FABP5 is correlated with poor prognosis and promotes tumour cell growth and metastasis in clear cell renal cell carcinoma

European Journal of Pharmacology 862 (2019) 172637

Contents lists available at ScienceDirect

European Journal of Pharmacology journal homepage: www.elsevier.com/locate/ejphar

Full length article

FABP5 is correlated with poor prognosis and promotes tumour cell growth and metastasis in clear cell renal cell carcinoma

T

Guangzhen Wua,b,1, Yingkun Xua,1, Qifei Wangb, Jianyi Lia, Lin Lic, Chenglin Hana, Qinghua Xiaa,∗ a

Department of Urology, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, China Department of Urology, The First Affiliated Hospital of Dalian Medical University, Dalian, China c Department of Orthopedics, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, China b

ARTICLE INFO

ABSTRACT

Keywords: Fatty acid binding protein 5 Lipid metabolism Clear cell renal cell carcinoma

To support proliferation, tumour cells often undergo a metabolic switch to aerobic glycolysis, and a large amount of fatty acids (FAs) is produced to provide conditions for the formation of cell membrane structures. This phenomenon is particularly prominent in clear cell renal cell carcinoma (ccRCC). FAs need to be combined with fatty acid binding proteins (FABPs) for transport. Fatty Acid Binding Protein 5 (FABP5) is an important chaperone protein of FAs that is upregulated in a variety of tumours. However, to date, the potential regulatory role and molecular mechanisms of FABP5 in the development and progression of cancers, including ccRCC, remain unknown. Herein, we demonstrate that FABP5 is upregulated in human ccRCC tissues and cell lines and is positively correlated with the progression of ccRCC. FABP5 deletion inhibits the proliferation, colony-forming ability and migration of ccRCC cells, suggesting that FABP5 may be a cancer-promoting protein in ccRCC. Mechanistically, FABP5 deletion significantly downregulated MMP9 and the transcription factor Snail1 in addition to upregulating E-cadherin and downregulating N-cadherin and Vimentin to inhibit epithelial-mesenchymal transition (EMT) in the ACHN cell line. In summary, our data suggest that FABP5 may be a potential therapeutic target in ccRCC.

1. Introduction Approximately 300,000 people worldwide are diagnosed with kidney cancer each year (Fitzmaurice et al., 2015). The most common histological subtype of renal cell carcinoma is clear cell renal cell carcinoma (ccRCC), a tumour that originates in the proximal tubular epithelial cells of nephrons (Frew and Moch, 2015; Mitchell et al., 2018). Approximately one-third of ccRCC patients develop metastatic disease, which is called metastatic renal cell carcinoma (mRCC) and has a 5-year survival rate of less than 10% (Han et al., 2018; M et al., 2005). Although great progress has been made in the genetic research and targeted therapy of ccRCC, the cell phenotypes and corresponding molecular mechanisms have not been fully identified (Du et al., 2017). ccRCC is histologically defined as having clear cytoplasm characteristics because ccRCC specimens contain large amounts of lipids and glycogen; thus, lipid metabolism may be the future direction of ccRCC research (Network, 2013). To meet energy metabolism needs and support rapid proliferation, tumour cells often undergo a metabolic

switch to aerobic glycolysis, also known as the Warburg effect. The glycolytic products produced as a result of the Warburg effect are used in de novo synthesis of fatty acids (FAs), which are necessary for the rapid growth of cancer cells. Recent studies have found that the Warburg effect is more pronounced in ccRCC than in other cancers (Courtney et al., 2018). However, the process by which fatty acids are used to form membrane structure requires the participation of fatty acid-binding proteins (FABPs) as chaperone proteins. FABPs are a family of low molecular weight, intracellular lipid-binding proteins involved in the uptake and transport of FAs (Currie et al., 2013; RL and DR, 2011) and play an important role in regulating the expression of genes related to cell proliferation, invasion and migration (Storch and Corsico, 2008; Storch and Thumser, 2010). Therefore, FABPs may be involved in the metabolic regulation of FAs during the process of carcinogenesis (S et al., 2018). Fatty Acid Binding Protein 5 (FABP5) is a low molecular weight protein (15 kD) that binds to a variety of FAs and long-chain FAs(Storch and Corsico, 2008). FABP5 transports lipids into cells for storage and cell biofilm synthesis (Furuhashi and Hotamisligil,

Corresponding author. Department of Urology, Shandong Provincial Hospital Affiliated to Shandong University, 9677 Jingshidong Road, Jinan City, 250001, Jinan, Shandong Province, China. E-mail address: [email protected] (Q. Xia). 1 Guangzhen Wu and Yingkun Xu contribute equally to the paper. ∗

https://doi.org/10.1016/j.ejphar.2019.172637 Received 20 June 2019; Received in revised form 28 August 2019; Accepted 2 September 2019 Available online 03 September 2019 0014-2999/ © 2019 Published by Elsevier B.V.

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

2008). Recent studies have shown that FABP5 contributes to cell migration and proliferation in a variety of cancers, such as prostate cancer, oesophageal cancer, squamous cell carcinoma and breast cancer (Alshalalfa et al., 2012; EA et al., 2008; Kowalewska et al., 2012; Levi et al., 2013; Ogawa et al., 2008; Schug et al., 2008). Recent literature also support our results in the 786-O cell line (Lv et al., 2019). Here, we consider that FABP5 has different roles in different ccRCC cell lines. In our study, we found that FABP5 expression is elevated in ccRCC tissues and cells and is positively correlated with the progression of ccRCC. Knockout of FABP5 in ccRCC cells significantly inhibited the proliferation, colony-forming and migration abilities. In addition, we found that FABP5 triggers EMT in the ACHN cell line by affecting the expression of Snail, MMP9, vimentin, E-cadherin, N-cadherin and other proteins.

