Absolute configuration of brevisamide and brevisin: Confirmation of a universal biosynthetic process for Karenia brevis polyethers

Article abstract of DOI:10.1021/np100159j

The discovery of brevisin, the first example of an interrupted polycyclic ether, obtained from the dinoflagellate Karenia brevis, posed some important questions regarding the mechanism of the cyclization process. Consequently, we have established absolute configurations of brevisin and its related metabolite brevisamide using a modified Mosher's esterification method. For brevisin, analysis was carried out on both the 31-monokis-and the 10,31-bis-MTPA esters. The results suggest that both metabolites, like other polyethers from K. brevis, result from polyepoxide precursors with uniform (S, S) configurations for all epoxides and provide further support for a universal stereochemical model for dinoflagellate polyether formation.

Full text of DOI:10.1021/np100159j

J. Nat. Prod. 2010, 73, 1177–1179
1177
Absolute Configuration of Brevisamide and Brevisin: Confirmation of a Universal Biosynthetic
Process for Karenia breWis Polyethers
Ryan M. Van Wagoner,† Masayuki Satake,‡ Andrea J. Bourdelais,† Daniel G. Baden,† and Jeffrey L. C. Wright*,†
Center for Marine Science, UniVersity of North Carolina, Wilmington, 5600 MarVin K. Moss Lane, Wilmington, North Carolina 28409, and
Department of Chemistry, School of Science, The UniVersity of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
ReceiVed March 10, 2010
The discovery of brevisin, the first example of an “interrupted” polycyclic ether, obtained from the dinoflagellate Karenia
breVis, posed some important questions regarding the mechanism of the cyclization process. Consequently, we have
established absolute configurations of brevisin and its related metabolite brevisamide using a modified Mosher’s
esterification method. For brevisin, analysis was carried out on both the 31-monokis- and the 10,31-bis-MTPA esters.
The results suggest that both metabolites, like other polyethers from K. breVis, result from polyepoxide precursors with
uniform (S, S) configurations for all epoxides and provide further support for a universal stereochemical model for
dinoflagellate polyether formation.
More than 20 years after their initial discovery, the fused-ring
polyethers of dinoflagellates remain subjects of intense study.
Although many recent studies have identified PKS gene fragments
in toxin-producing dinoflagellates, the inherent complexity of
dinoflagellate genome structure and posttranscriptional RNA pro-
cessing have proven formidable obstacles toward the goal of
identifying a complete biosynthetic pathway for a known toxin.1-5
In spite of this, the ongoing discovery of new polyether metabolites
Figure 1. Observed ∆δS-R values for the MTPA ester of brevisa-
mide (1).
and the development of new and innovative techniques for their
synthesis have continued to expand both the information available
for and the interest in the processes leading to their production. In
particular, many researchers have focused on the potential formation
of these compounds via epoxide hydrolase-mediated cascades from
polyepoxide precursors, by analogy to bacterial polyether com-
pounds.6-13 The recent discovery of brevisamide (1)12 from a
brevetoxin-producing strain of the dinoflagellate Karenia breVis has
shed light on the process of initiation of these cascades. The
subsequent discovery of brevisin (2)13 as an “interrupted” polyether
was surprising and posed some questions as to the universality of
the polyether cyclization process. It was reasoned that knowledge
of the absolute configuration of these compounds would provide
crucial details in this regard since stereochemical observations first
led to the polyepoxide hypothesis and subsequent extensions of
it.6-9,14 However, initial limitations in the amount of these
metabolites prevented the determination of the absolute configu-
ration of both products, but with the accumulation of additional
material we now report the determination of the absolute config-
uration of brevisamide (1) and brevisin (2).
