Serinolamides and Lyngbyabellins from an Okeania sp. Cyanobacterium Collected from the Red Sea

Article abstract of DOI:10.1021/acs.jnatprod.7b00449

NMR- and MS-guided fractionation of an extract of an Okeania sp. marine cyanobacterium, collected from the Red Sea, led to the isolation of four new metabolites, including serinolamides C (1) and D (2) and lyngbyabellins O (3) and P (4), together with the three known substances lyngbyabellins F (5) and G (6) and dolastatin 16 (7). The planar structures of the new compounds were determined using NMR and MS analyses. The absolute configurations of 1 and 2 were determined by Marfey's analysis of their hydrolysates. The absolute configuration of 3 was ascertained by chiral-phase chromatography of degradation products, while that of 4 was determined by comparison to 3 and 5. The cytotoxic and antifouling activities of these compounds were evaluated using MCF7 breast cancer cells and Amphibalanus amphitrite larvae, respectively. Compounds 3, 4, and 7 exhibited strong antifouling activity, and 3 and 7 were not cytotoxic. A structure-activity relationship was observed for the cytotoxicity of the lyngbyabellins with the presence of a side chain (4 is more active than 3) leading to greater activity. For the antifouling activity, the acyclic form without a side chain (3) was the most active.

Full text of DOI:10.1021/acs.jnatprod.7b00449

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX-XXX
Serinolamides and Lyngbyabellins from an Okeania sp.
Cyanobacterium Collected from the Red Sea
Julie G. Petitbois,† Loida O. Casalme,† Julius Adam V. Lopez,† Walied M. Alarif,‡
Ahmed Abdel-Lateff,§,⊥ Sultan S. Al-Lihaibi,‡ Erina Yoshimura,∥ Yasuyuki Nogata,▽ Taiki Umezawa,†,¶
Fuyuhiko Matsuda,†,¶ and Tatsufumi Okino*,†,¶
†Graduate School of Environmental Science and ¶Faculty of Environmental Earth Science, Hokkaido University, Sapporo 060-0810,
Japan
‡Department of Marine Chemistry, Faculty of Marine Sciences, King Abdulaziz University, P.O. Box 80207, Jeddah 21589, Saudi
Arabia
§Department of Natural Products and Alternative Medicine, Faculty of Pharmacy, King Abdulaziz University, P.O. Box 80260, Jeddah
21589, Saudi Arabia
⊥Department of Pharmacognosy, Faculty of Pharmacy, Minia University, Minia 61519, Egypt
∥CERES, Inc., 1-4-5 Midori, Abiko, Chiba 270-1153, Japan
▽Environmental Science Research Laboratory, Central Research Institute of Electric Power Industry, Abiko 270-1194, Japan
S
* Supporting Information
ABSTRACT: NMR- and MS-guided fractionation of an extract of an
Okeania sp. marine cyanobacterium, collected from the Red Sea, led to
the isolation of four new metabolites, including serinolamides C (1) and
D (2) and lyngbyabellins O (3) and P (4), together with the three
known substances lyngbyabellins F (5) and G (6) and dolastatin 16 (7).
The planar structures of the new compounds were determined using
NMR and MS analyses. The absolute configurations of 1 and 2 were
determined by Marfey’s analysis of their hydrolysates. The absolute
configuration of 3 was ascertained by chiral-phase chromatography of
degradation products, while that of 4 was determined by comparison to
3 and 5. The cytotoxic and antifouling activities of these compounds
were evaluated using MCF7 breast cancer cells and Amphibalanus
amphitrite larvae, respectively. Compounds 3, 4, and 7 exhibited strong
antifouling activity, and 3 and 7 were not cytotoxic. A structure−activity
relationship was observed for the cytotoxicity of the lyngbyabellins with the presence of a side chain (4 is more active than 3)
leading to greater activity. For the antifouling activity, the acyclic form without a side chain (3) was the most active.
yanobacteria are a bountiful source of novel bioactive
and dolastatin 16 (7).11−13 Serinolamides C and D were named
with reference to the fatty acid amides serinolamides A and B,
which were isolated from Moorea producens.14,15 The
antifouling and cytotoxic activities of these compounds were
measured using the cyprids of the barnacle Amphibalanus
amphitrite and MCF7 breast cancer cells, respectively.
