Bromo-spiroisoxazoline Alkaloids, including an Isoserine Peptide, from the Caribbean Marine Sponge Aplysina lacunosa

Article abstract of DOI:10.1021/acs.jnatprod.9b01286

Three new bromotyrosine spiroisoxazoline alkaloids, lacunosins A and B (1 and 2) and desaminopurealin (3), were isolated from a MeOH extract of the marine sponge Aplysina lacunosa that showed modest α-chymotrypsin inhibitory activity. The structures of 1-3 share the spirocyclohexadienyl-isoxazoline ring system found in purealidin-R and several other Verongid sponge secondary metabolites. Compounds 1 and 2 are coupled to a glycine and an isoserine methyl ester, respectively. Alkaloid 3 is linked, contiguously, to an O-1-aminopropyl 3,5-dibromotyrosyl ether and, finally, to histamine through an amide bond. The planar structures of all three compounds were obtained from analysis of MS and 1D and 2D NMR data. The absolute configuration of the SIO unit of 1-3 was assigned by electronic circular dichroism (ECD). The isoserine amino acid residue in 2 was found to be a 1:1 mixture of epimers using a new Marfey's type reagent, derived from Trp-NH₂. Allylic O-naphthoylation of the SIO subunit enhances the ECD spectrum of SIOs and improves discrimination of enantiomorphs. A unifying hypothesis is proposed that links the biosynthesis of several of the new compounds with previously reported analogues.

Full text of DOI:10.1021/acs.jnatprod.9b01286

pubs.acs.org/jnp
Article
Bromo-spiroisoxazoline Alkaloids, Including an Isoserine Peptide,
from the Caribbean Marine Sponge Aplysina lacunosa
Mariam N. Salib, Matthew T. Jamison, and Tadeusz F. Molinski*
Cite This: https://dx.doi.org/10.1021/acs.jnatprod.9b01286
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Three new bromotyrosine spiroisoxazoline alkaloids,
lacunosins A and B (1 and 2) and desaminopurealin (3), were isolated
from a MeOH extract of the marine sponge Aplysina lacunosa that showed
modest α-chymotrypsin inhibitory activity. The structures of 1−3 share
the spirocyclohexadienyl-isoxazoline ring system found in purealidin-R and
several other Verongid sponge secondary metabolites. Compounds 1 and
2 are coupled to a glycine and an isoserine methyl ester, respectively.
Alkaloid 3 is linked, contiguously, to an O-1-aminopropyl 3,5-
dibromotyrosyl ether and, finally, to histamine through an amide bond.
The planar structures of all three compounds were obtained from analysis
of MS and 1D and 2D NMR data. The absolute configuration of the SIO
unit of 1−3 was assigned by electronic circular dichroism (ECD). The
isoserine amino acid residue in 2 was found to be a 1:1 mixture of epimers
using a new Marfey’s type reagent, derived from Trp-NH2. Allylic O-naphthoylation of the SIO subunit enhances the ECD spectrum
of SIOs and improves discrimination of enantiomorphs. A unifying hypothesis is proposed that links the biosynthesis of several of
the new compounds with previously reported analogues.
mong the earliest marine natural products reported are
Aalkaloids and peptides containing modified bromotyr-
osine residues from marine sponges (Porifera), mainly in the
order Verongida. Approximately 300 bromotyrosine alkaloids
of marine origin have been reported in the literature to date.1,2
The formidable oxidative brominating capacity of Verongid
sponges is manifested in not only secondary metabolites but
also high levels of bromotyrosine amino acid residues in the
structural proteins (chitin) of Ianthella basta3 and Aplysina
cavernicola.4 Bromotyrosine secondary metabolites from
marine sponges show a variety of biological properties,
including antibacterial,5 anti-inflammatory,6 antineoplastic,7
and antifungal activities that clearly demonstrate bio- and
chemodiverse repertoires. A remarkable report by Berlinck and
co-workers shows that several bromotyrosine-derived natural
products, conventionally associated with Verongid sponges, are
produced by fermentation of a Verongid sponge-derived
bacterium, Pseudovibrio denitrificans Ab134.8
In screening for protease inhibitors, we found the MeOH
extract of the marine sponge Aplysina lacunosa from the
Bahamas exhibited significant inhibition of α-chymotrypsin
activity. Here we report the structures of three new brominated
spiroisoxazoline (SIO) alkaloidal peptides, lacunosins A (1)
and B(2) and desaminopurealin (3), from an assay-guided
survey of extracts with α-chymotrypsin inhibitory activity. The
structure of alkaloid 2 includes a rare isoserine residue.
RESULTS AND DISCUSSION
■
Extracts of the sponge A. lacunosa were found to induce 100%
inhibition of α-chymotrypsin at 50 μg·mL−1. The total MeOH
extract of A. lacunosa was separated by progressive solvent
partitioning into four fractions, A−D, and α-chymotrypsin
inhibitory activity was found to partition into the CH2Cl2/
MeOH-soluble “B fraction”. Further purification of the latter
by size exclusion chromatography further segregated activity
into mid-eluting and late-eluting fractions (Figure 1). HPLC
1
and examination by LCMS and H NMR delivered three new
bromotyrosine-derived metabolites (1−3) in addition to the
known compounds aeroplysinin-1,9 aerophobin-1,10 aeropho-
bin-2,11 purealidin-N,12 purealin,13 methyl ester 4a,14 and
aplysinin-B7a (for structures, see Supporting Information). MS
analysis of 1 and 2 showed a [M + Na]+ of 474.9113 and
504.9215, corresponding to molecular formulas of
C13H14Br2N2O6 and C14H16Br2N2O7, respectively. The iso-
topic patterns of the sodium adduct ions confirmed the
presence of two Br atoms in each of 1 and 2. Additionally,
FTIR absorptions at ∼3400 and 1680−1670 cm−1 revealed the
Received: December 31, 2019
© XXXX American Chemical Society and
American Society of Pharmacognosy
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX
A

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 1. α-Chymotrypsin inhibition by fractions from size-exclusion chromatography (Sephadex LH-20, MeOH elution) of Aplysina lacunosa
solvent-partitioned extract. See Experimental Section.
1
Table 1. H and 13C NMR Data for 1−3 in CD3OD
δC, type
δH (int., mult, J, Hz)
ab
,
a
a
c
d
d
no.
