Synthesis of squalamine utilizing a readily accessible spermidine equivalent

Full text of DOI:10.1021/jo981344z

J . Org. Chem. 1998, 63, 8599-8603
8599
The published syntheses of squalamine were not suit-
able for this purpose. The chemical synthesis of squal-
amine has been demonstrated in 17 steps from both 3â-
acetoxy-5-cholenic acid12 and 3â-hydroxy-5-cholenic acid.13
The published methods, however, afforded a mixture of
diastereomers at C-24 and required an expensive starting
material. A 20-step formal stereoselective synthesis of
squalamine (3) from the inexpensive starting material
stigmasterol (1) was also described.3 Stigmasterol is also
the starting point for the fifteen step synthesis of
squalamine described in this report (Scheme 1). The first
10 steps of this route have recently been reported (1 to
2).14,15
Syn th esis of Squ a la m in e Utilizin g a
Rea d ily Accessible Sp er m id in e Equ iva len t
Xuehai Zhang, Meenakshi N. Rao, Stephen R. J ones,
Bin Shao, Penina Feibush, Melissa McGuigan,
Nathan Tzodikov, Binyamin Feibush, Ilya Sharkansky,
Brad Snyder, Larry M. Mallis, Ani Sarkahian,
Susan Wilder, J oshua E. Turse, and William A. Kinney*
Magainin Pharmaceuticals, Inc., 5110 Campus Drive,
Plymouth Meeting, Pennsylvania 19462
Hans J ørgen Kjærsgaard
Niels Clauson-Kaas A/ S, Rugmarken 28,
DK-3520 Farum, Denmark
The final five steps involve the incorporation of the
polyamine and sulfate moieties. Steroid 2 is suitably
protected to allow the stepwise elaboration of the target
molecule. There remained some significant challenges
in completing a practical synthesis of squalamine. We
expected to couple a protected polyamine to a C-3 ketone
by a reductive process to afford the â-polyamine. Frye
has shown that there is no stereoselectivity in the
reductive amination in the presence of a C-7 benzyl13 or
benzoate.16 Additionally, the yield of sulfation of the C-24
hydroxyl group was low with a spermidine-functionalized
molecule. Therefore, our plan was to first introduce the
sulfate and then cleave the benzoate group, once its
function had been served; the stability of the sulfate to
conditions that cleave the benzoate has been demon-
strated.16 The unprotected C-7 hydroxyl group was
predicted to provide less steric hindrance to the approach
of the reducing agent from the R-face during reductive
amination. Supportive evidence that a small group at
C-7 is advantageous was provided by the good stereo-
control noted by Moriarty in the presence of a C-7 acetate
group3. Finally, we had to utilize a protecting group on
the polyamine that would be labile under conditions that
would not cleave the sulfate group. The BOC-protected
spermidine derivative used by Moriarty3 and Frye13 is
removed in neat trifluoroacetic acid, which would be
expected to cleave the sulfate group. The selection of an
appropriate protecting group is further complicated by
the fact that the group must be stable to the reductive
amination conditions. For instance, a trifluoroacetyl-
group would be unsatisfactory. Ideally the polyamine
should not require many steps or chromatography to
prepare in analytically pure form. The BOC-protected
spermidine, which is typical of described monoprotected
spermidines, is prepared in multiple steps and requires
chromatography.17
Ronald S. Michalak†
Ash-Stevens, 5861 J ohn C. Lodge Freeway,
Detroit, Michigan 48202
Received J uly 10, 1998
In tr od u ction
Squalamine (3, Scheme 1) is a natural aminosterol
found in the tissues of the dogfish shark, which was
initially identified based on its antimicrobial activity.1
Its structure was determined by spectroscopic2 and
chemical3 means. More recently, squalamine was found
to be antiangiogenic in that it inhibits endothelial cell
function and effects the growth of solid tumors.4-7 The
need for new chemotherapeutic approaches in the treat-
ment of cancer8,9 and the importance of angiogenesis in
tumor growth and in the development of metastatic
disease10,11 have both propelled clinical development of
antiangiogenic agents such as squalamine. Although the
majority of the squalamine used in preclinical studies
was obtained by extraction and purification of dogfish
livers, this source was projected to be too costly and
unreliable to provide adequate supplies for clinical trials.
Therefore, we directed our efforts to a cost-effective
chemical synthesis.
† Current address: Chemical Development, Wyeth-Ayerst Research,
64 Maple St., Rouses Point, NY 12979.
(1) Moore, K. S.; Wehrli, S.; Roder, H.; Rogers, M.; Forrest, J . N.
J r.; McCrimmon, D.; Zasloff, M. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 1354-1358.
(2) Wehrli, S. L.; Moore, K. S.; Roder, H.; Durell, S.; Zasloff, M.
Steroids 1993, 58, 370-378.
(3) Moriarty, R. M.; Enache, L. A.; Kinney, W. A.; Allen, C. S.;
Canary, J . W.; Tuladhar, S. M.; Guo, L. Tetrahedron Lett. 1995, 36,
5139-5142.
(4) Sills, A. K., J r; Williams, J . I.; Tyler, B. M.; Epstein, D. S.; Sipos,
E. P.; Davis, J . D.; McLane, M. P.; Pitchford, S.; Cheshire, K.; Gannon,
F. H.; Kinney, W. A.; Chao, T. L.; Donowitz, M.; Laterra, J .; Zasloff,
M.; Brem, H. Cancer Res. 1998, 58, 2784-2792.
(12) Moriarty, R. M.; Tuladhar, S. M.; Guo, L.; Wehrli, S. Tetrahe-
dron Lett. 1994, 35, 8103-8106.
(13) Pechulis, A. D.; Bellevue, F. H., III; Cioffi, C. L.; Trapp, S. G.;
Fojtik, J . P.; McKitty, A. A.; Kinney, W. A.; Frye, L. L. J . Org. Chem.
1995, 60, 5121-5126.
(14) J ones, S. R.; Selinsky, B. S.; Rao, M. N.; Zhang, X.; Kinney, W.
A.; Tham, F. S. J . Org. Chem. 1998, 63, 3786-3789.
(15) Rao, M. N.; McGuigan, M. A.; Zhang, X.; Shaked, Z.; Kinney,
W. A.; Bulliard, M.; Laboue, B.; Lee, N. E. J . Org. Chem. 1997, 62,
4541-4545.
(16) Bellevue, F. H., III, Ph.D. Thesis, 1994, Department of Chem-
istry, Rensselaer Polytechnic Institute, Troy, NY.
(17) Goodnow, R., J r.; Konno, K.; Niwa, M.; Kallimopoulos, T.;
Bukownik, R.; Lenares, D.; Nakanishi, K. Tetrahedron 1990, 46, 3267-
3286.
(5) Williams, J . I.; Mangold, G.; Zasloff, M.; von Hoff, D. D. Breast
Cancer Res. Treat. 1997, 46, 30 (poster #104).
(6) Schiller; J . H.; Bittner, G. N.; Williams, J . I.; Zasloff, M. Proc.
Am. Assoc. Cancer Res. 1997, 38, 205 (abstract no. 1378) and 1998,
39, 45 (abstract no. 307).
(7) Williams, J . I.; Ara, G.; Zasloff, M.; Teicher, B. Proc. Am. Assoc.
Cancer Res. 1979, 39, 313 (poster #2135).
(8) Denekamp, J . Cancer and Metastasis Rev. 1990, 9, 267-282.
(9) Kohn, E. C.; Liotta, L. Cancer Res. 1995, 55, 1856-1862.
(10) Teicher, B. A. Crit. Rev. Oncology/ Hematology 1995, 20, 9-39.
(11) Weidner, N.; Folkman, J . Important Advances in Oncology;
DeVita, V. T.; Hellman, S.; Rosenberg, S. A., Eds.; Lippincott-Raven
Publishers: Philadelphia, PA, 1996; pp 167-190.
10.1021/jo981344z CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/28/1998

8600 J . Org. Chem., Vol. 63, No. 23, 1998
Notes
Sch em e 1
Sch em e 2
/

Notes
J . Org. Chem., Vol. 63, No. 23, 1998 8601
Ch a r t 1
Ta ble 1
entry 2, % de 6, % de (after purification) 3, de (after HPLC)
borohydride. These conditions maximize production of
the desired â-orientation at the C-3 position (ratio of 6:1
of 3 to 11, Chart 1). The nitrile of 8 was hydrogenated
at low pH (1-2) using platinum oxide as catalyst to yield
3. When the hydrogenation reaction was attempted at
higher pH, significant amounts of products were obtained
from cyclization of the secondary nitrogen upon the nitrile
function as in compounds 9 and 10 (Chart 1). After
removal of catalyst, 3 was purified by both propyl sulfonic
acid ion exchange and reverse phase HPLC columns.
Synthetic 3, isolated as the trifluoroacetate salt, was
compared to natural 3 by 1H NMR, 13C NMR, and HPLC
and was found to be identical. Early fractions from the
HPLC purification were enriched in the C-24 epimer of
3. By discarding those fractions containing 12 (Chart 1),
it was possible to increase the diastereomeric excess (de)
of the final compound (Table 1).
1
93
94
97
Resu lts a n d Discu ssion
The completion of this 15-step synthesis of squalamine
began (Scheme 2) with the liberation of the protected
ketone functionality of intermediate 2. Acid hydrolysis
of 2 was achieved utilizing Amberlyst 15 ion-exchange
resin as catalyst, which could be easily removed by
filtration. Pyridine was added, and the acetone removed
by distillation. The intermediate ketone 4 was treated
with sulfur trioxide-pyridine complex and the reaction
mixture was worked up with saturated potassium chlo-
ride solution to afford 5 in excellent yield. Cleavage of
the benzoate at C-7 with potassium hydroxide afforded
steroid 6. The starting material 2 contained a small
amount of the C-24 epimer and its quantity was not
significantly altered during the transformation to 6
(Table 1).
To continue the synthesis, we chose the masked
spermidine equivalent 7, because the nitrile is stable to
reductive amination conditions, yet is easily converted
to an amino function (spermidine) by catalytic hydroge-
nation under conditions that do not affect the sulfate.
Reagent 7 has been utilized previously as a spermidine
equivalent.18 It was now prepared in one step by adding
4-bromobutyronitrile over 1.5 h to 1,3-diaminopropane
(5 equiv) at -6 °C. The crude reaction mixture was
converted to the free base using a Dowex 1 × 8-100 (-OH
form) column and then distilled to afford 7 in good yield
(73% versus 42% literature). The conditions of workup
and distillation are crucial, as compound 7 readily
cyclizes to 1,5-diazabicyclo[4.3.0]non-5-ene (DBN).
Squalamine (3) was produced by coupling steroid 6 to
the masked spermidine reagent 7 using trimethyl ortho-
formate and sodium borohydride. The primary amine
reacts preferentially to afford the imine with the elimina-
tion of water by trimethyl orthoformate. The reaction
mixture was cooled to -78 °C and treated with sodium
Con clu sion
Squalamine was efficiently prepared in five steps (37%
yield overall) from 2 utilizing the latent spermidine
reagent 7. This reagent was shown to be most appropri-
ate, as it is stable to reductive amination conditions, is
converted to spermidine under weakly acidic conditions,
and is easily prepared. The investigation of alternate
spermidine equivalents will be reported in due course.
The soybean-based sterol, stigmasterol, which is used as
the starting material to prepare key intermediate 2, is
both inexpensive and readily available. This stereose-
lective synthesis of squalamine from steroid 2 has
provided supplies for phase I clinical trials, which are
currently in progress.
Exp er im en ta l Section
Gen er a l. The 13C NMR spectra were generated at 100 MHz,
utilizing 49.15 (CD3OD) and 77.23 (CDCl3) ppm as the reference.
Elemental analyses were performed at Oneida Research Ser-
vices, Inc., Whitesboro, NY. Fast atom bombardment mass
spectral analysis was carried out at M-Scan Inc., West Chester,
PA.
P ota ssiu m (5r, 7r, 24R)-7-Ben zyloxy-3-k eto-ch olesta n -
24-yl Su lfa te (5). A three-necked 3 L flask equipped with a
mechanical stirrer was charged with a mixture of 2 (54 g, 95
mmol),14 acetone (2.7 L), and Amberlyst 15 ion-exchange resin
(18) Umeda, Y.; Moriguchi, M.; Kuroda, H.; Nakamura, T.; Fujii,
A.; Iinuma, H.; Takeuchi, T.; Umezawa, H. J . Antibiot. 1987, 40 9,
1303-1315.

8602 J . Org. Chem., Vol. 63, No. 23, 1998
Notes
(H+ form, 22 g) and stirred at room temperature for 4 h. After
filtration, pyridine (6 mL) was added to the filtrate. After
evaporation of the solvent in vacuo, pyridine (1.5 L) was added
to the residue at room temperature under nitrogen. Sulfur
trioxide-pyridine complex (30.0 g, 188 mmol) was added in one
portion, and the mixture was warmed to 80 °C for 30 min, after
which TLC showed completion of the reaction. The solvent was
removed in vacuo, and ethyl acetate (610 mL) was added to the
residue. The resulting suspension was filtered, and the filter
cake was washed with ethyl acetate (2 × 85 mL). A solution of
potassium chloride (22 g) in water (150 mL) was added to the
vigorously stirred organic solution. After a few minutes, a thick
suspension was obtained. tert-Butyl methyl ether (1500 mL) was
added, and the suspension was cooled to 0 °C, filtered, and
washed with cold water (2 × 100 mL) and tert-butyl methyl ether
(2 × 100 mL). The solid was dried (50 °C, 2 mmHg) to yield 5
as the potassium salt (48.5 g, 77% from 2); 1H NMR (400 MHz,
CD3OD): δ 8.02 (d, J ) 8 Hz, 2H), 7.62 (t, J ) 7 Hz, 1H), 7.51,
(t, J ) 7.5 Hz, 2H), 5.14 (br s, 1H), 4.08 (q, J ) 5 Hz, 1H), 2.04-
1.08 (m, 27H), 0.95-0.86 (m, 12H), 0.72 (s, 3H). Anal. Calcd
for C34H49O7SK‚1.2H2O (FW 662.54): C, 61.64; H, 7.82; H2O,
3.26. Found: C, 61.59; H, 7.85; H2O (KF), 2.94.
clear after the addition of aqueous 0.1 M silver nitrate. The
resin was washed with IPA (11 L) and the column was ready
for use.
The above filtrate, containing crude 7, was added to the top
of the ion exhange column. After the entire filtrate had passed
through the column, the column was washed with IPA (14 L). A
small aliquot of the combined eluent, when neutralized to pH 7
with acetic acid, was clear after the addition of aqueous 0.1 M
silver nitrate. The combined eluent was concentrated to a
weight of 1.39 kg using a water aspirator and a bath at 45-50
°C. Molecular sieves (3 Å, 100 g) were added to the residue,
which was stored at 0-10 °C overnight. The sieves were
removed by filtration, and the filtrate was distilled under
reduced pressure in a 2 L flask equipped with overhead stirring,
a thermometer, and a distillation head with a short Vigreaux
column. Fractions distilling at less than 114 °C (0.6 mmHg)
were collected and discarded (627 g). Two fractions of pure 7
were collected (total ) 699 g, 73%): fraction 1 (108 g) distilled
at 114-115 °C (0.6 mmHg) and fraction 2 (591 g) distilled at
110-112 (0.5 mmHg): 1H NMR (400 MHz, CDCl3): δ 2.71 (t, J
) 6.5 Hz, 2H), 2.68 (t, J ) 6.5 Hz, 2H), 2.61 (t, J ) 7 Hz, 2H),
2.39 (t, J ) 7 Hz, 2H), 1.75 (p, J ) 7 Hz, 2H), 1.56 (p, J ) 7 Hz,
2H); 13C NMR (100 MHz, CDCl3): δ 119.9, 48.1, 47.7, 40.5, 33.8,
25.8, 15.0; MS (+FAB): 142 ([M + H]+, 100); IR (neat, cm-1):
3280, 2930, 2244, 1592, 1470, 1128, 830. Anal. Calcd for
C7H15N3-0.15H2O: C, 58.42; H, 10.72; N, 29.20. Found: C,
58.60, H, 10.52, N, 28.86.
(3â, 5r, 7r, 24R)-3-[[3-[(4-a m in ob u t yl)a m in o]p r op yl]-
a m in o]-7-h yd r oxych olesta n -24-yl Hyd r ogen Su lfa te Bis-
(tr iflu or oa ceta te) Sa lt (Squ a la m in e, 3). The polyamine 7
(8.00 g, 56.7 mmol) was dissolved in anhydrous methanol (650
mL) at room temperature, and trimethyl orthoformate (50 mL,
457 mmol) was added. Steroid 6 (10.0 g, 18.4 mmol) was added,
and the reaction mixture was stirred for 18 h. The reaction
mixture was cooled to -78 °C, treated with sodium borohydride
(1.06 g, 28.0 mmol) over one min, and stirred for 3.5 h at -78
°C. The reaction was allowed to warm to room temperature and
was concentrated at 31 °C under a water aspirator vacuum. The
crude product was dissolved in 100% ethanol (290 mL), purged
with nitrogen, and acidified to pH 1-2 with neat trifluoroacetic
acid. Platinum oxide (1.00 g) was added and the mixture was
shaken on a Parr apparatus (40 psi) for 18 h. The reaction
mixture was filtered through paper, and the solids were washed
with methanol (620 mL). The filtrate was evaporated and then
dissolved in 50% ethanol in water.
A propyl sulfonic acid (PSA) ion exchange column was
prepared by suspending 80 g of resin (J . T. Baker, Phillipsburg,
NJ ) in 10% IPA in water to form a slurry and by adding 200
mL of 10% IPA to the column, followed by the slurry. At least
five column volumes of 10% IPA were eluted through the column
at a flow rate of 40 mL/min. The column was washed with 0.05%
trifluoroacetic acid (TFA) in 50% ethanol in water (150 mL)
at 20 mL per min. The steroid from above was loaded on the
column in two portions, and eluent was collected. For each
portion the column was washed with two column volumes of
0.05% TFA in 50% ethanol in water and two column volumes of
0.05% TFA in 10% IPA. The column was then eluted with 4.5
M KOAc/10% IPA (pH 5), and fractions were collected (150 mL
each). Fractions that contained squalamine by TLC (silica gel,
eluent 6:3:1, dichloromethane:methanol:ammonium hydroxide;
Rf ) 0.35-0.40) were combined.
The two batches of crude 3 from the ion exchange were
purified on a 25 × 5 cm plus 5 × 5 cm YMC ODS-AQ C18
reversed phase columns (YMC, Inc. Wilmington, NC). The
eluent from PSA was diluted with four volumes of deionized
water. The column was loaded and eluted with four column
volumes of buffer A (0.05% TFA in 1% acetonitrile in water).
Then the column was eluted (100 mL/min) with the following
gradient of increasing percentages of buffer B (B ) 0.05% TFA
in 1% water in acetonitrile) (Detector: UV λ ) 220 nm): 0-25%
(10 min), 25-40% (50 min), and 40-80% (20 min). Between 28
and 60 min (30-40% B) fractions were collected every 30 s (50
mL). All fractions were examined by TLC, and the early and
late fractions that contained 3 were analyzed by analytical HPLC
with o-phthaldialdehyde (OPA) derivitization (see below). Frac-
tions that were >95% pure were combined and lyophilized to
afford 97% pure 3 (10.3 g, 62% yield) as the trifluoroacetate salt.
P ota ssiu m (5r, 7r, 24R)-7-Hyd r oxy-3-k etoch olesta n -24-
yl Su lfa te (6). A 1 L three-necked flask equipped a mechanical
stirrer, thermometer, and reflux condenser was charged with
methanol (350 mL), potassium hydroxide (23.3 g of 85%, 353
mmol), and 5 (40 g, 60 mmol) under nitrogen. The mixture was
heated to 60 °C and stirred overnight. The suspension was
concentrated under reduced pressure, and water (275 mL) and
dichloromethane (275 mL) were added to the residue with
vigorous stirring. The suspension was cooled to 0 °C and Celite
(11 g) was added to facilitate the filtration. The mixture was
filtered, and the solid was washed with cold water (2 × 100 mL)
and dichloromethane (2 × 75 mL). The solid was vacuum-dried
at 50 °C to afford a solid (42.2 g). The solid was suspended in
methanol (400 mL) and triethylamine (7 mL), warmed to 60 °C
for 45 min, and filtered. The filter cake was washed with warm
methanol (110 mL), and the combined filtrates were concen-
trated at reduced pressure until a thick suspension was ob-
tained. tert-Butyl methyl ether (325 mL) was added to the
suspension, and the mixture was cooled to 0 °C, filtered, and
washed with tert-butyl methyl ether (100 mL). The solid was
vacuum-dried at 50 °C to afford 6 as a white solid (29.4 g, 90%,
mp 153-158 °C): 1H NMR (400 MHz, DMSO-d6): δ 4.16 (m,
1H), 3.78 (br q, J ) 5 Hz, 1H), 3.61 (m, 1H), 2.5-1.0 (m, 27H),
0.94 (s, 3H), 0.86 (d, J ) 6 Hz, 3H), 0.81 (d, J ) 7 Hz, 3H), 0.79
(d, J ) 7 Hz, 3H), 0.63 (s, 3H). Anal. Calcd for C27H45O6SK‚
0.04Na‚0.38H2O (FW 544.58): C, 59.55; H, 8.47; H2O, 1.26; K,
7.18; Na, 0.17. Found: C, 59.64; H, 8.39; H2O, 1.28; K, 7.20;
Na, 0.17; IR (KBr, cm-1): 3436, 2936, 1708, 1470, 1390, 1208,
1056, 1038, 950, 812; HPLC 94% de.
3-[(3-Cya n op r op yl)a m in o]p r op yla m in e (7). The following
is a modification of the published procedure.18 A flask containing
1,3-diaminopropane (2.50 kg, 33.7 mol) was stirred with a
mechanical stirrer, cooled (-6 °C), and treated with 4-bromobu-
tyronitrile (1.00 kg, 6.76 mol) over 1.5 h, maintaining the
internal temperature below 0 °C. The cold bath was removed,
and the reaction was allowed to stir without auxiliary temper-
ature control for 1 h. 2-Propanol (IPA, 11 L) was added in one
portion to the reaction. The mixture was stirred for 15 min after
the appearance of a precipitate and then stored at 0-10 °C
overnight. The solids were collected by filtration in a Buchner
funnel lined with a polypropylene felt filter pad. The solids were
washed with IPA (2 × 1.1 L). The combined filtrate was purified
using a Dowex 1 × 8-100 (-OH form) ion exhange column
prepared as described below.
A 2.2 kg sample of Dowex 1 × 8-100 (-Cl form) was combined
with aqueous 5 N sodium hydroxide (6 L). After standing for 1
h, the mixture was poured into a suitably sized chromatography
column with a coarse glass frit. The resin was washed with 5
N sodium hydroxide (44 L). A small aliquot of the eluent,
neutralized to pH 7 with acetic acid, appeared hazy when
aqueous 0.1 M silver nitrate was added. An additional wash
with 5 N sodium hydroxide (9 L) did not visibly improve clarity.
The resin was washed with deionized water (6.6 L), at which
time the pH of the eluent was 7. An aliquot of the eluent was

Notes
J . Org. Chem., Vol. 63, No. 23, 1998 8603
1
This synthetic material was identical to natural 3 by analytical
HPLC (OPA method); H NMR (400 MHz, CD3OD): δ 4.12 (br
9: H NMR (400 MHz, CD3OD): δ 4.08 (br q, J ) 3 Hz, 1H),
3.77-3.73 (m, 3H), 3.53 (t, J ) 7 Hz, 2H), 3.28-3.05 (m, 3H),
2.91 (t, J ) 8 Hz, 2H), 2.14 (p, J ) 7.5 Hz, 2H), 2.04-1.08 (m,
29H), 0.93-0.88 (m, 9H), 0.83 (s, 3H), 0.67 (s, 3H); MS (+FAB):
624.2 (M + 1, 30), 544.3 (27), 526.4 (33), 125.1 (100). Anal. Calcd
for C34H61N3O5S‚TFA‚3H2O: C, 54.59; H, 8.65; N, 5.31; S, 4.05.
Found: C, 54.96; H, 8.32; N, 5.31; S, 4.15.
10: 1H NMR (400 MHz, CD3OD): δ 4.10 (br q, 1H), 3.77 (br s,
1H), 3.67 (m, 2H), 3.32-3.28 (m, 2H), 3.13-3.09 (m, 5H), 2.13-
1.09 (m, 33H), 0.94-0.90 (m, 9H), 0.85 (s, 3H), 0.68 (s, 3H); MS
(+FAB): 611.3 (M + 1, 92%), 531.4 (32), 513.4 (100). Anal. Calcd
for C34H62N2O5S‚TFA‚2.5H2O: C, 56.16; H, 8.90; N, 3.64; S, 4.16.
Found: C, 55.93; H, 8.69; N, 3.73; S, 4.26.
HP LC An a lysis of 2 a n d (24S)-2. The diastereomeric excess
for compound 2, compared to the 24S-isomer of 2, was deter-
mined by HPLC. A sample of 2 (2 mg/mL) in acetonitrile was
prepared and injected (5 µL) on a Waters Nova-Pak Phenyl (3.9
× 150 mm) column at ambient temperature and eluted with 45%
water in acetonitrile (1 mL/min). Detection was at 230 nm.
Typical retention times for compound 2 and the 24S-isomer of
compound 2 are 21 and 19 min, respectively.
1
q, 1H), 3.76 (br s, 1H), 3.2-2.9 (m, 9H), 2.1-1.0 (m, 33H), 0.94-
0.90 (m, 9H), 0.84 (s, 3H), 0.67 (s, 3H); 13C NMR (100 MHz, CD3-
OD): δ 86.7, 68.4, 59.2, 57.7, 51.8, 46.8, 46.0, 43.9, 43.0, 41.2,
40.1, 38.7, 38.0, 37.8, 37.5, 37.0, 32.7, 32.2, 32.1, 29.5, 28.3, 26.1,
25.7, 24.7, 24.6, 24.3, 22.3, 19.6, 18.6, 18.3, 12.6, 11.7. Anal.
Calcd for C34H65N3O5S‚2TFA‚2.5H2O (FW 901): C, 50.65; H,
8.05; N, 4.66; S, 3.56; F, 12.65. Found: C, 50.71; H, 8.06; N,
4.70; S, 3.75; F, 12.13; HPLC 96% de.
The 3-R-isomer of 3 (11) eluted after squalamine on reversed
phase HPLC (ratio of 3 to 11 ) 6/1). 11: 1H NMR (400 MHz,
CD3OD): δ 4.14 (q, J ) 8 Hz, 1H), 3.81 (br s, 1H), 3.44 (br s,
1H), 3.18-2.97 (m, 8H), 2.20-2.16 (m, 2 H), 2.04-1.14 (m, 31H),
0.96-0.93 (m, 9H), 0.87 (s, 3H), 0.71 (s, 3H); 13C NMR (CD3OD,
100 MHz): δ 86.5, 68.4, 57.7, 56.9, 51.8, 48.4, 46.8, 46.1, 44.5,
43.8, 41.1, 40.2, 37.34, 37.29, 37.2, 33.4, 32.9, 32.5, 32.1, 30.2,
29.4, 28.3, 25.6, 24.6, 24.2, 24.1, 23.9, 21.8, 19.5, 18.5, 12.5, 11.0.
Anal. Calcd for C34H65N3O5S‚2TFA‚2H2O: C, 51.17; H, 8.02;
N, 4.71; S, 3.59; F, 12.78. Found: C, 51.12; H, 7.48; N, 4.66; S,
3.51; F, 12.72.
The fractions enriched in the 24S-isomer of 3 (12) were
combined, concentrated in vacuo to remove the organic modifier,
and purified on the same YMC column. The column was eluted
(100 mL/min) with the following gradient of buffers A and B
(see above): 0-25% (10 min), 25-35% (50 min), 35-60% (10
min), and 60-80 (10 min) (detector: UV λ ) 220 nm). Fractions
were collected every 30 s (50 mL) and analyzed by TLC and
HPLC (OPA method). Fractions which contained pure 12 were
combined and lyophilized: 1H NMR (400 MHz, CD3OD): δ 4.13
(br q, J ) 5 Hz, 1H), 3.80 (br s, 1H), 3.19-2.97 (m, 9H), 2.16-
HP LC An a lysis of 6 a n d 24S-6. This HPLC analysis
provides the relative quantity of 6 and the 24S-isomer of 6. The
analysis was performed at
a flow rate of 1 mL/min and
temperature of 35 °C on a Kromasil C4 RP (4.6 × 250 mm), 100
Å, 5 µm column (Phenomenex) using the following mobile
phases: A: 0.1% TFA in 90/10 water/acetonitrile and B: 0.1%
TFA in 10/90 water/acetonitrile and a gradient of 35-55% B (15
min), 55-85% B (10 min). The compound was observed by
ultraviolet spectroscopy at 200 nm. The sample (10 µL) was
prepared at a concentration of 4 mg/mL in 0.01% triethylamine
in methanol. Typical retention times: 24S-isomer of 6, 18 min,
and 6, 19 min.
OP A HP LC An a lysis of 3. This HPLC analysis provides
the relative quantities of squalamine (3), 12, and 11. Because
of the lack of a strong UV chromophore, 3 was converted to its
OPA derivative using o-phthaldialdehyde reagent (Sigma) prior
to analysis. The analysis was performed at a flow rate of 1 mL/
min and column temperature of 45 °C on a Kromasil C18 RP
250 × 4.6 mm, 100 Å, 5 µm column using the following mobile
phases: A: 0.1% TFA in 90/10 water/acetonitrile and B: 0.1%
TFA in 10/90 water/acetonitrile and a gradient of 30-50% B (12
min), 50-60% B (5 min), 60-85% B (2 min). The compound
was observed by fluorescence detection (excitation ) 230 nm,
emission ) 455 nm). The sample was prepared at a concentra-
tion of 100 µg/mL in 20 mM acetate buffer (pH ) 4.0). The
derivitization reaction was accomplished in the autosampler by
removing 5 µL of OPA reagent, removing 10 µL of sample,
removing 5 µL of OPA, and mixing in the injector. Average
retention times: 12, 11.5 min; 3, 12.2 min; and 11, 15.5 min.
1.06 (m, 33H), 0.99-0.93 (m, 9H), 0.88 (s, 3H), 0.72 (s, 3H); 13
C
NMR (100 MHz, CD3OD): δ 86.5, 68.3, 59.2, 57.7, 51.8, 48.4,
46.7, 46.0, 43.9, 43.0, 41.2, 41.1, 40.1, 38.7, 37.85, 37.77, 37.5,
37.0, 32.6, 32.0, 31.8, 29.6, 28.6, 26.1, 25.6, 24.7, 24.5, 24.2, 22.2,
19.6, 18.9, 18.0, 12.6, 11.7. Anal. Calcd for C34H65N3O5S‚2TFA‚
2H2O: C, 51.17; H, 8.02; N, 4.71; S, 3.59. Found: C, 51.31; H,
7.99; N, 4.76; S, 3.66.
P r ep a r a tion of 9 a n d 10. Compounds 9 and 10 are two of
the major impurities identified, which elute after 3 and 11 on
reversed phase HPLC. Late HPLC fractions from several
syntheses of 3 that contained 9 and 10 were combined and
lyophilized. Crude solid (2 g) was purified by flash chromatog-
raphy on a silica gel column (5 cm diameter, gradient elution
with 12/3/1 to 3/3/1 of chloroform/methanol/isopropylamine)
affording first 10 and then 9 in their free base forms. Com-
pounds 9 and 10 were converted to their trifluoroacetic acid
(TFA) salts on the polystyrene resin, amberchrom (CG-161,
TosoHaas, Montgomeryville, PA). The free base was dissolved
in methanol, treated with amberchrom (2 mL), and concentrated
in vacuo. This solid was loaded onto a column (20 × 1 cm) of
the same absorbent prepared in water. The column was eluted
with 0.1% TFA in water (200 mL), 0.1% TFA and 20% acetoni-
trile in water (200 mL), 0.1% TFA and 40% acetonitrile in water
(200 mL), 0.1% TFA and 60% acetonitrile in water (200 mL),
and 0.1% TFA and 80% acetonitrile in water (200 mL). Com-
pound 9 eluted with 40% acetonitrile (550 mg), and 10 eluted
with 60% acetonitrile (815 mg).
Ack n ow led gm en t. The authors gratefully acknowl-
edge the leadership of Michael Zasloff and Ze’ev Shaked
in supporting the discovery and development of the
squalamine manufacturing process.
J O981344Z