Thanos Halazonetis, D.D.S., Ph.D.
Professor
Department of Molecular Biology, Room 3031
University of Geneva, Sciences III
30, quai Ernest-Ansermet
CH-1211, Geneva 4, Switzerland
Tel: 41 22 379 61 12
Fax: 41 22 379 68 68
E-mail: Thanos.Halazonetis@unige.ch

The long-term goal of our research is to understand cancer at the molecular level and then use this knowledge to develop novel cancer therapies. This is a very ambitious goal. Yet, because it is shared by many laboratories world-wide, there is considerable progress and hope for new effective therapies in the coming decades. Our research at the molecular level of cancer may contribute to the development of these therapies.

Thanos Halazonetis joined the Department of Molecular Biology in March 2006. Previously he was at the Wistar Institute in Philadelphia, USA.

He is currently the coordinator of the GENICA FP7 research program

Research Projects

1. Activation of DNA Damage Checkpoint Pathway as a Universal Difference between Cancer and Normal Tissues

There is significant heterogeneity among cancers at the molecular level. In turn, this heterogeneity means that different therapies may need to be developed for each cancer type. For example, Gleevec, an inhibitor of the abl kinase, is very effective in treating a somewhat uncommon form of leukemia, referred to as CML, in which the abl gene is mutated, but in other cancers in which the abl gene is not mutated, Gleevec is usually not effective.

It follows from this that identifying differences that can distinguish all cancers from all normal tissues would be very important, because then such differences could be exploited for development of therapies that are effective against all types of cancer. Our laboratory, in collaboration with the laboratory of Vassilis Gorgoulis (University of Athens Medical School, Athens, Greece) and, independently, the laboratory of J. Bartek (Danish Cancer Society, Copenhagen, Denmark) have identified such a difference. The difference that we identified as distinguishing cancer cells from normal cells relates to the state of activation of the DNA damage checkpoint pathway. This is a pathway that monitors the presence of DNA damage and then, if damage is detected, elicits an appropriate cell cycle response. For example, in the presence of DNA double strand breaks (DSBs), the DNA damage checkpoint pathway will inhibit cell division until the breaks are repaired. What we found is that while in the majority of normal cells the DNA damage checkpoint pathway is not active (because in normal cells the DNA is not damaged), in human precancerous and cancerous tissues, the DNA damage checkpoint pathway is constitutively activated.

We further identified DNA replication stress as the cause for constitutive activation of the DNA damage checkpoint pathway. DNA replication stress is a term that refers to problems with DNA replication. Normally, replication forks replicate the genomic DNA without stalling. However, under certain conditions the replication forks may stall or even completely collapse, a condition referred to as DNA replication stress. Collapse of DNA replication forks leads to DNA DSBs, which then activate the DNA damage checkpoint pathway.

The reason why cancer cells have DNA replication stress is not known, but is something we are actively trying to understand. Irrespective of the reason, the presence of DNA replication stress and, subsequently, of DNA DSBs in cancer cells provides a therapeutic opportunity. Cells with DNA DSBs must repair these breaks before dividing or they will die during cell division. Cancer cells can detect the presence of the DNA DSBs and transiently stop dividing (for a few hours) so the DNA DSBs can be repaired. If the cancer cells could be made to fail to detect the presence of DNA DSBs (by inhibiting the DNA DSB checkpoint), then these cells would divide with broken DNA and die. Accordingly, our laboratory also tries to understand how the DNA DSB checkpoint works at the molecular level, so that approaches for inhibiting its function can be developed.

2. Analysis of the DNA Double Strand Break (DSB) Checkpoint

At the molecular level, the DNA DSB checkpoint involves 53BP1, a protein shown in our laboratory to be a sensor of DNA DSBs; Mre11, Nbs1 and Rad50, three proteins that form a complex and may also sense DNA DBSs; ATM, a DNA damage signaling kinase downstream of 53BP1 and Mre11/Nbs1/Rad50; Chk2, a kinase activated by ATM, and various ATM and Chk2 substrates, the most important of which is the p53 tumor suppressor protein, a transcription factor that induces cell cycle arrest and apoptosis and which is targeted by mutations in the majority of human cancers.

DSBs -> 53BP1 or Mre11/Nbs1/Rad50 -> ATM -> Chk2 -> p53 -> cell cycle arrest or apoptosis

Our laboratory has determined the mechanism by which 53BP1 senses DNA DSBs [25]. We first solved the three-dimensional structure of a 120 amino acid evolutionarily conserved domain within 53BP1 that is sufficient and necessary for recruitment to sites of DNA DSBs. The domain was found to consist of two tandem tudor folds with a deep pocket at their interface formed by evolutionarily conserved hydrophobic residues. We hypothesized that this deep hydrophobic pocket binds to methylated lysine or arginine residues, because methylated lysine/arginine side chains are long and hydrophobic and could fit well in the pocket. In turn, this suggested that 53BP1 may bind to histones, because histones are often methylated. Indeed, the tandem tudor domain of 53BP1 bound histone H3 and further analysis of histone H3 bound to 53BP1 identified methylation of lysine 79 (K79) as the single modification that correlated with 53BP1 binding. Because H3 methylation on K79 is not enhanced in response to DNA DSBs, we propose that preexisting methylated K79 residues become exposed after DNA damage. K79 of histone H3 maps to the histone core and would normally not be exposed in higher order chromatin structure due to nucleosome stacking. However, disruption of higher order chromatin structure by a DNA DSB would result in nucleosome unstacking, exposure of methylated K79 of histone H3 and recruitment of 53BP1 [25]. Interestingly, this model suggests a mechanism to inhibit the DNA DSB checkpoint by developing pharmaceutical compounds that bind to 53BP1 and inhibit its interaction with histone H3.

We intend to continue studying the DNA DSB checkpoint at the molecular level. One very interesting project would be to determine how the ATM kinase is activated by 53BP1 and the Mre11/Nbs1/Rad50 complex.