Qianfeng Chen, Yuxia Zhong, Bohan Li, Yucong Feng, Yuandie Zhang, Tao Wei, Margaret Zaitoun, Shuang Rong, Hua Wan, Qing Feng. Acrolein-triggered atherosclerosis via AMPK/SIRT1-CLOCK/BMAL1 pathway and a protection from intermittent fasting[J]. The Journal of Biomedical Research. DOI: 10.7555/JBR.38.20240025
Citation:
Qianfeng Chen, Yuxia Zhong, Bohan Li, Yucong Feng, Yuandie Zhang, Tao Wei, Margaret Zaitoun, Shuang Rong, Hua Wan, Qing Feng. Acrolein-triggered atherosclerosis via AMPK/SIRT1-CLOCK/BMAL1 pathway and a protection from intermittent fasting[J]. The Journal of Biomedical Research. DOI: 10.7555/JBR.38.20240025
Qianfeng Chen, Yuxia Zhong, Bohan Li, Yucong Feng, Yuandie Zhang, Tao Wei, Margaret Zaitoun, Shuang Rong, Hua Wan, Qing Feng. Acrolein-triggered atherosclerosis via AMPK/SIRT1-CLOCK/BMAL1 pathway and a protection from intermittent fasting[J]. The Journal of Biomedical Research. DOI: 10.7555/JBR.38.20240025
Citation:
Qianfeng Chen, Yuxia Zhong, Bohan Li, Yucong Feng, Yuandie Zhang, Tao Wei, Margaret Zaitoun, Shuang Rong, Hua Wan, Qing Feng. Acrolein-triggered atherosclerosis via AMPK/SIRT1-CLOCK/BMAL1 pathway and a protection from intermittent fasting[J]. The Journal of Biomedical Research. DOI: 10.7555/JBR.38.20240025
Unproofed Manuscript: The manuscript has been professionally copyedited and typeset to confirm the JBR’s formatting, but still needs proofreading by the corresponding author to ensure accuracy and correct any potential errors introduced during the editing process. It will be replaced by the online publication version.
Acrolein-triggered atherosclerosis via AMPK/SIRT1-CLOCK/BMAL1 pathway and a protection from intermittent fasting
Department of Nutrition and Food Hygiene, Key Laboratory of Toxicology, School of Public Health, Nanjing Medical University, Nanjing, Jiangsu 211166, China
2.
Department of nosocomial infection management, Suzhou Wujiang District Children's Hospital, Suzhou, Jiangsu 215299, China
3.
Department of Preventive Medicine, Department of Nutritional Health and Toxicology, School of Public Health, Wuhan University of Science and Technology, Wuhan, Hubei 430065, China
4.
Healthcare center, Sir Run Run Hospital, Nanjing Medical University, Nanjing, Jiangsu 211112, China
Hua Wan, Healthcare center, Sir Run Run Hospital, Nanjing Medical University, 109 Longmian Avenue, Nanjing, Jiangsu 211112, China. E-mail: wanhua2006@njmu.edu.cn
Qing Feng, Department of Nutrition and Food Hygiene, Key Laboratory of Toxicology, School of Public Health, Nanjing Medical University, 101 Longmian Avenue, Nanjing, Jiangsu 211166, China. E-mail: qingfeng@njmu.edu.cn
Circadian clock plays a vital role in the pathological progression of cardiovascular disease (CVD). Our previous studies showed that acrolein, an environmental pollutant, promoted atherosclerosis by reducing CLOCK/BMAL1 and disturbing circadian rhythm. Whereas, intermittent fasting (IF), a diet pattern, was able to ameliorate acrolein-induced atherosclerosis. In vivo, mice were fed acrolein 3 mg/kg/day via drinking water and IF for 18h (0:00-18:00). We observed that IF decreased acrolein-accelerated the formation of aortic lesion in ApoE−/− mice. Up-regulation of NF-κB, IL-1β and TNF-α levels were found in liver and heart tissue upon acrolein exposure, while was down-regulated by IF. Interestingly, IF treatment exhibited higher AMPK, p-AMPK and SIRT1and lower MAPK expression which was caused by acrolein. Besides, circadian genes Clock/ Bmal1 expression were suppressed and disturbed treated with acrolein, while were reversed by IF. Furthermore, consistent with that in vivo, short-term starvation as a fasting cell model in vitro could improve the disorders of CLOCK/BMAL1 and raised SIRT1 via regulating AMPK, as well as ROS-MAPK induced by acrolein. In conclusion, we demonstrated that IF repressed ROS-MAPK while activated AMPK to elevate the expression of circadian clock genes to ameliorate acrolein-induced atherogenesis, which shed a novel light to prevent cardiovascular diseases.
Colorectal cancer (CRC) is featured by extremely high morbidity and mortality rates. According to Global Cancer Statistics 2022, CRC accounted for approximately 10.0% of all cancer cases and accounted for 9.4% of all cancer-related deaths worldwide[1]. The metastatic rate of early-stage CRC patients ranges between approximately 10% and 20%, but it may be as high as approximately 50% for the advanced-stage CRC patients[2]. The survival rates for localized versus distant CRC differ significantly. In the United States, patients with localized CRC have a five-year survival rate of approximately 91%, while those with distant CRC have a five-year survival rate of only approximately 15%[3]. Therefore, the pathogenesis and molecular mechanisms of CRC have become the focus of CRC research to develop effective therapeutic agents with minimal side effects.
Neovascularization is an important channel for cancer cell metastasis, enabling cancer cells to reach distal metastatic sites via the blood vessels[4]. Angiogenesis is vital to sustaining tumor growth and facilitating metastasis, which is characterized by the proliferation and migration of vascular endothelial cells[5]. Thus, targeting the proliferation and migration process of vascular endothelial cells is a potential strategy for the CRC therapy[6]. Because the vascular endothelial growth factor (VEGF) is a key mediator of vascular endothelial cell proliferation and migration[7], anti-VEGF agents have been employed to treat metastatic CRC[8]. However, their use is constrained by a high risk of fatal bleeding complications, including severe side effects such as intestinal perforation and arterial embolism, limiting their widespread application in the CRC treatment[9, 10]. Therefore, identifying molecules that may be specifically targeted to inhibit tumor angiogenesis may offer significant clinical benefits to CRC patients.
Long non-coding RNAs (lncRNAs) are highly abundant RNA transcripts with limited or no protein-coding capacity[11]. They play crucial roles in immune responses, inflammation, and cancer progression through various pathways, such as serving as miRNA precursors and participating in DNA methylation, histone modification, and chromatin remodeling[12, 13]. Moreover, the high expression of lncRNAs in tumors or blood serves as a diagnostic biomarkers in a variety of cancers. Several lines of evidence have demonstrated that lncRNAs regulate angiogenesis, thereby influencing tumor initiation, development, and metastasis[14-16]. Specifically, lncRNAs such as FLANC and PVT1 have been identified as key players in cancer progression by regulating VEGF expression in metastatic colorectal cancer (mCRC)[17, 18].
LncRNA LINC01503 was reported to be aberrantly expressed in tumors and promoted tumor progress[19]. However, whether LINC01503 has an angiogenesis effect remains unexplored. In the current study, we aimed to investigate the role and underlying mechanisms by which LINC01503 may promote angiogenesis in CRC from human samples, as well as at both cellular and animal levels.
Materials and methods
Patients and clinical samples
In the current study, we collected 76 pairs of paraffin-embedded human CRC tissue samples and their corresponding adjacent normal mucosal samples at the First Affiliated Hospital of Nanjing Medical University. Patients were selected based on specific inclusion and exclusion criteria to ensure a well-defined and representative patient population. The inclusion criteria included patients aged 45–80 years with histologically confirmed colorectal cancer who had not received chemotherapy or radiotherapy before surgical treatment and had sufficient tumor tissue samples available for molecular analysis. Patients with a history of other malignancies, significant comorbid conditions that may influence study outcomes, or those who had undergone any surgical intervention or treatment that may have altered normal tissue structure before sample collection were excluded. Clinicopathological features of patients with CRC are summarized in Table 1. These 76 pairs of CRC samples were immediately snap-frozen in liquid nitrogen after collection and stored at −80 ℃ until RNA extraction. This project was approved by the Institutional Review Board of Nanjing Medical University (Approval No. [2016] 640).
Table
1.
