Abstract Background Riboflavin is an essential component of the human diet and its derivative cofactors play an established role in oxidative metabolism. Riboflavin deficiency has been linked with various human diseases. Objective The objective of this study was to identify whether riboflavin depletion promotes tumorigenesis. Methods HEK293T and NIH3T3 cells were cultured in riboflavin-deficient or riboflavin-sufficient medium and passaged every 48 h. Cells were collected every 5 generations and plate colony formation assays were performed to observe cell proliferation. Subcutaneous tumorigenicity assays in NU/NU mice were used to observe tumorigenicity of riboflavin-depleted HEK293T cells. Mechanistically, gene expression profiling and gene ontology analysis were used to identify abnormally expressed genes induced by riboflavin depletion. Western blot analyses, cell cycle analyses, and chromatin immunoprecipitation were used to validate the expression of cell cycle–related genes. Results Plate colony formation of NIH3T3 and HEK293T cell lines was enhanced >2-fold when cultured in riboflavin-deficient medium for 10–20 generations. Moreover, we observed enhanced subcutaneous tumorigenicity in NU/NU mice following injection of riboflavin-depleted compared with normal HEK293T cells (55.6% compared with 0.0% tumor formation, respectively). Gene expression profiling and gene ontology analysis revealed that riboflavin depletion induced the expression of cell cycle–related genes. Validation experiments also found that riboflavin depletion decreased p21 and p27 protein levels by ∼20%, and increased cell cycle–related and expression-elevated protein in tumor (CREPT) protein expression >2-fold, resulting in cyclin D1 and CDK4 levels being increased ∼1.5-fold, and cell cycle acceleration. We also observed that riboflavin depletion decreased intracellular riboflavin levels by 20% and upregulated expression of riboflavin transporter genes, particularly SLC52A3, and that the changes in CREPT and SLC52A3 correlated with specific epigenetic changes in their promoters in riboflavin-depleted HEK293T cells. Conclusion Riboflavin depletion contributes to HEK293T and NIH3T3 cell tumorigenesis and may be a risk factor for tumor development. riboflavin, SLC52A3, CREPT, cell cycle, tumorigenesis Introduction Riboflavin (vitamin B-2) is a water-soluble vitamin that serves as a precursor for flavin mononucleotide and flavin adenine dinucleotide (1). Because flavin mononucleotide and flavin adenine dinucleotide act as electron carriers for a range of redox reactions, including the oxidative metabolism of macronutrients and electron transport chains (2), riboflavin deficiency leads to reduced activity of flavin-dependent enzymes, such as glutathione reductase (3), flavin adenine dinucleotide–dependent endoplasmic reticulum oxidoreductin 1 (4), and sulfhydryl oxidases (5). Reduced enzymatic activity affects critical in vivo metabolic pathways, such as the oxidative metabolism of carbohydrates, amino acids, and fatty acids (6, 7). Riboflavin cannot be synthesized in mammals and must be obtained from exogenous sources, including diet and the normal microflora of the large intestine (8). The effects of riboflavin deficiency on humans are difficult to determine because riboflavin deficiency in isolation (in the absence of other deficiencies or conditions) is unusual in populations in which riboflavin deficiency is endemic. Previous reports have indicated that riboflavin deficiency in infants and children in developing countries is associated with poor growth (9). In addition, lack of dietary riboflavin is associated with an increased risk for esophageal cancer (10–12). In vitro experiments have shown that riboflavin may have specific functions associated with cell fate determination, as well as implications for cell growth and development. In Jurkat lymphoid cells, riboflavin depletion has an inhibitory effect on cell viability (13). In contrast, in HepG2 human liver cells, riboflavin depletion leads to an unfolded protein/stress response (14) and interferes with normal progression of the cell cycle (15–17), likely affecting the expression of regulatory genes at both the transcriptional and proteomic levels (18). Recent studies have provided insight into how riboflavin depletion is associated with an increase in expression of SLC52A3 at the protein and mRNA levels, and corresponds to epigenetic changes that enhance SLC52A3 promoter activity (19, 20). The relation between riboflavin and cancer is intriguing, but many gaps remain in our knowledge. Several studies indicate that riboflavin depletion inhibits tumor growth in experimental animals and possibly in humans (21, 22). However, there is currently no information regarding whether riboflavin depletion promotes tumorigenesis. Materials and Methods Cell culture HyClone DMEM (riboflavin, 1.1 µM), DMEM/F12 medium (riboflavin, 0.6 µM), riboflavin-free DMEM, and riboflavin-free DMEM/F12 were purchased from Thermo Fisher Scientific. FBS was obtained from GIBCO (Life Technologies). The FBS contained 0.165 µM riboflavin (Supplemental Figure 1) as determined by HPLC as described previously (23). NIH3T3 cells were cultured in DMEM/F12 (defined as 3T3 R+) or riboflavin-free DMEM/F12 (defined as 3T3 R–) media. HEK293T cells were cultured in DMEM (defined as 293T R+) or riboflavin-free DMEM (defined as 293T R–) media. All media were supplemented with 10% FBS (contributing 16.5 nM to the final concentration) and antibiotics (100 U/mL penicillin and 100 mg/L streptomycin). All cells were maintained at 37°C in a humidified 5% CO2 atmosphere. Cell plate colony formation assays Cell plate colony formation assays were performed as described previously (24). Briefly, cells were plated at a density of 500 cells per well in 6-well plates and incubated for 15 d at 37°C with 5% CO2. After washing with 4°C precooled PBS, cultures were fixed with 4°C precooled methanol