CXCL12/CXCR4 promotes inflammation-driven colorectal cancer progression through activation of RhoA signaling by sponging miR-133a-3p
- 322 Downloads
Activation of CXCL12/CXCR4 axis has been found to be associated with invasion and metastasis in many cancers. However, the underlying mechanism remains elusive. Increasing data highlight that non-coding RNAs are linked to CRC progression.
The effects of CXCR4 were investigated using villin-CXCR4 transgenic mice model by flow cytometry assay, immunohistochemistry, and Western blot. The mechanism was explored through bioinformatics, luciferase reporter assay and RNA immunoprecipitation assay.
We found that high CXCR4 expression exacerbated colitis-associated cancer in villin-CXCR4 transgenic mice. CXCR4+/−Apcmin/+ compound mutant mice demonstrated higher colorectal tumorigenesis than Apcmin/+ mice. Furthermore, overexpression of CXCR4 was found to promote the epithelial-mesenchymal transition (EMT) and infiltration of myeloid-derived suppressor cells (MDSCs) and macrophages in colonic tissue, accelerating colitis-associated and Apc mutation-driven colorectal tumorigenesis and progression. Notably, miR-133a-3p was found to be significantly decreased in HCT116 cells overexpressing CXCR4 by miRNA sequencing. miR-133a-3p was proved to target RhoA, which is involved in cytoskeletal reorganization that drive cell motility. Importantly, CXCL12/CXCR4-induced upregulation of lncRNA XIST functioned as a ceRNA to sponge miR-133a-3p, thereby liberating the repression of RhoA by miR-133a-3p. The negative correlation of miR-133a-3p with RhoA was also confirmed in human CRC tissues and CXCR4+/− mice.
Our findings revealed the critical role of CXCR4 in promoting progression of inflammatory colorectal cancer through recruiting immunocytes and enhancing cytoskeletal remodeling by lncRNA XIST/ miR-133a-3p/ RhoA signaling. These results provide novel potential therapeutic targets for hindering CXCL12/CXCR4-induced CRC progression.
KeywordsCXCR4 RhoA miR-133a-3p lncRNA XIST CAC
Azoxymethane/dextran sulfate sodium
Competitive endogenous RNAs
Chemokine (C-X-C motif) receptor 4
Hematoxylin & eosin
Intestinal epithelial cells
Long non-coding RNAs
Myeloid-derived suppressor cells
Polycomb repressive complex 2
RNA-induced silencing complex
Stromal cell-derived factor 1
LncRNA X-inactive-specific transcript
Colorectal cancer (CRC) is the third most common cancer worldwide. The mortality of CRC is principally attributable to the development of liver metastasis that is a critical determinant of poor prognosis of CRC patients . CRC involves multi-factorial and multistep progression in which abnormal expression of Chemokine (C-X-C motif) receptor 4 (CXCR4) has been demonstrated to play a crucial role in the invasion and liver metastasis of CRC . CXCR4 belongs to G protein-coupled receptor superfamily, which selectively binds to Stromal cell-derived factor 1 (SDF-1), also known as CXCL12 to promote cancer metastasis . However, the underlying mechanism has not been fully understood. Some signaling pathways such as Wnt/β-catenin and PI3K-AKT are reported to contribute to the CXCR4-mediated invasion and angiogenesis in cancers [4, 5]. Recent data highlight that microRNAs (miRNAs) are closely associated with CRC progression . MiRNAs belong to a cluster of highly conserved 22-nucleotide single-stranded RNAs that trigger mRNA degradation or translational inhibition by forming the RNA-induced silencing complex (RISC) . However, whether miRNAs are involved in CXCR4-driven invasion and metastasis of CRC remains poorly explored.
