Final Report Summary - MELANOMA GENES (Gene Discovery in Melanoma Progression and Therapeutic Resistance)
Aim 1: Identification of novel genes cooperating with Braf-V619E in melanoma formation.
Metastatic malignant melanoma has a grave prognosis and discovering driver genes in this disease is challenging because of the abundance of passenger mutations caused by UV light exposure. Small molecule inhibitors targeting the key melanoma oncogene BRAF-V600E or MAPK signalling have shown clinical efficacy. Acquired resistance, however, makes these therapies non-curative. Therefore the identification of additional genetic lesions in melanoma that drive tumor initiation, modulate tumor progression, and influence response to therapy must be identified.
I used the conditional mouse model LSL-Braf-V619E/+; Tyrosinase-CreERT2 (hereby designated as BC mice) in which expression of the oncogenic Braf-V619E mutation (orthologous to the human BRAF-V600E mutation) is specifically induced in melanocytes. This results in the development of skin hyperpigmentation, nevi, and melanomas with a prolonged latency analogous to other models.
The long latency and incomplete penetrance of melanoma in BC mice provides an ideal model system to screen for driver genes. To this end, I generated a quadruple mutant mouse model by intercrossing BC mice with mice carrying the Sleeping Beauty transposon (hereby designated as BCTSB13 mice). Tumorigenesis was dramatically accelerated by transposons, with a median survival of 131 days, compared to BC mice (p-value <0.0001) or the control cohort (p-value <0.0001). I harvested and isolated genomic DNA from a total number of 283 tumours coming from a cohort of 78 mice and the samples were subjected to DNA sequencing for the identification of transposon Common Insertion Sites (CISs). The list of genes linked to CISs included many with established roles in melanomagenesis such as Cdkn2a, Pten and Mitf. In order to identify the genes that are most relevant in the human disease I compared the CIS list with two different CGH datasets of human melanoma: one including 111 short-term melanoma cultures and cell lines, and another dataset of 24 melanoma short-term cultures.
Notably, the Enhancer of Polycomb 1 (EPC1) gene was one of the most commonly altered genes in our analysis, and is also mutated in ovarian, breast, colon carcinomas and glioma in the COSMIC database, thus I decided to focus my attention on this gene as a potential tumour suppressor. The distribution of transposon insertions inside Epc1 is concordant with a role as a tumor suppressor with insertions distributed with a pattern predicted to disrupt the gene. Furthermore, the deletion of EPC1 correlated with a reduction in mRNA expression using gene expression data derived from 26 short-term human melanoma cultures.
Then, in order to perform a functional validation, I engineered a cell line derived from a BC mouse melanoma in order to suppress Epc1 expression using a lentiviral delivered small hairpin RNA (shRNA). Interestingly, the suppression of Epc1 expression significantly increased melanoma cells proliferation, allowed cells to form colonies when seeded on plastic at low density, and imparted upon them anchorage-independent growth in soft-agar, all features consistent with a tumor suppressor.
Therefore, with this study I identified a list of potential new driver genes of melanoma and decided to follow up on one of them. These experiments allowed showing that EPC1 is a novel melanoma tumor suppressor gene whose loss causes increased proliferation and transformation of melanoma cells.
Aim 2. Identification of genes responsible for resistance to BRAF-V600E inhibitors in melanoma
The discovery that over 60% of melanomas harbour the activating BRAF-V600E mutation led to the development of several BRAF-V600E-specific inhibitors. Clinical experience with the inhibitor Vemurafenib showed a notable 80% anti-tumor response against BRAF-V600E-positive melanomas. However, development of acquired resistance impairs its clinical impact. Therefore, understanding of the molecular mechanisms of resistance to BRAF inhibitors operative in vivo is vital in order to establish new effective therapies for patients.
With this aim I used the BCTSB13 mouse model of BRAF-V600E melanoma to identify novel genetic determinants of melanoma resistance to BRAF inhibitors.