for 20 min and stained with crystal violet for 15 min. Colonies were photographed and their sizes calculated with a FluorChem 8900 image analysis system (Alpha Innotech). Each experiment was performed in triplicate. In vivo tumorigenicity assays All animal experiments were conducted with the approval of the Institutional Animal Care and Use Committee of Shantou University. Male NU/NU mice (Vital River), 4 wk of age, were pair-fed a regular diet containing sufficient riboflavin (6 mg/kg diet; AIN-93G) (25) or a riboflavin-deficient diet (0 mg/kg diet; riboflavin-deprived AIN-93G) (TROPHIC Animal Feed High-Tech Co., Ltd.) for 38 d. HEK293T cells (1 × 106, 1 × 105, or 1 × 104) were subcutaneously injected into the flanks of nude mice fed the riboflavin-defined diet for 2 d (48 h) (n = 3/group). Xenograft tumors were subjected to hematoxylin and eosin (H&E) staining. Ki67 immunohistochemical staining Immunohistochemistry was performed as described in our previous studies (26). Briefly, sections 4 μm thick were dewaxed in xylene, rehydrated in alcohol, and incubated in 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity. Sections were incubated with 10% normal goat serum in PBS for 15 min at room temperature to block nonspecific binding. Sections were then incubated overnight at 4°C with primary antibodies for Ki67 (ZSBIO). After rinsing with PBS, slides were incubated for 30 min at 37°C with HRP Polymer Conjugate (ZYMED). Subsequently, slides were stained with a Liquid DAB Substrate kit (ZSGB-BIO), then counterstained with hematoxylin, dehydrated, and mounted. Finally, tumorigenesis was confirmed by a pathologist according to cellular morphology and positive signals for Ki67 located in the nucleus of tumor cells. Riboflavin determination The erythrocyte glutathione reductase activity coefficient (GRAC) (27–31) was calculated in accordance with the Glutathione Reductase Activation Coefficient assay kit instructions (Nanjing Jiancheng Bioengineering Institute). The higher the GRAC value, the lower the riboflavin levels. RNA extraction and quantitative real-time PCR Total RNA was extracted from cells with the use of TRIzol (Invitrogen) in accordance with the manufacturer's instructions. cDNA was generated from total RNA (1 μL), in a final volume of 20 μL, with the use of the Reverse Transcription System (Promega). Quantitative real-time PCR assays were performed with the Rotor-Gene 6000 system (Corbett Life Science) and SYBR Premix Ex Taq (TaKaRa) in accordance with the manufacturers’ instructions. Primers for quantitative real-time PCR are shown in Supplemental Table 1 (ACTB was amplified as an internal control). The comparative delta-delta Ct (2−ΔΔCT) method (32) was used to calculate relative expression levels. Western blot analyses For SLC52A3 protein assays (33), cells were homogenized by sonication in buffer (250 mM sucrose and 5 mM HEPES, pH 7.4) and were centrifuged for 10 min at 2000 × g at 4°C. Supernatants were centrifuged again for 30 min at 15,000 × g at 4°C. Pellets were used for crude membrane samples and supernatants were used for cytoplasmic samples. Membrane and cytoplasmic samples were suspended in 1 × Laemmli sample buffer (Bio-Rad). Lysates were centrifuged for 5 min at 15,000 × g at 4°C. Protein concentrations were estimated with the use of a Pierce 660-nm protein assay (Thermo Fisher Scientific). Equal amounts of tissue lysate (30 μg) were subjected to electrophoresis on 10% polyacrylamide gels (40 V for 30 min, followed by 60 V for 3 h). Proteins were transferred to PVDF membranes (Millipore), which were then blocked for 1 h with 5% skim milk in PBS-Tween 20 (0.01 M PBS, 0.05% Tween 20). Membranes were then incubated for 1.5 h at room temperature with anti-rabbit SLC52A3 polyclonal antibody (1:500; Santa Cruz Biotechnology). Next, membranes were incubated for 1.5 h at room temperature with anti-rabbit peroxidase-conjugated secondary antibodies (1:5000; Santa Cruz Biotechnology). Anti-rabbit integrin α5 polyclonal antibody (1:500; Santa Cruz Biotechnology) was used as a control for equal extraction of cell membrane proteins. For total protein extraction, whole cell lysates were prepared in 1 × Laemmli sample buffer, heated for 5 min at 95°C, and subjected to immunoblot analyses. The following primary antibodies were used: anti-p18, anti-p21, anti-p27, anti-cyclin-dependent kinase (CDK) 2, anti-CDK4, anti-CDK6, anti-cyclinD1, and anti-cyclinD3 (Cell Signaling Technology); and anti-β-actin (Santa Cruz Biotechnology). The anti–cell cycle–related and expression-elevated protein in tumor (CREPT) antibody (3E10) (34) was a kind gift from Dr Zhi-Jie Chang (School of Life Science, Tsinghua University, China). Protein bands were visualized with the use of Western blotting Luminol reagent (Santa Cruz Biotechnology). Image acquisition and quantitative analyses were performed with the use a FluorChem 8900 image analysis system (Alpha Innotech). GeneChip Human Transcriptome Array 2.0 293T R+ and 293T R– cells samples were extracted with TRIzol (Invitrogen), and RNA was sent to the Shanghai Biotechnology Corporation (Shanghai) for gene expression profiling based on the use of the Human Transcriptome Array 2.0 (Affymetrix) in accordance with the manufacturer's instructions. Labeling, hybridization, scanning, and data extraction of the microarray were performed by the Shanghai Biotechnology Corporation in accordance with the recommended Affymetrix protocols. Briefly, the fluorescence signals of the microarray were scanned and saved as DAT image files. Command Console Software (Affymetrix) was used to transform DAT files into CEL files to change image signals into digital signals that recorded the fluorescence intensity of the probes. Next, the obtained CEL files were analyzed by Expression Console (Affymetrix) and Transcriptome Analysis Console (Affymetrix) software to detect differentially expressed genes. To identify significantly enriched gene