| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Research Articles: Therapeutics, Targets, and Development
Inhibition of phosphatidylinositol-3-kinase and mitogen-activated protein kinase kinase 1/2 prevents melanoma development and promotes melanoma regression in the transgenic TPRas mouse model
1 Division of Radiation and Cancer Biology, Stanford University, Stanford, California; 2 Division of Epidemiology and Biostatistics, and 3 Arizona Cancer Center, University of Arizona, Tucson, Arizona
Requests for reprints: Marianne Broome Powell, Division of Radiation and Cancer Biology, 269 Campus Drive, CCSR-S-1230, Stanford 94305, CA. Phone: 650-498-5874. E-mail: mbp{at}stanford.edu
Abstract
A number of human melanomas show hyperactivation of the Ras pathway due to mutations of the molecule or alteration of upstream or downstream effectors. In this study, we evaluated the effect of blocking the two Ras downstream pathways phosphatidylinositol-3-kinase/Akt and Raf/mitogen-activated protein kinase kinase/extracellular signal–regulated kinase on melanoma development and regression in the TPRas mouse model. The inhibition of these two signaling cascades by topically applied Ly294002 and U0126 significantly delayed melanoma development and significantly decreased the tumor incidence, particularly when the drugs were applied in combination. Treatment with the inhibitors of established melanomas resulted in complete remission in 33% of mice and partial regression in 46% of mice when drugs were delivered in combination. These responses correlated with increased apoptosis and decreased proliferation both in vitro and in vivo and reduced tumor angiogenesis. In conclusion, this study strongly supports the role of the phosphatidylinositol-3-kinase/Akt and Raf/mitogen-activated protein kinase kinase/extracellular signal–regulated kinase pathways in the development and maintenance of Ras-dependent melanomas and supports the notion that specific inhibition of these effectors may represent a very promising avenue for the treatment and prevention of the disease. [Mol Cancer Ther 2006;5(12):3071–7]
Introduction
Malignant melanoma is a very aggressive cancer with a high propensity to metastasize and with resistance to most current therapeutic regimens. Therefore, melanoma represents a challenging disease for the development of new effective chemopreventive and therapeutic approaches (1, 2).
Common genetic alterations in human melanomas include loss of the tumor suppressor locus CDKN2a, known as a melanoma susceptibility locus (3), and acquisition of activating mutations or hyperactivation of the Ras pathway.
NRas is mutated in 
30% to 50% of human melanomas and dysplastic nevi (4, 5). However, activation of the Ras pathway could also result from aberrant signaling upstream of Ras (6) or from deregulation of Ras downstream targets such as BRaf and Akt3 (7, 8).
Confirmation of the pivotal role played by Ras in melanomagenesis comes from the development of transgenic mice bearing an active mutant of the molecule. These TPRas mice express the human activated Ha-Ras gene under the mouse tyrosinase promoter, which restricts the expression of the molecule mainly in melanocytes. In the model developed by Powell et al. (9), mice develop cutaneous melanoma only after topical application of the carcinogen dimethylbenz(a)anthracene (DMBA) or after exposure of neonates to UV light (10, 11). In the model by Chin et al. (12), melanomas originate spontaneously only when the TPRas mouse is crossed with a CDKN2a knockout animal. Interestingly, in the DMBA-induced melanoma model, tumors often present loss of the INK4a-ARF proteins, which are codified by the CDKN2a locus (13). Both models, therefore, show genetic characteristics that are commonly observed in the human disease, thus representing a suitable system for testing chemopreventive and therapeutic agents for melanoma.
Inhibition of the Ras pathway represents a promising avenue in the treatment of melanoma. Indeed, inhibitors such as perillyl alcohol have shown some efficacy in reducing the incidence of melanoma in TPRas mice (14). However, Ras functions could also be affected by targeting downstream Ras effectors through which survival, proliferation, angiogenesis, and invasion are regulated (15).
