Multigram Synthesis of the C29-C51 Subunit and Completion of the Total Synthesis of Altohyrtin C (Spongistatin 2)

Article abstract of DOI:10.1021/ja030317+

A multigram synthesis of the C29-C51 subunit of altohyrtin C (spongistatin 2) has been accomplished. Union of this intermediate with the C1-C28 fragment and further elaboration furnished the natural product. Completion of the C29-C51 subunit began with th

Full text of DOI:10.1021/ja030317+

Published on Web 09/27/2003
Multigram Synthesis of the C29-C51 Subunit and Completion
of the Total Synthesis of Altohyrtin C (Spongistatin 2)
Clayton H. Heathcock,* Mark McLaughlin, Jesus Medina, Jed L. Hubbs,
Grier A. Wallace, Robert Scott, Michele M. Claffey, Christopher J. Hayes, and
Gregory R. Ott
Contribution from the Center for New Directions in Organic Synthesis,
Department of Chemistry, UniVersity of California, Berkeley, California 94720
Received May 27, 2003; E-mail: heathcock@cchem.berkeley.edu
Abstract: A multigram synthesis of the C29-C51 subunit of altohyrtin C (spongistatin 2) has been
accomplished. Union of this intermediate with the C1-C28 fragment and further elaboration furnished the
natural product. Completion of the C29-C51 subunit began with the aldol coupling of the boron enolate
derived from methyl ketone 8 and aldehyde 9. Acid-catalyzed deprotection/cyclization of the resulting
diastereomeric mixture of addition products was conducted in a single operation to afford the E-ring of
altohyrtin C. The diastereomer obtained through cyclization of the unwanted aldol product was subjected
to an oxidation/reduction sequence to rectify the C35 stereocenter. The C45-C48 segment of the eventual
triene side chain was introduced by addition of a functionalized Grignard reagent derived from (R)-glycidol
to a C44 aldehyde. Palladium-mediated deoxygenation of the resulting allylic alcohol was followed by
adjustment of protecting groups to provide reactivity suitable for the later stages of the synthesis. The
diene functionality comprising the remainder of the C44-C51 side chain was constructed by addition of an
allylzinc reagent to the unmasked C48 aldehyde and subsequent dehydration of the resulting alcohol.
Completion of the synthesis of the C29-C51 subunit was achieved through conversion of the protected
C29 alcohol into a primary iodide. The synthesis of the C29-C51 iodide required 44 steps with a longest
linear sequence of 33 steps. From commercially available tri-O-acetyl-D-glucal, the overall yield was 6.8%,
and 2 g of the iodide was prepared. The C29-C51 primary iodide was amenable to phosphonium salt
formation, and the ensuing Wittig coupling with a C1-C28 intermediate provided a fully functionalized,
protected seco-acid. Selective deprotection of the required silicon groups afforded an intermediate
appropriate for macrolactonization, and, finally, global deprotection furnished altohyrtin C (spongistatin 2).
This synthetic approach required 113 steps with a longest linear sequence of 37 steps starting from either
tri-O-acetyl-D-glucal or (S)-malic acid.
As discussed in the previous Article in this issue,1 the
spoingipyran natural products are potent marine cytotoxins that
have potential for use in cancer chemotherapy, and this
structurally unique family of compounds has attracted consider-
able synthetic interest.2 Our goal is to develop a total synthesis
of one of these molecules, altohyrtin C (spongistatin 2, 1), that
is efficient enough to provide multigram quantities of the natural
product for in vivo studies as a possible anticancer drug. Our
synthetic approach assembles the altohyrtin macrocycle from
suitably protected versions of two major building blocks, the
C1-C28 unit 2 and the C29-C51 unit 3. In the preceding
Article, we have described a greatly improved process for the
C1-C28 fragment of the altohyrtins, leading to almost 10 g of
a functional derivative of 2. In this Article, we report a
multigram synthesis of a protected version of the C29-C51
subunit 33 and the method whereby these two advanced
intermediates have been fashioned into 0.25 g of altohyrtin C.
As in the published syntheses, the synthetic strategy was
designed around macrolactonization as the final key step,
preceded by a Wittig reaction to form the C28-C29 (Z)-alkene
bond.
Our revised plan for the C29-C51 unit breaks the target down
into four units, the most complex of which is the methyl ketone
4, which we have previously obtained in 11 steps from
commercially available tri-O-acetyl-D-glucal.3 The other pro-
jected building blocks are aldehyde 5, prepared by the use of
Evans’ oxazolidone chemistry, vinyl bromide 6, and an allyl-
organometallic reagent that would be used to append the three
terminal carbons of the side chain.
