br Results and discussion Computer modelling

Results and discussion Computer modelling studies suggested that the endocyclic oxygens in furocoumarin derivatives, such as STA4783 3a (Fig. 1), could also coordinate metal ions in the HIV-1 IN enzyme active site and inhibit the integration of viral DNA. The coumarin moiety is found in various medicinally active natural products, including umbelliferone (7-hydroxycoumarin) [9] and warfarin [10]. In our own research on applications of Baylis–Hillman chemistry, we have reported several approaches to coumarins and their elaboration to novel analogues [11] of the clinically established HIV-1 PR inhibitor, ritonavir, and, more recently, the synthesis and evaluation of coumarin-AZT conjugates as potential dual-action HIV-1 PR/RT inhibitors [12], [13]. In our approach to the synthesis of the furocoumarins 3a–g as potential HIV-1 IN inhibitors, however, we decided to follow the five-step pathway outlined in Scheme 1. Resorcinol 4 was reacted overnight with diethyl 2-acetylglutarate 5 at 0°C in an ethanolic solution of HCl, generated in situ by the cautious addition of acetyl chloride to dry ethanol (CAUTION! exothermic reaction), and the known coumarin derivative 6 [14] was isolated in 64% yield (Scheme 1). Following a literature procedure [15], alkylation of the phenolic hydroxyl group in compound 6 was achieved by treatment with highly lachrymatory chloroacetone in the presence of freshly calcined K2CO3; the intermediate 7 was obtained, after recrystallisation, in 73% yield. The critical furanocoumarin scaffold 8 was obtained in good yield (79% after recrystallisation) by heating the ether 7 under basic conditions [15]. This mechanistically interesting cyclisation is considered [16] to proceed via initial opening of the coumarin ring to afford a phenoxide ion which, in turn, activates the para-position to intramolecular cyclisation and formation of the furan ring - processes presumably accompanied by saponification of the ester group; subsequent protonation and acid-catalysed dehydration then regenerates the coumarin moiety. With the common scaffold, 3- 3,5-dimethylfuro[3,2-g]coumarin-6-yl propanoic acid 8, in hand, access to the targeted furanocoumarin carboxamides 3a–g was achieved in two steps, viz., activation of the carboxylic acid group by treatment with N-hydroxysuccinimide and N,N′-diisopropylcarbodiimide (DIC) to generate the succinimido intermediates 9 (Scheme 1), followed by in situ reaction with each of the amines 10a–g. The resulting, novel carboxamides 3a–g were obtained in moderate to good yield (63–83% after recrystallisation; Table 1) and fully characterised. Chromatographed samples were tested for their inhibition of the HIV-1 IN strand-transfer process using the method adapted from Hazuda et al. [17]. Percentage enzyme inhibition values for compounds 3a–g and raltegravir (a known HIV-1 IN inhibitor) are reported in Table 1. While some of the synthetic ligands exhibited negligible inhibition activities at a concentration of 10μM, a number of compounds exhibited statistically significant inhibition (p>0.05) at this concentration [viz., 3a (16.61%); 3d (10.31%); 3e (11.12%); and 3g (15.92%)] and may warrant further investigation in the development of possible lead compounds. While compounds 3a and 3d contain pyridyl moieties with hydrogen-bonding potential, compounds 3e and 3g contain non-polar moieties (benzyl and propyl, respectively). In silico docking studies of these specific compounds is expected to guide future developments.
Conclusions The structure of a potential HIV-1 IN inhibitor has been identified from an examination of its virtual interactions with a model of the HIV-1 IN active site based on available sub-domain X-ray crystal structures and NMR structures. This compound and a series of six analogues have been successfully synthesised and HIV-1 IN inhibition studies reveal that some of the ligands exhibit statistically significant inhibition at 10μM concentrations. Ongoing research in this area at Rhodes is being focussed on the development of cinnamate ester-AZT and DKA analogue-AZT conjugates as potential dual-action HIV-1 integrase/reverse transcriptase inhibitors.