Metadherin Mediates Lipopolysaccharide-Induced Migration and Invasion of Breast Cancer Cells

Background Breast cancer is the most prevalent cancer in women worldwide and metastatic breast cancer has very poor prognosis. Inflammation has been implicated in migration and metastasis of breast cancer, although the exact molecular mechanism remains elusive. Principal Findings We show that the pro-inflammatory endotoxin Lipopolysaccharide (LPS) upregulates the expression of Metadherin (MTDH), a recently identified oncogene, in a number of breast cancer lines. Stable knockdown of MTDH by shRNA in human breast MDA-MB-231 cells abolishes LPS-induced cell migration and invasion as determined by several in vitro assays. In addition, knockdown of MTDH diminishes Nuclear Factor-kappa B (NF-κB) activation by LPS and inhibited LPS-induced IL-8 and MMP-9 production. Conclusions These results strongly suggest that MTDH is a pivotal molecule in inflammation-mediated tumor metastasis. Since NF-κB, IL-8 and MMP-9 play roles in LPS-induced invasion or metastasis, the mechanism of MTDH-promoted invasion and metastasis may be through the activation of NF-κB, IL-8 and MMP-9, also suggesting a role of MTDH in regulating both inflammatory responses and inflammation-associated tumor invasion. These findings indicate that MTDH is involved in inflammation-induced tumor progression, and support that MTDH targeting therapy may hold promising prospects in treating breast cancer.


Introduction
Breast cancer is the most prevalent cancer in women worldwide and metastatic breast cancer has very poor prognosis. Although breast cancer incidence rate has been decreasing in the past decades due to the early detection, breast cancer remained to be of the top incidence rate and it has the second highest cancer mortality in women [1]. Breast cancer is a heterogeneous disease and is stratified by race, stage, grade, and estrogen (ER)/ progesterone (PR) receptor status. Typically, there are two broad categories of genetic changes in the process of tumorigenesis: tumor suppressor genes and oncogenes. Tumor suppressor genes, including BRCA2, inhibit cell division, survival, or other properties of cells. They are often disabled in cancer cells therefore to promote the malignant changes. Oncogenes promote malignancy by expressing at inappropriately high levels, or being altered to have novel properties [2]. Metadherin (MTDH) is a recently identified oncogene [3]. Here we report the role of MTDH in promoting invasion and metastasis in breast cancers.
Current studies have revealed that MTDH could be a prognostic factor in breast cancer: its high expression is associated with poor survival [14]. Statistical analysis showed a significant correlation of MTDH expression with the clinical staging of the patients, tumor classification, node classification, and metastasis classification. Previous studies from our group have also shown a significant correlation between MTDH expression with patients' age, ER status and p53 status that are also poor prognostic features, further supporting the notion that MTDH expression is correlated with poor prognosis and high morbidity in breast cancer patients [15]. Previously, MTDH has been shown to induce the lung metastasis with a ''lung-homing domain'' selected from lung-homing Balb/c-derived 4T1 mammary tumor cell line phage cDNA library and has been related to tumor angiogenesis with the expression of vascular endothelial growth factor (VEGF) and microvessel density (MVD) [16,17]. Our group also demonstrated the role of MTDH in promoting metastatic seeding and enhancing chemoresistance [4]. Recently we found that MTDH enhanced EMT which drove the aggressive behavior of the breast cancer and identified novel SNPs of MTDH that are correlated to breast cancer susceptibility [12,18]. Recently the MTDH/AEG-1-based DNA vaccine was shown to increase chemosensitivity to doxorubicin in inhibiting breast cancer metastasis to the lung [19]. These studies suggested MTDH as a potential candidate of target therapy for cancer, especially for enhancing the efficacy of chemotherapy and reducing metastasis.
About 150 years ago, Virchow first discovered the relationship between chronic inflammation and carcinogenesis [20], which attracted extensive studies. The last decade witnessed increased interest and extensive studies on tumor microenvironment that contributes to neoplastic process, cell proliferation and migration [21]. Among those factors such as the recruitment of microenvironmental cells and the cytokines those cell secreted into the tumor microenvironment, the activation of TLR4 was considered as a two edged sword, which has both the anti-tumor and pro-tumor functions [22]. Lipopolysaccharide (LPS), the major component of the outer membrane of Gram-negative bacteria, also the ligand of TLR4, could up-regulate the expression of MTDH in human promonocytic cell line which takes part in the regulation of the TLR4 signaling pathway, including the activation of Nuclear Factor-kappa B (NF-kB). These observations suggest that MTDH might play an important role in the regulation of innate immunity [11]. NF-kB can promote inflammation-associated tumerigenesis and is related to the MTDH-associated tumor progression and metastasis in cervical cancers [23,24]. It remains elusive whether MTDH plays a causative role in LPS-induced tumor progression and metastasis. Here we determined the function of MTDH in promoting tumor progression in response to LPS and investigated how LPS regulated MTDH expression in breast cancer cells.