450 nm with a microplate reader.

2. Materials and methods

Cells were seeded in a 6-well plate. After starvation, cells were scratched with a 200 μl pipette tip. Cells were washed to remove debris, and serum-free medium was added. Images were acquired at regular intervals to measure wound healing rates.

2.5. Immunohistochemistry ccRCC specimens were fixed with 4% paraformaldehyde and embedded in paraffin for sectioning. Antigen retrieval was performed with 10 mM citrate buffer (pH 6.0, 13 min). For labelling, sections were incubated with rabbit polyclonal antibody (anti-FABP5) overnight at 4 °C. Sections were incubated with secondary antibodies for 30 min and detected using a DAB kit (Abcam, ab64238).), and nuclei were stained with haematoxylin (Abcam, ab143166) according to the manufacturer's protocol. The results were evaluated by two independent pathologists. 2.6. Wound healing assay

2.1. Cell lines, siRNA knockdown, antibodies and reagents The human ccRCC cell lines 786-O, ACHN, and OS-RC-2 and renal tubular epithelial HK2 cells were purchased from the Cell Bank of the Chinese Academy of Sciences. All cells were cultured according to the manufacturer's protocol. 786-O and OS-RC-2 cells were cultured in RPMI 1640 medium containing 10% foetal bovine serum, and ACHN and HK2 cells were cultured in high-glucose Dulbecco's Modified Eagle Medium (DMEM) containing 10% foetal bovine serum. All cells were cultured at 37 °C in humidified air with 5% CO2. 786-O and ACHN cells were transfected with 20 nmol/L siRNAs (Invitrogen, 12935–200 and 12935–300) using Lipofectamine RNAiMAX reagent (Invitrogen). The antibodies used included rabbit anti-FABP5 (Proteintech, 12348), rabbit anti-β-actin (Abcam, 8227), rabbit anti-MMP9 (Abcam, 38898), rabbit anti-N-cadherin (Abcam, 18203), rabbit anti-E-cadherin (Abcam, 15148), rabbit anti-snail + slug (Abcam, 85936), and rabbit antiVimentin (Abcam, 92547).

2.7. Transwell invasion assay Medium adapted to the cell line and containing 10% serum was added to a 24-well plate. Transwell chambers (Corning, USA) were arranged in the wells. A mixture of 50 μl of BD Matrigel and medium adapted to the cell line (1:8) was added to each Transwell chamber and incubated for 3 h in a cell culture incubator. Cells were suspended in medium completely adapted to the cell line at a final concentration of 5 × 104 cells/ml. A 200 μl aliquot of cell suspension was added to each chamber. The 24-well plate was placed in a cell culture incubator. The Transwell chambers were removed after 24 h of culture and were then stained and fixed. The field of view was randomly selected under a microscope, and cells were imaged and counted. 2.8. Western blotting

2.2. Colony formation assay

After cells were grown in a culture dish to 70% confluence, they were lysed with RIPA lysis buffer (CST, 9806), and the protein concentration was determined by the BCA method (Abcam, ab102536). Protein samples (30 μg) were subjected to 4–20% SDS-PAGE (Keygen Biotech, KGP113K), transferred to a PVDF membrane (Millipore, R8CA8257B) and incubated with primary antibodies overnight, with βactin as the loading control. Then, the membrane was incubated with peroxidase-conjugated secondary antibody, and the immunoreactive bands were observed by an ECL system.

A mixture of 500 cells in medium was added to a six-well plate. Cells were cultured for 21 days. The medium was changed every 3 days. After colonies with more than 50 cells formed, culture was terminated. Cells were fixed in 4% paraformaldehyde, stained with crystal violet (MCE, B0324A) and photographed after rinsing with double distilled water. 2.3. RNA extraction and qRT-PCR Total RNA was extracted with Trizol reagent and reverse transcribed to cDNA using a PrimeScript RT reagent kit (Takara, Japan). mRNA levels were measured using SYBR Premix Ex Taq (Takara, Japan). All procedures followed the manufacturer's instructions, and each experiment was repeated three times. The primer sequences used in this manuscript are as follows: FABP5 (forward, 5′-CCTGTCCAAAGTGATGATGG-3′ and reverse, 5′-CAGCATCAGGAGTGGGATG-3′); MMP2 (forward, 5′-TACAGGATCA TTGGCTACACACC-3′ and reverse, 5′-GGTCACATCGCTCCAGACT-3′); Snai2 (forward, 5′-ACTCCGAAGCCAAATGACAA-3′ and reverse, 5′-CTCTCTCTGTGGGTGTGTGT-3′); and β-actin (forward, 5′- TGGCAC CCAGCACAATGAA-3′ and reverse, 5′-CTAAGTCATAGTCCGCCTAGAA GCA-3′).

2.9. Statistics All experiments were repeated at least three times. At least two investigators independently completed each experiment and statistically analysed the corresponding data with Prism software (GraphPad, CA, USA). Statistical significance was evaluated by a two-tailed t-test and one-way ANOVA between the two groups. When the variance between the groups was nonuniform, Welch's correction was used; a Pvalue of < 0.05 was considered statistically significant. 3. Results

2.4. CCK8 assay

3.1. The mRNA expression of FABP5 is upregulated in a variety of tumours and is associated with progression in ccRCC

Cells were seeded in 96-well culture plates and cultured at 2 × 103 cells/well with 6 replicate wells in each group; blank wells were used as controls. After the indicated time, 10 μl of CCK8 solution (Dojindo, Japan) was added to each well, and after incubation for 2 h, the optical density (OD) of each well was measured at a wavelength of

To investigate the differential expression of FABP5 in multiple tumours and corresponding adjacent tissues, we investigated the difference in FABP5 expression between different tumours and the corresponding normal tissues through the UALCAN website (UALCAN: http://ualcan.path.uab.edu) (Chandrashekar et al., 2017) (Fig. 1A 2

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

(caption on next page)

3

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 1. FABP5 is upregulated in a variety of tumours and is associated with progression in ccRCC. A.B. Expression of FABP5 across TCGA tumours and expression of FABP5 across TCGA cancers (with tumour and normal samples) (Chandrashekar et al., 2017). C. Analysis of FABP5 mRNA expression levels in cancer tissues and paracancer tissues in the stomach adenocarcinoma (STAD), head and neck squamous cell carcinoma (HNSC), oesophageal carcinoma (ESCA), prostate adenocarcinoma (PRAD), lung squamous cell carcinoma (LUSC), and liver hepatocellular carcinoma (LIHC) datasets in the TCGA database (Chandrashekar et al., 2017). ***P < 0.001. D. The TCGA database accessed through the UALCAN website was used to analyse the difference in mRNA levels of FABP5 between ccRCC and paracancer tissues according to KIRC subtypes, tumour grades and individual cancer stages (Chandrashekar et al., 2017). **P < 0.01, ***P < 0.001. E. The TCGA database accessed through the UALCAN website was used to analyse the relationship between FABP5 expression and the patient survival rate (Chandrashekar et al., 2017) (P = 0.0026). F. The relationship between FABP5 expression and the patient survival rate was analysed using the TCGA database on THE HUMAN PROTEIN ATLAS website (Uhlen et al., 2017) (P = 4.7e-10).

and B). We also used the TCGA database to analyse stomach adenocarcinoma (STAD), head and neck squamous cell carcinoma (HNSC), oesophageal carcinoma (ESCA), prostate adenocarcinoma (PRAD), lung squamous cell carcinoma (LUSC), and liver hepatocellular carcinoma (LIHC) datasets. We found that the mRNA expression levels of FABP5 were significantly higher in the above tumour tissues than in the corresponding adjacent tissues (Fig. 1C). Because the Warburg effect is particularly prominent in clear cell renal cell carcinoma (ccRCC) and downstream products resulting from the Warburg effect, such as fatty acids (FAs), need to function through the chaperone protein FABP5(Furuhashi and Hotamisligil, 2008; S et al., 2018; Storch and Corsico, 2008), we hypothesized that FABP5 is also upregulated in ccRCC. Then, we analysed FABP5 mRNA expression in ccRCC through the UALCAN website (UALCAN: http://ualcan.path.uab.edu) (Chandrashekar et al., 2017; Uhlen et al., 2017) (Fig. 1D) and found that FABP5 is upregulated and closely related to tumour subtype, tumour stage and tumour grade in ccRCC. We used two different databases to analyse the correlation between FABP5 expression and survival in renal clear cell carcinoma (UALCAN: http://ualcan.path.uab.edu and THE HUMAN PROTEIN ATLAS: https://www.proteinatlas.org) (Chandrashekar et al., 2017; Uhlen et al., 2017) (Fig. 1E and F). The above data suggested that FABP5 is a critical gene during the carcinogenesis and malignant progression of ccRCC.

in the cytoplasm and that its expression level in cancer tissues is significantly higher than that in surrounding normal tissues (Fig. 3A, B, 3C). We further analysed the expression of FABP5 in cancer tissues and adjacent tissues by qRT-PCR and immunoblotting, and we found that the results were consistent with those of immunohistochemistry. The mRNA and protein levels of FABP5 in cancer tissues were significantly higher than those in adjacent tissues (Fig. 3E, F, 3G). To further validate the overexpression of FABP5 in ccRCC cells, we used three ccRCC cell lines (ACHN, 786-O, and OS-RC-2) and HK-2 cells (a proximal tubular epithelial cell line). qRT-PCR and western blotting were performed to measure the expression of FABP5 in these four cell lines. We found that the mRNA and protein levels of FABP5 in the three ccRCC cell lines was significantly higher than that in HK2 cells (Fig. 3H, I, 3J). Additionally, survival curve analysis showed that patients with higher FABP5 expression (> the median H score) had significantly lower survival rates than patients with lower FABP5 expression (< the median H score) (Fig. 3D). Correlation analysis showed that a higher expression level of FABP5 was positively correlated with a larger tumour size and higher grade of ccRCC (Table 1). Taken together, these results indicate that FABP5 is overexpressed in ccRCC tissues and cell lines and is positively correlated with the progression of human ccRCC. 3.4. FABP5 and EMT-related genes have co-expression relationships in ccRCC

3.2. FABP5 expression is upregulated in the ONCOMINE database and associated with disease-free survival in ccRCC

In the above studies, we found that FABP5 may be closely related to the progression of human ccRCC as an oncogene, but the mechanism by which it promotes tumour progression remained unclear. Therefore, we analysed nine genes co-expressed with the FABP5 gene via TCGA database analysis (Chandrashekar et al., 2017). We found that FABP5 is co-expressed with a variety of genes involved in tumour progression, such as the classical TGF pathway genes TGFB3 and TGF1I1; ANGPTL2, which increases tumour angiogenesis and increases tumour invasion capacity (Aoi et al., 2011; Lin et al., 2015); DPYSL3, which regulates mitotic transition, cell migration, and EMT (Matsunuma et al., 2018); CPT1C, which is associated with endoplasmic reticulum stress and EMT induction (N et al., 2016); and the classical genes related to EMT, such as MMP2, MMP14, SNAI2 and the cytokine IL6 (Fig. 4A and B) (DJ et al., 2018; Nguyen et al., 2017). We further analysed the genes associated with FABP5 through the String website (https://string-db.org) (Szklarczyk et al., 2019). We found that FABP5 is associated with ACACA, FASN and ACLY. These genes are key rate-limiting enzymes in FA synthesis and play an important role in this process. In addition, FABP5 was associated with the transcription factors PPARD and PPARG. However, the specific interaction mechanism still needs further proof. In summary, we found that FABP5 is likely to be associated with EMT in ccRCC, so we verified the levels of the classical EMT-related genes MMP2 and SNAI2 by qRT-PCR in ACHN cells. We found that the

To investigate the role of FABP5 in human ccRCC, we first mined the ONCOMINE database. We performed a meta-analysis of the following six studies in the ONCOMINE database: Gumz Renal, Higgins Renal, Jones Renal, Lenburg Renal, BMC cancer, TCGA Renal 2 and Yusenko Renal (Fig. 2A). In the development of ccRCC, FABP5 is an overexpressed gene that plays an important role. Similarly, we analysed the Jones Renal Statistics dataset and found that the expression of FABP5 in ccRCC tumour tissues was significantly higher than that in normal kidney tissues (Fig. 2B). We found the same expression levels of FABP5 in Bittner Multi-cancer Statistics, FABP5 Expression in Barretina CellLine and Garnett CellLine Statistics (Fig. 2C, D, and 2E, respectively). The relationship between FABP5 expression and the disease-free survival rate was analysed using the GEPIA2 website (http://gepia2. cancer-pku.cn/#index) (Tang et al., 2019) (Fig. 2F). 3.3. FABP5 expression is upregulated and associated with poor survival in ccRCC To investigate the role of FABP5 in human ccRCC, we first analysed the expression of FABP5 in 60 cancer tissues and adjacent tissues by immunohistochemistry. We found that the FABP5 protein is expressed

4

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 2. FABP5 expression is upregulated in the ONCOMINE database and associated with disease-free survival in ccRCC A. Comparison of all genes across 6 studies: Gumz Renal, Higgins Renal, Jones Renal, Lenburg Renal, BMC cancer, TCGA Renal 2 and Yusenko Renal. B. Analysis of Jones Renal Statistics: 1. Kidney (23 cases); 2. Clear Cell Renal Cell Carcinoma (23 cases). C, D, E. The expression of FABP5 in Bittner Multi-cancer Statistics, Barretina CellLine and Garnett CellLine Statistics. F. The relationship between FABP5 expression and the disease-free survival rate was analysed using the GEPIA2 website (Tang et al., 2019) (P = 0.024). 5

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 3. FABP5 expression is upregulated and associated with poor survival in ccRCC A. Immunohistochemical analysis of 3 typical ccRCC samples with FABP5 expression; the scale bar represents 100 μm. B.C. Three typical cases and all 90 cases were quantified by the H score, and the data are shown as the mean ± S.E.M. Statistical significance was assessed with Student's t-test. *P < 0.01. D. Survival curves of ccRCC patients with high (> the median H scores) (n = 31) and low (< the median) H scores (n = 29). Statistical significance was determined by the log-rank test (P = 0.007). E. qRT-PCR analysis of the difference in FABP5 mRNA levels in 9 samples of ccRCC tissue and adjacent tissues. The data are shown as the mean ± S. E.M. Statistical significance was assessed with a two-tailed Student's t-test. ***P < 0.001. F, G. Western blotting. The protein levels of FABP5 in 6 ccRCC tissues and adjacent tissues are shown in the immunoblot, and the data are shown as the mean ± S.E.M. Statistical significance was assessed with a two-tailed Student's t-test. ***P < 0.001. H. qRT-PCR analysis of the difference in FABP5 mRNA levels in HK2 cells and three ccRCC cell lines. The data are shown as the mean ± S. E.M. Statistical significance was assessed with a two-tailed Student's t-test. **P < 0.01, ***P < 0.001. I, J. Immunoblot analysis of the difference in FABP5 protein levels between HK2 cells and three ccRCC cell lines. The data are shown as the mean ± S.E.M. Statistical significance was assessed with a two-tailed Student's t-test. *P < 0.05, **P < 0.01.

6

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

0.5989

expression of the mesenchymal phenotype proteins N-cadherin and Vimentin and the matrix metalloproteinase MMP9 was significantly downregulated. We analysed the upstream transcription factors associated with EMT, snail and slug, by immunoblotting. The results showed that silencing of FABP5 significantly inhibited the expression of snail and slug compared with that in the control group. Since FABP5 knockdown did not inhibit EMT in 786-O cells, the 786-O related WB results are not shown. Taken together, these data indicate that deletion of FABP5 attenuates EMT in ACHN cells by affecting the levels of the abovementioned proteins (Fig. 6G and H).

0.8090

4. Discussion

Table 1 Association between FABP5 level in ccRCC tissues and clinicopathological features (n = 60). Features

Gender Male Female Age ≤60 > 60 Location side Left Right Tumor size (cm) ≤4.0 > 4.0 Furman grade G1+G2 G3+G4

Number

FABP5 expression

P

Low (n = 29)

High (n = 31)

41 19

19 10

22 9

36 24

16 13

20 11

28 32

14 15

14 17

35 25

22 7

13 18

47 13

28 1

19 12

0.7827

To meet the needs of membrane structure formation, tumour cells need to undergo metabolic rearrangement, that is, a switch to metabolizing glucose by aerobic glycolysis (Warburg effect). This effect is particularly pronounced in clear cell renal cell carcinoma (ccRCC), and the glycolytic metabolites can be used in de novo synthesis of lipids, resulting in the formation of large amounts of fatty acids (FAs) (Courtney et al., 2018; Fritz et al., 2010; Huang et al., 2012). These FAs are capable of being incorporated into the membrane structures and lipid rafts required for cell proliferation and constitute a second messenger in cellular signalling pathways. These substances are involved in the processes of cell proliferation, invasion and migration. However, the above effects of FAs require the participation of chaperones. Fatty acid binding protein 5 (FABP5) is an important chaperone protein through which fatty acids perform their functions and is involved in the binding and storage of hydrophobic ligands such as long-chain FAs. Therefore, we hypothesized that FABP5 also plays an important role in the development and progression of ccRCC (Aoi et al., 2011; S et al., 2018; Storch and Corsico, 2008). In this study, we identified the role of FABP5 in human ccRCC and the underlying mechanism of FABP5. We found that FABP5 is highly expressed in human ccRCC tissues and cell lines. FABP5 is closely related to tumour size and grade, and high expression of FABP5 is an important risk factor for poor survival in ccRCC patients. In addition, FABP5 promoted the growth of ccRCC tumour cells in vitro. Deletion of FABP5 inhibited the proliferation, colony-forming and migration ability of ccRCC cells in vitro, suggesting that FABP5 may be a tumour-promoting protein in ccRCC. In addition, we found that FABP5 is not only upregulated in ccRCC but also upregulated in stomach adenocarcinoma, head and neck squamous cell carcinoma, oesophageal carcinoma, pancreatic adenocarcinoma, lung squamous cell carcinoma and liver hepatocellular carcinoma. This finding is consistent with the observation that tumour cell proliferation requires an abundance of FAs and corresponding chaperone proteins participating in membrane structure formation. Moreover, we found that FABP5 is co-expressed with various oncogenes—genes involved in tumour progression, such as TGFB3, ANGPTL3, CPT1A and DPYSL3, as well as genes encoding the EMTassociated proteins MMP2, SNAI2, MMP14 and IL6. Therefore, we hypothesized that FABP5 may induce epithelial-mesenchymal transition (EMT) by activating a specific pathway or mechanism (Fig. 7). Of course, these hypotheses still require further supporting experimental evidence. As low molecular weight proteins in cells, fatty acid binding proteins (FABPs) can enter the nucleus under certain conditions and target transcription factors, such as the molecules PPAR-α, PPAR-δ and PPARγ, which are members of the peroxisome proliferator-activated receptor (PPAR) family. FABPs are controlled by these transcription factors.

0.0098* 0.0012*

Analysis was carried out with Fisher's exact t text, symbol * indicates statistically significant (p < 0.05).

mRNA levels of MMP2 and SNAI2 were significantly decreased after FABP5 knockdown (Fig. 4C and D). In summary, FABP5 may participate in the induction of EMT in ccRCC cells through a specific molecular mechanism. 3.5. FABP5 promotes the growth of ccRCC cells in vitro To confirm that FABP5 overexpression promotes the proliferation of ccRCC cells, we performed CCK8 assays using ACHN and 786-O cells. The results showed that knockdown of FABP5 significantly inhibited cell proliferation in vitro (Fig. 5A and B). To show whether FABP5 knockout can affect the long-term proliferation of ccRCC cells, we assessed colony formation at defined time points. We found that the ability of 786-O and ACHN cells to form colonies was significantly impaired after FABP5 knockdown (Fig. 5C). In general, these results indicate that FABP5 may play an important role in ccRCC cell proliferation as an oncogene. 3.6. Depletion of FABP5 suppresses migration and blocks EMT in ACHN but not 786-O cells After observing significant changes in cell proliferation following FABP5 depletion, we used a series of experiments to determine whether knockdown of FABP5 affects the migration ability of ccRCC cells. Wound healing assays showed that the migration of FABP5-depleted 786-O cells was similar to that of the vector-transfected control group (Fig. 6A and B). However, the wound healing assay showed that the migration rate of FABP5-depleted ACHN cells was significantly lower than that of the vector-transfected control group (Fig. 6C and D). The Transwell invasion assays showed the same experimental results as the wound healing assays (Fig. 6E and F). To determine the underlying molecular mechanism by which FABP5 deletion produces inhibitory effects on proliferation and migration in cancer cells, we investigated the effect of FABP5 knockdown on EMT in ACHN cells. Immunoblotting revealed that the epithelial phenotype protein E-cadherin was significantly upregulated in ACHN cells after knockdown of FABP5. The

7

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 4. FABP5 expression is upregulated and associated with poor survival in ccRCC A. The UALCAN website was used to analyse 9 genes co-expressed with the FABP5 gene (Chandrashekar et al., 2017). B. The Pearson CC score of the FABP5 gene and its co-expressed genes (Chandrashekar et al., 2017). C. The STRING website was used to analyse the interaction of proteins with FABP5-related proteins (Szklarczyk et al., 2019). D. qRT-PCR confirmed the difference in the mRNA levels of the FABP5, MMP2 and SNAI2 genes between the siFABP5 group and siControl group in the ACHN cell line. The data are shown as the mean ± S. E.M. of the values from triplicate experiments. Statistical significance was assessed with a two-tailed Student's t-test. **P < 0.01, ***P < 0.001.

These transcription factors act as ligands for fatty acids or other hydrophobic agonists (Furuhashi and Hotamisligil, 2008). In highly invasive prostate and breast cancer cells, FABP5 upregulates the expression of genes such as HSL, MAGL, ELOVL6 and ACSLL, which are involved in FA synthesis, triglyceride lipolysis and FA metabolism; in

addition, FABP5 activates the NFκB signalling pathway, thereby promoting tumour proliferation and the metastasis of prostate and breast cancer cells (S et al., 2018). Epithelial-mesenchymal transition (EMT) is composed of multiple dynamic transition states between the epithelial and mesenchymal

8

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 5. FABP5 promotes the growth of ccRCC cells in vitro A.B. CCK8 assay. The OD values of the siFABP5 group and siControl group in 786-O cells and ACHN cells were determined by a CCK8 assay, and the relative growth of each group was assessed. The data are shown as the mean ± S.E.M. ***P < 0.001. C. Colony formation assay. Colonies were counted with Image Pro Plus software. The data are shown as the mean ± S.E.M. The experiment was repeated three times. Statistical significance was assessed using a two-tailed t-test. *P < 0.05.

phenotypes and plays a crucial role in regulating tumour progression and metastasis (MA et al., 2016). One important finding reported here is that deletion of FABP5 effectively blocked the induction of EMT by upregulating the expression of E-cadherin and downregulating the expression of N-cadherin and Vimentin in ACHN cells. We further showed that FABP5 knockdown had a significant inhibitory effect on the expression of the EMT-induced transcription factors Snail and slug. Taken together, our results indicate that deletion of FABP5 attenuates EMT in ACHN cells by downregulating EMT-related proteins and inhibiting the expression of transcription factors. In addition, matrix metalloproteinase-9 (MMP9), which is closely linked to EMT, can enhance EMT by modifying the Extracellular matrix (ECM) and upregulating the expression of transcription factors associated with EMT, such as Snail and slug, thereby inducing EMT (P et al., 2012; Radisky et al., 2005). Upregulation of transcription factors such as snail and slug can activate MMP9 expression, which further upregulates the expression of EMTrelated transcription factors, resulting in positive feedback regulation (Craene and Berx, 2013). In the present study, we found that knockdown of FABP5 in the ACHN cell line significantly downregulated MMP9 protein expression, suggesting that knockdown of FABP5 may impair the abovementioned positive feedback regulatory loop, thereby blocking EMT. In addition, recent literature support our results in the 786-O cell line (Lv et al., 2019). Here, we believe that FABP5 has a

different role in different ccRCC cell lines. The specific mechanism requires further research. Above all, our study found that FABP5 is an important regulator of ccRCC cell proliferation and metastasis and indicated that FABP5 may regulate the development and progression of ACHN cells by mediating signalling pathways involved in EMT. Of course, the underlying mechanism still needs further study. In summary, these data suggest that FABP5 may be a new therapeutic target in ccRCC. Author contributions Qinghua Xia and Guangzhen Wu conceived the experiments; Guangzhen Wu and Yingkun Xu performed the experiments; Qifei Wang, Jianyi Li, Lin Li and Chenglin Han analysed the data; Guangzhen Wu and Yingkun Xu wrote the manuscript. All of the authors have reviewed the manuscript and approved the submitted version. Conflicts of interest The authors declare that they have no conflicts of interest.

9

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

(caption on next page)

10

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al.

Fig. 6. Depletion of FABP5 suppresses migration and blocks EMT in ccRCC but not 786-O cells. A. B. C. D. Wound healing assays. The wound closure rate was calculated with Image Pro Plus software. Scale bar: 200 μm. NS = nonsignificant, *P < 0.05. E. F. Transwell invasion assays. The number of invaded cells was calculated with Image Pro Plus software. Scale bar: 200 μm. NS = nonsignificant, ***P < 0.001. G. Immunoblotting revealed FABP5, MMP9, E-cadherin, N-cadherin, vimentin, Snail and slug protein levels in the ACHN cell line. β-Actin was used as the loading control. H. The content of the above proteins was determined with Image Pro Plus software, and all data are shown as the average ± S.E.M. The experiment was repeated at least three times. Statistical significance was assessed using a two-tailed t-test. **P < 0.01, ***P < 0.001.

Fig. 7. Hypothesized mechanism by which FABP5 induces EMT (requires further experimental proof).

Acknowledgments

Werdecker, A., Gessner, B.D., Te Ao, B., McMahon, B., Karimkhani, C., Yu, C., Cooke, G.S., Schwebel, D.C., Carpenter, D.O., Pereira, D.M., Nash, D., Kazi, D.S., De Leo, D., Plass, D., Ukwaja, K.N., Thurston, G.D., Yun Jin, K., Simard, E.P., Mills, E., Park, E., Catalá-López, F., DeVeber, G., Gotay, C., Khan, G., Hosgood, H.D., Santos, I.S., Leasher, J.L., Singh, J., Leigh, J., Jonas, J.B., Sanabria, J., Beardsley, J., Jacobsen, K.H., Takahashi, K., Franklin, R.C., Ronfani, L., Montico, M., Naldi, L., Tonelli, M., Geleijnse, J., Petzold, M., Shrime, M.G., Younis, M., Yonemoto, N., Breitborde, N., Yip, P., Pourmalek, F., Lotufo, P.A., Esteghamati, A., Hankey, G.J., Ali, R., Lunevicius, R., Malekzadeh, R., Dellavalle, R., Weintraub, R., Lucas, R., Hay, R., Rojas-Rueda, D., Westerman, R., Sepanlou, S.G., Nolte, S., Patten, S., Weichenthal, S., Abera, S.F., Fereshtehnejad, S., Shiue, I., Driscoll, T., Vasankari, T., Alsharif, U., RahimiMovaghar, V., Vlassov, V.V., Marcenes, W.S., Mekonnen, W., Melaku, Y.A., Yano, Y., Artaman, A., Campos, I., MacLachlan, J., Mueller, U., Kim, D., Trillini, M., Eshrati, B., Williams, H.C., Shibuya, K., Dandona, R., Murthy, K., Cowie, B., Amare, A.T., Antonio, C.A., Castañeda-Orjuela, C., van Gool, C.H., Violante, F., Oh, I., Deribe, K., Soreide, K., Knibbs, L., Kereselidze, M., Green, M., Cardenas, R., Roy, N., Tillmann, T., Li, Y., Krueger, H., Monasta, L., Dey, S., Sheikhbahaei, S., Hafezi-Nejad, N., Kumar, G.A., Sreeramareddy, C.T., Dandona, L., Wang, H., Vollset, S.E., Mokdad, A., Salomon, J.A., Lozano, R., Vos, T., Forouzanfar, M., Lopez, A., Murray, C., Naghavi, M., 2015. The global burden of cancer 2013. JAMA Oncol. 1, 505. Frew, I.J., Moch, H., 2015. A clearer view of the molecular complexity of clear cell renal cell carcinoma. Annu. Rev. Pathol. 10, 263–289. Fritz, V., Benfodda, Z., Rodier, G., Henriquet, C., Iborra, F., Avances, C., Allory, Y., de la Taille, A., Culine, S., Blancou, H., Cristol, J.P., Michel, F., Sardet, C., Fajas, L., 2010. Abrogation of de novo lipogenesis by stearoyl-CoA desaturase 1 inhibition interferes with oncogenic signaling and blocks prostate cancer progression in mice. Mol. Cancer Ther. 9, 1740–1754. Furuhashi, M., Hotamisligil, G.S., 2008. Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets. Nat. Rev. Drug Discov. 7, 489–503. Han, Z., Zhang, Y., Sun, Y., Chen, J., Chang, C., Wang, X., Yeh, S., 2018. Erβ-mediated alteration of circATP2B1 and miR-204-3p signaling promotes invasion of clear cell renal cell carcinoma. Cancer Res. 78, 2550–2563. Huang, W.C., Li, X., Liu, J., Lin, J., Chung, L.W.K., 2012. Activation of androgen receptor, lipogenesis, and oxidative stress converged by SREBP-1 is Responsible for regulating growth and progression of prostate cancer cells. Mol. Cancer Res. 10, 133–142. Kowalewska, M., Radziszewski, J., Goryca, K., Bujko, M., Oczko-Wojciechowska, M., Jarzab, M., Siedlecki, J.A., Bidzinski, M., 2012. Estimation of groin recurrence risk in patients with squamous cell vulvar carcinoma by the assessment of marker gene expression in the lymph nodes. BMC Canc. 12, 223. Levi, L., Lobo, G., Doud, M.K., von Lintig, J., Seachrist, D., Tochtrop, G.P., Noy, N., 2013.

This work was funded by a grant from the National Natural Science Foundation of China (grant serial number: 81672553). References Alshalalfa, M., Bismar, T.A., Alhajj, R., 2012. Detecting cancer outlier genes with potential rearrangement using gene expression data and biological networks. Adv. Bioinformat. 2012, 1–13. Aoi, J., Endo, M., Kadomatsu, T., Miyata, K., Nakano, M., Horiguchi, H., Ogata, A., Odagiri, H., Yano, M., Araki, K., Jinnin, M., Ito, T., Hirakawa, S., Ihn, H., Oike, Y., 2011. Angiopoietin-like protein 2 is an important facilitator of inflammatory carcinogenesis and metastasis. Cancer Res. 71, 7502–7512. Chandrashekar, D.S., Bashel, B., Balasubramanya, S.A.H., Creighton, C.J., PonceRodriguez, I., Chakravarthi, B.V.S.K., Varambally, S., 2017. UALCAN: a portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia 19, 649–658. Courtney, K.D., Bezwada, D., Mashimo, T., Pichumani, K., Vemireddy, V., Funk, A.M., Wimberly, J., McNeil, S.S., Kapur, P., Lotan, Y., Margulis, V., Cadeddu, J.A., Pedrosa, I., DeBerardinis, R.J., Malloy, C.R., Bachoo, R.M., Maher, E.A., 2018. Isotope tracing of human clear cell renal cell carcinomas demonstrates suppressed glucose oxidation in vivo. Cell Metab. 28, 793–800. Craene, B.D., Berx, G., 2013. Regulatory networks defining EMT during cancer initiation and progression. Nat. Rev. Cancer 13, 97–110. Currie, E., Schulze, A., Zechner, R., Walther, T.C., Farese, R.V., 2013. Cellular fatty acid metabolism and cancer. Cell Metab. 18, 153–161. DJ, M., GR, H., EA, M., HE, A., KL, L., 2018. Differential expression profiles of conserved Snail transcription factors in the mouse testis. Andrology-US 6, 362–373. Du, W., Zhang, L., Brett-Morris, A., Aguila, B., Kerner, J., Hoppel, C.L., Puchowicz, M., Serra, D., Herrero, L., Rini, B.I., Campbell, S., Welford, S.M., 2017. HIF drives lipid deposition and cancer in ccRCC via repression of fatty acid metabolism. Nat. Commun. 8. EA, M., SS, F., J, A., CS, F., H, F., M, I., C, B., PH, S., Y, K., 2008. Expression of cutaneous fatty acid-binding protein (C-FABP) in prostate cancer: potential prognostic marker and target for tumourigenicity-suppression. Int. J. Oncol. 32. Fitzmaurice, C., Dicker, D., Pain, A., Hamavid, H., Moradi-Lakeh, M., MacIntyre, M.F., Allen, C., Hansen, G., Woodbrook, R., Wolfe, C., Hamadeh, R.R., Moore, A.,

11

European Journal of Pharmacology 862 (2019) 172637

G. Wu, et al. Genetic ablation of the fatty acid-binding protein FABP5 suppresses HER2-induced mammary tumorigenesis. Cancer Res. 73, 4770–4780. Lin, M.I., Price, E.N., Boatman, S., Hagedorn, E.J., Trompouki, E., Satishchandran, S., Carspecken, C.W., Uong, A., DiBiase, A., Yang, S., Canver, M.C., Dahlberg, A., Lu, Z., Zhang, C.C., Orkin, S.H., Bernstein, I.D., Aster, J.C., White, R.M., Zon, L.I., 2015. Angiopoietin-like proteins stimulate HSPC development through interaction with notch receptor signaling. ELIFE 4. Lv, Q., Wang, G., Zhang, Y., Han, X., Li, H., Le, W., Zhang, M., Ma, C., Wang, P., Ding, Q., 2019. FABP5 regulates the proliferation of clear cell renal cell carcinoma cells via the PI3K/AKT signaling pathway. Int. J. Oncol. 54, 1221–1232. M, S., K, R., N, H., CG, S., M, S., 2005. Targeted agents for the treatment of advanced renal cell carcinoma. Curr. Drug Targets 6. MA, N., RY, H., RA, J., JP, T., 2016. EMT: 2016. Cell 166, 21–45. Matsunuma, R., Chan, D.W., Kim, B., Singh, P., Han, A., Saltzman, A.B., Cheng, C., Lei, J.T., Wang, J., Roberto Da Silva, L., Sahin, E., Leng, M., Fan, C., Perou, C.M., Malovannaya, A., Ellis, M.J., 2018. DPYSL3 modulates mitosis, migration, and epithelial-to-mesenchymal transition in claudin-low breast cancer. Proc. Natl. Acad. Sci. 115, E11978–E11987. Mitchell, T.J., Turajlic, S., Rowan, A., Nicol, D., Farmery, J.H.R., O Brien, T., Martincorena, I., Tarpey, P., Angelopoulos, N., Yates, L.R., Butler, A.P., Raine, K., Stewart, G.D., Challacombe, B., Fernando, A., Lopez, J.I., Hazell, S., Chandra, A., Chowdhury, S., Rudman, S., Soultati, A., Stamp, G., Fotiadis, N., Pickering, L., Au, L., Spain, L., Lynch, J., Stares, M., Teague, J., Maura, F., Wedge, D.C., Horswell, S., Chambers, T., Litchfield, K., Xu, H., Stewart, A., Elaidi, R., Oudard, S., McGranahan, N., Csabai, I., Gore, M., Futreal, P.A., Larkin, J., Lynch, A.G., Szallasi, Z., Swanton, C., Campbell, P.J., 2018. Timing the landmark events in the evolution of clear cell renal cell cancer: TRACERx renal. Cell 173, 611–623. N, C., V, Z., L., H., R, F., R, R., D, S., 2016. Carnitine palmitoyltransferase 1C: from cognition to cancer. Prog. Lipid Res. 61, 134–148. Network, C.G.A.R., 2013. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature 499, 43–49. Nguyen, A.T., Chia, J., Ros, M., Hui, K.M., Saltel, F., Bard, F., 2017. Organelle specific Oglycosylation drives MMP14 activation, tumor growth, and metastasis. Cancer Cell 32, 639–653. Ogawa, R., Ishiguro, H., Kuwabara, Y., Kimura, M., Mitsui, A., Mori, Y., Mori, R., Tomoda, K., Katada, T., Harada, K., Fujii, Y., 2008. Identification of candidate genes involved

in the radiosensitivity of esophageal cancer cells by microarray analysis. Dis. Esophagus 21, 288–297. P, N., MJ, B., DC, R., 2012. Epithelial-mesenchymal transition: general principles and pathological relevance with special emphasis on the role of matrix metalloproteinases. CSH Perspect. Biol. 4. Radisky, D.C., Levy, D.D., Littlepage, L.E., Liu, H., Nelson, C.M., Fata, J.E., Leake, D., Godden, E.L., Albertson, D.G., Angela Nieto, M., Werb, Z., Bissell, M.J., 2005. Rac1b and reactive oxygen species mediate MMP-3-induced EMT and genomic instability. Nature 436, 123–127. RL, S., DR, P., 2011. The human fatty acid-binding protein family: evolutionary divergences and functions. Hum. Genom. 5, 170–191. S, S., N, K., K, K., A., A., H, F., 2018. Fatty acid-binding protein 5 (FABP5) promotes lipolysis of lipid droplets, de novo fatty acid (FA) synthesis and activation of nuclear factor-kappa B (NF-κB) signaling in cancer cells. Biochim. Biophys. Acta Mol. Cell Biol. Lipids 1863, 1057–1067. Schug, T.T., Berry, D.C., Toshkov, I.A., Cheng, L., Nikitin, A.Y., Noy, N., 2008. Overcoming retinoic acid-resistance of mammary carcinomas by diverting retinoic acid from PPARbeta/delta to RAR. Proc. Natl. Acad. Sci. U. S. A. 105, 7546–7551. Storch, J., Corsico, B., 2008. The emerging functions and mechanisms of mammalian fatty acid-binding proteins. Annu. Rev. Nutr. 28, 73–95. Storch, J., Thumser, A.E., 2010. Tissue-specific functions in the fatty acid-binding protein family. J. Biol. Chem. 285, 32679–32683. Szklarczyk, D., Gable, A.L., Lyon, D., Junge, A., Wyder, S., Huerta-Cepas, J., Simonovic, M., Doncheva, N.T., Morris, J.H., Bork, P., Jensen, L.J., Mering, C.V., 2019. STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613. Tang, Z., Kang, B., Li, C., Chen, T., Zhang, Z., 2019. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res. 47, W556–W560. Uhlen, M., Zhang, C., Lee, S., Sjöstedt, E., Fagerberg, L., Bidkhori, G., Benfeitas, R., Arif, M., Liu, Z., Edfors, F., Sanli, K., von Feilitzen, K., Oksvold, P., Lundberg, E., Hober, S., Nilsson, P., Mattsson, J., Schwenk, J.M., Brunnström, H., Glimelius, B., Sjöblom, T., Edqvist, P., Djureinovic, D., Micke, P., Lindskog, C., Mardinoglu, A., Ponten, F., 2017. A pathology atlas of the human cancer transcriptome. Science 357 n2507.

12