ontology terms, the publicly available tools GO (Gene Ontology) (35) and DAVID (36) were used. The microarray data have been submitted to NCBI's Gene Expression Omnibus (GEO) (accession number GSE98927). Cell cycle analyses After 48 h of cell subculture, cells were harvested, washed once with cold PBS, fixed overnight in 70% ice-cold ethanol at 4°C, washed twice with PBS, and stained with propidium iodide for 30 min at 4°C. The DNA content of cells was analyzed by flow cytometry (Accuri C6, BD). Cell cycle distribution was determined via the use of FlowJo v. 7.6 software (Emerald Biotech Co., Ltd.). Chromatin immunoprecipitation Chromatin immunoprecipitation (ChIP) analysis was performed with the use of an EZ-Magna ChIP A/G Chromatin Immunoprecipitation kit (Millipore) in accordance with the manufacturer's instructions. Briefly, 293T R+ and 293T R– cells were incubated at a density of 106 cells/10 cm plate. Following formaldehyde crosslinking, cell lysates were sonicated on wet ice and then centrifuged to pellet the debris. The INPUT sample (5%) was removed before the immunoprecipiation. The supernatant containing the crosslinked chromatin preparation was incubated for overnight at 4°C with 1 μg of the specific antibody [H3 lysine 4 trimethylation (H3K4me3, a marker of repression, heterochromatin; Abcam), H3K9me3 (a marker of repression, heterochromatin; Abcam), H3K9 acetylation (ac; a marker of activation, euchromatin; Upstate Biotechnology), RNA polymerase II (positive control, Millipore) and normal rabbit IgG (negative control, Santa Cruz Biotechnology)]. After incubation, samples were subjected to DNA purification following the manufacturer's protocol. Finally, the purified DNA was analyzed by semiquantitative PCR and real-time PCR (defined as ChIP-qPCR) with the use of CREPT and SLC52A3 promoter-specific primers (Supplemental Table 1). The qPCR data was normalized to percentage of input and represented as percentage of enrichment relative to IgG, and compared with the control as described before (37, 38). Statistical analyses Data analyses were performed with SPSS 13.0 (SPSS, Inc.). Two-tailed independent sample t tests were used to determine whether differences were significant between groups. Differences were considered statistically significant if P < 0.05. Data shown are the means ± SDs. Results Riboflavin depletion promotes colony formation in NIH3T3 and HEK293T cells To assess the effects of riboflavin depletion on cell growth, we cultured NIH3T3 and HEK293T cells in riboflavin-deficient and riboflavin-supplemented media for 10 or 20 generations (where cells were passaged every 48 h), then examined the ability of cells to form colonies. As shown in Figure 1, the efficiency of cell colony formation in riboflavin-depleted NIH3T3 and HEK293T cells was significantly increased compared with control cells. These results suggest that riboflavin depletion promoted cell colony formation and proliferation of immortalized cells. FIGURE 1 View largeDownload slide Riboflavin depletion promotes colony formation in NIH3T3 (A) and HEK293T (B) cell lines. Cells subcultured in riboflavin-deficient medium for 10 or 20 generations show enhanced colony formation. Representative pictures (left) and quantitative analyses (right) of colony numbers. Values are means ± SDs, n = 3 (means of triplicates). **Means differ, P < 0.01. R+, riboflavin supplemented; R–, riboflavin depleted. FIGURE 1 View largeDownload slide Riboflavin depletion promotes colony formation in NIH3T3 (A) and HEK293T (B) cell lines. Cells subcultured in riboflavin-deficient medium for 10 or 20 generations show enhanced colony formation. Representative pictures (left) and quantitative analyses (right) of colony numbers. Values are means ± SDs, n = 3 (means of triplicates). **Means differ, P < 0.01. R+, riboflavin supplemented; R–, riboflavin depleted. Riboflavin depletion promotes HEK293T cell tumorigenesis in vivo Because riboflavin depletion promotes cell proliferation, we speculated that riboflavin depletion might induce tumorigenesis. To examine this hypothesis, we performed tumorigenicity assays in vivo. Riboflavin-depleted mice had decreased weights compared with the control group (Figure 2A) and 293T R– cells consistently formed larger tumors compared with 293T R+ cells, particularly in mice injected with 1 × 106 293T R– cells (Supplemental Figure 2A and B). H&E and Ki-67 immunohistochemical staining showed that most 293T R– tumors contained cancer cells and were positive for Ki-67 expression, whereas 293T R+ cells tumors consisted of large numbers of lymphocytes. Moreover, the total percentage of tumor formation was 55.6% in 293T R– cells (Figure 2B and Supplemental Table 2). Taken together, our results indicate that riboflavin depletion in HEK293T cells resulted in tumorigenicity in vivo. FIGURE 2 View largeDownload slide Riboflavin depletion promotes HEK293T cell tumorigenesis in nude mice. (A) Curve showing the weight increases of NU/NU mice that were fed AIN-93G or riboflavin-deprived AIN-93G for 38 d. Values are means ± SDs, n = 9 (means of nonduplicates). *,**,***Means differ: *P < 0.05; **P < 0.01; ***P < 0.001. (B) H&E and Ki-67 immunohistochemical staining of tumors formed from 293T R+ cells and 293T R– cells confirmed a malignant phenotype. Positive reactions with Ki-67 were defined as those showing brown (DAB) signals in tumor cells (×200). Scale bars: 50 μm. H&E, hematoxylin and eosin; R+, riboflavin supplemented; R–, riboflavin depleted. FIGURE 2 View largeDownload slide Riboflavin depletion promotes HEK293T cell tumorigenesis in nude mice. (A) Curve showing the weight increases of NU/NU mice that were fed AIN-93G or riboflavin-deprived AIN-93G for 38 d. Values are means ± SDs, n = 9 (means of nonduplicates). *,**,***Means differ: *P < 0.05; **P < 0.01; ***P < 0.001. (B) H&E and Ki-67 immunohistochemical staining of tumors formed from 293T R+ cells and 293T R– cells confirmed a malignant