LPS induced the upregulation of MTDH in the TLR4positive breast cancer cells
First, we examined which breast cancer cell lines were TLR4 positive. We evaluated the expression of TLR4 at mRNA level by Real Time RT-PCR. TLR4 expression was detected in three breast cell lines including MDA-MB-231, MCF-7 and MDA-MB-468. TLR4 was undetectable in the cell line T47D, which was used as negative control in our study. The mRNA level of TLR4 was higher in MDA-MB-231 cells than MCF-7 and MDA-MB-468 ( Figure 1A). Although the differences of TLR4 protein levels were not so obvious perhaps because of translation control, the expression levels of TLR4 protein in these four cells still showed similar pattern as determined by the Western-blot assays ( Figure 1B). Next, we examined the effects of LPS on the expression of MTDH in these cells. As shown in Figure 1C, the expression of MTDH protein changed in a concentrationdependent manner, the peak appeared in the three TLR4 positive cells (MDA-MB-231, MCF-7, and MDA-MB-468) when cells were stimulated with LPS at the concentration of 100 ng/ml. In the TLR4 negative T47D cell line, no significant change of MTDH expression was observed when the cells were treated with LPS. The basal total p65 expression was not significantly different among the four types of cell lines as shown in Figure 1B, suggesting that the different reactions to the LPS stimulation were not due to the lack of basal NF-kB expression. Since MTDH expression level peaked at a concentration of LPS 100 ng/ml, in the following experiments we selected MDA-MB-231 cells as the representative TLR4 positive cells and T47D as TLR4 negative cells treated with 100 ng/ml LPS. We evaluated MTDH expression at mRNA and protein levels at different time points in MDA-MB-231 cells which show clear induction of MTDH by LPS ( Figure 2). The results suggested that LPS upregulated the MTDH expression in timedependent and concentration-dependent manners in TLR4 positive breast cancer cells. There were reports that MTDH was activated over a long period through PI3K/AKT while it was activated within 30 min through the NF-kB signal pathway [11].

LPS induced up-regulation of MTDH and increased the migration ability of MDA-MB-231 cells
Since MTDH is considered to be involved in tumor migration and invasion [13], and the LPS induced inflammation response upregulates the MTDH expression in breast cancer cells, we carried out the in vitro scratch assays to investigate whether elevated MTDH levels are responsible for the enhanced tumor migration. We knocked down the MTDH expression by shRNA against MTDH. The decrease of MTDH was confirmed at both protein and mRNA levels ( Figure 3A). Knockdown of MTDH in MDA-MB-231 cell significantly reduced the migration ability. As shown in Figure

LPS induced MTDH expression regulated the TLR4 signaling pathways
The TLR4 signaling pathway includes MyD88 (myeloid differentiation primary response gene 88)-dependent and MyD88-independent pathways [25]. It has been reported that the early-phase NF-kB activation by LPS was dependent on MyD88, whereas late-phase NF-kB activation was dependent on TRIF (TIR domain-containing adaptor inducing IFN-b,mediating MyD88-independent signaling), and both phases of NF-kB activation were required for cytokine production [26]. We evaluated the MYD88, IRAK4 (IL-1 receptor-associated kinase 4) expressions which were in the MYD88-dependent pathway and TRIF which was involved in MYD88-independent pathway by Western-blot assays. The expression levels of TLR4, IRAK4, TRIF and P-p65 were decreased in MTDH knockdown cells ( Figure 5A). As shown in Figure 5A, all the protein mentioned above in the TLR4 signaling pathway were upregulated when stimulated by LPS, and in the MTDH knockdown cells the TLR4 and TRIF were slightly upregulated meanwhile MyD88, IRAK4 expression changes were not obvious. Those results implied that LPS activated MTDH in the TLR4 signaling pathway and the depletion of MTDH turned out to further impede this pathway.
The NF-kB family (also known as the Rel/NF-kB family) of transcription factors consists of five members: p50, p52, RelA (also known as p65), RelB and c-Rel. With the activation of NF-kB, the p65 was phosphorylated; the p50-p65 NF-kB translocated into the nucleus, which in turn bound to the specific sequence and augmented related downstream gene transcription. LPS treatment in prp-control cells increased the P-p65 level ( Figure 5A), suggesting that NF-kB was activated by LPS-induced upregulation of MTDH, which is consistent with the previous study that MTDH activates the NF-kB pathway [27]. In the MTDH knockdown cells, the P-p65 was barely detectable with or without the stimulation of LPS, providing evidence that MTDH mediates the activation of NF-kB by LPS.
Cytokines production changes detected with the stimulation of LPS

