Modeling of leukemia in flies

Acute myeloid leukemia (AML) is one of the most frequently observed forms of leukemia in the Western world. The current strategy to treat AML has hardly changed during the past four decades, and consists of aggressive chemotherapy. A large number of patients relapses, and 5-year survival rate is stagnant at approximately 60%. One major problem in development of new forms of AML treatment is our lack of understanding of how the mechanisms that establish the cancerous state of AML cells. Identifying the genes that support proliferation and survival of the tumor cells is a major task.

The goal of this project is to model human leukemia in the fruit fly Drosophila melanogaster. The advantage of fly models of leukemia is that we can perform unbiased, large-scale genetic screens to identify the genetic network that underlies the cancer state. We also plan to use this fly model of AML to perform high-throughput drug screens to identify compounds that slow down or even revert cancer progression. The results of these experiments will be validated in human AML cells.

The project will start in the summer of 2015 and will be headed by Helene Knævelsrud.

Cancer projects

 

1. Development of novel cancer therapies

We are particularly interested in developing novel treatment options for rare forms of cancer, as well as for cancers that do not respond to current forms of therapy. One example is mast cell leukemia, a rare but extremely aggressive form of leukemia with little treatment options and dismal prognosis. Another example is drug-resistant chronic myeloid leukemia (CML), which can also be difficult to treat. Using methods originally developed for yeast research, we have designed a powerful cell competition assay that we are using for high-throughput screening of large drug libraries. The goal of these screens is to identify compounds that specifically kill oncogene-expressing cells while leaving isogenic, untransformed cells unharmed. This project is sponsored by a persoanlized medicine grant from the Norwegian Cancer Society.

 

2. Personalized medicine for AML and CML patients

Acute myeloid leukemia (AML) is a common form of cancer. Unfortunately, AML has a relatively poor survival rate, especially for elderly patients who often do not tolerate the aggressive chemotherapy that is used to treat this disease. Further complicating effective treatment of patients is the fact that AML can be caused by a heterogeneous set of mutations, and it is believed that this heterogeneic nature of the disease is one of the underlying causes for chemotherapy failure. Sponsored by the personalized medicine grant from the Norwegian Cancer Society already mentioned above, we are collaborating with clinicians at Oslo University Hospital to develop a method for individualized treatment of AML patients. The ultimate goal is to provide treatment for patients that currently do not qualify for standard of care treatment, and for patients that have failed chemotherapy.

 

3. Oncogene addiction

Our cancer research projects are primarily focused on oncogene addiction. Oncogene addiction is the puzzling phenomenon in which cells which have been exposed to the activity of a specific oncogene for a certain period of time suddenly become critically dependent upon the continuous activity of that oncogene for their survival. In this process, the oncogene somehow re-wires cellular survival pathways. How this is accomplished is not well understood, although it has been suggested to involve irreversible alterations of transcriptional programs. We are particularly interested in unraveling the mechanisms by which these oncogenes re-wire the transcriptional circuitries of cells.

Yeast projects

 

Goal of our yeast work

Cells are continuously exposed to changes in their environment. This is not only a challenge for single-celled organisms like budding yeast, but it also holds true for for example tumor cells, which can be exposed to various forms of stress including low oxygen levels, chemotherapy, radiation, and low nutrient levels. For optimal growth and survival, cells have developed mechanisms that sense environmental alterations and generate responses that maintain cellular homeostasis. A major response to environmental stimuli is transcriptional reprogramming. We use the budding yeast Saccharomyces cerevisiae as a model organism to understand how cells cope with environmental changes, in particular in terms of regulation of transcription. We mainly focus on regulation of transcription by post-translational modifications, such as phosphorylation, ubiquitination and sumoylation.

 

1. Sumoylation and cell stress

We want to understand how cells modulate transcription in response to environmental changes, in particular nutrient starvation. We are particularly interested in the role of the ubiquitin family protein Sumo. Sumoylation is clearly important in rewiring of transcription in response to cell stress. However, which proteins are sumoylated and how Sumo regulates their function remains mysterious. By performing ChIP-seq experiments and mass spectrometry we have identified a number of proteins that become sumoylated during cell stress. We are currently studying the molecular mechanism by which Sumo modulates the activity of these proteins to regulate transcription.

