Main Content

Timofeev group

1. Tumor addiction to late p53 inactivation

Cancer development is driven by activated oncogenes and loss of tumor suppressors. While the addiction of tumors to oncogenes is well-established and provides the rationale for oncogene-targeted therapies, it is less clear whether tumors likewise depend on the absence of tumor suppressors. This issue has been studied most extensively for p53 – the most commonly inactivated tumor suppressor gene. Mouse studies using p53 alleles that can be switched from knockout to wildtype have provided first genetic proof-of-concept that restoring p53 function in a p53-deficient tumor induces therapeutic responses and has fueled the development of compounds targeted to reactivate inactive p53. As first compounds are currently evaluated in clinical trials, it becomes evident that many of the observed effects are explained by off-target activities, warranting more sophisticated investigation into the dependence of tumors on p53-loss.

It is a caveat of the initial mouse studies that p53 reactivation was investigated only in models where loss of p53 was the initiating driver of tumorigenesis. Tumor development in a p53-compromised background alleviates the requirement for additional mutations in p53-dependent pathways such as oncogene-induced apoptosis or senescence. Reactivation of p53 in this context restores coupling of oncogenic signals to intact apoptosis or senescence machineries and results in tumor regression. While these studies nicely model tumor development in Li-Fraumeni patients with p53-germline mutations, p53 mutations in sporadic patient tumors often occur at later stages of tumorigenesis. Such tumors initially evolve in the presence of wild-type p53 and acquire alterations in the p53 pathway to cope non-mutated p53. If such tumors mutate p53 at later time points, the preexisting alterations in the p53 pathway could limit the therapeutic efficacy of p53 reactivation approaches.

It is the goal of this project to evaluate whether tumors become dependent on the loss of p53 if it occurs at late stages of tumor development. These studies will be conducted in a mouse model of Myc-driven lymphoma in which p53 is a well-established barrier to tumorigenesis. To mimic pharmacological p53 reactivation, we will employ a tamoxifen-regulated p53-fusion protein that allows p53 to be reversibly switched from active to inactive and back. Lymphomas with early and late p53 inactivation will be compared to explore how the timing of p53 inactivation during tumorigenesis influences the response to therapeutic p53 reactivation approaches. The obtained findings will be validated in an independent mouse model of lung cancer and in human cancer cells. These studies are expected to provide fundamental support for or against p53 as a suitable target for therapeutic reactivation strategies.

Funding:
DFG TI 1028/2-1

2. Phosphorylation control of p53 DNA binding cooperativity in apoptosis and tumor suppression

 p53 protects us from cancer by transcriptionally regulating tumor suppressive programs leading to cell-cycle arrest or apoptosis. While a p53-driven cell cycle arrest is in principle reversible, apoptosis permanently eliminates an incipient cancer cell from the organism. How p53 decides between cell-cycle arrest and apoptosis is therefore of major importance in clinical oncology. Own previous work indicated that the activation of apoptotic, but not cell cycle inhibitory, target genes requires the interaction and cooperative DNA binding of p53 molecules. Loss of DNA binding cooperativity results in a pronounced predisposition to cancer development, emphasizing the importance of DNA binding cooperativity for p53's anti-cancer activity. Whether and how DNA binding cooperativity is regulated is presently unclear. Preliminary data suggest that DNA binding cooperativity is inhibited by phosphorylation of a serine residue at the p53 interaction interface. As phosphorylation signals can be therapeutically targeted by kinase inhibitors, our findings have the potential to open new avenues to enhance p53's tumor suppressor activity for cancer prevention and tumor therapy.

To explore this concept we have assembled an interdisciplinary team from Marburg and Gießen that comprises experts on p53 DNA binding cooperativity, p53 phosphorylation and p53 mouse models. Our primary goal is to identify and biochemically validate druggable kinases that regulate p53 DNA binding cooperativity. We will further explore how phosphorylation-dependent control of DNA binding cooperativity impacts on the genomic p53 binding pattern and chromatin status by ChIPseq, characterize the resulting gene expression changes by RNAseq and analyze the impact on p53-based cell fate decisions with a special focus on apoptosis. Importantly, we plan to investigate whether cooperativity can be increased with kinase inhibitors to stimulate chemotherapy-induced apoptosis and overcome therapy resistance in cancer patients. Last but not least, we plan to generate a phosphorylation-site mutant p53 knock-in mouse to address the implications for normal development and tumor suppression in a physiological context. Together this work program is aimed to provide comprehensive, pre-clinical evidence for targeting p53 cooperativity with kinase inhibitors as a promising, novel strategy for cancer therapy.

Funding:
Deutsche Krebshilfe (#111444 together with Prof. Dr. Lienhard Schmitz, Justus-Liebig University, Gießen, Germany)

3. Therapeutic targeting of non-hotspot TP53 mutants in AML

TP53 mutations are found in 7-8% of de novo and ~25% of therapy-related AML cases and associated with adverse outcome. TP53 mutations are mostly missense mutations that – in the case of classical hotspot mutations – destroy the normal tumor suppressive activity of p53 as a transcription factor (loss-of-function, LOF) and endow the mutant p53 (mutp53) protein with oncogenic properties (gain-of-function, GOF) that actively promote progression and therapy resistance. As tumors become addicted to stabilized GOF mutp53, elimination of mutp53 by destabilization is considered a promising therapeutic approach. However, the vast majority (~70%) of all TP53 mutations are not hotspot mutations and surprisingly little is known about their functional properties. There is substantial evidence that many of these non-hotspot mutp53 proteins are stabilized in tumor cells, but – in contrast to hotspot mutp53 – often retain residual transcriptional and non-transcriptional tumor suppressive activities (partial LOF). In this project, we plan to explore how leukemia with such non-hotspot TP53 mutations can be treated most effectively using a mouse model for a partial LOF p53 mutation that we have recently developed. Results from this mouse model will be validated with human AML cells engineered to express non-hotspot mutp53 with CRISPR/Cas9 nucleases. We will investigate whether (a) residual tumor suppressive activities of a non-hotspot mutp53 can be therapeutically enhanced with MDM2 inhibitors, (b) the presence of a non-hotspot mutp53 protein generates therapeutic vulnerabilities to non-transcriptional forms of cell death and (c) leukemia cells become addicted to non-hotspot mutp53 as a basis for therapy with mutp53 destabilizing drugs.