In our previous work,3 the aldol coupling of methyl ketone 4
with aldehyde 5 was conducted via the lithium enolate. In this
work, we discovered that the initial lithium aldolate rapidly
reacts with a second equivalent of aldehyde, leading to a
complex mixture of 2:1 adducts that could be converted into
the desired aldol 7 only by treatment with methanolic K2CO3.
Although the desired aldol product could be liberated by this
(1) Hubbs, J. L.; Heathcock, C. H. J. Am. Chem. Soc. 2003, 125, 12836.
(2) A comprehensive reference list for previous synthetic work, presented in
ref 1, is omitted here in the interest of conserving journal space.
(3) For our first-generation synthesis of a C29-C44 fragment, see: Wallace,
G. A.; Scott, R. W.; Heathcock, C. H. J. Org. Chem. 2000, 65, 4145.
9
12844
J. AM. CHEM. SOC. 2003, 125, 12844-12849
10.1021/ja030317+ CCC: $25.00 © 2003 American Chemical Society

Completion of the Total Synthesis of Altohyrtin C
A R T I C L E S
Scheme 1 a
a (a) Cy2BCl, Et3N, CH2Cl2, -78 °C, then aldehyde 5; (b) Ph3P‚HBr,
MeOH, THF (1:1.1 mixture of 8:9); (c) Dess-Martin periodinane, pyridine,
CH2Cl2; (d) L-Selectride, THF, -78 °C; (e) KHMDS, TBSCl, THF; (f)
TBAF, THF.
method in good yield (88%, 3.4:1.0 mixture of diastereomers
at C35, favoring the unwanted epimer), the wasteful nature of
this procedure with respect to the aldehyde prompted investiga-
tion of an alternative. To this end, generation of the boron
enolate of ketone 4 by the action of dicyclohexylboron chloride
and triethylamine, followed by the addition of 1.05 equiv of
the aldehyde 5, afforded the aldol 7 directly (Scheme 1). These
aldol products were then subjected to catalytic amounts of
triphenylphosphine hydrobromide in methanol/tetrahydrofuran
to effect deprotection of the triethylsilyl ether and cyclization
to the methyl ketals 8 and 9 in a single operation, in 85% overall
yield from the methyl ketone 4. The 1.1:1.0 mixture of
diastereomers obtained by this procedure was separable by flash
column chromatography, and the undesired epimer was recycled
in 83% yield through an oxidation-reduction sequence, which
employed the Dess-Martin periodinane4 and L-Selectride.
Having established the method by which the configuration at
C35 in the unwanted diastereomer could be rectified, the
resulting axial hydroxy group of 9 was protected as a TBS ether,
and the primary TIPS ether was selectively removed using
TBAF, providing alcohol 10 in 76% yield for the two steps.
With relatively large quantities of alcohol 10 in hand, attention
was turned to the synthesis of the vinyl bromide 6 that would
form the foundation for what would ultimately become the triene
side chain of altohyrtin C. It was envisaged that the C47
stereocenter could be derived from a suitably protected glycidol
by regioselective opening of the epoxide with an acetylide anion.
Such a reaction, utilizing (R)-glycidyl pivaloate and lithium
(trimethylsilyl)acetylide, had indeed been previously reported.5
On a larger scale, however, it was preferable to use (R)-glycidol
triethylsilyl ether 11 to avoid unwanted side reactions (Scheme
2). The crude product from the epoxide opening was treated
sequentially with pyridinium p-toluenesulfonate and K2CO3 to
cleave the TES and TMS groups in 73% overall yield. The
(4) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
(5) MacMillan, D. W. C.; Overman, L. E.; Pennington, L. D. J. Am. Chem.
Soc. 2001, 123, 9033 and references therein.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12845

A R T I C L E S
Heathcock et al.
Scheme 2 a
Scheme 3 a
a (a) TMSA, n-BuLi, BF3‚OEt2, THF, -78 °C; (b) PPTS, MeOH/THF;
(c) K2CO3, MeOH, THF; (d) PivCl, pyridine, CH2Cl2; (e) 9-Br-BBN,
CH2Cl2, 0 °C, then AcOH; (f) TBSCl, imidazole, DMF; (g) DIBAL-H,
toluene, -78 °C; (h) TESCl, EtNiPr2, DMF; (i) t-BuLi, THF, -78 °C, then
MgBr2‚OEt2.
primary hydroxyl of the resulting alkynediol 12 was then
selectively acylated with pivaloyl chloride, and the vinyl
bromide was introduced by bromoboration6 of the alkyne.