Association between LINC01503 expression and the clinicopathological characters of colorectal cancer patients
Characters
Cases (n)
LINC01503 expression
P-value
Low (n)
High (n)
Sex
0.245
Male
42
23
19
Female
34
15
19
Age (years)
0.08
≥60
39
16
23
＜60
37
22
15
Tumor diameter (cm)
0.169
≥5
27
11
16
＜5
49
27
22
Tumor location
0.5
Colon
41
21
20
Rectum
35
17
18
Clinical stage
0.04*
Ⅰ/Ⅱ
38
21
17
Ⅲ/Ⅳ
38
14
24
T stage
0.031*
Ⅰ
2
0
2
Ⅱ
19
14
5
Ⅲ
41
20
21
Ⅳ
14
4
10
N stage
0.323
Absent
37
20
17
Present
39
18
21
M stage
0.307
M0
72
37
35
M1
4
1
3
Vascular invasion
0.041*
Absent
61
34
27
Present
15
4
11
Perineural invasion
0.5
Absent
67
34
33
Present
9
4
5
Analyzed with either the Chi-square test. Abbreviations: T, tumor; N, node; M, metastasis.
Immunohistochemical staining was performed on paraffin-embedded tissue sections using anti-CD31 and vascular endothelial growth factor A (VEGFA) antibodies, as previously described[20]. Briefly, tissue specimens were paraffin-embedded and sliced into sections that were dewaxed with xylene and dehydrated with gradient alcohol. The slices were heated in citrate buffer, followed by blocking with 3% H2O2, and sealed with goat serum. The sections were incubated with primary antibody overnight at 4 ℃, including anti-CD31 antibody (1∶200 dilution Cat. #ab38624, Abcam, Cambridge, UK), anti-VEGFA antibody (1∶200 dilution; Cat. No. Ab171828, Abcam), and then incubated with HRP-labeled secondary antibody (50 μL; Cat. #AFIHC0003, AiFang biological, Hunan, China ). The slices were treated with 3,39-diaminobenzidine tetrahydrochloride reagent to develop color, and re-stained with hematoxylin. Each section was photographed under the microscope (Zeiss, Melville, NY, USA) at 200× magnification.
The VEGFA scores were analyzed using a modified H-score. Briefly, five fields at 400 × magnification were randomly selected and the staining intensity was rated as 0, 1, 2, or 3, corresponding to the presence of negative, low, moderate, and strong staining, respectively. The semi-quantitative assessment was performed by two independent investigators who were blinded to clinical parameters of the patients examined. The H-score was calculated by the formula: ([% of weak staining]× 1) + ([% of moderate staining] × 2) + ([% of strong staining] × 3)[21]. The expression levels by the median H-score for each patient was also categorized as low or high[22].
After Immunohistochemical staining with CD31, microvessels were quantified according to the method described by Weidner et al[23]. Single endothelial or clustered endothelial cells, with or without a lumen, were recognized to be individual vessels. Image J software (National Institutes of Health, Bethesda, MD, USA) was used to quantify the microvessel density (MVD) based on the CD31 staining[23].
Cells and cell culture
The HCT116, RKO, Lovo, and Caco-2 CRC cell lines were obtained from the Shanghai Institute of Cell Biology of the Chinese Academy of Sciences. The CX-1 cell line was purchased from Fu Hen Cell Center (Shanghai, China). The cell lines were authenticated by short tandem repeat (STR) profiling and confirmed to be free of mycoplasma. The cells were cultured in either RPMI 1640 or DMEM medium supplemented with 10% FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin (all reagents from Gibco, Grand Island, NY, USA), and grown in a humidified incubator at 37 ℃ with 5% CO2.
Human umbilical vein endothelial cells (HUVECs) were isolated from the umbilical cords of healthy donors at Nanjing Maternal and Child Health Hospital (Approval No. PJ2021-039-001) by incubating the veins with 0.1% collagenase D (Roche, Mannheim, Germany ) for 35 minutes. HUVECs were grown in an endothelial cell growth medium (Cat. #1001-b, ScienCell, Carlsbad, CA, USA) that was changed twice a week.
Cell Counting Kit-8 (CCK8) assay
The CCK8 kit (Medchem Express, Monmouth Junction, NJ, USA) was utilized to evaluate cell proliferation. CRC cells were seeded in 96-cell plates with a density of 1 × 103 cells per well followed by transfection with either small interfering RNA (siRNA) or plasmid. Absorbance at the wavelength of 450 nm was measured at different time points using a microplate reading device (Bio-Tek, Winooski, VT, USA). Each experiment was performed in triplicate and repeated at least three times.
Colony formation assay
Cells transfected with siRNA or plasmids were cultured in six-well plates at a density of 500 cells per well for 10 to 15 days with the medium changed every three days. The colonies were stained with a 0.1% solution of crystalline violet (Keygene, Nanjing, China), and then counted to determine the degree of colony formation. Each experiment was repeated at least three times.
Transwell assays
We performed the Transwell assays for the cell migration and invasion using 24-well transwell chambers with an 8-μm pore size polycarbonate membrane (Corning, NY, USA). A total of 3× 104 (CX-1 cells), 6 × 104 (RKO/HCT116 cells), and 1.6 × 104 HUVECs in 200 μL of serum-free medium were seeded into the upper chamber, which was either coated with Matrigel (Corning, Corning, NY, USA)(for invasion assays) or left uncoated (for migration assays). To the lower chamber, 600 μL of medium containing 10% FBS was added. For the assays involving HUVECs, the ratio of the conditional medium to the basal medium was 2∶1. After an incubation period of 10 to 48 h, the non-migration or non-invaded cells on the upper surface were removed. Migrating and invading cells on the bottom were fixed with 95% ethanol, stained with 0.1% crystal violet in methanol/phosphate-buffered saline (PBS), and imaged using a Zeiss microscope (Melville, NY, USA). In each group, five fields per filter were selected at random, and stained cells were counted to assess the degree of cell migration and invasion. Each experiment was repeated at least three times.
Collection of conditional medium
CRC cells were transfected with either siRNA or plasmids, and the culture supernatant was removed 48 h later. The cells were then washed three times with PBS, and the medium in the wells was replaced with serum-free basal medium. After 48h, the conditional medium was then collected and filtered through the filter (0.45-μm Merck Millipore, Darmstadt, Germany), and stored at −80 ℃ for further use.
HUVECs tube formation assay
Pre-cooled matrix gel (Corning) was added to the wells of a 96-well plate and polymerized for 30 minutes at 37 ℃. HUVECs (8 × 103 cells) were suspended in 100 μl medium (conditional medium to basal medium ratio = 2∶1), and incubated at 37 ℃, 5% CO2 for one to six hours. Quantitative analysis of the total length of the tubes in each well was performed using Image J software. Each experiment was repeated at least three times.
Tumor formation and metastasis assays in nude model mice
BALB/c nude male mice (4–5 weeks) (GemPharmatech, Nanjing, China) were divided into two groups (n = 5 for each group). Mice were maintained under pathogen-free conditions on a 12-h light/12-h dark cycle. LINC01503-knockdown HCT116 cells or control cells (1 × 107) were subcutaneously injected into the dorsal flank of mice. Tumor growth was evaluated every two days for four weeks. The tumor volume was measured using the following formula: volume = length × width2 × 0.5. After four weeks, all mice were euthanized, and the xenografts were extracted for further analysis.
For the in vivo lung metastasis assay, HCT116 cells with LINC01503 stable knockdown and control cells (4 × 106) were administered via tail vein injection into each nude mouse (n = 5 per group). After eight weeks, the mice were euthanized, and their lungs were examined for metastases. The study protocols were approved by the Animal Care and Use Committee of Nanjing Medical University (Approval No.1601080). All procedures adhered strictly to the Guide for the Care and Use of Laboratory Animals.
Chicken chorioallantoic membrane (CAM) assay
Fertilized White Leghorn chicken eggs were incubated at 37 ℃ with constant humidity. On the 8th day of incubation, a small window was carefully created in the shell above the air sac, and then sealed with paraffin. 1×107 in 50 μL CRC cells were mixed with 50 μL medium containing 50% High Concentration Matrigel (BD Biosciences, San Jose, CA, USA). A 0.1 mL aliquot of this cell suspension was applied to the CAM of 8-day-old embryos. Four to five days after implantation, photographs of the Matrigel implants were taken, and blood vessel counts were conducted by two independent observers who were blinded to the treatment groups.
siRNA, plasmid, and miRNA mimics/inhibitor transfection
Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) was employed to transfect cells with short interfering RNAs (siRNAs), plasmids, microRNA (miRNA) mimics, or miRNA inhibitors, following the manufacturer's instructions. Briefly, when 1 × 105 cells were seeded in a 6-well plate and incubated for 24 h, the cell density reached approximately 40%, optimal for transfecting 4 μL of siRNA or miRNA mimics and inhibitors. In contrast, seeding 3 × 105 cells in the same conditions resulted in a density of about 80%, suitable for transfecting 4 μg of plasmids. Six hours post-transfection, the medium was replaced with complete medium containing 10% FBS, and cells were cultured for an additional 24 to 48 h before further studies.
