Synthesis of cytotoxic aza analogues of jaspine B

Article abstract of DOI:10.1021/jo1014239

A straightforward access to pyrrolidine-based analogues of jaspine B was developed. Five stereoisomers were prepared including the all-cis derivatives presenting the configuration of the natural anhydrophytosphingosine. The synthesis of the latter relied

Full text of DOI:10.1021/jo1014239

pubs.acs.org/joc
Synthesis of Cytotoxic Aza Analogues of Jaspine B
Arnaud Rives,‡ Sonia Ladeira,§ Thierry Levade,^
^
,‡
ꢀ
Nathalie Andrieu-Abadie, and Yves Genisson*
‡
ꢀ
LSPCMIB, UMR 5068, CNRS-Universite Paul Sabatier,
Toulouse, France, SFTCM, FR 2599, Universite Paul
§
ꢀ
Sabatier, Toulouse, France, and ^I2MR, INSERM U858,
ꢀ
Universite Paul Sabatier, Toulouse, France
genisson@chimie.ups-tlse.fr
Received July 27, 2010
FIGURE 1. Structures of the sphingomyelin synthase inhibitor
jaspine B (1), the pyrrolidine-based glucosylceramide synthase
inhibitor 2 and the targeted aminopyrrolidines 3.
We recently identified jaspine B (1) as a new structural
archetype for the development of sphingomyelin synthase
inhibitors (Figure 1).4 On the other hand, we designed
pyrrolidine-derived sphingosine mimics such as 2 with po-
tent glucosylceramide synthase inhibitory activity.5 Both
chemical series display a marked pro-apoptotic behavior in
melanoma cells, correlated with an increased intracellular
ceramide concentration.
In search for new sphingolipid analogues interfering with
ceramide metabolism in cancer cells, we envisioned merging
the structural frameworks of 1 and 2.6 Such structural
hybridation would lead to novel pyrrolidine-based jaspine
B derivatives 3. Altered ionization state and hydrogen-bond
donor/acceptor capacity of such analogues may indeed
modulate their pharmacological profiles. We report here
our results concerning the enantioselective syntheses of these
derivatives and their first biological evaluations.
A straightforward access to pyrrolidine-based analogues
of jaspine B was developed. Five stereoisomers were
prepared including the all-cis derivatives presenting the
configuration of the natural anhydrophytosphingosine.
The synthesis of the latter relied on an original Staudinger-
type cyclization process. The compounds were evaluated
regarding their ability to alter tumor cells’ viability and to
interfere with the metabolism of sphingolipids.
Syntheses of 2-alkyl-4-amino-3-hydroxy pyrrolidine such
as 3 are scarce.7 Our synthetic approach relied on the use of
an enantioenriched R,β-epoxyaldehyde as a precursor of the
four-carbon head unit of the targeted structures (Scheme 1).8
We planned that the all-cis aminopyrrolidines 6 could be
accessed from a (2R,3S,4S)-anti-epoxyamine 4. Early intro-
duction of the lipophilic chain with creation of the (4S) stereo-
genic center was envisioned by means of the anti-selective
alkylation of an intermediate imine.9 A carbonatation/intra-
molecular cyclization sequence would in turn invert the
configuration at C-3 while leaving a free hydroxyl group at
Sphingolipids represent a wide family of bioactive lipids.1
They are key mediators of fundamental cell-signaling pro-
cesses related to tumorigenicity and cancer progression.2 In
balance with the “tumor-promoting” sphingosine-1-phosphate,
the pro-apoptotic ceramide lies at the center of a metabolic
manifold. In cancer cells, this “tumor-suppressing lipid” is
rapidly converted into sphingomyelin (an abundant plasma
membrane component) or glucosylceramide (involved in
chemoresistance of cancer cells). Inhibiting the enzymatic
pathways responsible for this accelerated consumption of
ceramide is thus expected to enhance the susceptibility of
cancer cells to apoptosis.3
ꢀ
(5) (a) Faugeroux, V.; Genisson, Y.; Andrieu-Abadie, N.; Colie, S.;
Levade, T.; Baltas, M. Org. Biomol. Chem. 2006, 4, 4437–4439. (b) Rives,
ꢀ
(1) (a) Hirabayashi, Y.; Igarashi, Y.; Merrill, A. H. J. Sphingolipid
Biology; Springer-Verlag: Tokyo, 2006. (b) Hannun, Y. A.; Obeid, L. M.
