protein)
IC50 (lM)a Mean ± SD
LS180
174.3 ± 16.6
9.64 ± 4.33
20.2 ± 0.3
8.67 ± 3.17
5.7 ± 1.7
15.7 ± 1.01
HT29/219 Caco2 SW742
Undetectable
21.9 ± 8.04
a
IC50 value is presented as a mean of two independent assays, each done in triplicate
HT29/219 cancer cell lines (Table 2). Similar results on cell death were obtained by trypan blue staining (data not shown). Antigenic heterogeneity within cell lines and variations in 5-FU chemosensitivity
Fig. 2 Sensitivity of CRC cell lines to growth inhibition by anticancer agent 5-FU Cell viability was determined using MTT assay after treatment of cells with various concentrations of anticancer agents at 37 °C for 72 h. Data are means of three experiments, each done in triplicate. a The linear scale of X-axes; b The log scale of X-axes
concentrations of 5-FU. MTT assay results indicated that all the cells undergo growth inhibition in a dose-dependent manner (Fig. 2). However, there was no direct correlation between 5-FU resistance and cellular CEA expression levels in these cell lines. The IC50 of 5-FU was 21.9 ± 8.04 lM for SW742, 9.64 ± 4.33 lM for LS180, 15.7 ± 1.01 lM for Caco-2, and 8.67 ± 3.17 lM for
123
Human CRC cell lines are composed of heterogeneous population of cells that express different amounts of CEA antigen [22]. Considering the heterogeneity of cells within each cell line in terms of CEA expression, we investigated the relationship between 5-FU resistance and CEA expression levels of cellular subpopulations in two CRC cell lines. To elucidate antigen expression profile of cells after selective resistance to chemotherapy, HT29/219 and LS180 cells were treated with 2 different concentrations of 5-FU and CEA levels were measured in CRC cells that survived chemotherapy. For this purpose 2 9 105 cells were seeded in T25 flasks. Twenty-four hours later, 5-FU was added at the concentrations indicated for 72 h and protein extracts of surviving cells were examined for CEA expression by ELISA assay (Table 3). The results indicated that 5-FU resisting cells display higher CEA contents than control cells (unexposed to 5-FU) and cells survived exposure to high concentrations of 5-FU display higher CEA expression than cells surviving exposure to low 5-FU
Mol Biol Rep (2014) 41:459–466
463
Table 3 CEA protein expression in 5-FU resistant subpopulations of CRC cells Cell lines
5-FU conc (lM) 0
LS180
HT29/219
ng CEA/mg proteina,c
ng CEA/mg proteinb,c
Table 4 Expression of CEA in 2D and 3D culture of CRC cells and growth inhibition by 5-FU Cell lines
174.3 ± 16.6
20
187.6 ± 18.3
40 0
223.8 ± 19.8 20.2 ± 0.3
12
34.6 ± 2.3
24
37.5 ± 10.2
192 ± 15.9
32.5 ± 3.1
CEA expression (ng/mg protein)a
IC50 (lM)b
2D culture
2D culture
3D culture
LS180
174.3 ± 16.6
207.8 ± 20.3
9.6 ± 4.3
12.1 ± 5.3
HT29/ 219
20.2 ± 0.3
36.3 ± 2.3
8.6 ± 3.1
25.2 ± 1.4
0.4 ± 0.22
1.77 ± 0.82
2.15 ± 0.56
5.7 ± 1.76
CHO
0.00
0.00
70
68
a
CHO/ CEA1
b
a, b
CEA expression levels after growing in medium containing 5-FU drug
CEA expression levels after growing the 5-FU pre-treated cells for two passages in 5-FU-free medium
3D culture
Mean ± SD, mean values were calculated from two independent experiments, each done in triplicate
c
Mean ± SD, mean values were calculated from two independent experiments, each done in triplicate
concentration. To verify whether the 5-FU-mediated increase of CEA was a stable characteristic, we re-cultivated the 5-FU pre-treated cells in 5-FU-free medium. The CEA expression did not change after growing cells for two passages in 5-FU-free medium (Table 3). These results suggest that within each cell line the subpopulations of cells that have retained high-level of CEA expression resist chemotherapy and survive longer. Multicellular spheroid culture induce CEA expression and 5-FU resistance in CRC cells Numerous studies have reported that the formation of multicellular aggregates of tumor cells (3D culture) increase their resistance to chemotherapy agents [23–25]. To investigate the effect of spheroid culture on the chemosensitivity of cells to 5-FU, HT29/219, SW742, LS180, and Caco-2 cells as well as CHO and CHO/CEA transfectants were cultured under 3D culture condition as described in ‘‘Materials and methods‘‘ section. SW742 and Caco-2 cells were unable to form spheroids, but remained
as free-floating single cells and died after a few days. However, on day 5–7 after seeding, LS180, HT29/219, and CHO/CEA transfectants produced large, approximately 500 lm aggregates (Fig. 3) and 5-FU added as described above. In comparison to 2D monolayer culture, the IC50 value of multicellular spheroids of LS180 and HT29/219 increased by 1.3 times (IC50 = 12.1 ± 5.3 lM) and 3 times (IC50 = 25.2 ± 1.4 lM), respectively (Table 4). Interestingly, the multicellular spheroids of CHO/CEA transfectant was also more resistant to 5-FU (IC50 = 5.7 ± 1.76 lM) than multicellular aggregates of untransfected CHO cells (IC50 = 1.77 ± 0.82 lM). We then measured the CEA expression in multicelluar aggregates of CRC cells. As shown in Table 4, CEA expression was upregulated in LS180 and HT29/219 spheroids by 20 and 80 %, respectively, compared with monolayer culture of these cells (P \ 0.05).