The siRNAs targeting LINC01503 (si-LINC01503-1, si-LINC01503-2, si-LINC01503-3), as well as those targeting CBP (si-CBP) and P300 (si-P300), along with a negative control siRNA (si-NC), were purchased from RiboBio Co., Ltd. (Guangzhou, Guangdong, China). For overexpression experiments, the cDNAs of LINC01503 and HSP60 were cloned into the pcDNA3.1 expression vector, generating the pcDNA-LINC01503 and pcDNA-HSP60 constructs, respectively, by GeneChem (Shanghai, China). Additionally, miRNA mimics and inhibitors specific to miR-342-3p were also obtained from RiboBio.
The sequences for the siRNAs targeting LINC01503, CBP, and P300 are shown in Supplementary Table 1 (available online).
Analysis of The Cancer Genome Atlas (TCGA)-CRC data
The transcriptome expression profiles of CRC tumor tissues and paired normal tissues and corresponding clinical data were downloaded from the TCGA data portal (https://cancergenome.nih.gov). Specifically, we utilized data from the TCGA Colon Adenocarcinoma Collection (TCGA-COAD)[24] to analyze differences in LINC01503 expression between CRC and normal tissues. The data were accessed in July 2020. Furthermore, to determine the relationship between LINC01503 expression levels and OS of the CRC patients, Kaplan–Meier analysis was performed. Corresponding analysis of LINC01503 with VEGFA and HSP60 was also performed with GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA).
RNA extraction and real-time reverse transcription-PCR (qRT-PCR)
Total RNA (1 μg) was reverse transcribed into cDNA using the PrimeScript™ 1st Strand cDNA Synthesis Kit (Takara Bio, Shiga, Japan). For miRNA, the TaqMan® MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) and specific primers were used for reverse transcription. The expression levels of the respective RNAs were quantified using SYBR Green Master Mix (Takara Bio) on an ABI 7300 sequence detector (Applied Biosystems, Foster City, CA, USA). Relative RNA expression fold changes were calculated using the ΔΔCt method, with GAPDH or U6 serving as normalization controls. The PCR-used primers are shown in Supplementary Table 2 (available online). Each experiment was repeated at least three times.
Enzyme-linked immunosorbent assay (ELISA)
ELISA was performed using a Kit following the manufacturer's protocol. Cells were transfected according to experimental groups for 48 h and replaced with serum-free basic medium, then cultured for 24 h. The supernatant was collected and filtered through a 0.45μm filter. Use Human VEGFA ELISA kit (Cat. #EHC108.96, NeoBioscience, Shenzhen, China) for detection, add supernatant for incubation and washing; add biotinylated antibody for incubation and washing; add enzyme conjugate for incubation and washing; add chromogenic substrate to incubate and finally add stop solution. Absorbance was measured at 450 nm using a microplate reader (Bio-Tek).
Generation of stable LINC01503 knockdown cells
Lentiviruses carrying sh-LINC01503 or scrambled shRNA were constructed by Genechem (Shanghai, China). HCT116 cells were infected with these lentiviruses at a multiplicity of infection (MOI) of 20. Post-infection, stable transductants were selected using 4 μg/mL puromycin (Sigma, St. Louis, MO, USA). The sequence for the shRNA targeting LINC01503 was 5′-CCACCTTTCTGGTAATGCA-3′.
Western blotting (WB) analysis
Lysates from the cells or tumor tissue were subjected to separation by SDS-PAGE and then transferred onto polyvinylidene fluoride membranes. The membranes were subsequently probed with primary antibodies, followed by incubation with appropriate secondary antibodies. Protein detection was carried out using chemiluminescent methods. Antibodies against the following antigens were used in the current study: VEGFA (1∶1000 dilution; Cat. #Ab71828, Abcam, Cambridge, UK), HSP60 (1∶1000 dilution; Cat. #SC-271215, Santa Cruz, CA, USA ), HSP60 (RIP:5 μg, Cat. #Ab19082, Abcam ), AKT (1∶1000 dilution , Cat. #4691, Cell Signaling Technology, Danvers, MA, USA), p-AKT(Thr308) (1∶1000 dilution; Cat. #13038, Cell Signaling Technology), PCNA(1∶1000 dilution; Cat. #SC-56, Santa Cruz ), Histone H3 (1∶1000 dilution; Cat. #D1H2, Cell Signaling Technology), histone H3 lysine 27 acetylation (1∶1000 dilution; Cat. #Ab4729, Abcam), CBP (1∶1000 dilution; Cat. #Ab253202 Abcam, ), p300 (1∶1000 dilution; Cat. #Ab275378,Abcam)and GAPDH (1∶2000 dilution; Cat. #SC-47724 Santa Cruz). Each experiment was repeated at least three times.
Duel-luciferase reporter assay
Sequences of LINC01503 with predicted miR-342-3p binding sites, along with their respective mutated sequences, were amplified by PCR and cloned into the pmirGLO vector (Genechem) to create LINC-WT (5′-GTGCAGGGATTACAGGTGTGAGC-3′) and LINC-MUT (5′-CCAGAGACCCTGAGCCCAGAGAA-3′) plasmids, respectively. The same procedure was used to construct the VEGFA plasmids. These plasmids were co-transfected into CRC cells along with either a miR-342-3p mimic or a miR-342-3p mimic negative control (mimic-NC). Dual-luciferase assays (Promega, Madison, WI, USA) were also performed following the manufacturer's protocol. Each experiment was repeated at least three times.
RNA immunoprecipitation (RIP)
RNA immunoprecipitation (RIP) was performed using the EZ-Magna RIP Kit (Millipore, MA, USA) according to the manufacturer's instructions. In brief, HCT116 cells were transfected with either a miR-342-3p mimic or a miR-342-3p mimic-NC for 24 h and 1 × 107 cells were pelleted and resuspended in RIP lysis buffer. The lysates were incubated with magnetic beads conjugated to anti-Ago2 (Abcam) or anti-IgG antibodies and rotated at 4 ℃ overnight. Then, cell lysates were incubated with 5μg of IgG (Millipore) and AGO2 antibody (Cat. #ab186733, Abcam) antibody-magnetic coated beads and rotated at 4 ℃ overnight.
The 1 × 107 untreated HCT116 and CX-1 cells were pelleted and resuspended in RIP lysis buffer and incubated with 5 μg of IgG (Millipore) and HSP60 antibody (Abcam) antibody-magnetic coated beads and rotated at 4 ℃ overnight.The immunoprecipitated proteins were digested with protein K. The purified RNA was then subjected to qRT-PCR to measure the levels of LINC01503 or VEGFA. Each experiment was repeated at least three times.
Chromatin immunoprecipitation (ChIP)
The ChIP assay was performed using a ChIP kit (Abcam) according to the manufacturer's instructions. Briefly, A total of 3 × 106 cells were fixed with formaldehyde, lysed, and sonicated to shear the DNA into fragments between 200 and 1000 bps. The DNA fragments were immunoprecipitated using 5 μg anti-H3K27Ac antibody (Abcam) or negative-control IgG. The precipitated DNA was purified and then analyzed by qRT-PCR. The primers used in the ChIP assay are shown in Supplementary Table 3 (available online).
Chromatin isolation by RNA purification (CHIRP) assay and mass spectrometry
The CHIRP assay was performed according to the previously described protocol[25]. Briefly, HCT116 cells were cross-linked with 1% formaldehyde for 10 min, equilibrated with glycine buffer for 5 min, washed with cold PBS three times, and then scraped with 1 mL lysis buffer. Cell lysates were then sonicated to generate RNA fragments of 200–1000 bp and then centrifuged at 10000g for 10 min. Next, 50 μL of the supernatant was transferred as the input, and the remaining supernatant lysate was incubated with 3 μL probes (100 μmol/L) at 37 ℃ with shaking. Samples were mixed well with 100 μL prepared beads and incubated with rotation. The bead-sample mixture was washed twice with wash buffer and then resuspended in 1 mL wash buffer. Next, 100 μL was set aside for RNA isolation with TRIzol (Invitrogen). The remaining 900 μL was centrifuged at 12 000 g for 10 min and the pellet was resuspended in 50 μl protein buffer and boiled for mass spectrometry detection (Shanghai Applied Protein Tech, Shanghai, China). The probes and primers used are shown in Supplementary Table 4 (available online)
Statistical analysis
Each experiment was repeated at least three times, and their means were used for the analysis. The statistical analysis was performed by SPSS20. 0 (IBM Corp. Armonk, USA). Data were presented as mean ± standard deviation. To compare the statistical difference between the two groups, the Student's t-test was used. To compare the statistical difference among the multiple groups, one-way ANOVA, followed by Dunnett's tests for multiple comparisons was employed. The relationship between the expression of LINC01503 and clinical pathological characteristics in patients with CRC was analyzed with either the Chi-square test or the Fisher test. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs) for univariable and multivariate analyses. The survival curves were plotted using the Kaplan-Meier method and compared by the log-rank test. P < 0.05 was considered a measure of statistical significance.
