High expression of Fatty Acid‐Binding Protein 5 promotes cell growth and metastatic potential of colorectal cancer cells

Fatty acid‐binding proteins (FABPs) are responsible for binding and storing hydrophobic ligands such as long‐chain fatty acids, and for transporting these ligands to the appropriate compartments within the cell. The present study demonstrates that the FABP5 gene is upregulated in colorectal cancer cells compared to normal colon cells in a manner that correlates with disease stage and that FABP5 significantly promotes colorectal cancer cell growth and metastatic potential. Thus, FABP5 might be a promising prognostic or therapeutic biomarker candidate in human colorectal cancer.

Fatty acid-binding proteins (FABPs) are responsible for binding and storing hydrophobic ligands such as long-chain fatty acids, and for transporting these ligands to the appropriate compartments within the cell. The present study demonstrates that the FABP5 gene is upregulated in colorectal cancer cells compared to normal colon cells in a manner that correlates with disease stage and that FABP5 significantly promotes colorectal cancer cell growth and metastatic potential. Thus, FABP5 might be a promising prognostic or therapeutic biomarker candidate in human colorectal cancer.
Colorectal cancer (CRC) is the third most frequent malignancy and fourth leading cause of cancer-related death worldwide [1]. Genetic mutations, epigenetic alterations, inflammation, and environmental factors such as diet are thought to be linked to CRC initiation and progression [2,3]. Despite growing clinical significance, the detailed molecular mechanisms underlying metastasis in CRC remain poorly understood. The major cause of death in CRC is metastasis. Therefore, it is very important to identify the critical genes involved in the progression of CRC in order to diagnose the disease at an earlier stage using an appropriate biomarker.
We previously reported that FABP5 was highly expressed and involved in metastasis in prostate cancer cells [4][5][6][7][8][9]. In addition, other studies reported that FABP5 was upregulated in oral squamous cell carcinoma [10], intrahepatic cholangiocarcinoma [11], pancreatic [12], bladder [13], and triple negative breast cancer [14,15]. Moreover, recent proteomic analyses indicated that FABP5 was also upregulated in hepatocellular carcinoma [16] and CRC cells [17]. Thus, as FABP5 is upregulated in several cancer types, it is expected to be a promising prognostic or therapeutic biomarker candidate in these cancers; however, the precise molecular mechanisms underlying FABP5 upregulation and its oncogenic effects in cancer cells remain unclear, despite the extensive efforts of many research groups attempting to clarify the mechanisms. It would be interesting to assess whether the regulatory mechanisms underlying the upregulation of FABP5 gene expression and the functions of FABP5 protein in cancer cells are mediated by a common signaling pathway. Further studies on the mechanisms regulating FABP5 gene expression in cancer cells are now in progress in our laboratory. In particular, although FABP5 is the most upregulated protein in the FABP family consisting of ten isoforms [18], the molecular functions of FABP5 in CRC cells remain poorly characterized. As CRC is a common cancer and a major cause of mortality in men and women, it is very important to elucidate these issues. Therefore, the present study attempted to characterize the functions of FABP5 in CRC cells.
Fatty acid-binding proteins (FABPs) are members of the intracellular lipid-binding proteins that bind intracellular hydrophobic ligands such as long-chain fatty acids. FABPs are involved in fatty acid uptake and transport [18,19]. Recent studies also report that FABPs play roles in the regulation of gene expression, cell growth, and differentiation [20,21]. Several FABPs are upregulated in cancer cells; however, the mechanisms that regulate FABP gene expression and function in cancer cells remain poorly characterized. Recent studies demonstrate that metabolic reprogramming is necessary to sustain cancer cell growth and survival. Alteration in fatty acid metabolism is a hallmark of cancer, and several lines of evidence showed that limiting fatty acid availability controls cancer cell proliferation [22,23]. As fatty acids are required for the formation of membrane components, energy sources, and the production of cellular signaling molecules during cancer cell proliferation, FABPs might play an important role in cellular proliferation.
The present study focuses on the physiological functions of FABP5 in CRC cells and assesses the effects of FABP5 expression on CRC cell progression. Results suggest for the first time that high-level FABP5 promotes cell proliferation and metastatic potential in CRC cells.