C
compounds.1 Okeania gen. nov. was recently delineated
from the genus Lyngbya as part of a phylogenetic study
reported by Engene et al. in 2013.2 Several compounds have
subsequently been isolated from O. hirsuta, including
malyngamide C3 and bastimolide A,4 as well as kurahyne B,5
polycavernoside D,6 janadolide,7 odoamide,8 and urumamide9
from Okeania spp. Notably, all seven of these compounds have
been reported to show cytotoxic effects toward various cancer
cell lines.
As part of our ongoing research toward the isolation and
identification of bioactive compounds from Saudi marine
organisms,10 we recently investigated the isolation of
compounds from a dark brown filamentous cyanobacterium
collected from the Red Sea, which was identified as an Okeania
sp. Four new compounds were isolated, including serinolamides
C (1) and D (2) and lyngbyabellins O (3) and P (4), together
with the known compounds lyngbyabellins F (5) and G (6)
RESULTS AND DISCUSSION
■
The Okeania sp. marine cyanobacterium was collected from the
Red Sea and identified based on its tuft-forming morphology2
and a phylogenetic analysis using 16S rRNA gene sequencing
(Supporting Information S1, S2, and S3). The morphology and
the presence of malyngamide C acetate, a metabolite
commonly found in Okeania hirsuta2 and confirmed by LC-
Received: May 25, 2017
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1
Table 1. NMR Spectroscopic Data (1H 400 MHz, 13C 100 MHz, CDCl3) for Serinolamides C (1) and D (2)
serinolamide C (1)
δH (J in Hz)
serinolamide D (2)
δH (J in Hz)
a
a
position
δC, type
169.9, C
HMBC
δC, type
169.2, C
HMBC
1
2
130.3, C
130.5, C
3
137.4, CH
28.4, CH2
28.7, CH2
29.3−29.6, CH2
31.9, CH2
22.6, CH2
14.0, CH3
12.6, CH3
50.8, CH
6.41, m
137.1, CH
28.3, CH2
28.7, CH2
29.329.6, CH2
31.9, CH2
22.6, CH2
14.0, CH3
12.6, CH3
48.1, CH
6.36, m
4
2.14, ddd (7,7,7)
1.43, m
2.14, ddd (7,7,7)
1.42, m
5
6−13
14
15
16
17
18
19
1.26, m
1.26, m
1.26, m
1.26, m
1.26, m
1.26, m
0.88, t (7.0)
1.86, s
14, 15
1, 2, 3
0.88, t (6.4)
1.84, s
14, 15
1, 2, 3
4.13, m
4.40, m
64.4, CH2
3.71, dd (11.8, 4.2)
3.86, dd (11.4, 3.8)
3.60, dd (9.6, 3.6)
3.63, dd (10.0, 3.6)
3.38, s
63.3, CH2
4.14, dd (11.4, 6.0)
4.28 (11.2, 6.4), dd
3.53, dd (9.8, 3.4)
20
73.3, CH2
59.3, CH3
71.1, CH2
21
22
20
59.2, CH3
171.3, C
3.37, s
20
22
3.46, dd (9.6, 4.8)
2.07, s
23
20.8, CH3
NH
6.41, m
6.11, d (7.6)
a
HMBC data optimized for 8 Hz are from proton(s) stated to the indicated carbon.
MS (Supporting Information S4), suggested that our strain was
more closely related to O. hirsuta than O. comitata.3
eluted with 100% EtOAc by preparative HPLC gave two new
compounds, lyngbyabellins O (3) and P (4), as well as three
known compounds, lyngbyabellins F (5) and G (6) and
dolastatin 16 (7), whose identities were confirmed by
comparison with the chemical shifts reported in the
literature11−13 (Figures S26−S29).