Discussion
Cancer-related inflammation is an important hallmark of cancer [28,29]. Inflammation induces the genetic instability in the tumor microenvironment [30]. Two pathways have been identified linking inflammation and cancer: the extrinsic ones and intrinsic ones [31]. In the extrinsic pathway, inflammatory environment promotes the malignancy of cancer. For example, the Helicobacter pylori infection increases the risk of gastric cancer and hepatitis virus infection is found in majority of liver cancers. Here, our study provided the evidence that LPS could upregulate the expression of MTDH which increases the invasiveness of breast cancer. In the intrinsic pathway, genetic changes lead to the expression of inflammation-related programs that guide the construction of an inflammatory microenvironment. The production of cytokines such as IL-8 and MMP-9 was increased, which could further help inducing angiogenesis and modifying the extracellular matrix in the tumor microenvironment. In our present study, we found that MTDH is a molecule linking inflammation and cancer.
Toll-like receptors (TLRs) are a family of pattern recognition receptors that are best-known for their role in initiation of the innate and adaptive immune responses. However, recent evidences showed that functional TLRs are also expressed in a wide variety of tumors suggesting that TLRs may play important roles in tumor biology. An earlier report revealed that TLR4 was expressed on many tumor cell surfaces and associated with the risk of prostate cancer, nasopharyngeal carcinoma and gastric carcinoma [22]. LPS, the ligand of TLR4, was reported to increase the lung metastasis in 4T1 mouse mammary carcinoma cells [32]. Furthermore, a recent study showed that samples of breast carcinomas with recurrence exhibit a significant increase in the mRNA levels of TLR4 and that TLR4 expression is significantly associated with a great rate of distant metastasis [33]. Also the activation of NF-kB, which is the downstream gene of TLR4, contributed to mammary epithelial tumor progression in vivo. Inhibition of NF-kB led to the decreased tumor burden and decreased numbers of lung metastases [34]. All these results provide a possibility that TLR4-NF-kB pathway is a connection between inflammation, immunity and cancer.
In our study, TLR4 expressed in breast cancer cell line MDA-MB-231, MCF-7 and MDA-MB-468. We demonstrated the activation of NF-kB with the stimulation of LPS in accordance with the overexpression of MTDH. Previously, MTDH was reported as a LPS-responded gene in proinflammatory cells which suggested that MDTH may play an important role in the regulation of innate immunity [11]. Moreover, MDTH is also involved in tumor progression, migration and invasion as a NF-kB dependent gene in Hela cells [24,27]. It functioned as a bridging molecule between NF-kB, CBP (cyclic AMP-responsive element binding protein-binding protein) and the basal transcription machinery. The present study demonstrated for the first time that LPS could increase the MTDH expression via activating the TLR4-NF-kB signaling pathway in human breast cancer cells, we hypothesized that there might be a positive feedback loop that LPS stimulation initiated a signaling cascade leading to upregulated expression of MTDH, which subsequently activated NF-kB, resulting in a second phase of TLR4 activation or TRIF activation, reinforcing and amplifying LPS-induced signaling. Our results provided evidence that MTDH correlated the inflammation, innate immunity and carcinogenesis together.
It has been reported that the LPS-induced metastatic growth was associated with increased angiogenesis vascular permeability and tumor cell invasion [35]. In this study we have shown the increased secretion of IL-8 and MMP-9 by LPS treatment. IL-8 is believed to function as a chemoattractant and a potent angiogenic factor. Furthermore, IL-8 can promote cell invasion and migration, and the capacity of IL-8 to induce tumor-associated macrophages to secrete additional growth factors will further increase the rate of cell proliferation and cancer cell invasion at the tumor site [36]. MMP-9, a 92 kDa type IV collagenase, 92 kDa gelatinase or gelatinase B (GELB), is a key effecter of ECM remodeling. MMP-9 is activated to degrade type IV and V collagens and promote the process of metastasis. In the present study, LPS upregulated the MTDH expression and further activated the NF-kB pathway, promoting the secretion of the downstream genes such as IL-8 and MMP-9 in prp-control-231 cells. With the knockdown of MTDH, the induction of the mRNA expression of IL-8 and MMP-9 decreased profoundly, and LPS treatment did not increase their secretion. This indicated that MTDH increased the metastatic ability of breast cancer via the secretion of these cytokines. Besides, there was study put forward the hypothesis that signaling through TLR4 is strongly inhibited by intact extracellular matrix and that inhibition is abrogated and endogenous agonist(s) are liberated when the matrix is degraded [37]. It is possible that with upregulation of MTDH induced by LPS, the secretion of cytokines such as MMPs increased greatly and could degrade the extracellular matrix which then further activated the TLR4 and enhanced the invasive ability of cancer cells.
In summary, induction of MTDH by LPS could increase the migration and invasiveness in the TLR4 positive breast cancer cells. To date, we are the first to show the role of MTDH in the TLR4 signaling pathway in breast cancer cell line, MDA-MB-231 cells. The results provide evidence that MTDH might be a pivotal molecule not only in the progression of cancers but also in relation of cancer, innate immunity and inflammation. However, the immune evasion possibly mediated by TLR4-NF-kB-MTDH pathway should be further studied. The relationship between MTDH and TLR4 signaling pathway might lead to the development of TLR4 and MTDH as novel therapeutic targets for breast cancer.  Western blot analysis Cells were washed three times with cold phosphate-buffer saline (PBS) and lysed on ice in RIPA (16PBS, 1% NP40, 0.1% sodium dodecyl sulfate (SDS), 5 mM EDTA, 0.5% sodium deoxycholate and 1 mM sodium orthovanadate) with protease inhibitors. The concentrations of protein were determined by BCA methods. Eighty micrograms of protein were separated by 10% SDS-PAGE and electro-blotted to a PVDF membrane using a semi-dry blotting apparatus (Bio-Rad, Hercules, CA, USA). After blocked in 5% non-fat milk, the membranes were incubated overnight at 4uC with the primary antibodies. Then membranes were incubated in the secondary antibodies for 2 hours at room temperature with slightly shock. The bands are visualized using a Pro-lighting HRP agent. b-actin was used as a loading control.