It is recently proposed that a large number of non-coding RNAs might function as molecular sponges for miRNAs and hence functionally liberate other RNA transcripts targeted by active miRNAs . Long non-coding RNAs (lncRNAs), longer than 200 nucleotides, control gene expression by diverse modes, including miRNA sponging, epigenetic modification and mRNA stabilization . In CRC, aberrant expression of lncRNAs promotes tumorigenesis and metastasis by acting as competitive endogenous RNAs (ceRNAs) and “miRNA sponges” to antagonize their functions and leads to the de-repression of their endogenous targets [10, 11]. Recently lncRNA X-inactive-specific transcript (XIST) has attracted more attention and it was demonstrated that lncRNA XIST was significantly increased in CRC and could be a potential oncogene in CRC . LncRNA XIST is transcribed from the X inactivation center of the X chromosome that directly binds to polycomb repressive complex 2 (PRC2), which is the epigenetic complex responsible for the trimethylation of histone H3 at Lys27 (H3K27me3), silencing the whole chromosome . LncRNA XIST was found to expedite metastasis by competing for miR-200b-3p to modulate the expression of ZEB1 . It could also sponge miR-34a to promote colon cancer progression via Wnt/β-catenin signaling pathway .
Increasing data indicated that Rho signaling contributes to tumor invasion and metastasis by actin cytoskeletal reorganization. RhoA is capable of mediating stress fiber formation and generating contractile force needed for retraction of the trailing edge during migration. ROCK is a major mediator of RhoA function. Inhibition of ROCK blocks the formation of RhoA-mediated actin cytoskeletal structures. Many studies have shown that RhoA/ROCK-mediated cytoskeletal regulation plays a key role in cancer metastasis. Therefore, targeting RhoA/ROCK signaling-associated miRNAs could be useful strategy for inhibiting migration of cancer cells in treatment of cancer metastasis.
In this study, using villin-CXCR4 transgenic mice, we found that overexpressing CXCR4 enhanced tumor load in azoxymethane/dextran sulfate sodium (AOM/DSS)-treated mice and Apcmin/+ mice. CXCR4 promotes the progression of colitis-associated cancer (CAC) by enhancing EMT and recruiting myeloid-derived suppressor cells (MDSCs) and macrophages. In vitro assay using cell model by overexpressing and silencing of CXCR4, we unraveled that miR-133a-3p was significantly reduced in colorectal cancer cells overexpressing CXCR4. We demonstrated that RhoA, a GTPase that facilitates actin polymerization, was the direct target gene of miR-133a-3p. Furthermore, lncRNA XIST functions as a ceRNA sponging miR-133a-3p, thereby de-repressing target gene of RhoA. Our results shed new lights on the progression of CRC driven by inflammation and cytoskeletal reorganization through the lncRNA XIST/ miR-133a-3p/ RhoA signaling pathway.
Materials and methods
Construction of mice models
Villin-CXCR4 transgenic mice (CXCR4+/− Tg) were generated by Cyagen Biosciences Inc. (Guangzhou, China) overexpressing CXCR4 in intestinal epithelial cells (IEC) under the control of villin promoter.
C57BL/6 J male mice heterozygous for Apc allele (ApcMin/+) were purchased from Jackson Laboratory (Shanghai, China). CXCR4+/−ApcMin/+ mice were established by mating ApcMin/+ males with CXCR4+/− females. Genotyping was performed by allele specific PCR assays using the primers described in Additional file 1: Table S1. Mice were maintained in the Animal Care Facility of Capital Medical University. Housing and care of the animals were in accordance with the NIH Guidelines for the Care and Use of Laboratory Animals. All experiments related to animals were approved by Animal Welfare Committee of Capital Medical University.
AOM/DSS-induced colonic tumorigenesis model
The establishment of AOM/DSS mouse model was conducted as previously described . In brief, 8 weeks old C57BL/6 mice (n = 8 per group) were treated by a single intraperitoneal injection of AOM (10 mg/kg, Sigma-Aldrich) and subsequent oral administration of 1% DSS (MP Biomedicals) in drinking water ad libitum for 7 consecutive days and then removed for 14 days intervals. Three cycles of DSS are performed to establish this model. To inhibit CXCR4, AMD3100 (2.5 mg/kg, Sigma-Aldrich) was administered intraperitoneally daily for nine consecutive weeks. Mice were monitored twice a week for clinical signs of illness including rectal bleeding, rectal prolapse and body weight loss. CXCR4+/−ApcMin/+ mice (aged at 12 weeks, n = 15) were treated with 3 cycles of 1% DSS and sacrificed at age of 22, 26 and 30 weeks respectively. DSS-induced colitis was scored as disease activity index (DAI) obtained based on weight loss, stool consistency, and bloody excreta as follows: weight loss score = 0: < 1%, 1: 1–5%, 2: 5–10%, 3:10–15%, 4: > 15%; stool consistency score = 0: normal, 2: loose, 4: diarrhea; blood in excreta score = 0: normal; 2: reddish, 4: bloody . Upon sacrifice, colons and intestines were excised and opened longitudinally to count tumor numbers. Size of tumors was measured with a caliper to evaluate intestinal tumor development. Colons were further processed for histopathological analysis.
Immunohistochemistry and immunofluorescence staining
Colorectal tissues and tumors were fixed in 10% formalin followed by paraffin embedding. Tissue sections of 5 μm were stained with hematoxylin & eosin (HE) for pathological evaluation using standard procedures. Immunohistochemistry (IHC) was performed using the antibodies of E-cadherin (3195), Vimentin (5741), active β-catenin (8814) (CST), CXCR4 (ab1670) (Abcam), RhoA (sc-418) (Santa Cruz), F4/80 (GB11207), CD4 (GB13064), CD8 (GB11068), CD68 (GB11067) (Servicebio) according to manufacturer’s instructions. The results of the IHC analysis were taken with a digital slide scanning system (Pannoramic Scan, 3DHISTECH Ltd.). The results were semi-quantified of mean density (IOD/area) by image-pro plus 6 software (IPP, USA).
For immunofluorescence staining of cytoskeleton, Rhodamine-labeled phalloidin and nuclear labeling with DAPI were performed as instructed by cytoskeleton Inc. briefly, SW620 cells overexpressing CXCR4 and vector control (GFP-labeled) were fixed in 4% paraformaldehyde, blocked with 2% bovine serum albumin, and incubated with Rhodamine-labeled phalloidin at 4 °C overnight. Nuclei were labeled with DAPI. Fluorescence images were captured using a Leica confocal scanning system.
Colon tissue fractions or cell pellets were lysed in RIPA buffer (Beyotime, Nanjing, China) supplemented with 1 mM PMSF and complete protease inhibitor mixture (Roche Pharmaceuticals) on ice. Total protein was quantified by using Pierce BCA Protein Assay Kit (Thermo Scientific, Germany). Equal amounts of protein lysates were resolved by SDS-PAGE and transferred to PVDF membranes (Millipore, Merck). The membranes were immunoblotted at 4 °C with primary antibodies against E-cadherin (3195), Vimentin (5741), Snail (3879), c-Myc (5605), MMP7 (3801), β-catenin (8480), active β-catenin (8814), IL-1 (12242), IL-6 (12912), COX-2 (12282), P-STAT3 (9145), STAT3 (12640), P-JAK2 (3776), JAK2 (3230), RhoA (2117), P-FAK (3284), FAK (13009), ROCK1 (4035), P-MLC (3671), MLC (8505), (CST) and CXCR4 (ab1670) (Abcam) followed by HRP-conjugated secondary antibodies. Immunoreactive products were visualized using Fluorchem FC3 system (Proteinsimple, USA) by chemiluminescence (Millipore, USA) and quantified by densitometry using Alphaview software. Densitometric analyses of bands were normalized with β -actin functioning as a loading control.
RNA extraction and RT-qPCR
Total RNA was extracted using TRIzol reagent (Invitrogen). Reverse transcription of total miRNA was performed by using a miScript reverse transcription kit (Qiagen) in accordance with manufacturer’s protocol. MiScript SYBR Green PCR kit (Qiagen) and miR-133a specific primers were used to determine the expression of mature miRNAs. RNU6B was used as an internal control.
To determine mRNA expression, reverse transcription was carried out using SuperScript III First-Strand Synthesis System (Invitrogen) according to manufacturer’s instruction. Real-time PCR was performed in triplicate using Powerup SYBR Green qPCR master mix in an ABI 7500 fast real-time PCR system (Applied Biosystems). The thermal cycling was initiated by polymerase activation step for 10 min at 95 °C followed by 40 cycles of denaturation (95 °C for 15 s) and annealing/extension (60 °C for 1 min). Target gene expression levels were normalized to GAPDH and determined by a previously described method .
Flow cytometry analysis
The immunocytes isolated from blood and inflamed colonic tumors were subjected to flow cytometry analysis according to previously described method . For multicolor flow cytometry analysis, cells were stained with indicated antibodies (Additional file 1: Table S2) and acquired on a BD LSR Fortessa flow cytometer. Briefly, the immnocytes were incubated for 30 min on ice with the appropriate combination of antibodies in staining buffer (Biolegend) and the data were analyzed using TreeStar Flowjo software.
Enzyme-linked immunosorbent (ELISA) assay
The level of IL-1, IL-6 and tumor necrosis factor alpha (TNF-α) was quantified by using ELISA kits (Immunoway, USA) according to manufacturer’s protocols.
Cell culture and cell transfection
Human colorectal adenocarcinoma cell lines SW620 and HCT116 were purchased from American Type Culture Collection (ATCC, Rockville, MD). Cells were grown in RPMI 1640 medium containing 10% fetal bovine serum (FBS) and penicillin/streptomycin at 37 °C in a humid atmosphere (5% CO2). Cancer cells were infected with lentivirus expressing CXCR4 (LV-CXCR4) or transfected with CXCR4 siRNA as described previously . To knockdown lncRNA XIST, 200 nM of XIST siRNA with oligonucleotide sequences 5′- GCUGCACUAAUUGACUAAUTT-3′ were used to transfect cells for 48 h (GenePharma, Shanghai, China). To determine the roles of miR-133a, cells were transfected with 100 nM of miR-133a mimics or inhibitors (GenePharma Shanghai, China) for 48 h by using lipofectamine 2000 (Invitrogen) according to manufacturer’s instructions.
Total RNA was extracted from HCT116 cells infected with lentiviral CXCR4 (Genechem, Shanghai) and stimulated with CXCL12 using Trizol reagent (Invitrogen). RNA integrity and concentration was assessed by using the RNA Nano 6000 assay kit of the Bioanalyzer 2100 system (Agilent, CA). Library construction and sequencing were performed on an Illumina HiSeq2500 sequencer according to manufacturer’s specifications (Illumina) at Annoroad Gene Technology (Beijing, China). miRNAs with P < 0.05 and |log2_ratio| ≥ 1 are identified as differentially expressed miRNAs.
Plasmid construction and luciferase reporter assay
To confirm that RhoA was the target gene of miR-133a, RhoA gene containing miR-133a binding sites was synthesized and digested by Xba I and cloned into GV272 vector (GeneChem, Shanghai). Similarly, XIST gene containing miR-133a binding sites was cloned into GV272 vector (GeneChem, Shanghai) to make the luciferase constructs. Wild type and mutant inserts were confirmed by sequencing. GV272 empty vector was used as control (Con081).
The luciferase reporter assay was performed as previously described . Briefly, HCT116 cells were co-transfected with wild type or mutant RhoA-3’UTR-Luc/ LncRNA XIST-Luc firefly luciferase constructs (100 ng) and 40 nM miR-133a mimics or inhibitors using lipofectamine 2000 reagent. 2 ng of pRL-SV40 plasmid was transfected to monitor transfection efficiency. Luciferase activity was determined by a dual-luciferase reporter assay system (Promega).
RNA immunoprecipitaion (RIP) assay
RIP assay was performed by the EZ-Magna RIP kit (Millipore) according to the instructions. Briefly, cells were lysed in complete RIP lysis buffer. Then, the cell lysate was incubated with RIP buffer containing magnetic beads conjugated to a human anti-Ago2 antibody (Millipore). The precipitates were incubated with proteinase K with shaking to digest proteins and the immunoprecipitated RNA was isolated. Subsequently, the purified RNA was subjected to RT-qPCR analysis.
Human CRC tissue specimens
The experiment was approved by the Institutional Review Board of Xuan Wu Hospital of Capital Medical University. The informed consents were obtained from all patients. Human CRC tissue specimens and adjacent normal mucosa were obtained from Xuan Wu Hospital of Capital Medical University (Beijing, China) and were confirmed by pathological analysis. The clinical information was shown in Additional file 1: Table S3.
Rho activation assay
RhoA GTP levels were assessed using a Rho-binding domain (RBD) affinity precipitation assay (Cytoskeleton, Denver, CO). Briefly, HCT116 and SW620 cells were grown to 80% confluence in 100 mm tissue culture plates and serum-starved overnight. Cells were infected with lentivirus expressing CXCR4 (LV-CXCR4) or transfected with CXCR4 siRNA or miR-133a-3p mimics before stimulation with 50 ng/ml CXCL12 for 12 h. Cells were solubilized in lysis buffer and cleared by centrifugation. Cleared lysates were incubated with RBD-GST beads for 1 h at 4 °C to bind RhoGTP. The beads were washed, eluted in SDS-sample buffer and analyzed by immunoblotting with anti-RhoA monoclonal antibody.
All data were presented as mean ± SD and statistical data were analyzed using SPSS 12.0. Statistical differences among multiple groups were evaluated by one-way analysis of variance (ANOVA) followed by Dunnett (multiple comparisons to the same control) post hoc tests. Student’s t test was used to compare differences between two groups. The Spearman’s correlation was used to evaluate potential correlations between miR-133a and RhoA expression in paired CRC tissues. A P < 0.05 was considered to indicate statistical significance.
Overexpression of CXCR4 exacerbated AOM/DSS-induced CAC in mice
CXCR4 +/− Apc min/+ compound mutant mice exhibited more tumor load than Apc min/+ mice
CXCR4 enhanced CRC progression by promoting EMT and Wnt-β-catenin activation
We next sought to determine whether high CXCR4 expression promotes tumor progression to an invasive phenotype in vivo. Immunohistochemistry was performed to detect the expression levels of key genes related to epithelial-to-mesenchymal transition (EMT) phenotype. EMT is a process by which epithelial cells lose their cell polarity and gain migratory and invasive phenotype. It is characterized by loss of epithelial markers E-cadherin and upregulation of mesenchymal marker vimentin. As expected, CXCR4+/− mice revealed a pronounced downregulation of E-cadherin in the cell membrane of intestine epithelial cells and upregulation of cytoplasmic vimentin in the mesenchymal cells. These results suggest that CXCR4 enhanced the invasion and metastasis of CAC by promoting EMT phenotype, which could be impaired by AMD3100 treatment (Fig. 3b, c). Consistently, we confirmed the loss of E-cadherin and upregulation of vimentin expressiom in colon tissue of CXCR4+/− mice by Western blot analysis (Fig. 3d, e). Furthermore, the marked activation of Wnt-β-catenin signaling pathway was also found in CXCR4+/− mice compared to that of WT mice by immunohistochemistry and Western blot analysis, showing more nuclear staining of β-catenin and upregulation of active β-catenin in colonic tissue of CXCR4+/− mice (Fig. 3b-e).
CXCR4 enhanced CRC progression by increasing inflammatory cytokines and recruiting immune suppressive cells
MDSCs and tumor-associated macrophages (TAM) could enhance tumor progression by facilitating the growth and migration of tumor cells and suppression of immune response . To determine whether CXCR4 promotes the recruitment of immune cells from the circulatory system into the inflamed intestinal tissue, we used cell surface markers to define the immunocyte profiles in blood and colonic tissues. Overexpression of CXCR4 dramatically increased the percentages of macrophage (CD11b+F4/80+), granulocytic MDSCs (G-MDSCs, CD11b+Ly6G+) and monocytic MDSCs (M-MDSCs, CD11b+Ly6C+) in the blood (Fig. 4c). Importantly, CXCR4+/− mice exhibited more accumulation of macrophages and MDSCs in the intraepithelial and lamina propria of colonic tissues than WT mice. AMD3100 strongly abrogated the AOM/DSS-induced the infiltration of these immunocytes in colonic tissues (Fig. 4d). The massive infiltration of macrophages (F4/80+) in colonic tissues of AOM/DSS-induced CXCR4+/− mice was also evaluated by IHC staining as presented in Fig. 4e. These results suggest that CXCR4 is responsible for the recruitment of macrophage and MDSCs from circulatory system to inflamed colonic mucosa and tumors. MDSCs have been shown to play an important role in immune evasion via suppressing the activation and proliferation of T cells and NK cells . The recruited MDSCs trigger tumor progression primarily via suppressing CD8+ T-cell cytotoxicity against tumor cells. Since AOM/DSS-induced massive infiltration of MDSCs in colonic tissue, thus, it was conceivable that a lower percentage of CD8+ T-cells were observed in CAC of CXCR4+/− mice compared with WT mice by flow cytrometry and IHC analysis (Fig. 4c, e).
Immune suppressive MDSCs and macrophages are key participants of inflammatory tumor microenvironment where they can promote tumor initiation and metastasis. To determine the effect of CXCR4 on inflammatory microenvironment in Apcmin/+ mice, flow cytometry was employed to examine MDSCs and macrophages in blood and colonic tissues of CXCR4+/−Apcmin/+ and Apcmin/+ mice. As presented in Additional file 4: Figure S3A, robust increased numbers of M-MDSCs (CD11b+Ly6C+) and macrophages (CD11b+F4/80+) and diminished numbers of CD8+ T-cells were observed in the blood of CXCR4+/−Apcmin/+ mice compared with Apcmin/+ mice. Notably, the significant increase of MDSCs and macrophages in blood was potentiated by DSS. Importantly, there were greater infiltration of both MDSCs and macrophages in colonic mucosa of CXCR4+/−Apcmin/+ than Apcmin/+ mice, which was strengthened by DSS (Additional file 4: Figure S3B). Expectedly, the percentage of CD8+ T-cells were significantly reduced in CXCR4+/−Apcmin/+ mice compared with Apcmin/+ mice in colonic cancer tissues by IHC staining (Additional file 4: Figure S3C). Collectively, these findings indicated that overexpression of CXCR4 promoted colitis-associated tumorigenesis and progression via recruiting macrophages and MDSCs from the circulatory system to inflamed colonic mucosa.
The downregulation of miR-133a-3p by activation of CXCL12/CXCR4 plays a crucial role in tumor invasion by targeting RhoA
CXCL12/CXCR4-induced lncRNA XIST functions as a ceRNA regulating RhoA expression by sponging miR-133a-3p
To further confirm the direct binding between lncRNA XIST and miR-133a-3p, we constructed lncRNA XIST luciferase reporter plasmids with miR-133a-3p binding site (lncRNA XIST-WT) and its mutant type (lncRNA XIST-MUT). Cotransfection of lncRNA XIST-WT with miR-133a-3p significantly reduced the luciferase activity compared with miR-NC, whereas co-transfection of lncRNA XIST-MUT with miR-133a-3p did not show significant change in luciferase activity (Fig. 6f). miRNA is known to be present in the cytoplasm in the form of miRNA ribonucleoprotein complex (miRNP) containing argonaute-2 (AGO2), the core component of RISC . To determine if lncRNA XIST is binding with miRNP, RIP experiment was conducted by using anti-AGO2 antibody. LncRNA XIST and miR-133a-3p were significantly enriched in AGO2-containing immunoprecipitates compared with control immunoglobulin G (IgG) immunoprecipitates. Successful immunoprecipitation of AGO2 was confirmed by Western blot (Fig. 6g).
CXCR4 enhanced CRC invasion by regulating cytoskeletal reorganization through activation of RhoA
ROCK inactivates myosin phosphatase by direct phosphorylation and activation of myosin light chain (MLC). As a consequence, RhoA/ROCK/p-MLC signaling enhances actomyosin contractility driving cell motility. To prove that CXCR4 plays a crucial role in colorectal cancer invasion through activation of RhoA/ROCK signaling in vivo, Western blot was performed to examine the expression of RhoA, p-FAK/FAK, ROCK1 as well as p-MLC/MLC in colonic tissues of AOM/DSS-induced CXCR4+/− mice and Apcmin/+ mice. The expression of RhoA, ROCK1 were markedly increased and p-FAK, p-MLC were significantly enhanced in CXCR4+/− mice, which can be attenuated by AMD3100. Consistently, CXCR4 enhanced the expression of RhoA, ROCK1 and phosphorylation of FAK, MLC in Apcmin/+ mice, particularly in response to DSS treatment (Fig. 7c, d). Importantly, the expression of miR-133a-3p was substantially reduced in the colonic mucosa of CXCR4+/− mice compared with that of WT mice (Fig. 7e, f). Further functional experiments of miR-133a demonstrated that overexpression of miR-133a suppressed EMT and cell invasion in CRC (Additional file 5: Figure S4). These results indicated that CXCR4 plays a critical role in the activation of RhoA-ROCK/p-MLC signaling that was responsible for actin polymerization and CRC cell invasion through downregulation of miR-133a-3p.
Chronic inflammation and inflammatory mediators promote tumor initiation, invasion and metastasis. MDSC and macrophages are important inflammatory drivers of immune suppression and established cross-talk impacting anti-tumor immunity in most tumors . MDSCs potentially inhibit T cell function and activation that correlates well with clinical cancer stage and metastatic tumor burden in CRC patients . Tumour associated macrophages (TAMs) have been reported to play a key role in the progression of CAC by secreting vascular endothelial growth factor (VEGF) and are linked with poor prognosis . Using villin-CXCR4 transgenic mice, we provide strong evidence that overexpression of CXCR4 enhanced the recruitment of MDSCs especially G-MDSCs and macrophages in the colonic epithelial mucosa to trigger the progression of CAC. Consistently, it was reported that CXCL12/CXCR4 axis triggered PGE2-induced accumulation of MDSCs in ovarian cancer microenvironment . Multiple myeloma cells recruit tumor-supportive macrophages through the CXCL12/CXCR4 axis and promote their polarization toward M2 phenotype . Our results unveiled a mechanism of the recruitment of MDSCs and TAM in colonic epithelial mucosa which contributed to the progression of CAC.
To explore the potential miRNAs and related lncRNAs that play an important role in CXCL12/CXCR4-induced invasion and metastasis of CRC, we screened the differentially expressed miRNAs upon the activation of CXCL12/CXCR4 axis in CRC cell models. Notably, miR-133a-3p was found to be significantly reduced in HCT116 cells overexpressing CXCR4. Increasing evidences demonstrated that CXCL12/CXCR4 promoted EMT and tumor invasion in CRC [4, 35], miR-133a-3p was markedly downregulated in a variety of cancers and associated with tumorigenesis and metastasis [36, 37]. We firstly demonstrated that RhoA was the direct target gene of miR-133a-3p. More importantly, the negative correlation of RhoA and miR-133a-3p was observed in human CRC specimens. Therefore, it is likely that the downregulated miR-133a-3p upon the activation of CXCL12/CXCR4 axis promoted the invasion and metastasis of CRC by regulation of target gene RhoA.
In addition, we investigated the mechanism of downregulation of miR-133a-3p by the activation of CXCL12/CXCR4 axis. LncRNAs were recently reported to function as ceRNAs that could sponge miRNAs. We predict that lncRNA XIST is the potential ceRNA of miR-133a-3p and further confirmed that lncRNA XIST regulated the expression of RhoA through sponging miR-133a-3p determined by the assays of luciferase activity and RIP. Lnc RNA XIST has been reported to exert the oncogenic functions in many cancers by acting as a molecular sponge of miR-34, miR-101 and miR-181a [15, 38, 39]. We unveiled a novel mechanism of downregulation and repression of miR-133a-3p by CXCL12/CXCR4-induced lnc RNA XIST.
