Stereoselective synthesis of N,O,O,O -tetraacetyl- D - ribo-phytosphingosine, N,O,O -triacetyl- D - erythro -sphingosine and N,O,O - triacetyl sphingonine from a common chiral intermediate derived from D -mannitol

An efficient protocol for the stereoselective synthesis of tetraacetyl-D - ribo -phytosphingosine, triacetyl-D - erythro -sphingosine and triacetyl sphinganine has been devised from a common chiral intermediate derived from commercially available D -mannitol. The key steps involved are Sharpless epoxidation, Miyashita C(2) selective endo mode azide opening of 2,3-epoxy alcohol, and selective E-Wittig olefination.


Introduction
Sphingoid bases are long-chain aliphatic compounds typically possessing 2-amino-1,3-diol and 2-amino-1,3,4-triol functionality (Figure 1). 1 They are the structural backbone of sphingolipids (cerebrosides, sphingomyelins, gangliosides and ceramides), which are an important membrane constituents of eukaryotic cells, plasma membranes and intracellular organelles, and responsible for many physiological processes including cell growth, adhesion, differentiation and also play a prominent role in cell signaling. 2Apart from this, many sphingolipids from marine organisms display pronounced antifungal, antitumor, antiviral, immunostimulatory, neuritogentic, antidiabetic, cytotoxic, and protein kinase inhibitor activities. 3mong the naturally occurring sphingoid bases D-erythro-sphingosine 2 is the first isolated compound from human brain by Thudichum in 1884. 4In addition to 2, several sphingoid bases were isolated, among them C18-phytosphingosines are the more biologically active sphingoid bases have that been isolated from plants, yeast, fungi, marine organisms 5 and mammalian tissues 6 such as brain, hair, intestine, uterus, liver, skin and blood plasma.D-ribo-Phytosphingosine (1) is the most frequently occurring phytosphingosine in nature and has been shown to play an important role as a potential heat stress signal in yeast cells 7 and as a cytotoxic agent against human leukemic cell lines 8 Furthermore, D-erythro-sphingosine 9 and D-ribophytosphingosines 10 are essential part of more complex bioactive molecules such as GalCer (4) and KRN7000 (5) respectively.Sphingolipids are available only in limited amounts from natural sources, and their isolation and purification from natural sources is expensive and difficult task.Because of the interesting biological properties of sphingolipids, there is growing interest in developing efficient methods for their synthesis. 1,11Although, many methods for the synthesis of sphingolipids have been reported, a simple and straight forward synthesis from inexpensive starting material with high level of stereocontrol is always in demand.In continuation of our efforts on natural product synthesis 12 and development of new methodologies 13 we herein, disclose a simple and convenient new approach for the asymmetric synthesis of title compounds 1, 2, and 3 from Dmannitol (6), involving the steps with high stereocontrol approach.

Results and Discussion
Our retrosynthetic analysis for target compounds 1, 2, and 3 is depicted in Scheme 1. Retrosynthetically, it was envisioned that, all the three titled sphingolipids 1, 2 and 3 to be obtained from a common intermediate 11, could be acquired by regioselective epoxide opening with azide nucleophile followed by protection of 1,3-diol as their benzylethers and consequent deprotection of cyclohexylidene group of epoxy alcohol 8 derived from the D-mannitol (6).Scheme 1. Retrosynthetic plan for the synthesis of targeted sphingolipids.
According to the above retrosynthetic analysis, the first target is to be synthesis of diastereomerically pure azido diol intermediate 11 (Scheme 2).Towards that object, the allylic alcohol 7, which was easily prepared from D-mannitol according to known procedures, 14 was subjected to Sharpless catalytic asymmetric epoxidation 15 using diethyl D-tartrate, Ti(O i Pr)4 and cumene hydroperoxide to afford epoxy alcohol 8 in 95% yield with high diastereoselectivity, (98% de, determined by NMR and GC-MS analysis).The next crucial step in our strategy is C(2) regioselective opening of the epoxy alcohol 8 by azide nucleophile.This was accomplished by using NaN3-(CH3O)3B system developed by Miyashita and co-workers. 16This reaction proceeds via an intramolecular boron chelate through a novel endo-mode epoxide opening with extremely high C(2) selectivity.The same thing was observed in our case.Under Miyashita conditions, the azide nucleophile selectively opening epoxy alcohol 8 at C(2) rather than C(3) position, which is sterically hindered by neighbouring protected vicinal diol moiety (>95% de, based on crude 1 H-NMR analysis).Furthermore, we have treated the crude product with NaIO4 to remove the C(3) opened compound easily in the form of aldehyde by the column chromatography.Gratifyingly, the desired azido diol 9 was isolated in 96% yield, and as a single diastereomer after purification.Benzyl ether protection of azido diol 9 followed by selective deprotection of cyclohexylidene group by treating with 10% HCl in CH3CN gave the desired intermediate 4-azido 1,2-diol 11 in 93% yield.After successful synthesis of the key intermediate 11, next we turned our attention towards the synthesis of D-ribo-phytosphingosine (1) from intermediate 11 (Scheme 3).Toward this objective, the primary alcohol of 11 was selectively protected as TBS ether and the secondary alcohol group was protected as benzyl ether to afford compound 13.At this juncture, it becomes necessary to free the compound 13 from the TBS group and it was removed by stirring a solution of 13 in dry THF in the presence of n-Bu4NF (TBAF), gave the compound 14 in 96% yield.The primary alcohol functionality of 14 was converted to the corresponding aldehyde moiety using 2-Iodoxybenzoic acid (IBX), and the obtained aldehyde was rather labile to column purification and therefore it was quickly subjected to Wittig olefination 17 using n-C13H27P + Ph3Br -ylide in presence of n-BuLi to afford alkene 15 as a mixture of trans and cis isomers (E/Z, 19:1; determined by 1 H-NMR spectra) in 81% yield.The geometrical isomer ratio is no relevance to the planned synthetic sequence as the double bond will be reduced in the next step.
Having the desired compound at the penultimate stage, attention was focused on the final deprotection and reduction step.One-pot reduction of azide group, saturation of double bond, and deprotection of the benzyl ethers was carried out by hydrogenation using Pd(OH)2/C on 15 affording the crude residue of target molecule 1 which was difficult to purify by column chromatography and was therefore converted it into its acetyl derivative using Ac2O and pyridine to afford tetraacetyl D-ribo-phytosphingosine (1a) in good yield.The analytical and spectroscopic data of this compound is in good agreement with the reported data.After successful synthesis of compound 1, then we turned our attention towards the synthesis of compounds 2 and 3.As per the Scheme 4, oxidative cleavage of the diol 11 with NaIO4 yielded corresponding aldehyde and without column purification it was subjected to the Wittig olefination 17 by using n-C14H29P + Ph3Br -ylide in the presence n-BuLi to afford alkene 16 as a inseparable mixture of trans and cis isomers (E/Z, 20:1, determined by 1 H-NMR spectra).The selective reduction of azide group of the compound 16 with Lindlar's catalyst 19 under hydrogen atmosphere provided the amine 17 in 83% yield.Deprotection of the benzyl groups in 17 by means of Birch reduction 20a gave the title compound D-erythro-sphingosine (2).For analytical purpose, compound 2 was acetylated with acetic anhydride in presence of pyridine and catalytic amount of DMAP to obtain triacetyl D-erythro-sphingosine (2a) in good yield.The analytical and spectroscopic data of both 2 and 2a were in good agreement with the reported data.9a-c,18b, 20 (a) i. NaIO4, aq.CH3CN (60%), rt, 30 min; ii.n-C14H29P + Ph3Br The third desired sphingoid base, sphinganine (3) was synthesized (Scheme 5) from compound 16 in a single step through the one-pot reduction of azide, hydrogenation of double bond, and deprotection of the benzyl ethers by catalytic hydrogenation using Pearlman's catalyst.For analytical reasons, compound 3 was acetylated with acetic anhydride in presence of pyridine and DMAP (catalytic) to obtain N,O,O-triacetyl sphinganine (3a) in good yield.The analytical and spectroscopic data of 3 and 3a were in good agreement with the reported data of the respective natural product.2) selective endo mode azide opening of epoxy alcohol, and preferential E-Wittig olefination were effectively utilized in accomplishing the synthesis.We believed that the key intermediate reported in this paper serves as a good synthon for making of other natural products.

Experimental Section
General.The solvents were dried over standard drying agents and freshly distilled prior to use.The reagents were purchased from Aldrich and Lancaster, and were used without further purification unless otherwise stated.All moistrure-sensitive reactions were carried out under N2 atmosphere.Column chromatography: silica gel (SiO 2 ; Acme's 60-120 mesh).Optical rotations: Perkin-Elmer P241 polarimeter and JASCO DIP-360 digital polarimeter at 25 o C, IR spectra: Perkin-Elmer IR-683 spectrophotometer.NMR: Recorded on Varian Gemini 200 or Bruker Avance 300 or Varian Unity 400 MHz spectrometer depends on their availability, using TMS as an internal standard for 1 H NMR, and CDCl 3 for 13 C NMR (chemical shift values in δ, J in Hz).MS: Recorded either on Thermo-Finnigan MAT1020B or Micromass 7070H spectrometer operating at 70 eV using direct inlet system.All high resolution mass spectra (HRMS) were recorded on QSTAR XL hybrid MS/MS system equipped with an ESI source.GC-MS were recorded on Agilent 6890 series GC-MS system, GC (Agilent Technologies, Palo Alto, CA) equipped with a model 5973N mass selective detector and a HP-5MS capillary column (5% phenyl, 95% PDMS, 30 m x 0.25 mm i.d.x 0.25 µm film thickness).