The esterification of brevisin (2) with MTPA was carried out at
4 °C using stoichiometric control of reagents. This was found to
be necessary after preliminary experiments yielded instead the
tetrakis-MTPA ester of brevisin, leading to interactions in ∆δS-R
effects among the different modification sites. Complete NMR
assignments of all hydrogen atoms were made for the 31-monokis-
and the 10,31-bis-MTPA esters based on 1H NMR, TOCSY, MQ-
COSY, ROESY, HSQC, and HMBC experiments. The esterification
sites were identified on the basis of downfield shifts in δC for the
esterified alcohol carbon atoms relative to brevisin (2) and by the
disappearance of the corresponding hydroxy proton signals. Com-
parison of the derived ∆δS-R values for the monokis-ester to those
of the bis-ester showed that the two corresponded within (0.02
ppm for positions 29-33 and 39, whereas for all other positions
the monokis-ester had ∆δS-R values of 0.00 ( 0.03 ppm. This
indicates that the effects of the two esterification sites do not
interfere with one another in any way. Figure 2 shows the observed
∆δS-R values for 10,31-bis-MTPA-brevisin. Each of the two sites
shows a clear pattern of ∆δS-R values with negative values on one
side and positive values on the other, following the expected pattern
for the Mosher method16 and allowing the unambiguous assignment
of an S configuration for both C-10 and C-31. Thus, by using the
relative stereochemical relationships noted previously for brevisin
The R- and S-MTPA monoesters of brevisamide (1) were readily
formed using DMAP and triethylamine as bases. Assignments of
δH were made for all protons in the two derivatives using a
combination of 1H NMR, TOCSY, MQ-COSY, HSQC, and HMBC
experiments. The resulting ∆δS-R values for each position are shown
in Figure 1. Using a traditional modified Mosher’s method analysis
of the results indicates an 11-S configuration, which, combined with
the relative configuration originally reported12 and subsequently
confirmed by synthesis,15 indicates the absolute configuration shown
in Figure 1. This configuration is identical to that of synthetic
brevisamide (1), confirming the earlier observation of the synthetic
compound having a specific rotation similar in sign to that of the
natural product, though differing in magnitude.15
* To whom correspondence should be addressed. Phone: 910 962 2397.
Fax: 910 962 2410. E-mail: wrightj@uncw.edu.
† University of North Carolina, Wilmington.
‡ The University of Tokyo.
10.1021/np100159j  2010 American Chemical Society and American Society of Pharmacognosy
Published on Web 06/07/2010

1178 Journal of Natural Products, 2010, Vol. 73, No. 6
Notes
Figure 2. Observed ∆δS-R values for the 10,31-bis-MTPA-ester of brevisin (2).
Figure 3. Illustration of how uniform configuration in epoxides might be achieved by a promiscuous epoxidase that forms contacts with an
acyl carrier protein (ACP).
(2),13 it is possible to assign the absolute configuration of brevisin
as shown in Figure 2.
is oxidized if one views the nascent chain with the side of
attachment to the PKS as an orienting landmark (e.g., always on
the left side; Figure 3). If our proposal for the direction of ring
formation with respect to the direction of chain formation is correct,
this in turn suggests that the formation of epoxides occurs at a time
when the polarity of direction of the nascent chain can be
distinguished, i.e., before the nascent chain has been cleaved from
the PKS. Similarly, the oxidative cyclization of precursors to
bacterial polyethers has been proposed to take place on a polyketide
chain tethered to a dedicated ACP in the monensin pathway.10 This
also suggests that the logic underlying epoxide formation is simple,
requiring only an E olefin of variable substitution with stereochem-
ical orientation being provided by attachment of one end of the
nascent chain to the PKS. As such, it is possible that a very small
set of oxygenases (possibly a single enzyme) is required for this
aspect of producing all polyethers in K. breVis.9
The absolute configurations for brevisamide and brevisin fit well
within the framework of polyether production in K. breVis. For both
compounds, as for brevenal, PbTx-1, and PbTx-2, the carbon atom
bearing the alcohol group (or ester in the case of the brevetoxins)
ꢀ to the ether oxygen on the terminal ring has an S configuration.
Indeed, for brevisin both such carbinol moieties have the same
configuration. In a ring-closing cascade based on a polyepoxide,
this position must retain the configuration of the precursor epoxide
because it occurs on the carbon atom that is not subject to
nucleophilic attack. This suggests that the stereochemical uniformity
originally noted for the brevetoxins and related compounds extends
to brevisamide (1) and brevisin (2),9 which appear to result from
polyepoxide precursors where every epoxide group has an (S, S)
configuration for its respective carbon atoms.9 Crucially, this result
rules out the “interrupted” frame of brevisin as resulting from a
stereochemical incompatibility of an isolated epoxide group with
continuous ring frame formation. Before the absolute configuration
of brevisin (2) was known, it was possible (however unlikely) to
conjecture that the epoxides from which rings A-C are formed
might have an absolute configuration opposite that of the epoxides
from which rings D-F are formed. Thus, this stereochemical
incompatibility might cause the epoxide ring cascade to falter,
favoring the interrupted frame over the continuous frame. However,
our results show that a putative polyepoxide precursor to brevisin
is completely uniform with respect to the configuration of all
epoxides, just as such precursors are for the other K. breVis
polyethers. This indicates that the formation of an interrupted frame
results from either some other property of the polyepoxide precursor
(such as spacing between hydroxy group nucleophiles and upstream
epoxides) or some property of the enzyme(s) catalyzing the cascade.
