Investigation of furo[2,3-h ]- and pyridazino[3,4-f ]cinnolin-3-ol scaffolds as substrates for the development of novel HIV-1 integrase inhibitors

With the aim to develop novel HIV-1 integrase inhibitors, we obtained a set of condensed ring systems based on the furo[2,3-h ]cinnolin-3(2 H )-one and pyridazino[3,4-f ]cinnolin-3-ol scaffolds bearing a potential chelating pharmacophore, which can be involved in the inhibition mechanism of the enzyme. Herein, we report the design, synthesis, structural investigation and preliminary biological results of these heteroaromatic systems


Introduction
HIV-1 integrase (IN), the viral enzyme that catalyzes the integration of proviral cDNA into the host cell genome, has emerged as an attractive target for novel anti-AIDS agents. [1][2][3] The first IN inhibitor (e.g. raltegravir, Isentress) was recently approved by the US FDA, 4 and other IN inhibitors are in clinical trials or under clinical investigation. 5,6 These compounds belong to a class of compounds bearing a β-diketo acid (DKA) pharmacophoric motif, which are the most promising lead in anti-IN drug discovery. 7 In the course of our drug discovery program focused on developing original IN inhibitors, 10-15 a series of polycyclic templates carrying a novel potential chelating pharmacophore has been designed and synthesized. Our attention was addressed to the furo [2,3-h]cinnolin-3(2H)-one scaffold I (Figure 2), as analogue of the previoulsy reported 6-aryl-5-methyl-4,5-dihydro-3-(2H)pyridazinones, 4,4a-dihydro-5H-indeno [1,2-c]pyridazinones II, III, [16][17][18][19] and more strictly to 5,6-dihydrobenzo[h]-, 5,6-dihydrothieno [2,3(3,2)-h]cinnolin-3(2H)-ones IV, V. [20][21][22] These condensed ring systems demonstrated a versatile platform to incorporate a pyridazinone ring, which has shown several pharmaceutical properties. 23 In this context, the N-NH-CO-C-R (R = OH, NH2) motif could be considered as a potential chelating fragment, eventually suitable of bioisosteric replacement of the β-diketo enol pharmacophore. Moreover, furan oxygen of the heteroaromatic backbone, can be involved in potential hydrogen bonding with amino acid residues on the active site. The synthetic routes for the preparation of cinnolinones 3-5 are depicted in Scheme 1. 5,6dihydrofuro [2,3-h]cinnolin-3(2H)-one 3 was synthesized (62% yield) by refluxing the hydroxy(4-oxo-4,5,6,7-tetrahydro-1-benzofuran-5-yl)acetic acid 6 in an excess of hydrazine hydrate for 30 minutes. The intermediate α-hydroxy acid 6 was obtained by reacting the 6,7dihydro-1-benzofuran-4(5H)-one 7 with equimolar amount of glyoxylic acid in an aqueous solution of sodium hydroxide at room temperature. Addition of chloroacetaldehyde to 1,3cyclohexanedione 8, following a previously reported procedure, 24 3-h]cinnolin-3(2H)-one 5 was obtained (38% yield) by reaction with hydrazine hydrate in refluxing ethanol for 3 hours of the (4-Oxo-4,5,6,7-tetrahydro-1benzofuran-5-yl)acetic acid 9, which was obtained in good yield by alkaline hydrolysis of the corresponding ester 10. The latter was prepared by alkylation of the ketone 7 with ethyl bromoacetate in THF solution of LiHMDS at -78 °C (Scheme 1). Compounds 3-5 were fully characterized by means of NMR spectroscopy, mass spectrometry and elemental analysis. In particular, these systems presented a common pattern of signals, constituted by broad singlets for the exchangeable NH protons in the range of 10-13 ppm, and two doublets at 7.69-7.63 and 6.75-6.61, for H8 and H9, respectively, of the furan ring. In addition to the other aromatic H4 proton, compound 3 revealed a multiplet in the range of 3.02-2.93 ppm, attributable to H5 and H6, whereas a more complicated signal pattern ranging from 3.94 to 1.73 ppm was observed for compound 4. The latter also displayed a deuterium oxide exchangeable proton of the OH function located in position 4. As for compound 5 H4 proton appeared upfield as a multiplet centered at 2.84 ppm.
Focusing on 13, in addition to the two expected doublets detected at 8.05 and 7.57 ppm for H5 and H6, respectively, the 1 H-NMR spectrum exhibited two other doublets at downfield centered at 9.34 and 8.47 ppm, attributed to the H9 and H10, respectively. The coupling constant values (J = 5.5 Hz) of these signals are in accordance with similar patterns in the cinnoline ring, further supporting its formation. Also, exchangeable broad singlets detected at 13.50, 7.64, and 6.20 were assigned to the enolic OH in position 3, the NH protons in position 1 and 2, and the NH2 group in position 4, respectively.
Then, when cinnolinone 3 was kept in refluxing hydrazine hydrate for 15-30 minutes, it was converted into 55% of its tautomer 1,2-dihydrofuro[2,3-h]cinnolin-3(4H)-one 14. In particular, the 1 H-NMR spectrum of 14 was characterized by a singlet at 3.59 ppm (H4), a singlet at 7.18 ppm (overlapping H8 and H9), two doublets centered at 7.94 and 7.20 ppm, for H6 and H5, respectively, and by an exchangeable broad singlet at 5.33 ppm, corresponding to the two NH in positions 1 and 2. Further prolonging of reaction time of 14, can reasonably give 11 according to a previously observed behavior for similar reaction 22 (data not shown). Further magnetic resonance techniques such as DEPT/APT, COSY and NOESY (i.e. compound 13, Scheme 2) support the assigned structures for title compounds. A mechanistic hypothesis for the formation of the above-mentioned compounds (11 and/or 14) is displayed in Figure 4. The reaction process can start through an initial tautomerization of 3 to [3a] and 14, which can evolve to give 11 by dehydrogenation in an oxidative step mediated by hydrazine according to the different experimental conditions, whose N-N bond is cleaved to give two ammonia molecules.  Amination at the 4-position of the pyridazinone moiety of 11 to give compounds 12 and 13 was explained according with the mechanism (A, Figure 5) proposed by Singh 26 and Cignarella et al. 25 Briefly, this reaction occurs through an initial 1,4 addition of hydrazine to the pyridazinone ring, to form the intermediate 11', which, by dehydrogenation and final amination to the 4,4a-5,6 conjugated system, leads to 12 (and stage one of 13, Figure 5). However, the alternative mechanism (B, Figure 5), with the initial formation of the intermediate 11", according with previously reported by Shemyakin et al. 27 and Cignarella et al. 28 can also be considered. Although these compounds can directly be obtained from 6, we can reasonably hypothesize that this reaction may proceed via compound 11. The complete formation of 13 can occur by a second-stage mechanism which involves another hydrazine addition to the furan ring of 12 (at position 8 of the furocinnolinone scaffold) to give the intermediate 12a, following by a ring-opening to give 12b ( Figure 5). The formation of the intermediate 12a can also involve a starting furan epoxydation, which is well-documented in the literature, 29 and that can facilitate the Michael addition of hydrazine. Intramolecular nucleophilic attack to the carbonyl group would be carried out by the N1-amino group of the hydrazine coupled to give a six-membered ring 12c, which led to 13 by a final prototropic rearrangement with loss of a molecule of water.
The cinnolin-3(2H)-ones 3-5, 11, 12, the cinnolin-3-ol 13, and the intermediate 6, were tested for their ability to inhibit IN catalytic activities in in vitro assays employing purified enzyme (Table 1). Inhibitors 1 and 2 were used as reference compounds. 15 With the exception of 12 and 13, all tested cinnolinone-derivatives, as well as the intermediate 6, did not show any anti-IN activities. Conversely, the 4-amino-derivatives 12 and 13, shared a certain inhibitory activity, thus demonstrating some inhibitory properties of this novel chemical scaffold. With a IC50 of 60 ± 13 μM against strand transfer reaction, 13 proved to be the most active compound of the series. Interestingly, when compared with reference compound 1, the derivative 13 demonstrated approximately the same inhibitory activities (IC50 values of 60 ± 13 and 69 ± 4 μM for 13 and 1, respectively), thus confirming that several features of these systems could be considered for a structural development. Furthermore, as expected, 13 proved to be more of 100-fold less active of 2, a well studied and validated DKA inhibitor. From a structural point of view, an amino functionality in position 4 (both for 12 and 13) and the enol OH in position 3 (only for 13) of the pyridazinone ring can be predicted as an additional point of chelation on this pharmacophoric fragment, and are important for the anti-IN activity.

Conclusions
In this work, a series of novel heterocycles have been designed and synthesized, to be used as versatile platform in drug design of IN inhibitors. The inhibition of IN enzyme as well as several different viral processes have been targeted via metal chelation. Since the central role of divalent metal ions in these transformations, inhibitors of such processes can be designed on pharmacophores that bind and/or interact to these divalent metal ions. This work has mainly focused in the designing and synthesis of novel chemical scaffold containing a chelating motif addressed toward metal-containing enzymatic sites, such as IN as virological target. Based on the data presented here, these novel prototypes might affect metal affinity in the context of the active site binding. These results prompted us to propose that these types of chromophore are suitable for extensive modifications and will be undertaken in future studies. Therefore, further synthetic and biological investigation for some related congeners are currently in progress and will be reported elsewhere.

Experimental Section
General. Anhydrous solvents and all reagents were purchased from Sigma-Aldrich, Merck or Carlo Erba. All reactions involving air-or moisture-sensitive compounds were performed under a nitrogen atmosphere using oven-dried glassware and syringes to transfer solutions. Melting points (mp) were determined using an Electrothermal melting point or a Kofler apparatus. Nuclear magnetic resonance ( 1 H-NMR, 13 C-NMR, DEPT, COSY, and NOESY) spectra were determined in CDCl3, DMSO-d6 or CDCl3/DMSO-d6 (in 3/1 ratio) and were recorded at 200 MHz and 500 MHz on a Varian XL-200 and a Bruker Avance 500, respectively. Chemical shifts (δ scale) are reported in parts per million (ppm) downfield from tetramethylsilane (TMS) used as an internal standard. Splitting patterns are designated as follows: s, singlet; d, doublet; t, triplet; q, quadruplet; m, multiplet; brs, broad singlet; dd, double doublet. The assignment of exchangeable protons (OH and NH) was confirmed by the addition of D2O. Electron ionization and MALDI-TOF mass spectra (70 eV) were recorded on a Hewlett-Packard 5989 Mass Engine Spectrometer and on a MALDI micro MX (Waters, micromass) equipped with a reflectron analyser, respectively. Analytical thin-layer chromatography (TLC) was carried out on Merck silica gel F-254 plates. Flash chromatography purifications were performed on Merck Silica gel 60 (230-400 mesh ASTM) as the stationary phase. Elemental analyses were performed on a Perkin-Elmer 2400 spectrometer at Laboratorio di Microanalisi, Dipartimento di Chimica, Università di Sassari (Italy), and were within ±0.4% of the theoretical values.

Furo[2,3-h]cinnolin-3(2H)-one (11).
A solution of α-hydroxy acid 6 (0.21 g, 0.0010 mol) in hydrazine hydrate (5 mL, 0.10 mol) was refluxed for 48 hours. After cooling the solid was filtered and washed with ethanol. The solid was then triturated with acetone, filtered to give a beige solid. Yield: 35%; mp 300 ºC dec. 1 (12). From 6. A suspension of 6 (0.23 g, 0.0011 mol) in hydrazine hydrate (5 mL, 0.10 mol) was refluxed for 72 hours. After cooling the excess of hydrazine hydrate was evaporated and the residue was triturated with acetone. The solid formed was purified by flash chromatography eluting with dichloromethane/methanol 9.5/0.5 to give a brown solid. Yield: 40%. From 11. A suspension of 11 (0.20 g, 0.0011 mol) and hydrazine hydrate (5 mL, 0.10 mol) was refluxed for 24 hours. After cooling the excess of hydrazine hydrate was evaporated and the residue was triturated with acetone. The solid formed was purified by flash chromatography eluting with dichloromethane/methanol 9.5/0.5 to give a brown solid. Yield: 45%, mp 300 ºC dec. 1  Integrase assays. Inhibition of IN catalytic activities, 3'-processing (3'-proc) and strand transfer (ST), were evaluated by oligonucleotide-based assays in in vitro assays employing purified enzyme as previously described. 15