phenotype. Positive reactions with Ki-67 were defined as those showing brown (DAB) signals in tumor cells (×200). Scale bars: 50 μm. H&E, hematoxylin and eosin; R+, riboflavin supplemented; R–, riboflavin depleted. Riboflavin depletion decreases intracellular riboflavin levels and negative feedback upregulates riboflavin transporter expression To explore mechanisms of riboflavin depletion–induced tumorigenesis, we examined colony formation as shown in Figure 3A. 293T R– cells maintain strong colony-forming ability when subcultured out to 30–40 generations in riboflavin-deficient media. We next assessed the cellular GRAC to characterize intracellular riboflavin levels. The results showed that 293T R– cells have higher GRAC values, indicative of lower riboflavin levels, compared with 293T R+ cells (Figure 3B). We next examined the mRNA levels of riboflavin transporters (SLC52A1, SLC52A2, and SLC52A3), as well as the protein levels of SLC52A3. There were 1.3- to 2.7-fold increased levels of riboflavin transporter mRNAs in 293T R– cells compared with 293T R+ cells (Figure 3C). Moreover, SLC52A3 protein in 293T R– cell membranes were significantly enhanced (Figure 3D). These data suggest that riboflavin depletion decreases intracellular riboflavin levels, and the negative feedback upregulates riboflavin transporter expression. FIGURE 3 View largeDownload slide Intracellular riboflavin levels and riboflavin transporter gene expression in 293T R+ and 293T R– cells. (A) Colony formation assays and quantitative analyses of colony numbers of 293T R+ and 293T R– cells subcultured with riboflavin-deficient media for 40 generations. (B) Glutathione reductase activity in 293T R+ and 293T R– cells. (C) Quantitative real-time PCR assays of SLC52A mRNA expression levels. (D) Expression of SLC52A3 was analyzed by immunoblotting. Integrin α5 (a plasma membrane marker) was used as a control for equal extraction of cell membrane proteins (pellet). Bands were semiquantified according to the band gray values, and the numbers under the bands represent the gray value ratios of SLC52A3/integrin α5 (R+ set as 1). Values are means ± SDs, n = 3 (means of triplicates). **,***Means differ: **P < 0.01; ***P < 0.001. GRAC, erythrocyte glutathione reductase activity coefficient; R+, riboflavin supplemented; R–, riboflavin depleted. FIGURE 3 View largeDownload slide Intracellular riboflavin levels and riboflavin transporter gene expression in 293T R+ and 293T R– cells. (A) Colony formation assays and quantitative analyses of colony numbers of 293T R+ and 293T R– cells subcultured with riboflavin-deficient media for 40 generations. (B) Glutathione reductase activity in 293T R+ and 293T R– cells. (C) Quantitative real-time PCR assays of SLC52A mRNA expression levels. (D) Expression of SLC52A3 was analyzed by immunoblotting. Integrin α5 (a plasma membrane marker) was used as a control for equal extraction of cell membrane proteins (pellet). Bands were semiquantified according to the band gray values, and the numbers under the bands represent the gray value ratios of SLC52A3/integrin α5 (R+ set as 1). Values are means ± SDs, n = 3 (means of triplicates). **,***Means differ: **P < 0.01; ***P < 0.001. GRAC, erythrocyte glutathione reductase activity coefficient; R+, riboflavin supplemented; R–, riboflavin depleted. Genes affected by riboflavin depletion overlap with cell cycle–related genes and DNA damage Transcriptomic changes were evaluated, in the riboflavin-depleted cells, with the use of the Human Transcriptome Array 2.0 expression microarray. Differentially expressed genes (≥2-fold change) were identified in 293T R+ and 293T R– cells (GSE98927). The analysis identified 660 differentially expressed genes, of which 287 (43%) genes were upregulated and 373 (57%) genes were downregulated. The top (based on fold change ranking) up- and downregulated genes following riboflavin depletion are shown in Supplemental Table 3. The 660 differentially expressed genes were subjected to GO enrichment analysis to determine the associated gene networks, and cellular functions. The affected genes most strongly correlated with biological processes that directly or indirectly relate to cell cycle and DNA damage (Supplemental Table 4). Riboflavin depletion regulates expression of cell cycle genes Flow cytometry assays were employed to determine whether riboflavin depletion enhances cell growth and promotes tumorigenesis via cell cycle alterations. The number of cells in S phase was significantly decreased and the number of cells in the G2/M phase was increased following riboflavin depletion (Figure 4A). By determining the length of each phase, we found that riboflavin depletion shortened the cell cycle in 293T R– and 3T3 R– cells (Figure 4B). Because the cell cycle was altered by riboflavin depletion, we addressed whether riboflavin depletion affects the expression of cell cycle–related genes. We focused on genes for activators (CDK2, CDK4, CDK6, cyclin D1, and cyclin D3), inhibitors (p18INK4c, p21CIP1, and p27KIP1), as well as the upstream gene CREPT (39). Immunoblotting analyses indicated that cyclin D1 and CDK4 protein levels were dramatically higher in 293T R– and 3T3 R– cells. In contrast, CDK2 and CDK6 expression remained constant, whereas p18INK4c and p21CIP1/27KIP1 levels decreased. Additionally, CREPT was also increased upon riboflavin depletion (Figure 4C). These results indicate that riboflavin depletion alters the cell cycle. FIGURE 4 View largeDownload slide Riboflavin depletion alters the cell cycle and expression of cell cycle-associated genes. (A) Cell cycle distribution of 293T R+, 293T R–, 3T3 R+, and 3T3 R– cells. (B) The percentages of cells in the G1, S, G2/M, and sub-G1 phases of the cell cycle. (C) Expression of cell cycle–related genes were evaluated with the use of Western blot analysis. Bands were semiquantified according the band gray values, and the numbers under the bands represent the