Vanadium bromoperoxidase-catalyzed biosynthesis of halogenated marine natural products

Article abstract of DOI:10.1021/ja047925p

Marine red algae (Rhodophyta) are a rich source of bioactive halogenated natural products. The biogenesis of the cyclic halogenated terpene marine natural products, in particular, has attracted sustained interest in part because terpenes are the biogenic precursors of many bioactive metabolites. The first enzymatic asymmetric bromination and cyclization of a terpene, producing marine natural products isolated from red algae, is reported. Vanadium bromoperoxidase (V-BrPO) isolated from marine red algae (species of Laurencia, Plocamium, Corallina) catalyzes the bromination of the sesquiterpene (E)-(+)-nerolidol producing α-, β-, and γ-snyderol and (+)-3β-bromo-8-epicaparrapi oxide. α-Snyderol, β-snyderol, and (+)-3β-bromo-8-epicaparrapi oxide have been isolated from Laurencia obtusa, and each have also been isolated from other species of marine red algae. γ-Snyderol is a proposed intermediate in other bicyclo natural products. Single diastereomers of β-snyderol, γ-snyderol, and mixed diastereomers of (+)-3β-bromo-8-epicaparrapi oxide (de = 20-25%) are produced in the enzyme reaction, whereas two diastereomers of these compounds are formed in the synthesis with 2,4,4,6-tetrabromocyclohexa-2,5-dienone (TBCO). V-BrPO likely functions by catalyzing the two-electron oxidation of bromide ion by hydrogen peroxide producing a bromonium ion or equivalent in the active site that brominates one face of the terminal olefin of nerolidol. These results establish V-BrPO's role in the biosynthesis of brominated cyclic sesquiterpene structures from marine red algae for the first time.

Full text of DOI:10.1021/ja047925p

Published on Web 10/27/2004
Vanadium Bromoperoxidase-Catalyzed Biosynthesis of
Halogenated Marine Natural Products
Jayme N. Carter-Franklin and Alison Butler*
Contribution from the Department of Chemistry and Biochemistry,
UniVersity of California, Santa Barbara, California 93106-9510
Received April 10, 2004; E-mail: butler@chem.ucsb.edu
Abstract: Marine red algae (Rhodophyta) are a rich source of bioactive halogenated natural products.
The biogenesis of the cyclic halogenated terpene marine natural products, in particular, has attracted
sustained interest in part because terpenes are the biogenic precursors of many bioactive metabolites.
The first enzymatic asymmetric bromination and cyclization of a terpene, producing marine natural products
isolated from red algae, is reported. Vanadium bromoperoxidase (V-BrPO) isolated from marine red algae
(species of Laurencia, Plocamium, Corallina) catalyzes the bromination of the sesquiterpene (E)-(+)-nerolidol
producing R-, â-, and γ-snyderol and (+)-3â-bromo-8-epicaparrapi oxide. R-Snyderol, â-snyderol, and (+)-
3â-bromo-8-epicaparrapi oxide have been isolated from Laurencia obtusa, and each have also been isolated
from other species of marine red algae. γ-Snyderol is a proposed intermediate in other bicyclo natural
products. Single diastereomers of â-snyderol, γ-snyderol, and mixed diastereomers of (+)-3â-bromo-8-
epicaparrapi oxide (de ) 20-25%) are produced in the enzyme reaction, whereas two diastereomers of
these compounds are formed in the synthesis with 2,4,4,6-tetrabromocyclohexa-2,5-dienone (TBCO).
V-BrPO likely functions by catalyzing the two-electron oxidation of bromide ion by hydrogen peroxide
producing a bromonium ion or equivalent in the active site that brominates one face of the terminal olefin
of nerolidol. These results establish V-BrPO’s role in the biosynthesis of brominated cyclic sesquiterpene
structures from marine red algae for the first time.
Introduction
The sheer abundance of halogenated compounds isolated from
the marine environment distinguishes marine natural products
from terrestrial natural products. In particular, halogenated
compounds from marine red algae (Rhodophyceae) comprise
some of the most frequently reported metabolites.1-3 Brominated
compounds such as thyrsiferyl acetate (I, see Figure 1),4-6
halomon (II),7,8 ma’ilione (III),9,10 furanones (e.g., IV),11-13
cyclized sesquiterpenes (e.g., V, VI),14 and acetogenins (e.g.,
(1) Fenical, W. J. Phycol. 1975, 11, 245-259.
(2) Gribble, G. W. Acc. Chem. Res. 1998, 31, 141-152.
(3) Faulkner, D. J. Nat. Prod. Rep. 2002, 19, 1-48 and references therein.
(4) Suzuki, T.; Suzuki, M.; Furusaki, A.; Matsumoto, T.; Kato, A.; Imanaka,
Y.; Kurosawa, E. Tetrahedron Lett. 1985, 26, 1329-1332.
(5) Matsuzawa, S. I.; Suzuki, T.; Suzuki, M.; Matsuda, A.; Kawamura, T.;
Mizuno, Y.; Kikuchi, K. FEBS Lett. 1994, 356, 272-274.
(6) Schonthal, A. H. Front. Biosci. 1998, 3, 1262-1273.
(7) Fuller, R. W.; Cardellina, J. H., II; Kato, Y.; Brinen, L. S.; Clardy, J.;
Snader, K. M.; Boyd, M. R. J. Med. Chem. 1992, 35, 3007-3011.
(8) Fuller, R. W.; Cardellina , J. H., II; Jurek, J.; Scheurer, P. J.; Alvarado-
Lindner, B.; McGuire, M.; Gray, G. N.; Steiner, J. R.; Clardy, J.; Menez,
E.; Shoemaker, R. H.; Newman, D. J.; Snader, K. M.; Boyd, M. R. J. Med.
Chem. 1994, 37, 4407-4411.