Discussion The usual treatment of CRC is surgery, which in many cases is followed by chemotherapy using 5-FU and its derivatives to reduce tumor recurrence. However, human
Fig. 3 Micrograph of LS80 and HT29/219 spheroid
123
464
colon carcinomas are heterogeneous at the molecular and cellular levels and a considerable heterogeneity in the response to chemotherapy are observed in clinical studies [3]. The identification of molecular variables that are involved in tumor resistance to chemotherapy is of major interest in predicting treatment response and avoiding adverse events. CEA is a widely used tumor marker for monitoring of the course of CRC progression and a popular molecular target for cancer immunotherapy. Clinical studies suggest that CEA overexpression is associated with poor clinical outcome and reduced survival of colon cancer patients [26, 27]. Soeth et al. [16] reported that a CEA-targeted ribozyme that silenced CEA expression in human HT-29 colon cancer cells increases apoptosis in response to various apoptotic stimuli including 5-FU treatment. However, no previous studies have directly examined the effect of CEA overexpression on resistance of CRC cells to anticancer drugs. Therefore, the aim of this study was to determine the correlation between CEA expression levels and response to treatment with anticancer drugs in CRC cells. For this purpose the full length CEA cDNA, cloned in the eukaryotic expression vector pcDNA3.1, was transfected into human colon carcinoma SW742 and into CHO cell line. Transfection experiments were done in three replicates and repeated at least two times. We obtained several pools of CEA-transfectants that expressed different levels of the transfected cDNA. Several of the moderate- and high-expressing pools of clones were examined for drug resistance. CEA transfected cells were more resistant to 5-FU than untransfected parental cells (Table 1). Moreover, there was a significant correlation between the CEA expression levels in transfected cells and their resistance to 5-FU cytotoxity. When we examined the cytotoxic effects of methotrexate and cisplatin, the sensitivity of SW742and CHO–CEA transfectants was not significantly different from that of parental cells. The results suggest that the selective resistance to 5-FU cytotoxicity was a direct effect of the elevated CEA expression in the transfected cells rather than a general effect due to selection process. Results presented here confirm and extend the findings of Soeth et al. [16] who demonstrated that decreasing CEA expression in HT-29 cells by an inducible hammerhead ribozyme to CEA, increases apoptosis in response to5-FU. When we compared the effect of 5-FU on a panel of four CRC cell lines, the correlation between CEA expression and sensitivity to 5-FU did not bear a simple relationship (Table 2). Among the cell lines examined, the maximal IC50 value for 5-FU induced cytotoxicity was observed with SW742 cells, in which no CEA protein was detected. Obviously, other differences between human colon carcinoma cells other than CEA expression levels contribute to the drug resistance in neoplastic cells. These include
123
Mol Biol Rep (2014) 41:459–466
mechanisms involved in drug influx and efflux, processing of drug-induced damage, drugs inactivation, and induction of cell apoptosis etc. The genotypes and activity of several enzymes including thymidylate synthase, thymidine phosphorylase, and MTHFR enzyme have been shown previously to correlate with cancer response to 5-FU in vitro and in vivo [4, 28, 29]. On the basis of the observed heterogeneity within CRC cell lines in terms of the amount of CEA expression [22], we anticipated that the high CEA-expressing fraction of CRC cell lines would be more resistant and selected by 5-FU treatment. To test this hypothesis, we cultured two CRC cell lines HT29/219 and LS180 with two different concentrations of 5-FU and the cellular CEA contents were analyzed before and after 5-FU treatment. In both cell lines, surviving cells displayed an elevated CEA expression in comparison to untreated cells and the fractions survived at high drug concentration expressed higher levels of CEA protein than cells survived at low drug concentration (Table 3). Some authors previously suggested that treatment of neoplastic cell lines with 5-FU may induce CEA up-regulation transiently [30, 31]. When we re-cultivated the 5-FU-pretreated