Stereoselective Synthesis of the C9-C19 Fragment of Lyngbyaloside B and C via Ether Transfer

Abstract

A stereoselective synthesis of the C9-C19 fragment of lyngbyaloside B and C highlighted, by an extension of our ether transfer methodology, enables the formation of tertiary ethers. 2-Naphthylmethyl ethers have been shown to proceed efficiently through ether transfer with high stereoselectivity and are easily deprotected by DDQ oxidation. Variation of the workup conditions results in the stereoselective formation of syn-1,3-diol mono- or diethers.

Polyketide metabolites of marine cyanobacteria are increasingly attracting the attention of synthetic groups due to their unique structural features and diverse biological activity. These compounds have demonstrated a broad spectrum of biological activities, including antiviral, antibiotic, antifungal, anticancer, and protein synthesis inhibition. For example, investigations of L. bouillonii collections from the Ulong Channel in Palau led to the discovery of lyngbyaloside B by Moore (Figure 1).¹ More recently, Luesch and co-workers isolated congeners, 18E- and 18Z-lyngbyaloside C, from Papua New Guinea.²

Figure 1.

The Structure of Lyngbyalosides B and C from L. bouillonii.

Many polyketides from marine cyanobacteria have been shown to possess β-branching, methyl substituents at odd numbered carbons, resulting from an HMG-CoA synthase (HCS) cassette within the polyketide synthase.³ The lyngbyalosides are structurally interesting due to a lactone generated by acylation of a tertiary alcohol. Despite initial efforts, lyngbyaloside B and C have yet to be synthesized.⁴

We have recently developed a novel approach to the formation of syn-1,3-diol mono- and diethers through electrophilic activation of homoallylic alkoxymethyl ethers.⁵ As shown in Scheme 1, activation of alkene 1a with iodine monochloride, in toluene at low temperature, initiates a stereoselective cyclization to form oxonium ion 2a through a chair-like transition state. NMR evidence supported the ultimate generation of chloromethyl ether 3a, which can be selectively functionalized with a variety of nucleophiles. Hydrolytic workup directly generates a 1,3-syn-diol monoether 4a with excellent stereocontrol.

Scheme 1.

Syn-1,3-Diol Mono- and Diethers via Ether Transfer.

More recently, we envisaged extension of this methodology to the preparation of tertiary ether 4b from 1,1-disubstituted alkene 1b. The effect of axial methyl substituent (R’) in 2b on the stereoselectivity of the cyclization was an important question. However, good diastereoselectivities were observed in related iodine-induced carbonate cyclization previously demonstrated by Cardillo.⁶

The homoallylic methoxymethyl ether 5a was readily obtained as a racemate from ring-opening of the corresponding terminal epoxide with isopropenyl magnesium bromide followed by protection with commercially available methoxymethyl chloride. Initial studies focused on subjecting 5a to a variety of electrophilic activation conditions followed by a hydrolytic workup and the results are listed in Table 1. The choice of activating agent and solvent proved critical for high yield and diastereoselectivity of the desired methyl ether transfer product and suppression of side reactions. As in our previous study,^5a iodine monochloride and toluene at low temperature were found to be the best activation conditions (entry 1) and increased diastereoselectivity was observed by lowering the temperature to −95 °C (entry 6). With favorable reaction conditions in hand, the analogous benzyloxymethyl ether 5b was investigated. Benzyl ether transfer provides access to orthogonally protected syn-1,3-diol units since benzyl groups can typically be removed by hydrogenolysis. The choice of the activating agents and solvents were again critical in the formation of the ether transfer product 6b over acetal 7a which results from benzyl cleavage of intermediate 8.^5a The use of ICl in toluene provided predominantly the ether transfer product 6b (entry 8), while activation in DCM led to preferential formation of the 1,3-dioxane 7a as the major product (entry 10). Interestingly, the use of IBr in toluene (entry 9) also resulted in benzyl cleavage although rationale for the sensitivity of the reaction to halide counter ion and solvent is currently unclear.

Table 1.

Electrophile-Induced Methoxy Transfer

	sm	conditions	yield (dr)α	yield 7a (dr)
1	5a	ICl, PhCH3, −78 °C	6a 78% (5:1)	not obs.
2	5a	IBr, PhCH3, −78 °C	6a 62% (1.1:1)	not obs.
3	5a	I2, PhCH3, −78 °C	trace	not obs.
4	5a	IPy2BF4, PhCH3, −78 °C	NR	not obs.
5	5a	ICl, CH2Cl2, −78 °C	6a 63% (2.4:1)	not obs.
6	5a	ICl, PhCH3, −95 °C	6a 75% (8.5:1)	not obs.
7	5a	ICl, pentane, −115 °C	6a 53% (7.5:1)	not obs.
8	5b	ICl, PhCH3, −78 °C	6b 63% (8.5:1)	15%
9	5b	IBr, PhCH3, −78 °C	6b 5%	71% (5:1)
10	5b	ICl, CH2Cl2, −78 °C	trace	60% (3:1)
11	5b	ICl, CH3CN, −78 °C	6b 50% (2.5:1)	trace
12	5b	ICl, PhCH3, −95 °C	6b 60% (10:1)	not obs.
13	5c	ICl, PhCH3, −95 °C	6c 65% (11:1)	not obs.
14	5c	IBr, PhCH3, −78 °C	6c 51% (20:1)	20%

^αDiastereomeric ratio was determined by ¹H and ¹³C NMR.

The use of an alternative arylmethyl ether in the transfer, a 2-naphthylmethyl group, improved the selectivity for ether transfer product 6c over dioxane 7a even under the IBr activation conditions (entries 13 and 14). The stereochemistry of ether transfer products 6a-c were all inferred from 7a, unambiguisly assigned by ROESY (Rotating-frame Overhauser Effect Spectroscopy) experiments, since each of these compounds arise from intermediate 8.

Our original publication on ether transfer provided in situ NMR evidence to support the intermediacy of chloromethyl ether intermediate 3 prior to aqueous workup and loss of the acetal methylene as formaldehyde.^5a This enabled the development of a number of nucleophilic workup conditions to provide syn-1,3-diethers.⁵ Of particular interest, was the regeneration of the methoxymethyl ether through a basic methanol quench. As shown in Table 2, ICl-induced ether transfer followed by basic methanol quench provided orthogonally protected diethers 9a-c from 5a-c in high yields and stereoselectivities. The 2-naphthylmethyl ether was found to be a particularly useful transferable group as the resulting product could be easily deprotected with aqueous DDQ providing 10 in 84% yield.

Table 2.

Ether Transfer – Methanol Quench

	sm	R	product	yield (dr)
1	5a	−CH3	9a	88% (7:1)
2	5b	−CH2Ph	9b	65% (12:1)
3	5c	−CH2(2-Naphth)	9c	74% (12:1)

Despite its critical role in the ether transfer, the use of iodine monochloride has also shown some limitations. With some substrates, an appreciable amount of ICl addition products have been observed leading to low yields of the desired ether transfer. Recently, Snyder reported polyene cyclizations affected by new electrophilic halogen sources such as bromodiethylsulfoniumbromopentachloroantimonate (BDSB) as well as the iodine version (IDSI).⁷ These new conditions were successfully applied to the ether transfer, Table 3, to provide methoxy and benzyloxy transfer products in higher yield (entries 11 and 12) but unfortunately, at the expense of the diastereomeric ratio.

Table 3.

New Activating Conditions

	sm	conditions	yield (dr)
1	5a	BDSB, CH3NO2, −29 °C	11 75% (4:1)
2	5a	BDSB, CH3CH2NO2, −78 °C	11 72% (6.1:1)
3	5a	BDSB, CH3CN, −45 °C	11 62% (7.4:1)
4	5a	BDSB, CH3CH2CN, −78 °C	11 46% (17:1)
5	5b	BDSB, CH3NO2, −29 °C	12 71% (4.3:1)
6	5b	BDSB, CH3CH2NO2, −78 °C	7b only
7	5b	BDSB, CH3CN, −45 °C	12 52% (6.3:1)
8	5b	BDSB, CH3CH2CN, −78 °C	7b only
9	5a	IDSI, CH3NO2, −29 °C	6a 78% (3.2:1)
10	5a	IDSI, CH3CH2NO2 , −78 °C	6a 76% (3.3:1)
11	5a	IDSI, CH3CH2CN, −78 °C	6a 95% (2.8:1)
12	5b	IDSI, CH3NO2, −29 °C	6b 75% (4:1)
13	5b	IDSI, CH3CH2CN, −78 °C	6b 67% (5:1)

The oxygenation pattern found in ether transfer products such as 6a-c and 9a-c is of significant utility in the synthesis of polyketides with unique patterns of methylation such as those derived from cyanobacterial PKS biosynthetic pathways. Having accomplished the development of a reliable ether transfer protocol to stereochemically defined tertiary ethers, we next focused our efforts on an application to the synthesis of the C9-C19 fragment of lyngbyaloside B and C, Scheme 3. The epoxy alcohol 13, obtained in three steps from propargyl alcohol and 2-methyl-2-propen-1-ol,⁸ was regioselectively opened with methylmagnesium bromide in the presence of copper bromide dimethyl sulfide complex.⁹ BPS protection of the primary alcohol followed by incorporation of the transferable naphthylmethyloxymethyl group provided 15.

Scheme 3.

Application to the C9-C19 Fragment Synthesis

Upon treatment with our optimized conditions followed by methanolic quench, the MOM protected 16 was obtained in 88% yield as a 10:1 mixture of diasteromers. Mild oxidative deprotection, which selectively liberated the tertiary alcohol, was followed by coupling to allylmagnesium bromide in the presence of copper iodide. Carbon-carbon bond formation and generation of 17 likely occurs through nucleophilic addition to an in situ generated epoxide. Confirmation of the 1,3-syn-stereochemical assignment was obtained by conversion of MOM ether 17 to the corresponding methylene acetal under acidic conditions (see supporting information). Toward our natural product target, the tertiary alcohol was protected as a triethylsilyl ether and subjected to an E-selective cross-metathesis homologation. Finally, Wittig olefination with (bromomethyl) triphenylphosphonium bromide gave the C9-C19 fragment of lyngbyaloside 18 as a 1:2 mixture of E:Z isomers.

In summary we have successfully applied our ether transfer conditions to the stereoselective formation of tertiary ethers. Electrophilic activation of homoallylic alkoxymethyl ethers enabled the formation of syn-1,3-diol mono- and orthogonally protected diethers through ether transfer. As an additional advancement of our previous described method, 2-naphthylmethyl ethers have been shown to proceed efficiently through ether transfer with high stereoselectivity and are easily deprotected by DDQ oxidation. This new, stereoselective route to structural units containing tertiary ethers has found application in the synthesis of the C9-C19 fragment of lyngbyaloside B and C. Additional efforts towards the total synthesis of these cyanobacteria-derived polyketides are currently underway in our laboratories and will be reported in due course.