Nat. Rev. Mol. Cell Biol. 2008, 9, 139–150.
A.; Genisson, Y.; Faugeroux, V.; Zedde, C.; Lepetit, C.; Chauvin, R.; Saffon,
N.; Andrieu-Abadie, N.; Colie, S.; Levade, T.; Baltas, M. Eur. J. Org. Chem.
2009, 2474–2489.
(2) (a) Ogretmen, B.; Hannun, Y. A. Nat. Rev. Cancer 2004, 4, 604–616.
ꢀ
(6) For a review of sphingolipids metabolism modulators: Delgado, A.;
Casas, J.; Llebaria, A.; Abad, J. L.; Fabrias, G. ChemMedchem 2007, 2, 580–
606.
ꢀ
(b) Segui, B.; Andrieu-Abadie, N.; Jaffrezou, J.-P.; Benoist, H.; Levade, T.
Biochim. Biophys. Acta 2006, 1758, 2104–2120.
(3) (a) Lin, C. F.; Chen, C. L.; Lin, Y. S. Curr. Med. Chem. 2006, 13, 1609–
1616. (b) Modrak, D. E.; Gold, D. V.; Goldenberg, D. M. Mol. Cancer Ther.
2006, 5, 200–208. (c) Morales, A.; Fernandez-Checa, J. C. Mini-Rev. Med.
Chem. 2007, 7, 371–382.
(7) Boto, A.; Hernandez, D.; Hernandez, R. Tetrahedron Lett. 2009, 50,
3974–3977.
(8) For the use of such epoxyaldehydes in the synthesis of iminosugars
see: Ayad, T.; Genisson, Y.; Baltas, M.; Gorrichon, L. Chem. Commun.
ꢀ
2003, 582–583. Ayad, T.; Genisson, Y.; Baltas, M. Org. Biomol. Chem. 2005,
(4) (a) Salma, Y.; Lafont, E.; Therville, N.; Carpentier, S.; Bonnafe, M. J.;
ꢀ
Levade, T.; Genisson, Y.; Andrieu-Abadie, N. Biochem. Pharmacol. 2009,
78, 477–485. (b) Salma, Y.; Ballereau, S.; Ladeira, S.; Andrieu-Abadie, N.;
3, 2626–2631.
(9) For the use of a related vinylation in the synthesis of iminosugars see:
ꢀ
Genisson, Y. Org. Biomol. Chem. 2010, 8, 3227–3243.
ꢀ
Ayad, T.; Genisson, Y.; Baltas, M.; Gorrichon, L. Synlett 2001, 866–868.
7920 J. Org. Chem. 2010, 75, 7920–7923
Published on Web 10/18/2010
DOI: 10.1021/jo1014239
r
2010 American Chemical Society

Rives et al.
JOCNote
SCHEME 1. General Synthetic Approach to all-cis and all-trans
Pyrrolidine-Based Jaspine B Analogues 6 and 8
SCHEME 2. Synthesis of Aminopyrrolidines 22 and 23
C-2 (route a).10 Introduction of a nitrogenated group with
inversion of configuration in this position, leading to 5,
would set the stage for the construction of the all-cis pyrro-
lidine ring of 6. Variation of this route via epimerization at
C-2 would also give ready access to trans-cis aminopyrroli-
dine derivatives (vide infra). On the other hand, preparation
of all-trans analogues 8 would rely on the regioselective oxirane
opening of epoxypyrrolidine intermediates 7 (route b).11 Finally,
preparation of another set of steroisomers was planned
starting from enantiomeric epoxide precursors.
SCHEME 3. Synthesis of Aminopyrrolidine ent-23
The assembling of the C18 skeleton thus relied on the
preparation of epoxyamine 10 (Scheme 2). The (2R,3R)-cis-
epoxyaldehyde 910 was first treated with benzylamine, and
the resulting imine was reacted without isolation with tetra-
decyl magnesium bromide in the presence of Et2OBF3. The
unstable anti-epoxyamine 10 was isolated as a single diatereo-
isomer in 70% yield. Treatment of the latter with ammonium
carbonate smoothly delivered oxazolidinone 11 in good yield.
At this stage, the choice of the nitrogenated group was decisive.
The presence of a nucleophilic amine was anticipated to interfere
with the final cyclization process while the phtalimide moiety
proved labile under preliminary saponification attempts. The
use of an azido group was thus considered. Azide 14 was
prepared in good yield from secondary alcohol 11 via the
corresponding triflate. Desilylation of the primary hydroxyl
group followed by carefully monitored saponification led to
the linear azido aminoalcohol 18 in 74% yield (based on the
starting material recovering, bsmr, at 63% of conversion).
Appel12 and Mitsunobu reactions13 are two complemen-
tary processes to prepare hydroxylated pyrrolidine from
linear aminopolyols. Fast trapping of the Ph3P by the vicinal
azido group was however expected here. We thus thought of
exploiting the 1,2-azidoalcohol moiety in 18 to activate the
primary hydroxyl group. After some optimization, it was found
that treatment of 18 with one equivalent of Ph3P at 50 ꢀC in
THF gratifyingly delivered the corresponding amino pyrro-
lidine 20 in 78% yield (Scheme 4).14 Finally, hydrogenolysis
of the N-benzyl group gave the targeted all-cis (2S,3S,4S)
jaspine B analogue 22 (Scheme 2).
We also used this route to prepare trans-cis analogues 23
(Scheme 2). The secondary hydroxyl group in 11 was epi-
merized to give 13. The diastereoisomeric azido amino-
alcohol 15 obtained after desilylation and saponification of 13
was treated with Ph3P under the previous conditions to give
the desired aminopyrrolidine in 87% yield. The trans-cis
(2R,3S,4S) pyrrolidine-based jaspine B analogue 23 was
obtained after a final hydrogenation step. In order to further
illustrate the flexibility of our approach, we targeted the
trans,cis pyrrolidine enantiomer of 23 from the trans-epoxide
precursor 24,10 thus avoiding the epimerization step (Scheme 3).
ꢀ
(10) Ayad, T.; Faugeroux, V.; Genisson, Y.; Andre, C.; Baltas, M.;
Gorrichon, L. J. Org. Chem. 2004, 69, 8775–8779.
(11) For an in-depth study of this oxirane opening reaction see ref 5b.
(12) (a) Appel, R. Angew. Chem., Int. Ed. 1975, 14, 801–811. (b) Casiraghi,
G.; Rassu, G.; Spanu, P.; Pinna, L.; Ulgheri, F. J. Org. Chem. 1993, 58, 3397–
3400. (c) Razavi, H.; Polt, R. J. Org. Chem. 2000, 65, 5693–5706.
(14) The structural assignment of the cyclisation products as pyrolidines
was based on NMR analyses. HMBC experiments on 21, the N-Bn precursor
of 23, showed a 3J correlation between the benzylic methylene protons and
the C-1 (sphingolipid numbering), thus indicating an N-C-1 connection.
€
(13) (a) Veith, U.; Schwardt, O.; Jager, V. Synlett 1996, 1181–1183.
(b) Schwardt, O.; Veith, U.; Gaspard, C.; Jager, V. Synthesis-Stuttgart
1999, 1473–1490.
J. Org. Chem. Vol. 75, No. 22, 2010 7921

Rives et al.
JOCNote
TABLE 1. Cytotoxicity of 22, 23, ent-23, 31 and ent-31 toward Cancer Cells (MTT Experiments, see Supporting Information)
a1.1 μM for natural jaspine B. b4.7 μM for natural jaspine B. c2.4 μM for natural jaspine B.
SCHEME 4. Proposed Mechanism for the Staudinger-Type
Cyclization/Reduction Process
SCHEME 5. Synthesis of all-trans Aminopyrrolidine 31
The Grignard reagent addition onto the intermediate epoxy-
imine delivered the expected anti-epoxyamine 25 in 53%
isolated yield (anti/syn 80:20, based on the 1H NMR of the
crude reaction mixture). The oxazolidinone ent-13 was then
directly obtained by mean of the carbonatation/intramole-
cular cyclization sequence. The rest of the sequence leading
to the jaspine B analogue ent-23 paralleled the one used for
the preparation of its enantiomer. The X-ray diffraction
analysis of a crystalline sample of the oxazolidinone ent-17
allowed us to assign the final (2S,3R,4R) configuration in
this series (see Supporting Information).
The synthesis of the three aminopyrrolidines 22, 23 and
ent-23 thus relied on an original Ph3P-promoted cyclization/
reduction of azido aminoalcohols to aminohydroxy pyrroli-
dines. A Staudinger-type mechanism may be proposed for
this transformation (Scheme 4).15 From 18, the formation of
iminophosphorane 26 would lead to oxazophospholidine 27
through intramolecular trapping by the vicinal primary
alcohol. Such a pathway is reminiscent to the phosphine-
promoted aziridine formation introduced by Zwanenburg
and coll..16 However, the presence of the secondary amine
would allow here formation of the pyrrolidine ring, concomi-
tant with the reduction of the azido group and liberation
of Ph3PO, yielding 20.17 Noteworthy, this transformation
offers a direct entry to the important 3-amino-4-hydroxy-
pyrrolidine framework present in several 5-membered ring
iminosugars including potent inhibitors of the lysosomal
catabolism of glycosphingolipids.18
The epoxypyrrolidine intermediate 29 required for the prepa-
ration of the all-trans aminopyrrolidines was obtained in
72% yield through desilylation of 10 and Appel cyclization
(Scheme 5). We then took advantage of a regio- and stereo-
controlled oxirane-opening reaction to introduce the nitro-
genated group in the correct location and configuration.
Thus, treatment of the epoxypyrrolidine 29 with benzyl-
amine in methoxyethanol/water at 65 ꢀC smoothly led to
the formation of aminopyrrolidine 30, isolated in 77% yield
as a single isomer. A final hydrogenolysis delivered the
targeted (2S,3R,4S) jaspine B analogue 31.
The enantiomeric (2R,3S,4R) jaspine B ent-31 analogue
was obtained through a similar sequence starting from the
(2S,3S)-cis-epoxyaldehyde ent-9. Crystallographic analysis
of the epoxypyrrolidine ent-29 allowed confirmation of its
structure (see Supporting Information).
The ability of aminopyrrolidines 22, 23, ent-23, 31, and
ent-31 to inhibit tumor cells growth was evaluated in various
melanoma cell lines (Table 1). All compounds showed a
dose-dependent effect with IC50 in the low micromolar
range. In murine B16 cells, the pyrrolidine 22, presenting
the all-cis configuration of the parent natural compound, as
well as the all-trans derivatives 31 and ent-31 displayed an
IC50 inferior to 4 μM. This trend was also observed in human
A375 and WM115 cells were the aza analogues 22, 31 and
ent-31 proved as cytotoxic as the natural jaspine B with IC50
values around 2.5 μM.
To determine whether the cytotoxic aminopyrrolidines 22
and 31 affect sphingomyelin synthesis, as we previously
reported for the natural jaspine B, we next evaluated the
changes in the sphingolipid pattern in B16 cells treated for 6 h
by jaspine B aza analogues.4a We observed that 22 and 31
elicited an increase in [3H]sphingosine-labeled ceramide content,
whereas the level of radiolabeled sphingomyelin decreased
as compared to untreated control cells (see Supporting
(15) For a review on the Saudinger reaction see: Gololobov, Y. G.;
Kasukhin, L. F. Tetrahedron 1992, 48, 1353–1406.
(16) (a) Legters, J.; Thijs, L.; Zwanenburg, B. Tetrahedron Lett. 1989, 30,
4881–4884. (b) Thijs, L.; Porskamp, J. J. M.; Vanloon, A. A. W. M.; Derks,
M. P. W.; Feenstra, R. W.; Legters, J.; Zwanenburg, B. Tetrahedron 1990, 46,
2611–2622. (c) Legters, J.; Thijs, L.; Zwanenburg, B. Recl. Trav. Chim. Pays-Bas
1992, 111, 1–15. For a related process see: (d) Chun, J.; He, L. L.; Byun, H. S.;
Bittman, R. J. Org. Chem. 2000, 65, 7634–7640.
(17) The generation of Ph3PO was monitored by 31P NMR with the
appearance of a signal at δ = 29.3 ppm in CDCl3.
(18) For recent studies see: (a) Rountree, J. S. S.; Butters, T. D.; Wormald,
M. R.; Dwek, R. A.; Asano, N.; Ikeda, K.; Evinson, E. L.; Nash, R. J.; Fleet,
G. W. J. Tetrahedron Lett. 2007, 48, 4287–4291. (b) Choubdar, N.; Bhat,
R. G.; Stubbs, K. A.; Yuzwa, S.; Pinto, B. M. Carbohydr. Res. 2008, 343,
1766–1777. (c) Curtis, K. L.; Evinson, E. L.; Handa, S.; Singh, K. Org.