Results
LINC01503 was highly expressed in CRC and correlated with an advanced stage, vascular invasion, and poor survival
To identify the lncRNAs potentially involved in CRC progression, we collected primary tumors and corresponding adjacent normal mucosal tissues from three patients with stage Ⅰ/Ⅱ and three patients with stage Ⅲ/Ⅳ CRC and performed an lncRNA microarray analysis[25]. The results showed that LINC01503 was not only more highly expressed in tumors, compared with normal tissues, but also significantly upregulated in stage Ⅲ/Ⅳ CRCs compared with stage Ⅰ/Ⅱ CRCs (Supplementary Fig. 1A, available online). This finding indicates that LINC01503 may be involved in the progression of CRC.
To validate the microarray results, LINC01503 levels were measured in 76 matched CRC tumors and normal colorectal tissues by qRT-PCR. The results showed that LINC01503 was significantly elevated in CRC tissues, compared with normal colorectal tissues (P < 0.01; Fig. 1A). Further Kaplan-Meier survival analysis showed that CRC patients with high LINC01503 expression level had a shorter overall survival (OS) than those with low LINC01503 expression level (Fig. 1B). These results aligned with the analysis outcomes from The Cancer Genome Atlas (TCGA) database (Supplementary Fig. 1B–1D, available online).
Figure
1.LINC01503 was specifically overexpressed in colorectal cancer (CRC) tissues and was associated with a poor prognosis.
A: The expression of LINC01503 in clinical CRC samples was evaluated by qRT-PCR analysis (n = 76). ACTB was used as an internal reference gene. P < 0.01 by paired Student's t-test. B: Kaplan-Meier survival curve of overall survival (OS) in CRC patients with low LINC01503 expression versus those with high LINC01503 expression. P < 0.01 were determined by log-rank test. HR indicates hazard ratio; 95% CI indicates 95% confidence interval. C: Representative images of CD31 immunohistochemistry in clinical CRC samples with low or high LINC01503 expression. CD31 protein expression was assessed in the form of microvessel density in tumor tissues of the two groups by immunohistochemistry assay (n = 38). Original magnification, 100×. D: The correlation between LINC01503 expression and MVD-CD31+. Abbreviations: CRC, colorectal cancer; qRT-PCR, real-time reverse transcription-PCR; MVD, microvessel density; HR, hazard ratio; CI, confidence interval.
To investigate clinicopathological significance of LINC01503 in CRC, we analyzed the correlation between LINC01503 expression levels and clinicopathologic characteristics of our CRC cohort. The results in Table 1 showed that high LINC01503 expression levels were significantly associated with clinical stage (P = 0.04), T stage (P = 0.031), and vascular invasion (P = 0.041). We further analyzed the correlation between LINC01503 expression levels and angiogenesis in CRC tissues, and found that high LINC01503 expression levels were significantly associated with high MVD (Fig. 1C and 1D). Consistent with this finding, a positive correlation was found between the expression levels of LINC01503 and CD31 based on the TCGA data (Supplementary Fig. 1E, available online). Collectively, these findings indicate that the high expression of LINC01503 in CRC tissues may promote disease progression and is associated with tumor angiogenesis.
LINC01503 promoted tumorigenesis in CRC
To investigate biological function of LINC01503 in CRC, we transfected CRC cell lines of HCT116 and CX-1 with siRNAs targeting LINC01503 or of RKO and CX-1 with pcDNA-LINC01503 plasmids. The siRNA transfection effectively knocked down LINC01503 expression, while the pcDNA-LINC01503 transfection led to increased LINC01503 levels (Supplementary Fig. 2A, available online). The CCK-8 and colony formation assays demonstrated that silencing LINC01503 significantly decreased the proliferative capacity of CRC cells, while overexpressing LINC01503 significantly enhanced CRC cell viability (Fig. 2A and 2B). The transwell assays further revealed that LINC01503 knockdown significantly impaired CRC cell migration and invasion, compared with control cells, whereas LINC01503 overexpression promoted these processes (Fig. 2C and 2D).
Figure
2.LINC01503 promoted cell proliferation and invasion in vitro.
A: HCT116 and CX-1 cell viability after transfection of LINC01503 siRNAs according to CCK-8 assay. RKO and CX-1 cell viability after transfection of LINC01503 overexpressing plasmid according to CCK-8 assay. B: HCT116, CX-1 and RKO cell proliferation after transfection of LINC01503 siRNAs or overexpressing plasmid evaluated by colony formation. C and D: Transwell assay were performed to detected the cell migration (C) and invasion (D) of HCT116, CX-1, and RKO cell after the transfection of LINC01503 siRNAs or overexpressing plasmid. Two-tailed Student's t-test was performed to examine the significance except that statistical analysis for A , which was performed by two-way ANOVA followed by Dunnett's tests. Abbreviations: CRC, colorectal cancer; siRNA, short interfering RNA; CCK-8, cell counting kit-8; ANOVA, analysis of variance.
To examine the effect of LINC01503 on tumorigenesis in vivo, we constructed the stable LINC01503-knockdown HCT116 cells (sh-LINC01503) and control HCT116 cells (sh-NC). In the xenograft experiments, the nude mice inoculated with sh-LINC01503 cells had lower tumor weights and volumes than the control group (Fig. 3A–3D). Notably, silencing LINC01503 reduced PCNA expression and CD31+ MVD in tumor tissues (Fig. 3E). We next examined the effects of LINC01503 on CRC metastasis by injecting cancer cells into mice via tail veins. Silencing LINC01503 markedly suppressed the formation of pulmonary metastatic nodules in the mice (Fig. 3F and 3G). Collectively, these results indicate that the LINC01503 may promote CRC tumorigenesis.
Figure
3.LINC01503 promoted cell proliferation and invasion in vivo.
A: Photographs of xenograft tumors from groups of BALB/c-nude mice 28 days after injection with stable LINC01503-konnockdown (sh-LINC01503) or control (sh-NC) HCT116 cells (n = 5). B: The subcutaneous tumor weight of nude mice was compared between the two groups (n = 5). C: The tumor growth curve of nude mice was based on tumor volume in the two group (n = 5). D: The expression of LINC01503 in xenograft tumor tissues was evaluated by qRT-PCR analysis (n = 5). E: Representative images of PCNA and CD31 immunohistochemistry in xenograft tumor tissues. PCNA and CD31 protein expression were assessed by immunohistochemistry assay (n=5). Original magnification, 400× for PCNA and 100× for CD31. F: Representative images of tumor metastasis in lung tissues (left) and the corresponding quantification of the number of metastatic foci in lung tissues (right) after tail vein injection with stable LINC01503-konnockdown (sh-LINC01503) or control (sh-NC) HCT116 cells. The white arrows show the metastases. G: Hematoxylin and eosin staining of lung tissues, original magnification, 100×. Data were presented as mean ± standard deviation. *P < 0.05 and **P < 0.01 by unpaired Student's t-test. Abbreviations: CRC, colorectal cancer; qRT-PCR, real-time reverse transcription-PCR.
LINC01503 promoted angiogenesis by increasing VEGFA expression
To investigate the function of LINC01503 in angiogenesis and tube formation, we performed a transwell assay with HUVECs. We found that the conditioned medium from LINC01503-knockdown CRC cells markedly reduced the number of branch points formed by HUVECS and their migration ability; meanwhile, the conditioned medium from LINC01503-overexpressing CRC cells yielded the opposite results (Fig. 4A and 4B). The pro-angiogenic effect of LINC01503 was next verified in the chick embryo CAM assay. The results showed that silencing LINC01503 shortened the length of the second and third vessels of CRC cells cultured on the CAM (Fig. 4C). These results indicate that LINC01503 may have a pro-angiogenic function.
Figure
4.LINC01503 promoted the proliferation and migration of vascular endothelial cells.
A: After LINC01503 knockdown or overexpression for 48 h in HCT116, CX-1, and RKO cells, the medium was replaced with serum-free medium. Conditional media were collected after an additional incubation of 24 h. The effects of LINC01503 in CRC cell-conditioned medium on HUVECs was evaluated using a tube formation assay. B: A Transwell assay was performed to assess the migration ability of HUVECs influenced by LINC01503 expression in CRC cells. C: The effects of LINC01503 on the length of secondary and tertiary vessels were analyzed using the chick embryo chorioallantoic membrane assay. Data are presented as mean ± standard deviation from three independent experiments. Statistical significance was determined by one-way ANOVA, followed by Dunnett's test for multiple comparisons and unpaired Student's t-test . *P < 0.05, **P < 0.01. Abbreviations: CRC, colorectal cancer; HUVECs, human umbilical vein endothelial cells; CAM, chick embryo chorioallantoic membrane.
