RAD001 targeted HUVECs reverses 12‐lipoxygenase‐induced angiogenesis in oesophageal squamous cell carcinoma

Abstract 12‐LOX plays an important role in the progression of various malignancies. However, the underlying mechanisms of the action of 12‐LOX and tumour treatment strategies remain not fully defined. In this study, we investigated the possible roles of 12‐LOX in ESCC and explored the new therapeutic target. Approximately 73% of ESCC tissues showed marked up‐regulation of 12‐LOX, which was associated with poor prognosis. 12‐LOX overexpression was positively correlated with the malignant progression of ESCC as demonstrated both in vitro and in vivo. Up‐regulation of 12‐LOX significantly increased the proliferation of ESCC cells and the xenograft volume. Moreover, 12‐LOX up‐regulation promoted tube formation of HUVECs and tumour angiogenesis in xenografts. Mechanism investigation indicated that 12‐LOX overexpression led to activation of the PI3K/AKT/mTOR pathway and the up‐regulation of VEGF in ESCC cells. Subsequent analysis indicated that the RAD001 could reverse the 12‐LOX‐induced promoting effect on ESCC. Specifically, the application of RAD001 inhibited the proliferation of ESCC cells and the tube‐forming ability of HUVECs. In the drug group, the xenografts exhibited significant volume reduction and angiogenesis inhibition. We demonstrated that RAD001 could inhibit HUVEC migration. These findings presented the evidence that RAD001 had distinct roles on HUVECs and could exert anti‐tumour effects by targeting not only the PI3K/AKT/mTOR pathway but the angiogenesis in ESCC.

billion, and there were 375 000 deaths every year. 4 Due to the atypical early symptoms of EC and the lack of specific tumour markers for early diagnosis, the majority of the patients are diagnosed with an advanced stage and cannot be actively treated, resulting in poor prognosis. The 5-year survival rate is only 15% to 25%. 5,6 Therefore, there is an urgent need to explore biomarkers for early diagnosis of ESCC and the treatment strategies to improve patient prognosis.
Arachidonic acid (AA) is a long-chain polyunsaturated fatty acid (PUFA) widely distributed in mammalian cell membranes. The lipoxygenase (LOX) pathway is the main pathway of AA metabolism, as well as the cyclooxygenase (COX) and the cytochrome P450 (CYP450) pathways. The metabolites of AA obtained through these pathways are collectively called eicosanoids, which are essential components of the cell membrane, participating in cell membrane function and cell signalling transduction. The dysregulation of bioactive lipids, especially LOX and its metabolites, is implicated in a variety of pathophysiological processes and diseases, including inflammation and cancers. It is known that LOX and HETE can promote the progression of various human malignant tumours, and the mechanisms include the inhibition of apoptosis, 7 induction of cell proliferation [7][8][9] and metastasis of cancer [10][11][12] cells and activation of tumour angiogenesis. 13,14 Tumorigenesis is often driven by maladjustment of cellular information processing and signal transduction. As normal cells evolve progressively to a neoplastic state, they acquire a succession of hallmark capabilities that contribute to tumorigenesis.
Inducing angiogenesis and maintaining proliferative signals are important signs of cancer. 15 Tumour microvascular structure has long been viewed as a core direction of tumour research. Tumour microvasculature is a crucial compartment in TME, which plays a significant role in tumour energy and oxygen supply and tumour metabolism. During tumour development, persistent proangiogenic signals lead to the continued growth of new blood vessels in the quiescent vascular system, which in turn helps maintain tumour growth. However, tumour neovascularization caused by chronic stimulation and unbalanced pro-angiogenic signals is usually abnormal. Compared with normal blood vessels, tumour blood vessels show tortuous and disordered structures, uneven thickness, arteriovenous and vascular malformation and incomplete tube wall formation. This aggravates the hypoxic environment of the tumour, making tumour cells less responsive to chemoradiotherapy and eventually leading to chemoradiotherapy resistance. 16 In addition, abnormal blood vessels may promote appropriate conditions for tumour metastasis. Therefore, the investigation of tumour microvasculature is of great significance for understanding the process of tumour development, metastasis and drug resistance, as well as the prevention, treatment and prognosis of cancer-related diseases.
The phosphatidylinositol 3-kinase (PI3K)/AKT/mTOR pathway (mammalian target of rapamycin) is activated in most human cancers, [17][18][19] which plays an important role in numerous cellular functions including metabolism, proliferation, adhesion, migration, invasion, survival and angiogenesis. 20,21 Therefore, the PI3K/AKT/ mTOR pathway is an important target for cancer treatment, and several inhibitors of this pathway have been developed. 22,23 The mTOR pathway plays an important role in regulating cell metabolism, proliferation and survival in response to environmental signals. [24][25][26] Accumulated clinical evidence has shown that the mTOR inhibitors exhibit effective suppression effect against malignant tumours. 27 Everolimus (RAD001) is the most commonly used mTOR inhibitor, which directly inhibits mTORC1, but has limited or no effect on mTORC2 activity. 28 Currently, everolimus is the only mTOR inhibitor approved by the Food and Drug Administration (FDA) for the treatment of certain types of breast cancer, pancreatic neuroendocrine tumours and papillary renal carcinoma. RAD001 has been shown to exhibit anti-tumour effects in several cancers by inhibiting tumour growth and reducing tumour vascularization. [29][30][31] 2 | MATERIAL S AND ME THODS