Biomol. Chem. 2007, 5, 3544–3553. (d) Rountree, J. S. S.; Butters, T. D.;
Wormald, M. R.; Boomkamp, S. D.; Dwek, R. A.; Asano, N.; Ikeda, K.;
Evinson, E. L.; Nash, R. J.; Fleet, G. W. J. ChemMedChem 2009, 4, 378–392.
7922 J. Org. Chem. Vol. 75, No. 22, 2010

Rives et al.
JOCNote
Information). These effects were comparable with those
produced by the natural jaspine B.
(m, 2H), 2.59 (dd, 2J = 10.2 Hz, 3J = 3.6 Hz, 1H), 3.10 (d, 2J =
13.2 Hz, 1H), 3.27-3.32 (m, 1H), 3.92 (pseudot, 3J ≈ 3J ≈
5.4 Hz, 1H), 4.00 (d, 2J = 13.2 Hz, 1H), 7.18-7.30 (m, 5H); 13
C
In conclusion, C-alkyl aminopyrrolidines embedding a
sphingolipid-like C18 carbon skeleton were designed as aza
analogues of jaspine B. Five stereoisomers were prepared
and an original Ph3P-promoted cyclization/reduction of
azido aminoalcohols to aminohydroxypyrrolidines was de-
veloped. All compounds displayed cytotoxicity against va-
rious melanoma cells with a potency comparable to that of the
parent natural compound for the all-cis compound 22 and
the all-trans derivatives 31 and ent-31. The active aza ana-
logues 22 and 31 were shown to impair the conversion of
ceramide into sphingomyelin in B16 cells, as the natural
jaspine B.
NMR (75 MHz, CDCl3) δ 14.1, 22.6, 26.6, 28.1, 29.3-30.1 (8C),
30.1, 31.9, 51.4, 57.9, 60.1, 68.2, 71.6, 126.9, 128.1, 128.7, 139.1;
MS (ES): m/z = 389 (100) [M þ Hþ]; HRMS (ESIþ): calcd for
C25H45N2O 389.3532, found. 389.3518.
(2S,3S,4S)-4-amino-2-tetradecylpyrrolidin-3-ol (22). Twenty
percent Pd(OH)2/C (11.1 mg, 23% w/w) and 12 N HCl aqueous
solution (1-2 drops) were successively added to a solution of N-
benzylpyrrolidine 20 (47.1 mg, 0.12 mmol) in MeOH (1.2 mL).
The flask was purged with N2 and then loaded with H2 (10-12
bar). The mixture was stirred at room temperature until disap-
pearance of the starting material (24-90 h). The catalyst was
then removed by filtration through Celite and the filtrate evapo-
rated to dryness. The residue was taken up in 2:1 MeOH/water
(3 mL) and Dowex 50WX8-200 ion-exchange resin (1.44 g) was
added. After being stirred for 1 h, the resin was successively
filtered and washed with water and MeOH. A 3 M ammonium
hydroxide solution was then added (6 mL) and the suspension
was stirred for 1 h before being filtered and rinsed with a 3 M
ammonium hydroxide solution (60 mL). The resulting solution
was evaporated to dryness under reduced pressure. The crude
product was purified by flash column chromatography on silica
gel (MeOH/EtOH/NH4OH/CH2Cl2, 6:12:6:76) to give 22 (24.6 mg,
Experimental Section
(2S,3R,4S)-2-Azido-4-(benzylamino)octadecane-1,3-diol (18).
NaOH (68.0 mg, 1.7 mmol) was added to a solution of oxazoli-
dinone 16 (76.7 mg, 0.17 mmol) in a 8:2 EtOH/H2O mixture
(3.5 mL). The reaction mixture was heated at 85 ꢀC for 6 h before
being cooled to room temperature. The mixture was extracted
three times with EtOAc and the combined extracts were concen-
trated in vacuo. The crude product was purified by flash column
chromatography on silica gel (PE/EtOAc 80:20, 0.8% NH4OH)
togive 18 (49.5mg, 74% bsmrat63% of conversion) as acolorless
25
70%) as a white amorphous solid. [R]D þ18.8 (c 1.3; MeOH);
IR (neat) 3321 (O-H), 1601 (N-H) cm-1; 1H NMR (300 MHz,
CD3OD) δ 0.90 (t, 3J = 6.7 Hz, 3H), 1.25-1.44 (m, 24H),
1.47-1.68 (m, 2H), 2.63 (dd, 2J = 10.8 Hz, 3J = 9.2 Hz, 1H),
2.94 (td, 3J = 7.0 Hz, 3J = 2.8 Hz, 1H), 3.06 (dd, 2J = 10.8 Hz,
25
oil. [R]D þ25.1 (c 1.1; CHCl3); IR (neat) 3436 (O-H), 2099 (N3)
cm-1; 1H NMR (300 MHz, CDCl3) δ 0.88 (t, 3J = 6.7 Hz, 3H),
1.20-1.40 (m, 24H), 1.54-1.67 (m, 2H), 2.76 (td, 3J = 6.4 Hz,
3J = 2.5 Hz, 1H), 3.05 (brs, 2H), 3.35 (pseudoq, 3J ≈ J ≈ J ≈ 5.7
3
3
3J = 8.6 Hz, 1H), 3.35 (pseudotd, 3J ≈ J ≈ 8.8 Hz, 3J = 4.0 Hz,
3
3
3
1H), 3.79 (pseudotd, 3J ≈ J ≈ 9.0 Hz, 1H); 13C NMR (75 MHz,
3
Hz, 1H), 3.54 (dd, J = 6.4 Hz, J = 2.6 Hz, 1H), 3.81 (AB
system, 2J = 12.6 Hz, δa - δb = 47.4 Hz, 2H), 3.86 (AB part of
an ABX, 2J = 11.6 Hz, 3J = 5.9 Hz, 3J = 4.7 Hz, δa-δb = 45.6
Hz, 2H), 7.25-7.36 (m, 5H); 13C NMR (75 MHz, CDCl3) δ 14.1,
22.7, 26.1, 29.3-29.7 (9C), 31.9, 32.2, 52.3, 57.7, 61.9, 65.2,
72.6, 127.5, 128.4, 128.6, 139.1; MS (ES): m/z = 433 (100)
[M þ Hþ]. HRMS (ESIþ): calcd forC25H45N4O2 433.3543, found
433.3547.
CD3OD) δ 14.5, 23.8, 28.1, 30.6-30.9 (10C), 31.1, 33.1, 51.6,
56.7, 64.0, 73.8; MS (ES): m/z = 299 (100) [M þ Hþ]; HRMS
(ESIþ): calcd for C18H39N2O 299.3062, found 299.3054.
Acknowledgment. ThePierreFabregroupandtheCERPER
are gratefully acknowledged for their support and a PhD Grant
to A.R. We are grateful to Dr. C. Debitus for a gracious gift of
ꢀ
Jaspis sp. sponge extracts. We thank Dr. Stephanie Ballereau for
insightful discussions about the cyclization process.
(2S,3S,4S)-4-Amino-1-benzyl-2-tetradecylpyrrolidin-3-ol (20).
Triphenylphosphine (42.0 mg, 0.16 mmol) was added to a
solution of azidoalcohol 18 (67.2 mg, 0.16 mmol) in anhydrous
THF (1.6 mL). The mixture was allowed to react at 50 ꢀC for
16 h. The THF was then evaporated in vacuo. The crude product
was purified by flash column chromatography on silica gel (EtOAc/
MeOH, 95:5; 1.6% NH4OH) to give 20 (47.1 mg, 78%) as a
Supporting Information Available: Experimental proce-
dures and characterization data for all new compounds; copies
of 13C NMR spectra for 10-23, 25; crystallographic data for
ent-17 and ent-29; experimental procedures and graphs for
biological evaluations. This material is available free of charge
via the Internet at http://pubs.acs.org.
25
colorless oil. [R]D þ65.8 (c 1.8; CHCl3); IR (neat) 3430 (O-H),
1604 (N-H) cm-1; 1H NMR (300 MHz, CDCl3) δ 0.86 (t, 3J =
6.7 Hz, 3H), 1.18-1.38 (m, 24H), 1.47-1.69 (m, 2H), 2.29-2.40
J. Org. Chem. Vol. 75, No. 22, 2010 7923