Examination of the olefin-olefin ring closing metathesis to prepare Latrunculin B

Article abstract of DOI:10.1016/j.tetlet.2008.10.144

Three subunits of the potent actin polymerization inhibitor Latrunculin B were synthesized and assembled using olefin-olefin ring closing metathesis chemistry to close the 14-membered macrocycle. Metathesis reactions of substrates with various remote protecting group patterns were examined and gave 6,7-E-lactones as the preferred products.

Full text of DOI:10.1016/j.tetlet.2008.10.144

Tetrahedron Letters 50 (2009) 298–301
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Examination of the olefin–olefin ring closing metathesis
to prepare Latrunculin B
*
Jin She, John W. Lampe, Alexandra B. Polianski, Paul S. Watson
Department of Chemistry, Inspire Pharmaceuticals Inc., 4222 Emperor Boulevard, Suite 470, Durham, NC 27703, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Three subunits of the potent actin polymerization inhibitor Latrunculin B were synthesized and assem-
bled using olefin–olefin ring closing metathesis chemistry to close the 14-membered macrocycle.
Metathesis reactions of substrates with various remote protecting group patterns were examined and
gave 6,7-E-lactones as the preferred products.
Received 24 October 2008
Accepted 30 October 2008
Available online 5 November 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Elegant work reported by Kashman1 in the late 1970’s and early
1980’s documenting the isolation and characterization of Latrun-
culin A (1) and Latrunculin B (2) led to a body of synthetic work2,3
to address this family’s interesting chemical architecture as well as
an effort to understand their unique biological mechanism (Fig. 1).
The initial wave of synthetic work was dedicated to understanding
the unique challenges that the new array of functional diversity
embedded in this family of natural products might present in a to-
tal synthesis campaign; specifically (using Latrunculin B as an
example), (1) construction of the 14-membered macrocyclic lac-
tone with its specific set of olefin geometries; (2) control of the
configuration (both relative and absolute) of the C8/C11 stereogen-
ic centers; (3) control of the formation of the C15 stereogenic cen-
ter; (4) conservation of the stereochemical integrity of the C16
stereogenic center throughout the creation of the C14/C15 bond
and protection/deprotection strategies for the 2-thiazolidinone
ring; and (5) formation of the tetrahydropyran ring system. These
pioneering efforts of Kashman, White, and Smith led to the total
synthesis of Latrunculins A and B and relied on the basic strategy
of initial C13/14 bond formation using aldol reactions, ketalization
to form the tetrahydropyran (THP), and macrocycle formation
(Fig. 2). The combined efforts of these research groups were able
to expertly define some of the synthetic limitations of these sys-
tems and paved the way to synthetic improvements and/or alter-
native construction strategies. The advent of ring closing
metathesis technology has invited the application of this powerful
new tool to the synthesis of many complex ring systems.4 Accord-
ingly, Fürstner5 has successfully applied an alkyne/alkyne ring
closing metathesis/semi-hydrogenation strategy to the total syn-
thesis of Latrunculins A and B. Our interest in this family of natural
O
H
R
Me
Me
PgN
S
R
R
OH
Me
S
acid
O
Me
Me
H
O
H
PgN
O
N
OH
CHO
aldol
Pg
OPg OH
Me
O
OPg
S
O
O
I-Ph3P+
Me
O
OH
OH
Fürstner
6
White
Smith
3
Me
Me
7
8
macrolactonization
esterification
Wittig olefination
macrolactonization
O
alkyne/alkyne RCM
semi-hydrogenation
O
2
Me
Me
1
11
O
O
Me
Me
13
O
H
15
O
H
O
O
OH
OH
17
Me
Me
HN
HN 16
O
O
S
S
O
H
O
H
1
O
O
2
OH
OH
1
2
HN
HN
Figure 1. Latrunculins A and B.
S
S
O
O
* Corresponding author. Tel.: +1 919 287 1268.
E-mail address: pwatson@inspirepharm.com (P. S. Watson).
