Abstract
Background/Aim: Riboflavin transport in enterocytes is mediated by three translocators: RFVT3 located on the apical membrane, and RFVT1 and RFVT2 on the basolateral membrane. The aim of this study was to investigate whether the expression levels of RFVTs are altered in human colorectal cancer (CRC). Materials and Methods: In human colon adenocarcinoma cell lines (CaCo2, DLD-1, HT-29) and in tissues of patients with CRC, gene and protein expression levels were evaluated by real time-polymerase chain reaction and western blotting. Intracellular flavin content was determined by high-performance liquid chromatography. Results: RFVT3 and RFVT2 gene and protein expression levels were higher in DLD-1 and HT-29 compared to Caco2 cells. In HT-29 cells, the RFVT1 protein level was drastically lower. These differences are presumably responsible for the higher total flavin content in DLD-1 and HT-29 cells. In tumor tissues of patients with CRC, RFVT1 content was reduced at both protein and mRNA levels compared to normal mucosa. RFVT3 and RFVT2 gene expression levels were increased, while protein expression was reduced, with a small reduction in riboflavin amount. Conclusion: This study provides first evidence that transcription/translation of RFVTs are profoundly altered in CRC.
Riboflavin, otherwise known as vitamin B2, is an essential dietary component and represents the precursor of flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), that are important enzymatic cofactors required for carbohydrate, amino acid and lipid metabolism, and other cellular regulatory roles (1-5). Humans, being unable to synthesize riboflavin, must obtain it from food, in particular from milk, meats, fatty fish and green vegetables, and, to a lesser amount, from intestinal microflora; after intestinal absorption, mainly occurring in the small intestine, riboflavin is distributed from the blood to several tissues (1, 2, 6).
In different cells, riboflavin uptake occurs via specialized carrier-mediated processes supported by three specific members of the solute carrier family 52 (SLC52A), identified and named riboflavin transporter 1 (RFVT1; SLC52A1), RFVT2 (SLC52A2), and RFVT3 (SLC52A3), respectively (6-9). Alterations of these proteins were recently correlated with rare inherited neuromuscular diseases (10). Inside the cells, riboflavin is phosphorylated to FMN by riboflavin kinase (E.C. 2.7.1.26) and it is subsequently metabolized to FAD by FAD synthase (E.C. 2.7.7.2) (11, 12), which also participates in cofactor delivery to the appropriate apo-flavoenzymes during holoenzyme biogenesis (13).
At the apical membrane of enterocytes, riboflavin is absorbed by the action of RFVT3, which is expressed at a higher level compared with the other riboflavin transporters. The vitamin is then released into blood by RFVT2 and RFVT1, which are mainly localized at the basolateral membrane domain (2, 8, 14, 15).
Studies have shown that a low intake of dietary riboflavin can lead to negative health consequences, that include the development of several cancer such as colorectal cancer (CRC) (16, 17). Riboflavin nutritional status and riboflavin availability in intestinal cells have been shown to be relevant to normal cellular functions, including proliferation, growth and survival (18). Interestingly, plasma and tissue riboflavin levels were found decreased in patients with cervical intraepithelial neoplasia and cervical squamous cell carcinoma as compared with normal controls. In the tumor tissues of these patients, high RFVT3 expression levels were found and these levels were positively correlated with the tumor stage, suggesting a protective role of RFVT3 against riboflavin deficiency associated with cervical carcinogenesis (19). Different results were obtained in gastric cancer, where a decrease in RFVT3 mRNA and protein expression were reported (20). To date, in human CRCs, the profile of expression of RFVTs has been under investigated and it is still not known whether there are other factors, such as tumor stage and tumor grade differentiation, which can influence flavin absorption and transport.
Therefore, the aim of the present study was to evaluate gene expression and protein levels of RFVTs in three human colon adenocarcinoma cell lines with different degrees of differentiation: CaCo2 cell line (well-differentiated); DLD-1 cell line (well-differentiated with Dukes' C characteristics) and HT-29 cell line (moderately differentiated). Moreover, the gene and protein expression levels of RFVTs were also measured in tumor and in surrounding normal mucosa of patients with CRC. To gain further insight into intestinal homeostasis of flavins in cancer, the basal levels of riboflavin, FAD and FMN were also assessed both in three cell lines and in intestinal tissues of patients with CRC.