cells for two passages in 5-FU-free medium, the CEA expression did not decline. Moreover, previous investigations showed that 5-FU was not able to induce CEA expression in normal or in CEA-negative tumor cells [31, 32]. These observations support the hypothesis that the mechanism of 5-FU-mediated increase of cellular CEA levels might be due to the selection process rather than drug-induced CEA expression. Sorbye and Dahl also reported a transient elevation of serum CEA during FOLFOX therapy in CRC patients [33]. But, they did not measure CEA expression in tumor tissues. There are many factors affecting serum CEA concentrations in cancer patients with CRCs [34]. Indeed, the plasma CEA values in response to chemotherapy may show an initial delay or even a transient surge due to cytolysis in patients responding to chemotherapy before demonstrating the expected pattern of change [35]. Human tumors, although often treated as a single entity, comprise a population of cells with heterogeneous expression of several surface molecules including CEA antigen [22, 28, 36]. Therefore, 5-FU treatment may result in the selection of high CEA-expressing tumor cells that would tolerate high drug concentrations. Additional experimental and clinical studies are needed to determine the correlation between levels of CEA expression and dose of 5-FU drug required for treatment of CEA expressing tumors. Three dimensional cultures of cancer cells are better representation of solid tumors when compared to monolayer cultures. In agreement with previous studies [37, 38], we
Mol Biol Rep (2014) 41:459–466
found that 3D growth of CRC cells was associated with an increased 5-FU resistance (LS180: IC50 = 12.1 ± 5.3 lM, 1.3-fold increase; HT29/219: IC50 = 25.2 ± 1.4 lM, threefold increase). As shown in Table 4, CEA levels are increased in 3D culture of CRC cell lines when compared to monolayer cultures of these cells. Compared with monolayer culture, the 3D culture of LS180 and HT29/219 cells resulted in a 20 and 80 % increase in CEA expression levels, respectively. This observation is similar to another investigation which found that the CEA expression is stimulated by 3D growth of poorly differentiated MIP-101 human colorectal carcinoma cells that do not produce CEA in monolayer cultures [39]. It has been reported that it is more difficult for drug to penetrate into the center cell mass of 3D spheroids. Some prior studies have suggested that 5-FU specifically targets proliferating cells, and thus would not kill the quiescent cells in the spheroids. Whereas in 2D monolayer cultures, cells proliferate at a faster rate and thus 5-FU inhibits cellular growth more effectively [40]. In our study, the 3D spheroids of CHO/CEA transfectants were significantly more resistant to 5-FU when compared to untransfected parental CHO cells cultured under 3D condition. Although above mentioned factors are likely to play some roles in increasing chemo-resistance of multicellular spheroids of cells, other factors, such as up-regulated CEA expression and cell–cell interactions may also influence the survival of aggregated CRC cells to cytotoxic drugs. The actual mechanisms underlying CEA-mediated 5-FU resistance is unknown. It has been reported that CEA can protect tumor cells from undergoing anoikis, i.e., apoptosis induced by loss of cell contact with the extracellular matrix, by directly binding to TRAIL-R2 (DR5) [13, 15]. Previous analysis of 273 genes in HT29 cells by microarray experiment also revealed that CEA down-regulate expression of various groups of cancer-related genes, in particular apoptotic genes [16]. Definition of the biological characteristics of resistant tumor cells is important for developing treatment strategies and selecting patients who are most likely to benefit from the treatment. CEA is an important biomarker in the diagnosis and monitoring of relapse in CRC. We believe that the results of our study offer useful information regarding the potential use of CEA expression levels as a marker for tumor response to fluorouracil-based chemotherapy in CRC. Acknowledgments This study was a part of the dissertation of Ebrahim Eftekhar, submitted to Shiraz University of Medical Sciences in partial fulfillment of the requirements for the Ph. D in biochemistry. This work was supported by a grant from the Vice Chancellor for Research, Shiraz University of Medical Sciences.