| Lentiviral vector infection
The control and 12-LOX overexpression lentiviral vectors were constructed by GeneChem (Shanghai, China). Kyse150 cells were seeded in 6-well plates with a density of 5 × 10 4 cells/well. 12-LOX overexpression vectors or control vectors were infected into Kyse150 cells according to the relevant manufacturer's instructions when the confluency reached 30%-40%, and the medium was replaced at 24 hours after infection. The infected cells were then selected with 3 µg/ml of puromycin to establish cells stably expressing 12-LOX at 72 hours. After 2 passages, the cells were used for subsequent experiments and divided into two groups as follows: cells infected with control vector (LV-Ctrl group) and 12-LOX overexpression vector (LV-12-LOX group).

| Colony formation assay
Cells were seeded in 6-well plates (800 cells/well). The adherent 12-LOX-Kyse150 cells were treated with gradient concentration inhibitors (LY294002 and RAD001, Selleckchem; Houston, TX, USA) or control DMSO for two weeks prior to staining. The colonies were fixed in methanol and stained with crystal violet, and then, the number of colonies was measured by counting more than 50 cells (>50 cells was defined as a colony) (ImageJ 1.47v software; NIH; National Institutes of Health, Bethesda, MD, USA).   HUVECs were seeded on coated plates (4 × 10 4 cells/well) and cultured with serum-free medium at 37℃ for 4 hours. Tube formation was observed and photographed using an Olympus DP71 microscope. Five random fields were selected in each well, and the total tube length was quantified using the NIH ImageJ 1.47v software.

| Tube formation assay
Each group was tested in triplicate.

| Transwell assay
HUVEC cells were seeded into 6-well plates. After RAD001 stimulation for 24 hours, cells were trypsinized, counted and plated into a transwell chamber (pore size: 8 μm, 24-well) with serum-free medium (5 × 10 4 cells/well). 20% FBS-containing medium was used to induce cell migration. The migrating cells were fixed after 12 hours, stained with crystal violet, photographed and counted.

| Western blot analysis
Cells were seeded into 6-well plates. Following overnight incubation, the cells were treated with LY294002, RAD001 or DMSO and harvested at 48 hours. Western blot analysis was performed as previously described. 32 The membranes were exposed with an enhanced chemiluminescence reaction (ECL) kit, and the grey levels of the protein bands were analysed using the ImageJ 1.47v software. The sources of the primary antibodies applied were as follows: p-S6 S235/236 (#4858) and S6 (#2217) were purchased from Cell Signaling Technologies; polyclonal rabbit anti-12-LOX was from Novus Biologicals (NBP2-29941; Novus Biologicals; Centennial, CO, USA); VEGF antibody was obtained from Santa Cruz (sc-7269; Santa Cruz; Dallas, TX, USA).

| Immunohistochemical/ Immunofluorescence analysis
Immunohistochemical staining and qualitative scoring were performed as previously described. 32 All antibodies applied were validated by immunohistochemistry (IHC) and immunofluorescence (IF) in paraffin-embedded tissues as determined by the manufacturer.

| In vivo experiments
All animal experiments were approved by the Ethics Committee of Qilu Hospital of Shandong University and were carried out in accordance with the national regulations for animal research in China.
Female BALB/c nude mice aged 4 weeks were randomly divided into 2 groups and maintained under standard conditions according to institutional guidelines for animal care. LV-Ctrl-Kyse150 or LV-12-LOX-Kyse150 cells (1 × 10 6 ) were suspended in PBS and injected subcutaneously into the flank of the mice. RAD001 intervention (diluted with NS, intraperitoneal injection, 3mg/kg) was performed once the tumours had reached a volume of 100 mm 3 , and the compound was administered daily for 7 days. Tumour size was measured every five days using callipers and the tumour volumes were calculated using the following equation: V = (length × width 2 )/2. The mice were killed at the end of the study, and the tumours were dissected four weeks after appearance and stored at −80℃ for subsequent analysis. The flow chart showed the entire process of the in vivo experiment ( Figure 6A).

| Statistical analysis
Unpaired t test was conducted using the SPSS software 24.0 (IBM Corp., Armonk, NY, USA). Kaplan-Meier method and log-rank test were used for survival analysis. The results were presented as mean ±standard error (SE). P value <0.05 was considered a significant difference.

| 12-LOX overexpression was associated with poor prognosis of ESCC
IHC detection was conducted in all 153 ESCC cases (see Table S1 for the basic characteristics) to assess the expression levels of 12-LOX.
The results indicated that 112 cases were included in the 12-LOX high-expression group (score >8) and 41 cases in the 12-LOX lowexpression group (score ≤8). IHC staining results of 12-LOX was shown in Figure 1A. TCGA database mining indicated that the expres- Data are presented as the mean ±SEM. *P < 0.05; **P < 0.01; ***P < 0.001 of 12-LOX in squamous cell carcinoma was significantly higher than that in normal oesophageal and adenocarcinoma tissues ( Figure 1B).
The survival outcome of ESCC patients was consistent with TCGA data ( Figure 1C; P = 0.015). The OS and PFS of the patients in the 12-LOX high-expression group were significantly lower than those in the low-expression group ( Figure 1D, E, P < 0.001, respectively).

| 12-LOX promoted proliferation of ESCC cells
The basal expression of 12-LOX in Eca109 and TE-1 cells was higher than that in Kyse150 and Eca9706 cells (Figure 2A To explore the role of 12-LOX in ESCC cell proliferation, EdU, colony formation and the CCK-8 assays were conducted. The EdU assay

| 12-LOX promoted migration of HUVECs and tube formation
Previous studies have shown that lipoxygenase is an important factor required for VEGF-mediated angiogenesis. 33,34 Therefore, the expression of VEGF in ESCC cells was assessed. Western blot analysis ( Figure 3A, B) and ELISA ( Figure 3C) were used to detect the content of VEGF. The results indicated that VEGF level in the 12-LOX overexpression group was significantly higher than that in the control group.

Migration of endothelial cells is another key step in angiogene-
sis, which allows cells to expand from existing vessels. Subsequently, it was demonstrated that 12-LOX could promote cell migration and tube formation of HUVECs, which could be used to establish a model to assess endothelial function and angiogenesis. As expected, conditioned medium used in LV-12-LOX group significantly promoted HUVECs migration and tube formation compared with the control group ( Figure 3D-G). As shown in Figure 3D, E, conditioned medium significantly increased the number of migrating HUVEC cells in LV-12-LOX group compared with the control cells. Moreover, following stimulation under the conditioned medium, tube formation of LV-12-LOX group was highly increased compared with that of the control group ( Figure 3F). The conditioned medium led to a significant advantage of mesh, master segment and branch in tubes ( Figure 3G).
Specifically, the number and length of mesh, master segment and branch in the 12-LOX overexpression group was higher than those in the control group (P < 0.001, respectively). Overall, these results indicated that 12-LOX may promote angiogenesis in vitro by accelerating endothelial cell migration and tubular structure formation.

| Overexpression of 12-LOX activated the PI3K-AKT-mTOR pathway
In order to explore the intrinsic biological function of 12-LOX in ESCC, we further examined the PI3K-AKT-mTOR pathway. The results indicated that the phosphorylation levels of AKT and mTOR and of the downstream substrate proteins of the mTOR signalling pathway (P70S6K/S6/4EBP1) were specific activated and increased significantly in 12-LOX up-regulated cell lines. And the activation of the pathway was significantly inhibited with the application of Baicalein ( Figure 3H). The conclusion was replicated in patients' tissues, and IHC staining showed that patients with high expression of 12-LOX also had higher mTOR expression ( Figure 3I).

| 12-LOX exerted a tumour-promoting effect in vivo
To further verify the pro-tumour effect of 12-LOX in vivo, a xenograft

| mTOR inhibitor RAD001 could reverse the pro-tumour effects of 12-LOX
A number of inhibitors of the PI3K/AKT/mTOR pathway have been discovered, and these agents have been shown to reduce VEGF secretion and restrain angiogenesis. 35 To confirm the interaction between the tumour-promoting effect of 12-LOX in the development of cancer phenotype and the activation of the PI3K/AKT/ mTOR pathway, the latter pathway was suppressed with inhibitors and the cancer-associated phenotype was further examined. The biologically active doses of the inhibitors in Kyse150 cells were initially established. LY294002 at 20 μM completely abrogated p-AKT and p-mTOR activity. RAD001 at 1 μM could inhibit the PI3K/AKT/mTOR pathway and at 10 μM completely abrogated the expression levels of p-mTOR, p-P70S6K and p-S6. Our functional tests were carried out under these concentration stimuli. In addition, the inhibitory effect of LY294002 and RAD001 on VEGF was dose-dependent ( Figure 5A, B).
Subsequently, the proliferation of ESCC cells and the induction of angiogenesis following the use of LY294002 and RAD001 were measured. The proliferation of ESCC cells following treatment with the inhibitors was examined using EdU ( Figure 5C, D), CCK-8 ( Figure   S3A) and colony formation assays ( Figure S3B Figure   S4A, B), the crawling ability of HUVECs stimulated by RAD001 was significantly lower than that of the controls and the creeping distance was shorter at 12, 24 and 48 hours. Consistently, the conditioned medium containing inhibitors reversed the catalytic effect of 12-LOX on tube formation, which was not noted in the absence of the inhibitors (Figure 5H, I). The conditioned medium containing inhibitors led to a significant suppression of mesh, master segment and branch in HUVECs, which was reflected in the number and length of mesh, master segment and branch. These results indicated that 12-LOX promoted angiogenesis and cell proliferation in ESCC through the PI3K/AKT/mTOR pathway and that RAD001 might be a potentially powerful cancer-preventing agent in ESCC.

| RAD001 reversed the pro-tumour effect of 12-LOX in vivo
Previous in vivo studies demonstrated that RAD001 exerted a significant anti-tumour effect. 29,36,37 To further verify the anti-tumour effect of RAD001 in ESCC, an additional xenograft model with LV-12-LOX-Kyse150 cells was established. As expected, nude mice treated with normal saline (NS) developed large neoplasms reaching an average size of 786mm 3 at the end of the study period, while mice treated with RAD001 showed a smaller neoplasm size ( Figure 6B-D), confirming that RAD001 significantly inhibited tumour growth.
Subsequently, the inhibitory effects of RAD001 were examined on mTOR signalling by staining tumour sections for p-mTOR expression ( Figure 6E). Following successful inhibition of the pathway, blood vessel formation in tumour tissues was assessed. The results of IF co-staining with 12-LOX and CD31 ( Figure 6F-G) indicated that the expression level of 12-LOX in both groups was high, while the expression level of CD31 in the RAD001-treated group was lower than that in the saline group. These findings revealed that RAD001 patients with high expression of 12-LOX was generally lower than patients who had relatively low expression of 12-LOX, which was also consistent with the statistical data in the TCGA database. Taken together, the results highlighted that 12-LOX could effectively promote ESCC, and 12-LOX might be a potential biological marker for prognostic assessment and treatment of ESCC patients.
The vasculature system in TME is maintained by persistent induction of angiogenic signals 15 and is characterized by structural abnormalities. Although the abundant vascular system is the key to energy supply and growth of tumour cells, the abnormal vascular structure may cause poor blood flow and elevated blood pressure, leading to hypoxic TME and production of HIF-1α. Hypoxia will increase VEGF-A production and the formation of abnormal vessels. The hypoxic regions within the TME may affect the efficacy of radiotherapy or chemotherapy. 35,41 Furthermore, HIF-1α is more stable under hypoxic conditions, leading to X-ray resistance and up-regulation of genes involved in cell survival. 41 Meanwhile, emerging studies have reported that hypoxic TME is related to tumour immune tolerance.
Hypoxia can stimulate the release of damage-associated pattern molecules that trigger the immune system's rejection of the tumour.
However, the counter-activation of the tolerance mechanism at tumour hypoxia sites can lead to the tumour's immune escape. It has been found that tumour hypoxia can promote recruitment of regulatory T(Treg) cells by inducing the expression of the chemokine CC-chemokine ligand 28 (CCL28), which in turn facilitates immune tolerance and angiogenesis. 42 Hence, certain therapeutic strategies aim to reduce the complexity of tortuous vessels and correct hypoxic TME, which are essential steps in the treatment of malignant tumours.