Funding:
Deutsche José Carreras Leukämie Stiftung (DJCLS 09 R/2018)

4. Establishment of iPSC-based lung organoid system as an alternative for animal-based research models

Lung diseases are the most common human disorders worldwide, which result in 4 million premature deaths each year. Due to the enormous socio-economic importance of this problem the research activity in the field of chronic respiratory diseases, in particular in cancer studies, is constantly growing. Because of the high complexity of lung as an organ, which cannot be accurately reproduced in cell culture, the most common experimental models used in such studies are laboratory animals, mostly laboratory mice. However, work with laboratory mice is slow, cost- and labor-intensive and, importantly, mouse data cannot be directly translated to human biology and must be validated with human cells and tissues. Use of tissue organoids is an attractive alternative. An organoid is an in vitro produced, three-dimensional multicellular structure that accurately reproduces micro-anatomical features of a functional subunit of an organ – in other words, this is a miniaturized and simplified organ. Organoids can be generated from induced pluripotent cells using directed differentiation, where the key steps of organogenesis during embryonic development are recapitulated in vitro. Using our previous experience in lung organoids, cancer research and gene editing, in this project we want to establish an iPSC-based platform for generation of human lung organoids with desired genetic modifications. This platform can be used for basic and applied research as a replacement for transgenic mouse models, primarily as an alternative for lung cancer mouse models. To achieve this goal, first we plan to generate genetically engineered human iPSC lines with inducible expression of CRISPR components. Besides integrated CRISPR-machinery these cell lines are modified to improve the efficiency of gene editing and to permit conditional expression of targeted genes. Next, we use these cell lines for generation of lung organoids. Finally, we plan to generate the organoids with known cancerous mutations for phenotypical and molecular characterization and comparing the results with our previous data and published works obtained in mouse models and human tumors. The iPSC lines and technologies that will be developed in the frame of this project are important to replace mouse models of lung diseases and potentially can be used for other organoid models, thus providing more possibilities for further reduction of using laboratory animals in basic and translational research.

Funding:
Bundesministerium für Bildung und Forschung (Alternativmethoden - Verbund: IPSELON - Teilprojekt A, 161L0279A)

5. Understanding and controlling of the tumor microenvironment in p53-mutated lung cancer

Lung cancer is the leading cause of cancer deaths in both men and women with a devastating 5-year survival rate of ≤15%. Different therapeutic options are currently available including classic chemo- and radiotherapy as well as modern molecular treatments with oncogene inhibitors or immunotherapeutics such as immune checkpoint inhibitors. While oncogene inhibitors primarily target cell-autonomous functions of the tumor cells, immunotherapies act on the tumor microenvironment to engage specific immune cells and direct them at the tumor cells. Response to immunotherapies is still poorly predictable and apparently dependent on the architecture of the tumor microenvironment. 

The most frequently mutated gene in all subtypes of lung cancer is the powerful tumor suppressor gene TP53. In contrast to other tumor suppressors, where cancer-associated frameshift or nonsense mutations lead to the absence of the protein product, TP53 is mostly affected by missense mutations. These result in expression of an aberrant protein which has not only lost its tumor suppressor functionality (loss of function, LOF) but in many cases acquired novel oncogenic properties (gain of function, GOF) that contribute to tumor progression, metastasis and drug resistance. While the cell-autonomous effects of p53 mutations are under intensive investigation since decades, only very little is known about the role of mutant p53 in modulation of the tumor microenvironment.

In this project we aim to investigate the effect of p53 missense mutations on the tumor microenvironment using lung adenocarcinoma (LUAD), the most frequent type of lung cancer, as a model. Pathogenic TP53 mutations are identified in at least 50% of cases with clear prognostic implications. Using knock-in mouse models we generate autochthonous LUAD driven by KrasG12C oncogene in combination with either a p53 knock-out, as a model for a LOF mutation, or GOF p53 mutations. We combine these cancer gene mutations also with inactivation of DNA mismatch repair to increase heterogeneity and tumor mutational burden, responsible for neoantigen production and immunogenicity. Tumors of all genotypes are comprehensively analyzed by deep microenvironmental profiling using bulk and single-cell transcriptomics, multispectral flow cytometry and multiplexed immune-fluorescence imaging with spatial analysis. Our primary goal is to characterize non-cell-autonomous mechanisms hijacked by mutant p53 upon remodeling of the tissue microenvironment. Using single-cell sequencing we aim to identify stromal and immune cells present in the genetically distinct tumor entities and infer receptor-ligand interactions that mediate pro-tumorigenic effects of mutant p53 on the tumor microenvironment. Since p53 mutants are key drivers of lung cancer progression and therapy resistance but still considered difficult to target directly, this work program is aimed to identify critical, and potentially better druggable, non-cell autonomous effector mechanisms of mutant p53 that operate in the lung cancer microenvironment.

Funding:
Deutsche Krebshilfe (DKH 70115830, in cooperation with Justus-Liebig-Universität Gießen, Prof. Dr. Rajkumar Savai)