Figure 2
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.10.144

J. She et al. / Tetrahedron Letters 50 (2009) 298–301
299
O
7
8
b, c
Me
Me
Me
Me
a
Me
Me
O
Me
Me
Me
Me
Me
O
O
Me
9
10
11
Me
O
O
O
Bn
O
S
Bn
O
O
H
HN
O
Me
Me
8
O
H
O
H
d
e
11
OH
Me
N
S
N
S
OH
N
S
OH
OMe
7
Et3SiO
O
S
HN
Bn
S
PMBN
S
S
12
13
14
2
3
4
S
O
Me
8
O
O
13
O
11
f
H
7
Et3SiO
Me
O
H
PMBN
Me
Me
5
H
O
Et3SiO
O
Scheme 2. Reagents and conditions: (a) CH2Cl2, NaHCO3, CH3CO3H, 0 °C; (b) p-TSA,
THF, H2O, rt; (c) NaHCO3, H2O, CH2Cl2, NaIO4, rt, 54% for 3 steps; (d) 12, TiCl4,
(i-Pr)2NEt, CH2Cl2, À78 °C then 11, 76%; (e) DMF, Et3SiCl, (i-Pr)2NEt , 0 °C, 98%; (f)
(i-Bu)2AlH, toluene, À78 °C, 92%.
S
OH
5
7
6
O
Figure 3
genic center is used to establish the relative and absolute configu-
rations of the C13 and C15 tetrahydropyran centers. Our plan was
to control this C11 configuration using an auxiliary-mediated ace-
tate aldol of aldehyde 11 with thiazolidinethione 12.8 Aldehyde 11
could conveniently be produced from readily available citronellene
(see Scheme 2). Selective epoxidation of (S)-citronellene with per-
acetic acid followed by hydrolysis of crude epoxide 10 and oxida-
tive cleavage provided aldehyde 11.9 Thiazolidinethione 12,
products was the discovery6 that in addition to their extensive
application as biochemical probes, Latrunculins A and B have also
shown potential therapeutic value by exhibiting the ability to in-
crease the aqueous humor outflow facility in normal tensive mon-
keys and may be useful for the treatment of glaucoma. As part of a
program to evaluate the latrunculins as potential clinical candi-
dates, we wanted to develop dependable methods to prepare gram
quantities of Latrunculins A and B. Encouraged by the work of
Fürstner, which demonstrated that a strained latrunculin macrocy-
cle could be formed containing a C6/C7 acetylenic bond and that
this bond could be reduced to the Z-olefin, we were interested in
establishing what the stereochemical outcome would be for the di-
rect olefin–olefin metathesis reaction of compound 4 (Fig. 3). A po-
sitive (Z-selective) outcome to the olefin–olefin metathesis
reaction would allow for direct access to the C8 stereogenic center
from the citronellene chiral pool and would avoid the unnecessary
use of low molecular weight alkynes as intermediates. This report
describes the preliminary results of this metathesis study and the
total synthesis of Latrunculin B (2) and its 6,7-E-isomer (3).
The retrosynthetic plan for the construction of metathesis pre-
cursor 4 (aldol condensation, THP ring formation, and intermolec-
ular ester formation, Fig. 3) required the assembly of three
subunits (5, 6, and 7) of fairly equal complexity. For the execution
of this study, thiazolidinone ketone 7 was obtained using the pub-
lished procedure of Smith.2a This decision obligated us to use the
PMB protecting group for the thiazolidinone nitrogen and to accept
the published challenges of removing this group late in the synthe-
sis with less than ideal yields.
readily available from D
-phenylalanine10 or commercially available
in kilogram quantities, was treated with titanium tetrachloride
(1.05 equiv) and diisopropyl-ethylamine (1.10 equiv) at À78 °C
for 1 h to generate the corresponding enolate. Addition of aldehyde
11 (1.00 equiv) delivered alcohol 13 in 76% isolated yield. The dia-
stereomeric ratio generated in this sequence was typically >7:1
and the minor isomer could easily be removed by flash chromatog-
raphy. Protection of the secondary alcohol as its triethylsilyl ether
and reductive removal of the chiral thiazolidinethione auxiliary
afforded aldehyde 5 directly.
With the three required fragments in hand, their assembly into
metathesis precursor 4 was undertaken (Scheme 3). Following the
established precedent of Fürstner and Smith, it was anticipated
that a reaction sequence entailing a Lewis acid mediated-aldol
addition followed by an acid catalyzed deprotection/cyclization/
equilibration step would lead to hemiketal 15 enriched in 15e. For-
tuitously, formation of the chlorotitanium enolate of ketone 7
Me
H
Me
Me
O
H
Me
The dienic acid 6, ultimately the C1/C6 portion of the macrocy-
cle, was readily prepared as shown in Scheme 1. A one-pot substi-
tution/deprotonation sequence with propargyl chloride and two
equivalents of allyl magnesium bromide provided the magnesium
acetylide intermediate,7 which cleanly produced ynoic ester 8
upon addition to ethyl-chloroformate. Stereoselective addition of
dimethyl cuprate followed by saponification of the resulting ester
afforded acid 6 (see Scheme 1).