Materials and Methods
Cell culture conditions. The human colon adenocarcinoma cell lines CaCo2, DLD-1 and HT-29 were obtained from the Interlab Cell Line Collection (Genoa, Italy). The cells were seeded at a density of 2×105 cells/10 ml of RPMI 1640 medium for CaCo2 and DLD-1 cells, and McCoy's 5A medium for HT-29 cells in 100 mm tissue culture dishes (Corning Costar Co., Milan, Italy), supplemented with 10% fetal bovine serum, 1% non essential amino acids, 2 mM glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, in monolayer cultures, and incubated at 37°C in a humidified atmosphere containing 5% CO2 in air. At confluence, the grown cells were harvested and serially subcultured with a 1:4 split ratio. All cell culture components were purchased from Sigma Aldrich (Milan, Italy).
Patients. Twenty-four consecutive patients (nine men and 15 women, mean age=68.58±11.37 years) undergoing surgery for CRC were enrolled in the study. They all gave informed consent to take part in the study. Colorectal normal mucosa and cancer tissue were obtained from each according to a standardized procedure. Samples of mucosa were taken from macroscopically normal areas of intestine at 10 cm from the neoplastic lesion. Specimens were taken within 1 h after the surgical procedure and stored at −80°C until assayed. The laboratory analyses were performed blindly with respect to the clinical characteristics and histopathologic al features of the samples. Clinical and histopathological features of each patient were recorded and are reported in Table I.
RFVT gene expression analysis. Human RFVT mRNA levels were analyzed by real time-polymerase chain reaction (RT-PCR) assay. Total tissue RNA, isolated with TRI-Reagent (Mol. Res. Centre Inc., Cincinnati, OH, USA), was reverse transcribed in 20 μl final volume at 41°C for 60 min using 30 pmol antisense primer for human RFVT1, RFVT2, RFVT3 and β-actin gene (Table II). Real-time PCRs were performed in 25 μl final volume containing 2 μl of cDNA, master mix with SYBR Green (iQ SYBR Green Supermix Bio-Rad Laboratories, Milan, Italy) and sense and antisense primers for each target gene and the β-actin gene (Table II). Real-time PCR was carried out as previously described (21). All expression data were normalized using β-actin as an internal control. Gel electrophoresis was used to confirm the specificity of PCR products.
Protein extraction and western blotting analysis. Total proteins from cell lines and tissue samples were extracted using RIPA buffer (Sigma-Aldrich). After quantization of protein concentration by a standard Bradford assay (Bio-Rad Laboratories), equal amounts of protein (50 μg) were separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis and subsequently transferred onto a polyvinylidene difluoride membrane (Bio-Rad Laboratories). The primary antibodies were directed against the following proteins: SLC52A1-RFVT1, SLC52A2-RFVT2 (Thermo Fisher Scientific, Rockford, IL, USA), SLC52A3 (MyBioSource, San Diego, CA, USA), β-actin (Cell Signaling, Beverly, MA, USA) and α-tubulin (Sigma-Aldrich). After overnight incubation, a horseradish peroxidase-conjugated secondary antibody (Bio-Rad Laboratories) was used. The immunoreactive bands were visualized and analyzed using chemiluminescence detection system (ChemiDoc XRS apparatus and software; Bio Rad Laboratories) and the proteins detected were normalized against β-actin.
Determination of flavin levels. Flavin content was measured in human colon adenocarcinoma cell lines and in tissues of tumor and normal mucosa of 11 patients with CRC. Cell pellets and about 45 mg of tissue samples were re-suspended in 150 and 300 μl of lysis buffer respectively (50 mmol/l Tris-HCl pH 7.5, 1% Triton X-100, 5 mmol/l β-mercaptoethanol, 1 mmol/l NaF, 0.1 mmol/l phenylmethylsulfonyl fluoride) and protease-inhibitor cocktail (Roche Diagnostics GmbH, Mannheim, Germany; 1 tablet/10 ml of lysis buffer). After homogenization and centrifugation at 13000 × g for 10 min at 4°C, the supernatant was recovered as cell lysate and protein concentration was measured by a standard Bradford assay. Riboflavin, FMN, and FAD in neutralized perchloric acid extracts of cell lysates (0.45 mg protein) were measured by high-performance liquid chromatography (HPLC), essentially as previously described (22). Quantitative determinations of flavins were carried out with a calibration curve made in each analysis with standard solutions diluted in the extraction solution.