Subsequent protection of the secondary hydroxyl of 13 as a TBS
ether, followed by DIBAL-H mediated cleavage of the pivaloyl
group and reprotection of the primary alcohol as a TES ether,
provided vinyl bromide 6.7 The Grignard reagent 6a, required
for coupling with the EF bis-pyran subunit, was available by
successive treatment with tert-butyllithium and magnesium
bromide etherate.
The union of the bis-pyran subunit 10 with the functionalized
Grignard reagent 19 was performed as shown in Scheme 3.
Oxidation of the primary alcohol 11 under Moffat-Swern
conditions8 yielded a rather unstable aldehyde, which was used
in crude form. Treatment of the oxidation reaction mixture with
an excess of Grignard reagent 6a afforded the desired alcohol
15 in 87% yield as a single diastereomer. Although the
configuration of the product alcohol was assumed to be that
shown, based on literature precedent9 for similar additions of
vinyl Grignard reagents to aldehydes, the stereochemical
outcome of this transformation was ultimately inconsequential
because the ensuing synthetic steps would eliminate this
stereocenter. Along these lines, formylation using Katritzky’s
N-formylbenzotriazole10 provided the allylic formate 16 in 98%
yield and set the stage for a deoxygenation reaction in which it
was required that the exoskeletal position of the alkene was
maintained. We sought to accomplish this goal by a modification
of the π-allyl reduction method of Tsuji.11 Through systematic
investigation, it was found that less polar solvents in combination
with a high ratio of tributylphosphine to palladium (200:1) gave
a >20:1 ratio of the desired exoskeletal alkene 17, relative to
the internal alkene isomer. The observed effect of the excess
exogenous ligand on the ratio of alkene products suggests that
greater steric bulk at the palladium promotes intramolecular
a (a) (COCl), DMSO, CH2Cl2, -78 °C, then EtNiPr2; (b) Grignard 6a,
THF, -78 °C; (c) N-formylbenzotriazole, EtNiPr2, DMAP, CH2Cl2; (d) 10
mol % Pd(PPh3)4, 2000 mol % Bu3P; ammonium formate, cyclohexane,
80 °C; (e) DDQ, CH2Cl2/pH 7 buffer, 89%; (f) Ph3P‚HBr,MeOH/THF, 0
°C; (g) TESCl, imidazole, DMF.
hydride transfer to the more substituted position, leaving the
Pd coordinated to the less substituted double bond. Under the
optimized reaction conditions, deoxygenation occurred ef-
ficiently to give alkene 17 in 86% yield, together with a minor
1
side product, the H NMR spectrum of which was consistent
with loss of the TES protecting group. Treatment of the
recovered primary alcohol with TESCl led to an additional 4%
of alkene 17 and brought the total yield up to 90%. Thus, the
Grignard addition, allylic deoxygenation protocol gives alkene
17 in 77% overall yield for the three steps.
Having demonstrated a successful approach to a C29-C48
fragment 17 of the altohyrtins, effort was then focused on the
elaboration of this substrate to include the three carbons atoms
that form the end of the triene side chain of altohyrtin C. A
concern at this point, however, was the stability of a fully formed
diene functionality to the oxidative conditions necessary to effect
removal of the three PMB groups at a late stage in the synthesis.
Indeed, our studies on model systems12 had indicated competing
(6) (a) Hara, S.; Dojo, H.; Takinami, S.; Suzuki, A. Tetrahedron Lett. 1983,
24, 731. (b) Hara, S.; Satoh, Y.; Ishiguro, H.; Suzuki, A. Tetrahedron Lett.
1983, 24, 735.
(7) Although this sequence begins and ends with a triethylsilyl protecting group
on the primary hydroxyl, an intermediate pivaloyl group was necessary
because the acidic conditions of the bromoboration reaction were not
compatible with the silyl ether.
(8) For reviews, see: (a) Tidwell, T. T. Synthesis 1990, 857. (b) Tidwell, T.