The qRT-PCR analysis results showed that, among the cell proliferation and migration-related genes expressed in vascular endothelial cells, the VEGFA expression level was the most significantly correlated with the variation in LINC01503 levels (Fig. 5A). We also found that high LINC01503 expression in CRC tissues was significantly associated with high VEGFA protein levels (Fig. 5B), which was consistent with the results of TCGA data analysis showing that LINC01503 expression was correlated with VEGFA mRNA levels (Supplementary Fig. 2B, available online). The WB and qRT-PCR results confirmed that LINC01503 overexpression promoted the protein and mRNA expression of VEGFA, while LINC01503 knockdown had the opposite effect (Fig. 5C and 5D). The ELISA results also showed that the VEGFA protein level in the supernatant of CRC cells was significantly affected by changes in LINC01503 expression (Fig. 5E).
Figure
5.LINC01503 induced vascular endothelial growth factor A (VEGFA) expression and secretion.
A: HCT116 and CX-1 cells were infected with LINC01503 knockdown lentiviruses, while RKO and CX-1 cells were transfected with LINC01503 overexpression plasmids. qRT-PCR analysis was conducted to assess the expression of multiple angiogenesis-associated genes, with pseudo-color scale values log2 transformed. Transcript levels were normalized to ACTB expression. B: Representative images of VEGFA immunohistochemistry in clinical colorectal cancer (CRC) samples (left) are shown, with original magnification at 100×. The correlation between LINC01503 and VEGFA is presented on the right. C: Western blotting analysis demonstrated the effect of LINC01503 on VEGFA protein levels. D: qRT-PCR analysis assessed the impact of LINC01503 expression on VEGFA mRNA levels. E: The effect of LINC01503 on VEGFA secretion from CRC cells was analyzed using enzyme-linked immunosorbent assay (ELISA). F: Following 48 h of LINC01503 knockdown or overexpression in HCT116, CX-1, and RKO cells, the medium was replaced with serum-free medium. Conditional media were collected after an additional 24 h. The effect of supernatant from CRC cells with high or low LINC01503 expression on the Akt signaling pathway in HUVECs was evaluated by Western blotting. Data are presented as mean ± standard deviation from three independent experiments. Statistical significance was determined by one-way ANOVA, followed by Dunnett's test for multiple comparisons and unpaired Student's t-test. *P < 0.05 and **P < 0.01. Abbreviations:HUVECs, human umbilical vein endothelial cells; qRT-PCR, quantitative real-time PCR.
VEGFA-AKT is a classic pathway, which promotes endothelial cell proliferation and new blood vessels formation. Hence, we stimulated HUVECs with culture supernatant from CRC cells with low or high LINC01503 expression levels. The results showed that the supernatant from CRC cells with high LINC01503 expression activated the AKT signaling pathway in HUVECs, while that from CRC cells with low LINC01503 expression inhibited the AKT signaling pathway in these cells (Fig. 5F).
Taken together, these findings demonstrate that LINC01503 may increase the capacity of tube formation and migration of vascular endothelial cells by promoting the expression and secretion of VEGFA.
LINC01503 promoted VEGFA expression by binding to miR-342-3p
We next investigated how LINC01503 increased VEGFA expression. LncRNAs may promote the expression of related mRNAs by binding to miRNAs. We therefore used bioinformatics to predict miRNAs that bound to LINC01503 using miRDB (http://mirdb.org/) and Starbase (https://starbase.sysu.edu.cn/); miRNAs with binding scores ≥ 60 were selected. We found that miR-342-3p was the only miRNA that overlapped between the two databases (Fig. 6A). Analysis of miRanda algorithms revealed that both LINC01503 and VEGFA contained miR-342-3p-binding sequences (Fig. 6B). We then speculated that LINC01503 might promote VEGFA expression by binding to miR-342-3p. The following RIP assays demonstrated a direct interaction between miR-342-3p and the LINC01503 transcript (Fig. 6C). Furthermore, a luciferase reporter assay showed that upregulating miR-342-3p inhibited the luciferase reporter activity of LINC01503, but not that of mutant LINC01503 (Fig. 6D). Similar results were observed for VEGFA (Fig. 6E). These suggested that LINC01503 bound to miR-342-3p, and that miR-342-3p bound to VEGFA mRNA at the same site. Significantly, a miR-342-3p mimic abolished the tube formation and migration abilities of HUVECs induced by LINC01503 overexpression in CRC cells. Conversely, a miR-342-3p inhibitor reversed the inhibitory effects of LINC01503 knockdown in CRC cells on the tube formation and migration capacities of HUVECs in CRC cells (Supplementary Fig. 3A–3D, available online). Moreover, the miR-342-3p mimic abrogated the increased expression and secretion of VEGFA induced by LINC01503 overexpression in CRC cells, while the miR-342-3p inhibitor reversed the decreased VEGFA expression and secretion induced by LINC01503 knockdown in CRC cells (Fig. 6F–6I ).
Figure
6.LINC01503 promoted VEGFA expression by binding to miR-342-3p.
A: Predicted miRNAs that bind to LINC01503 were identified using the miRDB and Starbase databases. B: Sequences of miR-342-3p predicted to target both LINC01503 and VEGFA are shown. C: RNA immunoprecipitation (RIP) was performed in HCT116 cells transfected with the miR-342-3p mimic or mimic-NC, using anti-IgG or anti-Ago2 antibodies; the level of LINC01503 was quantified by qRT-PCR. D: HCT116 and RKO cells were co-transfected with the miR-342-3p mimic or mimic-NC and a reporter plasmid containing either the wild-type (LINC-WT) or mutant LINC01503 binding sequence. Luciferase reporter assays measured the activity. E: HCT116 and RKO cells were similarly co-transfected with the miR-342-3p mimic or mimic-NC and a luciferase reporter plasmid containing either the wild-type (VEGFA-WT) or mutant VEGFA 3′ UTR (VEGFA-MUT). Relative luciferase activity assays assessed the activity. F and G: Western blotting was conducted to measure VEGFA levels after co-transfection of the miR-342-3p inhibitor with si-LINC01503 in HCT116 and CX-1 cells (F), or the miR-342-3p mimic with the LINC01503 overexpression plasmid in RKO and CX-1 cells (G). H and I: ELISA was used to quantify secreted VEGFA levels following co-transfection of the miR-342-3p inhibitor with si-LINC01503 in HCT116 and CX-1 cells (H), or the miR-342-3p mimic with the LINC01503 overexpression plasmid in RKO and CX-1 cells (I). Data were presented as mean ± standard deviation from three independent experiments. *P < 0.05 and **P < 0.01 by one-way analyses of variance, followed by Dunnett's tests for multiple comparisons and unpaired Student's t-test. Abbreviations: CRC, colorectal cancer; qRT-PCR, quantitative real-time PCR.
These results indicate that LINC01503 promotes VEGFA expression and secretion by binding to miR-342-3p, thereby increasing the tube formation and migration ability of HUVECs.
LINC01503 enhanced the protein stability of VEGFA by binding to HSP60
We next showed that treating LINC01503-knockdown CRC cells with cycloheximide (CHX) significantly increased VEGFA protein degradation (Fig. 7A). Several studies have shown that lncRNAs may affect protein stability by binding to proteins[26], we therefore hypothesized that LINC01503 regulated the protein stability of VEGFA by binding to specific proteins.
Figure
7.LINC01503 enhanced the protein stability of VEGFA by binding to HSP60.
