|Home | About | Journals | Submit | Contact Us | Français|
Neurotensin receptor-1 (NTSR1) is a G-protein coupled receptor that has been recently identified as a mediator of tumorigenicity and metastasis. NTSR1, as well as its endogenous ligand, neurotensin (NTS), are co-expressed in several breast cancer cell lines and breast cancer tumor samples but not in normal breast tissue. We have previously published that ceramide mimetics could inhibit breast cancer growth in vitro and in vivo. Thus, understanding the biochemical and biophysical regulation of NTSR1 by ceramide can help further define NTSR1 as a novel target in breast cancer. Our results demonstrate that nano-liposomal formulations of ceramide inhibit NTSR1-mediated MDA-MB-231 breast cancer progression (mitogenesis, migration and MMP-9 activity). In addition, liposomal ceramide inhibited NTSR1-mediated, but not PMA-mediated activation of the MAP kinase pathway. Mechanistically, nano-liposomal short chain ceramide reduces NTSR1 interaction with Gαq/11 subunits within structured membrane microdomains (SMDs), consistent with diminished NTS-induced translocation of NTSR1 into SMDs. Collectively, our findings suggest that exogenous short chain ceramide has the potential to be used as an adjuvant therapy to inhibit NTS-dependent breast cancer progression.
Neurotensin (NTS), originally characterized as a tridecapeptide neurotransmitter (1) or gastrointestinal hormone and mitogen (2, 3), has recently been demonstrated to be expressed in metastatic and drug resistant cancers (4, 5). The G-protein coupled neurotensin receptor, NTSR1, is also overexpressed in many types of cancer (6), including breast cancer (7). 91% of invasive ductal breast tumors were positive for NTSR1 (7). In fact, NTSR1 and NTS were coexpressed in 30% of these ductal breast tumors (7). In normal mammary tissue, NTSR1 and NTS are not expressed (7). NTSR1 activation has been observed in both ER-positive and ER- negative breast cancer cells, stimulating BCl-2 (8) and matrix metalloprotease-9, respectively (7). In addition, siRNA knockdown of NTSR1 in xenografted MDA-MB-231 cells resulted in a 70% fold decrease in tumor growth compared with wild type cells (7). Taken together, these studies identify NTSR1 as a key target in breast cancer.
Structured membrane microdomains (SMDs), also known as lipid rafts, are regions in the membrane enriched in cholesterol, glycolipids and sphingomyelin (9). Enrichment of SMDs with polyunsaturated fatty acids (PUFA) (10), sphingolipid or glycosphingolipid metabolites, such as ceramide (11, 12) alter the lipid microenvironment in a way that may interfere with receptor-mediated signaling within microdomains. In fact, the structure of SMDs is highly unstable as these lipid metabolites are dynamically and temporarily formed and degraded (13). Often, these structured membrane microdomains contain scaffolding elements, such as caveolin-1, which interacts with promitogenic signaling elements including glycophosphatidyl-anchored proteins, acylated proteins, G protein-coupled receptors (GPCRs), trimeric and small G-proteins and their effectors (14). We, and others, have previously demonstrated that short chain ceramide localizes to SMDs and causes an increase in the phosphorylation and localization of PKCzeta within caveolin-enriched microdomains to inactivate promitogenic and prosurvival Akt (15, 16). As nano-scale ceramide formulations have previously been shown to selectively inhibit breast and ovarian cancer growth in vitro and in vivo (17-19), we now investigate if exogenous nanosized formulations of liposomal C6 ceramide alters NTSR1-induced prosurvival, promitogenic signal transduction cascades in breast cancer cells.