Figure 1. Halogenated natural products from marine red algae. Thyrsiferyl
acetate (I),4-6 halomon (II),7,8,11 ma’ilione (III),9,10 brominated furanone
(IV),11-13 brominated cyclized sesquiterpenes (V, VI)14 and brominated
cyclized acetogenin (VII).14
(9) Juagdan, E. G.; Kalidindi, R.; Scheuer, P. J. Tetrahedron 1997, 53, 521-
528.
(10) Davyt, D.; Fernandez, R.; Suescun, L.; Mombru’, A. W.; Saldana, J.;
Dominguez, L.; Coll, J.; Fujii, M. T.; Manta, E. J. Nat. Prod. 2001, 64,
1552.
VII),14 all from marine red algae, exhibit wide-ranging biologi-
cal activities including protein phosphatase inhibition (I),
antitumor and cytotoxic function (II), anthelminthic properties
(11) Kazlauskas, R.; Murphy, P. T.; Quinn, R. J.; Wells, R. J. Tetrahedron Lett.
1977, 1, 37-40.
(12) De Nys, R.; Wright, A. D.; Konig, G. M.; Sticher, O. Tetrahedron 1993,
49, 11213-11220.
(13) Maximllien, R.; De Nys, R.; Holmstrom, C.; Gram, L.; Givskov, M.; Crass,
K.; Kjelleberg, S.; Steinberg, P. D. Aquat. Microb. Ecol. 1998, 15, 233-
246.
(14) Suzuki, M.; Daitoh, M.; Vairappan, C. S.; Abe, T.; Masuda, M. J. Nat.
Prod. 2001, 64, 597-602.
9
15060
J. AM. CHEM. SOC. 2004, 126, 15060-15066
10.1021/ja047925p CCC: $27.50 © 2004 American Chemical Society

Biosynthesis of Halogenated Marine Natural Products
A R T I C L E S
Scheme 1
(III), anti-fouling properties (IV), as well as broad spectrum
antimicrobial (V-VII) and antiviral (I) activities (Figure 1).
Within the vast array of halogenated compounds, it is the
biosynthesis of halogenated cyclic terpenes that has attracted
sustained interest, in part because terpenes are the biogenic
precursors of many bioactive natural products. Early investiga-
tions on the synthesis of halogenated terpenes were carried out
at a time when the existence of halogenating enzymes had only
been hypothesized. Van Tamelen and Hessler demonstrated that
N-bromosuccinamide reacts with methyl farnesate in an aqueous
tetrahydrofuran mixture to produce a bromo-bicyclic ester of
methyl farnesate. This reaction was the first demonstration of
a bromonium-ion-induced cyclization of a terpene forming a
bromocyclic product.15,16 Subsequently, reagents have been
developed to produce brominated cyclized terpenes including
bromine in the presence of Lewis acids such as AlBr3, SnBr4,
or AgBF4,17-19 2,4,4,6-tetrabromocyclohexa-2,5-dienone
(TBCO),20-22 the acid-catalyzed cyclization of terpene-contain-
ing terminal bromohydrins,23,24 as well as mercuric trifluoro-
acetate in the presence of molecular bromine.25-27 Through these
investigations a consensus developed that brominated cyclized
terpene natural products likely resulted from a bromonium-ion-
induced cyclization of an acyclic terpene precursor.15-17,19,28-31
Since the time of the initial investigations described above,
vanadium haloperoxidase32-34 and FeHeme haloperoxidase35-37
enzymes have been isolated from marine organisms. Vanadium
bromoperoxidase (V-BrPO), which appears to be the more
prevalent of the two classes of enzymes, is an abundant and
robust enzyme found in all classes of marine algae, including
species that produce chiral halogenated natural products. In the
active site of V-BrPO, the vanadate ion (VV) is coordinated to
the protein scaffold by a single histidine residue positioned at
the bottom of a deep active-site channel.38,39 Multiple amino
acid side chains hydrogen bond to the vanadate oxygen atoms.
The ligand histidine and the amino acids that participate in
hydrogen bonding are conserved in V-BrPO isolated from both
red (Rhodophyta) and brown (Phaeophyta) marine algae,
although little sequence homology exist in the rest of the active-
site channel.
V-BrPO catalyzes the oxidation of halides (I-, Br-, Cl-) by
the peroxo complex of V-BrPO (Scheme 1). The oxidized
halogen intermediate can halogenate an appropriate organic
substrate or oxidize a second equivalent of hydrogen peroxide,
producing dioxygen in the singlet excited state to complete the
catalytic cycle (Scheme 1).34,38,40 Bromination reactions cata-
lyzed by V-BrPO proceed via an electrophilic mechanism (i.e.,
Br+), as opposed to a radical (Br•) process.41
While the numbers of halogenated metabolites isolated from
marine organisms continue to increase, the biochemical path-
ways by which these products are synthesized within the algae
have not been fully elucidated.42-44 We recently reported that
V-BrPO isolated from marine red algae (e.g., Corallina offici-
nalis, Laurencia pacifica, and Plocamium cartilagineum) cata-
lyzes the bromination and cyclization of monoterpenes geraniol
and nerol to cyclic structures that are found in many bromocyclic
terpene marine metabolites,45 although these bromocyclic
monoterpene compounds are not known, marine natural prod-
ucts. Several bromosesquiterpene metabolites isolated from the
red alga L. obtusa provide an attractive target for biosynthetic
studies because all of these compounds could originate from a
single substrate. Herein we report the first enzymatic asymmetric
bromination and cyclization of a sesquiterpene, (E)-(+)-nerolidol
(1), by V-BrPO producing marine natural products R-snyderol
(2), â-snyderol (3), and (+)-3â-bromo-8-epicaparrapi oxide (5),
and we establish V-BrPO’s likely role in the biosynthesis of
brominated cyclic sesquiterpene structures from marine red
algae.