465
References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61(2):69–90 2. Malet-Martino M, Jolimaitre P, Martino R (2002) The prodrugs of 5-fluorouracil. Curr Med Chem Anticancer Agents 2(2):267–310 3. Isacoff WH, Borud K (1997) Chemotherapy for the treatment of patients with metastatic colorectal cancer: an overview. World J Surg 21(7):748–762 4. Furuta T (2009) Pharmacogenomics in chemotherapy for GI tract cancer. J Gastroenterol 44(10):1016–1025 5. Peters GJ, Backus HH, Freemantle S, van Triest B, CodacciPisanelli G, van der Wilt CL, Smid K, Lunec J, Calvert AH, Marsh S, McLeod HL, Bloemena E, Meijer S, Jansen G, van Groeningen CJ, Pinedo HM (2002) Induction of thymidylate synthase as a 5-fluorouracil resistance mechanism. Biochim Biophys Acta 1587(2–3):194–205 6. Chan CH, Stanners CP (2007) Recent advances in the tumour biology of the GPI-anchored carcinoembryonic antigen family members CEACAM5 and CEACAM6. Curr Oncol 14(2):70–73 7. Goldstein MJ, Mitchell EP (2005) Carcinoembryonic antigen in the staging and follow-up of patients with colorectal cancer. Cancer Invest 23(4):338–351 8. Ilantzis C, DeMarte L, Screaton RA, Stanners CP (2002) Deregulated expression of the human tumor marker CEA and CEA family member CEACAM6 disrupts tissue architecture and blocks colonocyte differentiation. Neoplasia 4(2):151–163 9. Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, Somerfield MR, Hayes DF, Bast RC Jr (2006) ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 24(33):5313–5327 10. Park IJ, Kim HC, Yu CS, Yoo JH, Kim JC (2005) Cutoff values of preoperative s-CEA levels for predicting survivals after curative resection of colorectal cancer. J Korean Med Sci 20(4):624–627 11. Lin JZ, Zeng ZF, Wu XJ, Wan DS, Chen G, Li LR, Lu ZH, Ding PR, Pan ZZ (2010) Phase II study of pre-operative radiotherapy with capecitabine and oxaliplatin for rectal cancer and carcinoembryonic antigen as a predictor of pathological tumour response. J Int Med Res 38(2):645–654 12. Li M, Li JY, Zhao AL, He JS, Zhou LX, Li YA, Gu J (2009) Comparison of carcinoembryonic antigen prognostic value in serum and tumour tissue of patients with colorectal cancer. Colorectal Dis 11(3):276–281 13. Ordonez C, Screaton RA, Ilantzis C, Stanners CP (2000) Human carcinoembryonic antigen functions as a general inhibitor of anoikis. Cancer Res 60(13):3419–3424 14. Chan CH, Camacho-Leal P, Stanners CP (2007) Colorectal hyperplasia and dysplasia due to human carcinoembryonic antigen (CEA) family member expression in transgenic mice. PLoS ONE 2(12):e1353 15. Samara RN, Laguinge LM, Jessup JM (2007) Carcinoembryonic antigen inhibits anoikis in colorectal carcinoma cells by interfering with TRAIL-R2 (DR5) signaling. Cancer Res 67(10):4774–4782 16. Soeth E, Wirth T, List HJ, Kumbhani S, Petersen A, Neumaier M, Czubayko F, Juhl H (2001) Controlled ribozyme targeting demonstrates an antiapoptotic effect of carcinoembryonic antigen in HT29 colon cancer cells. Clin Cancer Res 7(7):2022–2030 17. Li HL, Wang TS, Li XY, Li N, Huang DZ, Chen Q, Ba Y (2007) Overexpression of cyclin L2 induces apoptosis and cell-cycle arrest in human lung cancer cells. Chin Med J (Engl) 120(10):905–909
123
466 18. Friedrich J, Seidel C, Ebner R, Kunz-Schughart LA (2009) Spheroid-based drug screen: considerations and practical approach. Nat Protoc 4(3):309–324 19. Naghibalhossaini F, Pakdel A, Ghaderi AA, Saberi Firoozi M (2005) Effective production of carcinoembryonic antigen by conversion of the membrane-bound into a recombinant secretory protein by site-specific mutagenesis. Pathol Oncol Res 11(4):211–217 20. Pakdel A, Naghibalhossaini F, Mokarram P, Jaberipour M, Hosseini A (2012) Regulation of carcinoembryonic antigen release from colorectal cancer cells. Mol Biol Rep 39(4):3695–3704 21. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 22. Han JS, Nair PP (1995) Flow cytometric identification of cell surface markers on cultured human colonic cell lines using monoclonal antibodies. Cancer 76(2):195–200 23. Green SK, Francia G, Isidoro C, Kerbel RS (2004) Antiadhesive antibodies targeting E-cadherin sensitize multicellular tumor spheroids to chemotherapy in vitro. Mol Cancer Ther 3(2):149–159 24. St Croix B, Kerbel RS (1997) Cell adhesion and drug resistance in cancer. Curr Opin Oncol 9(6):549–556 25. Smalley KS, Lioni M, Herlyn M (2006) Life isn’t flat: taking cancer biology to the next dimension. In Vitro Cell Dev Biol Animal 42(8–9):242–247 26. Ballesta AM, Molina R, Filella X, Jo J, Gimenez N (1995) Carcinoembryonic antigen in staging and follow-up of patients with solid tumors. Tumour Biol 16(1):32–41 27. Jantscheff P, Terracciano L, Lowy A, Glatz-Krieger K, Grunert F, Micheel B, Brummer J, Laffer U, Metzger U, Herrmann R, Rochlitz C (2003) Expression of CEACAM6 in resectable colorectal cancer: a factor of independent prognostic significance. J Clin Oncol 21(19):3638–3646 28. Bandres E, Zarate R, Ramirez N, Abajo A, Bitarte N, GariiaFoncillas J (2007) Pharmacogenomics in colorectal cancer: the first step for individualized-therapy. World J Gastroenterol 13(44):5888–5901 29. Longley DB, Johnston PG (2005) Molecular mechanisms of drug resistance. J Pathol 205(2):275–292 30. Prete SP, Turriziani M, Massara MC, De Rossi A, Correale P, De Vecchis L, Torino F, Bonmassar L, Aquino A (2008) Combined effects of 5-fluorouracil, folinic acid and oxaliplatin on the
123
Mol Biol Rep (2014) 41:459–466
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
expression of carcinoembryonic antigen in human colon cancer cells: pharmacological basis to develop an active antitumor immunochemotherapy. J Exp Clin Cancer Res 270:5 Aquino A, Prete SP, Guadagni F, Greiner JW, Giuliani A, Orlando L, Masci G, De Santis S, Bonmassar E, Graziani G (2000) Effect of 5-fluorouracil on carcinoembryonic antigen expression and shedding at clonal level in colon cancer cells. Anticancer Res 20(5B):3475–3484 Cappelletti D, Cardillo A, Bonanno E, Prete SP, Cucchiara G, Turriziani M, Greiner JW, Cottarelli A, Breda E, Aquino A, Bonmassar E, De Vecchis L (2000) Drug-induced modulation of carcinoembryonic antigen (CEA) expression in neoplastic cells from a patient with rectal cancer. J Exp Clin Cancer Res 19(4):467–469 Sørbye H, Dahl O (2004) Transient CEA increase at start of oxaliplatin combination therapy for metastatic colorectal cancer. Acta Oncol 43(5):495–834 Duffy MJ (2001) Carcinoembryonic antigen as a marker for colorectal cancer: is it clinically useful? Clin Chem 47(4):624–630 Bidart JM, Thuillier F, Augereau C, Chalas J, Daver A, Jacob N, Labrousse F, Voitot H (1999) Kinetics of serum tumor marker concentrations and usefulness in clinical monitoring. Clin Chem 45(10):1695–1707 Leteurtre E, Gouyer V, Rousseau K, Moreau O, Barbat A, Swallow D, Huet G, Lesuffleur T (2004) Differential mucin expression in colon carcinoma HT-29 clones with variable resistance to 5-fluorouracil and methotrexate. Biol Cell 96(2):145–151 Graham CH, Kobayashi H, Stankiewicz KS, Man S, Kapitain SJ, Kerbel RS (1994) Rapid acquisition of multicellular drug resistance after a single exposure of mammary tumor cells to antitumor alkylating agents. J Natl Cancer Inst 86(13):975–982 Sutherland RM (1988) Cell and environment interactions in tumor microregions: the multicell spheroid model. Science 240(4849):177–184 Jessup JM, Brown D, Fitzgerald W, Ford RD, Nachman A, Goodwin TJ, Spaulding G (1997) Induction of carcinoembryonic antigen expression in a three-dimensional culture system. In Vitro Cell Dev Biol Animal 33(5):352–357 Tung YC, Hsiao AY, Allen SG, Torisawa YS, Ho M, Takayama S (2011) High-throughput 3D spheroid culture and drug testing using a 384 hanging drop array. Analyst 136(3):473–478
