Synthesis and stereochemistry of the terminal spiroketal domain of the phosphatase inhibitor dinophysistoxin-2

Article abstract of DOI:10.1016/j.bmcl.2008.01.002

An expedient synthesis of both axially and equatorially C35 methyl substituted spiroketals representing the C28-C38 domain of the potent and selective protein serine/threonine phosphatase inhibitor dinophysistoxin-2 (DTX-2) was developed to enable comparative stereochemical analyses and a stereochemically correct total synthesis of DTX-2. Comparison of proton and carbon NMR data of the synthetic diastereomers with those published for DTX-2 indicates that DTX-2 possesses the (30S*,34R*,35S*)-relative configuration with an axial C35 methyl substituent.

Full text of DOI:10.1016/j.bmcl.2008.01.002

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3043–3046
Synthesis and stereochemistry of the terminal spiroketal
domain of the phosphatase inhibitor dinophysistoxin-2
Craig J. Forsyth* and Ce Wang
Department of Chemistry, The Ohio State University, 100 W. 18th Avenue, Columbus, OH 43210, USA
Received 7 November 2007; revised 27 December 2007; accepted 2 January 2008
Available online 8 January 2008
Abstract—An expedient synthesis of both axially and equatorially C35 methyl substituted spiroketals representing the C28–C38
domain of the potent and selective protein serine/threonine phosphatase inhibitor dinophysistoxin-2 (DTX-2) was developed to
enable comparative stereochemical analyses and a stereochemically correct total synthesis of DTX-2. Comparison of proton and
carbon NMR data of the synthetic diastereomers with those published for DTX-2 indicates that DTX-2 possesses the
(30S ,34R ,35S )-relative configuration with an axial C35 methyl substituent.
*
*
*
Ó 2008 Elsevier Ltd. All rights reserved.
Diarrhetic shellfish poisoning (DSP) continues to afflict
humans worldwide via the consumption of contami-
nated shellfish.1 Marine dinoflagellates of the genus
Dinophysis are among the primary producers of DSP
toxins,2 which include okadaic acid (OA, 1),3 dinophy-
sistoxin-1 (2, DTX-1), and dinophysistoxin-2 (3, DTX-
2, Fig. 1).4–6 We have previously documented the
importance of the terminal spiroketal domain of 1 for
the differential inhibition of serine/threonine protein
phosphatases 1 and 2A, the primary targets of DSP tox-
ins.7 Accordingly, the recent stereochemical clarification
of the terminal spiroketal domain of DTX-2 and an
interpretation of its selective phosphatase binding re-
ported by Larsen et al.8 prompted us to independently
examine the stereochemical assignments, as a prelude
to total syntheses of DTX-2 and its congeners. For this
purpose the syntheses and analyses of potential DTX-2
terminal spiroketal domain diastereomers are summa-
rized here.
Me
O
OH
O
O
H
H
H
O
R1
O
14
1
HO
O
O
27
HO Me
H
15 H
O
28
Me
35
OH
H
H
R2
Me
HO
R3
R1
Me
Me
H
R2
H
Me
H
R3
H
H
OA (1)
DTX-1 (2, 35
R)
DTX-2 (3, 35
S
)
Me
O
O
DTX-2 C28-C38
Domain Models:
28
HO
O
HO 28
O
35
35
H
H
Me
Me
(35S)-5
(35R)-4
Figure 1. Structures of major diarrhetic shellfish toxins and DTX-2
synthetic spiroketal models.
data, the C35 methyl substituent of DTX-1 is equatori-
ally oriented.4
The structure of DTX-1 was originally published by
Murata et al. in 1982, wherein the carbon skeleton of
DTX-1 was found to match that of OA, but with one
additional methyl substituent at C35 (Fig. 1).4 Accord-
A decade later DTX-2 was isolated and characterized by
the Wright group.6 Their published structural assign-
ment mirrored that of DTX-1, but without the methyl
substituent at C31 that is resident in both 1 and 2. Based
on NOE correlations, they assigned the DTX-2 C35
methyl group as being axially oriented. However, they
indicated that the same NOE data regarding the C35
methyl group of DTX-2 were obtained with DTX-1.4
The assumption that 2 and 3 bore the same C35 axial
substitution was propagated in the primary and second-
ary literature. In 1998, Sasaki et al. assigned the absolute
1
ing to the original H NMR vicinal coupling constant
Keywords: Dinophysistoxin-II; Phosphatase inhibitor; Stereochemis-
try; C35-configuration; Synthesis; Spiroketal.
*
Corresponding author. Tel.: +1 614 688 5347; fax: +1 614 292
1685; e-mail: forsyth@chemistry.ohio-state.edu
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.002

3044
C. J. Forsyth, C. Wang / Bioorg. Med. Chem. Lett. 18 (2008) 3043–3046
configuration of DTX-1, designating C35 as 35R, which
corresponds to equatorial substitution at C35.9 Most re-
cently, the published reports that DTX-1 and DTX-2
possess the same relative configuration at C35 have been
re-evaluated by Larsen et al. based upon comprehensive
de novo NMR spectroscopic analyses of the natural
products.8 Larsen and coworkers concluded that
DTX-1 and DTX-2 have equatorial and axial methyl
substituents at C35, respectively.8
9. Protecting group manipulations and subsequent oxi-
dation provided aldehyde 10. This aldehyde was coupled
with the boron enolate of N-propionyl oxazolidinone 11
to yield the syn aldol product 12 in 78% yield.12 Treat-
ment of 12 with pyridinium p-toluenesulfonate in meth-
anol initiated both deprotection of the latent primary
alcohol and subsequent thermodynamic dehydrative
cyclization to form spiroketals 13, epimeric at C35
(DTX-2 numbering). Attempts to chromatographically
separate the two C35 epimers at this stage were unsatis-
factory. However, after reductive excision of the oxazo-
lidinone auxiliary, the resultant diastereomeric primary
alcohols were separated successfully to provide the
equatorially C35-methyl-substituted spiroketal 4 and
the axially C35-methyl-substituted 5.
To independently examine the relative stereochemistry
about the C29–C38 domain of DTX-2 and provide ster-
eochemically correct intermediates for projected total
syntheses, synthetic fragments representing C28–C38
of DTX-2 with both equatorially (4) and axially (5) ori-
ented methyl-bearing stereogenic centers at C35 (Fig. 1)
were synthesized and analyzed. This approach allows a
direct spectroscopic comparison between synthetic
probes of known relative stereochemical configurations
and the terminal spiroketal domains of DTX-1 and
DTX-2 that have been in question. Moreover, the C28
hydroxyl functionalization of the stereochemical probes
4 and 5 parallels that of the C28–C8 intermediates used
in total syntheses of 110 and 7-deoxy-1.11
The relative configurations at C34 and C35 of 4 and 5 were
assigned on the basis of NOE data (Fig. 2). Irradiation of
the C35 methyl group of 4gave rise toNOE enhancements
of the resonances of both geminal C36 protons, as well as
of the C33 axial proton. Upon irradiation of the C35
methyl group of 5, NOE enhancements of the resonance
intensities of both protons at C33, the equatorial proton
at C36, and the axial proton at C37 were observed. Com-
bined, these experimental results indicate that the C35
methyl group of 4 is equatorially disposed on the thermo-
dynamically defined (34R)-spiroketal, whereas that of 5 is
axially oriented. Given the established aldol chemistry12
used to access 4 and 5 via 12, the absolute configurations
of our stereochemical probes are (29S,30S,34R,35R)-4
and (29S,30S,34R,35S)-5.
