Targeting MYC and CDK2 in cancer : impacts on senescence, apoptosis and immune modulation

Abstract

Cancer is a disease in which cells acquire the ability to divide uncontrollably and invade other tissues. An important component of this process is the activation of oncogenes and inactivation of tumor suppressor genes. The MYC family of oncogenes, consisting of MYC, MYCN and MYCL, is frequently deregulated in human cancer, and is often associated with aggressive disease, resistance to treatment and poor outcome. MYC encodes a transcription factor that controls many essential cellular processes of relevance for tumor development. In this thesis, two such MYC-regulated processes have been in focus: cellular senescence and immune modulation. Previous work has shown that MYC suppresses cellular senescence, which together with apoptosis is an important barrier against tumor development. Recent studies have also shown an important role for MYC in escape from immune surveillance. Due to the central role of MYC in cancer there is an increasing interest in targeting MYC, which for a long time have been considered “undruggable”. However, to function as a transcription factor, MYC needs to interact with several partner proteins. Two such proteins in the focus of this thesis are MAX and cyclin-dependent kinase 2 (CDK2). Targeting these interactions or the activities of partner proteins are plausible ways of inhibiting MYC function. This thesis centres around evaluating these two different strategies to inhibit MYC functions and develop new treatments for cancer therapy. We reported previously that suppression of senescence by MYC requires CDK2-mediated phosphorylation of MYC at serine 62, and that ablation of CDK2 activity results in senescence induction in vitro. In the first part of this thesis, encompassing paper I and II, we evaluated the role of CDK2 for MYC-mediated suppression of senescence and for MYC-driven tumor development in vivo using depletion and pharmacological inhibition of CDK2 in two MYC-driven mouse tumor models representing acute myeloblastic leukemia (AML) and lung tumors.

In paper I, we utilised a mouse AML model driven by MYC and the anti-apoptotic factor BCL-XL, Hematopoietic stem cells (HSC) purified from bone marrow cells of mice were transduced with lentiviral vectors containing MYC and BCL-XL and then transplanted into lethally irradiated mice. Untreated mice developed massive AML-like leukemia after 2-3 weeks of transplantation. The mice were treated with the CDK2 inhibitor CVT2584 or vehicle upon disease onset, either by daily intraperitoneal injections or through continuous delivery via mini-pumps. Analysis of isolated bone marrow, liver and spleen cells of mice by flow cytometry revealed a significant decrease of the leukemic AML cell population in the treated mice. CDK2 targeting reduced phosphorylation of MYC at Ser 62 in leukemic cells, significantly delayed onset of disease and improved mice survival. This was associated with induction of senescence, as measured by the decrease in proliferation, increase in senescence associated β-gal activity, H3K9me3 heterochromatin foci, p19ARF and p21CIP1, and activation of pRb.

In paper II, we utilized a conditional immunocompetent mouse lung tumor model induced by CRE-mediated recombination of mutant BRAFV600E and the MycER fusion protein through inhalation of Ad-CRE virus. In addition, MYC activity in this system is regulated by tamoxifen. These mice were cross-bred with conditional CDK2 knockout mice, to create a model where CDK2 is deleted by Ad-CRE simultaneously with activation of the two other loci. Depletion of CDK2 resulted in delayed/inhibited tumor development and significant enhancement of overall survival. This was associated with induction of senescence in tumors lacking CDK2, evidenced by reduced tumor cell proliferation and pRB phosphorylation, and induction of p21CIP1 and H3K9me3. Similar results were obtained by pharmacological inhibition of CDK2. RNA-Seq analysis of whole lung tissue revealed that CDK2 loss counteracted MYC activity and induced the expression of genes representing the senescence-associated secretory phenotype (SASP) and genes involved in immune cell recruitment and activation. Supporting this notion, immunofluorescence staining demonstrated an increase infiltration of T-, and B-cells and macrophages in the lungs of CDK2-deficient mice irrespective of MycER status. Further, FACS analysis of immune cell populations in the lung revealed a significant increase in CD8+ T cells, macrophages and activated NK cells, and a reduction in CD4+ T cells, in response to CDK2 depletion. Finally, examination of the effect of MYC deactivation in CDK2-depleted tumors showed a massive reduction in tumor burden. Collectively, our results demonstrate that inhibition/depletion of CDK2 inhibited AML and lung tumor development linked to induction of cellular senescence. Additionally, we uncover a promising potential role for CDK2 in modulating immune surveillance.

MAX is a dimerization partner that is essential for MYC’s function as a transcription factor, and this interaction is therefore a promising target to abrogate MYC activity in cancer cells. In the second part of the thesis, covering paper III and IV, we identified and characterized small molecules inhibiting MYC:MAX interaction. Using a cell-based MYC:MAX interaction inhibitor screen employing bimolecular fluorescence complementation (BiFC) assay, we identified six compounds, named MYC:MAX inhibitors (MYCMIs), specifically reducing the BiFC signal. The effect of the compounds on MYC:MAX interaction was further validated in cells utilizing split Gaussia luciferase (GLuc), in situ proximity ligation assay (isPLA), co-immunoprecipitation as well as functional studies using chromatin immunoprecipitation and analysis of MYC target gene expression. This showed that MYCMI-6 and MYCMI-7 were the most potent and selective MYC:MAX inhibitors with IC50s in the single digit micromolar range. Next, we examined direct binding of MYCMI-6 and MYCMI-7 to recombinant bHLHZip domains of MYC and MAX in vitro, using microscale thermophoresis (MST) and surface plasmon resonance (SPR) assays. The results showed that MYCMI-6 and MYCMI-7 both bound directly and selectively to MYC with a KD of 1.6 and 4.3 μM, respectively, but did not bind MAX to any greater extent. Analysis of MYC expression showed that MYCMI-7 reduced MYC protein levels while MYCMI-6 did not. Evaluation of the inhibitory effects of MYCMI-6 and MYCMI-7 on MYC-dependent cell growth showed that MYC-expressing transformed cells were highly sensitive, while MYC-null cells were largely unaffected. Further, MYCMI-6 and MYCMI-7 strongly reduced growth/viability of different MYC-driven cancer cell lines, including MYCN-amplified neuroblastoma and Burkitt’s lymphoma cells at single digit μM concentrations, and significant correlation between MYC mRNA/protein levels and the sensitivity to treatment was observed in the NCI-60 human tumor cell line panel. Importantly, both compounds caused growth arrest but did not have cytotoxic effects on normal cells. Finally, treatment with MYCMI-6 and MYCMI-7 in vivo using an MYCN-amplified neuroblastoma xenograft mouse model resulted in reduced MYCN activity/expression, reduced tumor burden and massive apoptosis and necrosis in the tumor tissue, without causing severe side effects. Further experiments using MYCMI-7 treatment showed similar effects in mouse tumor models of AML and breast cancer, resulting in increased survival in the neuroblastoma and breast cancer models. The results obtained from MYCMI-6 and MYCMI-7 are encouraging and warrant further investigation with respect to their mechanism of action and the improvement of their efficacy and bioactivity for further pre-clinical/clinical development.

In conclusion, the two approaches to interfere with MYC function presented in this thesis - inhibition of CDK2 and of MYC:MAX interaction - showed promising results and can potentially pave the way for new advancements in anti-MYC cancer therapy as well as new immunotherapies.