Abstract

The use of nucleic acid molecules in cancer therapy and diagnosis represents a field in continuous growth. During this thesis, bioconjugation and click chemistry techniques were applied to biological system in order to provide new tools for cancer therapy (chapters 1-4) or diagnosis (chapter 5). In the first part the development of a series of STING agonists is described. STING is a key protein in the regulation of the innate immune system. The activation of the STING pathway begins when DNA is released in the cytosol. This DNA is seen by cells as a clear danger sign upon which the DNA sensor cGAS specifically recognizes and binds cytosolic DNA. Using ATP and GTP, cGAS can synthesise the second messenger 2’,3’-cGAMP (Figure 1). 2’,3’-cGAMP is then recognized by STING and this leads to a conformational change of the protein structure which ultimately triggers interferon expression. Because of their ability to activate the immune system, the use of 2’,3’-cGAMP analogues and STING agonists in medicine is gaining interest, with a constantly growing number of molecules currently in preclinical and clinical trials in the field of immunotherapy or vaccines. Because of the negative charge of 2’,3’-cGAMP, which impairs its uptake by cells and because of its metabolic instability, there is high need for new 2’,3’-cGAMP analogues that can cross the cell membrane or that are more stable towards the action of human or viral enzymes that are known to specifically degrade this cyclic dinucleotide. Therefore, we developed a series of cGAMP analogues that are based on a 2’,3’-cyclic dinucleotide scaffold and that contain adenosine or guanosine nucleosides. In addition to the natural cGAMP, we synthesised the 2’,3’-cyclic adenosine monophosphate-adenosine monophosphate (cAAMP), the dehydroxylated analogues 1, 2 and 3 and the methylated analogues 4 and 5 (Figure 2). In collaboration with Dr. D. Drexler (Hopfner group), the synthesised compounds were tested in thermal shift assays with the soluble portion of STING and then in isothermal titration calorimetry experiments in order to calculate the binding affinities of the synthesised analogues as well as the thermodynamic parameters of their interaction with the protein. Compounds 1, 2, 3, 4 and 5 were then further tested in cellular assays using THP-1 dual reporter cells, which allowed to measure the interferon expression triggered by these analogues and determine their EC50 value, reflecting the potency of the synthesised STING agonists. With these assays, the most potent STING agonist resulted to be compound 1 (EC50= 8.5 μM) while we calculated an EC50 value of 10.6 μM for cGAMP and a value of 60.5 and 106.5 μM for compounds 2 and 3. The methylated analogues 4 and 5 did not induce STING activation in our assays, likely because of the conformation of their ribose moieties. In the second part of this project we developed prodrug derivatives of compounds 2 and 5 described before in order to improve the cell permeability of these molecules and to achieve their efficient internalization. To do this, we modified the S-acylthioethyl (SATE) moiety, which is frequently used in prodrugs, to include a terminal alkyne which allows further late - stage functionalization by click chemistry to improve drug uptake (Figure 3). The synthesised analogue 31 was tested in THP-1 cells to determine its activity. With the introduction of the phosphate caging groups, we calculated an EC50 value of 47.6 nM indicating that 31 is approximately 200-fold more potent than cGAMP itself. In collaboration with W. Greulich (Hornung group) we studied the phosphorylation of the key proteins involved in the STING and interferon pathways (STING, TBK1 and STAT1) by western blot and we detected a much higher phosphorylation level of these proteins using compound 31 compared to cGAMP. Furthermore, the masked derivatives containing an alkyne were further functionalized by click chemistry with an anandamide azide leading to compound 44 and 45. Compound 44 was also tested in THP-1 cells, but, even if it proved to be more active than cGAMP, we measured a lower activity of derivative 44 compared to 31. In chapter 5 published work is presented, in which clickable dendrimers have been used in order to enhance the signal of a cell proliferation assay based on EdU incorporation. After cell feeding with EdU, a tetraazide dendrimer was employed to increase the number of reactive sites per each incorporated alkyne (Figure 5). In a second step, we reacted the multiple azides with a dye alkyne (double click procedure) or with another dendrimer with four alkynes, esponentially increasing the number of reactive alkynes for each incorporated EdU (triple click procedure). In this last procedure, we finally performed a click reaction with a dye azide to allow detection of proliferating cells by fluorescence microscopy. By employing these clickable dendrimers it was finally possible to achieve a 6-fold enhancement in the fluorescent signal.