Dioleoyl phosphatidylethanolamine (DOPE), dioleoyl phosphatidylcholine (DOPC), d-erythro-hexanoyl-sphingosine (C6-ceramide), polyethyleneglycol-450-C8-ceramide (PEG-C8) were purchased from Avanti Polar Lipids (Alabaster, AL). Di-hydro-erythro-hexanoyl-sphingosine (DHC6) was purchased from BIOMOL Research Laboratories (Plymouth Meeting, PA). Polyacrylamide gel electrophoresis gradient gels and gelatin containing gels were purchased from Invitrogen (Carlsbad, CA), and enhanced chemiluminescence reagent was purchased from Thermo Fisher Scientific Inc. Human MDA-MB-231 breast adenocarcinoma cells and HEK-293T were obtained from American Type Culture Collection (Manassas, VA) and grown at 37°C in RPMI 1640 medium and DMEM, respectively, supplemented with 10% fetal bovine serum (FBS). The MDA-MB-231 cell line is a highly aggressive metastatic, estrogen receptor-negative, EGFR positive and NTSR1 positive model of human breast cancer. BrdU cell proliferation ELISA colorimetric kit was obtained from Roche (Indiana). NTS was purchased from Sigma-Aldrich and was dissolved in 0.05% acetic acid just before use in experiments. JMV-449, a synthetic, stable NTS analogue was purchased from NeoMPS (Strasbourg, France). All antibodies were purchased from Santa Cruz Corp, CA. Peptide-N-glycosidase F (PNGaseF) was obtained from Sigma, Saint Louis, Missouri. Phorbol ester (PMA) was obtained from Promega Inc (Madison, WI). Human NTSR1 and HA-tagged cDNA was purchased from Missouri S&T cDNA Resource Center, Rolla, MO (Catalogue Number: NTSR100000).
Cells were grown in 10 mm plates until 50% confluency. Transfection was performed using Lipofectamine 2000 (Invitrogen) transfection agent according to the manufacturer recommendations. After 48 hours, the cells were harvested using RIPA buffer (100 mM Tris pH 8.0, 300 mM NaCl, 1 % v/v Nonidet P40,1% sodium deoxycholate, 0.2% SDS) and NTSR1 expression was assayed using western blot analysis.
To ensure specificity and reliability of a goat polyclonal IgG Anti-NTSR1 (sc25042, Santa Cruz, Inc.), we have utilized several experimental approaches. First, we purchased a positive control from Santa Cruz, IMR-32 cell lysate: sc-2409, which is a human neuroblastoma cell line. The antibody recognized a band in human IMR-32 cell lysate at an analogous molecular weight to human MDA-MB-231 breast cancer cells (Supplementary Fig. 1A), a molecular weight consistent with previous reports (20). To further confirm specificity, we overexpressed NTSR1 in HEK-293T cells which normally do not express this receptor (this was confirmed using RT-PCR, data not shown) using a non-HA-tagged construct. Western blot analysis of the lysates obtained from tranfected cells recognized bands at comparable molecular weight relative to MDA-MB-231 cells as well as IMR-32 cells, while no bands were detected in the non transfected cells (Supplementary Fig. 1B). Additionally, we repeated the same overexpression experiment using an HA-tagged NTSR1 construct (Supplementary Fig. 1C). Cell lyasates containing equal amount of total protein were immunoprecipitated using Anti-HA antibody. Band corresponding to molecular weight ~52−54 kDa was observed when the blots were probed with both Anti-HA and anti-NTSR1 antibodies. The selectivity of the NTSR1 antibody to the epitope that was used to generate the antibody was confirmed by a competition study using a blocking peptide, where the bands completely disappeared (data not shown). In all supplemental and manuscript data, we used a 1:500 dilution along with the SuperSignal West Femto Chemiluminescent Kit from Pierce for optimal signal detection. Therefore, together, the data confirm that the antibody was able to selectively and specifically detect NTSR1 of human origin.