Statistical analysis. The significance of the differences between the three groups of experimental cells were evaluated with one-way analysis of variance (ANOVA) and Tukey's multiple comparison test. The differences in RFVT1, RFVT2 and RFVT3 mRNA levels, as well as the differences in flavin content, between normal colon mucosa and cancer were analyzed by paired t-test. The Mann–Whitney test was used to analyze the differences in RFVT mRNA and protein levels in relation to clinical parameters such as age, sex, tumor site, stage of disease and histological differentiation. Differences were considered statistically significant with a p-value of less than 0.05.
Results
Figure 1 shows mRNA levels of RFVTs in three human colon adenocarcinoma cell lines with a different degree of differentiation, namely CaCo2, DLD-1 and HT-29. In particular, no significant difference was detected in RFVT1 mRNA levels among the three cell lines (Figure 1A), whilst the levels of RFVT2 and RFVT3 gene expression were significantly higher in HT-29 cells than CaCo2 and DLD-1 cells (p<0.001 and p<0.0001, ANOVA and Tukey's multiple comparison test, respectively; Figure 1B and C). As regards the protein expression levels of RFVTs in cell lines, a lack of expression of RFVT1 was found in HT-29 cells (p<0.001 and p<0.0001, ANOVA and Tukey's multiple comparison test; Figure 2A). In concordance with gene expression, protein levels of RFVT2 and RFVT3 were also significantly higher in HT-29 with respect to CaCo2 and to DLD-1 cells (p<0.001 and p<0.0001, ANOVA and Tukey's multiple comparison test; Figure 2B and C). In addition, RFVT2 protein levels was significantly higher in DLD-1 compared to CaCo2 cells (p<0.0001, ANOVA and Tukey's multiple comparison test; Figure 2B).
In order to confirm that alteration in the expression of flavin transporters was associated with CRC, gene and protein expression levels were evaluated for RFVTs in tumor tissue and in surrounding normal mucosa from 24 patients with CRC, whose clinical and histopathological features are described in Table I. Figure 3 shows RFVT1 gene expression level to be significantly lower in cancer tissue compared to normal mucosa; on the contrary, levels of RFVT2 and RFVT3 gene expression were significantly higher in tumor tissue (p<0.0001, ANOVA and Tukey's multiple comparison test), in accordance with changes observed in cell models. Nevertheless, in the tissues, RFVT1 and RFVT2 protein expression was significantly lower in tumor as compared to normal mucosa (p<0.001 and p<0.0001, ANOVA and Tukey's multiple comparison test, respectively; Figure 4A and B); while a small reduction in RFVT3 protein levels was found, although it was not statistically significant (Figure 4C). No association was found between both mRNA and protein levels of RFVTs and age, sex, tumor site, disease stage and histological differentiation (data not shown).
To understand whether changes in the expression levels of RFVTs may alter the flavin content or its distribution in enterocytes, the basal levels of FAD, FMN and riboflavin were measured by HPLC in extracts obtained both human colon adenocarcinoma cell lines and tissue samples. Total flavin amounts in DLD-1 and HT-29 were significantly higher compared with CaCo2 cells (p<0.0001, ANOVA and Tukey's multiple comparison test; Figure 5A). In particular FAD content, which represents the major contribution to the total acid-extractable flavins in cells, was significantly enhanced in DLD-1 and HT-29 compared with CaCo2 cells as shown in Figure 5B (p<0.0001, ANOVA and Tukey's multiple comparison test). Moreover, a statistically significantly lower FMN content was found in DLD-1 compared to CaCo2 and HT-29 cell lines (p<0.0001, ANOVA and Tukey's multiple comparison test; Figure 5C), while no significant differences in riboflavin amount were found (Figure 5D).
In the tissues, from patients with CRC, the content of total flavins, as well as FMN, did not show any variation between tumor and normal mucosa (Figure 6A and C). FAD was the most representative flavin, but no statistically significant differences in its content were found between normal and tumor tissues (Figure 6B), despite the slightly lower RFVT3 content. Nevertheless, a small but statistically significant reduction in riboflavin amount was observed in cancer tissue compared to normal surrounding mucosa of patients with CRC (p<0.001, ANOVA and Tukey's multiple comparison test; Figure 6D).
Discussion
In the present study, we demonstrated, for the first time as far as we are aware, changes in expression levels of riboflavin transporters in human CRC.