The synthesis of 4 and 5 commenced with d-valerolac-
tone 6 (Scheme 1). a-Methylation followed by nucleo-
philic opening of the lactone with the organolithium
derived from primary iodide 8 generated racemic ketone
O
O
The NMR spectral data8 of the terminal spiroketal do-
main of DTX-1 and DTX-2 were then compared with
those of 4 and 5. First, the H NMR chemical shifts
LDA, MeI, THF
O
O
1
-78 oC, 81% yield
O
6
7
were compared among these four compounds (Fig. 3),
which revealed some striking correlations. The chemical
shifts of the C36 axial protons of 5 and DTX-2 resonate
at 2.09 and 2.12 ppm, respectively, whereas the C36 ax-
ial protons of 4 and DTX-1 resonate at 1.50 and
1.52 ppm, respectively (CDCl3). Similarly, chemical
shifts of the C37 methylene protons of 5 (1.32 and
1.79 ppm, Dd = 0.47 ppm) and 4 (1.50 and 1.63 ppm,
Dd = 0.13 ppm) correlate best with those of DTX-2
(1.25 and 1.78 ppm, Dd = 0.53 ppm) and DTX-1 (1.42
and 1.61 ppm, Dd = 0.19 ppm), respectively. The third
set of diagnostic 1H NMR correlations involved the
1) t C
-BuLi, THF, -78 o
OPMB
I
9
2) 7, 78% yield
PMBO
OTBS
OH
8
1) TBSCl, Im, DMAP
CH2Cl2, 92%
O
2) DDQ, CH2Cl2
pH 7 buffer, 84%
3) NMO, TPAP
10
CH Cl , 81%
4
MS,
2
2
H
O
n
-Bu2BOTf, Et3N
O
O
O
O
OH
OTBS
0 oC, CH2Cl2
O
N
O
N
O
Me
Me
10, -78 oC to 0 oC, 71%
28
28
Bn
Bn
OH
OH
29
30
29
30
12
11
O
O
LiBH4, H2O
Et2O, 0 oC, 78%
PPTs, MeOH
78%
O
O
O
O
N
O
H
Ha
34
34
Bn
O
O
13
Ha
37
Ha
37
36
He
33
35
33
36
Me
35
He
He
O
O
Ha
+
28
HO 28
O
HO
O
35
35
Me
H
H
ax
eq
(29S
,30
S,34R,35R)-4
(29S,30S
,34R,35S)-5
(29
S,30S,34R
,35R
)-4
(29S,30S,34R,35S)-5
Scheme 1. Synthesis of DTX-2 terminal spiroketal models.
Figure 2. Observed NOEs involving the C35 methyl group of 4 and 5.

C. J. Forsyth, C. Wang / Bioorg. Med. Chem. Lett. 18 (2008) 3043–3046
3045
19.1
1.60
1.73
Ha
He
26.8
32.6
He
1.69
1.17
O
1.24
Ha
He
O
98.3
36.8
73.5
Ha
33
60.6
20.2
1.51
H
34.8
Ha
He
He 1.25
H
3.63
30
O
1.44
1.36
36.0
34
35
38
37
3.54
H
28
O
29
25.9
H
15.3
14.5
36
3.46
Ha
1.78
Me
H
Me
0.95
1.64
Ha He
0.99
DTX-2 C28-C38 Domain
18.7
1.29
2.12
C28-C38 Domain of DTX-2
25.9
32.2
1.74
Ha
1.65
He
O
98.4
66.5
73.3
60.6
19.9
1.79
1.42
He
Ha
38.9
1.28
HO
O
Ha
1.48
33
H
35.9
He
Ha
O
3.67
H
H
30
3.73
HO
3.63
25.9
11.9
(29
34
35
38
37
14.5
He
He
3.60
1.32
28
O
29
H
S,30
S
,34R,35S)-5
36
3.84
Me
Me
Ha 1.79
H
1.02
H
a
0.95
He
1.65
26.5
2.09
R
1.34
10.8
35.2
26.0
27.5
74.9
(29
S,30
S
,34
,35
S)-5
O
97.9
60.0
26.4
31.3
2.01 1.40
O
Ha
He
39.1
H
1.15
He
0.90
Me
He
1.90
27.6
16.0
Ha
O
1.76
H
33
16.9
Ha
3.59
H
30
0.99
DTX-1 C28-C38 Domain
18.5
34
38
37
3.51
He
He
28
O
29
1.61
35
H
36
3.25
Me
Me
1.42
26.4
Ha
31.6
H
0.90
0.92
1.50
He
Ha
O
98.4
66.9
73.0
1.63
1.52
60.1
26.0
38.9
C28-C38 domain of DTX-1
HO
O
38.6
H
1.87
1.63
27.6
11.5
16.9
Ha
He
H
He
Ha
1.44
1.40
1.70
O
a
(29S
,30S,34R,35R)-4
1.36
H
33
He
Figure 4. Comparative 13C NMR data of synthesized fragments 4 and
5 with natural DTX-18 and DTX-28 (CDCl3).
Ha
3.58
H
3.74
HO
30
3.68
34
35
38
37
3.58
1.63
28
He
He
O
29
H
36
3.89
Ha
1.50
Me
Me
H
0.96
1.55
0.90
He
ration at C35 of DTX-1 best matches that of 4 based
upon these H NMR correlations.
Ha
1
1.50 1.55
(29S,30S,34R,35R)-4
The 13C NMR data of 4, 5, DTX-1, and DTX-2 (Fig. 4)
were found to be fully in accord with the above analyses
of the 1H NMR data. The 13C NMR chemical shifts of 5
map very well upon those of DTX-2, whereas the 13C
NMR data for 4 correspond to those of DTX-1 at
C35-38, including the C35 methyl substituent. The dif-
ferences in 13C NMR chemical shifts between DTX-2
and 5 among C30–C38 are less than 0.5 ppm. Similarly,
the differences between the 13C NMR chemical shifts of
C34–C38 of DTX-1 and 4 are 60.5 ppm. In contrast,
there are clearly distinctive differences between the 13C
NMR chemical shift data of 5/DTX-2 versus 4/DTX-1.
Figure 3. Comparative 1H NMR data of probes 4 and 5 with natural
DTX-18 and DTX-28 (CDCl3).
methylene protons at C33. The relative chemical shifts
of the C33 axial and equatorial protons are inverted be-
tween 4 (C33-Haxial at 1.70 ppm, and C33-Hequatorial at
1.40 ppm, Dd = 0.30 ppm) and
5
(C33-Haxial at
1.28 ppm, and C33-Hequatorial at 1.79 ppm, Dd =
À0.47 ppm). In the chair conformation of cyclohexanes
the axial protons generally resonate downfield of the
geminal equatorial protons by ca 0.5 ppm.13 However,
in both DTX-2 (C33-Haxial at 1.17 ppm, and C33-Hequa-
torial at 1.69 ppm, Dd = À0.52 ppm) and 5, the C33 equa-
torial proton resonates downfield of the C33 axial
proton resonance, reflecting a strong differential stereo-
electronic effect of the C35 axial methyl substituent on
the C33 protons.14 The absence of a C35 axial methyl
substituent in both 4 and DTX-1 is associated with the
C33 methylene protons resonating at the typical differ-
ential chemical environments of axial versus equatorial
geminal protons. The relative stereochemistry in the
C29–C38 region of the natural product DTX-2 corre-
sponds best with that of 5, whereas the relative configu-
Combined, the spectral correlation data among 2–5 fully
support the relative steroechemical assignments of
(29S,30S,34R,35S) to 3 and 5, and of (29S,30S,
34R,35R) to 2 and 4. It is generally accepted that1–3 share
the same overall absolute configurations,3,6 beyond the
C35 epimers of 2 and 3. These stereochemical results are
in agreement with those of Larsen et al., who arrived at
the same conclusions based upon spectroscopic and com-
putational analyses of the natural products without the
aid of diastereomeric probes of known configurations.8

3046
C. J. Forsyth, C. Wang / Bioorg. Med. Chem. Lett. 18 (2008) 3043–3046
PP2A is less sensitive to DTX-2 than to OA.15 These
differential sensitivities have been ascribed to the
orientation of the C35 axial methyl substituent in
DTX-2 which incurs less favorable steric interactions
in the PP2A hydrophobic groove.8 On-going efforts in
our laboratories toward total syntheses of 2 and non-
natural structural variants thereof may allow the further
exploitation of differential contacts about the terminal
spiroketal domain to enhance selective phosphatase
binding.