1
2
3
1
2
3
1
2
3
4
5
6
7
75.2, CH
113.9, C
148.7, C
122.1, C
132.3, CH
91.9, C
74.0, CH
112.6, C
147.8, C
121.3, C
130.6, CH
91.0, C
74.1, CH
112.9, C
147.7, C
121.3, C
130.8, CH
90.9, C
4.08, s
4.11, s
4.08, s
6.42, s
c
6.41, s
6.443 [6.437 ], s
39.9, CH2
38.7, CH2
38.7, CH2
3.76, d (18.0)
3.09, d (18.0)
3.79, d (18.6)
3.11, d (18.6)
3.78, d (18.6)
3.10, d (18.6)
8
9
10
154.7, C
160.3, C
41.3, CH2
153.4, C
160.2, C
42.5, CH2
153.8, C
160.2, C
36.6, CH2
e
4.01, s
3.600 [3.589 ], dd (13.0, 6.0, 4.8)
3.58, t (6.9)
e
3.600 [3.589 ], dd (13.0, 6.0, 4.8)
11
12
13
14
170.4, C
69.1, CH
172.8
29.2, CH2
70.8, CH2
151.2, C
4.34, dd (6.0, 4.8)
2.10, pent (6.6)
4.04, t (6.0)
117.5, C
14′
117.5, C
15
15′
16
133.1, CH
133.1, CH
136.2, C
7.46, s
7.46, s
17
18
27.4, CH2
150.6, C
3.81, s
19
164.2, C
20
21
22
37.6, CH2
24.4, CH2
133.1, C
3.56, t (6.7)
2.93, t (6.7)
23
24
3-OMe
11-OMe
12-OMe
116.1, CH
133.3, CH
59.0, CH3
7.30, s
f
8.77, s (1JCH = 218 Hz )
60.2, CH3
52.3, CH3
58.9, CH3
51.3, CH3
3.71, s
3.72, s
3.75, s
3.77, s
3.72, s
a
b
c
d
e
13C chemical shifts were obtained indirectly from HSQC and HMBC data. 125 MHz, (CD3)2CO. 500 MHz. 600 MHz. Signal for the C-11
f
epimer. See text for discussion of multiplet complexity and Figure 3b for NMR simulation. From “13C satellite” measurements.
presence of OH/NH and amide carbonyl groups, respectively.
Comparison of the 1H and 13C NMR data of 1 and 2 (Table 1)
with reported literature values for (+)-purealidin-R12 (4b,
Figure 2) demonstrated that both 1 and 2 contained the SIO
moiety with different side chains attached to the amide NH.
The 1H and 13C NMR spectra of 1 showed the characteristic
features of the O-Me-spiroisoxazoline ring system common to
purealidin-R (4b) and many other bromotyrosine-derived
3.72, s; δC 52.3) and the deshielded methylene CH2-10 (δH
4.01, 2H, s; δC 41.3). The C-9 signal (δC 170.4) in 1 was
assigned to a carboxamide which was linked, sequentially by
HMBC correlations, to the deshielded CH2-10 (δH 4.02, 2H)
and the terminal methyl ester group at δH 3.72, therefore
constituting a Gly-OMe residue. Compound 1 is the N-Gly-
OMe-extended derivative of 4b.
The H and HSQC spectra of 2 indicated a diastereotopic
CH2 group (δH 3.64, m and 3.60, m) with a complex multiplet
structure (see Figure 3 and discussion, below), an O-CH-11
1
1
natural products from Verongid sponges. Additionally, the H
and HSQC spectra of 1 revealed an extra C-11 OMe group (δH
B
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
(δH 4.34, dd, 6.0, 4.8), and the 13-MeO signal (δH 3.75, s; δC
58.9) in addition to the anticipated SIO ring. A COSY cross-
peak between the CH2-10 and CH-11 signals, together with
HMBC correlations between CH2-10 and C-9/C-11/C-12,
between CH-11 and C-11/C-12, and between C-13-MeO and
the carbonyl C-12 (δC 172.8) suggest 2 comprises an SIO
joined to an isoserine (isoSer) methyl ester residue through an
amide bond.
Compound (−)-3, with the formula C27H28Br4N6O7, is the
desamino analogue of purealin, a potent inhibitor of myosin
and dynein ATPase isolated from an Okinawan specimen of
Psammaplysilla purea.13,14b The 1H and 13C NMR spectra of 3
were almost identical with those of purealin;13 the major
differences were the presence of an imidazolyl 1H NMR signal
Figure 2. COSY and HMBC correlations of (−)-3 (600 MHz,
CD3OD).
1
H-23 (δH 8.77, s, JCH = 218 Hz) and other minor chemical
shift changes near the histamine group. Full 1H and 13C NMR
assignment of 3 was secured through DQF-COSY, HSQC, and
HMBC (Figure 2). HMBC correlations were observed from
H-21 (δH 2.93, 2H, t, 6.7) to C-23 (δC 116.1) and C-24 (δC
133.3). As with almost all natural products containing α-
ketoxime-bromotyrosine amides, (−)-3 has the E-oxime
configuration.15 The low specific rotation ([α]D −11) and
weak Cotton effect [λ 253 (Δε +0.8), 284 (+0.8)] of (−)-3
suggest this sample is nearly racemic (see below).
The configuration of the isoSer in 2 was addressed by a
variant of Marfey’s method.16 Total hydrolysis of 2 (6 M HCl,
110 °C, 16 h), followed by treatment of the residue with our
newly described L-FDTA reagent (5, Scheme 1)17 gave DTA
Scheme 1. Hydrolysis of 2 and Derivatization of isoSer by L-
FDTA (5)
derivatives with superior resolutions on reversed-phase
HPLC.18 Unexpectedly, two peaks corresponding to L-isoSer-
L-DTA (tR = 13.63 min) and D-isoSer-L-DTA (tR = 13.94 min)
were observed, suggesting the isoSer residue in 2 was a mixture
of epimers. When the hydrolysis was repeated under milder
conditions (4 M HCl, 110 °C, 12 h), followed by
derivatization by L-FDTA, HPLC revealed the same mixture.
To exclude the possibility of epimerization during acid
hydrolysis of 2, authentic L-isoSer was subjected to the latter
conditions of hydrolysis, and upon L-FDTA derivatization and
HPLC analysis, a single peak was obtained corresponding to L-
isoSer-L-DTA. Consequently, the above results confirm that 2
is a mixture of C-11 epimers (1:1).
1
Figure 3. (a) H NMR spectra of the diastereotopic 10-CH2 of 2
1
(600 MHz, CD3OD) and (b) simulated spectrum of ABX. Line width,
1.15 Hz, δA 3.589, J = 13.0, 6.0, 4.8 Hz; δB 3.644, J = 13.0, 6.0, 4.8 Hz.
Epimer: δA 3.600, J = 13.0, 6.0, 4.8 Hz; δB 3.652, J = 13.0, 6.0, 4.8 Hz
(δX is not shown). Ref 20.
Given the results above, the complexity of the H NMR
spectrum of 2, particularly the 10-CH2 multiplet structure,
could now be interpreted in a different, consistent manner.