 

2. Regulation of transcription by signal transduction kinases and phosphatases

Another major focus of the group is to determine how kinases and phosphatases regulate transcription. In response to specific environmental cues, signaling pathways rewire transcriptional programs to make sure the cell optimally responds to changes in conditions. These signaling pathways often consist of kinases and phosphatases. However, how these enzymes control transcription is not well understood. Using a combination of ChIP-seq, mass spectrometry and fluorescence microscopy we are unraveling the molecular mechanisms by which signaling kinases and phosphatases control transcription.

 

Previous findings

Rad6 in transcription and cell cycle regulation

Using a high-throughput chemical-genetic screen, we found that the E2 ubiquitin conjugase Rad6 is temporally recruited to promoter regions of cyclin genes, where it ubiquitinates histone H2B (Fig. 1). This promotes transcription of cyclins, thereby activating the cyclin dependent kinase Cdk1 to promote efficient cell cycle entry.


Figure 1. Cell cycle regulation by Rad6. (A) Cell cycle entry is initiated by phosphorylation of Whi5 by Cln3-Cdc28, which leads to nuclear exclusion of Whi5 and activation of SBF. SBF activates the G1 transcriptional program, which includes the cyclins CLN1 and CLN2, which in a positive feedback loop further phosphorylate and inhibit Whi5 to induce efficient cell cycle entry. (B) Doa1 supplies ubiquitin to the pool of free ubiquitin. Rad6 forms a link between the pool of free ubiquitin and the cell cycle by promoting transcription of cyclins. (C) Cdc48 promotes cell cycle progression in at least two ways. It promotes degradation of Far1, presumably through its binding partners Ufd1 and Npl4. This relieves the inhibition of Cln-Cdc28 complexes, in particular when cells recover from pheromone-induced cell cycle arrest. In addition, Cdc48 promotes cell cycle progression by increasing the expression of cyclins through the Doa1-ubiqtuin-Rad6-Bre1 pathway.

 

Cdk1 in regulation of transcription

We discovered that Cdk1 has a kinase-dependent function in regulation of the basal transcription machinery. Using ChIP-seq, we found that Cdk1 localizes to highly transcribed genes, primarily involved in housekeeping. At these genes, Cdk1 cooperates with another CDK, Kin28 (the catalytic subunit of TFIIH), to phosphorylate RNA polymerase II (Fig. 2). RNA polymerase II has a long C-terminal domain that consists of 28 repeats of the sequence Y1S2P3T4S5P6S7. Cdk1 and Kin28 promote phosphorylation of the serine residue at position 5 (Ser5). This in turn is important for efficient recruitment of the capping machinery, which places a cap on the mRNA as soon as it protrudes from the polymerase.


 

Figure 2. A model for cooperation between Cdk1 and Kin28 in regulation of RNA polymerase II. (A) Kin28 phosphorylates the C-terminal domain of RNA pol II, which may result in recruitment of Cdk1, which further phosphorylates RNA pol II (B).

 

Cell reponses to nutrient stress: Regulation of fatty acid and ceramide metabolism by TORC1 and GSK3

Very long chain fatty acids (VLCFAs) are essential fatty acids with multiple functions, including ceramide synthesis. Although the components of the VLCFA biosynthetic machinery have been elucidated, how their activity is regulated to meet the cell’s metabolic demand remains unknown. We recently found that the fatty acid elongase Elo2 is regulated by phosphorylation. Elo2 phosphorylation is induced by nutrient stress and inhibition of TORC1 and requires GSK3. We found that expression of nonphosphorylatable Elo2 profoundly altered the ceramide spectrum, which reflects aberrant VLCFA synthesis. Furthermore, depletion of VLCFAs resulted in constitutive activation of autophagy. This constitutive activation of autophagy diminished cell survival, indicating that VLCFAs serve to dampen the amplitude of autophagy. Together, our data revealed a function for TORC1 and GSK3 in the regulation of VLCFA synthesis with important implications for autophagy and cell homeostasis.

 

Regulation of long chain fatty acid synthesis

 

Figure 3. The rate of long-chain fatty acid synthesis depends on the cell's nutrient status. The activity of the long chain fatty acid synthase Elo2 is stimulated by phosphorylation by GSK3, which is counteracted by PP2A. Failure to phosphorylate Elo2 leads to accumulation of long chain bases and phosphorylated long chain bases, which activate autophagy.

