Molecular basis of melanoma progression and drug resistance

In the News:

Cancer Drug Works by Overactivating Cancer Gene

Why don't all moles progress to melanoma?

Melanoma is the most aggressive form of skin cancer. The median survival of metastatic patients is generally limited to 6 to 10 months, and it has not been significantly improved since the 1970s. A main contributor to this poor prognosis is an extreme resistance to standard modalities of anticancer treatment, ranging from immuno-, to radio or chemotherapy. Therefore, the melanoma field is in urgent need of new therapeutic strategies.

The identification of the molecular basis of melanoma progression and chemoresistance are largely unknown. In part, this is due to a notorious inter- and intra-tumor heterogeneity, and by the fact that during their transformation into metastatic cells, human melanocytes acquire a large number of genetic and epigenetic alterations in multiple survival and cell death signaling cascades. Understanding which alterations are causative of the disease (and which a byproduct), what are their specific impact at each step of melanoma development (i.e., from pre-malignant to in situ and metastatic lesions) and specifically how do they affect treatment response remain major challenges in cutaneous oncology.

The working hypothesis of our laboratory is that inactivation of apoptotic pathways plays a direct role in the initiation, invasion and metastasis of melanoma. Moreover defects in cell death factors underlie the multi-drug-resistant phenotype of advanced melanomas. The long-term objective of our research projects is to identify the key modulators of melanoma survival in vivo to develop new therapeutic strategies. Defects in apoptosis are not limited to melanoma, as increased cell survival is a general concept in tumor biology. Therefore, the approaches and results of our studies may be relevant as well to a variety of other aggressive tumor types and may provide the basis for a rational design of improved antineoplastic treatments.

Main Areas of Research

Figure 1 (below) depicts the main areas of research in our laboratory and their connection to melanoma progression and chemoresistance.

Research Aims

1. TUMOR MARKERS - Inactivation of apoptotic pathways during melanoma progression. Identification of tumor markers and drug response genes.
2. TRANSFORMATION - Development of model systems to address the contribution of tumor-stromal interactions to melanoma initiation and progression.
3. CHEMORESISTANCE/TREATMENT - Treatment alternatives to bypass melanoma chemoresistance

Experimental Approaches

Research in our laboratory integrates genetic and functional analyses on normal melanocytes, melanoma cell lines and tumor specimens isolated from all stages of melanoma progression. An important consideration of our experimental approaches is that the malignant phenotype of human melanoma is not only determined by genetic or epigenetic defects of tumor cells, but the environment and neighboring stromal cells are critical contributors as well. Therefore, we consider essential to analyze tumor progression and drug response in the context of the three-dimensional structure of the natural organ where melanoma arises (i.e. skin) or frequently metastatizes (i.e., lymph nodes, lungs or brain). To this end, we use an artificial skin model where melanocytic tumors are engineered with specific genetic alterations in normal or tumor cells. In addition, highly sensitive imaging systems available at the Univ. of Michigan Tumor Imaging Center (fluorescence , bioluminiscence and nuclear magnetic resonance imaging [MRI]) are exploited to monitor melanoma growth and drug response in vivo.

Current Projects

1. Melanoma Tissue Microarrays
2. Artificial Human Skin as a Physiological Platform to Address Cell-Cell Interactions
3. GFP Fluorescence Techniques: Imaging Tumor Growth and Drug Response Regression
4. Visualization of internal metastasis. Nuclear Magnetic Resonance Imaging (MRI)
5. Caspase Activation in vivo
6. Pharmacological Strategies to Overcome Melanoma Chemoresistance
   1. Proteasome Inhibitor Bortezomib
   2. BH3 mimetics

Melanoma Tissue Microarrays

Taking advantage of being part of one of the largest Melanoma Clinics in the United States, (with more than 1,000 new cases per year) we decided to develop microarrays of melanoma tissues containing specimens from all stages of the melanoma disease from premalignant nevus, to metastatic melanoma (See example in Fig.). Those arrays are generated and analyzed in collaboration with Dr. Gruber and the UMCCC Department of Pathology. Large-scale analyses of changes of gene expression of modulator and effector apoptotic genes (Bcl-2, Bcl-xL, Mcl-1 and Apaf-1) during melanoma progression are underway.

Artificial Human Skin as a Physiological Platform to Address Cell-Cell Interactions

Melanocytes provide protective functions to neighboring dermal cells (for example after UV irradiation). However, they also depend on paracrine stimulation from fibroblasts and keratinocytes. To determine the impact of the microenvironment on melanoycte transformation and melanoma cell survival, our laboratory uses a 3D-model that recapitulates the basic architecture of the human skin. In this system, normal melanocytes transduced with specific genes (via lentiviral infection) and/or melanoma cells can be reconstituted either in the epidermis (Fig. 3A) or in the dermis (Fig. 3B). In addition, this model provides an excellent platform for pharmacological analyses of drug response and drug selectivity (see example in Fig. 3C,D).