Quantitative real-time PCR (Q-PCR)
Total RNA was extracted using the TRI Reagent (Molecular Research Center, Cincinnati, OH, USA), and cDNAs were synthesized from 1 lg of total RNA using the Rever-Tra Ace qPCR RT Master Mix (Toyobo, Osaka, Japan). Quantitative real-time PCR (Q-PCR) analyses were performed with the StepOne Real-Time PCR system (Applied Biosystems, Foster City, CA, USA) using THUNDER-BIRD SYBR qPCR Mix (Toyobo).

Western blotting
Cells were lysed in RIPA buffer with protease inhibitor cocktail (Nacalai Tesque). Equivalent amounts of protein were fractionated by SDS/PAGE. Immunoblotting was carried out using the appropriate antibodies. Signals were detected using chemiluminescent substrate (Thermo Scientific) with the Image Quant LAS4000 Mini (GE Healthcare Life Sciences, Pittsburgh, PA, USA).

Cell proliferation assay
Cells were counted to assess proliferation. HCT116 cells were plated onto six-well plates at a density of 2 9 10 5 cells/well and transfected with control or FABP5 siRNA, or treated with GSK-3787. Cells were counted at 24, 48, and 72 h after transfection or treatment.

Cell cycle analysis
Cell cycle distribution was assessed by flow cytometry after staining cells with propidium iodide. Briefly, floating and adherent cells were collected, washed with ice-cold PBS and fixed with 70% ethanol. The cells were then treated with stain solution (100 lgÁmL À1 RNase A, 1% Triton X-100, 40 mM sodium citrate, and 50 lgÁmL À1 propidium iodide) for 1 h at room temperature in the dark. The stained cells were analyzed using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA).

Apoptosis assay
The cells transfected with FABP5 or negative control siRNA were collected, diluted in PBS and fixed with BD Cytofix/Cytoperm Fixation and Permeabilization Solution for 20 min on ice in the dark. The cells were then washed with washing buffer and reacted with anti-active caspase-3 antibody (BD Biosciences) for 1 h at room temperature. Next, the cells were washed with washing buffer and reacted with FITC-conjugated anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA, USA) for 30 min at room temperature in the dark. After the reaction, the cells were diluted in PBS, and flow cytometry was performed with a FACSCalibur flow cytometer. The data were analyzed using the CellQuest software (BD Biosciences). For each sample, 3 9 10 4 cells were recorded.

In vitro invasion assay
The in vitro invasion assay was performed using BioCoat Matrigel invasion chambers (24-well plate, 8 lm pore size; BD Biosciences). Briefly, HCT116 cells transfected with FABP5 or negative control siRNA were cultured in DMEM supplemented with 5% FBS for 48 h before the assay. 5 9 10 4 cells were resuspended in serum-free DMEM and seeded into the upper chamber, and the lower compartment was filled with DMEM supplemented with 10% FBS. After 24 h, the invasive cells were fixed and stained using the Diff-Quik Stain Kit (Sysmex, Kobe, Japan). The invasive cells were counted in five random fields per chamber (9 200 magnification). Assays were conducted in duplicates for each experiment, and repeated three times.

Statistical analysis
Data were expressed as the means AE standard deviation (SD) from at least three independent experiments. Statistical analysis was performed using the Student's t-test. P < 0.05 was considered statistically significant.

High expression of FABP5 in CRC cells
Although several FABPs are upregulated in cancer cells, the FABP isoforms expressed in CRC cells have not been identified. We first examined the expression of FABP subtypes in normal colon fibroblast cells (CCD-18Co) and CRC cell lines (HCT116, LoVo, DLD-1 and Caco-2). As shown in Fig. 1A, FABP1 (L-FABP) and FABP3 (H-FABP) were expressed in Caco-2 cells. FABP6 (IBABP) was expressed in all cell lines at lower levels, which is consistent with recent studies that demonstrated that FABP6 was upregulated in CRC [25], as was a longer FABP6 variant (IBABP-L) with 49 additional NH 2 -terminal amino acid residues [26]. The functional differences between IBABP and IBABP-L in CRC cells remain unclear. Further studies on this issue are now in progress in our laboratory. Interestingly, one of the FABP isoforms, FABP5, is predominantly expressed in all CRC cell lines tested, especially in malignant CRC lines (HCT116, LoVo, and DLD-1), suggesting that FABP5 might play a critical role in the malignancy of CRC. Next we measured the levels of FABP5 mRNA and protein in CCD-18Co cells and CRC cell lines. As shown in Fig. 1B,C, FABP5 is markedly upregulated in CRC cells, especially in the highly metastatic HCT116 cells, compared to the nonmetastatic Caco-2 cells [27,28] and normal colon fibroblasts (CCD-18Co). These results clearly suggest that the FABP5 gene is upregulated in CRC cells and that its level of expression may correlate positively with disease stage.