I randomly enrolled a cohort of 4-OHT treated BCTSB13 mice for treatment with the BRAF inhibitor PLX4720. Mice with melanomas of at least 200 mm3 in size were transferred onto a standard lab diet containing PLX4720. An acute response to PLX4720 treatment was evident; melanomas initially regressed during the first 1-4 weeks of treatment. Stable disease was maintained in BCTSB13 mice for an average of 76 days until relapse at the site of the initial tumor formation was observed. After relapse tumors were collected before reaching 10% of body weight. I harvested 53 PLX4720-resistant tumors from 30 mice, I isolated genomic DNA from them and used for DNA sequencing to isolate CISs, which were defined using Gaussian kernel convolution statistics (GKC). This analysis yielded a list of 8 significant CISs corresponding to unique genes. Importantly, this list included Braf, Mitf and Cdkn2a, all of which have previously been described as mediators of resistance to BRAF inhibitors in melanoma patients, while the remaining genes of the list have not previously been implicated in resistance to BRAF inhibitors.
One of the top genes of the list of candidates was the proto-oncogene ERas (ES cell-expressed Ras), a small gene spanning ~4 Kb on the X chromosome. In order to validate the role of ERAS in resistance to PLX4720 I ectopically expressed human ERAS in BRAFV600E-mutant melanoma cell lines by retroviral transduction and assessed survival in response to PLX4720 concentrations ranging from 0 to 100 μM. Accordingly, ERAS expression reduced sensitivity to PLX4720 in melanoma cell lines.
Since previous work has shown that ERas activates the Phosphatidylinositol 3-kinase pathway (PI3K), but not the MAPK pathway, I examined AKT, ERK and MEK phosphorylation in melanoma cells overexpressing ERAS. I found robust AKT phosphorylation while MAPK activity was unaffected, confirming previous published data. Given this result, I then assessed the effect of the combination of PLX4720 and the allosteric AKT inhibitor MK2206 in ERAS-expressing melanoma cells. Dual PLX4720/MK2206 treatment negated ERAS-mediated resistance of melanoma cells to PLX4720, confirming that drug resistance is due to activity of the PI3K pathway.
Having established that ERAS functions through the PI3K/AKT pathway in melanoma cells, I reasoned that it might confer resistance to PLX4720 by activating AKT-dependent survival signals. Therefore, I assessed the effect of PLX4720 and MK2206 treatment of melanoma cells on members of the Bcl-2 family of apoptotic proteins, which have been established as important mediators of melanoma cell survival. These experiments allowed determining that ERAS-expression did increase the survival of PLX4720 treated cells, a phenotype that was negated by combined treatment with PLX4720 and MK2206. Furthermore, increased resistance of ERAS-expressing cells to PLX4720 is mediated through upregulation of the PI3K/AKT pathway and phosphorylation/inactivation of the pro-apoptotic protein BAD.
A correlation between stromal production of hepatocyte growth factor (HGF) in patients with BRAF mutant melanoma and resistance to BRAF inhibitors was recently reported. This is due to concomitant re-activation of both MAPK and PI3K pathways in melanoma cells expressing the HGF receptor MET. Since HGF is known to regulate cell survival through PI3K-mediated phosphorylation/inhibition of BAD, we asked whether HGF causes resistance to vemurafenib through the regulation of BAD. Therefore, I determined the levels of BAD phosphorylation in response to PLX4720 with the addition of HGF under conditions previously shown to elicit resistance, and without HGF as a control. In concordance with the hypothesis, BAD was hyperphosphorylated following PLX4720 treatment in the presence of HGF. Hence, the mechanism of resistance induced by ERas in the mouse model activates a survival pathway that is the same operative when human melanoma cells are treated with stromal HGF. Since HGF-dependent resistance has been shown to operate in patients where stromal expression of HGF can be detected, our experiments confirm that regulation of the activity of BAD is relevant in determining the response of melanoma to BRAF inhibition.
The results of this research will have an impact not only on scientific knowledge and academia but also in the economic and societal sectors. For example, it will impact:
- the industrial sector and attract R&D investments to generate drugs against new melanoma drivers or a new generation of drugs for melanoma to avoid the development of resistance;
- the industrial sector involved in the search for diagnostic melanoma biomarkers;
- new research or clinical trials to examine novel drug combinations;
- the project will also increase the excellence and competitiveness of UK and Europe by contributing to rationalise the design and development of new and more efficient drugs for the therapy of melanoma;
- this research will also have impact on different categories of the general public, such as the non-scientific audience reachable through non specialised publications, the broader audience of the media, the locals which are keen to participate to the local scientific festivals. This will contribute to increase the understanding of science as well as the awareness of scientific research and its achievements in the public.