GFP Fluorescence Techniques: Imaging Tumor Growth and Drug Response Regression

To analyze tumor growth and drug response in vivo we use whole-body imaging systems in animal models. To this end, eGFP-tagged human melanoma lines are injected s.c. or i.v. in mice to address local and metastatic melanomas by fluorescence imaging (Fig. 4).

Visualization of internal metastasis

Nuclear Magnetic Resonance Imaging (MRI). In collaboration with Dr. Alnawaz Rehemtulla and the Small Animal Tumor Center, internal mestastasis are analyzed by MRI. Shown in Fig. 5. are representative images of melanoma cells grown as brain implants (Fig. 5, c-h) compared to other brain tumor cells (glioblastoma; Fig. 5, a-b). Melanoma cells were found to expand in an aggressive manner recapitulating the invasive nature of the human disease. The approximate doubling time of 2-3 days (Fig. 5B) offers an amenable time-window for large-scale studies of tumor growth and drug response.

Caspase Activation in vivo

A chimeric luciferase reporter construct (has been generated by Dr. Rehentulla's lab for in vivo bioluminescent imaging of caspase activation 37 . The firefly luciferase gene is fused at both termini to the estrogen receptor regulatory domain (ER). Luciferase remains inactive until cleavage at internal DEVD sequences, preferentially by casp-3 and -7. Our melanoma cells have been stably transfected with the luciferase constructs and are currently being injected in immunosupressed mice. Fig. 6 [link to graphic} depicts in vivo imaging of caspase activation using this luciferase construct.

Pharmacological Strategies to Overcome Melanoma Chemoresistance

One of the main objectives of our research program is to identify the molecular basis of melanoma chemoresistance and, more importantly, how to bypass or overcome this resistance in vivo. To this end, we are currently testing two novel sets of compounds: (i) a broad-spectrum agent (the proteasome inhibitor Bortezomib), and (ii) a series of small molecule inhibitors that specifically bind and inactivate anti-apoptotic members of the Bcl-2 family (by interfering with their BH3 domain).

1. Proteasome Inhibitor Bortezomib
   We have recently determined that the proteasome inhibitor Bortezomib (also known as Velcade or PS-341) was significantly more potent and selective in melanoma cells than standard chemotherapeutic agents such as adriamycin or cisplatin. More importantly, Bortezomib was able to engage an effective apoptotic program in a panel of melanoma cell lines isolated from early, intermediate and late stages of the disease. Botezomib's toxicity was independent from the functional status of Apaf-1, Bcl-2, Bcl-xL, Mcl-1 and other genes associated with melanoma progression (including p16INK4a, p19ARF or BRAF; not shown). More importanly, fluorescence-based whole-body tumor imaging of mice injected with aggressive melanoma cells showed a significant effect of PS-341 in controlling tumor growth and metastasis in vivo (Fig. 7).
   
   We are currently determining the molecular basis underlying Bortezomib's toxicity to identify putative new targets for therapeutic intervention.
2. BH3 mimetics
   
   
   To determine the specific impact of modulators of the intrinsic (mitochondrial) pathway in melanoma chemoresistance, we are collaborating with Dr. Shaomeng Wang's laboratory (Dept Biological Chemistry) to test a series of compounds specifically act as BH3 mimetics. One such compound, termed BL-193, was found to have a high affinity for the BH3 groove of Bcl-2/Bcl-x L (Fig. 8) and a high killing activity in melanoma cells in vitro and in vivo (see above in Fig.4 [anchor link]). Although it had a preferential effect on melanoma cells, BL-193 reduced the viability of normal melanocytes.
   
   A second generation of BH3 mimetics has been designed with the objective of increasing the affinity for BH3 domains and reducing the toxicity in normal cells. Fig 7A shows the structure of two of thee compounds (TW-37 and TW-97) and an inactive derivative (TW-98A). The higher affinity of TW-37 and TW-90 for Bcl-2 indicated in Fig. 8B. Direct binding to Bcl-2 was determined by NMR (15 N-HSQC) (Fig. 8D). Preliminary in vitro studies in our laboratory have indicated that both TW-37 and TW-90 are highly selective for our melanoma lines in vitro (Fig. 8E). Experiments underway are aimed to establish the mode of action of both compounds in vivo. In addition, particular emphasis is being dedicated to determine whether BH3 mimetics can cross the blood-brain barrier and thus treat brain metastasis (currently incurable).

Associated Programs

• Cellular and Molecular Biology Graduate Program
• Biological Sciences Scholars Program

Back to Cancer Biology