gray value ratios of target protein/β-actin (R+ set as 1). Results are representative of 3 independent experiments. CDK, cyclin-dependent kinase; CREPT, cell cycle–related and expression-elevated protein in tumor; R+, riboflavin supplemented; R–, riboflavin depleted. FIGURE 4 View largeDownload slide Riboflavin depletion alters the cell cycle and expression of cell cycle-associated genes. (A) Cell cycle distribution of 293T R+, 293T R–, 3T3 R+, and 3T3 R– cells. (B) The percentages of cells in the G1, S, G2/M, and sub-G1 phases of the cell cycle. (C) Expression of cell cycle–related genes were evaluated with the use of Western blot analysis. Bands were semiquantified according the band gray values, and the numbers under the bands represent the gray value ratios of target protein/β-actin (R+ set as 1). Results are representative of 3 independent experiments. CDK, cyclin-dependent kinase; CREPT, cell cycle–related and expression-elevated protein in tumor; R+, riboflavin supplemented; R–, riboflavin depleted. Diverse histone modifications on gene promoters for CREPT and SLC52A3 regulate CREPT and SLC52A3 expression Several previous studies have shown that different histone epigenetic markers and their modifications give rise to differential gene expression under different physiological contexts of cells (40–42). To investigate a plausible molecular explanation for CREPT and SLC52A3 overexpression by riboflavin depletion, we carried out ChIP and analyzed histone H3 tail modifications at promoter regions for the CREPT and SLC52A3 genes. The results showed that H3K4me3 and H3K9me3 were present at low levels, whereas high levels of H3K9ac enrichment were observed on the CREPT promoter 31-P2 region in 293T R– cells (Figure 5A and Supplemental Figure 3). We also observed similar results for the SLC52A3 promoter region in 293T R– cells, which were combined with low levels of H3K9me3 enrichment (Figure 5B and Supplemental Figure 3). This suggests that the induction in CREPT and SLC52A3 promoter activity observed following riboflavin deficiency may involve epigenetic mechanisms. FIGURE 5 View largeDownload slide ChIP-qPCR analysis of H3K4me3, H3K9me3 and H3K9ac in immunoprecipitated DNA fragments for the CREPT (A) and SLC52A3 (B) promoters in 293T R+ and 293T R– cells. Top, schematic of the human CREPT or SLC52A3 promoter. The right arrows mark the TSS. Bottom, ChIP was carried out with antibodies recognizing H3K4me3, H3K9me3, or H3K9ac. Anti-RNA polymerase II was used as a positive control and normal rabbit IgG was used as a negative control. qPCR analysis was conducted with the primer sets described in the Methods. Values are means ± SDs, n = 3 (means of triplicates). *,**Means differ: *P < 0.05; **P < 0.01. ChIP, chromatin immunoprecipitation; CREPT, cell cycle–related and expression-elevated protein in tumor; H3K4me3, H3 lysine 4 trimethylation; H3K9ac, H3 lysine 9 acetylation; R+, riboflavin supplemented; R–, riboflavin depleted; TSS, transcriptional start site. FIGURE 5 View largeDownload slide ChIP-qPCR analysis of H3K4me3, H3K9me3 and H3K9ac in immunoprecipitated DNA fragments for the CREPT (A) and SLC52A3 (B) promoters in 293T R+ and 293T R– cells. Top, schematic of the human CREPT or SLC52A3 promoter. The right arrows mark the TSS. Bottom, ChIP was carried out with antibodies recognizing H3K4me3, H3K9me3, or H3K9ac. Anti-RNA polymerase II was used as a positive control and normal rabbit IgG was used as a negative control. qPCR analysis was conducted with the primer sets described in the Methods. Values are means ± SDs, n = 3 (means of triplicates). *,**Means differ: *P < 0.05; **P < 0.01. ChIP, chromatin immunoprecipitation; CREPT, cell cycle–related and expression-elevated protein in tumor; H3K4me3, H3 lysine 4 trimethylation; H3K9ac, H3 lysine 9 acetylation; R+, riboflavin supplemented; R–, riboflavin depleted; TSS, transcriptional start site. Discussion Riboflavin is essential for normal cellular function. Mammals are unable to synthesize riboflavin and thus must acquire it via intestinal absorption (43). Riboflavin transporters are likely essential for the maintenance of riboflavin homeostasis in the intestine and kidney (44). Recently, a family of novel riboflavin transporters, RFVT/SLC52, has been identified (33, 45–47). Of these, SLC52A3 is predominantly expressed in the testis and small intestine (33). Our results show that riboflavin deficiency in HEK293T cells decreases intracellular riboflavin levels and promotes proliferation. Moreover, the expression of riboflavin transporter genes is significantly higher in HEK293T cells following riboflavin depletion, most likely because of a negative feedback mechanism between riboflavin deficiency and riboflavin transporters, enabling cells to obtain more riboflavin. However, when riboflavin transporter proteins become overexpressed and cannot be degraded due to disturbances in cell metabolism or cell cycle dysregulation, tumors are induced. A recent study showed that the induction in riboflavin uptake under riboflavin deficiency is associated with an increase in expression of SLC52A2 and SLC52A3 at the protein and mRNA levels, and the adaptive regulation in riboflavin uptake by substrate concentration is mediated at both the transcriptional and post-transcriptional levels (20). Notably, a few reports indicate that SLC52A3 mRNA and protein are upregulated in esophageal squamous cell carcinoma and glioma (48, 49). Moreover, statistical analyses of esophageal squamous cell carcinoma patients classified by the Tumor, Node, Metastasis staging system have determined that SLC52A3 is upregulated in the early stages and persists throughout tumor progression, suggesting that SLC52A3 may be important for the entire spectrum of esophageal squamous cell carcinoma tumorigenesis (48). Together, these findings indicate that riboflavin depletion may contribute to tumorigenesis via upregulation of riboflavin transporter