Real-Time RT-PCR
Total RNA was extracted with TRIZOL reagents according to the manufacturer's protocol (invitrogen, Carlsbad, CA, USA). The cDNA was synthesized from 800 ng of total RNA by PrimerScript RT Reagent Kit (TaKaRa BIO Inc, Otsu, Shiga, Japan). The real-time RT-PCR was carried out by using a SYBR green PCR mix in Applied Biosystems StepOne and StepOnePlus Real-Time PCR Systems. The gene expression values were calculated by normalizing with endogenous control b-actin with DataAssist software. Each experiment was repeated three times to confirm the conclusions.

Plasmids construction and Cell transfection
The plasmids used in the experiment were described previously (Hu, Chong et al. 2009). Briefly, two different short-hairpins RNA (shRNA) were constructed to knockdown the expression of MTDH in MDA-MB-231 cell line, generating the psuper-retropuro-MTDH (prp-MTDH) cells, psuper-retro-puro vector was used as control (prp-control). Cells were seeded in six-well plates 24 hours before the transfection and were transfected with psuperretro-puro-MTDH or psuper-retro-puro vectors by Lipofectamine 2000 transfection reagent (Invitrogen, USA) according to the manufacturer's instructions. Cells were selected for 2 weeks in 0.5 ug/ml puromycin and maintained in 0.2 ug/ml puromycin. The expression levels of MTDH in surviving colonies were evaluated by western blot and RT-PCR.

In vitro scratch assays
The cells were seeded in 24-well plates at a density 1610 5 cells/ well. After incubation overnight, cells were stimulated with LPS (100 ng/ml) for 4 hours, and then scraped by a p20 pipette tips to create a straight-line cell-free scratch. Each well was washed with PBS three times to remove the remaining unattached cells and debris. The scratch area were marked and photographed in every 12 hours. The distances were measured by the software imageJ and cell motility was quantified by measuring the distance between the migrating cell boundaries [38]. Data were analyzed statistically.

Invasion assays
The transwell system (24 wells, 8 um pore size, BD Biosciences, Heidelberg, Germany) coated with 2 mg/ml basement membrane Matrigel (BD Biosciences) were used for the in vitro invasion assays. 1610 5 cells were suspended in serum-free DMEM in the upper chamber of each well. The lower well of each chamber was filled with 750 ul DMEM supplemented with 20% FBS. LPS (100 ng/ml) was added to both the upper and the lower chambers. After 18 hours of LPS treatment, filters were fixed with methanol and stained with 0.1% crystal violet. Then the number of cells in at least five randomly selected microscope fields were counted and analyzed statistically.

Statistical evaluation
Statistical differences were analyzed by Student's t-test, p,0.05 was considered significant. Independent repeated experiments were carried out at least three times.