Previously we suggested that the ring-forming cascade from a
polyepoxide precursor flows in the opposite direction of nascent
polyketide chain biosynthesis.12 Such a polyepoxide precursor might
be derived from an all E-polyene.9 Although the degree of
substitution of the olefinic groups to be modified varies from unit
to unit within a metabolite, without exception the same olefin face
A second (logically, not necessarily temporally)9,14 set of events
crucial for polyether assembly is the formation of the initial ring,
after which frame extension by endo-tet ring openings could in
principle proceed spontaneously in water11,17 or under the control
of a single epoxide hydrolase.14,18,19 Here again, there is a
remarkable consistency among the polyethers of K. breVis. We have
previously argued that the occurrence of ring frame initiation can
be reliably predicted from a putative polyepoxide precursor structure
based solely on the spacing between candidate nucleophile hydroxy
groups and their nearest upstream epoxides.13 Our results indicate
that the two initial epoxide groups in the putative precursor to
brevisin (at C-11/C-12 and C-24/C-25, respectively) would have
identical absolute configurations, a property known to be crucial
for bacterial enzymes in favoring the otherwise disfavored 6-endo-
tet cyclization.20
In conclusion, the absolute configuration of the anomalous
metabolites brevisamide (1) and brevisin (2) highlights commonali-
ties with the brevetoxins in the enzymatic reactions occurring en
route to polyether formation. Epoxidation occurs with universal
stereochemical regularity.9 Initial ring formation occurs invariably
and exclusively when spacing requirements are met between alcohol
and epoxide functionalities.13 As more information regarding the

Notes
Journal of Natural Products, 2010, Vol. 73, No. 6 1179
Acknowledgment. This work was supported by grant 5P41GM076300-
01 from NIH (J.L.C.W.) and by funding from the state of North Carolina
MARBIONC program for which the authors are grateful.
various polyether metabolites of K. breVis becomes available, more
inferences can be drawn on the principles underlying their assembly.
Supporting Information Available: 1H 1D NMR spectra and
TOCSY, HSQC, HMBC, DQF-COSY, and ROESY 2D NMR spectra.
This material is available free of charge via the Internet at http://
pubs.acs.org.
Experimental Section
General Experimental Procedures. NMR spectra were acquired
on a 500 MHz Bruker Avance spectrometer with a 1.7 mm TXI probe.
NMR data were analyzed using Topspin 2.0 (Bruker Biospin, Inc.).
Preparative HPLC was accomplished using a system with two Waters
515 HPLC pumps, a gradient controller, and a Waters 2487 dual-
wavelength UV detector. All solvents used were HPLC grade.
Formation of MTPA Esters of Brevisamide (1). A solution of 4.4
mg of DMAP and 3.4 µL of triethylamine in 50 µL of dry CH2Cl2 was
prepared. Two 500 µg portions of brevisamide (1) were dried in vacuo;
then each was dissolved in 20 µL of the aforementioned solution, to
which was immediately added 1.7 µL of S-(+)- or R-(-)-MTPA
chloride (Fluka, Inc.).21 The solutions were left at room temperature
for 1 h and quenched by addition of 30 µL of CH2Cl2 and 50 µL of
H2O. The solutions were separately agitated by vortex and the organic
layers extracted by syringe. Each reaction product was purified by
HPLC using a Gemini-NX 150 × 4.6 mm, 3 µ C18 column
(Phenomenex, Inc.) with a binary mobile phase system consisting of
0.1% formic acid (A) and MeCN (B). Each sample was injected onto
the column under isocratic elution at 20% B (0.8 mL/min) followed
by a linear gradient to 100% B over 80 min. The MTPA esters eluted
at 45 min and were detected by absorbance at 290 nm.
Formation of 31-Monokis- and 10,31-Bis-MTPA Esters of Brevi-
sin (2). For each reaction, 1.4 mg of brevisin (2) was dried in vacuo,
then equilibrated at 4 °C. Each sample was dissolved in 120 µL of a
solution prepared by dissolving 5.8 mg of DMAP in 290 µL of dry
CH2Cl2. To this solution was added 6 µL of a solution prepared by
mixing 1 µL of R-(-)- or S-(+)-MTPA chloride with 12 µL of dry
CH2Cl2. The reactions were maintained at 4 °C for 3 h and quenched
by addition of 100 µL of chilled H2O to each vessel. The reactions
were agitated by vortex and the organic layers removed by syringe.
