Chemistry of vanadium and zinc antiviral tetra-aza macrocycles
Some bicyclam compounds are highly active as HIV inhibitor agents with high selectivity for the CXCR4 coreceptor, an important protein receptor for the human immunodeficiency virus. One particular bicyclam, AMD3100 (1,1’-(1,4-phenylenebismethylene)-bis-1,4,8,11-tetraazacyclotetra- decane, known as xylyl-bicyclam) reached Phase II clinical trials in 2001. More recently, it has been relaunched commercially as Mozobil, a stem cell mobiliser. The activity of this macrocycle and configurationally restricted analogues as antivirals against HIV, have been shown to increase on complexation with zinc. A range of metals has been studied with varying degrees of activity seen, but none show activity comparable with the zinc complexes. In order to investigate the chemistry of macrocycles in their role as anti-HIV agents, the syntheses of vanadium-cyclam, vanadium-bicyclam and zinc constrained cyclam complexes have been carried out. Of the many oxovanadium(IV) cyclam complexes synthesised, those with a sulfate or chloride axial ligand trans- to the V=O group were chosen for comparison purposes with oxovanadium(IV) bicyclam analogues for both solid state and solution analyses. X-ray crystallography showed that the oxovanadium(IV) cyclam complexes crystallised in the thermodynamically stable trans-III configuration. Aqueous solution studies using EPR spectroscopy revealed a different behaviour for the oxovanadium(IV) cyclam and bicyclam complexes studied. Results suggested that the 6th axial ligand trans- to the V=O group could undergo facile exchange in solution for the cyclam complexes but may be retained in the bicyclam complexes. Antiviral test results for oxovanadium(IV) cyclam complexes revealed these to be completely inactive against HIV. Xylyl-bicyclam was synthesised and complexed with vanadium, as oxovanadium(IV) sulfate and chloride structures. EPR spectroscopy results suggested that for the bicyclam chloride complex some displacement of the 6th axial ligand may occur, but the bicyclam sulfate appeared to retain both axial ligands. More importantly, these ligands also appear to have a significant effect on the hyperfine coupling constant, A. Antiviral test results for oxovanadium(IV) bicyclam sulfate and chloride complexes revealed these to be highly active against HIV, with the chloride structure showing similar activity to AMD3100. Possible interactions between the V=O group and the protein backbone of CXCR4 were studied using molecular modelling techniques, incorporating the X-ray crystal structure obtained for the oxovanadium(IV) cyclam sulfate complex. This revealed the possibility of a weak interaction between the oxygen of the V=O group and the proton of the alpha-carbon of a tryptophan residue, Trp195. No metal-carboxylate interaction was seen between the vanadium centre and aspartate residue Asp171 in the binding site investigated, possibly due to the distance between them being unfavourable for formation of a coordination bond (3.4 Å). Possible hydrogen bonding was seen between the carboxylate oxygens of Asp171 and the NH protons of the trans-III configured ring. These studies may demonstrate how oxo-metals interfere with the protein backbone in such a way as to distort the molecule from an optimum binding geometry. It may also indicate the importance of a second binding site known to support a cis-V configuration of the macrocycle, bearing in mind that vanadium bicyclam complexes do show activity as antivirals against HIV. It is possible that one ring of the vanadium bicyclam complexes interacts with the binding site for the trans-III configuration in a similar fashion to the cyclam complexes and the antiviral activity may result, in this case, from interactions formed through binding of the second ring. The aromatic linker is also thought to be important. Cyclams can be constrained by adding bulky substituents and/or additional rings to the structure and this has the effect of stabilising one configuration. NMR studies show how a zinc constrained cyclam retains the trans-III configuration in solution and also shows high activity against HIV, comparable with the vanadium bicyclam sulfate complex. This emphasises how the oxo-vanadium cyclams must be in an unfavourable position and unable to form crucial interactions known to be important for antiviral activity. A 4-coordinate zinc constrained macrocycle showed unusual binding to three ring nitrogens and may reveal an intermediate species in the uptake and release of metals by cyclams. Further complexation by exchange of the axial ligand revealed a 5-coordinate zinc bound to all four nitrogens on the ring. This thesis provides the first investigation into the role of oxo-vanadium complexes as antiviral agents against HIV looking at possible interactions of the oxo-metal group with the protein backbone of the CXCR4 receptor. An investigation into configurational change in solution for zinc constrained cyclams is also described.