(15) vanTamelen, E.; Hessler, E. J. Chem. Commun. 1966, 411-413.
(16) vanTamelen, E. Acc. Chem. Res. 1968, 1, 111-120.
(17) Wolinksky, L. E.; Faulkner, D. J. J. Org. Chem. 1976, 41, 597-600.
(18) Hoye, T. R.; Kurth, M. J. J. Org. Chem. 1978, 43, 3693-3697.
(19) Gonzalez, A. G.; Darias, J.; Diaz, A.; Fourneron, J. D.; Martin, J. D.; Perez,
C. Tetrahedron Lett. 1976, 35, 3051-3054.
(20) Kato, T.; Ishii, K.; Ichinose, I.; Nakai, Y.; Kumagai, T. J. Chem. Soc.,
Chem. Commun. 1980, 1106-1108.
(21) Shieh, H.; Prestwich, G. D. Tetrahedron Lett. 1982, 23, 4643-4646.
(22) Gonzalez, I. C.; Forsyth, C. J. J. Am. Chem. Soc. 2000, 122, 9099-9108.
(23) Murai, A.; Abiko, A.; Kato, K.; Tadashi, M. Chem. Lett. 1981, 1125-
1128.
Methods and Materials
(24) Murai, A.; Abiko, A.; Masamune, T. Tetrahedron Lett. 1984, 25, 4955-
4958.
General Methods. Vanadium bromoperoxidase from the marine red
algae C. officinalis, P. cartilagineum, and L. pacifica were purified as
previously described for C. officinalis with minor modifications.46,47
C. officinalis and P. cartilagineum were collected off the coast of Santa
(25) Hoye, T. R.; Kurth, M. J. J. Org. Chem. 1979, 44, 3461-3467.
(26) Gopalan, A.; Prieto, R.; Mueller, B.; Peters, D. Tetrahedron Lett. 1992,
33, 1679-1682.
(27) Hoye, T. R.; Caruso, A. J.; Kurth, M. J. J. Org. Chem. 1981, 46, 6, 3550-
3552.
(28) Faulkner, D. J. Pure Appl. Chem. 1976, 48, 25-28.
(29) Gonzalez, A. G.; Martin, J. D.; Perez, C.; Ramirez, M. A. Tetrahedron
Lett. 1976, 2, 137-138.
(40) Everett, R. R.; Soedjak, H. S.; Butler, A. J. Biol. Chem. 1990, 265, 15671-
15679.
(30) Kato, T.; Kanno, S.; Kitahara, Y. Tetrahedron 1970, 26, 4287-4292.
(31) Kato, T.; Ichinose, I.; Kamoshida, A.; Kitahara, Y. J. Chem. Soc., Chem.
Commun. 1976, 518-519.
(41) Soedjak, H. S.; Walker, J. V.; Butler, A. Biochemistry 1995, 34, 12689-
12696.
(42) Fukuzawa, A.; Aye, M.; Nakamura, M.; Tamura, M.; Murai, A. Chem.
Lett. 1990, 1287-1290.
(32) Vilter, H. Bot. Mar. 1983, 26, 429-435.
(33) Wever, R.; de Boer, E. Biochem. Biophys. Acta 1985, 830, 181-186.
(34) Butler, A.; Walker, J. V. Chem. ReV. 1993, 93, 1937-1944.
(35) Manthey, J. A.; Hager, L. P. J. Biol. Chem. 1985, 260, 9654-9659.
(36) Manthey, J. A.; Hager, L. P. Biochemistry 1989, 28, 3052-3057.
(37) Roach, M. P.; Chen, Y. P.; Woodin, S. A.; Lincoln, D. E.; Lovell, C. R.;
Dawson, J. H. Biochemistry 1997, 36, 2197-2202.
(43) Fukuzawa, A.; Takasugi, Y.; Murai, A.; Nakamura, M.; Tamura, M.
Tetrahedron Lett. 1992, 33, 2017-2018.
(44) Fukuzawa, A.; Aye, M.; Takasugi, Y.; Nakamura, M.; Tamura, M.; Murai,
A. Chem. Lett. 1994.
(45) Carter-Franklin, J.; Parrish, J. D.; Tschirret-Guth, R. A.; Little, R. D.; Butler,
A. J. Am. Chem. Soc. 2003, 125, 3688-3689.
(38) Weyand, M.; Hecht, H. J.; Kiess, M.; Liaud, M. F.; Vilter, H.; Schomburg,
D. J. Mol. Biol. 1999, 293, 595-611.
(39) Isupov, M. N.; Dalby, A. R.; Brindley, A. A.; Izumi, Y.; Tanabe, T.;
Murshudov, G. N.; Littlechild, J. A. J. Mol. Biol. 2000, 299, 1035-1049.
(46) Carter, J. N.; Beatty, K. E.; Simpson, M. T.; Butler, A. J. Inorg. Biochem.
2002, 91, 59-69.
(47) Brindley, A. A.; Dalby, A. R.; Isupov, M. N.; Littlechild, J. A. Acta
Crystallogr. 1998, D54, 454-457.
9
J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004 15061

A R T I C L E S
Carter-Franklin and Butler
Barbara, CA. L. pacifica was collected from an intertidal region off
the coast of La Jolla, CA. Bromoperoxidase activity was determined
by monitoring the bromination of 50 µM monochlorodimedone (MCD)
spectrophotometrically at 290 nm under conditions of 0.1 M KBr, and
1 mM H2O2 in 0.1 M sodium phosphate buffer pH 6.00. The extinction
coefficient difference at 290 nm between brominated MCD and MCD
under the same reaction conditions as described above for V-BrPO-
catalyzed reactions (i.e., 0.04 M KBr in 0.15 M phosphate buffer (pH
5.7) containing 40% ethanol), although without addition of V-BrPO
or H2O2. Aqueous bromine was added in a controlled manner via
syringe pump at a rate approximating the bromination rate of MCD by
V-BrPO (all reactions were performed in the dark with stirring).