 

Genetic network of the cell cycle

Aberrant CDK activity underpins growth of all tumors. Several CDK inhibitors are, or have been, in clinical trial for cancer chemotherapy, with mixed outcome. A major problem with current chemotherapeutics is their toxicity and side-effects, such as immunosuppression, myelosuppression, nausea, and in some cases secondary neoplasms.

One goal of my lab is to identify (a) the pathways that are controlled by CDKs, and (b) the pathways that function in parallel to these CDK-controlled pathways to promote robust cell cycle progression.

Unraveling this genetic network of the cell cycle will facilitate development of pharmaceutical drugs that specifically and efficiently block cell cycle progression of tumor cells. We are make use of several cdk1 mutant alleles to perform genome-wide genetic screens in budding yeast to map all the genetic pathways that involve CDK1.

Cdk1 and regulation of transcription

Cells display a remarkable ability to adapt to changes in their environment. Many external stimuli evoke cellular responses that result in reprogramming of transcription profiles. These changes in gene transcription may in turn affect important cellular decisions, such as e.g. proliferation, senescence, or apoptosis. Exactly how cells relay signals from their environment to induce activation or repression of genes is not well understood, especially in terms of cell cycle control. My lab is interested in unraveling the molecular mechanisms of transcriptional responses to environmental stimuli, and how these changes affect cell cycle progression. 

 

Cdk1, transcription, and basic cell cycle regulation

A major funtion of Cdk1 is to control transcriptional programs. Cell cycle entry is well conserved among eukaryotes, and involves the activation of genes involved in cell cycle progression. In higher eukaryotes, the transcription factor E2F induces transcription of a variety of genes that facilitate the G1-S transition. During G1 phase, E2F forms a complex with the Retinoblastoma (Rb) tumor suppressor, which prevents binding of E2F to its target sites in promoters of cell cycle genes. Rb also recruits histone deacetylases (HDACs) to repress transcription. However, when Rb is phosphorylated by the CyclinD-Cdk4/6 and CyclinE-Cdk2 complexes, it no longer recruits HDACs and it dissociates from E2F, resulting in activation of E2F-controlled genes.
A similar mechanism operates in budding yeast. While Rb does not exist in yeast, another protein, Whi5, serves a similar function. Whi5 binds the transcription factor SBF (consisting of Swi4 and Swi6), thus keeping it inactive. Whi5, like Rb, also recruits HDACs that repress transcription. Phosphorylation of Whi5 by Cdk1 leads to dissociation of the complex, allowing SBF to activate a transcriptional program that induces the G1-S transition. Cdk1 also directly phosphorylates transcription factors involved in cell cycle progression. For example, it phosphorylates and activates the forkhead transcription factor Fkh2 as well as its binding partner, the transcriptional co-activator Ndd1.

In addition to phosphorylating transcription factors to induce transcription, Cdk1 may also facilitate transcription by recruiting the proteasome to ORFs, which appears to be independent of its kinase activity.

 

Transcriptional responses to external stimuli and cell cycle regulation

The cellular response to carbon sources

We make use of several well-defined models to unravel the molecular mechanism of transcription regulation upon environmental stimuli. In particular, we study the cellular responses to changes in external carbon sources, such as glucose and galactose. While the response to these carbon sources has been studied extensively, the interplay between the cell cycle and the genes involved in carbon metabolism remains poorly understood. We make use of a variety of techniques to unravel the role of Cdk1 in regulation of transcription of these genes, including chromatin immunoprecipitation (ChIP), RT-PCR, and ChIP followed by sequencing (ChIP-seq). Together with the genetic network of CDK1, this provides insight in the cellular response to carbon metabolism and feedback to cell cycle control.

The cellular response to DNA damaging agents

We recently identified a genetic interaction between CDK1 and RAD6 (Enserink JM et al, 2009). Rad6 is an E2 ubiquitin conjugase that interacts with several E3 ubiquitin ligases: Ubr1, involved in N-end rule protein degradation; Rad18, involved in post-replication DNA repair; and Bre1. The Rad6-Bre1 complex monoubiquitinates histone H2B and Swd2, leading to methylation oh histone H3, which affects transcription and which has a role in activation of the DNA damage checkpoint that induces cell cycle arrest. Especially the BRE1 branch of the RAD6 pathway strongly genetically interacts with CDK1, indicating an involvement of Cdk1 in Rad6-mediated transcriptional control. We are currently unraveling the role of Cdk1 in regulation of transcription, especially in during DNA damage.