expression. Previous studies reported that short-term riboflavin depletion inhibits cell growth (15, 50, 51). In our study, in order to continuously observe the growth of cells in riboflavin-deficient medium, we performed proliferation assays every 5 generations. After 10 generations, the proliferation of cells in riboflavin-deficient medium was clearly greater than that of cells grown in normal medium. Unlike short-term riboflavin depletion, our results indicate that long-term riboflavin depletion can promote tumorigenesis and maintain tumor characteristics. Tumor development is highly related to uncontrolled cell growth (52). Cell cycle–related proteins, such as cyclin D and CDKs, play a major role in tumorigenesis (53). Chang and coworkers identified the CREPT gene and unraveled the mechanisms by which CREPT accelerates tumorigenesis by regulating cyclin D1. Their data indicate that CREPT prevents RNA polymerase II from moving beyond the poly(A) site of cyclin D1 during transcription and enhances RNA polymerase II recycling from the terminator to promoter region via formation of a chromatin loop (39). Our study reveals that riboflavin deficiency in HEK293T and NIH3T3 cells enhances the function of cell cycle transition regulators CDK4, cyclin D1, and cyclin D3 by upregulating their expression and downregulating expression of their inhibitors p21 and p27. Moreover, upregulation of the upstream CREPT gene causes more rapid S phase exit and accelerates tumorigenesis. We provide here experimental evidence that histone modification is an important mechanism for regulation of CREPT and SLC52A3 gene expression. The increase in expression levels of both CREPT and riboflavin transporters suggests possible involvement of transcriptional regulatory mechanism(s) affecting the respective genes. Based on these results, we propose the following model for riboflavin depletion in HEK293T cells: intracellular riboflavin deficiency enhances expression of CREPT and riboflavin transporters (such as SLC52A3) by histone modification (removal of H3K9me3 from histone tails), and subsequently regulates expression of cell cycle–related genes (CDK4, cyclin D1, cyclin D3, p21, and p27) (48, 49), resulting in uncontrolled cell growth and tumorigenesis (Supplemental Figure 4). There are still some limitations to our findings. In our experiment, we only selected 2 nontumorigenic cell lines to study tumorigenesis dependent on riboflavin deficiency, and we need to choose more immortal cells in future studies, especially riboflavin-susceptible epithelial cells such as esophageal epithelial cells. Furthermore, we only used a model of extreme riboflavin depletion and merely took complete lack of riboflavin into consideration for tumorigenesis. In further studies, we will examine riboflavin depleted at various concentrations, and combine different environmental factors such as zinc depletion and nitrosamine supplementation. Lastly, we did not perform direct riboflavin-deficient animal experiments. Some questions remain concerning the role of riboflavin depletion. For example, it is not clear if the key factor is riboflavin itself or overexpressed riboflavin transporters and CREPT in riboflavin depletion–induced tumorigenesis. It is also unknown what causes histone modification at the CREPT and SLC52A3 genes. Finally, it is still unclear whether there are more genes undergoing histone modification in riboflavin-depleted cells. These questions will be addressed in future studies to increase our understanding of the mechanisms involved in riboflavin depletion–induced tumorigenesis. In conclusion, our data show that riboflavin depletion promotes tumorigenesis in HEK293T and NIH3T3 cells. We find that riboflavin depletion promotes tumor development by sustaining cell proliferation and regulating transcription of cell cycle–related genes, including CREPT. We suggest that riboflavin deficiency should be considered another risk factor for tumorigenesis. Acknowledgments We thank Stanley Li Lin, Department of Cell Biology and Genetics, Shantou University Medical College, for assistance in revising the manuscript, and the Central Laboratory at Shantou University Medical College for their assistance, including Wen-Hong Luo and Hong-Jun Luo, for obtaining the riboflavin concentration data by HPLC. The authors’ responsibilities were as follows—L-YX and E-ML: designed the study; LL: performed the experiments and analyzed the data; LL and L-YX: wrote the manuscript and contributed to the discussion; L-DL: performed the cell culture; YC and Y-MX: performed the animal experiments; J-ZH, YC and X-EX: undertook the pathological review; and all authors: read and approved the final manuscript. Notes Supported by a grant from the Natural Science Foundation of China–Guangdong Joint Fund (No. U1301227) and the Department of Education, Guangdong Government under the Top-tier University Development Scheme for Research and Control of Infectious Diseases. Author disclosures: LL, J-ZH, YC, X-EX, L-DL, Y-MX, E-ML and L-YX, no conflicts of interest. Supplemental Figures 1–4 and Supplemental Tables 1–4 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/jn/. Abbreviations used: CDK, cyclin-dependent kinase; ChIP, chromatin immunoprecipitation; CREPT, cell cycle–related and expression-elevated protein in tumor; GRAC, erythrocyte glutathione reductase activity coefficient; H&E, hematoxylin and eosin; H3K4me3, H3 lysine 4 trimethylation; H3K9ac, H3 lysine 9 acetylation. References 1. Lienhart WD , Gudipati V , Macheroux P . The human flavoproteome . Arch Biochem Biophys 2013 ; 535 : 150 – 62 . Google Scholar CrossRef Search ADS PubMed 2. Powers HJ . Riboflavin (vitamin B-2) and health . Am J Clin Nutr 2003 ; 77 : 1352 – 60 . Google Scholar CrossRef Search ADS PubMed 3. Hornbeck CL , Bradley DW . Letter: Concentrations of FAD and glutathione as they affect values for erythrocyte glutathione reductase . Clin Chem 1974 ; 20 : 512 – 3 . Google