Lipids, dissolved in chloroform (CHCl3), were combined according to the following molar ratios: 1,2-Dioleolyl-sn-Glycero-3-Phosphoethanolamine (DOPE): 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC): C8 mPEG 750 Ceramide: 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (Ammonium salt): C6-Ceramide or Dihydro-C6-ceramide (dC6-ceramide) (1.75: 3.75: 0.75: 0.75: 3). The biologically inactive dihydro-C6-ceramide was used as a negative control. Ceramide-free liposomes (ghost) containing similar concentrations and ratios of the other lipid components were also used as a negative lipid control. Lipid mixtures were dried under a stream of nitrogen above lipid transition temperatures, and hydrated with sterile phosphate-buffered saline (PBS). The resulting solution underwent sonication for 10 minutes followed by extrusion through 100-nm polycarbonate membranes. The amount of total lipids in C6 ceramide, dihydroceramide, and ghost liposomal formulations remained constant in all experiments. In all experiments, C6 ceramide was delivered in a non-toxic nano-liposomal formulation as previously described by us (18) as well as previously characterized by the National Cancer Institute Nanotechnology Characterization Laboratory (http://ncl.cancer.gov/MK_022207_073007.pdf). For short term signaling and sucrose gradient experiments, we used 10 μM liposomal C6 ceramide to maximize biochemical and biophysical effects. For “endpoint” analysis (migration, mitogenesis), we utilized 1μM liposomal C6 ceramide to role out any long term potential non-specific toxic effects of ceramide. Previous studies have demonstrated that the IC50 value for liposomal C6 ceramide is 5μM(18).
Highly invasive estrogen negative MDA-MB-231 cells were seeded at 2.0 × 105 cells/well in 96-well plates and grown overnight prior to 24-h serum starvation. Cells were treated with the indicated concentration of JMV-449 alone or in the presence of 1μM C6 ceramide liposomes, di-hydro-C6-ceramide liposomes or ghost liposomes. After 48 hours, cells were labeled with BrdU (20 μl/well) for 5 hours then BrdU incorporation into DNA was measured according to the manufacturer's protocol (Roche Applied Biosciences fluorescent ELISA kit).
MDA-MB-231 cells were seeded at 4.0 × 105 cells/well in 60-mm plates and grown overnight prior to 24-h serum starvation, In some experiments, it was necessary to starve cells for up to 3 days to downregulate basal ERK phosphorylation. Cells were then treated with liposomal C6 for 1 hour prior to 100 nM nt or PMA stimulation for 15 minutes. Cells were washed once with cold PBS followed by the addition of 150 μl of cold lysis buffer (1% Triton X-100, 20 mM Tris, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 2 mM Na4P2O7, 1 mM glycerolphosphate, 1 mM Na3VO4, and 1 μg/ml leupeptin in ddH2O, pH 7.5) on ice. Cells were lysed for 15 min on ice, and cell lysate was harvested and centrifuged at 15,000g for 15 min. Thirty-five micrograms of protein were loaded in 4 to 12% precasted SDS-polyacrylamide gel electrophoresis gradient gels and probed for pErk 1/2. Blots were stripped and re-probed for Erk 2 to demonstrate equal loading. Protein bands were visualized using enhanced chemiluminescence and quantified using Image-J software.
MDA-MB-231 cells were seeded at 4.0 × 105 cells/well in 100-mm plates and allowed to grow till 80% confluency. Cells were then treated with 100 nM NTS in the presence or absence of 1 μM C6 Ceramide, di-hydro-C6- Ceramide or ghost liposomes for 48 hours. Conditioned media were then collected and concentrated using Centriprep concentrators (YM-30). Aliquots of the concentrated media containing equal amounts of total proteins were then subjected to non-reducing SDS-gel elctrophoresis using gels that contain 1% gelatin (Invitrogen). The gels were then developed according to the manufacturer protocol.
MDA-MB-231 cells were seeded at 4.0 × 105 cells/well in 6 well plates marked with a line grid and allowed to grow until confluency was reached. The confluent cell layer were scratched using 200 μl pipette tip and the cells were treated with 100 nM NTS in the presence or absence of 1μM C6 ceramide in 0.1 % serum containing media. Photomicrographes were taken using a digital camera (Diagnostic instruments, inc.) installed on a microscope (Olympus), 10× magnification power. Image-J software was used to measure the wound width.