Effect of FABP5 expression on cell proliferation, cell cycle, and apoptosis in CRC
Although the expression of FABP5 in CRC cells has been reported [17], the molecular mechanisms underlying the function of FABP5 in CRC cell progression are unknown. Thus, to evaluate the possible roles of FABP5 in CRC cells, we first examined the effect of FABP5 on cell proliferation. HCT116 cells were transfected with FABP5-specific or control siRNA, and a cell proliferation assay was performed. We confirmed that FABP5 expression was significantly reduced at both the mRNA and protein levels 72 h post-transfection ( Fig. 2A,B). Knockdown of FABP5 expression significantly suppressed cell proliferation (Fig. 2C,D). Similar results in LoVo cells were obtained (data not shown). To further investigate the physiological functions of FABP5 in cell cycle progression, we conducted a cell cycle analysis and an apoptosis assay. Cell cycle analysis revealed that knockdown of FABP5 expression resulted in a marked increase in G1/G0 phase population (Fig. 2E). To clarify the functions of FABP5 in cell cycle progression, we assessed FABP5 knockdown-induced p53 Ser15 phosphorylation and p21 WAF1/Cip1 upregulation, a key regulator of cell cycle arrest [29]. As expected, phospho-p53 (Ser15) and p21 WAF1/Cip1 protein were dramatically elevated (about threefold) in HCT116 cells transfected with FABP5 siRNA compared to cells transfected with negative control siRNA (Fig. 2F). We also found that the transcription factor c-MYC, a negative regulator of p21 WAF1/Cip1 [30], was downregulated by FABP5 knockdown (Fig. 2F). To assess whether knockdown of FABP5 expression induces apoptotic cell death, a cell apoptosis assay based on the evaluation of caspase-3 activity by flow cytometry was performed. As illustrated in Fig. 2G, knockdown of FABP5 expres-sion significantly activated caspase-3, up to 2.2 times compared to control. Cell growth suppression upon knockdown of FABP5 can be attributed to p21mediated G1 arrest and apoptosis. These results indicate that FABP5 might play a critical role in cell proliferation, and this is the first report showing a relationship between overexpression of FABP5 and CRC progression.