We have previously shown that disruption of the phosphatidylinositol-3-kinase (PI3K)/Akt and Raf/mitogen-activated protein kinase kinase (MEK)/extracellular signal–regulated kinase (Erk) pathways by the use of topically applied specific inhibitors leads to the reduced development of a Ras-dependent melanoma cell line in severe combined immunodeficient mice (SCID; ref. 16). Tumor xenografts are used as models for preclinical anticancer drug development because they are predictive indicators of the probable clinical activity of new therapeutic drugs (17). However, these models do not recapitulate the stages of tumor development and primarily allow the study of drug response from previously established tumors, but not the outcome of tumor-preventive approaches. On the other hand, the TPRas mouse represents a model for melanoma where the tumors show similar characteristics to the human disease and present a moderate latency period and a high incidence of tumor development, providing an ideal system for evaluating both chemopreventive and therapeutic agents.
In this study, we show that inhibition of PI3K/Akt and Raf/MEK/Erk pathways by topical application of the specific inhibitors, Ly294002 and U0126, significantly reduces the incidence of melanoma and promotes tumor regression in DMBA-treated TPRas mice. These effects are associated with increased apoptosis and reduced proliferation both in vivo and in vitro and with reduced tumor angiogenesis.
Taken together, these findings show that topical treatment with inhibitors of the PI3K/Akt and Raf/MEK/Erk pathways can be an effective means for both the prevention and treatment of melanoma and shows that both signaling cascades are required in Ras-dependent melanoma development and maintenance.
Materials and Methods
Chemicals and Plasmids
DMBA was purchased from Sigma and dissolved in acetone (Sigma, St. Louis, MO). The PI3K inhibitor, Ly294002, and the MEK1/2 inhibitor, U0126, were obtained from LC Laboratories (Woburn, MA). The final concentrations used for Ly294002 and U0126 were 50 and 10 µmol/L, respectively, for both in vitro and in vivo studies. The PI3K and MEK1/2 dominant negative constructs have been previously described (16).
TPRas Mice
The TPRas transgenic mice have also been previously described (9). This mouse contains a mutated T24 Ha-Ras gene driven by a 2.5 kb fragment of the mouse tyrosinase promoter.
In vivo Experiments and Immunohistochemistry
Mice were housed in an Association for the Assessment and Accreditation of Laboratory Animal Care–approved facility with 12 h light cycles. Food and water were provided ad libitum. In the first experiment, 68 mice were treated with DMBA (50 µg) once a week for 5 weeks to induce melanoma growth and randomly divided into four treatment groups: ethanol (control), Ly294002, U0126, and Ly294002 + U0126. Treatments with the inhibitors started a week before DMBA treatment and were done three times a week for the duration of the experiment (36 weeks). The primary outcome was time to first tumor. Tumor incidence was examined using nonparametric survival analysis (Kaplan-Meier). In the second experiment, 62 mice already bearing melanomas (size range, 150–400 mm3), were divided into four treatment groups as above and followed for 35 weeks to see if tumors progressed, partially regressed, or completely regressed. These data were analyzed using Stata v9's ologit procedure, an ordinal logistic regression model to designate tumors that stopped growing or decreased in size.
For immunohistochemical studies, tumors were excised and frozen in O.C.T. Compound (Sakura, Tokyo, Japan). Five-micron section were cut and stained with antibodies according to the manufacturer's instructions. Briefly, primary antibodies were incubated overnight followed by incubation with biotinylated secondary antibody (1:200; The Jackson Laboratories, Bar Harbor, ME) and with either FITC-conjugated avidin D (1:500; Vector Labs) or streptavidin/horseradish peroxidase (The Jackson Laboratory). Antibodies used were: rat anti-CD31 (platelet/endothelial cell adhesion molecule 1, 1:50; BD PharMingen, Bedford, MA), rat anti-Ki67 (1:50; Dako Corporation, Santa Barbara, CA), rabbit anti-cleaved caspase 3 (1:100), rabbit anti-phosphor-Akt (1:50, immunohistochemistry specific), and rabbit anti-phosphor-MEK (1:50; Cell Signaling Technology, Danvers, MA).