MDA-MB-231 cells were seeded at 4.0 × 105 cells/well in 100-mm plates and allowed to grow till 80% confluency. Cells were lysed using modified lysis buffer (50mM Tris-HCl, pH 7.5; 150 mM NaCl; 1mM EGTA; 10 mM MgCl2; 0.5 % Triton X-100; protease inhibitor cocktail). NTSR1 was immuno-precipitated as previously described (16). Immuno precipitated protein was subjected to deglycosylation using peptide-N-glycosydase (PNGase F) (Sigma) according to the manufacturer recommended protocol. The samples were then analysed using SDS-PAGE electrophoresis
We chose a detergent-free protocol based upon high reproducibility and lack of potential experimental artifacts. MDA-MB-231 cells were seeded at 4.0 × 105 cells/well in 100-mm plates and allowed to grow till 80% confluency. Cells were then treated with 10 μM C6 ceramide nano-liposomes or Ghost nano-liposomes (1hour) followed by 100 nM NTS (15 minutes) and then washed with ice-cold PBS twice. Sucrose gradient fractionation was performed according to Song et al. protocol (21). Due to the low concentration of NTSR1 receptor protein in the fractions, trichloroacetic (TCA) acid-acetone protein precipitation was required in order to be able to analyze the fractions using western blot analysis. In brief, 250 μl of freshly prepared TCAstock (100% w/v) was added to 1 ml sucrose gradient fraction and incubated on ice for at least 30 minutes. The samples were then microcentrifuged at 14K rpm for 5 minutes. The supernatants were then removed and the pellet washed twice with ice-cold acetone. After drying the pellets, the pellets were dissolved in 20 μl PBS and then 10 μl 3X SDS sample buffer was added followed by analysis using SDS-PAGE electrophoreses as described in western blot analysis.
We chose this procedure, which is performed at pH 7, in order to preserve protein-protein interactions. The high pH required for detergent-free fractionation can disrupt such interactions. This procedure was performed as described in the previous detergent-free fractionation method except that the cells were lysed using 1% Triton-X in MBS buffer (25 mM MES, pH 6.5, and 150 mM NaCl). Fractions 4, 5 and 6 that represent the caveolin-enriched fractions were separated and subjected to immuno-precipitation using anti Gαq/11 antibody. The immuno-precipitates were separated using SDS-PAGE electrophoresis and the blots were probed using anti-NTSR1 antibody. The blots were then developed as described previously in the western blot section.
MDA-MB-231 cells were grown in 6-well plates until 80% confluency and then starved for 24 −72 hours. To remove cholesterol, serum-starved cells were incubated in RPMI containing the indicated concentration of methyl-β-cyclodextrin (MβCD) (Sigma) for 1 h at 37°C. For cholesterol repletion, depleted cells were incubated with a solution containing 0.2 mM of preformed MβCD/cholesterol (10:1 mol/mol) complexes, similar to a previously described procedure (16). The cells were then stimulated with 1 μM nt or PMA for 15 minutes and then lysed using NP-40 lysis buffer. Total proteins were separated on SDS-PAGE and the immunoblots were probed for Phospho-ERK and ERK-2.
Differences among treatment groups were statistically analyzed using a two-tailed Student's t test for statistical analyses. Where appropriate, One-way analysis of variance with Bonferroni multiple comparison post-hoc test and t tests analysis were performed using GraphPad Prism 4.0 software. A statistically significant difference was reported if p < 0.05 or less. Data are reported as mean ± S.E. from at least n = 3 separate experiments.
Neurotensin (NTS) has been shown to induce cellular proliferation of several cancer cell lines including prostate cancer PC3 cells and breast cancer (4). We utilized a BrdU incorporation assay to initially assess the effect of NTS on DNA synthesis. Treatment of cells with increasing doses of JMV-449, a synthetic stable analogue of NTS, for 48 hours after a 24-hour serum starvation resulted in a dose-dependent increase in DNA synthesis (Fig. 1A). The observed increase in DNA synthesis after treatment with 1 and 10 μM NTS analogue was not statistically significant compared to 100 nM treatment indicating a level of saturation. Using the lowest saturating dose for NTS, 100 nM, we evaluated the effect of liposomal ceramide on NTS-mediated mitogenesis. Cells were serum-starved for 24 hours before being stimulated with 100 nM JMV-449 in the presence or absence of 1 μM liposomal C6 ceramide (Fig. 1B), a dose that was previously shown to inhibit breast cancer proliferation. NTS-induced BrdU incorporation was significantly reduced in the presence of liposomal C6 ceramide, but not liposomal di-hydro C6 ceramide. In addition, these liposomal formulations did not have any effect on basal level DNA synthesis under these experimental conditions (Fig. 1C). 10 % fetal bovine serum was used as a positive control and produced an effect consistent with that observed with NTS treatment.