From inspection of the structure of 2, the expected multiplet
pattern of NH−CH2−CH(OH)− (Figure 3a) should be that
C
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
of an ABX system (the couplings of OH and NH signals are
absent due to deuterium exchange in CD3OD: see DQF-
COSY, 600 MHz, Figure S6). The CH2-10 signal should,
therefore, consist of a maximum of eight lines; yet clearly 16
∼250 nm observed in samples of fistularin-3 (S2) range from
Δε +16 to +32.1 (Table 2). We were left to consider the
possibility that 1−3 are not homochiral; that is, the SIO ring
system is formed with imperfect stereofidelity.30 There is
precedence for this phenomenon; for example, ianthesine-A
(S14) and -B−D, from an Australian Ianthella sp.,31 purealidin
B (S15) from an Okinawan Psammaplysilla purea,32 and
pseudoceratinines A and C (S16 and S17) from P. verrucosa33
have SIO rings of the antipodal (1S,6R) configuration. Capon
and co-workers isolated ( )-purealin (S18) from Southern
Australian specimens of Pseudoceratina spp. and convincingly
demonstrated that coisolated (−)-pseudoceratinine A (S16)
and (−)-aerophobin-2 were partial racemates after conversion
(R-MTPA, DCC) to a mixture of their respective diastereo-
meric R-Mosher’s esters (dr 2:1).34 From a sample of P.
verrucosa collected in Dampier, Western Australia, Karuso and
co-workers found the (−)-enantiomer of (+)-purpuroceratic
acid B (8b)35,36 along with other antipodal SIO compounds.37
Finally, the curious case of 4c offers an even more interesting
study: The EC of 4c (Table 2) is of lower magnitude and
opposite in sign to that of fistularin-3 (S2), both isolated from
the same specimen of Pseudoceratina sp. from the Bahamas.38
The 1,3-diene twist of SIOs gives rise to a modest CE, which
becomes difficult to detect in “near racemic” samples. In order
to enhance the ECD signal in SIOs and provide an
independent indicator of chirality, we exploited the useful
cyclic allylic benzoate method developed by Nakanishi and co-
workers.39 The exciton coupling (EC) observed between the
two nondegenerate chromophoresa benzoyloxy group and
the adjacent CC bond−gives rise to a split Cotton effect, of
which the sign of the longer wavelength component is
determined reliably from the helicity. When extended to
acyclic 1,3-dienyl naphthoate esters, the magnitude of the EC
is enhanced significantly. From energy-minimized models of 4a
(Figure 4), it is apparent the diene helicity (θ (H-1−C-1−C-
6−H6 = 17.2°) and allylic helicity (θ (H-1−C-1−C-6−H6) =
102.2°) are of the same sign. We reasoned that the
superposition of CEs arising from helicities of the 1,3-diene
and 1,3-dienyl naphthoate moieties would reinforce the
magnitude of the CE.
1
lines are present. The most likely interpretation is that the H
NMR signals for CH2-10 of the epimeric mixture 2 are two
overlapping eight-line spin systemsthe AB parts of two ABX
spin systemswith virtually the same vicinal and geminal J
values, but each slightly offset in chemical shift by different
amounts (on average, Δδ ∼0.01, or 6.6 Hz at 600 MHz).19
Because the nearest stereocenter C-6 is six bonds away, the 10-
CH2 J-splitting pattern would be expected to be similar for
each epimer, dictated only by local torsional parameters,
1
solvation, and hydrogen bonding. The simulated H NMR
spectrum20 (Figure 3b) supports these conclusions. In short,
the deceptive spectral complexity of the 10-CH2 1H NMR
signal of 2 arises from overlapping spin systems of
diastereomers.21
The absolute configurations of the brominated cyclo-
hexadienyl-SIOs were originally assigned by Rinehart and co-
workers, who correlated the ECD spectrum and X-ray crystal
structure of the degenerate skewed 1,3-dienes in (+)-aero-
thionin,22,23 based on earlier studies of aeroplysinin-1 by
Fulmor and co-workers24 and pioneering chiroptical studies of
skewed dienes by Moscowitz25 and others.26 The 1,3-diene in
the canonical SIO structure, verongidoic acid (4c), deviates
from planarity by approximately +17°27 (Figure 4) and is
Figure 4. Energy-minimized conformers (MMFF) of (1S,6R)-
verongidoic acid methyl ester (4a). (a) The six lowest E conformers.
(b) Model of the lowest E conformer (iSpartan).
Treatment of synthetic (+)-4a14c with 6-methoxy-2-naph-
thoyl chloride (6)40 (Et3N, DMAP, CH2Cl2, Scheme 2) gave
naphthoate ester (−)-7. The ECD spectrum of naphthoate
ester (−)-7 (Figure 5) showed additional complexity not seen
in the ECD spectrum of the parent SIO including a red-shifted
positive Cotton effect (λ 310 nm, Δε +7.6) corresponding to
the forbidden R-band (n → π*) of the naphthoate
chromophore. Although the magnitude of the CE in (−)-7 is
no greater than native SIOs, the “fingerprint” CE of similar 6-
methoxy-2-naphthoate esters may be useful for fine-scale
chiroptical analyses where longer wavelength features are
preferable.
Brominated SIOs arise from bromination−oxidation of
tyrosine, as shown by Rinehart and co-workers using 14C-
labeling.47 Much speculation has appeared on the biosynthesis
of the SIO and subsequent transformation, the consensus
being that the SIO ring system arises from oxidation of a 3,5-
Br2Tyr-α-ketoxime to an 1,2-arene epoxide followed by
intramolecular SN2 attack−ring opening by the CNOH
group. The dominance of (+)-(1R,6S)-SIO stereoisomers in
Nature attests to re-facial selectivity of the putative arene
epoxidase, with the less common (−)-(1S,6R) enantiomorphs
arising from variant enzymes exhibiting preferential si-facial
expected to manifest a positive Cotton effect (CE). In fact, two
positive CEs are observed in the C2 dimer, (+)-aerothionin
(Table 2): λ 245 nm (Δε +23.7) and 284 (+21.4).23 The
optical rotations of the commonly configured (1R,6S)-SIOs are
also consistently dextrorotatory (Table 2). Similar trends are
observed in other brominated skewed 1,3-cyclohexadienes,
(−)-aeroplysillin-1 (S5) and aplysinafulvin (S8), which lack an
SIO ring system. The ECD spectra of 1−3 (Figure S14) each
showed the same positive CEs, albeit of lower magnitudes than
(+)-aerothionin. The strongest CEs were those of 1 [λ 253
(Δε +5.2), 284 (+3.6)] and the weakest, those of 3. It can be
concluded that the dominant chirality of the SIO ring in 1−3 is
(1R,6S), as depicted, but 3 is nearly racemic. Further
chiroptical determination of the %ee of 1−3 is impeded by
lack of reliable standard SIOs of known %ee.28
Concern was raised about the relatively lower magnitude
CEs observed in 1−3. Given that Δε is a molar quantity, the
magnitude of the CE should be relatively invariant across all
similar SIOs that lack additional stereogenic centers,29 yet this
is not the case. For example, the magnitudes of the CE at λ
D
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 2. Chiroptical Data ([α]D, ECD) for Three Chemotypes (A−C) of Dibromocyclohexa-1,3-dienes
a
a
compound
type
origin
Bahamas
Bahamas
Bahamas
Mexico
Belize
species
[α]D
λ/nm (Δε)
λ/nm (Δε)
ref
c
b
b
b
d
1
2
3
C
C
C
C
C
C
C
A
A
C
C
C
C
C
C
B
Ap. lacunosa
Ap. lacunosa
Ap. lacunosa
+71.8
+32
253 (+5.2)
253 (+3.1)
253 (+0.8)
245 (+23.7)
273 (+32.1)
284 (+3.6)
284 (+2.8)
289 (+0.8)
c
c
−11
+210
e
(+)-aerothionin (S1)
fistularin-3 (S2)
araplysillin III (S3b)
hexadellin C (S4)
(−)-aeroplysinin-1 (S5)
(+)-aeroplysinin-1 (S6)
“verongidoic acid” (4c)
S2
Ap. fistularis
284 (+21.4)
287 (+27.6)
22
5c
5c
5c
24
24
48
41
48
52
52
42
43
44
44
45
45
50
50
37
31
32
33
13
34
46
46
f
Ai. crassa
Ai. crassa
Ai. crassa
Ia. ardis
f
g
Belize
Belize
+96.3
+102
−198
+182
f
h
ij
k
Caribbean
Bahamas
Bahamas
Virgin Islands
Bahamas
Brazil
Bahamas
Australia
Brazil
282 (−15)
245 (−9.1)
k
Ia. sp.