Effect of FABP5 expression on cell invasion in CRC
We examined the effect of FABP5 expression on invasion in CRC cells using an in vitro invasion assay. As shown in Fig. 3A,B, knockdown of FABP5 significantly decreased the number of cells invading through the Matrigel-coated membrane. This finding strongly suggests that FABP5 might regulate the invasiveness of CRC cells during metastasis. Although FABP5 promotes invasion in cancer cells, the molecular mechanism remains unknown. Cancer cells change lipid metabolism to lipogenesis, which promotes malignancy [31]. Triacylglycerides are stored in lipid droplets and are used to generate energy or act as second Lysates from HCT116 cells were analyzed by western blotting using specific antibodies to FABP5, p21, c-MYC, p53, and phospho-p53 (Ser15). a-tubulin was used as a loading control. The western blot data shown are representative of three independent experiments. (G) Assay for caspase-3 activity after transfection with FABP5 siRNA or negative control siRNA was performed. The values represent the rate of induction of apoptosis compared to control (siControl). *, significantly different from siControl, P < 0.05. messengers in cancer cells. A recent report showed that monoacylglycerol lipase (MAGL) plays a critical role in liberating stored fats to promote cancer progression [32]. In addition, this enzyme produces free fatty acids, leading to the induction of specific lipids such as prostaglandin E2 (PGE2) and lysophosphatidic acid (LPA), which act as signaling molecules in the migration and invasion of cancer cells. Interestingly, this enzyme is highly expressed in aggressive cancer cells and regulates a fatty acid network that promotes migration, invasion, and survival [32]. As FABP5 is involved in binding and sorting free fatty acids, as well as in transporting them to the appropriate compartments in the cell, it would be interesting to examine whether FABP5 functionally associates with MAGL-dependent signaling pathway components to promote invasion in CRC cells. Furthermore, as shown in Fig. 3C, knockdown of FABP5 expression significantly suppressed MAGL and hormone-sensitive lipase (HSL) expression levels. Importantly, HSL, a rate-limiting enzyme in diacylglycerol (DG) hydrolysis, generates the MAGL substrate monoacylglycerol. This finding suggests that FABP5 might be involved in MAGL-dependent signaling in CRC cells. A recent study reported that the overexpression of FABP5 in oral squamous cell carcinoma increased cell invasiveness by increasing the expression of matrix metalloprotease-9 (MMP-9) [10]; however, knockdown of FABP5 expression did not significantly suppress MMP-9 gene expression in the present experimental conditions (data  not shown), suggesting that FABP5 may not be involved in the invasiveness mediated by MMP-9.
The functional association of FABP5 with PPAR b/d signaling might not be a major growthpromoting pathway in CRC cells Tumor progression caused by FABP5 overexpression may be the result of the activation of the PPAR b/d signaling pathway [33][34][35]. Briefly, FABP5-PPAR b/d signaling induced the expression of genes involved in cell growth and survival; however, it is unclear whether FABP5-dependent cell proliferation in CRC cells is attributable to this mechanism. To assess whether this signaling pathway functions in CRC cells, we examined the effect of FABP5 on PPAR b/d signaling using GW0742, a synthetic high-affinity PPAR b/d agonist. In HCT116 cells, activation of PPAR b/d with GW0742 resulted in an increase in the mRNA level of adipocyte differentiation-related protein (ADRP), a well-characterized PPAR (b/d and c) target gene [36], regardless of the FABP5 expression level (Fig. 4A). Furthermore, overexpression of FABP5 and treatment with GW0742 in Caco-2 cells did not influence ADRP expression level (Fig. 4B). These results suggest that FABP5 does not enhance PPAR b/d signaling and is not required for PPAR b/d ligand binding in CRC cells. GW0742 also failed to increase the expression of 3-phosphoinositide-dependent protein kinase-1 (PDPK1), which is a PPAR b/d target gene [37] and phosphorylates AKT/PKB (protein kinase B), leading to activation of survival signaling (Fig. 4A,B). This result is consistent with recent studies showing no changes in the expression of PDPK1 and phosphorylation of AKT in response to GW0742 in human HaCaT keratinocytes and CRC cells [38][39][40]. In addition, the levels of phosphorylated AKT were unchanged by knockdown of FABP5 (Fig. 4C,D) and high concentration of GW0742 could not rescue the siFABP5-mediated cell growth inhibition (data not shown). To further address the question of whether PPAR b/d mediates the oncogenic activities of FABP5, we investigated whether treatment with PPAR b/d antagonist affects CRC cell proliferation. Treatment with PPARb/d antagonist (GSK-3787) resulted in no significant decrease in cancer cell proliferation (Fig. 4E) and the expression levels of FABP5 (Fig. 4F). Thus, these results strongly suggest that PPAR b/d does not mediate the pro-oncogenic activities of FABP5. Indeed, several lines of evidence have shown that PPAR b/d signaling does not potentiate the growth of human cancer cell lines and attenuates colon carcinogenesis [39][40][41][42]. As shown in Fig. 4A,B, F, the FABP5 expression level did not significantly change in response to GW0742 or GSK-3787, suggesting that FABP5 gene expression is upregulated by a PPAR b/d-independent signaling pathway in CRC cells. Thus, these data are contradictory to the report showing that FABP5 is involved in PPAR b/d-dependent prostate cancer cell growth [34]. Moreover, knockdown of FABP5 expression resulted in the downregulation of fatty acid metabolizing genes, such as acetyl-CoA carboxylase a (ACCa) and isocitrate dehydrogenase 1 (IDH1) (Fig. 4G), suggesting that FABP5 is involved in metabolic alterations in CRC cells. In HCT116 cells, treatment with GW0742 or GW1929 (PPAR c agonist) had no significant effect on the expression levels of ACCa, FASN, and HSL (Fig. 4H). This result suggests that these lipid-metabolizing genes were not regulated by PPAR b/d or PPAR c in HCT116 cells. As ACCa is a rate-limiting enzyme in fatty acid biosynthesis, FABP5 might play a critical role in fatty acid metabolism in CRC cells. Importantly, as fatty acids are required for the formation of membrane components, energy sources, and the production of cell signaling molecules during cancer cell proliferation, FABP5 might play a pivotal role in cell proliferation by regulating fatty acid metabolism in CRC cells. Further mechanistic studies of FABP5-dependent growth in CRC cells are now in progress.
In conclusion, the present study demonstrated that the FABP5 gene is upregulated in CRC compared to normal colon in a manner that correlates with disease stage. FABP5 significantly promotes cell growth and invasion by a PPAR b/d-independent signaling pathway in CRC cells.