It has been shown previously that NTS induces migration of breast cancer cells (7). Using a cellular scratch assay, 100 nM NTS significantly enhanced cellular migration compared to untreated cells (Fig 2A). Pretreatment of cells with liposomal C6 ceramide for 1 hour mitigated NTS-induced breast cancer migration compared with ceramide-free liposomes. Ceramide-free liposomal formulations (ghost liposomes) had no effect on cellular migration.
Previous studies have shown that NTSR1 stimulation induces the expression and activity of matrix metalloproteinase-9 (MMP-9), which has been implicated in tumor invasion of blood vessels (7). We used gelatin zymography to study the effect of C6-ceramide on MMP-9 activity. Treatment of cells with NTS enhanced MMP-9 gelatinolytic activity compared to untreated cells (Fig. 2B). Pretreatment of cells with liposomal C6-ceramide significantly reduced NTSR1 mediated activation of MMP-9 while liposomal di-hydro C6 ceramide (Fig. 2B) and ceramide-free liposomes (data not shown) had only a minimal non-significant effect. As a positive MMP-9 standard control, conditioned media from MCF7 cells showed a band at the same molecular weight (~92 kDa) as our zymographed samples as well as a recombinant standard (data not shown).
To assess the effect of NTS on the promitogenic MAP Kinase pathway, we performed a time response study and measured the increase in ERK1/2 phosphorylation using western blot analysis. After serum starvation for 24 hours, MDA-MB-231 cells were treated with 100 nM NTS for the indicated periods of time (Fig. 3A). Phospho ERK 1/2 was transiently activated reaching maximal activation after 15 minutes treatment. Based on this study, we chose 15 minutes as a time point to study the effect of liposomal C6 ceramide on NTS-mediated ERK 1/2 activation. Pretreatment of cells with 10 uM liposomal C6 ceramide inhibited NTS-induced ERK 1/2 activation (Fig. 3B). As an additional control, we also showed that the MEK inhibitor U0124 also reduced NTS-stimulated ERK stimulation. These results suggest that NTS effect on breast cancer cells is mediated, in part; through the activation of the mitogenic ERK 1/2 pathway and ceramide can inhibit this biochemical signaling cascade.
In order to determine the specificity of C6 ceramide-induced inhibition of NTSR1-mediated MAP Kinase activation, we tested the ability of C6 ceramide to inhibit PMA-induced activation of ERK phosphorylation. PMA was chosen in these studies due to its ability to directly activate downstream PKC isozymes, bypassing NTSR1, as well as its direct effectors (Gαq/11, and phospholipase C). MDA-MB-231 cells were serum starved before a 1 hour pretreatment with 10 μM C6 ceramide. Cells were then stimulated with 1 μM PMA for 15 minutes and lysates were assayed using western blot analysis. C6 ceramide did not alter basal ERK phosphorylation as well as PMA-induced ERK phosphorylation (Fig. 3C). This suggests that C6 ceramide treatment selectively disrupts NTSR1 interaction with signaling intermediates that lie upstream of PKC isozymes.
In order to examine the importance of SMDs for NTSR1-mediated mitogenic signaling, we used Methyl-β-cyclodextrin (MCD) to deplete cholesterol, a major constituent of SMDs. After 24 hour serum starvation, cells were pretreated with Methyl-β-cyclodextrin in the presence or absence of cholesterol for 30 minutes, before being stimulated with 100 nM NTS or 1 μM PMA. MCD treatment resulted in inhibition of ERK1/2 activation, while cholesterol repletion restored NTSR1 activity (Fig. 3D). These results suggest that NTSR1 requires intact SMDs to signal properly. In contrast, MCD did not disrupt PMA-induced ERK activation, alluding to the specificity of receptor-mediated, but not non-receptor-mediated signaling within intact SMDs. Exogenous C6-Ceramide Inhibits Agonist-Dependent Translocation of NTSR1 into Structured Membrane Microdomains
Given that NTSR1-signaling responses require functional SMDs, we next analyzed if there was a biophysical component to ceramide inhibited NTSR1 signaling. We initially examined if exogenous ceramide, which localizes within SMDs (16), could alter localization of NTSR1 within SMDs. MDA-MB-231 cells were serum starved for 24 hours and then treated with 10 μM C6 ceramide for 1 hour before 1 μM NTS stimulation for 15 minutes. Due to the low sensitivity of sucrose gradient fractionation experiments in systems where endogenously expressed low abundant protein membrane localization is under study, we have utilized a higher NTS treatment concentration to ensure a detectable response. The lysates were subjected to detergent-free sucrose gradient fractionation and the isolated fractions were subsequently analyzed by western blotting. Fractions 4 and 5 (Fig. 4A) are enriched with caveolin, a protein that is known to mainly localize within SMDs and is extensively used in the literature as a marker for these domains (22). Fractions 8 to 12 contain very low level of caveolin and represent the buoyant non-SMD fractions. Caveolin content in the SMD fractions, 4 &5, does not change upon any treatment (Fig. 4A). After western blot analysis, we found that both the non-N-glycosylated and N-glycosylated isoforms of NTSR1 localize mainly to fraction 8−12 outside the caveolin-enriched microdomains (Fig. 4A). Upon receptor stimulation, mature NTSR1 isoform translocates into caveolin-enriched SMDs (Fig. 4A). This movement is transient as the receptor either internalizes or moves out of the rafts after 20 minutes treatment (data not shown). Most strikingly, we observed an approximate 2 kDa increase in the receptor molecular weight upon entrance into SMD, fractions 4 and 5. The identity of the band of higher molecular weight (~54 kDa) was confirmed to be the N-glycosylated isoform as shown in Fig. 4B, where treatment with PNGase F casued a shift of the band to a level similar to that of the lower molecular weight band (~52 kDa). It is apparent that the NTSR1 antibody identified the more glycosylated NTSR1 isoform under these immunoprecipitation conditions. This may reflect that the protein is not heat-denatured, in contrast to regular western blot analysis where the protein is heat denatured. It is most likely that this change in the mobility of the protein upon enrichment into SMDs is due to receptor hyperglycosylation. Pretreatment of cells with liposomal C6 ceramide before NTS stimulation resulted in inhibition of NTS-dependent NTSR1 translocation into SMDs, while treatment of cells with C6 ceramide alone does not affect the receptor localization (Fig. 4A). The percentage of NTSR1 receptor that translocates into the SMDs fraction was quantitated relative to total NTSR1 in all fractions using ImageJ software and was found to be about 30% (Fig.4A, ,5B).5B). These results suggest that accumulation of C6 ceramide within structure membrane microdomains prevents NTS-induced translocation of NTSR1 into these caveolin-rich microdomains.
Several reports suggest that NTSR1 can couple to multiple G proteins including Gαq and Gαs (23). Additionally, Souaze et al. suggested the participation of Gαq/GEγ and PKC signaling pathways in NTS-dependent breast cancer progression (7). Therefore, we examined if NTS-induced NTSR1 translocation to SMDs regulates NTSR1/Gαq/11 interaction. We again performed detergent-free sucrose gradient fractionation and observed an approximate 37% increase in NTSR1 localization to SMD fractions upon NTS receptor stimulation, which was abrogated by ceramide pretreatment (Fig. 5A). Gαq/11 mainly localize to SMDs and receptor stimulation with 1 μM NTS did not alter membrane localization (Fig. 5A). We also observed that ERK-2 mainly localizes to the non-caveolar-rich domains, however, a fraction of ERK-2 exists within SMDs and could be sufficient for signal transduction upon NTSR1 stimulation. Thus, it is suggested that NTS-induced NTSR1 translocation within SMD may lead to enhanced interaction of NTSR1 with Gαq/11.