A: Stable sh-LINC01503 and negative control (sh-NC) HCT116 cells were treated with 100 μg/mL cycloheximide (CHX) and harvested at specified time points. VEGFA protein levels were analyzed by Western blotting, quantified by densitometry, and plotted against time to assess VEGFA stability. B: The RNA immunoprecipitation (RIP) assay was performed using anti-HSP60 antibody in HCT116 cells, with IgG as a negative control. Enrichment of LINC01503 was quantified by qRT-PCR. C: Western blotting was conducted to measure HSP60 levels in LINC01503 knockdown HCT116 and CX-1 cells, as well as in LINC01503-overexpressing RKO and CX-1 cells. D: HCT116 cells transfected with LINC01503-overexpression plasmid or pcDNA-3.1 were treated with 100 μg/mL CHX for indicated time points; HSP60 protein levels were detected by Western blotting and plotted against time to evaluate HSP60 stability. E: Stable sh-LINC01503 and sh-NC HCT116 cells were treated with 5 μmol/L proteasome inhibitor (MG132) for 12 h, and HSP60 protein levels were measured by Western blotting. F and G: Immunoprecipitation was performed using anti-HSP60 (F) and anti-Flag (G) antibodies to investigate the interaction between HSP60 and VEGFA in HCT116 cells. H: Western blotting was used to quantify VEGFA levels in HSP60-overexpressing RKO and CX-1 cells. I: HSP60 overexpression plasmid (HSP60) or control plasmid (pcDNA3.1) was transfected into LINC01503-knockdown stable HCT116 cells, followed by treatment with 100 μg/mL CHX at various time points after 48 h of transfection; VEGFA protein levels were detected by WB and plotted against time to assess stability. J and K: CX-1 and RKO cells were co-transfected with si-LINC01503 and HSP60 overexpression plasmid, with VEGFA levels in cell lysates evaluated by Western blotting (J) and secreted VEGFA levels assessed by ELISA (K). Data are presented as mean ± standard deviation from three independent experiments. Statistical significance was evaluated using one-way ANOVA, followed by Dunnett's test for multiple comparisons and unpaired Student's t-test .*P < 0.05 and **P < 0.01. Abbreviations: CRC, colorectal cancer; qRT-PCR, quantitative real-time PCR.
Therefore, we performed the CHIRP assays using biotin-labeled probes specific for LINC01503 (Supplementary Fig. 4A, available online), before subjecting these samples to mass spectrometry analysis. While LINC01503 did not directly bind to VEGFA, we showed that it interacted with the HSP60 chaperone; we next selected this protein for further study (Supplementary Fig. 4B, available online). The RIP assays demonstrated that LINC01503 might directly bind to HSP60 (Fig. 7B). Moreover, LINC01503 knockdown decreased HSP60 expression, while LINC01503 overexpression increased HSP60 levels (Fig. 7C). However, LINC01503 knockdown or overexpression did not significantly affect the HSP60 mRNA levels (Supplementary Fig. 4C, available online), indicating that LINC01503 might instead affect HSP60 protein stability. To verify this, we examined the stability of HSP60 in CRC cells following CHX treatment, and found that overexpressing LINC01503 in CHX-treated CRC cells reduced the rate of HSP60 degradation (Fig. 7D). Additionally, the results of the MG132 treatment experiments showed that LINC01503 affected HSP60 stability in a proteasome dependent-degradation manner (Fig.7E). These results indicate that LINC01503 may increase HSP60 levels by inhibiting its degradation via the proteasomal degradation pathway.
As a chaperone, HSP60 binds to proteins to increase their stability[27]. We therefore believed that the stabilizing effect of LINC01503 on the VEGFA protein might be mediated by HSP60. To validate this hypothesis, we first performed the IP and Co-IP experiments using the anti-HSP60 and anti-Flag antibodies, and found that HSP60 bound to VEGFA (Fig. 7F and 7G). We then examined the effects of HSP60 on VEGFA protein levels, and found that high HSP60 expression significantly increased the protein levels of VEGFA (Fig. 7H). Meanwhile, HSP60 rescued the VEGFA protein from LINC01503-knockdown-mediated degradation (Fig. 7I), indicating that HSP60 increased VEGFA protein stability. Rescue experiments also showed that high HSP60 expression restored the expression and secretion of VEGFA in CRC cells following LINC01503 silencing (Fig. 7J and 7K). Finally, HSP60 reversed the inhibitory effect of LINC01503 downregulation in CRC cells on the tube formation and migration ability of HUVECs (Supplementary Fig. 4C–4E, available online).
Taken together, our results demonstrated that LINC01503 increased the stability of VEGFA by binding to HSP60, thereby promoting the growth and metastasis of vascular endothelial cells.
LINC01503 was transcriptionally activated by H3K27Ac
Increased levels of the H3K27Ac are commonly found in CRC versus normal tissues[28]. To investigate if the increased expression of LINC01503 in CRC cell lines was linked to abnormal histone modifications, we conducted a genomic bioinformatics analysis using UCSC Genome Browser (http://genome.ucsc.edu/), and found a significant enrichment of the H3K27Ac peak in the promoter region of LINC01503 (Supplementary Fig. 5A, available online), indicating that LINC01503 transcription might be regulated by H3K27Ac modification. We further assessed the enrichment of H3K27Ac in the LINC01503 promoter across various CRC cell lines with differing LINC01503 expression levels. The results showed that CRC cells with high H3k27Ac enrichment had correspondingly higher LINC01503 expression levels (Fig. 8A, Supplementary Fig. 5B and 5C [available online]). The H3K27Ac modification is mediated by the CREB-binding protein (CBP)/p300 complex.[29] C646, a histone acetyltransferase inhibitor targeting p300, significantly reduced H3K27Ac and LINC01503 levels in CRC cells (Fig. 8B and 8C). Moreover, transfecting CRC cell lines with siRNAs targeting CBP or p300 lowered their H3K27Ac and LINC01503 levels (Fig. 8D–8G). Furthermore, knocking down CBP or p300 significantly reduced the enrichment of H3K27Ac in the promoter region of LINC01503 (Fig. 8H). These results suggested that the expression of LINC01503 was regulated by the H3K27Ac modification.
Figure
8.LINC01503 was transcriptionally activated by histone H3 lysine 27 (H3K27) acetylation.
A: Chromatin immunoprecipitation (ChIP) followed by qRT-PCR was performed to assess H3K27Ac enrichment in various CRC cells with differing LINC01503 levels. B and C: HCT116 and CX-1 cells were treated with C646 (20 μmol/L) or DMSO for 48 h. LINC01503 expression levels were measured by qRT-PCR (B). Protein levels of H3K27Ac and total H3 were analyzed by Western blotting (C). D–G: HCT116 and CX-1 cells were transfected with specific siRNAs targeting CBP and P300. After 24 h, the mRNA levels of CBP, P300, and LINC01503 were quantified by qRT-PCR (D and F); protein levels of H3K27Ac and total H3 were measured by Western blotting (E and G). H: Following CBP or P300 siRNA transfection in HCT116 cells, the ChIP assay combined with qRT-PCR was utilized to detect H3K27Ac enrichment within the LINC01503 promoter region. Data were presented as mean ± standard deviation from three independent experiments. *P < 0.05 and **P < 0.01 by one-way analyses of variance, followed by Dunnett's tests for multiple comparisons and unpaired Student's t-test. Abbreviations: CRC, colorectal cancer; qRT-PCR, quantitative real-time PCR.
Collectively, these findings demonstrated that the H3K27Ac-mediated transcriptional regulation of LINC01503 promotes angiogenesis in CRC by inducing VEGFA expression in the miR-342-3p and HSP60-dependent manner (Fig. 9).
Figure
9.LINC01503, which is transcriptionally regulated by histone acetylation, promotes angiogenesis in colorectal cancer (CRC) by inducing the expression of VEGFA through both the miR-342-3p/VEGFA and hsp60/VEGFA signaling pathways.
In the current study, we demonstrated that an elevated LINC01503 expression in CRC tumor tissues was relevant to tumor progression, vascular invasion, clinical stage, and poor prognosis. These findings implied that LINC01503 might be an independent prognostic marker for CRC. Additionally, LINC01503 expression was positively correlated with CD31 and VEGFA levels. Functional assays, both in vitro and in vivo, demonstrated that LINC01503 enhanced the proliferation, migration, and invasion of CRC cells. Furthermore, LINC01503 was found to promote angiogenesis by upregulating VEGFA expression. Collectively, these results indicate that LINC01503 may contribute to CRC development through the promotion of angiogenesis.
Under physiological conditions, angiogenesis is delicately regulated by pro-angiogenic and anti-angiogenic regulatory factors. In the tumor tissue, however, this balance is disrupted, resulting in an increased angiogenesis[30]. Vascular endothelial cell function within the tumor microenvironment is controlled by VEGF, a cytokine secreted by tumor cells. In addition, other angiogenic signaling pathways supply tumors with oxygen and nutrients to maintain tumor cell proliferation; this environment also creates some favorable conditions for tumor metastasis[8]. Many studies have shown that lncRNAs affect tumor progression by regulating angiogenesis in CRC. For instance, lncRNA SNHG17 accelerated CRC cell proliferation and migration by sponging miR-23a-3p to regulate CXCL12-mediated angiogenesis[31]. The levels of angiogenesis regulators HIF-1α and VEGF were increased in CRC cells with high lncRNA SH3pXD2A-AS1 expression[32]. LncRNA ZFAS1 upregulated the expression of VEGFA through ceRNA[33], while lncRNA TPT1-AS1 promoted the secretion of VEGFA, which plays a key role in CRC angiogenesis[9]. These studies provide a theoretical basis for lncRNA as targets in anti-angiogenic cancer therapy.