In the cell models used, both mRNA and protein levels of RFVT3 and RFVT2 were higher in HT-29 compared to Caco2 and DLD-1 cells. This finding suggests an association between the increase in the expression level of RFVT3 and RFVT2 and the degree of tumor differentiation. Interestingly, the behavior of RFVT1 expression was different from that of the other two transporters: no significant difference in RFTV1 mRNA level was observed in any of the tested cell lines. This result is in line with a previous study, conducted on NCM460 cells maintained under riboflavin deficiency, in which no changes in RFVT1, but an increase of RFVT3 and RFVT2 expression levels were found (23). In addition, we found a loss of RFVT1 protein expression in HT-29 cells. RFVT1 is expressed at the basolateral membrane of intestinal cells, therefore the disappearance of this protein in HT-29 cells may be due either to a loss in cellular polarity or to a higher rate of RFVT1 protein degradation in these cells.
Moreover, in less differentiated cells, the total flavin amount was statistically higher, probably because of an adaptive mechanism of tumor cells, which apparently become greedy for this vitamin. Thus, tumor cells appear to increase the rate of flavin intake and, at the same time, reduce the rate of its export. In fact, we hypothesize that RFVT3 is the major factor responsible for the increase of the observed intracellular flavin content. Furthermore, the increase in flavin content also occurred in DLD-1 cells, where there was no decrease in RFVT1 expression level.
The increase of flavin content in DLD-1 and HT-29 cells is essentially linked to the contribution of FAD, the main enzymatic cofactor, which is synthesized by FAD synthase, operating at a rate which overcomes that of riboflavin transport (2). The decrease in FMN content observed in DLD-1 cells compared with the other cell lines might be explained by a concomitant variation in intracellular enzyme activities of riboflavin kinase/FAD synthase. Therefore, it can be hypothesized that at the transcriptional/post-transcriptional level, cancer induces conditions of intracellular stress, mimicking those occurring in the absence of the vitamin, leading to adaptation of flavin intestinal homeostasis.
In tumor tissues of patients with CRC, the adaptive response of RFVTs to tumorigenesis was confirmed. In particular, we found a reduction in both protein and gene expression levels of RFVT1, thus suggesting a reduced ability to export riboflavin in the blood, while a different behavior was observed between protein and gene expression of RFVT3 and RFVT2 in tumor and in surrounding normal mucosa. For both of these translocators, the protein expression was reduced in tumor, whereas mRNA levels were increased, suggesting a deregulation of the transcriptional and translational response of RFVTs to tumorigenesis. The high cell turnover, or deregulation in amino acid homeostasis, which occur in tumor tissues (24) may result in a reduction in the protein levels of these transporters. These findings fit well with the apparent balance in intracellular concentration of FAD and FMN, which was found not to be significantly changed in cancer tissues. The only sign of homeostasis derangement found in the tumor tissues compared to the surrounding mucosa of patients was the low intracellular level of riboflavin, probably deriving from an increased rate conversion into flavin cofactors.
In conclusion, we demonstrated that the transport of riboflavin through the intestinal barrier can be profoundly altered during cancer progression and this may explain the decrease in blood riboflavin level in patients (19, 20). The mechanisms involved, in both the transcriptional and post-transcriptional levels, are different for each transporter. Generally, the different behavior of the isoforms of these transporters in cell models, or, in the case of tissues, with respect to the normal mucosa, may reflect a marked alteration in the orientation/polarity of neoplastic cells (25).
Further studies are needed to understand the molecular mechanisms leading to deregulation of riboflavin transport and ascertain whether nutritional therapy with riboflavin may improve the status of patients with cancer.
Acknowledgements
The Authors thank V. Intini for her support in molecular experiments and M. A. Indiveri for linguistic consultancy. The helpful collaboration of G. Esperti, who participated as student in the early stages of this work, is also gratefully acknowledged.
This work was supported by PO Puglia FESR 2007-2013, Asse I, Linea 1.2. Accordo di Programma Quadro in materia di Ricerca Scientifica. Intervento Reti di Laboratori Pubblici di Ricerca, Progetto L.A.I.F.F.- RETE DI LABORATORI PER L'INNOVAZIONE NEL CAMPO DEGLI ALIMENTI FUNZIONALI (codice n. 47) to M.G.C. and also by PON01_00937- Modelli sperimentali Biotecnologici integrati per lo sviluppo e la selezione di molecole di interesse per la salute dell'uomo MIUR (Italian Ministry of Education, University and Research) to M.B.
- Received March 9, 2018.
- Revision received March 29, 2018.
- Accepted April 2, 2018.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved