Neurotrophic peptide aldehydes: Solid phase synthesis of fellutamide B and a simplified analog

Article abstract of DOI:10.1016/j.bmcl.2006.04.029

A combination of solid phase and solution phase synthetic methods have been used to complete the total synthesis of the neurotrophic lipopeptide aldehyde fellutamide B (2). The β-hydroxy aliphatic tail was prepared by regioselective reductive opening of a cyclic sulfate, and later coupled to a solid phase resin. The synthetic compound was then examined in cytotoxicity and nerve growth factor (NGF) induction assays. A simplified analog of fellutamide B also showed activity.

Full text of DOI:10.1016/j.bmcl.2006.04.029

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3855–3858
Neurotrophic peptide aldehydes: Solid phase synthesis
of fellutamide B and a simplified analog
b
b
a
a,b,c,
*
John S. Schneekloth, Jr., John L. Sanders, John Hines and Craig M. Crews
a
Department of Molecular, Cellular, and Developmental Biology, Yale University, PO Box 208103, New Haven, CT 06520-8103, USA
b
Department of Chemistry, Yale University, PO Box 208103, New Haven, CT 06520-8103, USA
Department of Pharmacology, Yale University, PO Box 208103, New Haven, CT 06520-8103, USA
c
Received 3 March 2006; revised 5 April 2006; accepted 5 April 2006
Available online 11 May 2006
Abstract—A combination of solid phase and solution phase synthetic methods have been used to complete the total synthesis of the
neurotrophic lipopeptide aldehyde fellutamide B (2). The b-hydroxy aliphatic tail was prepared by regioselective reductive opening
of a cyclic sulfate, and later coupled to a solid phase resin. The synthetic compound was then examined in cytotoxicity and nerve
growth factor (NGF) induction assays. A simplified analog of fellutamide B also showed activity.
Ó 2006 Elsevier Ltd. All rights reserved.
Interest in neurotrophic small molecules has increased
total synthesis of fellutamide B, a simplified analog, and
confirmation of both molecules’ biological activities.
1
–3
significantly in recent years. It is hoped that the study
of such low molecular weight factors will provide further
insight into the biochemical mechanisms of neuronal cell
proliferation/differentiation and lead to drug candidates
for neurodegenerative conditions such as Alzheimer’s
disease. As part of a program directed toward the total
synthesis of biologically active natural products and the
identification of their relevant intracellular binding pro-
tein(s), we became interested in the group of lipopeptide
aldehydes known as the fellutamides.
As both fellutamides A and B were reported to have
comparable IC50 values for cytotoxicity, we initially pur-
sued the synthesis of the more accessible fellutamide B
(2). It was envisioned early on that a solid phase con-
struction of the peptide chain would be facilitated by a
solution phase preparation of the (R)-(À)-3-hydroxydo-
decanoic acid synthon 6. Solid phase synthesis of the
peptide chain was a particularly attractive strategy, as
it would allow us to access analogs of the natural prod-
uct by coupling different amino acids to the resin. Addi-
tionally, the mild resin cleavage conditions would be
unlikely to epimerize the sensitive aldehyde functionality
(See Fig. 1).
Despite displaying potent activity in both the NGF
induction and cytotoxicity assays, the fellutamide class
of natural products have attracted little attention since
4
their isolation in 1991. Fellutamides A (1) and B (2)
were originally identified for their cytotoxic properties,
and later found to stimulate NGF synthesis and secre-
Our route to the requisite b-hydroxy side chain is de-
scribed below (Scheme 1). Decanal (3) was homologated
to a,b-unsaturated ester 4 by means of a Horner–Wads-
worth–Emmons condensation with triethyl phospho-
5
tion. The structures of fellutamides A and B were eluci-
dated by degradation studies, but have not yet been
confirmed by total synthesis. In an effort to better under-
stand the mechanism of action of these compounds, we
initiated a program directed toward synthesizing felluta-
mide B, studying its neurotrophic activity, and identify-
ing its target binding protein. In this letter, we report the
Keywords: Natural product; Peptide aldehyde; Nerve growth factor
(NGF); Synthesis.
*
Corresponding author. Tel.: +1 203 432 9364; fax: +1 203 432
161; e-mail: craig.crews@yale.edu
6
Figure 1. Structures of fellutamides A and B.
0
960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.029

3856
J. S. Schneekloth, Jr. et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3855–3858
8
the b-hydroxy ester. At this stage, the enantiomeric ex-
cess of the asymmetric dihydroxylation was confirmed
to be 90% by conversion of the alcohol to its corre-
9
sponding Mosher’s ester. Lithium hydroxide mediated
saponification of the ester in a MeOH/THF/H O mix-
2
ture provided (R)-(À)-3-hydroxydodecanoic acid 6 in
7
0% over two steps. The preparation of acid 6 was
accomplished in six steps with 47% overall yield, closely
following the methods previously reported in a synthesis
10
of sulfobacin A.
Scheme 1. Reagents and conditions: (a) LiOH, triethyl phosphonoac-
etate, THF, reflux, 4 h, 85%, 95:5 E:Z; (b) AD-Mix-b, MeSO NH , 1:1
t-BuOH/H , pyridine, 0 °C,
O, 0 °C, 36 h, 80%, 90% ee; (c) SOCl
0 min, 99%; (d) NaIO , 1 mol % RuCl , 3:1 MeCN/H O, rt, 30 min,
0%; (e) NaBH
, DMA, 0 °C to rt, 1 h, then 20% aq HCl, rt, 18 h,
2%; (f) LiOH, MeOH/H O/THF 1:1:1, rt, 3 h, 97%.
2
2
2
2
3
9
7
4
3
2
With acid 6 in hand, we next attempted to complete the
solid phase synthesis of fellutamide B (Scheme 2). Using
Fmoc solid phase peptide synthesis (SPPS) proce-
4
2
1
1,12
13
dures,
leucinal loaded beads 7 were coupled first
with protected glutamine (Fmoc-Gln(Trt)-OH), fol-
lowed by protected asparagine (Fmoc-Asn(Trt)-OH),
and finally with (R)-(À)-3-hydroxydodecanoic acid (6).
Global deprotection of the side chain protecting groups
noacetate in 85% yield. Sharpless asymmetric dihydr-
6
oxylation of 4 was carried out using AD-Mix-b with ex-
cess methane sulfonamide in a 1:1 t-BuOH/H O mixture
2
in 80% yield. The resultant diol was treated with thionyl
chloride, and the cyclic sulfite was immediately oxidized
to the sulfate by treatment with sodium periodate and
1
4
with anhydrous trifluoroacetic acid (TFA) was fol-
lowed by cleavage of the peptide from the resin with
1
5
0
.1% TFA in 2:3 MeCN/H O. Fellutamide B was iso-
2
7
catalytic ruthenium(III) chloride. Cyclic sulfate 5 was
lated in analytically pure form after lyophilization from
the resin cleavage solution. Synthetic fellutamide B was
found to be identical in all respects to the isolated natu-
isolated in 79% yield over three steps. Regioselective
reductive cleavage of sulfate 5 was affected by treatment
with sodium borohydride, and acidic workup provided
1
6
ral product.
Encouraged by our success with fellutamide B, we next
prepared an N-octanoyl analog of fellutamide B 10
1
7
using an analogous strategy (Scheme 2). This analog
contains a shorter lipophilic tail and lacks the b-hydroxy
amide functionality present in the natural product. It
was hoped that this analog would allow us to determine
the importance of the N-acyl chain for NGF induction.
Having prepared the two aldehydes 2 and 10, we exam-
ined their biological activities. Cytotoxicity IC values
5
0
1
8
were determined for three cell lines (Table 1). In our
hands, fellutamide B exhibited a potency similar to ear-
4
,5
lier reports. To our surprise, the N-octanoyl analog 10
was approximately an order of magnitude less cytotoxic
than 2 against all cell lines tested, suggesting that mod-
ification of the lipophilic tail has a significant effect on
the activity of these compounds.
With data in hand regarding the cytotoxicity of the two
compounds, we next examined the capacity of both
1
9
aldehydes to induce NGF secretion (Fig. 2). In this as-
say, L-M cells were treated with 50 lM fellutamide B or
octanoyl fellutamide for 24 h to allow for drug-induced
NGF secretion, after which the medium was removed,
filtered, and dialyzed. The conditioned medium was then
added to rat preneuronal PC12 cells for 48 h, after
which the cells were examined for neurite outgrowth.
Scheme 2. Reagents and conditions: (a) Fmoc-Gln(Trt)-OH, HBTU,
HOBt, 35 min; (b) 20% piperidine in DMF, 2 min; (c) Fmoc-Asn(Trt)-
OH, HBTU, HOBt, 35 min; (d) 6, HBTU, HOBt, 35 min; (e) TFA,
2
30 min; (f) 0.1:40:60 TFA/MeCN/H O, 30 min; (g) octanoic acid,
HBTU, HOBt, 35 min.
a
Table 1. Cytotoxicity assay results for fellutamide B (2) and octanoyl fellutamide analog 10
Cell line
Fellutamide B (2) (nM)
N-Octanoyl fellutamide (10) lM
KB (human epithelial carcinoma)
PC12 (rat pheochromocytoma)
L-M (mouse fibroblast)
349 (±31.5)
437 (±26.0)
482 (±81.5)
4.83 (±0.20)
1.67 (±0.23)
3.33 (±0.39)
a
Values are means of three to five experiments, standard error is given in parentheses.

J. S. Schneekloth, Jr. et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3855–3858
3857
Pharmaceuticals for a predoctoral fellowship. We
acknowledge Stanley Lo for help with initial experi-
ments. We acknowledge the NIH (GM062120) for
funding.
References and notes
1
2
3
4
. Pettus, T. R. R.; Chen, X. T.; Danishefsky, S. J. J. Am.
Chem. Soc. 1998, 120, 12684.
. Cho, Y. S.; Carcache, D. A.; Tian, Y.; Li, Y. M.;
Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 14358.
. Waters, S. P.; Tian, Y.; Li, Y. M.; Danishefsky, S. J.
J. Am. Chem. Soc. 2005, 127, 13514.
. Shigemori, H.; Wakuri, S.; Yazawa, K.; Nakamura, T.;
Sasaki, T.; Kobayashi, J. Tetrahedron 1991, 47, 8529.
Figure 2. PC12 cells differentiate in response to fellutamide B-
conditioned medium. (A) Undifferentiated PC12 cells. (B) PC12 cells
treated with DMSO-conditioned medium, 48 h. (C) PC12 cells treated
with NGF (100 ng/mL), 48 h. (D) PC12 cells treated with 50 lM
fellutamide B-conditioned medium, 48 h.
5. Yamaguchi, K.; Tsuji, T.; Wakuri, S.; Yazawa, K.;
Kondo, K.; Shigemori, H.; Kobayashi, J. Biosci. Biotech-
nol. Biochem. 1993, 57, 195.
6. Becker, H.; Sharpless, K. B. Angew. Chem., Int. Ed. 1996,
3
. Kim, B. M.; Sharpless, K. B. Tetrahedron Lett. 1989, 30,
5, 448.
7
Neurotrophic activity is defined here as neurite out-
growth in response to the NGF present in the condi-
655.
8. Gao, Y.; Sharpless, K. B. J. Am. Chem. Soc. 1988, 110,
7538.
2
0
tioned medium. Importantly, Figure 2D indicates
that fellutamide B does induce the secretion of NGF
from L-M cells. In an identical assay, 50 lM 10 was
equally effective in inducing NGF secretion (not shown).
These data indicate that even though 10 is significantly
less cytotoxic than 2, the simplified analog retains neu-
rotrophic activity. While the capacity of these com-
pounds to induce NGF secretion is intriguing, it
should be noted that concentrations necessary to
achieve the neurotrophic effect are currently still above
the IC50 value for cytotoxicity. However, it should be
noted that the secretion of NGF in this assay can be
attributed directly to the action of fellutamide on the
L-M cells. In a similar assay, conditioning of medium
by treatment of L-M cells with other general toxic
agents such as ion channel inhibitors (veratridine and
9
. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95,
12.
0. Gupta, P.; Naidu, S. V.; Kumar, P. Tetrahedron Lett.
004, 45, 9641.
5
1
1
2
1. General procedure for SPPS coupling: H-Leu-H Nov-
aSyn beads (1.0 g, 0.23 mmol) were added to an oven-
Ò
dried solid phase peptide synthesis reactor vessel and
allowed to soak three times under DMF in 10 min cycles.
To the dried beads was added a solution of a carboxylic
acid (0.64 mmol), 243 mg HBTU (0.64 mmol), 86 mg
HOBt (0.64 mmol), and 0.14 mL N-methylmorpholine in
4
3
mL DMF. The reaction was agitated with nitrogen for
5 min, and the solution was drained from the vessel.
After washing the beads with DMF, the reaction was
repeated for 35 min (to ensure complete conversion) and
the beads were drained and washed with DMF. Depro-
tection was carried out using the general procedure
described below, after which the next coupling was
immediately carried out.
2
1
ouabain) or an actin binding molecule (cytochalasin
2
D) did not produce neurite outgrowth in PC12 cells.
2
1
2. General procedure for SPPS Fmoc deprotection: beads
were covered with 20% piperidine in DMF and reacted for
2 min. The solution was drained, and the process repeated
twice to ensure complete deprotection. The beads were
washed five times with DMF to remove excess piperidine.
In conclusion, we have completed the first chemical syn-
thesis of the natural product fellutamide B (2) and an N-
octanoyl analog (10). Both compounds were verified to
be cytotoxic and induce NGF secretion in L-M cells.
Fellutamide B behaves in a manner similar to previous
reports. Simplification of the lipophilic tail of felluta-
mide B in 10 significantly decreases the cytotoxicity of
the compound, but the analog retains its neurotrophic
activity. As other peptide aldehydes are known to inhib-
Ò
1
1
3. H-Leu-H NovaSyn beads were obtained from Novabio-
chem (<http://www.novabiochem.com>).
4. General procedure for global deprotection: beads were
washed with DMF five times and dichloromethane five
times. The beads were covered with neat trifuoroacetic
acid, and reacted for 30 min, after which the solution was
drained. The process was repeated until no yellow color
appeared in the eluent, after which the beads were rinsed
five times with dichloromethane.
2
3
24
it proteases, including the proteasome, it is possible
that protease or proteasome inhibition could be a poten-
tial mechanism of action for these compounds. Work is
ongoing in our laboratory to elucidate the specific intra-
cellular protein target of fellutamide B and its mecha-
nism of NGF induction.
1
1
5. General procedure for cleavage from resin: the beads were
2
covered with a solution of 60:40:0.1 H O/MeCN/TFA and
agitated for 30 min. The solution was collected, and the
process repeated three times. The combined eluents were
lyophilized to afford the desired product.
1
6. Spectral data for fellutamide B (2): H NMR dH
Acknowledgments
(
1
500 MHz; DMSO-d ) 0.7–0.9 (m, 9H), 1.2–1.4 (m,
6H), 1.46 (m, 2H), 1.55 (m, 1H), 1.67 (m, 1H), 1.92
6
J.S.S. gratefully acknowledges the American Chemical
Society, Division of Medicinal Chemistry and Aventis
(m, 1H), 2.00 (m, 2H), 2.05 (m, 2H), 2.40 (m, 1H), 2.49 (m,
1H), 3.94 (m, 1H), 4.06 (m, 1H), 4.12 (m, 1H), 4.43 (m,

3858
J. S. Schneekloth, Jr. et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3855–3858
1
H), 6.68 (br s, 1H), 6.82 (br s, 1H), 7.17 (br s, 1H), 7.41
C
were analyzed using PRISM software (GraphPad
Software).
1
3
(
br s, 1H), 8.00 (m, 2H), 8.24 (br s, 1H), 9.28 (s, 1H).
NMR dC (125 MHz; DMSO-d ) 13.8, 21.3, 22.0, 23.0,
19. Neuronal cell differentiation: L-M cells were grown to
confluency and incubated for 24 h in serum-free Medium
199 containing DMSO (vehicle) or fellutamide B or
octanoyl fellutamide. This conditioned medium was then
collected, filtered, and dialyzed against RPMI 1640
medium at 4 °C for 24 h. Horse serum and fetal bovine
serum were then added to the dialyzed conditioned
medium to a final concentration of 2% and 1%, respec-
tively. Finally, the conditioned medium was applied to
undifferentiated PC12 cells that had been plated onto
collagen. After 48 h, the PC12 cells were photographed
and neurite outgrowth assessed.
20. Obin, M.; Mesco, E.; Gong, X.; Hass, A. L.; Joseph, J.;
Taylor, A. J. Biol. Chem. 1999, 274, 11789.
21. Shimojo, R. Y.; Iwaoka, W. T. Toxicology 2000, 154, 1.
22. Castilla, M. A.; Carmelo, C.; Gazapo, R. M.; Martin, O.;
Gonzalez-Pacheco, F. R.; Tejedor, A. Life Sci. 2000, 6,
1003.
6
2
6
3.9, 28.7, 29.0, 29.1, 31.3, 31.4, 36.9, 43.5, 49.8, 52.4, 56.6,
7.5, 171.0, 171.2, 171.8, 171.9, 201.6.
1
1
7. Spectral data for octanoyl fellutamide (10): H NMR dH
(
500 MHz; DMSO-d ) 1.05 (m, 9H), 1.47 (m, 2H), 1.35 (m,
6
8
4
1
H), 1.51 (m, 3H), 1.62 (m, 1H), 1.75 (m, 1H), 2.10 (m,
H), 2.41 (m, 2H), 4.04 (m, 1H), 4.19 (m, 1H), 4.50 (m,
H), 6.77 (br s, 1H), 6.94 (br s, 1H), 7.21 (br s, 1H), 7.41
(
(
br s, 1H), 8.06 (d, J = 7.2, 1H), 8.12 (d, J = 7.2, 1H), 8.31
1
3
d, J = 6.6, 1H), 9.35 (br s, 1H). C NMR dC (125 MHz;
) 14.3, 21.7, 22.4 (2C, i-Pr), 23.4, 24.3, 25.5, 28.8,
9.0, 31.5, 31.6, 36.1, 36.6, 36.9, 50.2, 52.7, 57.0, 171.5,
72.1, 172.9, 174.1, 202.0.
DMSO-d
6
2
1
3
1
8. [ H]thymidine incorporation: cells were seeded into 96-
well plates at a density of 2000–3500 cells/well. The
medium was replaced after 24 h with fresh medium
containing various concentrations of fellutamide B or
octanoyl-fellutamide B. After 20 h, the cells received
2
3
lCi/well of [ H]thymidine, and after another 4 h were
23. Ede, N. J.; Eagle, S. N.; Wickham, G.; Bray, A. M.;
Warne, B.; Shoemaker, K.; Rosenberg, S. J. Pept. Sci.
2000, 6, 11.
24. Myung, J.; Kim, K. B.; Crews, C. M. Med. Res. Rev. 2001,
21, 245.
detached from the wells and passed through glass fiber
filters. The filters were transferred to vials and the
amount of radioactivity incorporated into the cells was
quantified by scintillation counting. The resulting data