T. Org. React. 1990, 39, 297-572.
(9) (a) Guo, J.; Duffy, K. J.; Stevens, K. L.; Dalko, P. I.; Roth, R. M.; Hayward,
M. M.; Kishi, Y. Angew. Chem., Int. Ed. 1998, 37, 187-192. (b) Hayward,
M. M.; Roth, R. M.; Duffy, K. J.; Dalko, P. I.; Stevens, K. L.; Guo, J.;
Kishi, Y. Angew. Chem., Int. Ed. 1998, 37, 192-196.
(10) Katritzky, A. R.; Chang, H. X.; Yang, B. Synthesis 1995, 503.
(11) (a) Tsuji, J.; Yamakawa, T. Tetrahedron Lett. 1979, 7, 613. (b) Tsuji, J.;
Shimizu, I.; Minami, I. Chem. Lett. 1984, 1017.
(12) Ott, G. R.; Heathcock, C. H. unpublished results. These model studies were
carried out on the substrate reported in: Ott, G. R.; Heathcock, C. H. Org.
Lett. 1999, 1, 1475.
9
12846 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Completion of the Total Synthesis of Altohyrtin C
A R T I C L E S
Scheme 4 a
Scheme 5 a
a (a) Ph3P‚HBr, MeOH/CH2Cl2, 0 °C; (b) (COCl)2, DMSO, CH2Cl2, -78
°C, then EtNiPr2; (c) allylMgBr/ZnCl2, Et2O, -78 °C; (d) Martin’s sulfurane,
CHCl3; (e) DIBAL-H, toluene, -78 °C; (f) MsCl, Et3N, CH2Cl2, 0 °C; (g)
NaI, NaHCO3, Na2SO3, acetone, 56 °C.
diene oxidation to a dieneone under the influence of DDQ, and
Paterson had reported a similar problem in related systems.13
On the basis of these observations, we decided to exchange the
PMB ethers for TES ethers because this would direct our
synthesis toward late-stage persilylated intermediates very
similar to those already proven to be viable by other synthetic
groups. Thus, before commencing with the installation of the
diene functionality, PMB-protected intermediate 17 was treated
with 4 equiv of DDQ in dichloromethane/pH 7 buffer to provide
triol hemiketal 18 in 89% yield. Under the aqueous reaction
conditions, the methyl ketal suffered hydrolysis, a result
observed previously by Kishi9 and Paterson.13 Treatment of the
hemiketal with catalytic acid in methanol simultaneously
regenerated the methyl ketal and cleaved the primary TES ether
to give a tetraol, which upon exposure to excess TESCl afforded
compound 19.
With an appropriately protected C29-C48 fragment in hand,
completion of the diene construction was now possible and
began with selective deprotection of the primary TES ether of
19 under mild acidic conditions (Scheme 4). Although the
success of this deprotection was extremely sensitive toward acid
concentration and reaction time, careful monitoring by TLC
allowed us to obtain the desired primary alcohol 20 in 80%
yield. Ammonium fluoride in methanol was examined as a
potential alternative reagent for this delicate transformation but
proved too sluggish and gave lower yields. Oxidation of primary
alcohol 20 was conducted using Moffat-Swern conditions, and
the crude aldehyde was treated with an excess of allylzinc
a (a) Ph3P, MeCN, EtNiPr2, 80 °C; (b) MeLi‚LiBr, THF, -78 °C, then
aldehyde 24, 55-60% (two steps).
reagent14 to provide a 2:1 diastereomeric mixture of alcohols.
When these alcohols were subjected to Martin’s sulfurane,15
dehydration occurred to give the (E)-alkene 21 in 84% overall
yield from alcohol 20. Finally, in preparation for the Wittig
coupling with the C1-C28 (ABCD) portion of altohyrtin C,
the C29 pivaloyl group was cleaved with DIBAL-H, and the
resulting primary alcohol 22 was converted into iodide 23 by
mesylation, followed by the modified Finkelstein conditions
utilized by Evans.16
Conversion of iodide 23 into a phosphonium salt required
the use of 3 equiv of triphenylphosphine. Purification of the
phosphonium salt was briefly investigated, and, although it was
possible to remove the excess triphenylphosphine by flash
column chromatography, significant material losses were suf-
fered. Thus, for practical reasons, the phosphonium salt was
used crude in the Wittig coupling with C1-C28 aldehyde 24,1
a reaction we anticipated might be problematic based on the
experience of other synthetic groups. Moderate to low yields
for similar Wittig reactions have been observed in all of the
other completed spongistatin total syntheses, with the exception
of the most recently reported synthesis of altohyrtin C by
Crimmins,17 in which a Wittig reaction between a closely related
phosphonium salt and a truncated aldehyde comprising C16-
C28 of altohyrtin C occurred in 86% yield when MeLi‚LiBr
(13) Paterson, I.; Cowden, C. J.; Rahn, V. S.; Woodrow, M. D. Synlett 1998,
915. In contrast, a similar deprotection of a bis-PMB ether was achieved
without difficulty in the syntheses of altohyrtin A by Kishi (ref 13) and by
Paterson (Paterson, I.; Chen, D. Y.-K.; Coster, M. J.; Acen˜a, J. L.; Bach,
J.; Gibson, K. R.; Keown, L. E.; Oballa, R. M.; Trieselmann, T.; Wallace,
D. J.; Hodgson, A. P.; Norcross, R. D. Angew. Chem., Int. Ed. 2001, 40,
4055-4060). It appears that the presence of the side-chain halogen retards
the oxidation process that we had observed, permitting conventional removal
of the PMB groups.