Cell Line
The 1984-1 cell line, derived from a cutaneous melanoma on a TPras mouse, was described previously (14).
Cell Proliferation
Cells (105) were plated in triplicate in 6 cm plates and were counted every 3 to 4 days using an electronic particle counter (Beckman Coulter, Fullerton, CA). Ly294002 and U0126, alone or in combination, were added each time cells were reseeded.
Detection of Apoptosis
Flow cytometry analysis of propidium iodide–stained cells was used to quantify the percentage of apoptotic cells as the fraction of cells with a hypodiploid amount of DNA (sub-G1) as previously described (18). Cleaved caspase 3, a marker of caspase-mediated cell death, was detected by immunoblot analysis.
Northern Blotting
Total RNA from 1984-1 cells treated for 24 h with inhibitors or from cells overexpressing the PI3K or MEK1/2 dominant negative constructs were harvested using TRIzol Reagent (Life Technologies, Inc., Carlsbad, CA) as per the manufacturer's instructions. RNA samples were run on a 1% formaldehyde gel and transferred to Hybond-N+ membrane (Amersham Biosciences, Piscataway, NJ). Blots were hybridized with 32P-labeled human vascular endothelial growth factor (VEGF) cDNA and imaged with a Storm phosphor-screen (Amersham Biosciences).
Immunoblot Analyses
1984-1 cells were plated in RPMI 1640 plus 10% fetal bovine serum, allowed to adhere, then treated with the inhibitors in low serum media (0.5% fetal bovine serum). Cells expressing the dominant negative constructs were placed in low serum media and protein lysates were harvested 24 h later. Protein lysates were made using 9 mol/L of urea, 0.075 mol/L of Tris (7.6) buffer, quantified with the Bradford assay, and run on SDS-PAGE using standard methods. Antibodies used were: anti-phospho-p44/42 mitogen-activated protein kinase (Thr202/Tyr204, 1:1,000), anti-phospho-Akt (Ser473, 1:1,000), anti-cyclin D1 (1:1,000), anti-cleaved caspase 3 (1:1,000; Cell Signaling), anti-cyclin E (1:1,000, Upstate Biotechnology, Lake Placid, NY), anti–basic fibroblast growth factor (bFGF; 1:500; Transduction Laboratories, Lexington, NY), anti–hypoxia-inducible factor 1
(HIF1; 1:1,000; Bethyl Laboratories, Montgomery, TX), and anti-ß-actin (1:200, Santa Cruz Biotechnology, Santa Cruz, CA).
Results
PI3K and MEK1/2 Inhibition Reduces Melanoma Incidence
To test whether selective inhibition of the Ras downstream pathways PI3K/Akt and Raf/MEK/Erk had an effect on melanoma development, we treated TPRas mice with the specific inhibitors Ly294002 and U0126. The treatment was started when mice were free of tumor, a week before five weekly treatments with the carcinogen DMBA. Control mice received DMBA in combination with ethanol, used as a vehicle for the inhibitors. At the end of the study, only 9% of mice were melanoma-free in the control (vehicle) group. In the group of animals treated with U0126, mice showed melanoma development starting at week 10, whereas in Ly294002-treated mice, tumor development was delayed by about 8 weeks with respect to controls. Both inhibitors, when applied alone, reduced the overall melanoma incidence by 40%. When the agents were delivered in combination, 70% to 75% of mice did not develop melanoma (Fig. 1A
). All treatment groups showed significantly lower tumor incidences than the control group, with the combined agents presenting a higher survival rate (P < 0.0001). Importantly, during the prolonged treatment, no systemic toxicities or skin irritations were observed.