Bromination of 1 with TBCO was performed in nitromethane with
stirring in the dark.20-22
Competitive Substrate Kinetics. Competitive kinetics for the
bromination of phenol red (50 µM) catalyzed by V-BrPO (5 nM) or
carried out by the stoichiometric addition of aqueous bromine as a
function of the concentration of 1 were performed at 25 °C in the
presence of 40 mM KBr in 0.1 M phosphate buffer (pH 5.7) containing
40% ethanol. V-BrPO-catalyzed experiments were initiated by the
addition of 0.5 mM H2O2. Nonenzymatic competition experiments were
initiated by the addition of aqueous bromine (5 mM stock solution) in
10-µL aliquots at 30 s intervals. UV absorbance was measured 20 s
after each addition of aqueous bromine. Production of bromophenol
blue was monitored at 596 nm.
is 19 900 cm-1 M-1 48
.
1H NMR and 13C NMR spectra were recorded on a Varian Inova
instrument (500 and 125 MHz, respectively) using deuteriochloroform
as the solvent (internal reference δ ) 7.27 ppm, CHCl3). Consumption
and formation of reaction substrates and products were monitored by
gas chromatography (GC). Chiral GC analysis was performed using a
Cyclosil B capillary column (J & W Scientific).
Aqueous bromine was prepared by dilution of bromine vapors in
0.1 M NaOH and standardized by tri-iodide formation (I3-; λmax 353
nm, ∆ꢀ ) 26 600 cm-1 M-1) by reaction with 0.05 M KI in 0.1 M
citrate phosphate buffer pH 4.5. The predominant species in solution
at pH 6-7 is HOBr; however, multiple oxidized bromine species are
present in solution which are best represented as OBr- ) HOBr ) Br2
- 49
) Br3
, hereafter referred to as “aqueous bromine”.
Results and Discussion
Normal-phase HPLC was used to purify products 2, 3, and 4 which
often eluted as overlapping peaks in reversed-phase mode. Using a silica
HPLC column (semipreparative 250 mm × 10 mm, YMC-Sil, YMC
Inc.), products were eluted isocratically with 0.3% 2-propanol/hexanes,
and detected by UV absorbance (214 nm). Reversed-phase HPLC was
used to purify product 5. Samples were applied to a C18 column
(semipreparative 250 mm × 10 mm or analytical 250 mm × 4.5 mm,
ODS-AQ, YMC Inc.), and products were eluted using gradient mixtures
of acetonitrile and water. The UV absorbance of the eluent was
monitored at 214 nm. Flash column chromatography was performed
using 230-400 mesh silica gel (EM Science) and mixtures of ether
and pentane as eluent. (E)-(+)-Nerolidol (1) was purchased from
Indofine Chemical Company and used as received. TBCO was used
as obtained from Acros.
V-BrPO-Catalyzed Reaction with (E)-(+)-Nerolidol (1). (E)-(+)-
Nerolidol (1) (0.5 mM) was dissolved in ethanol and added to 0.15 M
phosphate buffer (pH 5.7) containing 40% v/v ethanol and 40 mM KBr.
Enzymatic reactions containing 23 nM V-BrPO, were initiated by
addition of 1 or 2 mol equiv of H2O2 via syringe pump with respect to
1. After 2.5 h, the reactions were extracted with 3 volumes of hexanes.
Alcohol and ether products were separated by elution on a silica Sep-
pak (Waters Corp.) with 10% ether/pentane. Bromo alcohol and
bromoether products were resuspended in 100% ethanol and injected
directly on a C18 analytical column for product profile and retention
time analysis by reversed-phase HPLC.
V-BrPO-Catalyzed Reaction with (E)-(+)-Nerolidol (1)
versus Reaction with Aqueous Bromine or TBCO. To explore
whether V-BrPO could be involved in the biosynthesis of
halogenated marine natural products, the reactivity of V-BrPO
toward (E)-(+)-nerolidol (1) was investigated. When (E)-(+)-
nerolidol (7 mM) (1) reacts with V-BrPO (55 nM) in phosphate
buffer (pH 6.5) containing 40% organic cosolvent (e.g., EtOH,
iPrOH), bromide (40 mM), and 1 or 2 equiv of hydrogen
peroxide, a mixture of bromoether, bromo alcohol, bromohydrin,
and epoxide species is produced.
V-BrPO-Catalyzed Biosynthesis of r-, â-, and γ-Snyderols.
The V-BrPO-catalyzed reaction with 1 produces the known
bromo alcohols R-, â-, and γ-snyderols (2, 3, and 4), where 2
and 3 are natural products isolated from the marine red alga L.
obtusa.50 Snyderol products 2-4 are not produced in the absence
of V-BrPO or of one or more reaction substrates (e.g., Br- or
H2O2), indicating that oxidation of bromide by a peroxo-V-
BrPO complex is necessary for the bromination and cyclization
of 1 to produce 2-4. In addition, the presence of organic
cosolvent (g40%) is necessary for the formation of snyderols,
where the use of 2-propanol is preferred to minimize competing
nucleophilic reactions by the cosolvent.
To obtain products 2, 3, 4, and 5 for spectroscopic characterization,
reactions were run on a scale of 7 mM 1 with 55 nM V-BrPO. Upon
reaction completion (ca. 2.5 h), products were extracted with 3 volumes
of hexanes, washed with brine, dried with MgSO4, and reduced to a
small volume in vacuo. Crude extracts were first fractioned by flash
chromatography followed by HPLC purification as described above.