(14) Scarlato, G. R.; DeMattei, J. A.; Chong, L. S.; Ogawa, A. K.; Lin, M. R.;
Armstrong, R. W. J. Org. Chem. 1996, 61, 6139.
(15) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327.
(16) Evans, D. A.; Trotter, B. W.; Coleman, P. J.; Coˆte´, B.; Dias, L. C.;
Rajapakse, H. A.; Tyler, A. N. Tetrahedron 1999, 55, 8671-8726.
(17) Crimmins, M. T.; Katz, J. D.; Washburn, D. G.; Allwein, S. P.; McAtee,
L. F. J. Am. Chem. Soc. 2002, 124, 5661.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12847

A R T I C L E S
Heathcock et al.
Scheme 6 a
a (a) TBAF, THF, 0 °C; (b) 2,4,6-trichlorobenzoyl chloride, EtNiPr2, then DMAP, toluene, 90 °C; (c) aqueous HF, MeCN, -15 °C.
was used as the base to generate the ylide. In a similar fashion,
and despite poor and irreproducible results during initial attempts
with alternative bases (LiHMDS, KHMDS, n-BuLi), the adop-
tion of Crimmins’ method with our system led to a repeatable
55-60% overall yield of the (Z)-alkene Wittig product 25 from
iodide 24 (Scheme 5).
possible reactive alcohols were allayed by previous observations
with the spongistatins by other groups, following the initial
synthesis by Evans and co-workers.16 Thus, following a modi-
fied Yamaguchi procedure, the intermediate active mixed
anhydride was cyclized in toluene at 90 °C to the 42-membered
ring 27 in 80% yield.
With the complete carbon skeleton of altohyrtin C now
assembled, the tasks of selective deprotection, macrolactoniza-
tion, and final, global deprotection could be addressed. As
alluded to previously, our synthesis had now arrived at a late-
stage intermediate very similar to those featured in published
syntheses of altohyrtin C, and, consequently, it was hoped that
similar reaction conditions would provide success. For the
selective deprotection of a C1-TIPS ester and a TES ether at
C41, Evans had reported the use of HF‚pyridine buffered with
pyridine.16 Protected seco-acid 25 also required cleavage of the
same two silicon groups and an additional TES ether at C42.