Tumor invasion is a complex and multistep program involved in the interplay of tumor cells and the microenvironment, and in turn tumor cells acquire the capacity of migration and invasion. RhoA signaling is engaged in at least two distinct types of motility in three-dimensional matrix, amoeboid motility and mesenchymal motility . RhoA regulates cell motility through cytoskeletal reorganization, and has been reported to be dysregulated in CRC [41, 42]. GTP binding and hydrolysis switches RhoA between a GTP-bound active and a GDP-bound inactive state. ROCK is a principal mediator of cytoskeletal tension downstream of RhoA. ROCK enhances actomyosin contractility by direct phosphorylation and activation of myosin light chain (MLC) . Compared with WT mice, the upregulation of RhoA, ROCK1 and p-MLC in CXCR4+/− transgenic mice induced by AOM/DSS could be impaired by AMD3100. These results demonstrated that RhoA/ROCK/p-MLC signaling played an important role in the cytoskeletal assembly and migration of CAC in CXCR4+/− transgenic mice. Accumulating evidences have indicated a prominent role of RhoA/ROCK signaling in the progression and metastasis of multiple tumors [44, 45]. To confirm that activation of CXCL12/CXCR4 is truly involved in the activation of RhoA, active form of GTP-RhoA was found to be regulated by the activation of CXCL12/CXCR4 axis in colorectal cancer cell model. Importantly, CXCL12/CXCR4-induced actin polymerization and reorganization was driven by the active RhoA, which can be blocked by ROCK inhibitor Y27632.
Taken together, our results showed that overexpressing CXCR4 in the intestine epithelial mucosa accelerates the tumorigenic process in inflammation-driven tumorigenesis, resulting in massive infiltration of inflammatory MDSCs and macrophages and increased tumor size and multiplicity. To further explore the underlying mechanism, miRNA sequencing results showed that miR-133a-3p was significantly downregulated in response to CXCL12/CXCR4 axis. We then used in vitro models to demonstrate that in CRC cells, miR-133a-3p directly targeted the regulation of RhoA that triggered tumor motility/invasion. More importantly, lncRNA XIST function as a ceRNA that sponged miR-133a-3p and liberated the repression of RhoA by miR-133a-3p. The regulation of lncRNA XIST/ miR-133a-3p/RhoA/ROCK/P-MLC signaling pathway was further supported in both human CRC specimens and CXCR4+/− transgenic mice models.
Our results highlight the important role of CXCR4 in promoting inflammation-driven colorectal cancer progression through activation of RhoA/ROCK signaling by lncRNA XIST mediated sponging of miR-133a. These results shed a new insight into the role of miR-133a-3p in CXCL12/CXCR4-induced invasion and metastasis of CRC, which provides potential therapeutic targets for hindering the metastatic behavior.
The authors would like to thank Jingjing Wang (Center for Laboratory animals of Capital Medical University) for her great assistance with the animal experiments.
This study was supported by grants from the National Natural Science Foundation (81201731, 91629303), Beijing Natural Science Foundation (7162024) and Basic and Clinical grants of Capital Medical University (16 JL56).
Availability of data and materials
Please contact the corresponding author for all data requests.
XY designed the research. XY, DW, XW, SS, YZ, SW, RM, XX performed the experiments and analyzed data. XY and XQ drafted the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
This study was approved by the Ethics Committee of Capital Medical University. Informed consent was obtained from all the patients.
Consent for publication
We have obtained consents to publish this paper from all the participants of this study.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 31.Diaz-Montero CM, Salem ML, Nishimura MI, Garrett-Mayer E, Cole DJ, Montero AJ. Increased circulating myeloid-derived suppressor cells correlate with clinical cancer stage, metastatic tumor burden, and doxorubicin-cyclophosphamide chemotherapy. Cancer Immunol Immunother. 2009;58:49–59.CrossRefGoogle Scholar
- 34.Beider K, Bitner H, Leiba M, Gutwein O, Koren-Michowitz M, Ostrovsky O, Abraham M, Wald H, Galun E, Peled A, Nagler A. Multiple myeloma cells recruit tumor-supportive macrophages through the CXCR4/CXCL12 axis and promote their polarization toward the M2 phenotype. Oncotarget. 2014;5:11283–96.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.