OH
OH
5
Et3SiO
O
+
O
H
O
H
PMBN
a
S
OH
OH
O
PMBN
PMBN
7
S
S
O
O
15e
15a
b
Critical to the synthesis of the C7/C13 aldehyde 5 was establish-
ing the (S)-configuration for the C11-hydroxyl group. As in all other
syntheses of the latrunculins, the configuration of the C11 stereo-
Me
Me
O
Me
OH
c
O
O
H
O
H
PMBN
OMe
OMe
PMBN
Me
S
a
b
S
O
O
MgBr
16
4
+
Cl
O
CO2Et
8
6
OH
Scheme 3. Reagents and conditions: (a) TiCl4, (i-Pr)2NEt, CH2Cl2, À20 °C then 5, 60%
for 15a and 15e; (b) CSA, MeOH, toluene, 88%; (c) 2,6-lutidine, (CF3SO2)2O, CH2Cl2,
À15 °C; 6, NaH, 15-Crown-5, THF, rt, 61% for 2 steps.
Scheme 1. Reagents and conditions: (a) Et2O, À10 °C then EtO2CCl, 40%; (b)
Me2CuLi, THF, À78 °C; LiOH, MeOH, H2O, 92%.

300
J. She et al. / Tetrahedron Letters 50 (2009) 298–301
Table 1
Me
Me
R1
R2
Triene
Z-olefin
E-olefin
Ratio (Z:E)
Conversion (%)
O
O
Me
Me
PMB
PMB
H
Me
H
Me
H
4
18
19
17
20
22
24
2
21
23
25
3
22:78
39:61
25:75
33:67
100
100
97
O
O
a
O
H
O
H
H
96
OMe
OH
PMBN
HN
S
S
4
17
O
O
ation Grubbs catalyst led to no ring closing products. However,
treatment with the stronger second generation Grubbs catalyst
(G2) provided a 60% isolated yield of a 22:78 mixture of Z/E mac-
rolactones, respectively (see Fig. 4 and Table 1). It was also ob-
served that the Z/E ratio did not change during the complete
time course of the cyclization. Cyclization of the partially protected
substrates 18 (natural hemiketal functionality) and 19 (natural
thiazolidinone functionality) in dichloromethane provided E-die-
nyl lactones as the major products (39:61 and 25:75 Z/E ratios,
respectively). Ring closing metathesis of the ‘natural’ acyclic sub-
strate 17 gave a 33:67 mixture of Latrunculin B (2) and its E-isomer
(3), see Table 1. Analytically pure 2 and 3 were obtained by HPLC
separation of the crude reaction products generated from 17, or
alternatively from pure lactone precursors 20 and 21 using the
established CAN deprotection conditions published by Smith.
Interestingly, when 2 and 3 were independently re-exposed to a
higher loading (50 mol %) of the second generation Grubbs catalyst
in dichloromethane, isomerization of the C6/C7 olefin geometry
was not detected. This observation suggests that the generally ac-
cepted reversible mechanism of RCM reaction is not involved in the
chemistry of substrate 17. The lack of reactivity of 2 and 3 under
ring opening metathesis conditions might be attributed to the rigid
conformation of the 14-membered lactone and the steric hin-
drance of the neighboring methyl group. Kashman has previously
observed that the C6/C7 olefin of Latrunculin B was inert to vigor-
ous hydrogenation conditions.1b
b
c
Me
Me
O
O
Me
Me
O
O
O
H
O
H
OMe
OH
HN
PMBN
S
S
18
19
O
O
Scheme 4. Reagents and conditions: (a) CAN, CH3CN, H2O, 64%; (b) PTSA, CH3OH,
91%; (c) 1 N HCl, CH3CN, 57%.