To further confirm the mechanistic importance of NTSR1 translocation into SMDs for NTS mitogenic signaling, we designed an experimental protocol that combines both detergent-based sucrose fractionation followed by co-immunoprecipitation of SMDs fractions. It was necessary to perform detergent-based sucrose gradient fractionation in order to maintain protein-protein interactions; the alkaline pH necessary for detergent-free fractionation can disrupt protein-protein interaction. The SMD fractions were separated and subjected to immuno-precipitation using anti-Gαq/11 antibody, followed by immuno-blotting using anti-NTSR1 antibody (Fig. 5B). The results show that Gαq/11 interacts with NTSR1 within SMDs only upon its stimulation with 1μM NTS for 15 minutes. It is important to note that the higher molecular weight, hyperglycosylated, NTSR1 is the major receptor population that interacts with Gαq/11. Pretreatment with 10 μM nano-liposomal ceramide for 1 hour before NTS stimulation disrupted this critical interaction (Fig. 5B). This result suggests that the receptor membrane translocation is essential for NTS coupling to Gαq/11 within SMDs.
G protein-coupled receptors (GPCRs) are very attractive targets for drug development (24). Recently, many GPCRs; such as neurotensin, endothelin, chemokine and lysophosphatidic acid receptors have been implicated in the tumorigenesis and metastasis of multiple human cancers (24). NTSR1 is overexpressed in several types of cancer and correlate with tumor metastasis and acquired resistance to conventional chemotherapeutic regimens (3). SMDs and caveolae have been shown to be involved in the regulation of various essential cell functions including fine-tuning of components of the cell signaling machinery (9). It is not well understood how GPCRs localize within SMDs (14). Some GPCRs localize exclusively into SMDs such as the gonadotrophin-releasing hormones receptor (GnRH) (25) and alpha1a-adrenergic receptor (26) while others are restricted from SMDs such as the oxytocin receptor (22, 25, 27). Several GPCRs such as the angiotensin II type 1 receptor translocate into SMDs only upon agonist stimulation (28). On the other hand, CB1 cannabinoid receptor translocates outside SMDs upon agonist binding (14). Such receptor translocations might be essential for the formation of an active receptor-signaling complex. Recently, reports from several labs, including ours (15, 16) have shown that ceramide tends to localize within SMDs, which serves to assemble large scale membrane domains and/or trigger signaling cascades (15, 16). Mehga and London showed that ceramide selectively displaces cholesterol from ordered lipid domains to form ceramide-rich SMDs (11). Yu et al. also supported this finding and additionally showed that such displacement of cholesterol can alter multiple proteins restricted within the SMDs (29). Such findings are supported by Gidwani at al. (12) who showed that incorporation of short-chain ceramides into SMDs correlates with inhibition of phospholipase D and downstream signaling by FcεRI (12).
These findings prompted us to investigate the effect of short-chain ceramide on NTSR1 signaling in breast cancer. The results of the functional studies including cellular proliferation, migration and MMP-9 expression suggested that liposomal short-chain ceramide antagonized NTSR1-mediated functions associated with breast cancer cell tumorigenesis. Previous studies showed that the MAP kinase pathway might, in part, mediate NTSR1 mitogenesis and/or oncogenesis (7). We found that liposomal C6 ceramide reduce NTSR1-dependent MAP kinase signaling (ERK 1/2), correlating with diminished breast cancer progression. The inability of C6 ceramide to disrupt PMA-induced ERK phosphorylation suggests that its effect on NTSR1 signaling is receptor specific. Consistent with this observation, cholesterol depletion using MCD reduced NTSR1-dependent ERK phosphorylation but had no significant effect on PMA- induced ERK phosphorylation, indicating that MDA-MB-231 cells are still responsive to non-receptor mediated signaling even in the presence of C6 ceramide. Taken together, these results suggest that C6 ceramide altered SMDs disrupts receptor-mediated, but not non-receptor-mediated, signaling pathways.