The MeOH extract of this sample was partitioned between
EtOAc and H2O, and the aqueous fraction was further
partitioned with n-BuOH. The EtOAc fraction was separated
into nine subfractions by column chromatography over normal-
phase (NP) silica gel. The fractionation process was monitored
by LC-ESIMS analysis. The major compound in the EtOAc
extract was determined to be the new compound serinolamide
C (1), which was eluted from the column using a 2:8 (v/v)
mixture of hexane/EtOAc. ESIMS analysis of this fraction
showed an [M + H]+ ion with an m/z value of 356.3165.
Further purification of the fraction that eluted with a 6:4 (v/v)
mixture of hexane/EtOAc by preparative HPLC led to the
isolation of additional 1, together with the related compound
serinolamide D (2). Further purification of the fraction that
The planar structure of 1 was elucidated using NMR and MS
analyses. The molecular formula of 1 was determined to be
C21H41NO3 by ESITOFMS, indicating two degrees of
1
unsaturation. H and 13C NMR analyses (Table 1) revealed
the characteristic signals of an unsaturated acyl chain composed
of eight methylene groups (δH 1.26, δC 29.3−29.6 ppm) and a
terminal methyl group (δH 0.88, δC 14.0), as well as signals
consistent with one O-methyl (δH 3.38, δC 59.3), one carbonyl
(δC 169.9), one trisubstituted olefinic methyl (δH 1.86, δC
B
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. NMR Spectroscopic Data (1H 600 MHz, 13C 150 MHz, CDCl3) for Lyngbyabellins O (3) and P (4)
lyngbyabellin O (3)
δH (J in Hz)
lyngbyabellin P (4)
δH (J in Hz)
a
a
position
δC, type
COSY
HMBC
δC, type
COSY
HMBC
1
171.0, C
43.8, CH
75.2, CH
31.6, CH2
21.6, CH2
49.1, CH2
90.1, C
171.0, C
43.7, CH
75.0, CH
31.4, CH2
21.6, CH2
49.1, CH2
90.1, C
2
3.00, dq (4.2, 6.6)
5.43, dt (4.8, 12.0)
1.81, m
3, 9
2, 4
3, 5
4, 6
5
1, 9
5
3.03, dq (4.8, 7.2)
5.43, dt (7.8, 4.2)
1.82, m
9
4
3
6
5
1
3
4
5
1.81, m
1.79, m
6
2.18, 2.25, m
5
2.18, 2.26, m
7
8
37.4, CH3
13.8, CH3
166.4, C
146.2, C
129.0, CH
179.7, C
71.7, CH
66.0, CH2
2.12, s
6, 7
37.4, CH3
13.6, CH3
160.9, C
146.4, C
129.0, CH
173.0, C
70.0, CH
67.8, CH2
2.12, s
6, 7
1, 3
9
1.28, d (6.6)
1, 2, 3
1.29, d (6.6)
10
11
12
13
14
15
8.24, s
11, 13
8.22, s
11, 13
24
5.09, br
15
14
5.29, dd (3.6, 7.8)
4.55, dd (11.4, 3.6)
4.59, dd (11.4, 7.8)
15
14
4.02, brd (16.8)
16
17
18
19
20
21
22
23
24
25
26
27
27-NH
28
29
30
31
32
33
34
35
36
161.8, C
146.1, C
128.7, CH
167.0, C
78.3, CH
72.2, C
161.7, C
146.2, C
128.5, CH
167.4, C
78.4, CH
72.0, C
8.17, s
6.16, s
20
18
17, 19
1, 19
8.17, s
6.13, s
17, 19
1, 19, 22, 23
25.3, CH3
26.9, CH3
52.6, CH3
1.18, s
1.41, s
3.95, s
23
22
20, 23
26.6, CH3
25.4, CH3
171.9, C
37.5, CH2
71.8, CH
51.6, CH
1.38, s
1.19, s
20, 21, 23
20, 21, 11
20, 21, 22
16
2.56, d (5.4)
3.90, ddd (6.0, 6.0, 6.0)
4.07, m
26
24, 26, 27
25
27-NH, 28
5.52, d (8.4)
1.34, 1.50, m
1.63, m
27
39.0, CH2
25.0, CH
23.5, CH3
21.5, CH3
174.3, C
27
30, 31
29
0.95, d (6.0)
0.90, d (6.0)
28, 29, 31
28, 29, 30
29
38.6, CH2
19.2, CH2
13.7, CH3
52.5, CH3
2.20, m
34
32
32
34
16
1.67, m
33, 35
34
0.97, t (7.8)
3.95, s
a
HMBC optimized for 8 Hz are from proton(s) stated to the indicated carbon.