Scholar PubMed 4. Tu BP , Weissman JS . The FAD- and O(2)-dependent reaction cycle of Ero1-mediated oxidative protein folding in the endoplasmic reticulum . Mol Cell 2002 ; 10 : 983 – 94 . Google Scholar CrossRef Search ADS PubMed 5. Thorpe C , Hoober KL , Raje S , Glynn NM , Burnside J , Turi GK , Coppock DL . Sulfhydryl oxidases: emerging catalysts of protein disulfide bond formation in eukaryotes . Arch Biochem Biophys 2002 ; 405 : 1 – 12 . Google Scholar CrossRef Search ADS PubMed 6. Nagao M , Tanaka K . FAD-dependent regulation of transcription, translation, post-translational processing, and post-processing stability of various mitochondrial acyl-CoA dehydrogenases and of electron transfer flavoprotein and the site of holoenzyme formation . J Biol Chem 1992 ; 267 : 17925 – 32 . Google Scholar PubMed 7. Tu BP , Ho-Schleyer SC , Travers KJ , Weissman JS . Biochemical basis of oxidative protein folding in the endoplasmic reticulum . Science 2000 ; 290 : 1571 – 4 . Google Scholar CrossRef Search ADS PubMed 8. Bowman BB , McCormick DB , Rosenberg IH . Epithelial transport of water-soluble vitamins . Annu Rev Nutr 1989 ; 9 : 187 – 99 . Google Scholar CrossRef Search ADS PubMed 9. Bates CJ , Prentice AM , Paul AA , Prentice A , Sutcliffe BA , Whitehead RG . Riboflavin status in infants born in rural Gambia, and the effect of a weaning food supplement . Trans R Soc Trop Med Hyg 1982 ; 76 : 253 – 8 . Google Scholar CrossRef Search ADS PubMed 10. Zou XN , Taylor PR , Mark SD , Chao A , Wang W , Dawsey SM , Wu YP , Qiao YL , Zheng SF . Seasonal variation of food consumption and selected nutrient intake in Linxian, a high risk area for esophageal cancer in China . Int J Vitam Nutr Res 2002 ; 72 : 375 – 82 . Google Scholar CrossRef Search ADS PubMed 11. Siassi F , Ghadirian P . Riboflavin deficiency and esophageal cancer: a case control-household study in the Caspian Littoral of Iran . Cancer Detect Prev 2005 ; 29 : 464 – 9 . Google Scholar CrossRef Search ADS PubMed 12. Khan NA , Teli MA , Mohib-ul Haq M , Bhat GM , Lone MM , Afroz F . A survey of risk factors in carcinoma esophagus in the valley of Kashmir, Northern India . J Cancer Research Ther 2011 ; 7 : 15 – 8 . Google Scholar CrossRef Search ADS 13. Camporeale G , Zempleni J . Oxidative folding of interleukin-2 is impaired in flavin-deficient Jurkat cells, causing intracellular accumulation of interleukin-2 and increased expression of stress response genes' . J Nutr 2003 ; 133 : 668 – 72 . Google Scholar CrossRef Search ADS PubMed 14. Manthey KC , Chew YC , Zempleni J . Riboflavin deficiency impairs oxidative folding and secretion of apolipoprotein B-100 in HepG2 cells, triggering stress response systems . J Nutr 2005 ; 135 : 978 – 82 . Google Scholar CrossRef Search ADS PubMed 15. Werner R , Manthey KC , Griffin JB , Zempleni J . HepG2 cells develop signs of riboflavin deficiency within 4 days of culture in riboflavin-deficient medium . J Nutr Biochem 2005 ; 16 : 617 – 24 . Google Scholar CrossRef Search ADS PubMed 16. Manthey KC , Rodriguez-Melendez R , Hoi JT , Zempleni J . Riboflavin deficiency causes protein and DNA damage in HepG2 cells, triggering arrest in G1 phase of the cell cycle . J Nutr Biochem 2006 ; 17 : 250 – 6 . Google Scholar CrossRef Search ADS PubMed 17. Nakano E , Mushtaq S , Heath PR , Lee ES , Bury JP , Riley SA , Powers HJ , Corfe BM . Riboflavin depletion impairs cell proliferation in adult human duodenum: identification of potential effectors . Dig Dis Sci 2011 ; 56 : 1007 – 19 . Google Scholar CrossRef Search ADS PubMed 18. Powers HJ , Corfe BM , Nakano E . Riboflavin in development and cell fate . Subcell Biochem 2012 ; 56 : 229 – 45 . Google Scholar CrossRef Search ADS PubMed 19. Aili A , Hasim A , Kelimu A , Guo X , Mamtimin B , Abudula A , Upur H . Association of the plasma and tissue riboflavin levels with C20orf54 expression in cervical lesions and its relationship to HPV16 infection . PLoS One 2013 ; 8 : e79937 . Google Scholar CrossRef Search ADS PubMed 20. Subramanian VS , Ghosal A , Kapadia R , Nabokina SM , Said HM . Molecular mechanisms mediating the adaptive regulation of intestinal riboflavin uptake process . PLoS One 2015 ; 10 : e0131698 . Google Scholar CrossRef Search ADS PubMed 21. Morris HP . Effects on the genesis and growth of tumors associated with vitamin intake . Ann N Y Acad Sci 1947 ; 49 : 119 – 40 . Google Scholar CrossRef Search ADS 22. Rivlin RS . Riboflavin and cancer: a review . Cancer Res 1973 ; 33 : 1977 – 86 . Google Scholar PubMed 23. Petteys BJ , Frank EL . Rapid determination of vitamin B(2) (riboflavin) in plasma by HPLC . Clin Chim Acta 2011 ; 412 : 38 – 43 . Google Scholar CrossRef Search ADS PubMed 24. Zhang HF , Chen Y , Wu C , Wu ZY , Tweardy DJ , Alshareef A , Liao LD , Xue YJ , Wu JY , Chen B et al. The opposing function of STAT3 as an oncoprotein and tumor suppressor is dictated by the expression status of STAT3beta in esophageal squamous cell carcinoma . Clin Cancer Res 2016 ; 22 : 691 – 703 . Google Scholar CrossRef Search ADS PubMed 25. Reeves PG . Components of the AIN-93 diets as improvements in the AIN-76A diet . J Nutr 1997 ; 127 : S838 – 41 . Google Scholar CrossRef Search ADS 26. Zhao Q , Shen JH , Shen ZY , Wu ZY , Xu XE , Xie JJ , Wu JY , Huang Q , Lu XF , Li EM et al. Phosphorylation of fascin decreases the risk of poor survival in patients with esophageal squamous cell carcinoma . J Histochem Cytochem 2010 ; 58 : 979 – 88 . Google Scholar CrossRef Search ADS PubMed 27. Blot WJ , Li JY , Taylor PR , Guo W , Dawsey S , Wang GQ , Yang CS , Zheng SF , Gail M , Li GY . Nutrition intervention trials in Linxian, China: supplementation with specific vitamin/mineral combinations, cancer incidence, and disease-specific mortality in the general population . J Natl Cancer Inst 1993 ; 85 : 1483 – 92 . Google Scholar CrossRef Search ADS PubMed 28. He Y , Ye L , Shan B , Song G , Meng F , Wang S . Effect of riboflavin-fortified salt nutrition intervention on esophageal squamous cell carcinoma in a high incidence area, China . Asian Pac J Cancer Prev 2009 ; 10 : 619 – 22 . Google Scholar PubMed 29. Tillotson JA , Baker EM . An enzymatic measurement of the riboflavin status in man . Am J Clin Nutr 1972 ; 25 ( 4 ): 425 – 31 . Google Scholar CrossRef Search ADS PubMed 30. Sauberlich HE , Judd JH Jr , Nichoalds GE , Broquist HP , Darby WJ . Application of the erythrocyte glutathione reductase assay in evaluating riboflavin nutritional status in a high school student population . Am J Clin Nutr 1972 ; 25 ( 8 ): 756 – 62 . Google Scholar CrossRef Search ADS PubMed 31. Prentice AM , Bates CJ . Biochemical evaluation of the erythrocyte glutathione reductase test for riboflavin deficiency . Proc Nutr Soc 1978 ; 37 ( 2 ): 49A . Google Scholar PubMed 32. Livak KJ , Schmittgen TD . Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method . Methods 2001 ; 25 : 402 – 8 . Google Scholar CrossRef Search ADS PubMed 33. Yao Y , Yonezawa A , Yoshimatsu H , Masuda S , Katsura T , Inui K . Identification and comparative functional characterization of a new human riboflavin transporter hRFT3 expressed in the brain . J Nutr 2010 ; 140 : 1220 – 6 . Google Scholar CrossRef Search ADS PubMed 34. Ren F , Wang R , Zhang Y , Liu C , Wang Y , Hu J , Zhang L , Chang Z . Characterization of a monoclonal antibody against CREPT, a novel protein highly expressed in tumors . Monoclon Antib Immunodiagn Immunother 2014 ; 33 : 401 – 8 . Google Scholar CrossRef Search ADS PubMed 35. Gene Ontology C . Gene Ontology Consortium: going forward . Nucleic Acids Res 2015 ; 43 ( Database issue ): D1049 – 56 . PubMed 36. Dennis G Jr. , Sherman BT , Hosack DA , Yang J , Gao W , Lane HC , Lempicki RA . DAVID: Database for Annotation, Visualization, and Integrated Discovery . Genome Biol 2003 ; 4 : P3 . Google Scholar CrossRef Search ADS PubMed 37. Xiang N , Zhao R , Song G , Zhong W . Selenite reactivates silenced genes by modifying DNA methylation and histones in prostate cancer cells . Carcinogenesis 2008 ; 29 : 2175 – 81 . Google Scholar CrossRef Search ADS PubMed 38. Chen T , He S , Zhang Z , Gao W , Yu L , Tan Y . Foxa1 contributes to the repression of Nanog expression by recruiting Grg3 during the differentiation of pluripotent P19 embryonal carcinoma cells . Exp Cell Res 2014 ; 326 : 326 – 35 . Google Scholar CrossRef Search ADS PubMed 39. Lu D , Wu Y , Wang Y , Ren F , Wang D , Su F , Zhang Y , Yang X , Jin G , Hao X et al. CREPT accelerates tumorigenesis by regulating the transcription of cell-cycle-related genes . Cancer Cell 2012 ; 21 : 92 – 104 . Google Scholar CrossRef Search ADS PubMed 40. Wu J , Wang SH , Potter D , Liu JC , Smith LT , Wu YZ , Huang TH , Plass C . Diverse histone modifications on histone 3 lysine 9 and their relation to DNA methylation in specifying gene silencing . BMC Genomics 2007 ; 8 : 131 . Google Scholar CrossRef Search ADS PubMed 41. Fu Y , Nachtigal MW . Analysis of epigenetic alterations to proprotein convertase genes in disease . Methods Mol Biol 2011 ; 768 : 231 – 45 . Google Scholar CrossRef Search ADS PubMed 42. Shanmugam MK , Sethi G . Role of epigenetics in inflammation-associated diseases . Subcell Biochem 2013 ; 61 : 627 – 57 . Google Scholar CrossRef Search ADS PubMed 43. Daniel H , Wille U , Rehner G . In vitro kinetics of the intestinal transport of riboflavin in rats . J Nutr 1983 ; 113 : 636 – 43 . Google Scholar CrossRef Search ADS PubMed 44. Spector R . Riboflavin transport by rabbit kidney slices: characterization and relation to cyclic organic acid transport . J Pharmacol Exp Ther 1982 ; 221 : 394 – 8 . Google Scholar PubMed 45. Yonezawa A , Masuda S , Katsura T , Inui K . Identification and functional characterization of a novel human and rat riboflavin transporter, RFT1 . Am J Physiol Cell Physiol 2008 ; 295 : C632 – 41 . Google Scholar CrossRef Search ADS PubMed 46. Yamamoto S , Inoue K , Ohta KY , Fukatsu R , Maeda JY , Yoshida Y , Yuasa H . Identification and functional characterization of rat riboflavin transporter 2 . J Biochem 2009 ; 145 : 437 – 43 . Google Scholar CrossRef Search ADS PubMed 47. Yonezawa A , Inui K-i . Novel riboflavin transporter family RFVT/SLC52: Identification, nomenclature, functional characterization and genetic diseases of RFVT/SLC52 . Mol Aspects Med 2013 ; 34 : 693 – 701 . Google Scholar CrossRef Search ADS PubMed 48. Jiang XR , Yu XY , Fan JH , Guo L , Zhu C , Jiang W , Lu SH . RFT2 is overexpressed in esophageal squamous cell carcinoma and promotes tumorigenesis by sustaining cell proliferation and protecting against cell death . Cancer Lett 2014 ; 353 : 78 – 86 . Google Scholar CrossRef Search ADS PubMed 49. Fu T , Liu Y , Wang Q , Sun Z , Di H , Fan W , Liu M , Wang J . Overexpression of riboflavin transporter 2 contributes toward progression and invasion of glioma . Neuroreport 2016 ; 27 : 1167 – 73 . Google Scholar CrossRef Search ADS PubMed 50. Lee ES , Corfe BM , Powers HJ . Riboflavin depletion of intestinal cells in vitro leads to impaired energy generation and enhanced oxidative stress . Eur J Nutr 2013 ; 52 : 1513 – 21 . Google Scholar CrossRef Search ADS PubMed 51. Manthey KC , Rodriguez-Melendez R , Hoi JT , Zempleni J . Riboflavin deficiency causes protein and DNA damage in HepG2 cells, triggering arrest in G1 phase of the cell cycle . J Nutr Biochem 2006 ; 17 : 250 – 6 . Google Scholar CrossRef Search ADS PubMed 52. Hanahan D , Weinberg RA . The hallmarks of cancer . Cell 2000 ; 100 : 57 – 70 . Google Scholar CrossRef Search ADS PubMed 53. Semczuk A , Jakowicki JA . Alterations of pRb1-cyclin D1-cdk4/6-p16(INK4A) pathway in endometrial carcinogenesis . Cancer Lett 2004 ; 203 : 1 – 12 . Google Scholar CrossRef Search ADS PubMed © 2018 American Society for Nutrition. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Journal of Nutrition – Oxford University Press
Published: May 7, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera