Inhibition of carnitine palmitoyl transferase 1A-induced fatty acid oxidation suppresses cell progression in gastric cancer

Background: Gastric cancer (GC) has a high rate of metastasis which thereason leading to death. Carnitine palmitoyltransferase 1a (CPT1A) has been reported to play a critical obstacle to various types of cancer progression, which is an attractive focus in anti-cancer therapy. However, the underlying molecular mechanisms of CPT1A involved in GC have not been clarified unclear. Methods: To determine the expression of CPT1A in human GC tissues and cells and illustrate whether it is correlated with the clinical pathologic characteristics and prognosis in GC patients. Its roles and potential mechanisms in regulating tumor growth and invasion were evaluated by CPT1A knockdown/overexpression of GC cells in vitro . Results: Marked upregulation of CPT1A protein expression was observed in GC cells and tissues, which was associated with grade, pathological stage, lymph node metastasis and poor prognosis in patients with GC. CPT1A overexpression also promoted the proliferation, invasion, EMT process of GC cells. In addition, CPT1A upregulation activated GC cell FAO via increasing NADP + /NADPH ratio, whereas inhibiting of FAO abolished the effects of CPT1A on GC cell proliferation and migration. Conclusion: Our results examine that CPT1A-mediated FAO activation increases GC cell proliferation and migration, supporting that CPT1A is a useful prognostic biomarker and an attractive focus for GC.


Introduction
Globally, gastric cancer (GC) is a common malignancy and has the highest incidence and mortality in the digestive tract [1]. Although much headway has been made in terms of clinical treatment in the late years, the prognosis of GC patients is still unsatisfactory, and the 5-year survival rate is poor [2,3]. Although several molecular markers have been proposed to detect GC diagnosis and metastasis, such as Long Noncoding RNA GMAN [4], and ER gamma [5], IRTKS [6], none has been approved for clinical use. Thus, exploring novel targets and the underlying mechanisms of GC development and metastasis is helpful for enhancing the early diagnosis of GC, even providing instructional clinical value for the treatment of metastatic GC.
Recently, the reprogramming of cellular energy metabolism, which support the unrestricted proliferation and metastatic progression of cancer cells, is widely accepted to be an emerging hallmark of cancer [7][8][9]. FA are fundamental cellular components and energy sources and catabolized mainly by fatty acid oxidation (FAO, also called βoxidation), which produces more than three-fold as much ATP per mole as oxidation of carbohydrates [10][11][12]. Recent evidence underscores the idea that abnormal lipid metabolism mediated by FAO plays an significant part in tumor progression [13,14].
Nieman et al. discovered that FAO-derived energy can guide ovarian cancer cells to metastasize to the omentum [15]. Wang et al. reported that CPT1A-induced FAO activation increased metastatic ability [16]. These findings reveals that inhibiting FAO might be the potential of strategy in cancer therapy.
Carnitine palmitoyl transferase 1 (CPT1) limits the rate-limiting step of FAO, which localizes to the outer mitochondrial membrane, and is responsible for FA transport into mitochondria for further oxidation by changing acyl-CoAs into acylcarnitines [17,18]. Recent studies showed that over-expression CPT1A (as a subtype of CPT1) accelerates tumor development in colorectal cancer, hepatocellular carcinoma and prostatic cancer, and is closely bound up with a poor overall survival (OS) of ovarian cancer patients [19][20][21][22]. However, the mechanistic detail of CPT1Amediated FAO activation on GC cell progression, remain obscure.
In the present study, we evaluated the CPT1A expression and attempted clarify its prognostic value and molecular mechanism in GC. Effects of CPT1A over-expression was considered to facilitate GC cells proliferation and invasion via FAO activation and conducive to an adverse outcome of GC patients. Our study supplies special insights into the potential molecular mechanisms of CPT1A involvement in FAO, and supports new tactics of molecule-targeting treatment to improve on the curative effects in GC.

Cell cultures and treatments
The gastric mucosal epithelial cell line GES-1 and human GC cell lines, were acquired from the American Type Culture Collection (ATCC, Manassas, VA, USA).
According to their guidelines, cells were cultured in a humid atmosphere with 5% CO2 at 37 °C. The cell lines were tested for authenticity in 2015 using short tandem repeat (STR) genotyping and screened for mycoplasma contamination.

Lentivirus transfection
To knockdown/ overexpress CPT1A levels, short hairpin RNA (shRNA)/fulllength targeted to human CPT1A or a non-target oligonucleotide was built into the LV-

Colony formation assay
GC cells (1×10 3 ) of each group were reseeded in 6-well plates and incubated (37 °C, 5% CO2) for 2 weeks. Then rinsed with cold PBS, colonies were fixed in methanol and stained with Giemsa. Light microscope was used to image and count.

Ethynyldeoxyuridine analysis
To detect the cell proliferation, Cell-Light™ EdU Apollo®488 (RiboBio, China) In Vitro Imaging Kit (RiboBio) was utilized according to the manufacturer's protocol after seeding cervical cancer cells in 96-well plate for 24h. The result was photographed using a microscope (Olympus, Tokyo, Japan).

Cell migration and invasion assay
Transwell chamber assay was performed to detect migratory and invasive abilities of GC cells. GC cells of each group were seeded on the upper chambers with 8-μmpore size polycarbonate membrane coated with 60 μL Matrigel, and serum-free DMEM media was added to it. Then 20% fetal bovine serum/DMEM media was put into bottom chamber serving as attractant. After 24h of incubation, cells were fixed with 4% formaldehyde and stained with 1% crystal violet to count the invading cells in 10 random fields by the microscope. Cell migration was investigated by transwell assay either, the only difference was no need for Matrigel. This experiment was performed 3 times for accuracy.

Wound-healing assay
Briefly, GC cells of each group were reseeded, and a linear wound was generated within the confluent monolayers by scraping the cells with sterile 10 μL pipette tips.
Using an inverted microscope imaged the progression of migration at 0h, 24 h.

Western blot
WB analysis was cultured as previously described [23]. Immunoreactive bands were visualized using ECL Western blot kit (Amersham Biosciences, Buckinghamshire, UK).

Statistical procedures
Statistical analysis was performed using SPSS 25 Pearson's correlation were also performed. Using the Kaplan-Meier method calculated the survival curves, and the differences were assessed by a Log-rank test. One and two asterisks indicated P<0.05, and P<0.01, respectively.

CPT1A is highly expressed in GC cell lines and tumor tissues
To investigate the role of CPT1A in the development of GC, we discovered that CPT1A was markedly increased in GC tissues from the Cancer Genome Atlas (TCGA) database and GEPIA datasets ( Fig. 1a and 1b). Similarly, the protein expression of CPT1A level in GC cells was remarkably higher than the expression of GES-1 (Fig.   1c). Moreover, we detected CPT1A protein expression in 45 adjacent non-tumor tissues and 129 GC tissues by IHC (Fig. 1d) 2e). These results supported that CPT1A may be served as a potential genetic marker for human GC.

CPT1A enhances the proliferation of GC cells in vitro
To elucidate the biological functions of CPT1A in GC, we overexpressed CPT1A in SGC7901/BGC803 cells, which exhibited relatively low endogenous CPT1A levels.
In addition, we knocked down CPT1A in AGS/BGC823 cells, which exhibited relatively high endogenous CPT1A levels (Fig. 3a). Subsequently, GC cells proliferation were evaluated. MTT assays shown that CPT1A overexpression or silencing markedly enhanced or inhibited cell growth in corresponding GC cells, respectively (Fig. 3b). EdU incorporation assays indicated that the percentage of EdUpositive cells were risen in AGS/BGC823 cells with CPT1A overexpression and were decreased in SGC7901/BGC803 cells with CPT1A knockdown (Fig. 3c). In addition, two-dimensional colony formation assays found that CPT1A overexpression or silencing markedly enhanced or impaired colony formation ability in corresponding GC cells, respectively (Fig. 3d). Therefore, these in vitro findings indicated that CPT1A enhances the proliferation of GC cells.

CPT1A inhibits the migration and invasion of GC through EMT
Based on our clinical pathological analyses supported that a positive association between CPT1A levels and LN metastatisis (Tables 2), we started to examine whether CPT1A affects GC migration and invasion. We found that CPT1A overexpression promoted the migration and invasion capability of these two cells in vitro (Fig. 4a, 4b); in contrast, knockdown of CPT1A expression levels resulted in cell wound healing inhibition (Fig. 4c), simultaneously a significant reduction of the migration and invasion capability.
Increasing evidence has demonstrated that epithelial mesenchymal transition (EMT) impels the metastatic potentials of tumor cells [24]. So, we detected the potential roles of CPT1A on EMT related marker molecules. As shown in Fig. 4d, E-cadherin expression was decreased and of Vimentin, and Snail were increased in GC cells, which downregulated CPT1A expression. In contrast, overexpression of CPT1A had reverse functions on EMT related proteins. These results indicated that CPT1A enhances the migration and invasion capability of GC through activating EMT.

CPT1A activates FAO in GC cells
Accumulation of lipid is necessary for tumor cells to satisfy the high-energy demand of their rapid proliferation and metastasis. We next appraised alterations in lipid accumulation using BODIPY 493/503 staining, and indicated that CPT1A silencing markedly promoted lipid droplet accumulation in SGC7901/BGC803 cells, whereas CPT1A overexpression suppressed it in AGS/BGC823 (Fig. 5A). Evaluation of changes in FAO level and NADP + /NADPH ratio also revealed, CPT1A silencing significantly suppressed FAO level and NADP + /NADPH ratio in SGC7901/BGC803 cells. Conversely, the results were opposite (Fig. 5b and 5c). These data indicated that CPT1A has involved in lipid accumulation of GC cells. Furthermore, we focus on the regulatory effects of CPT1A on key enzymes (ACSL1, ACSL5, and ACC1) involved in FAO. Western blot showed that CPT1A silencing could significantly reduced the expression levels of key enzymes, whereas CPT1A overexpression enhanced the expression levels (Fig. 5d). These results indicated that CPT1A induces FAO in GC cells.

CPT1A promotes the tumor progression of gastric cancer cells by regulating FAO
To further explore the function of CPT1A in GC cells, we treated with the FAO inhibitors-Etomoxir (ETX) in CPT1A-overexpressed GC cells, and measured cell proliferation and migration. We observed that ETX significantly decreased cell proliferative capacity in CPT1A-overexpressed GC cells in comparison with control cells (Fig. 6a). Colony formation assays showed a reduction in both the size and number of colonies in ETX-treated cells (Fig. 6b). Similarly, inhibition of FAO markedly inhibited cell migration capacity as determined by Transwell (Fig. 6c). In addition, ETX significantly reduced FAO and NADP + /NADPH ratio ( Fig. 6d and 6e). Western blot shown that ETX reversed the CPT1A-induced upregulation of Vimentin, ACSL1, and ACC1 and downregulation of Ecadherin (Fig. 6f). These results expounded that CPT1A regulates cell growth, migration, and the EMT process in a manner dependent on the FAO pathway.

Discussion
Previous studies reported that CPT1A serves a key role in the development of numerous cancers, and contributes to tumor growth and metastasis [25][26][27]. Xiong et al. found that CPT1A knockdown attenuates invasion and lymphangiogenesis through the regulation of VEGF signaling in breast cancer cell [28]. A recent study demonstrated that CPT1A essential for HCC cell survival, and Diet-induced hepatic steatosis activates Ras to promote hepatocarcinogenesis via CPT1A [29]. Furthermore, the PGC1α/CEBPB/CPT1A signaling axis advances the resistance to radiation in NPC via activating FAO [30]. Despite these accumulating evidence on the role of CPT1A in cancer progression, whether CPT1A could mediate GC growth and invasion, and by which underlying mechanisms, remain unclear. The present study aimed therefore to clarify the role of CPT1A in GC growth and invasion.
Shi et al. expounded that elevated CPT1A expression is closely bound up with the shorter survival of AML [31]. In the present study, CPT1A was found to be frequently upregulated in GC using bioinformatics and experimental analysis (Fig. 1, 2). IHC shown that the positive stain intensity of CPT1A was mostly expressed in GC tissues, and weak staining was expressed in adjacent non-tumor tissue (Fig. 1d, e). More interestingly, the up-regulation level of CPT1A was closely related to some clinical pathological factors (stage, grade, and LN metastasis) (P<0.01) (Fig. 2d). However, no differences were found with age, sex, or tumor size (Table 2). In particular, we found that GC patients exhibiting high CPT1A expression had poorerm OS, PPS and FP than patients with low expression by Kaplan-Meier Plotter datasets (Fig. 2e). Then, we performed survival analyses to evaluate that CPT1A is a prognostic marker for clinical outcomes. As a result, CPT1A may function as an important indicator during the germination and progression of GC.
Previous studies reported that CPT1A is abundant in advanced prostate cancer (PCa), and CPT1A knockdown decreased clonogenic growth and invasion in PCa cell [32]. Wang et al. demonstrated that the molecular functions of CPT1A as a succinyl transferase by targeting S100A10 and activating its succinylation, which promotes cancer invasion and metastasis [33]. Consistent with these findings, our reasearch have confirmed an important mechanism of CPT1A in GC proliferation. Then, we appraised the variations of cell migration and invasion in GC cells transfect with or without CPT1A, and found that silencing of CPT1A results in cell migration and invasion arrested, simultaneously overexpression of CPT1A exposed the contrary results.
Besides, we assessed the variations of EMT-related marker proteins, which were in charge of the transition of the cell from the epithelium-like shape to the mesenchymal phenotype and their up/down-regulation would lead to cell EMT process [34,35]. As expounded by our data, silencing of CPT1A lead to down-regulate expression of the Vimentin, Snail and increased E-cadherin. Conversely, the results were opposite, in accordance with previous fndings [28]. These results supported that CPT1A can influence the cancer cell progression by regulating the expression of EMT marker proteins, thus affected GC cell process.
Increasing evidence has placed that a novel emphasis on FAO as an important metabolic crux in cancer, which is the main approach for the degradation of FA and enhances the production of NADPH and ATP [36]. For instance, advanced FAO supplies glioblastoma cells metabolic malleability to accommodate to its rich nutrient microenvironment [37]. And Wang et al. indicated a distinctive mechanism by which IL-17A mediates FAO stimulates tumor angiogenesis of endothelial cells by enhancing endothelial mitochondrial respiration [38]. In the present study, we observed that of CPT1A knockdown increased lipid droplet accumulation, decreases FAO and NADP + /NADPH ratio in GC cells, and vice versa (Fig. 5a-c). Consistently, we found that the expression ratios of FAO-related proteins were decreased/increased following cell treatment with CPT1A down-regulate/up-regulate (Fig. 5d). Unsurprisingly, we confirmed that CPT1A could increase FAO level in GC cell by NADPH homeostasis.
A previous reasearch identified that targeting with metabolic reprogramming of cancer cells might be a potential therapeutic strategy and CPT1A has been indicated as a target of FA metabolism [39]. Wang et al. CPT1A-mediated FAO accelerates colorectal cancer cell metastasis by avoiding anoikis [16]. Hence, we hypothesized that CPT1A could be involved in GC proliferation and invasion through the FAO. To do so, the effects of Etomoxir (an inhibitor of FAO) on AGS/BGC823 cells with CPT1A overexpression was assessed. A few studies have reported that Etomoxir had an influence on CPT1A expression which could irreversibly inhibit the function of CPT1 [40]. Although recently some studies identified an off-target effect of etomoxir [41,42], still we could see the FAO inhibition of Etomoxir in research. Consistently, the FAO inhibition of etomoxir could be validated by decreased FAO and NADPH NADP + /NADPH ratio in the present study (Fig. 6d-e). In addition, our results showed that FAO inhibition by Etomoxir could suppress growth, migration and EMT process both in CPT1A overexpression GC cells (Fig. 6a-c, 6f). Taken together, these findings indicated that CPT1A stimulates GC growth and migration by activating FAO.

Conclusions
Through this study, our data clearly identified that FAO confers tumor progression via a CPT1A-mediated mechanism in GC (Fig. 7). The results confirmed that CPT1A expression was signifcantly elevated in GC tissues and cells, and that high expression

Acknowledgement
We wish to acknowledge that the project was supported by ChinaJapan Union Hospital of Jilin University. Disclosure statement: All the authors declare that they have no conflicts of interest.

Funding
Not applicable.

Availability of data and materials
Datasets used and/or analyzed data are available from the corresponding author upon reasonable request.

Consent for publication
Not applicable.    Western blot analysis of EMT marker (E-cadherin, Vimentin and Snail) in indicated cells with overexpressed or silenced of CPT1A. β-Actin was used as a loading control.

Figure Legend
Data are represented as mean ± standard error of mean of at least three independent experiments, *P<0.05; **P<0.01.