In the current study, we found that LINC01503 promoted VEGFA expression by binding to miR-342-3p and increasing the tube formation and migration ability of HUVECs. In addition to regulating VEGFA at the mRNA level, we found that LINC01503 stabilized the VEGFA protein. ChIRP and mass spectrometry identified HSP60 as the binding partner of LINC01503, while RIP assays demonstrated the direct binding between HSP60 and LINC01503. We further demonstrated that LINC01503 promoted HSP60 protein expression by inhibiting its ubiquitin-dependent degradation, and that HSP60 might bind and stabilize the VEGFA protein. Rescue experiments further showed that LINC01503 stabilized VEGFA protein by binding to HSP60, thereby promoting the growth and metastasis of vascular endothelial cells.
Heat shock proteins (HSPs) are molecular chaperones, which exert their biological functions by affecting protein stability[34]. HSPs such as HSP60, HSP70, HSP90, and HSP110 are widely and highly expressed in CRC, liver cancer, and other cancers[34, 35]. High HSP expression may contribute to the establishment of an immunosuppressive tumor microenvironment, and it may promote tumorigenesis, poor prognosis, and treatment resistance[34, 36]. HSPs also exert pro-cancer effects by maintaining the stability and function of angiogenic and oncogenic proteins[36, 37]. For instance, HSP90 promoted tumor angiogenesis and growth by binding to and stabilizing the protein kinase D2[38] and the macrophage migration inhibitory factor[39]. Therefore, efficient regulation of HSP60 by LINC01503 may be critical to prevent the excessive stabilization of major pro-angiogenic oncogenic proteins.
CRC tumors induce neovascularization, which provides a strong rationale for antiangiogenic strategies as a therapy. Indeed, anti-VEGF therapy has already been approved by the FDA for the treatment of metastatic CRC[40]. However, the high risk of fatal bleeding complications, including catastrophic side effects such as intestinal perforation and arterial embolism, has limited its widespread use in CRC. Additionally, the mechanisms of adaptive anti-angiogenic therapy on its evasion and resistance to VEGF inhibition are unclear[41, 42]. LncRNAs have been reported to be involved in tumor progression by regulating the expression of VEGF in metastatic CRC[17, 18]. Therefore, the search for upstream regulatory molecules of VEGF to provide new targets for anti-VEGF therapy has attracted an increasing attention. In the current study, we found that LINC01503 promoted the expression of VEGFA by simultaneously regulating both mRNA and protein stability of VEGFA. Targeted silencing of LINC01503 not only reduced the expression of VEGFA in CRC cells, but also significantly inhibited tumor growth and metastasis of transplanted tumors in nude mice. Moreover, histone acetylation in tumors has previously been shown to promote VEGFA transcription[43]. Similarly, our results demonstrated that histone acetylation also promotes LINC01503 transcription. These results suggest that histone acetylation may, on the one hand, upregulate VEGFA by promoting its RNA level, and on the other hand, it may also increase the expression levels of LINC01503 epigenetically, therefore up-regulating VEGFA expression through LINC01503. These findings imply that LINC01503 may potentially be targeted in the anti-angiogenesis therapy and serve as a VEGFA regulatory molecule to evaluate the effect of an anti-VEGF therapy.
Although there are some important findings from our present study, there are also some limitations. We lack enough evidence to confirm whether the oncogenic role of LINC01503 in CRC is mainly dependent on the VEGFA protein. In addition to the epigenetic regulation of LINC01503, we also need to investigate whether some small molecules, such as vitamin D, can be used to regulate the expression of LINC01503[44, 45], which will also be our subsequent research.
In summary, the current study demonstrates that LINC01503 may regulate VEGFA expression through the miR-342-3p/VEGFA and HSP60/VEGFA axes, activate the AKT signaling pathway downstream of VEGFA in vascular endothelial cells in a paracrine manner, and promote angiogenesis, which in turn drives CRC progression.
Fundings
The present study is supported by the Priority Academic Program Development of Jiangsu Higher Education Institutions and the Key Project of Jiangsu Commission of Health (Grant No. ZD2022012).
Acknowledgments:
We acknowledge and appreciate our professor and institutional colleagues for their experimental technical support.
Kaufman JD, Adar SD, Barr RG, et al. Association between air pollution and coronary artery calcification within six metropolitan areas in the USA (the Multi-Ethnic Study of Atherosclerosis and Air Pollution): a longitudinal cohort study[J]. Lancet, 2016, 388 (10045): 696–704. doi: 10.1016/S0140-6736(16)00378-0
[2]
de Bont J, Jaganathan S, Dahlquist M, et al. Ambient air pollution and cardiovascular diseases: an umbrella review of systematic reviews and meta-analyses[J]. J Intern Med, 2022, 291 (6): 779–800. doi: 10.1111/joim.13467
[3]
DeJarnett N, Conklin DJ, Riggs DW, et al. Acrolein exposure is associated with increased cardiovascular disease risk[J]. J Am Heart Assoc, 2014, 3 (4): e000934. doi: 10.1161/JAHA.114.000934
[4]
Moghe A, Ghare S, Lamoreau B, et al. Molecular mechanisms of acrolein toxicity: relevance to human disease[J]. Toxicol Sci, 2015, 143 (2): 242–255. doi: 10.1093/toxsci/kfu233
[5]
Zirak MR, Mehri S, Karimani A, et al. Mechanisms behind the atherothrombotic effects of acrolein, a review[J]. Food Chem Toxicol, 2019, 129: 38–53. doi: 10.1016/j.fct.2019.04.034
[6]
Chen L, Wu X, Zeb F, et al. Acrolein-induced apoptosis of smooth muscle cells through NEAT1-Bmal1/Clock pathway and a protection from asparagus extract[J]. Environ Pollut, 2020, 258: 113735. doi: 10.1016/j.envpol.2019.113735
[7]
Xu Y, Wu L, Chen A, et al. Protective effects of olive leaf extract on acrolein-exacerbated myocardial infarction via an endoplasmic reticulum stress pathway[J]. Int J Mol Sci, 2018, 19 (2): 493. doi: 10.3390/ijms19020493
[8]
Zhou Y, Jin W, Wu Q, et al. Acrolein: formation, health hazards and its controlling by dietary polyphenols[J]. Crit Rev Food Sci Nutr, 2023, doi: 10.1080/10408398.2023.2214625.
[9]
Vasim I, Majeed CN, DeBoer MD. Intermittent fasting and metabolic health[J]. Nutrients, 2022, 14 (3): 631. doi: 10.3390/nu14030631
[10]
Varady KA. Intermittent versus daily calorie restriction: which diet regimen is more effective for weight loss?[J]. Obes Rev, 2011, 12 (7): e593–e601.
[11]
Malinowski B, Zalewska K, Węsierska A, et al. Intermittent fasting in cardiovascular disorders—an overview[J]. Nutrients, 2019, 11 (3): 673. doi: 10.3390/nu11030673
[12]
Chen Y, Su J, Yan Y, et al. Intermittent fasting inhibits high-fat diet-induced atherosclerosis by ameliorating hypercholesterolemia and reducing monocyte chemoattraction[J]. Front Pharmacol, 2021, 12: 719750. doi: 10.3389/fphar.2021.719750
[13]
Schilperoort M, van den Berg R, Bosmans LA, et al. Disruption of circadian rhythm by alternating light-dark cycles aggravates atherosclerosis development in APOE*3-Leiden. CETP mice[J]. J Pineal Res, 2020, 68 (1): e12614. doi: 10.1111/jpi.12614
[14]
Huang T, Mariani S, Redline S. Sleep irregularity and risk of cardiovascular events: the multi-ethnic study of atherosclerosis[J]. J Am Coll Cardiol, 2020, 75 (9): 991–999. doi: 10.1016/j.jacc.2019.12.054
[15]
Kervezee L, Shechter A, Boivin DB. Impact of shift work on the circadian timing system and health in women[J]. Sleep Med Clin, 2018, 13 (3): 295–306. doi: 10.1016/j.jsmc.2018.04.003
[16]
Crnko S, Du Pré BC, Sluijter JPG, et al. Circadian rhythms and the molecular clock in cardiovascular biology and disease[J]. Nat Rev Cardiol, 2019, 16 (7): 437–447. doi: 10.1038/s41569-019-0167-4
[17]
Sun L, Ma J, Turck CW, et al. Genome-wide circadian regulation: a unique system for computational biology[J]. Comput Struct Biotechnol J, 2020, 18: 1914–1924. doi: 10.1016/j.csbj.2020.07.002
[18]
Xie M, Tang Q, Nie J, et al. BMAL1-downregulation aggravates porphyromonas gingivalis-induced atherosclerosis by encouraging oxidative stress[J]. Circ Res, 2020, 126 (6): e15–e29.