The expectation that compound 25 would be more robust than
the Evans intermediate due to the increased steric encumbrance
around C41/C42 proved correct, and the use of HF‚pyridine in
pyridine led to unsatisfactory results; even after long reaction
times, starting material remained, and the selectivity was
compromised. After experimenting with various reagents for
silyl ether cleavage (triethylamine trihydrofluoride, ammonium
fluoride), we determined that the optimum method for conduct-
ing this transformation involved the slow addition of 3 equiv
of TBAF to a solution of the protected seco-acid 25 at 0 °C
(Scheme 6). This procedure cleanly afforded partially depro-
tected seco-acid 26 in 80% yield and allowed progress to the
macrolactonization. Concerns regarding regioselectivity in the
macrolactonization of a substrate such as 26 containing two
All that remained was the global desilylation and hydrolysis
of the methyl ketal to the hemiketal. Once again, there was
ample literature precedent, all of which suggested the employ-
ment of relatively dilute aqueous HF in acetonitrile. However,
subjection of protected macrolactone 27 to the same conditions
produced a mixture of starting material and products from which
altohyrtin C was isolated in relatively low yield. Given our long-
term goal of synthesizing multigram quantities of the material,
such inefficiency in this crucial ultimate reaction was unac-
ceptable. Consequently, it was pleasing to discover alternative
conditions that bring about clean deprotection of macrolactone
27. The initial result with the dilute HF prompted testing of
more concentrated acid solutions and shorter reaction times, with
a view to increasing the yield of the natural product. These
conditions provided total conversion of the starting material,
but, despite a relatively clean TLC, the crude isolated yield was
only around 50%. Moreover, it was clear that this more
promising yield would be further diminished upon recognition
that what appeared to be a single compound by TLC analysis
was actually a mixture of products by 1H NMR. Further
chromatographic analysis revealed a minor component (ca. 5%)
that could be separated on a flash column and was subsequently
1
identified from its H NMR and mass spectra as an E-ring
dehydrated analogue 28. Paterson recently reported a similar
observation in his work on spongistatin 1 and, furthermore,
9
12848 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Completion of the Total Synthesis of Altohyrtin C
A R T I C L E S
found that the dehydrated version of spongistatin 1 was even
more biologically active than the parent natural product in in
vitro tests.18 Preliminary results in the altohyrtin C series suggest
that this dehydration product of altohyrtin C is exceedingly
potent, possibly even more potent than altohyrtin C itself in a
number of tumor cell lines.19
In summary, the synthesis of the C29-C51 subunit of
altohyrtin C requires a total of 44 steps and delivers the subunit
in 6.8% yield over the longest linear sequence starting from
the commercially available tri-O-acetyl-D-glucal. The full total
synthesis of altohyrtin C that is reported here requires 113 total
steps with a longest linear sequence of 37 steps from either
(S)-malic acid or tri-O-acetyl-D-glucal. A total of 0.25 g of the
marine natural product has been prepared by this route, and,
although still far from our long-term goal, this quantity is more
altohyrtin C (spongistatin 2) than has ever existed from the
original isolation and all previous syntheses combined. Continu-
ing efforts in our laboratory have the goal of further refining
our synthetic route such that we may ultimately achieve the
synthesis of multigram quantities of altohyrtin C (spongistatin
2), thereby facilitating meaningful in vivo studies with this
fascinating family of natural cytotoxins.
While these findings were certainly of interest, the yield of
the desired natural product was still less than satisfactory.
Further work on this final deprotection led to the optimum
conditions in which the use of fairly concentrated HF in
combination with lower temperature, a shorter reaction time,
and a low-temperature quench furnished clean altohyrtin C (1,
spongistatin 2) in 92% isolated yield. Relative to our initial
efforts, this represented a 3-fold increase in the yield of this
valuable macrolide and should be of significant benefit in future
attempts to prepare even greater quantities of altohyrtin C. The
synthetic (+)-altohyrtin C obtained from this work was identical
1
in all respects (500 MHz H NMR, LR-MS, HR-MS, optical
Acknowledgment. This work was supported by a research
grant from the United States Public Health Service (AI15027).
The Center for New Directions in Organic Synthesis is supported
by Bristol-Myers Squibb as a Sponsoring Member and Novartis
as a Supporting Member.
rotation, TLC, and HPLC) to an authentic sample kindly
provided by Professor Evans. Additionally, 13C NMR data for
synthetic (+)-altohyrtin C (spongistatin 2) were in agreement
with data originally obtained from the natural product by
Professor Pettit.
Supporting Information Available: Experimental procedures
and characterization for all new compounds, 1H and 13C NMR
spectra of synthetic altohyrtin C, 1H NMR spectrum of dehydra-
tion product 28, charts showing the convergency of the synthesis
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
(18) Paterson, I.; Acen˜a, J. L.; Bach, J.; Chen, D. Y.-K.; Coster, M. J. Chem.
Commun. 2003, 462.
(19) Note Added in Proof. Compound 28 exhibits picomolar GI50 against
several tumor cell lines (e.g., 2 × 10-11 in HL-60 leukemia; 5 × 10-11 in
HCT-116 colon cancer; 3 × 10-11 in UACC-62 melanoma; 3 × 10-11 in
MDA-MB-435 breast cancer). In addition, this analogue has nanomolar
GI50 against a number of other cancer lines, including eight nonsmall cell
lung cancers, five other colon cancers, four CNS cancers, six other
melanomas, six ovarian cancers, seven renal cancers, and two other breast
cancers. Edward Sausville, National Cancer Institute, private communica-
tion, August 14, 2003.
JA030317+
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12849