Towards a Synthesis of Naphthalene Derived Natural Products

Dieckmann-type cyclization reactions have been employed in the synthesis of the alkyl substituted naphthoquinone 11 and the naphthalenes 10 and 12. Various conditions for the benzylic oxidation of these compounds have been investigated with a view towards the synthesis of some naphthalene based natural products.


Introduction
A large number of naphthalene derived natural products are based on a substituted naphthalene skeleton as represented by 1. These include: the naphthyl ketone derivative guieranone A (2) [1], heterocyclic derivatives like the linear naphtho-γ-pyrone (NGP) rubrofusarin (3) [2][3], and naphthoquinones such as the crinoid derived pigment 4 [4]. Given that 3 is the only linear NGP based natural product to have been synthesized, [5][6][7] and that many structurally related NGPs possessing biological properties have been isolated, we are interested in developing a general synthesis of NGPs and as a corollary this synthesis should be extendable to related naphthalene derived natural products represented here by 2 and 4. Synthetic methods to access these substituted naphthalene based compounds are limited. While the 2-acetyl naphthalene 1 could serve as a starting material in the synthesis of the target compounds, the synthesis of substituted 2-acyl naphthalenes is not trivial. Traditional substitution of naphthalenes using Friedel-Crafts type reactions generally leads to peri-substituted products making the regioselective synthesis of compounds such as 1 from naphthalene precursors difficult. Shibata and co-workers circumvented this by cyclising the β-diketoester 5 at 200-220°C under high vacuum, [5] to give 1, albeit in a low 9% yield (misreported as 18%) while some years later Rideout and colleagues increased the yield in the synthesis of 1 to 18% by cyclising 5 in refluxing 1-methylnaphthalene [8]. Interestingly concurrent attempts by Shibata to cyclize 5 employing polyphosphoric acid gave the alkyl naphthalenic ester 6, representing initial attack at the distal ketone carbonyl. This paper outlines recent work directed at developing an improved synthesis of substituted 1,3,6,8-tetraoxygenated-2acyl-naphthalenes with a view towards the synthesis of naphthalene based natural products.

Results and Discussion
We envisaged a possible route to the desired 2-acyl-naphthalenes via a benzylic oxidation of an alkyl-substituted naphthalene. In this direction the ketones 8 and 9 were obtained in excellent yields through acylating the phenylacetic acid derivative 7 [9] with hexanoic or pentanoic anhydride, respectively, in toluene containing catalytic perchloric acid (Scheme 1). Base induced cyclization of 8 in refluxing sodium ethoxide gave the desired (metastable) 1,3,6,8tetraoxygenated-2-alkyl-naphthalene 10 in 25% yield, which readily oxidized to afford the major isolated product from the reaction, the naphthoquinone 11 (52% to 77% yield). The propensity for 10 to oxidize to 11 suggested it would be unwise to attempt benzylic oxidations on 10 without first protecting the phenol groups.
The MEM protected alkyl-naphthalene 12 was therefore synthesized by treating 9 with 3.1 equivalents of sodium hydride in DMF followed by the addition of MEM-Cl. The first equivalent of NaH promotes cyclization to the naphthalene, and the remaining NaH prevents protonation of the phenol groups, trapping the naphthalene as its di-sodium salt.
It was anticipated that benzylic oxidations on both the naphthoquinone 11 and the naphthalene 12 would lead ultimately to the 2-acyl-naphthoquinones and the 2-acyl-naphthalenes respectively, thereby providing access to some of our target compounds. We had initially thought that 12 could be brominated in the benzylic position however, reactions with NBS returned only the ring-brominated product 13. To date conditions suitable for effecting the desired benzylic oxidations have not been found with the majority of the conditions attempted resulting in decomposition of the starting material (Table 1). Future work will center on replacing the MEM group with alternative protecting groups in order to ascertain if the steric interactions are responsible for the lack of reactivity in the benzylic position.

(CAN) Decomposition
Experimental General 1 H-and 13 C-NMR spectra were recorded at 300 and 75 MHz, respectively, on either a Varian Gemini 300 or on a Varian Inova 300 spectrometer. 1 H spectra were referenced to residual protonated solvent (7.26 ppm) while 13 C spectra were referenced to the central peak of the CDCl 3 triplet (77.0 ppm). HMQC and HMBC spectra recorded on a Varian Inova 300 spectrometer were routinely measured to assist in spectra assignment and structural identification. Results are reported in the form δ(ppm), integration, mult, J, assignment (for 1 H) or δ(ppm), assignment (for 13 C). IR spectra were recorded on a Shimadzu FTIR 8400 Series spectrometer as thin films on NaCl discs. ESI-MS spectra were recorded on a FISONS VG QUATTRO II mass spectrometer, operating at a cone voltage between 20 and 70 V, with positive ion detection. EI-MS and HREI-MS were recorded on a VG autospec mass spectrometer, operating at 70 eV using positive ion detection. UV spectra were recorded on a Varian Cary 4G UV-Vis spectrophotometer in a 1cm cell. Melting points (mp) are uncorrected and were recorded on an Electrothermal digital melting point apparatus (Electrothermal Eng. Ltd.). Most reagents were obtained from Aldrich Chemical Company and used as supplied.