Ps. crassa
li
284 (−6.4)
m
p
Ap. f. fulva
+104.2
il
S2
S2
S2
Ps. crassa ,
245 (+16)
284 (+15)
Ap. cauliformiso
252 (+19.7)
251 (+19.0)
248 (+23.7)
251 (+1.63)
250 (+18.2)
251 (+18.3)
248 (+24)
286 (+19.7)
285 (+17.8)
284 (+20.6)
285 (+0.79)
288 (+15.2)
288 (+15.0)
284 (+24)
p
‘Ap. fulva’
q
r
s
11-epi-fisturalin-3 (S7)
Ag. oroides
+65.2
+130
t
u
aplysinafulvin (S8)
Ap. fulva
d
(12S)-hydroxy-11-oxo-aerothionin (S9)
(12R)-hydroxy-11-oxo-aerothionin (S10)
11-hydroxyaerothionin (S11)
Caissarine C (S12)
C
C
C
C
C
C
Bahamas
Bahamas
Brazil
Ap. fistularis
Ap. fistularis
Ap. caissara
Ap. caissara
+152.5
+162.7
d
v
w
+178
+175
+100
+140
v
x
Brazil
248 (+26)
244 (+8.3)
243 (+7.0)
290 (+27)
285 (+8.0)
289 (+5.2)
t
y
N-sulfatoaraplysillin I (S13)
(+)-purpuroceratic acid B (8b)
(−)-purpuroceratic acid B (8b)
Bahamas
Bahamas
Australia
Australia
Japan
New Caledonia
Japan
Australia
Egypt
‘Ap. fulva’
t
z
‘Ap. fulva’
aa
ab
Ps. verr
−9
ac
ad
ag
ianthesine A (S14)
purealdin B (S15)
C
C
C
C
C
C
C
Ianthella sp.
−118
248 (−10.2)
252 (−2.5)
248 (−9.66)
245 (−9.51)
285 (−9.94)
290 (−2.4)
290 (−8.48)
284 (−9.15)
ac
ae
af
Psa. pur.
Ps. verr.
−4.5
z
aa
pseudoceratinine A (S16)
−158
z
ae
ah
(−)-purealin (S18)
Psa. pur.
−85
ai
( )-purealin (S18)
subereamolline A (S19a)
subereamolline A (S19a)
Ps. sp
Suberea mollis
Suberea mollis
aj
+156.5
ak
Egypt
+22.9
a
MeOH. Literature values of molar ellipticities, [θ], are normalized: Δε = 3300[θ]. See Supporting Information for the structures S1−S19b.
b
c
d
e
f
g
h
Aplysina lacunosa. This work. See Experimental Section. Aplysina fistularis. (c 1.0, MeOH). Aiolochroia crassa. (c 0.19, MeOH). (c 0.067,
i
j
k
l
m
MeOH). Renamed Aiolochroia crassa. Ianthella ardis. (c 0.5, acetone). Pseudoceratina crassa. Pseudovibrio denitrificans Ab134 isolated from the
n
o
p
sponge Arenosclera braziliensis. Aplysina forma fulva. Aplysina cauliformis. Originally identified as Aplysina fulva, but upon review, it is more likely
Aplysina fistularis. The configuration at C-11 is variable.52 Agelas oroides. (c 1.04, acetone). Aplysina fulva. (c 0.002, MeOH). Aplysina caissara.
aa ab ac
(c 0.0014, MeOH). (c 0.0021, MeOH). (c 0.06, MeOH). (c 0.04, MeOH). Pseudoceratina verrucosa. (c 0.4, MeOH). Less common
ad ae af ag ah ai
(1S,6R) configuration. (c 1.02, MeOH). Psammaplysilla purea (likely P. purpurea). (c 1.3, MeOH). (c 1, MeOH). (c 2.1, MeOH). No
q
r
s
t
u
v
w
x
y
z
aj
ak
chiroptical data; presumably “0”. (c 0.55, MeOH). (c 6.25, CH2Cl2).
selectivity or even nonselectivity in the cases of near-racemic
SIOs. Another way to view this outcome is imperfect
desymmetrization by O atom addition to the aryl ring of the
3,5-Br2Tyr-α-ketoxime precursor.
Scheme 2. Naphthoate Ester (−)-7 from Acylation of (+)-4a
In this report, our interest was on the origin of isoSer, a rare
nonproteinogenic amino acid observed in 2 and the first time
in an SIO. Although it is possible that isoSer derives from an
independent, discrete biosynthetic pathway like most other
AAs, the rac-modification seen in 2 suggests an alternative
explanation: a postassembly oxidation of the SIO-containing
compound (Scheme 3). Condensation of 4c (reported from
Pseudoceratina sp.48 and as a fermentation product of
Pseudovibrio denitrificans Ab134849 and given the name
“verongidoic acid”) with β-alanine would give purpuroceratic
acid A (8a),36 a likely intermediate also found with
E
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
tions of nonspecific hydroxylation of spiroisoxazolines from
Aplysina sp. support the notion that such “biotransformations”
may be commonplace in the biosyntheses of brominated
SIOs.52 Finally, a unifying hypothesis comes to mind. The
structure of purealidin-R (4b), itself, may arise from a well-
known pathway to C-terminal primary amides:53 α-hydrox-
ylation of the terminal Gly residue of ii (again, 1 arises from ii
by adventitious methylation of ii) to give the carbinolamine iii
followed by hydrolytic loss of glyoxylic acid to give 4b.
Several peptides with isoSer residues have been shown to be
protease inhibitors. For example, a series of synthetic isoSer-
containing peptides showed significant inhibition of amino-
peptidase-N.54 Upon assay, none of the new compounds 1−3
or the other seven known secondary metabolites from the
active extract of A. lacunosa showed inhibition of α-
chymotrypsin at concentrations up to 25 μg·mL−1, suggesting
the chemical entity responsible for inhibitory activity is a very
minor, potent congener that escaped detection or decomposed
during purification.
In conclusion, three new brominated spiroisoxazoline
alkaloids (1−3) were isolated from extracts of Aplysina
lacunosa, and their structures determined by integrated analysis
of MS, NMR, and ECD data. The rare amino acid ( )-isoSer
was found as a residue in 2, a 1:1 mixture of epimers.
Configurational heterogeneity is noted in spiroisoxazolines,
both in ECD and the range of magnitude of the Cotton effects
(Δε) and preparation of diastereomeric amides from 4c. None
of the new or known compounds were responsible for activity
in the α-chymotrypsin inhibition assay.
Figure 5. UV−vis and ECD spectra (c 2.49 × 10−5 M, MeOH, 23 °C)
of 6′-methoxy-2′-naphthoate ester (−)-7 derived from (+)-4a.
Scheme 3. Biosynthetic Hypothesis for 1, 2, and 4b
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured on a JASCO P-2000 polarimeter at the D-double emission
line of Na. UV−vis spectra were measured on a JASCO V-630
spectrometer. ECD spectra were measured on a JASCO J-810
spectropolarimeter at 23 °C in quartz cells of 1, 2, or 5 mm path
length. FTIR spectra were collected from thin film samples using a
JASCO FTIR-4100 fitted with an ATR ZnSe plate. 1D and 2D NMR
spectra were measured on a JEOL ECA (500 MHz) spectrometer,
equipped with a 5 mm 1H{13C} room-temperature probe, or a Bruker
Avance III (600 MHz) NMR spectrometer with a 1.7 mm 1H-
{13C/15N} microcryoprobe (23 °C). Chemical shifts were referenced
1
to internal solvent or residual H signals (CDCl3, δH 7.26 ppm; δC
77.16. CD3OD, δH 3.31; δC 49.00). LC-MS measurements were
performed with a Thermoelectron Surveyor UHPLC coupled to an
MSD single-quadrupole detector. HR-ESI-TOF mass spectroscopy
analyses were conducted on an Agilent 1200 HPLC connected to an
Agilent 6350 TOF-MS at the Small Molecule Mass Spectrometry
Facility at the Department of Chemistry and Biochemistry at UCSD.
Preparative, semipreparative, and analytical HPLC purifications were
carried out on a JASCO system consisting of a UV−vis detector (UV-
2075), dual pumps (PU-2086 Plus), and a dynamic mixer (MX-2080-
32).
Biological Material. The sponge Aplysina lacunosa (11-14-018)
was collected in 2011 from Little San Salvador Island in the Bahamas
(lat. 24° 35.242′ N, long. 75° 58.495′ W) at a depth of −23 m using
scuba and kept frozen (−20 °C) until needed. A type sample
(MeOH) is archived in the Department of Chemistry and
Biochemistry, UCSD.
Extraction and Isolation. A sample of A. lacunosa (11-14-018)
was lyophilized (dry wt 24.0 g) and extracted twice with MeOH (2 ×
400 mL). The combined, filtered extracts were concentrated under
reduced pressure to yield an extract (5.17 g), which was dissolved in
MeOH/H2O (9:1, 400 mL) before repeated partitioning against
hexanes (3 × 400 mL) to provide the hexane-soluble “A” fraction
(0.2517 g). The MeOH/H2O solution (6:4, 600 mL) was adjusted to
homologues (e.g., purpuroceratic acid B, 8b), in at least two
other Verongid sponges.14a,50 Compound 8a would then suffer
stereorandom α-hydroxylation by a monooxygenase enzyme
(Scheme 3), giving carboxylic acid i; the corresponding methyl
ester 2 arises as an artifact of adventitious Fischer−Speier
esterification during extraction with MeOH.51 Earlier observa-
F
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
40% v/v H2O, and the mixture repeatedly partitioned against CH2Cl2
(2 × 400 mL). The combined CH2Cl2 layers were concentrated to
deliver the CH2Cl2-soluble “B-fraction” (0.8412 g) after removal of
volatiles. The MeOH/H2O layer was concentrated under reduced
pressure, and the residual aqueous layer extracted with n-BuOH (2 ×
300 mL) to yield the n-BuOH-soluble “C” layer (1.2460 g) and H2O-
soluble “D” layer (2.8300 g) after removal of volatiles.
MS (Hypersil Gold, C18 column, 50 × 2.1 mm, 1.9 μm, linear
gradient, initial conditions 15−45% CH3CN/H2O−0.1% HCOOH
over 25 min, 0.5 mL·min−1 flow rate). L-DTA derivatives of authentic
standards of L- and DL-isoSer eluted gave peaks at tR = 13.63 and 13.94
min, respectively. The LC-MS analysis of the L-DTA derivative of the
acid hydrolysate of 2 gave both L-isoSer-L-DTA and D-isoSer-L-DTA
(ratio ∼1:1).
6′-Methoxy-2′-naphthoate Ester ( )-7. A solution of ( )-4a14c
(1.0 mg, 2.5 μmol) in CH2Cl2 (0.2 mL) was treated sequentially with
6-methoxy-2-naphthoyl chloride (0.55 mg, 10 μmol, prepared from
the corresponding carboxylic acid and oxalyl chloride), Et3N (1.0 mg,
10 μmol), and DMAP (∼0.5 mg) at 0 °C, then stirred for 16 h at 23
°C. The mixture was concentrated, applied to a preparative TLC plate
(silica, 20 × 20 × 0.1 cm), and the plate eluted (40:60 EtOAc/
hexanes) to give a nonpolar UV-active fraction (1.4 mg) containing
( )-7 (Rf = 0.26, 35:65 EtOAc/hexanes). The latter was further
purified by HPLC (normal phase, Rainin Dynamax 10 × 250 mm, 5
μ, 35% EtOAc/hexanes, 4.0 mL·min−1) to give ( )-7 as a straw-
yellow solid (0.73 mg, 49%): UV−vis (MeOH) λmax 209 nm (log10 ε
4.16), 241 (4.17), 255 (4.18), 310 (3.48); 1H NMR (500 MHz,
CDCl3) δ 8.49 (1H, s), 7.97 (1H, dd, J = 8.4, 1.2 Hz), 7.85 (1H, d, J =
9.0 Hz), 7.77 (1H d, J = 8.4 Hz), 7.21 (1H, dd, J = 9.0, 2.4 Hz), 7.16
(1H, d, J = 2.4 Hz), 6.40 (1H, s), 6.18 (1H, s), 3.96 (3H, s, 6′-OMe),
3.85* (3H, s, 3-OMe), 3.82* (3H, s, COOMe), 3.61 (1H, d, J = 18.6
Hz, H-7b), 3.10 (1H, d, J = 18.6 Hz, H-7a); HRMS (ESI-TOF) m/z
601.9418 [M + Na]+ (calcd for C14H16Br2N2O7Na+, 601.9420).
(1R,6S)-6′-Methoxy-2′-naphthoate Ester (1R,6S)-7). A sample of
( )-4a (4.2 mg) was separated by semipreparative chiral-phase
HPLC (Phenomenex, Amylose-2, 10 × 250 mm, 30:70 CH3CN/
H2O, 4.0 mL·min−1) to give pure (+)-4a (tR = 18.2 min, 1.5 mg) and
(−)-4a (tR = 30.6 min, 1.5 mg). A sample of (+)-4a (0.78 mg, 2.0
μmol) was acylated with freshly prepared 6-methoxy-2-naphthoyl
chloride (7.9 μmoL) and Et3N (7.9 μmol) in CH2Cl2 (80 μL) to give
(1R,6S)-7a (0.29 mg) after purification by HPLC (normal phase,
Rainin Dynamax 10 × 250 mm, 5 μ, 35% EtOAc/hexanes, 4.0 mL·
The “B-fraction” (0.7342 g) was separated by size-exclusion
chromatography (Sephadex LH-20, MeOH), and the eluate grouped
into 16 fractions by TLC profiling (9:1 of CH2Cl2/MeOH, UV-
activity, p-anisaldehyde stain). The tenth fraction (18.5 mg) was
further purified by reversed-phase preparative HPLC (Phenomenex
C18 column, 150 × 21.2 mm, 5 μm, under gradient elution: initial
conditions 10% CH3CN/H2O−0.1% TFA for 3 min to 60% CH3CN/
H2O−0.1% TFA over 20 min, 13 mL·min−1 flow rate) to yield
lacunosin A (1, 3.2 mg) eluting at tR = 17.19 min. The 11th fraction
(10.0 mg) was also purified by reversed-phase semipreparative HPLC
(Phenomenex C18 column, 250 × 10 mm, 5 μm, step gradients, initial
conditions: 30% CH3CN/H2O−0.1% TFA for 5 min to 75%
CH3CN/H2O−0.1% TFA for an additional 25 min, 2.5 mL·min−1
flow rate) to yield 13 fractions. Fraction 6 (0.3 mg) contained
lacunosin B (2) eluting at tR = 18.95 min. Fraction 9 (0.2 mg), which
eluted at tR = 21.38 min, was further purified by reverse-phase
analytical HPLC (Synergi Hydro-RP column, 150 × 4.6 mm, 4 μm,
step gradients, initial conditions 20% CH3CN/H2O−0.1% TFA for 3
min to 70% CH3CN/H2O−0.1% TFA for 32 min, 0.7 mL·min−1 flow
rate) to yield desaminopurealin 3 (0.1 mg), which eluted at tR = 24.50
min. An additional 0.7 mg of compound 3 was purified from the 12th
fraction (11.7 mg) of the LH-20 column under similar semi-
preparative HPLC conditions as the 11th fraction. Additional
repeated HPLC separations delivered aeroplysinin-1,9 aerophobin-
1,10 aerophobin-2,11 purealidin-N,12 methyl ester 4a,14a,b and
aplysinin-B7a and whose identities were confirmed by MS and H
1
NMR.
Lacunosin A, 1: pale yellow solid; [α]23.8D +71.8 (c 1.00, MeOH);
ECD (c 3.93 × 10−4 M, MeOH) λmax (Δε) 214 (−1.8), 227 (0), 253
(+5.08), 282 (+3.7) nm; UV (MeOH) λmax (log ε) 194 (3.48), 230
(3.37), and 282 (3.09); FTIR (ATR, film, ZnSe) ν 3353, 1744, 1671,
1599, 1541, 1438, 1205, 1133, 1047, 1025, 988, 916, 837, 801, 766,
min−1 tR = 7.4 min). (1R,6S)-7: colorless solid; [α]23 −21 (c 0.058,
D
MeOH); UV−vis (MeOH) λmax (log10 ε) 210 (4.11), 237 (4.37), 253
(4.20), and 310 (3.82) nm; ECD (2.49 × 10−5 M, MeOH) λ (Δε)
1
217 (−5.7), 267 (+4.7), and 310 (+7.6) nm; H NMR (600 MHz),
1
and 722 cm−1; H and 13C NMR data, Table 1; HRMS (ESI-TOF)
identical with that of ( )-7; HRMS (ESI-TOF) m/z 596.9860 [M +
+
NH4]+ (calcd for C27H19Br2N2O7 , 596.9867).
m/z 474.9113 [M + Na+] (calcd for C13H14Br2N2O6Na+, 474.9111).
Lacunosin A, 2: white solid; [α]23.9 +31.8 (c 0.15, MeOH); UV
D
(MeOH) λmax (log ε) 194 (3.68), 225 (3.48), and 281 (3.15) nm;
ECD (c 3.56 × 10−4 M, MeOH) λmax 214 (Δε −1.4), 227 (0), 251
(+3.1), and 282 (+2.7) nm; FTIR (ATR, film, ZnSe) ν 3427, 1685,
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.9b01286.
1
1207, 1141, 805, and 724 cm−1; H and 13C NMR data, Table 1;
HRMS (ESI-TOF) m/z 504.9215 [M + Na]+ (calcd for
C14H16Br2N2O7Na+, 504.9216).
Desaminopurealin, 3: white solid; [α]23.5D −10.5 (c 0.13, MeOH);
UV (MeOH) λmax (log ε) 205 (3.88), 221 (4.56), and 275 (3.04) nm;
ECD (c 4.63 × 10−4 M, MeOH) λmax (Δε) 214 (−0.34), 227 (0), 251
(+0.8), and 282 (+0.7); FTIR (ATR, film, ZnSe) ν 3405, 1684, 1207,
Structures of several known cyclohexa-1,3-dienyl SIOs
(see Table 2), UV-ECD, 1D and 2D NMR, and HRMS
data for compounds 1−3, 1H HNMR of (−)-7, UV−vis,
ECD for 4a, S-6, and R-18, HRMS data, and description
of the α-chymotrypsin assay (PDF)
1
1142, 804, and 722 cm−1; H and 13C NMR data, Table 1; HRMS
+
(ESI-TOF) m/z 864.8828 [M + H]+ (calcd for C27H29Br4N6O7 ,
864.8826).
Compound 4a: white solid; [α]21.4 +145 (c 0.44, MeOH), lit.14c
D
AUTHOR INFORMATION
[α]27.9 +165.5 (c 0.325, C6H6); ECD (1.07 × 10−3 M, MeOH) λ
■
D
1
Corresponding Author
(Δε) 222 (4.48), 252 (12.58), and 289 (−4.83) nm; H NMR and
HRMS of 4a were consistent with literature values.14b,c
Tadeusz F. Molinski − Department of Chemistry and
Biochemistry and Skaggs School of Pharmacy and
Pharmaceutical Sciences, University of California, San Diego, La
Jolla, California 92093-0358, United States; orcid.org/
0000-0003-1935-2535; Phone: +1 (858) 534-7115;
Email: tmolinski@ucsd.edu; Fax: +1 (858) 822-0386
Acid Hydrolysis of 2. A sample of 2 (0.1 mg, 0.2 μmol) was
dissolved in 4 M HCl (100 μL, 400 μmol) and stirred in a sealed vial
at 110 °C for 12 h. The solution was cooled to room temperature,
dried under a stream of N2, and derivatized with L-FDTA (5)17 to
yield the isoSer-DLT derivatives of 2 for LC-MS analysis (see below).
Absolute Configuration of isoSer in Compound 2. To the acid
hydrolysate of 2 in H2O (50 μL) were added 1 M NaHCO3 (50 μL,
50 μmol) and L-FDTA17 (5, 0.25 mg, 0.65 μmol) in acetone (75 μL).
The reaction was stirred at 85 °C for 30 min, cooled, and neutralized
with 1 M HCl (50 μL, 50 μmol). The derivatized hydrolysate (20 μL)
was diluted with MeOH (80 μL), centrifuged, and analyzed by LC-
Authors
Mariam N. Salib − Department of Chemistry and Biochemistry,
University of California, San Diego, La Jolla, California 92093-
0358, United States
G
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
49, 2063−2076. (c) Shearman, J. W.; Myers, R. M.; Brenton, J. D.;
Ley, S. V. Org. Biomol. Chem. 2011, 9, 62−65.
Matthew T. Jamison − Department of Chemistry and
Biochemistry, University of California, San Diego, La Jolla,
California 92093-0358, United States
(15) Two lines of evidence to support the presence of
thermodynamically more stable E-geometry include the C-17
chemical shift (CCCCCCCC13C 27.4; cf. E,E-psammaplyin A, δ
27.1). (a) Arabshahi, L.; Schmitz, F. J. J. Org. Chem. 1987, 52, 3584−
3586 and the consistent presence of a hydrogen bond between the
oxime N and the amide NH observed in X-ray structures.
(b) Kazlauskas, R.; Lidgard, R. O.; Murphy, P. T.; Wells, R. J.;
Blount, J. F. Aust. J. Chem. 1981, 34, 765−786. Natural products with
the rare less-stable Z-geometry (viz. bastadin isomers) have been
documented. (c) Calcul, L.; Inman, W. D.; Morris, A. A.; Tenney, K.;
Ratnam, J.; McKerrow, J. H.; Valeriote, F. A.; Crews, P. J. Nat. Prod.
2010, 73, 365−372.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.9b01286
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank M. Cabrera-Abad, K. Planck, and R. Hendra for
assistance in HPLC purification of 1−4a and (−)-7, E. P. Stout
for assistance with collection of marine sponge samples, A.
Mrse and B. Duggan for NMR support, X. Su for HRMS
measurements, and J. Pawlik (UNC Wilmington) and the crew
of the R.V. Walton Smith for the logistics of sample collection
in the Bahamas. M.C.-A. and K.P. were supported by the
STARS summer research program, UC San Diego. Acquis-
itions of the Agilent TOF mass spectrometer and the 500 MHz
NMR spectrometer were made possible with funds from the
NIH Shared Instrument Grant program (S10RR025636) and
the NSF Chemical Research Instrument Fund (CHE0741968),
respectively. We are grateful for support for this research from
the NIH (AI100776, AT009783).
(16) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591−196.
(17) Salib, M. N.; Molinski, T. F. J. Org. Chem. 2017, 82, 10181−
10187.
(18) The Marfey’s derivatives L- and D-isoSer-L-DAA,16 prepared
from standard isoSer samples and L-FDAA, failed to separate under
the same reversed-phase HPLC conditions.
1
(19) Duplication of the H-5 vinyl H NMR signal (Δδ = 0.003;
Table 1) is also observed.
(20) Castillo, A. M.; Patiny, L.; Wist, J. J. Magn. Reson. 2011, 209,
123−130.
(21) This is not contradictory to our independent findings that the
epimer ratio of isoSer in 2 is 1:1 but 67:33 in the SIO heterocycle (see
Note 27 below and discussion in the text). For simplicity of argument,
if the latter ratio is approximated to 2:1 and the components
permuted, the four outcomes are two enantiomeric pairs of
diastereomers: (1S,6R,11S), (1S,6R,11R) and (1R,6S,11R),
(1R,6S,11S) in the proportions 3:3:2:2. The NMR-distinguishable
diastereomers, therefore, are present in the ratio of 5:5 or 1:1.
(22) McMillan, J. A.; Paul, I. C.; Goo, Y. M.; Rinehart, K. L.;
Krueger, W. C.; Pschigoda, L. M. Tetrahedron Lett. 1981, 22, 39−42.
(23) Fattorusso, E.; Minale, L.; Sodano, G. Chem. Commun. 1970,
752−753.
REFERENCES
■
(1) Peng, J.; Li, J.; Hamann, M. T. Alkaloids Chem. Biol. 2005, 61,
59−262.
(2) Lira, N. S.; Montes, R. C.; Tavares, J. F.; da Silva, M. S.; da
Cunha, E. V.; de Athayde-Filho, P. F.; Rodrigues, L. C.; da Silva Dias,
C.; Barbosa-Filho, J. M. Mar. Drugs 2011, 9, 2316−2368.
(3) Ueberlein, S.; Machill, S.; Schupp, P. J.; Brunner, E. Mar. Drugs
2017, 15, 34−49.
(4) Ueberlein, S.; Machill, S.; Niemann, H.; Proksch, P.; Brunner, E.
Mar. Drugs 2014, 12, 4417−4438.
(5) (a) Kobayashi, J.; Tsuda, M.; Agemi, K.; Shigemori, H.;
Ishibashi, M.; Sasaki, T.; Mikami, Y. Tetrahedron 1991, 47, 6617−
6622. (b) Andersen, R. J.; Faulkner, D. J. Tetrahedron Lett. 1973, 14,
1175−1178. (c) Gao, H.; Kelly, M.; Hamann, M. T. Tetrahedron
1999, 55, 9717−9726.
(6) De Medeiros, A.; Gandolfi, R. C.; Secatto, A.; Falcucci, R. M.;
Faccioli, L. H.; Hajdu, E.; Peixinho, S.; Berlinck, R. G. S.
Immunopharmacol. Immunotoxicol. 2012, 34, 919−924.
(7) (a) Shaala, L. A.; Youssef, D. T. A.; Badr, J. M.; Sulaiman, M.;
(24) Fulmor, W.; Van Lear, G. E.; Morton, G. O.; Mills, R. D.
Tetrahedron Lett. 1970, 11, 4551−4552.
(25) Moscowitz, A.; Charney, E.; Weiss, U.; Ziffer, H. J. Am. Chem.
Soc. 1961, 83, 4661−4663.
(26) Burgstahler, A. W.; Ziffer, H.; Weiss, U. J. Am. Chem. Soc. 1961,
83, 4660−4661.
(27) MMFF. Molecular modeling of aeroplysillin-1 shows the diene
has similar helicity (this work, MMFF, dihedral θ = 17.6°), in good
correspondence with the X-ray crystal structure assignments of both
(+)- and (−)-aeroplysillin-1 (θ = 17.3 1). Cosulich, D. B.; Lovell, F.
M. J. Chem. Soc. D 1971, 397−398.
̈
Khedr, A. Mar. Drugs 2015, 13, 1621−1631. (b) Gothel, Q.; Sirirak,
(28) If one assumes, for example, highly dichroic (+)-aerothionin
(S2, Table 2) is optically pure (“100% ee”, also see ref 38), the ratio
of CE magnitudes of S2 and 2 implies the latter is only 19.3% ee (er =
63:37). Lack of precision of ECD measurements on a sub-milligram
sample of 2 limits the accuracy of this conclusion.
(29) We have found the effects of more distal stereocenters in SIOs
upon the net ECD spectrum, e.g., fistularin-3 and 11-hydroxyfistular-
in-3, to be negligible.
(30) Another possibility we considered is that 4c is formed as a
racemic or near-racemic modification and kinetically resolved in
downstream enzymatic coupling reactions. This possibility would
explain enantiomorphic 4c and fistularin-3 in the same sample.48
(31) Okamoto, Y.; Ojika, M.; Kato, S.; Sakagami, Y. Tetrahedron
2000, 56, 5813−5818.
(32) Kobayashi, J. i.; Tsuda, M.; Agemi, K.; Shigemori, H.; Ishibashi,
M.; Sasaki, T.; Mikami, Y. Tetrahedron 1991, 47, 6617−6622.
(33) Benharref, A.; Païs, M.; Debitus, C. J. Nat. Prod. 1996, 59, 177−
180.
̈
T.; Kock, M. Beilstein J. Org. Chem. 2015, 11, 2334−2342.
́
́
(8) Nicacio, K. J.; Ioca, L. P.; Froes, A. M.; Leomil, L.; Appolinario,
L. R.; Thompson, C. C.; Thompson, F. L.; Ferreira, A. G.; Williams,
D. E.; Andersen, R. J.; Eustaquio, A. S.; Berlinck, R. G. S. J. Nat. Prod.
2017, 80, 235−240.
(9) Fattorusso, E.; Minale, L.; Sodano, G. J. Chem. Soc. D 1970,
751−752.
(10) Gunasekera, M.; Gunasekera, S. P. J. Nat. Prod. 1989, 52, 753−
756.
(11) Cimino, G.; De Rosa, S.; De Stefano, S.; Self, R.; Sodano, G.
Tetrahedron Lett. 1983, 24, 3029−3032.
(12) Kobayashi, J.; Honma, K.; Sasaki, T.; Tsuda, M. Chem. Pharm.
Bull. 1995, 43, 403−407.
(13) (a) Nakamura, H.; Wu, H.; Kobayashi, J. i.; Nakamura, Y.;
Ohizumi, Y.; Hirata, Y. Tetrahedron Lett. 1985, 26, 4517−4520.
(b) Nakamura, Y.; Kobayashi, M.; Nakamura, H.; Wu, H.; Kobayashi,
J.; Ohizumi, Y. Eur. J. Biochem. 1987, 167, 1−6.
(14) (a) Ciminiello, P.; Fattorusso, E.; Magno, S.; Pansini, M. J. Nat.
Prod. 1994, 57, 1564−1569. (b) Zhu, G.; Yang, F.; Balachandran, R.;
(34) Salim, A. A.; Khalil, Z. G.; Capon, R. J. Tetrahedron 2012, 68,
9802−9807.
̈
̈
Hook, P.; Vallee, R. B.; Curran, D. P.; Day, B. W. J. Med. Chem. 2006,
H
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
(35) Kijjoa, A.; Bessa, J.; Wattanadilok, R.; Sawangwong, P.;
Nascimento, M. S. J.; Pedro, M.; Silva, A. M. S.; Eato, G.; van Soest,
R.; Herz, W. Z. Naturforsch., B: J. Chem. Sci. 2005, 60, 904−908.
(36) From values of [α]D, we estimate the Karuso sample of (−)-8b
from NW Australia to be only 7% eerather than a different
ionization state, as speculated by the authorsto our sample of
(+)-8b from the Bahamas.48
(37) Ragini, K.; Fromont, J.; Piggott, A. M.; Karuso, P. J. Nat. Prod.
2017, 80, 215−219.
(38) It is perhaps no coincidence that earlier-reported SIOs, purified
by crystallization (e.g., (+)-aerothionin20), are optically enriched
through the process and have the highest values.
(39) (a) Gonnella, N. C.; Nakanishi, K.; Martin, V. S.; Sharpless, K.
B. J. Am. Chem. Soc. 1982, 104, 3775−3776. (b) Harada, N.;
Iwahuchi, J.; Yokota, Y.; Uda, H. A. J. Am. Chem. Soc. 1981, 103,
5590−5591.
(40) The 6-methoxy-substituted naphthoic acid was chosen for its
stronger electronic transition dipole moment (donor−acceptor
effect).
(41) Gopichand, Y.; Schmitz, F. J. Tetrahedron Lett. 1979, 20, 3921−
3924.
̈
(42) Konig, G. M.; Wright, A. D. Heterocycles 1993, 36, 1351−1358.
(43) Nunez, C. V.; de Almeida, E. V. R.; Granato, A. C.; Marques, S.
̃
O.; Santos, K. O.; Pereira, F. R.; Macedo, M. L.; Ferreira, A. G.;
Hajdu, E.; Pinheiro, U. S.; Muricy, G.; Peixinho, S.; Freeman, C. J.;
Gleason, D. F.; Berlinck, R. G. S. Biochem. Syst. Ecol. 2008, 36, 283−
296.
(44) Ciminiello, P.; Costantino, V.; Fattorusso, E.; Magno, S.;
Mangoni, A.; Pansini, M. J. Nat. Prod. 1994, 57, 705−712.
(45) Lira, T. O. d.; Berlinck, R. G. S.; Nascimento, G. G. F.; Hajdu,
E. J. Braz. Chem. Soc. 2006, 17, 1233−1240.
(46) Abou-Shoer, M. I.; Shaala, L. A.; Youssef, D. T. A.; Badr, J. M.;
Habib, A.-A. M. J. Nat. Prod. 2008, 71, 1464−1467.
(47) Carney, J. R.; Rinehart, K. L. J. Nat. Prod. 1995, 58, 971−985.
(48) Aiello, A.; Fattorusso, E.; Menna, M.; Pansini, M. Biochem. Syst.
Ecol. 1995, 23, 377−281.
(49) Compound 4c is likely the precursor to all SIOs.
(50) Rogers, E. W.; Molinski, T. F. J. Nat. Prod. 2007, 70, 1191−
1194.
(51) Counter examples of biosynthesized carboxylic acid methyl
esters have been described. (a) Molinski, T. F.; Ireland, C. J. J. Org.
Chem. 1988, 53, 2103−2105. (b) Ferreira, E. L. F.; Williams, D. E.;
́
Ioca, L. P.; Morais-Urano, R. P.; Santos, M. F. C.; Patrick, B. O.; Elias,
L. M.; Lira, S. P.; Ferreira, A. G.; Passarini, M. R. Z.; Sette, L. D.;
Andersen, R. J.; Berlinck, R. G. S. Org. Lett. 2015, 17, 5152−5155.
(52) Rogers, E. W.; Fernanda de Oliveira, M.; Berlinck, R. G. S.;
̈
Konig, G. M.; Molinski, T. F. J. Nat. Prod. 2005, 68, 891−896.
(53) Owen, T. C.; Merkler, D. J. Med. Hypotheses 2004, 62, 392−
400.
(54) Yang, K.; Fen, J.; Fang, H.; Zhang, L.; Gong, J.; Xu, W. J.
Enzyme Inhib. Med. Chem. 2012, 27, 302−310.
I
https://dx.doi.org/10.1021/acs.jnatprod.9b01286
J. Nat. Prod. XXXX, XXX, XXX−XXX