Metastatic malignant melanoma has a grave prognosis and discovering driver genes in this disease is challenging because of the abundance of passenger mutations caused by UV light exposure. Small molecule inhibitors targeting the key melanoma oncogene BRAF-V600E or MAPK signalling have shown clinical efficacy. Acquired resistance, however, makes these therapies non-curative. Therefore the identification of additional genetic lesions in melanoma that drive tumor initiation, modulate tumor progression, and influence response to therapy must be identified.
I used the conditional mouse model LSL-Braf-V619E/+; Tyrosinase-CreERT2 (hereby designated as BC mice) in which expression of the oncogenic Braf-V619E mutation (orthologous to the human BRAF-V600E mutation) is specifically induced in melanocytes. This results in the development of skin hyperpigmentation, nevi, and melanomas with a prolonged latency analogous to other models.
The long latency and incomplete penetrance of melanoma in BC mice provides an ideal model system to screen for driver genes. To this end, I generated a quadruple mutant mouse model by intercrossing BC mice with mice carrying the Sleeping Beauty transposon (hereby designated as BCTSB13 mice). Tumorigenesis was dramatically accelerated by transposons, with a median survival of 131 days, compared to BC mice (p-value <0.0001) or the control cohort (p-value <0.0001). I harvested and isolated genomic DNA from a total number of 283 tumours coming from a cohort of 78 mice and the samples were subjected to DNA sequencing for the identification of transposon Common Insertion Sites (CISs). The list of genes linked to CISs included many with established roles in melanomagenesis such as Cdkn2a, Pten and Mitf. In order to identify the genes that are most relevant in the human disease I compared the CIS list with two different CGH datasets of human melanoma: one including 111 short-term melanoma cultures and cell lines, and another dataset of 24 melanoma short-term cultures.
Notably, the Enhancer of Polycomb 1 (EPC1) gene was one of the most commonly altered genes in our analysis, and is also mutated in ovarian, breast, colon carcinomas and glioma in the COSMIC database, thus I decided to focus my attention on this gene as a potential tumour suppressor. The distribution of transposon insertions inside Epc1 is concordant with a role as a tumor suppressor with insertions distributed with a pattern predicted to disrupt the gene. Furthermore, the deletion of EPC1 correlated with a reduction in mRNA expression using gene expression data derived from 26 short-term human melanoma cultures.
Then, in order to perform a functional validation, I engineered a cell line derived from a BC mouse melanoma in order to suppress Epc1 expression using a lentiviral delivered small hairpin RNA (shRNA). Interestingly, the suppression of Epc1 expression significantly increased melanoma cells proliferation, allowed cells to form colonies when seeded on plastic at low density, and imparted upon them anchorage-independent growth in soft-agar, all features consistent with a tumor suppressor.
Therefore, with this study I identified a list of potential new driver genes of melanoma and decided to follow up on one of them. These experiments allowed showing that EPC1 is a novel melanoma tumor suppressor gene whose loss causes increased proliferation and transformation of melanoma cells.
Aim 2. Identification of genes responsible for resistance to BRAF-V600E inhibitors in melanoma
The discovery that over 60% of melanomas harbour the activating BRAF-V600E mutation led to the development of several BRAF-V600E-specific inhibitors. Clinical experience with the inhibitor Vemurafenib showed a notable 80% anti-tumor response against BRAF-V600E-positive melanomas. However, development of acquired resistance impairs its clinical impact. Therefore, understanding of the molecular mechanisms of resistance to BRAF inhibitors operative in vivo is vital in order to establish new effective therapies for patients.
With this aim I used the BCTSB13 mouse model of BRAF-V600E melanoma to identify novel genetic determinants of melanoma resistance to BRAF inhibitors.