Me
Me
Me
O
Me
O
Me
Me
O
O
Grubbs-G2
CH2Cl2, 40 o
O
+
O
H
N
O
O
H
O
H
R2
C
R2
R2
O
O
O
R1
R1
R1
N
N
S
S
S
O
O
O
Triene
Z-Olefin
Figure 4
E-Olefin
In summary, an evaluation of the potential to easily prepare
Latrunculin B from a direct olefin–olefin ring closing metathesis
reaction has been completed. The major product of the ring closing
metathesis reaction of several substrates with various remote pro-
tecting group patterns was the 6,7-E-geometry. A convergent syn-
thesis of metathesis precursor 4 has been developed from readily
available starting materials. Based on the current results, the selec-
tivity of the ring closure to form Latrunculin B is under kinetic
control.
(1.05 equiv TiCl4, 1.00 equiv DIPEA, CH2Cl2) at À20 °C followed by
the addition of aldehyde 5 resulted in the direct production of a
1:1 mixture of 15a and 15e in 60% overall yield. Use of the slightly
less stable triethylsilyl protecting group led to carbon-carbon bond
formation, 1,5-silyl transfer, ring closure, and subsequent loss of
the protecting group. Compounds 15a and 15e were readily sepa-
rated by flash chromatography. For the purpose of completing our
investigation, equatorial alcohol 15e was converted to methyl ketal
16 by stirring in mildly acidic methanol. Formation of the C13
equatorial triflate with its ensuing displacement with the sodium
salt of acid 6 gave axial ester 4.
As part of this study, we were also interested in examining the
effects of different protecting group patterns on the results of the
metathesis cyclization. Several ring closing metathesis approaches
to build medium-sized macrolactones have been published in the
literature.11 These previous studies reveal that a mixture of Z-
and E-isomers are often obtained using this approach. However,
in certain circumstances the E/Z selectivity could often be con-
trolled by changing the catalyst, the reaction conditions, and/or
the protecting group substitution pattern. Toward this end, com-
pound 4 could be completely deprotected by treatment with ceric
ammonium nitrate (CAN) in aqueous acetonitrile to provide 17.
Subsequently, the hemiketal of 17 could be converted to the
methyl ketal 19 by exposure to mild acid in methanol. Compound
4 could also be converted to the N-protected precursor 18 by ex-
tended exposure to aqueous acid (see Scheme 4).
Acknowledgments
The authors wish to thank Michael T. Crimmins for all of his
contributions to this work. The assistance of Dr. Wanda M. Bodnar
(Inspire Pharmaceuticals Inc.) and Dr. Jonathan L Bundy (Research
Triangle Institute) in obtaining HRMS data is also acknowledged.
Supplementary data
Experimental procedures as well as 1H and 13C NMR spectra for
all new compounds and Latrunculin B. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2008.10.144.
References and notes
1. (a) Oliveira, C. A.; Kashman, Y.; Mantovani, B. Chem.-Biol. Interact. 1996, 100,
141–153; (b) Blasberger, D.; Carmely, S.; Cojocaru, M.; Spector, I.; Shochet, N.
R.; Kashman, Y. Liebigs Ann. Chem. 1989, 12, 1171–1188; (c) Kashman, Y.;
Spector, I.; Carmely, S.; Blasberger, D. U.S. 4857538 A, 1989, 6 p; (d) Spector, I.;
Shochet, N. R.; Blasberger, D.; Kashman, Y. Cytoskeleton 1989, 13, 127–144; (e)
Blasberger, D.; Green, D.; Carmely, S.; Spector, I.; Kashman, Y. Tetrahedron Lett.
1987, 28, 459–462; (f) Kashman, Y.; Lidor, R.; Blasberger, D.; Carmely, S.
Exposure of the completely protected olefin–olefin metathesis
substrate 4 in refluxing dichloromethane to the milder first gener-

J. She et al. / Tetrahedron Letters 50 (2009) 298–301
301
Tetrahedron Lett. 1986, 27, 1367–1370; (g) Kashman, Y.; Groweiss, A.; Lidor, R.;
Blasberger, D.; Carmely, S. Tetrahedron 1985, 41, 1905–1914; (h) Groweiss, A.;
Shmueli, U.; Kashman, Y. J. Org. Chem. 1983, 48, 3512–3516; (i) Spector, I.;
Shochet, N. R.; Kashman, Y.; Groweiss, A. Science 1983, 219, 493–495; (j)
Kashman, Y.; Groweiss, A.; Shmueli, U. Tetrahedron Lett. 1980, 21, 3629–3632;
(k) Nèeman, I.; Fishelson, L.; Kashman, Y. Mar. Biol. 1975, 30, 293–296.