3. Tachibana, K.; Scheuer, P. J.; Tsukitani, Y.; Kikuchi, H.;
Engen, D. V.; Clardy, J.; Gopichand, Y.; Schmitz, F. J.
J. Am. Chem. Soc. 1981, 103, 2469.
4. Murata, M.; Shimatani, M.; Sugitani, H.; Oshima, Y.;
Yasumoto, T. Bull. Jpn. Soc. Sci. Fish. 1982, 48, 549.
5. Carmody, E. P.; James, K. J.; Kelly, S. S. Toxicon 1996,
34, 351.
6. Hu, T.; Doyle, J.; Jackson, D.; Marr, J.; Nixon, E.;
Pleasance, S.; Quilliam, M. A.; Walter, J. A.; Wright, J. L.
C. J. Chem. Soc, Chem. Commun. 1992, 39.
7. Frydrychowski, V. A.; Urbanek, R. A.; Dounay, A. B.;
Forsyth, C. J. Bioorg. Med. Chem. Lett. 2001, 11, 647.
8. Larsen, K.; Petersen, D.; Wilkins, A. L.; Samdal, I. A.;
Sandvik, M.; Rundberget, T.; Goldstone, D.; Arcus, V.;
Hovgaard, P.; Rise, F.; Rehmann, N.; Hess, P.; Miles, C.
O. Chem. Res. Toxicol. 2007, 20, 868.
Acknowledgments
We thank C.O. Miles for sharing DTX-1 and 2 spectro-
scopic data and helpful discussion, Dr. T. Young for
assistance with NMR experiments, L. Weyer for edito-
rial assistance, and The Ohio State University for gener-
ous unrestricted financial support.
9. Sasaki, K.; Onodera, H.; Yasumoto, T. Enantiomer 1998,
3, 59.
10. Forsyth, C. J.; Sabes, S. F.; Urbanek, R. A. J. Am. Chem.
Soc. 1997, 119, 8381.
11. Dounay, A. B.; Urbanek, R. A.; Frydrychowski, V. A.;
Forsyth, C. J. J. Org. Chem. 2001, 66, 925.
12. Evans, D. A.; Nelson, J. V.; Vogel, E.; Taber, T. R. J. Am.
Chem. Soc. 1981, 103, 3099.
References and notes
13. Friebolin, H. Basic One- and Two-dimensional NMR
Spectroscopy; Wiley-VCH: Weinheim, 1998.
14. Klod, S.; Koch, A.; Kleinpeter, E. J. Chem. Soc., Perkin
Trans. 2002, 2, 1506.
15. Aune, T.; Larsen, S.; Aasen, J. A. B.; Rehmann, N.;
Satake, M.; Hess, P. Toxicon 2007, 49, 1.
1. Yasumoto, T.; Murata, M.; Oshima, Y.; Sano, M.;
Matsumoto, G. K.; Clardy, J. Tetrahedron 1985, 41, 1019.
2. Yasumoto, T.; Oshima, Y.; Sugawara, W.; Fukuyo, Y.;
Oguri, H.; Igarashi, T.; Fujita, N. Bull. Jpn. Soc. Sci. Fish.
1980, 46, 1405.