|
|
Inhibition of PI3K and MEK1/2 Correlates with Reduced Cell Proliferation and Increased Apoptosis
Treatment with Ly294002 and U0126 reduces cell proliferation in vivo. The average number of Ki67-positive cells was significantly reduced in tumors treated with the combined agents (Fig. 3A
). Treatments with Ly294002 or U0126 alone resulted in reduced proliferation overall, but with intermediate effects with respect to the inhibitors applied in combination (data not shown). These in vivo data correlate with the results in vitro, in which treatment of the TPRas mouse melanoma cell line 1984-1 with Ly294002 and U0126 significantly reduced cell proliferation (Fig. 4A and B
). Also in this case, the effect of the inhibitors delivered in combination was more pronounced than the drugs applied singularly. Even though proliferation was significantly reduced, cells continued to grow, although slower than in control cells. This partial effect of the drugs is likely due, at least in part, to the instability of the compounds in culture.
|
|
Inhibition of both PI3K/Akt and Raf/MEK/Erk pathways increases tumor apoptosis (Fig. 3B). In fact, the number of cells showing positive staining for cleaved caspase 3 was significantly higher in Ly294002/U0126-treated tumors than in controls. This result correlates with the data obtained in vitro on 1984-1 melanoma cells. Acute treatment with Ly294002 induces cell death in 30% of the cells, whereas U0126 does not seem to exert a significant apoptotic effect. The drugs in combination induced cell death in 40% of the treated cells (Fig. 5A
), only slightly higher than the effect produced by Ly294002 alone, suggesting that cell death was mainly due to PI3K inhibition. These data correlate with an increased amount of cleaved caspase 3 in cells treated with Ly294002 and Ly294002/U0126 (Fig. 5B).
|
Most melanoma cells express VEGF and bFGF, which have been shown to contribute to melanoma development and progression by both supporting tumor growth and angiogenesis (22). Reduced vascularization in the TPRas tumors correlates with reduced VEGF and bFGF expression in vitro in 1984-1 cells treated with the chemical inhibitors or overexpressing dominant negative PI3K and dominant negative MEK1/2 (Fig. 6A and B
). This inhibition in angiogenic factor productions is likely to depend on the effect of the inhibitors, both chemical and genetic, on the transcription factor HIF1, whose expression has been shown, in other tumor models, to be, at least in part, controlled by the PI3K/Akt and Raf/MEK/Erk cascades (23–26).
|
In this study, we show that topical treatment with the PI3K and MEK1/2 inhibitors, Ly294002 and U0126, appreciably reduces the incidence of melanoma in the TPRas mouse model and also contributes to the regression of the tumor.
The Ras downstream signaling cascades, PI3K/Akt and Raf/MEK/Erk, regulate survival and growth pathways that are normally shared by both normal and tumor cells (27, 28). The regulation of these functions in physiologic conditions is strictly controlled by the availability of extracellular signals that converge on Ras and its effectors. In pathologic situations, such as neoplasms, these molecules can be constitutively activated, rendering cells able to grow indefinitely and to survive proapoptotic stimuli (29). A tumor cell that acquires such a phenotype is often resistant to common cytotoxic drugs and radiation. Down-regulation of PI3K/Akt and/or Raf/MEK/Erk pathways may therefore contribute to the inhibition of tumor development and progression, and improve responses to common therapies (30, 31).
We show that Ly294002 and U0126 significantly delay melanoma development and reduce the overall incidence of melanoma in TPRas mice, with the greatest effect seen when drugs were applied in combination. Inhibition of tumor development may be due in part to the cytostatic function exerted by the two drugs. In fact, we observed a reduced number of proliferating cells both in vivo and in vitro.
Previous data obtained in a xenograft model showed no effect of U0126 on melanoma proliferation in vivo (16). This is in apparent disagreement with the results obtained here in which both drugs seem to exert a similar inhibitory effect on melanoma development. In the previous study, Ly294002 and U0126 were applied on an already established melanoma cell line injected in SCID mice. Here, the drugs are delivered to melanocytes that were initiated by the presence of Ras and DMBA treatment but not yet transformed. Proliferation is essential for the establishment and perpetration of mutations. Both PI3K and MAPK contribute to Ras-dependent cell proliferation, therefore, it is very likely that both pathways are necessary in the initial steps of melanomagenesis.