I randomly enrolled a cohort of 4-OHT treated BCTSB13 mice for treatment with the BRAF inhibitor PLX4720. Mice with melanomas of at least 200 mm3 in size were transferred onto a standard lab diet containing PLX4720. An acute response to PLX4720 treatment was evident; melanomas initially regressed during the first 1-4 weeks of treatment. Stable disease was maintained in BCTSB13 mice for an average of 76 days until relapse at the site of the initial tumor formation was observed. After relapse tumors were collected before reaching 10% of body weight. I harvested 53 PLX4720-resistant tumors from 30 mice, I isolated genomic DNA from them and used for DNA sequencing to isolate CISs, which were defined using Gaussian kernel convolution statistics (GKC). This analysis yielded a list of 8 significant CISs corresponding to unique genes. Importantly, this list included Braf, Mitf and Cdkn2a, all of which have previously been described as mediators of resistance to BRAF inhibitors in melanoma patients, while the remaining genes of the list have not previously been implicated in resistance to BRAF inhibitors.
One of the top genes of the list of candidates was the proto-oncogene ERas (ES cell-expressed Ras), a small gene spanning ~4 Kb on the X chromosome. In order to validate the role of ERAS in resistance to PLX4720 I ectopically expressed human ERAS in BRAFV600E-mutant melanoma cell lines by retroviral transduction and assessed survival in response to PLX4720 concentrations ranging from 0 to 100 μM. Accordingly, ERAS expression reduced sensitivity to PLX4720 in melanoma cell lines.
Since previous work has shown that ERas activates the Phosphatidylinositol 3-kinase pathway (PI3K), but not the MAPK pathway, I examined AKT, ERK and MEK phosphorylation in melanoma cells overexpressing ERAS. I found robust AKT phosphorylation while MAPK activity was unaffected, confirming previous published data. Given this result, I then assessed the effect of the combination of PLX4720 and the allosteric AKT inhibitor MK2206 in ERAS-expressing melanoma cells. Dual PLX4720/MK2206 treatment negated ERAS-mediated resistance of melanoma cells to PLX4720, confirming that drug resistance is due to activity of the PI3K pathway.
Having established that ERAS functions through the PI3K/AKT pathway in melanoma cells, I reasoned that it might confer resistance to PLX4720 by activating AKT-dependent survival signals. Therefore, I assessed the effect of PLX4720 and MK2206 treatment of melanoma cells on members of the Bcl-2 family of apoptotic proteins, which have been established as important mediators of melanoma cell survival. These experiments allowed determining that ERAS-expression did increase the survival of PLX4720 treated cells, a phenotype that was negated by combined treatment with PLX4720 and MK2206. Furthermore, increased resistance of ERAS-expressing cells to PLX4720 is mediated through upregulation of the PI3K/AKT pathway and phosphorylation/inactivation of the pro-apoptotic protein BAD.
A correlation between stromal production of hepatocyte growth factor (HGF) in patients with BRAF mutant melanoma and resistance to BRAF inhibitors was recently reported. This is due to concomitant re-activation of both MAPK and PI3K pathways in melanoma cells expressing the HGF receptor MET. Since HGF is known to regulate cell survival through PI3K-mediated phosphorylation/inhibition of BAD, we asked whether HGF causes resistance to vemurafenib through the regulation of BAD. Therefore, I determined the levels of BAD phosphorylation in response to PLX4720 with the addition of HGF under conditions previously shown to elicit resistance, and without HGF as a control. In concordance with the hypothesis, BAD was hyperphosphorylated following PLX4720 treatment in the presence of HGF. Hence, the mechanism of resistance induced by ERas in the mouse model activates a survival pathway that is the same operative when human melanoma cells are treated with stromal HGF. Since HGF-dependent resistance has been shown to operate in patients where stromal expression of HGF can be detected, our experiments confirm that regulation of the activity of BAD is relevant in determining the response of melanoma to BRAF inhibition.
The results of this research will have an impact not only on scientific knowledge and academia but also in the economic and societal sectors. For example, it will impact:
- the industrial sector and attract R&D investments to generate drugs against new melanoma drivers or a new generation of drugs for melanoma to avoid the development of resistance;
- the industrial sector involved in the search for diagnostic melanoma biomarkers;
- new research or clinical trials to examine novel drug combinations;
- the project will also increase the excellence and competitiveness of UK and Europe by contributing to rationalise the design and development of new and more efficient drugs for the therapy of melanoma;
- this research will also have impact on different categories of the general public, such as the non-scientific audience reachable through non specialised publications, the broader audience of the media, the locals which are keen to participate to the local scientific festivals. This will contribute to increase the understanding of science as well as the awareness of scientific research and its achievements in the public.