F I G U R E 6 RAD001 reverses the pro-tumour effect of 12-LOX in vivo. A, Diagram of intervention in vivo experiments. B, Representative
images of subcutaneous tumours with Kyse150 cells in the two different treatment groups indicated. C, Tumour growth curves in nude mice from the two groups. D, Tumour weight of the two groups. E, IHC analyses of p-mTOR expression were performed using xenografts sections. Scale bar = 50 µm. F, Representative images of IF performed on xenografts treated as indicated with 12-LOX (green) and CD31 (red) antibodies. Nucleus was labelled with DAPI (blue), and images were merged. Scale bar = 50 µm. G, The expression levels of 12-LOX and CD31 in the two groups indicated. 12-LOX, lipoxygenase; IHC, immunohistochemical; IF, immunofluorescence. Data are presented as the mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 The main purpose of using anti-angiogenic drugs in cancer treatment is to destroy tumour angiogenesis. Elevated VEGF-A levels will increase vascular permeability, leading to vascular leakage, poor blood flow and increased interstitial pressure, thereby exacerbating hypoxia in TME. 35 Therefore, the use of certain drugs that interfere with the expression or action of VEGF may paradoxically lead to vascular normalization and hypoxic improvement in specific tumour types. VEGFR2 inhibitors can cause the temporary formation of a more regular vascular system. 43 However, there is a bottleneck in the clinical application of this strategy, given that the induction of vascular normalization is not permanent and that higher exposure to antiangiogenic drugs may increase hypoxia and potentially neutralize the beneficial effects caused by these agents. 44 Previous studies have explored the possibility of inhibiting carcinogenic signals in cancer cells as another strategy to regulate tumour blood vessels. 44 of key tumour-associated proteins, such as cyclin D1, VEGF and matrix metalloprotease 9. 47 As an inhibitor of the mTOR pathway, RAD001 can inhibit the phosphorylation of 4EBP1, thus preventing the synthesis of tumour-related proteins. Tang et al have shown that integrinβ4 is a key partner of 12-LOX and the disruption of integrin β4-12-LOX interaction can reduce the pro-inflammatory and carcinogenic activity of 12-LOX. 48 Based on this evidence, we speculated that RAD001 might affect the action of integrin β4 or interfere with integrin β4-12-LOX. Therefore, the specific mechanism of RAD001, by which reverses the cancer-promoting role of 12-LOX, needs to be further verified.
The present study demonstrated for the first time the role of 12-LOX in promoting cell proliferation and angiogenesis in ESCC, which provides important information of the role of eicosanoids in malignancy. Concomitantly, the present study demonstrated for the first time that RAD001, an inhibitor of mTOR pathway, played a significant anti-cancer role in ESCC by targeting the vascular system, which provides a possible target and drug choice for the treatment of this cancer type. However, our analysis did not explore in detail the specific mechanism of the action of RAD001 in ESCC and did not detect the normalization of the vascular architecture, but only detect VEGF expression and tube formation, which was the limitations of the current study.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
This study was approved by the Ethics Committee of Qilu Hospital of Shandong University. The above participants agree with the research results described in the article and agree to be responsible for all aspects of the work.

ACK N OWLED G EM ENTS
None.

CO N FLI C T S O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available in the supplementary material of this article. The data set supporting the results of this article is included within the article. The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request.