2. (a) Smith, A. B., III; Leahy, J. W.; Noda, I.; Remiszewski, S. W.; Liverton, N. J.;
Zibuck, R. J. Am. Chem. Soc. 1992, 114, 2995–3007; (b) Smith, A. B., III; Noda, I.;
Remiszewski, S. W.; Liverton, N. J.; Zibuck, R. J. Org. Chem. 1990, 55, 3977–3979;
(c) Smith, A. B. III; Zibuck, R.; Liverton, N. J. Studies in Organic Chemistry
(Amsterdam, 1986), 25 (New Synth. Methodol. Funct. Interesting Compd.),
183–202; (d) Zibuck, R.; Liverton, N. J.; Smith, A. B., III. J. Am. Chem. Soc. 1986,
108, 2451–2453.
3. (a) White, J. D.; Kawasaki, M. J. Org. Chem. 1992, 57, 5292–5300; (b) White, J. D.;
Kawasaki, M. J. Am. Chem. Soc. 1990, 112, 4991–4993.
4. For selected reviews see: (a) Gradillas, A.; Périz-Castells, J. Angew. Chem., Int. Ed.
Engl. 2006, 45, 6086–6101; (b) Grubbs, R. H.; Chang, S. Tetrahedron 1998,
54, 4413–4450; (c) Armstrong, S. K. J. Chem. Soc., Perkin Trans. 1 1998, 371–
388.
135–149; (c) Fürstner, A.; De Souza, D.; Turet, L.; Fenster, M. D. B.; Parra-
Rapado, L.; Wirtz, C.; Myott, R.; Lehmann, C. W. Chem.-A Eur. J. 2006, 13, 115–
134; (d) Fürstner, A.; Kirk, D.; Fenster, M. D. B.; Aissa, C.; De Souza, D.; Mueller,
O. PNAS 2005, 102, 8103–8108; (e) Fürstner, A.; Turet, L. Angew. Chem., Int. Ed.
Engl. 2005, 44, 3462–3466; (f) Fürstner, A.; De Souza, D.; Parra-Rapado, L.;
Jensen, J. T. Angew. Chem., Int. Ed. Engl. 2003, 42, 5358–5360.
6. (a) Kaufman, P. L.; Geiger, B. PCT Int. Appl. WO 9730701, 1997, 65 p; (b)
Peterson, J. A.; Tian, B.; Geiger, B.; Kaufman, P. L. Exp. Eye Res. 2000, 70, 307–
313.
7. Priebe, H.; Hopf, H. Angew. Chem., Int. Ed. Engl. 1982, 21, 286.
8. (a) Crimmins, M. T.; Shamszad, M. Org. Lett. 2007, 9, 149–152; (b) Zhang, Y.;
Phillips, A. J.; Sammakia, T. Org. Lett. 2004, 6, 23–25; (c) Romero-Ortega, M.;
Colby, D. A.; Olivo, H. F. Tetrahedron Lett. 2002, 43, 6439–6441; (d) Nagao, Y.;
Hagiwara, Y. J. Org. Chem. 1986, 51, 2391–2393.
9. Paquette, L. A.; Maynard, G. D. J. Am. Chem. Soc. 1992, 114, 5018–5027.
10. Thiazolidinethione acetate 12 can be produced by following the procedure to
make thiazolidinethione propionate: Crimmins, M. T.; Chaudhary, K. Org. Lett.
2000, 2, 775.
11. (a) Furstner, A.; Langemann, K. Synthesis 1997, 792–803; (b) Lee, C. W.; Grubbs,
R. H. Org. Lett. 2000, 2, 2145–2147; (c) Meng, D.; Su, D.-S.; Balog, A.; Bertinato,
P.; Sorenson, E. J.; Danishefsky, S. J.; Zheng, Y.-H.; Chou, T.-C.; He, L.; Horwitz, S.
B. J. Am. Chem. Soc. 1997, 119, 2733–2734.
5. (a) Fürstner, A.; Nagano, T.; Mueller, C.; Seidel, G.; Mueller, O. Chem.-A Eur. J.
2007, 13, 1452–1462; (b) Fürstner, A.; Kirk, D.; Fenster, M. D. B.; Aiessa, C.; De
Souza, D.; Nevado, C.; Tuttle, T.; Thiel, W.; Mueller, O. Chem.-A Eur. J. 2006, 13,