Purification of each was achieved using a Gemini-NX 150 × 4.6 mm,
3 µ C18 column (Phenomenex, Inc.) with a binary mobile phase system
consisting of H2O (A) and MeCN (B). The reactions yielded a mixture
of esterified products consisting approximately of 50% 31-monokis-
MTPA ester and 25% 10,31-bis-MTPA ester. Smaller amounts (15%
and 10%, respectively) were obtained of a second bis- and a tris-MTPA
ester, but these minor products were not characterized. The product
mixture was fractionated using a stepped gradient at 0.8 mL/min of
67% B from 0-10 min, 85% B from 10-20 min, and 94% B from
20-30 min, yielding the four esters eluting at 12, 21, 22, and 25 min,
respectively.
References and Notes
(1) Plumley, F. G. Limnol. Oceanogr. 1997, 42, 1252–1264.
(2) Snyder, R. V.; Gibbs, P. D.; Palacios, A.; Abiy, L.; Dickey, R.; Lopez,
J. V.; Rein, K. S. Mar. Biotechnol. 2003, 5, 1–12.
(3) Snyder, R. V.; Guerrero, M. A.; Sinigalliano, C. D.; Winshell, J.; Perez,
R.; Lopez, J. V.; Rein, K. S. Phytochemistry 2005, 66, 1767–1780.
(4) Kubota, T.; Iinuma, Y.; Kobayashi, J. I. Biol. Pharm. Bull. 2006, 29,
1314–1318.
(5) Monroe, E. A.; Van Dolah, F. M. Protist 2008, 159, 471–482.
(6) Cane, D. E.; Celmer, W. D.; Westley, J. W. J. Am. Chem. Soc. 1983,
105, 3594–3600.
(7) Nakanishi, K. Toxicon 1985, 23, 473–479.
(8) Shimizu, Y. In Natural Toxins: Animal, Plant, and Microbial; Harris,
J. B., Ed.; Oxford University Press: New York, 1986; pp 115-125.
(9) Gallimore, A. R.; Spencer, J. B. Angew. Chem., Int. Ed. 2006, 45,
4406–4413.
(10) Harvey, B. M.; Mironenko, T.; Sun, Y.; Hong, H.; Deng, Z.; Leadlay,
P. F.; Weissman, K. J.; Haydock, S. F. Chem. Biol. 2007, 14, 703–
714.
(11) Vilotijevic, I.; Jamison, T. F. Science 2007, 317, 1189–1192.
(12) Satake, M.; Bourdelais, A. J.; Van Wagoner, R. M.; Baden, D. G.;
Wright, J. L. C. Org. Lett. 2008, 10, 3465–3468.
(13) Satake, M.; Campbell, A.; Van Wagoner, R. M.; Bourdelais, A. J.;
Jacocks, H.; Baden, D. G.; Wright, J. L. C. J. Org. Chem. 2009, 74,
989–994.
(14) Gallimore, A. R.; Stark, C. B. W.; Bhatt, A.; Harvey, B. M.;
Demydchuk, Y.; Bolanos-Garcia, V.; Fowler, D. J.; Staunton, J.;
Leadlay, P. F.; Spencer, J. B. Chem. Biol. 2006, 13, 453–460.
(15) Kuranaga, T.; Shirai, T.; Baden, D. G.; Wright, J. L. C.; Satake, M.;
Tachibana, K. Org. Lett. 2009, 11, 217–220.
(16) Seco, J. M.; Quinoa, E.; Riguera, R. Chem. ReV. 2004, 104, 17–117.
(17) Morten, C. J.; Jamison, T. F. J. Am. Chem. Soc. 2009, 131, 6678–
6679.
(18) Gallimore, A. R. Nat. Prod. Rep. 2009, 26, 266–280.
(19) Liu, T.; Cane, D. E.; Deng, Z. Methods Enzymol. 2009, 459, 187–
214.
(20) Matsuura, Y.; Shichijo, Y.; Minami, A.; Migita, A.; Oguri, H.;
Watanabe, M.; Tokiwano, T.; Watanabe, K.; Oikawa, H. Org. Lett.
2010, 12, 2226–2229.
(21) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092–4096.
NP100159J