We envisioned that the biophysical mechanism by which ceramide leads to a decrease in NTSR1 signaling is mainly due to the modulation of lipid microenvironment within SMDs. We found that upon NTSR1 stimulation, the receptor translocates into SMDs and this localization could be reversed with exogenous C6 ceramide. We then went further and examined the effect of NTS stimulation on NTSR1 membrane localization, and interaction with G proteins that are candidate transducers of NTSR1-dependent MAP kinase activation. Previous reports show that NTSR1 can couple to Gαq/11 and Gαs (23). However, Souaze et al. suggested that Gαq signaling pathway is more likely to be responsible for NTSR1 mitogenic signaling in breast cancer (7). Using sucrose gradient fractionation, we found that Gαq is preferentially localized to SMDs, in agreement with Oh and Schnitzer (30), suggesting that Gαq interaction with NTSR1 may occur within SMDs. We confirmed this interaction by analyzing the SMD fractions using co-immunoprecipitation and noted that C6 ceramide disrupts NTSR1/Gαq/11 interactions. Based on these data, we believe that the intact SMDs are essential for the formation of the receptor signalplex and disruption of these domains, via C6 ceramide, can prevent the signal initiation. Our previously published data show that exogenous ceramide accumulates into caveolin-rich micordomains (16). We believe that this causes a change in the lipid microenvironment and prevents the movement of NTSR1 into SMDs, thus decreasing interactions with G protein dependent signaling cascades. Although, the data suggest that the G protein coupling for NTSR1 in breast cancer occurs mainly SMDs, we cannot role out the possibility that SMD-localization may also regulate receptor desensitization, trafficking and/or recycling of NTSR1/Gαq complex. Inhibition of NTSR1 localization by ceramide could also disrupt mitogenic signaling in breast cancer due to the reduction in available and responsive NTSR1. However, it is important to note that C6 ceramide treatment did not have any significant effect on caveolin-1 (Fig 4A, ,5B),5B), or NTSR1 expression level (data not shown).
In conclusion, our finding suggests a novel pharmacological approach to target NTSR1 in breast cancers. We demonstrate that nano-liposomal short chain ceramide inhibits NTSR1 induced breast cancer progression via biochemical (ERK), and biophysical (SMDs) mechanisms, findings consistent with our previous in vitro and in vivo work (17). We also provide a mechanism by which modulating SMD-dependent NTSR1 signaling inhibits the receptor ability to interact with Gαq subunits. Collectively, our studies implicate nanoliposomal C6 ceramide as a potential systemic adjuvant for breast cancer therapy. In support of our conclusion that short chain ceramide nano-formulations may have broad valuable clinical applications, we have recently reported that short chain ceramide incorporated into calcium phosphate nanocomposite particles diminish growth of MCF7 cells, a highly differentiated NTSR1 overexpressing and responsive breast cancer cellular model (31)
Suplementary Figure 1. Anti-NTSR1 antibody recognizes NTSR1 in human cell lines. A. Anti-NTSR1 goat polyclonal antibody recognizes NTSR1 in human MDA-MB-231 breast cancer cell lysates and human IMR32 neuroblastoma cell lysates. An apparent 52 kDa band is observed in both lysates, while an additional 54 kDa band is observed in the MDA-MB-231 cell line. B. Anti-NTSR1 antibody recognizes NTSR1 (52−54 kDa) in HEK293T cells transiently transfected with NTSR1 construct. No corresponding bands were visualized in untrasfected HEK293T control cells. Equal protein loading was confirmed by detecting GAPDH expression level. C. Anti-NTSR1 antibody recognizes NTSR1 (52−54 kDa) in HEK293T cells transiently transfected with HA-tagged NTSR1 construct; the same band was detected using Anti-HA tag antibody. No corresponding bands were visualized in untrasfected HEK293T control cells. The 51 kDa legend corresponds to the position of Glutamic Dehydrogenase (SeeBlue Plus2 Pre Stained Standard, Invitrogen).
This work was supported by National Institutes of Health Grant R01HL076789 (to M. K.) and Tobacco Settlement Fund (TSF), State of Pennsylvania, to M. K.
Conflict of interest
Penn State Research Foundation (PSRF) has licensed ceramide nanoliposomes to Tracon Pharmaceuticals Inc. San Diego, CA for development and commercialization. Mark Kester has licensed other non-liposomal nanotechnology for ceramide delivery to Keystone Nano Inc. Boalsburg, PA and MK is CMO of Keystone Nano Inc.