Interestingly, established TPRas melanomas also respond to Ly294002 and U0126. The response ranges between complete regression and partial regression, with treatment using both drugs having the greatest effect. The response may depend on the tumor size when treatments are initiated. In fact, not surprisingly, tumors with an average size of 150 mm3 tended to respond better than tumors that were double the size by showing a higher rate of regression.
We have previously observed that although inhibition of MEK1/2 did not inhibit tumor growth in a xenograft model for melanoma, treatment with U0126 contributed to the inhibition of tumor progression in part by reducing invasion and angiogenesis (16). In the studies on the TPRas model, we observed that both treatment with U0126 and Ly294002 contributed similarly to tumor regression (both total and partial) confirming the role of MAPK and PI3K pathways in the tumor maintenance. Nevertheless, we also observed that the percentage of melanomas that progressed in the U0126 group was higher compared with the Ly294002 group. Ly294002, therefore, more efficiently inhibits tumor growth and progression compared with U0126 in this system. A possible explanation for this phenomenon is that whereas the MEK1/2 inhibitor mainly exerts a cytostatic effect on these tumors, blocking PI3K results in both cell growth arrest and increased apoptosis.
Treatments with these drugs resulted in the inhibition of tumor angiogenesis. Treated tumors showed a dramatic decrease in the number of vessels with respect to vehicle-treated melanomas, and this correlates with a strong decrease in VEGF and bFGF expression in 1984-1 cells treated with the inhibitors or expressing dominant negative constructs for PI3K and MEK1/2.
Both angiogenic factors are important mediators of the tumor "angiogenic switch" and have been found to be highly expressed in human melanomas as well as being associated with melanoma progression (32). Oncogenes such as Ras can increase the expression of both VEGF and bFGF. This effect is likely due to increased stabilization/activity of HIF1, a transcription factor that plays an essential role in tumor angiogenesis and tumor progression (33). Both PI3K and MAPK have been shown, in other tumor models, to be implicated in the regulation of HIF1 function downstream of Ras (23–26). Indeed, we have observed that treatments with the chemical inhibitors or the dominant negative molecules reduce HIF1 protein, and this correlates with the reduction in VEGF and bFGF expression in 1984-1 Ras-dependent melanoma cells. Altogether, these effects on proliferation, survival, and angiogenesis contribute to the overall response of Ras-dependent melanomas to the treatments.
Great efforts have been put into the development of new drugs aimed at targeting specific oncogenes that are necessary for tumor development, maintenance, and progression. In melanomas that present BRaf mutations, inhibition of the molecule by small interfering RNA or by chemical inhibitors results in decreased proliferation and increased apoptosis (34, 35). On the other hand, we have shown that Akt-dependent melanomas respond well to rapamycin, a specific inhibitor of mTOR, a downstream target of Akt (36).
In this study, we present evidence that Ras-dependent melanomas require the activity of both PI3K/Akt and Raf/MEK/Erk pathways for their development and maintenance. We show that to achieve a maximal response, such tumors should receive combined treatment with the inhibitors of these molecules. We also show that a topical approach virtually eliminates the systemic toxicity that would likely occur when using inhibitors of pathways crucial for both normal and neoplastic cells.
Taken together, these findings underline the importance of choosing a therapeutic strategy that better counteracts the tumorigenic effect of specific oncogenes. We propose that combined treatment with Ly294002 and U0126 may represent a very promising new therapeutic approach in the prevention and treatment of human melanoma.
Footnotes
Grant support: NIH grants CA090897 and CA27502.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 5/ 9/06; revised 9/13/06; accepted 10/25/06.
References
This article has been cited by other articles:
|
|
 |
|
  H. Hao, V. M. Muniz-Medina, H. Mehta, N. E. Thomas, V. Khazak, C. J. Der, and J. M. Shields Context-dependent roles of mutant B-Raf signaling in melanoma and colorectal carcinoma cell growth Mol. Cancer Ther., August 1, 2007; 6(8): 2220 - 2229. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Cell Growth & Differentiation |