12.6), and one amide proton (δH 6.41), forming a fatty acid
amide with an O-methylserinol.
retention time of the hydrolysate of 1 matched that of the
synthetic (S)-O-methylserinol, leading to the assignment of C-
18 in compound 1 as the S-configuration.
Fatty acid amides are common metabolites of the genera
Moorea,14,15 Lyngbya,16 and Okeania.17 Moreover, two fatty acid
amides bearing an O-methylserinol moiety have previously
been isolated from Moorea producens,14,15 namely, serinola-
mides A and B. HMBC analysis of 1 allowed for the assignment
of the spin system formed by H-17, C-1, as well as the double
bond formed by C-2 and C-3. The configuration of the double
bond of 1 was determined based on the characteristic chemical
shift of the C-17 methyl group. As the chemical shifts of α-
carbons are dependent on the cis−trans isomerism of the
double bond, the shift of C-17 (δC 12.6 ppm) indicated an E-
configuration. Indeed, in cis alkenes, the α-carbons are shielded
from their positions in the alkane, whereas in trans alkenes, they
are deshielded.18 The absolute configuration of the stereogenic
center of the serinol moiety was determined by the synthesis of
the (S)- and (R)-enantiomers of O-methylserinol, followed by
the hydrolysis of 1 in conjunction with Marfey’s analysis.19 The
The molecular formula of serinolamide D (2) was
determined to be C23H43NO4 by ESITOFMS analysis,
1
indicating three degrees of unsaturation. The H NMR data
for 2 were quite similar to those obtained for 1 (Table 1),
except for the presence of an additional methyl singlet (δH
2.07), as well as a shielded shift for the amide proton (δH 6.11
instead of 6.41 for 1) and deshielded shifts for H-19 (δH 4.14
and 4.28 instead of 3.71 and 3.86 for 1) and H-18 (δH 4.40
instead of 4.13 for 1). Moreover, the 13C NMR spectrum of this
compound revealed the presence of an extra carbonyl group
(δC 171.3), as well as an extra methyl group (δC 20.8)
compared with 1. HMBC analysis (Figure S14) showed that
the serinol was O-acetylated. The configuration of the double
bond was also determined to be E based on the chemical shift
of the C-17 methyl (δC 12.6). Marfey’s analysis of the
hydrolysate of 2 showed a peak with a retention consistent
C
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. 2D NMR correlations for the partial structures of 3 (fragments A−E) and 4 (fragments A′−E′).
with synthetic (S)-O-methylserinol. In light of the O-acetylated
serinol in 2, the CIP rules led to the C-18 position being
assigned as the R-configuration.
Unfortunately, we did not observe any correlations
1
connecting fragments A−E. However, based on H and 13C
NMR data similar to lyngbyabellins and detailed consideration
of the many different combinations, the proposed structure
(Figure 1) was recognized as the best fit for the available data.
The chemical shift of C-3 (δC 75.2) showed that this carbon
should be attached to an oxygen atom. This proves that
fragments C and D are attached. If C-17 of fragment D was
attached to fragment A, it would be more deshielded, but
attached to fragment E, its chemical shift corresponds to the
one observed in the spectrum (Table 2). As the chemical shift
of C-11 is almost identical to that of C-17, we can consider that
it is also attached to a carboxyl, therefore to fragment C. The
remaining fragment A is attached to fragment B at C-13.
Similar observations were recorded for 4, except for the
presence of an extra component corresponding to C12H22NO3,
which was determined to be attached to C-15 via an ester
linkage based on an HMBC correlation from H-15 to C-24
(fragment A′) (Figure 1). HMBC analysis also showed that C-
24 was attached to a methylene moiety (C-25), which was
linked to an oxygenated methine (δC 71.8). A correlation was
also observed between H-25 and a methine group at C-27 (δC
51.6) by HMBC analysis, whereas ROESY analysis revealed a
correlation (Figure 1) to a methylene (C-28). A signal
corresponding to an amide proton at δH 5.52 (27-NH) showed
a COSY correlation to H-27 (Figure 1), which was expanded
into a modified leucine unit (C-24 to C-31) based on 2D NMR
correlations, in a similar manner to lyngbyabellins E, F, H, and
I.13 The 27-NH signal also showed a ROESY correlation to H-
33 of an aliphatic chain (C-33 to C35), which was bonded to a
carbonyl (C-32). Once again, no correlations were observed
linking C-13 to C-14 (fragments A′ and B′), C-10 to C-
11(fragments B′ and C′), C-3 to C-10 (fragments C′ and D′)
or C-17 to C-16 (fragments D′ and E′). The linkages between
the different fragments were therefore established using a
similar logic to that applied to 3.
NMR and MS data analyses were used to determine the
planar structures of 3 and 4. ESITOFMS analysis revealed that
the molecular formulas of 3 and 4 were C24H32Cl2N2O4S2 and
C36H52Cl2N3O12S2 with nine and 11 degrees of unsaturation,
respectively. Both of these compounds contained two chlorine
atoms, as exemplified by the 9:6:1 isotope peaks observed in
the mass spectra of 3 and 4 with m/z values of 627/629/631
and 854/856/858, respectively. The NMR data for 3 and 4 are
summarized in Table 2.
The 1H NMR spectrum of 3 (Table 2) showed two
deshielded signals (δH 8.17 and 8.24), whereas the 13C NMR
spectrum contained signals consistent with eight sp2 carbons
(C-10 to C-13 and C-16 to C-19). Taken together, these data
suggested the presence of two 2-alkylthiazole-4-carboxylates,
which are typical of lyngbyabellins and dolabellin.13 HMBC
analysis (Figure 1) also confirmed the presence of two thiazole
rings with correlations from H-12 to C-11 and C-13 and H-18
to C-17 and C-19. The remaining sp2 carbon (C-1 δC 171.0)
was assigned to the carbonyl group of a saturated ester.
Between them, the two thiazole rings and the three carbonyl
groups accounted for the nine degrees of unsaturation in 3.
The 1H NMR spectrum of 3 also showed a series of shielded,
coupled resonance signals indicative of an aliphatic chain. A
singlet corresponding to a methyl group at δH 2.12 was also
detected, which showed a correlation to a methylene at δC 49.1
(C-6) and a dichlorinated carbon at δC 90.1 (C-7) by HMBC
analysis, indicating an unusual gem-dichloro group. This group
has been reported previously for lyngbyabellins A−K and M−
N, as well as dolabellin and the hectochlorins.13 HMBC and
COSY analysis (Figure 1) indicated that this gem-dichloro
group was attached to the aliphatic chain, with the C-9 methyl
group attached to the C-2 methine.
Furthermore, HMBC analysis (Figure 1) revealed correla-
tions between H-20 and C-1, as well as correlations with C-19
of one of the thiazole ring systems and two methyl groups (C-
22 and C-23), both of which were attached to an oxygenated
nonprotonated carbon (C-21) (forming fragment D). COSY
and HMBC data (Figure 1) also revealed two partial structures:
−CH(OH)CH2OH (fragment A) and a methoxy attached to a
carbonyl group (fragment E).
The planar structure of 3 resembled that of lyngbyabellin G
(6), which was also isolated from this Okeania sp. extract. The
key difference between these two molecules is the open
structure between C-15 and the ester attached to the second
thiazole ring system. Likewise, the planar structure of
compound 4 bears a striking resemblance to that of
lyngbyabellin F (5), isolated from the same fraction, with the
only major difference being an OH group at C-26 instead of an
D
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
acetate group. Given that compounds 3 and 4 were both found
in the same fraction as 5 and 6 and share similar structural
features, it is highly likely that they also have the same
configuration.
intact even after 120 h. The observed antifouling activity of the
Okeania sp. extract was therefore mainly attributed to
compounds 3, 4, and 7 because the 100% EtOAc fraction
from which they were isolated was the most active of all of the
fractions tested.
1
Identical H and 13C NMR data for 6 compared to the
literature (the largest difference in 13C chemical shifts is 0.028
The cytotoxicities of the isolated compounds were evaluated
in MCF7 breast cancer cells. It is noteworthy that the two fatty
acid amides 1 and 2 precipitated out of the medium, precluding
their activities from being assessed. The GI50 values of
compounds 3, 4, and 6 toward the MCF7 cells were >160, 9,
and 120 μM, respectively. These differences in the cytotoxicity
could most likely be attributed to the differences in their
structures.13 The side chain in compound 4 could play an
important role in its superior cytotoxicity (GI50 9 μM) over 3
and 6, forming additional interactions with a common target,
providing a different mode of action, or simply improving its
permeability. Compound 6 appeared less active on MCF7 cells
than on H460 and Neuro-2a cells (2.2 and 4.8 μM,
respectively).13 Its mechanism of action in MCF7 cells is
probably different than in H460 and Neuro-2a cells or MCF7
cells, which are less sensitive to 6.
The results of a previous study showed that 7, isolated from
the sea hare Dolabella auricularia, inhibited the growth of
several human cancer cell lines (GI50 0.0012−0.00096 μg/
mL).22 Surprisingly, a total synthesis of this cyclodepsipeptide
revealed that synthetic 7 did not exhibit any discernible
cytotoxicity.23 The authors attributed this lack of activity to a
change in the conformation of the synthetic compound or the
presence of a chemically undetected (yet extremely active)
compound in the sample isolated from the natural source. A
new synthesis of 7 confirmed its inactivity.24 In this study, it
was confirmed that 7 is noncytotoxic but potently active toward
barnacle larvae.
ppm), together with similar specific rotation ([α]26 −28.6 for
D
6 and −26 reported for lyngbyabellin G), proved that 6 is the
known lyngbyabellin G. Mild methanolysis converted 6 into 3,
so 3 has the same absolute configuration as 6. Chiral-phase
chromatography experiments supported this information. After
ozonolysis and subsequent base hydrolysis, the obtained α,β-
dihydroxyisovaleric acid (Dhiv) residue was methylated
(CH2N2). MTPA derivatization with (R)-MTPA was con-
ducted on both synthetic (R)- and (S)-Dhiv methyl esters and
the ozonolyzed/hydrolyzed/methylated natural product 3.
Chiral-phase HPLC with comparison of the retention times
of the diastereomeric synthesized (R)- and (S)-Dhiv methyl
esters (14.6 and 16.5 min, respectively) established the absolute
configuration of the α,β-dihydroxyisovaleric acid methyl ester
(16.6 min) of 3 as the S-configuration at C-20. The
ozonolyzed/hydrolyzed natural product 3 was confirmed to
contain D-glyceric acid (3.8 min) by comparison of the
retention times by ESI-LC-MS using a chiral-phase column of
the commercially available L- and D-glyceric acids (3.5 and 3.8
min, respectively), giving an R-configuration at C-14. Identical
to 6, 3 has a 2S, 3S configuration.
Chemical shifts and the specific rotation of 5 ([α]26 −7.8
D
for 5 and −6.5 reported for lyngbyabellin F; the largest
difference in 13C chemical shifts is 0.015 ppm) were similar to
those reported for lyngbyabellin F. Deacetylation (NaOCH3) at
C-26 of 5 led to the conversion into 4. Degradation of 4 and 5
observed by ESI-TOF-MS analysis showed that both were
converted into 3, showing that these three compounds indeed
share the same configuration. As a result, 4 has a 2S, 3S, 14R,
20S, 26R, and 27S configuration.
Okeania spp. appear to be a promising source of biologically
active compounds based on the results obtained for the
compounds isolated from the genus Lyngbya or Moorea. The
cytotoxicity of the Okeania sp. extract in particular was
attributed to the presence of several lyngbyabellins, which are
well-known cytotoxic compounds. The antifouling activity of
this extract was mainly attributed to the known dolastatin 16
(7) and the new lyngbyabellins O (3) and P (4).