Spectroscopic data for 2, 3, 4, 5, and 5′ are identical to reported literature
values.17,50,51
The relative stereochemistries of 2, 3, 4, and 5 were determined
from coupling constants and by one-dimensional nOe analysis. Dia-
stereomeric excesses were established by chiral GC analysis using an
isotherm of 155 °C for 2, 3, 4, and 165 °C for 5. Chiral GC reten-
tion times for 3 and 4 were 75.7 and 91.0 min, respectively. Reten-
tion times for diastereomers 5 and 5′ were 27.6 and 26.8 min,
respectively.
The formation of 2-4 in the V-BrPO-catalyzed reaction
results from selective bromination of the C10-C11 olefinic bond
of 1, producing a bromonium-nerolidol adduct (Scheme 2).
The proposed bromocarbenium ion intermediate is subsequently
attacked by the electron-rich internal olefin, followed by one
of three different elimination reactions leading to 2, 3, and 4.
Brominated snyderols 2-4 comprise 10% of the total isolated
products of the reaction.52 The proposed mechanism of forma-
tion of 2, 3, and 4 is analogous to the V-BrPO-catalyzed
reactions with geraniol and geranyl acetate, producing bro-
mocyclized monoterpenes.17,45
In contrast to the V-BrPO reaction, only trace quantities of
2, 3, and 4 are formed in the reaction of 1 with aqueous bromine
(1-2 equiv) (Figure 2). Similar results were observed previously
with the monoterpenes, nerol and geraniol, where brominated
cyclized products were only produced in the enzyme-catalyzed
reactions and absent from reactions with aqueous bromine.45
Nonenzymatic Reactions with (E)-(+)-Nerolidol (1). Reaction of
1 (0.5 mM) with 1 or 2 mol equiv of aqueous bromine was carried out
(48) Hager, L.; Morris, D.; Brown, F.; Eberwein, H. J. Biol. Chem. 1966, 241,
1769-1777.
(49) Ziderman, I. Isr. J. Chem. 1972, 11, 7-20.
(50) Howard, B. M.; Fenical, W. Tetrahedron Lett. 1976, 1, 41-44.
(51) Faulkner, D. J. Phytochemistry 1976, 15, 1993-1994.
(52) The remaining 90% of isolated product included brominated cyclic ethers
(30%) and bromohydrin and epoxide species (60%).
9
15062 J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004

Biosynthesis of Halogenated Marine Natural Products
A R T I C L E S
Figure 2. Reversed-phase HPLC separation of bromo alcohol containing
fractions formed in the reactions of V-BrPO and aqueous bromine with 1.
(a) Mixture of authentic sample of R-, â-, γ-snyderol (2, 3, and 4) (i.e.,
synthesized using TBCO in nitromethane20,31). (b) Bromo alcohol containing
fraction of the V-BrPO-catalyzed reaction. (c) Bromo alcohol containing
fraction of the aqueous bromine reaction. Peaks are denoted with numbers
corresponding to their structures. The reaction mixtures contained 0.04 M
KBr and 0.5 mM 1 in 0.1 M sodium phosphate, pH 5.7 with 40% v/v
ethanol. Enzymatic reactions were initiated by the addition of H2O2 (1 mM)
and allowed to react at 24 °C for 60 min. Aqueous bromine reactions were
initiated by the addition of NaOBr to a final concentration of 1mM and
reacted at 24 °C for 60 min.
Figure 3. Chiral GC-MS chromatogram for 3 isolated from the (a)
V-BrPO-catalyzed reaction and the (b) TBCO reaction with 1. The retention
time for the single diastereomer of 3 isolated from V-BrPO reactions was
75.7 min. Diastereomers isolated from TBCO reactions eluted with retention
times of 74.1 and 75.7 min, respectively. Gas chromatography was
performed using 90 min, 155 °C isotherm runs.
Scheme 2. Proposed Mechanism for V-BrPO-Catalyzed
Biosynthesis of the Marine Natural Products R-, â-, and γ-Snyderol
(2, 3, and 4), and (+)-3â-Bromo-8-epicaparrapi Oxide (5 and 5′)
Figure 4. Chiral GC-MS chromatogram for 4 isolated from the (a)
V-BrPO-catalyzed reaction and the (b) TBCO reaction with 1. The retention
time for the single diastereomer of 4 isolated from V-BrPO reactions was
91.0 min. Diastereomers isolated from TBCO reactions eluted with retention
times of 89.6 and 91.0 min, respectively. Gas chromatography was
performed using 90 min, 155 °C isotherm runs.
arrangement of the bromonium-terpene adduct appears to
control the conformation of the transition state in the cyclization
reaction.21,23
On the other hand, reaction of 1 with TBCO carried out in non-
nucleophilic solvents (e.g., methylene chloride or nitromethane)
produces 2-4 as previously reported.20,31
Chiral GC-MS analysis of snyderols 2-4 demonstrates that
V-BrPO catalyzes the asymmetric bromination of the C10-
C11 double bond of 1 (Figures 3 and 4). A single peak is
observed in the chiral GC chromatogram of â-snyderol (3) and
γ-snyderol (4) formed in the enzyme reaction, with a retention
time equal to that of one of the two diastereomers formed in
the TBCO reaction (Figures 3 and 4). Snyderols 3 and 4 obtained
from TBCO reactions, as expected, showed the formation of
equal amounts of the two diastereomers.20,31 The diastereomers
of R-snyderol (2) could not be separated by chiral GC using
the conditions employed for 3 and 4 nor under any of the other
conditions investigated. The fact that single diastereomers of 3
and 4 are obtained in the V-BrPO reactions implies that
The relative stereochemistry of the proton adjacent to the
bromine in 2-4 is indicated by a doublet of doublets signal at
4.17 ppm for R-snyderol (J ) 9, 7 Hz), 4.10 ppm for â-snyderol
(J ) 11, 5 Hz), and 4.22 ppm for γ-snyderol (J ) 10, 4 Hz) in
1
the H NMR spectrum, which is indicative of an axial proton
and thus an equatorial bromine.17 The stereochemistry of 2-4
is consistent with our previous results of a V-BrPO-catalyzed
reaction with geraniol to produce bromocyclogeraniol, in which
the bromine is also found in the equatorial position.45 In addition,
the stereochemistry of 2-4 is in agreement with the stereo-
chemistry determined for the isolated marine natural products.50
The equatorial position of bromine in cyclized terpenes has been
observed in many cationic cyclization reactions, where the
9
J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004 15063

A R T I C L E S
Scheme 3
Carter-Franklin and Butler
bromination of the terminal olefin of 1 occurs within the active-
site cavity, and not outside the enzyme active site via a freely
diffusible brominating species (e.g., HOBr, Br2, Br3-, etc). Thus,
we propose that 1 orients in a favorable conformation within
the active site suitable for asymmetric bromination at the C10-
C11 double bond, possibly assisted by some protein interaction
with the allylic alcohol moiety of 1. The ability of V-BrPO to
specifically bind and orient substrates has been reported and
has resulted in the regioselective brominative oxidation of
substituted indoles,53 as well as the non-halide-dependent
enantiospecific oxidation of sulfides.54-56
Figure 5. Reversed-phase HPLC separation of bromoether-containing
fractions from the reactions of V-BrPO and aqueous bromine with 1. (a)
Authentic sample of (+)-3â-bromo-8-epicaparrapi oxide (5) and 5′ (i.e.,
synthesized using TBCO20,31). (b) V-BrPO-catalyzed reaction. (c) Aqueous
bromine reaction. Peaks are denoted with numbers corresponding to their
structures. The reaction mixture contained 0.04 M KBr and 0.5 mM 1 in
0.1 M sodium phosphate (pH 5.7) with 40% v/v ethanol. Enzymatic reactions
were initiated by the addition of H2O2 (1 mM) and allowed to react at 24
°C for 60 min. Aqueous bromine reactions were initiated by addition of
NaOBr (1mM) and reacted at 24 °C for 60 min.
The diastereoselective formation of γ-snyderol (4) is striking
when one considers that this is the first report of an enzymatic
bromonium-ion-induced cyclization forming a tetra-substituted
olefin. V-BrPO-catalyzed formation of 4 supports previous
proposals of an enzyme-catalyzed cyclization of farnesyl pyro-
phosphate to 10-bromo-R-chamigrene via a tetra-substituted
monobromocyclofarnesyl pyrophosphate intermediate1,28,57
(Scheme 3). Cyclization of the brominated monocyclofarnesyl
pyrophosphate is assumed to proceed through an allylic cation.
Previously, Wolinksky and Faulkner demonstrated in a biomi-
metic synthesis of 10-bromo-R-chamigrene the conversion of
racemic 4 to the chamigrene skeleton.17 10-Bromo-R-chamigrene
was not detected in our biosynthetic studies with V-BrPO;
however, reactions with (E)-farnesyl pyrophosphate are under
investigation.
V-BrPO-Catalyzed Biosynthesis of (+)-3â-Bromo-8-epi-
caparrapi Oxide. The marine natural product (+)-3â-bromo-
8-epicaparrapi oxide (5), and its corresponding diastereomer (5′),
was obtained from the bromoether-containing fractions of the
V-BrPO-catalyzed reaction with 1. As was observed with the
snyderol products, 5 and 5′ are produced only when all substrate
components of the enzyme reaction are present (i.e., V-BrPO,
KBr, H2O2, 40% cosolvent). The presence of the C-8 epimer
of 5 and 5′ is not observed, indicating that epimerization of the
allylic alcohol during the reaction does not occur. The optical
rotation measurements of 5, (i.e., [R]D + 31.8; c ) 1.52 in
EtOH) isolated from the V-BrPO-catalyzed reaction are similar
to those of the natural product isolated from L. obtusa, (i.e.,
[R]D + 30.4; c ) 1.46 in EtOH).51
The V-BrPO-catalyzed formation of 5 and 5′ is proposed to
occur via bromination of the terminal olefin of 1 followed by
stereospecific cyclization to ring closure (Scheme 2). Nucleo-
philic trapping of the proposed bromocarbenium ion by the
allylic alcohol leading to 5 and 5′ is supported by cationic
cyclization reactions of terpenes previously reported and used
to study the concertedness of the overall ring-formation
process.26,27,58,59
Reaction of 1 with aqueous bromine (1 or 2 equiv) produces
only trace quantities of 5 and 5′ (Figure 5). TBCO reactions
with 1 produces 5 and 5′ as previously described.20 In addition,
five- and six-membered cyclic bromoether species (i.e., 25%)
were also present within the bromoether-containing fractions.
These brominated cyclic ethers have been previously identi-
fied;20 the specificity of bromination of these products is under
investigation. (+)-3â-Bromo-8-epicaparrapi oxide (5) was iso-
lated in ∼5% yield, and the remaining cyclic bromoether
compounds were produced in 25% isolated yield.
The stereochemistry of 5 at the bromine bearing carbon was
determined by the presence of Hax signal as a doublet of doublets
at δ 3.99 (J ) 12.5 Hz). The stereochemistry of the bridgehead
methyl was determined by the presence of a nOe signal to the
vinylic proton at 6.0 ppm, thus indicating that the cyclization
reaction produces a trans-fused bicyclo[4.4.0] system. The
stereochemistry of 5 is consistent with that previously reported
in cyclization reactions where the four chiral centers of 5 are
introduced stereospecifically following bromination of 1, in
(53) Martinez, J. S.; Carroll, G. L.; Tschirret-Guth, R. A.; Altenhoff, G.; Little,
R. D.; Butler, A. J. Am. Chem. Soc. 2001, 123, 3289-3294.
(54) Andersson, M.; Willetts, A.; Allenmark, S. J. Org. Chem. 1997, 62, 2,
8455-8458.
(55) ten Brink, H. B.; Tuynman, A.; Dekker, H.; Hemrika, W.; Izumi, Y.; Oshiro,
T.; Schoemaker, H. E.; Wever, R. Inorg. Chem. 1998, 37, 6780-6784.
(56) Andersson, M.; Allenmark, S. Tetrahedron 1998, 54, 15293-15304.
(57) Martin, J. D.; Palazon, J. M.; Perez, C.; Ravelo, J. L. Pure Appl. Chem.
1986, 58, 395-406.
(58) Garst, M.; Cheung, Y.; Johnson, W. J. Am. Chem. Soc. 1979, 101, 4404-
4406.
(59) Snowden, R.; Eichenberger, J.; Linder, S.; Sonnay, P.; Vial, C.; Schulte-
Elte, K. J. Org. Chem. 1992, 57, 955-960.
9
15064 J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004

Biosynthesis of Halogenated Marine Natural Products
A R T I C L E S
Figure 6. Chiral GC-MS chromatogram for 5 and 5′ isolated from the
(a) V-BrPO-catalyzed reaction and the (b) TBCO reactions with 1. The
retention time of 5′ and 5 isolated from V-BrPO reactions was 26.8 and
27.6 min, respectively. Diastereomers isolated from TBCO reactions eluted
with retention times of 26.8 and 27.6 min, respectively. Gas chromatography
was performed using 90 min, 165 °C isotherm runs.
Scheme 4
which the set configuration of the carbon alcohol (C-3) bond
uniquely determines the remaining three stereocenters.20,21,25 The
biosynthesis of the trans-fused ring system with exclusion of
the cis-fused ring system can be rationalized by assuming the
planar bromocarbenium ion is quenched by internal nucleophilic
trapping on the empty p-orbital by the allylic alcohol in an
equatorial position (Scheme 4). Formation of the cis-fused ring
system is hindered by unfavorable isomerization of the bromine
from the equatorial to the axial position leading to 1,3-diaxial
interactions.18 V-BrPO-catalyzed bromination and cyclization
of 1 to 5 is considered consistent with the Stork-Eschenmoser
hypothesis for a synchronous cyclization mechanism initiated
by bromination at the C10-C11 olefin of 1.60,61
V-BrPO catalyzed the asymmetric bromination of the terminal
olefin of 1 followed by successive cyclization to produce (+)-
3â-bromo-8-eqicaparrapi oxide (5) as a mixture of diastereomers
(de ) 20-25%), whereas equal amounts of the two diastere-
omers are observed for the corresponding TBCO reaction20
(Figure 6). In an analogous reaction, enzymes such as terpene
cyclases control the stereospecific cyclization of isoprenoids by
the asymmetric protonation of the terminal olefin, followed by
successive cyclizations and stabilization of intermediates and
conformations within enzyme cavities lined with hydrophobic
residues.62-65 These results suggest a “terpene cyclase”-like
cyclization along the hydrophobic substrate channel in V-BrPO.
Cyclization is initiated by the asymmetric bromination of the
Figure 7. Time course for the bromination of phenol red as a function of
the (E)-(+)-nerolidol (1) concentration (a) catalyzed by V-BrPO or (b) by
reaction with aqueous bromine. The reactions were performed at 24 °C in
the presence of 50 µM phenol red and 40 mM KBr in 0.1 M sodium
phosphate buffer (pH 5.7) with 40% v/v ethanol. V-BrPO reactions were
initiated by addition of 0.5 mM H2O2 and 5 nM V-BrPO. Aqueous bromine
reactions were initiated by addition of NaOBr (stock solution 5 mM in
0.02 M NaOH) in 10 µL aliquots at 30 s intervals. Production of
bromophenol blue was monitored at 596 nm. Concentration of 1: b, 0
µM; O, 50 µM; 1100 µM.
terminal olefin to generate a bromonium-ion adduct of nerolidol
on the terminal olefin which controls the stereoselectivity of
the cyclization reaction.
Competition Kinetics for the Bromination of (E)-(+)-
Nerolidol (1) versus Phenol Red. Vanadium bromoperoxidase
preferentially catalyzes the bromination of 1 when presented
with a mixture of 1 and phenol red as organic substrates (Figure
7). The preferential bromination of 1 is concentration dependent
as indicated by the increase in the lag phase for the bromination
of phenol red as the concentration of 1 is increased (Figure 7a).
Following the consumption of 1, the rate of bromination of
phenol red occurs at the same rate as in the absence of added
1. Furthermore, a concentration-dependent lag phase is absent
in reactions with aqueous bromine under the same conditions
as V-BrPO, indicating that 1 and phenol red are brominated
simultaneously (Figure 7b). The difference in reactivity of 1 in
the V-BrPO/H2O2/Br- system versus reaction with aqueous
(60) Stork, G.; Burgstahler, A. W. J. Am. Chem. Soc. 1955, 77, 5068-5077.
(61) Eschenmoser, A.; Ruzicka, L.; Jeger, O.; Arigoni, D. HelV. Chim. Acta
1955, 38, 1890-1904.
(62) Croteau, R.; Cane, D. E. Methods Enzymol. 1985, 110, 383-405.
(63) Cane, D. E. Chem. ReV. 1990, 90, 1089-1103.
(64) Dougherty, D. A. Science 1996, 271, 163-168.
(65) Davis, E. M.; Croteau, R. Top. Curr. Chem. 2000, 209, 53-95.
9
J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004 15065

A R T I C L E S
Carter-Franklin and Butler
bromine, is consistent with an enzyme-trapped or enzyme-bound
brominating species, and not a freely diffusible brominating
species.53,66
bromination reaction outside the enzyme active-site.67-69 Sub-
sequent kinetic studies showing sequential bromination or
brominative oxidation of different substrates,66 as well as
different selectivity,53,70 suggested that some substrates bind to
the V-BrPO active site and block the release of the oxidized
halogen intermediate. The discovery of direct asymmetric
oxidation of bicyclic sulfides catalyzed by V-BrPO, producing
the corresponding sulfoxides with high enantioselectivity,54,56
bolstered the hypothesis that V-BrPO can participate in asym-
metric halogenation reactions. Finally, the bromination and
cyclization of model terpene precursors45 demonstrated that
V-BrPO can catalyze the bromination and cyclization of
terpenes, producing cyclic structures that make up parts of
halogenated marine natural products. Now, as we demonstrated
above, V-BrPO has been shown to catalyze the asymmetric
bromination and cyclization of (E)-(+)-nerolidol (1), forming
known marine natural products. While compounds 2, 3, and 5
are all natural products that have been isolated from Laurencia
obtusa,50,51,71 these compounds have each also been isolated
singly from other species of Laurencia (i.e., L. intricata, L.
synderiae), and the product distribution of these compounds
within different algae is not yet understood. Additional experi-
ments are under way to further understand the interactions
between (E)-(+)-nerolidol (1) and V-BrPO that promote the
cyclization process, both for sesquiterpenes as reported here and
also for longer-chain terpenes. Other cyclic brominated terpenes
are also metabolites produced by marine red algae that likely
contain haloperoxidases with different substrate reactivity or
substrate selectivity or that work in concert with other enzymes
such as terpene cyclases, etc. Thus, with the discovery of the
asymmetric halogenation catalyzed by V-BrPO, many interesting
questions can now be investigated.
Conclusions
Vanadium bromoperoxidase catalyzes the asymmetric bro-
mination and cyclization of (E)-(+)-nerolidol (1) to produce
single diastereomers of the marine natural products â- and
γ-snyderol (3 and 4) and a mixture of diastereomers of (+)-
3â-bromo-8-epicaparrapi oxide (5). In contrast, reaction of 1
with aqueous bromine produced only minimal quantities of these
brominated cyclized products, and reaction of 1 with TBCO in
nitromethane produced an equal mixture of each diastereomeric
product of 3, 4, and 5. Diastereomers of 2 could not be resolved
by chiral GC. The observed diastereoselectivity is the first report
of V-BrPO-catalyzed enantiospecific bromination and cycliza-
tion of sesquiterpenes, forming chiral brominated marine natural
products, and establishes a role for V-BrPO in the biogenesis
of halogenated metabolites in marine algae. The high specificity
of these V-BrPO-catalyzed reactions suggests that 1 docks
within the active-site channel of V-BrPO in a specific orientation
and is not randomly binding within the active-site cavity. In
the case of random substrate binding, one would expect
symmetric bromination of the C10-C11 olefinic bond of 1,
leading to equal distribution of both diastereomers of the
different natural products. V-BrPO functions by coordination
of H2O2 followed by oxidation of Br- to produce the equivalent
of a bromonium ion (Br+) that attacks one face of the terminal
olefin of 1. The nature of the brominating species as enzyme
bound (e.g., V-OBr) or enzyme trapped (e.g., Br+, HOBr,
OBr-, etc.) is not known and cannot be addressed by these
experiments, although the competitive kinetic data shows that
the brominating species is not consistent with release from the
active site.
Acknowledgment. We are particularly thankful to Prof. R.
Daniel Little (UCSB) for thoughtful discussions on the pro-
ject.We are grateful for support from NSF CHE 0213523 (A.B.),
California Sea Grant NA06RG0142, Project R/MP-94 (A.B.),
and a Sea Grant Traineeship for J.C.F. funded by a grant from
the National Sea Grant College Program, National Oceanic and
Atmospheric Administration, and the U.S. Department of
Commerce.
The X-ray structure of V-BrPO from C. officinalis shows that
the vanadium site resides at the bottom of a 20-Å deep substrate
channel.39 Hydrophobic patches of residues dominate the
substrate channel leading to the vanadium site. With the
exception of the amino acids involved in hydrogen bonding to
the vanadate ion, only three other hydrophilic residues (Glu124,
Arg395, Asp292) are positioned within 7.5 Å of the vanadate
oxygen atoms. Thus, the hydrophobic surface of the substrate
cavity, as well as certain charged residues, likely provides the
environment necessary for docking and bromination of 1.
In early studies of V-BrPO, the apparent lack of selectivity
led to the suggestion that V-BrPO produced a diffusible oxidized
halogen intermediate such as the equilibrium mixture of HOBr
JA047925P
(67) Itoh, N.; Izumi, Y.; Yamada, H. J. Biol. Chem. 1987, 262, 11982-11987.
(68) de Boer, E.; Wever, R. J. Biol. Chem. 1988, 263, 12326-12332.
(69) Itoh, N.; Hasan, A.; Izumi, Y.; Yamada, H. Eur. J. Biochem. 1988, 172,
477-484.
(70) Butler, A.; Tschirret-Guth, R. A. In Mechanisms of Biohalogenation and
Dehalogenation; Janssen, D. B., Soda, K., Wever, R., Eds.; Royal
Netherlands Academy of Arts and Sciences: Amsterdam, The Netherlands,
1996; pp 55-68.
-
) OBr- ) Br2 ) Br3 that would carry out a molecular
(71) Topcu, G.; Zeynep, A.; Imre, S.; Goren, A. C.; Pezzuto, J. M.; Clement, J.
A.; Kingston, D. G. J. Nat. Prod. 2003, 66, 1505-1508.
(66) Tschirret-Guth, R. A.; Butler, A. J. Am. Chem. Soc. 1994, 116, 411-412.
9
15066 J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004