[19]
Wu X, Chen L, Zeb F, et al. Clock-Bmal1 mediates MMP9 induction in acrolein-promoted atherosclerosis associated with gut microbiota regulation[J]. Environ Pollut, 2019, 252: 1455–1463. doi: 10.1016/j.envpol.2019.06.042
[20]
Ge Y, Zhou M, Chen C, et al. Role of AMPK mediated pathways in autophagy and aging[J]. Biochimie, 2022, 195: 100–113. doi: 10.1016/j.biochi.2021.11.008
[21]
Masri S. Sirtuin-dependent clock control: new advances in metabolism, aging and cancer[J]. Curr Opin Clin Nutr Metab Care, 2015, 18 (6): 521–527. doi: 10.1097/MCO.0000000000000219
[22]
Li H, Kilgallen AB, Münzel T, et al. Influence of mental stress and environmental toxins on circadian clocks: implications for redox regulation of the heart and cardioprotection[J]. Br J Pharmacol, 2020, 177 (23): 5393–5412. doi: 10.1111/bph.14949
[23]
Sundar IK, Sellix MT, Rahman I. Redox regulation of circadian molecular clock in chronic airway diseases[J]. Free Radic Biol Med, 2018, 119: 121–128. doi: 10.1016/j.freeradbiomed.2017.10.383
[24]
Duan L, Zhao Y, Jia J, et al. Myeloid-restricted CD68 deficiency attenuates atherosclerosis via inhibition of ROS-MAPK-apoptosis axis[J]. Biochim Biophys Acta Mol Basis Dis, 2023, 1869 (5): 166698. doi: 10.1016/j.bbadis.2023.166698
[25]
Adafer R, Messaadi W, Meddahi M, et al. Food timing, circadian rhythm and chrononutrition: a systematic review of time-restricted eating's effects on human health[J]. Nutrients, 2020, 12 (12): 3770. doi: 10.3390/nu12123770
[26]
Di Tano M, Raucci F, Vernieri C, et al. Synergistic effect of fasting-mimicking diet and vitamin C against KRAS mutated cancers[J]. Nat Commun, 2020, 11 (1): 2332. doi: 10.1038/s41467-020-16243-3
[27]
Wolf D, Ley K. Immunity and inflammation in atherosclerosis[J]. Circ Res, 2019, 124 (2): 315–327. doi: 10.1161/CIRCRESAHA.118.313591
[28]
Matsubara Y, Kiyohara H, Teratani T, et al. Organ and brain crosstalk: the liver-brain axis in gastrointestinal, liver, and pancreatic diseases[J]. Neuropharmacology, 2022, 205: 108915. doi: 10.1016/j.neuropharm.2021.108915
[29]
González A, Hall MN, Lin SC, et al. AMPK and TOR: the yin and yang of cellular nutrient sensing and growth control[J]. Cell Metab, 2020, 31 (3): 472–492. doi: 10.1016/j.cmet.2020.01.015
[30]
Pan X, Mota S, Zhang B. Circadian clock regulation on lipid metabolism and metabolic diseases[M]//Jiang X. Lipid Transfer in Lipoprotein Metabolism and Cardiovascular Disease. Singapore: Springer, 2020: 53–66.
[31]
Alfarhan M, Jafari E, Narayanan SP. Acrolein: a potential mediator of oxidative damage in diabetic retinopathy[J]. Biomolecules, 2020, 10 (11): 1579. doi: 10.3390/biom10111579
[32]
Rom O, Korach-Rechtman H, Hayek T, et al. Acrolein increases macrophage atherogenicity in association with gut microbiota remodeling in atherosclerotic mice: protective role for the polyphenol-rich pomegranate juice[J]. Arch Toxicol, 2017, 91 (4): 1709–1725. doi: 10.1007/s00204-016-1859-8
[33]
Hong J, Song Y, Xie J, et al. Acrolein promotes aging and oxidative stress via the stress response factor DAF-16/FOXO in Caenorhabditis elegans[J]. Foods, 2022, 11 (11): 1590. doi: 10.3390/foods11111590
[34]
Jiang L, Luo S, Qiu T, et al. Bidirectional role of reactive oxygen species during inflammasome activation in acrolein-induced human EAhy926 cells pyroptosis[J]. Toxicol Mech Methods, 2021, 31 (9): 680–689. doi: 10.1080/15376516.2021.1953204
[35]
Yu J, Jiang Q, Liu N, et al. Apigenin and apigenin-7, 4'-O-dioctanoate protect against acrolein-aggravated inflammation via inhibiting the activation of NLRP3 inflammasome and HMGB1/MYD88/NF-κB signaling pathway in Human umbilical vein endothelial cells (HUVEC)[J]. Food Chem Toxicol, 2022, 168: 113400. doi: 10.1016/j.fct.2022.113400
[36]
Srivastava S, Sithu SD, Vladykovskaya E, et al. Oral exposure to acrolein exacerbates atherosclerosis in apoE-null mice[J]. Atherosclerosis, 2011, 215 (2): 301–308. doi: 10.1016/j.atherosclerosis.2011.01.001
[37]
D'Onofrio N, Servillo L, Balestrieri ML. SIRT1 and SIRT6 signaling pathways in cardiovascular disease protection[J]. Antioxid Redox Signal, 2018, 28 (8): 711–732. doi: 10.1089/ars.2017.7178
[38]
Jordan SD, Lamia KA. AMPK at the crossroads of circadian clocks and metabolism[J]. Mol Cell Endocrinol, 2013, 366 (2): 163–169. doi: 10.1016/j.mce.2012.06.017
[39]
Imai SI, Guarente L. NAD+ and sirtuins in aging and disease[J]. Trends Cell Biol, 2014, 24 (8): 464–471. doi: 10.1016/j.tcb.2014.04.002
[40]
Reustle A, Torzewski M. Role of p38 MAPK in atherosclerosis and aortic valve sclerosis[J]. Int J Mol Sci, 2018, 19 (12): 3761. doi: 10.3390/ijms19123761
[41]
Goldsmith CS, Bell-Pedersen D. Diverse roles for MAPK signaling in circadian clocks[J]. Adv Genet, 2013, 84: 1–39.
[42]
Li B, Li D, Ni H, et al. The circadian clock regulator Bmal1 affects traumatic brain injury in rats through the p38 MAPK signalling pathway[J]. Brain Res Bull, 2022, 178: 17–28. doi: 10.1016/j.brainresbull.2021.11.003
[43]
Duregon E, Pomatto-Watson LCDD, Bernier M, et al. Intermittent fasting: from calories to time restriction[J]. GeroScience, 2021, 43 (3): 1083–1092. doi: 10.1007/s11357-021-00335-z
[44]
Naous E, Achkar A, Mitri J. Intermittent fasting and its effects on weight, glycemia, lipids, and blood pressure: a narrative review[J]. Nutrients, 2023, 15 (16): 3661. doi: 10.3390/nu15163661
[45]
Jamshed H, Beyl RA, Della Manna DL, et al. Early time-restricted feeding improves 24-hour glucose levels and affects markers of the circadian clock, aging, and autophagy in humans[J]. Nutrients, 2019, 11 (6): 1234. doi: 10.3390/nu11061234
[46]
Pignatta S, Cortesi M, Arienti C, et al. Effects of radiotherapy and short-term starvation combination on metastatic and non-tumor cell lines[J]. DNA Repair (Amst), 2020, 95: 102949. doi: 10.1016/j.dnarep.2020.102949
[47]
Chen Y, Su J, Yan Y, et al. Intermittent fasting inhibits high-fat diet-induced atherosclerosis by ameliorating hypercholesterolemia and reducing monocyte chemoattraction[J]. Front Pharmacol, 2021, 12: 719750.
[48]
Kavanagh K, Bashore AC, Davis MA, et al. Early time-restricted feeding improves high-density lipoprotein amount and function in nonhuman primates, without effects on body composition[J]. Obesity (Silver Spring), 2023, 31 Suppl 1 (Suppl 1): 75–84.
[49]
Dorighello GG, Rovani JC, Luhman CJF, et al. Food restriction by intermittent fasting induces diabetes and obesity and aggravates spontaneous atherosclerosis development in hypercholesterolaemic mice[J]. Br J Nutr, 2014, 111 (6): 979–986. doi: 10.1017/S0007114513003383
[50]
Okoshi K, Cezar MDM, Polin MAM, et al. Influence of intermittent fasting on myocardial infarction-induced cardiac remodeling[J]. BMC Cardiovasc Disord, 2019, 19 (1): 126. doi: 10.1186/s12872-019-1113-4
Authors and Reviewers
DownLoad:
