Syphonoside, a Unique Macrocyclic Glycoterpenoid
) 0.13, CHCl3); IR (liquid film): 3390, 2930, 1725 cm-1; UV
(MeOH): λmax 206 (ꢀ ) 6700); selected 1H NMR values (300 MHz,
CD3OD): δ 6.57 (1H, t, J ) 3 Hz, H-3), 5.61 (1H, app. t, J ) 7
Hz, H-14), 4.36 (1H, d, J ) 8 Hz, H-1′′), 1.77 (3H, bs, H3-16),
1.33 (3H, s, H3-19), 0.99 (3H, d, J ) 6 Hz, H3-17), 0.90 (3H, s,
H3-20); HRESIMS (m/z): (M + Na)+ calcd for C27H44O9Na,
535.2883; found, 535.2859.
containing an ent-clerodane diterpenoid moiety, two D-glucose
units, and a 3-hydroxy-3-methylglutaric residue. This structure
displays an original conformation exhibiting two long nine-atom
extended strands joined by two “cyclohexane-chairlike” two-
atom junctions. The resulting three-dimensional shape of this
22-membered ring is a moderately curved turtle-shell structure
characterized by an inhomogeneous distribution of polar vs
hydrophobic regions. In fact, a prevalently polar rim (also
including the perpendicular extracyclic glucose moiety) sur-
rounds a mostly hydrophobic central region, also including a
cavity that could easily host either an all-hydrophobic group or
a small hydrophobic chain ending with a H-bond donor group
(pointing toward O1′ atom). This cavity, together with the two
long, almost parallel eight-residue strands, could represent
interaction sites with other biomolecules and, at the same time,
provide potential technological applications. In a preliminary
cytotoxicity test, syphonoside (1) did not exhibit any cytotoxic
activity against the employed cell lines but was able to inhibit
high density induced apoptosis. Therefore, this compound may
play an important role in regulating cell survival and cell death
under specific conditions but additional studies are required for
the elucidation of this role.
The aqueous phase from the workup was filtered on a Sep-Pak
and dried to afford pure D-glucose [6.0 mg, Rf ) 0.4 (70:20:10,
EtOAc:MeOH:H2O); [R]D 45.4 (c ) 0.37, H2O)]. The absolute
1
stereochemistry of D-glucose was determined by analysis of H
NMR and CD spectra of the corresponding benzoate derivatives.14
Compound 3. Glycosyl terpene 2 (2.0 mg) was dissolved in
dry pyridine (2 mL), acetic anhydride (0.5 mL) was added to the
resulting solution, and the reaction mixture was stirred overnight
at room temperature. Pyridine was removed in Vacuo, and the
residue was dissolved in diethyl ether and worked up as usual to
afford pure 3 (1.8 mg), Rf ) 0.7 (95:5, CHCl3:MeOH); [R]D -41.4
(c ) 0.1, CHCl3); IR (liquid film): 2940, 1758, 1232 cm-1; UV
1
(MeOH): λmax 206 (ꢀ ) 5250); H NMR (400 MHz, CDCl3): δ
6.52 (1H, app. t, J ) 3 Hz, H-3), 5.41 (1H, app. t, J ) 7 Hz, H-14),
5.16 (1H, app. t, J ) 9 Hz, H-3′′′), 5.00 (1H, app. t, J ) 9 Hz,
H-4′′′), 4.93 (1H, dd, J ) 8, 9 Hz, H-2′′′), 4.62 (1H, d, J ) 8 Hz,
H-1′′′), 4.75 (1H, m, H-15a), 4.46 (1H, m, H-15b), 4.21 (1H, m,
H-6′′′a), 4.13 (1H, m, H-6′′′b), 4.09 (1H, m, H-11), 3.68 (3H, s,
OMe), 3.61 (1H, m, H-5′′′), 2.40 (1H, m, H-12a), 2.22 (1H, m,
H-6a), 2.20 (1H, m, H-12b), 2.17 (1H, m, H-2a), 2.12 (3H, s, OAc),
2.10 (1H, m, H-2b), 2.07 (3H, s, OAc), 2.04 (3H, s, OAc), 2.01
(3H, s, OAc), 1.98 (3H, s, OAc), 1.75 (3H, bs, H3-16), 1.69 (1H,
m, H-1a), 1.57 (1H, m, H-1b), 1.55 (1H, m, H-8), 1.38 (2H, m,
H2-7), 1.31 (1H, m, H-10), 1.30 (3H, s, H3-19), 1.08 (1H, m, H-6b),
0.84 (3H, s, H3-20), 0.81 (3H, d, J ) 7 Hz, H3-17); 13C NMR (300
MHz, CDCl3): δ 171.4 (s, OAc), 170.3 (s, 2C, OAc), 169.4 (s,
2C, OAc), 167.8 (s, C-18), 142.8 (s, C-4), 139.4 (s, C-13), 136.1
(d, C-3), 121.6 (d, C-14), 98.8 (d, C-1′′′), 77.4 (d, C-11), 73.1 (d,
C-3′′′), 71.9 (d, C-2′′′), 71.4 (d, C-5′′′), 69.2 (d, C-4′′′), 62.4 (t,
C-6′′′), 61.3 (t, C-15), 51.2 (q, OMe), 46.6 (d, C-10), 43.5 (s, C-9),
42.5 (t, C-12), 38.0 (s, C-5), 35.7 (t, C-6), 27.9 (t, C-7), 27.2 (t,
C-2), 21.1 (2C, q, C-19 and OAc), 20.6 (4C, q, OAc), 18.8 (t, C-1),
17.5 (d, C-17), 16.2 (q, C-16), 12.5 (q, C-20); HRESIMS (m/ z):
(M + Na)+ calcd for C37H54O14Na, 745.3411; found, 745.3430.
Experimental Section
General experimental procedures were performed as previously
described.9
Collection and Extraction. The collection and the extraction
of S. geographica individuals (seven animals) has already been
reported.9 A sample of H. stipulacea was collected from the same
site as the mollusk in November 2003 and was frozen at -20 °C.
This material (dry residue 103.0 g) was extracted according to the
standard procedures.9 The n-BuOH layer from the acetone extract
was evaporated under reduced pressure to afford 1.31 g of crude
extract. Analogously, the n-BuOH layer of the digestive gland
extract of the mollusk was evaporated to dryness, yielding 618.0
mg of residue. Both extracts were analyzed by TLC (chloroform/
methanol in variable ratios) and compared.
Isolation of 1. The n-BuOH extract (618 mg) of the mollusk
was subjected to Sephadex LH-20 chromatography using a mixture
of chloroform/methanol in 1:1 ratio as eluent, to yield a fraction
(115 mg) which was further purified on preparative TLC (SiO2,
7:3, CHCl3:MeOH) to give 49.2 mg of pure metabolite 1 (ca. 8%
of the extract). An aliquot (500 mg) of the n-BuOH soluble part of
the extract of the sea-grass was chromatographed under the above-
described conditions to give18.8 mg of pure 1 (ca. 4% of the
extract).
Acid Methanolysis of 2. Compound 2 (10 mg) was dissolved
in a 1 N HCl solution in MeOH (2 mL), and the obtained solution
was stirred for 12 h at 40 °C. After the usual workup, the reaction
mixture was dried and partitioned between CHCl3 and H2O/MeOH,
8:2. The aqueous layer was concentrated, filtered on a Sep-Pak
cartridge, and dried. The 1H NMR spectrum of this fraction
indicated that it was a mixture of R-and â-methyl glucopyranose,
the absolute stereochemistry of which was determined by analysis
Syphonoside (1). Rf ) 0.50 (7:3, CHCl3:MeOH); [R]D -7.0 (c
) 0.54, MeOH); CD (EtOH) [θ]209 2754, [θ]237 -8574; IR (liquid
film): 3500, 2887, 1714 cm-1; UV (MeOH): λmax 206 (ꢀ )
1
11 660); H and 13C NMR in Table 1; HRESIMS (m/z): (M +
1
Na)+ calcd for C38H58O17Na, 809.3572; found, 809.3545.
of H NMR and CD spectra of the corresponding tetra-benzoate
derivatives.14
Alkaline Methanolysis of 1. A 25 mg amount of 1 was dissolved
in a 0.5 M solution of MeONa in MeOH (2.5 mL), and the resulting
mixture was stirred at room temperature for 1.5 h. The reaction
mixture was first chromatographed on a DOWEX-50W column
eluted with methanol (100 mL) and then extracted with Et2O to
afford 15.0 mg and 4.5 mg of residue from the ethereal and aqueous
phases, respectively. An aliquot of the ethereal phase was analyzed
by HPLC (Phenomenex, Kromasil 5 µm 100A C18; eluent CH3-
CN/H2O 4:6; flow 1 mL/min). The dimethyl ester of 3-hydroxy-
3-methylglutaric acid was identified by comparison of its retention
time (tR 4 min) with that of an authentic sample. An additional
LC-ESIMS analysis was conducted confirming the identity of
3-hydroxy-3-methylglutaric dimethyl ester. The remaining ether
soluble residue was purified on preparative TLC (SiO2, 9:1, CHCl3:
MeOH) to give 12.0 mg of compound 2 (Rf ) 0.4); [R]D -4.8 (c
Conformational Analysis of 1. Syphonoside structures were
obtained by restrained SA/EM.15 Experimental NOE intensities were
converted into 22 proton-proton distance restraints classified into
two ranges: 1.7-3.0 Å and 1.7-4.0 Å, corresponding to strong/
strong-medium and medium/medium-weak NOE peaks, respec-
tively. Weak and very weak interactions, as well as peaks
corresponding to fixed distances in glucose rings, were not included.
Methylene protons in the NMR spectra of syphonoside were not
resolved, but they were described using an “ambiguous restraint”
approach in which, for each restraint, all the possible distances
between the two methylene protons and their NOE counterpart
proton(s) were calculated and weighted by the inverse of the sixth
power of their values. Calculations were performed with Sander
J. Org. Chem, Vol. 72, No. 15, 2007 5629