As observed for lyngbyabellin E that converted to
lyngbyabellin G,13 one ester linkage of 4 is apparently prone
to methanolysis and a regioselective ester cleavage occurred at
C-14, giving 3, which can be considered as an artifact of 4 and
also 6, as mild methanolysis gave 3. Conversion of 6 into 3 is
possible by regioselective ester cleavage at C-16. This kind of
conversion was observed for lyngbyabellin C, which was
converted into homohydroxydolabellin.20
The isolated compounds, except 5, were subjected to an
antifouling assay using Amphibalanus amphitrite larvae. After an
exposure time of 48 h, compounds 1 and 2 inhibited the
settlement of the larvae at concentrations below 1 μg/mL.
However, after an exposure time of 120 h at similar
concentrations, barnacle larvae were able to settle. Although
the two fatty acid amides 1 and 2 slowed the settlement process
considerably, they were unable to inhibit it completely. A
recovery test for 2 showed that after 7 days almost all of the
exposed larvae settled regardless of the tested concentration.
Compound 6 showed low activity (EC50 7.4 μM), while 3 and
4 exhibited potent activity, with EC50 values of 0.38 and 0.73
μM, respectively. These results suggest that the larvae are more
sensitive to the acyclic structure of 3 than the cyclic structure of
6. The flexibility of the acyclic structure may facilitate
interactions with a common target, but addition of a side
chain decreases them, as observed with 4. Compound 7, which
is well known as a powerful antifoulant,21 exhibited high
inhibition activity (EC50 0.09 μM), and its activity remained
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotation data were
acquired on a HORIBA SEPA-300 polarimeter using a 0.1 dm cuvette
with chloroform or methanol as the solvent. UV spectra were recorded
on a JASCO MD-1510 multiwavelength UV detector. IR spectra were
recorded on a JASCO FT/IR-4100 type A. The H and 13C NMR
1
spectra for compounds 1, 2, and 6 were recorded on a JEOL 400 MHz
spectrometer, whereas the 2D NMR spectra of 1 and 2 were recorded
on a Bruker Avance 300 MHz. The 1H, 13C, and 2D NMR spectra for
compounds 3 and 4 were recorded on a JEOL 600 MHz spectrometer.
1
The H NMR spectrum of 5 was recorded on a JEOL 400 MHz
spectrometer, whereas its 13C NMR spectrum was recorded on a
Bruker AMX500. All of the NMR spectra were recorded in CDCl3
(Cambridge Isotope Laboratories, Inc.) using the residual solvent peak
(CHCl3) as an internal reference (δH 7.27 and δC 77.00 ppm). LC-MS
data were obtained on an Agilent 1100 Series HPLC system coupled
with a Bruker Daltonics micrOTOF-HS mass spectrometer, which was
configured for electrospray ionization (ESI). The HPLC system was
equipped with a Cadenza CD-C18 column (2 × 150 mm, 3 μm, 0.2
mL/min, 25 °C). The HPLC system was operated under the following
conditions: 0−20 min, linear gradient from 50% to 100% MeCN with
0.1% (v/v) formic acid in Milli-Q H2O; 20−35 min, isocratic 100%
MeCN with 0.1% (v/v) formic acid. Preparative HPLC purifications
E
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
were conducted over a Cosmosil Cholester column (10 × 250 mm, 5
μm) at a flow rate of 3 mL/min, with UV detection at 210 nm.
Biological Material. The dark brown cyanobacterial sample S1501
(Figure S1) was collected by scuba diving at a depth of 10−15 m on
the Algetah Alkabira reef near Jeddah, Saudi Arabia (N 21°41′23.98″;
E 39°00′52.94″) in April 2015. Foreign particles were removed by
hand, and the samples were squeezed by hand to remove any seawater
before being stored in MeOH for transportation. A small portion of
this material was also preserved in 10 mL of RNAlater solution for
genetic analysis. The sample was identified as an Okeania sp. by
phylogenetic analysis (Figure S3) of its 16S rRNA gene sequence
(GenBank KY889150). The morphological characteristics of this
material (hair-like, dark brown color, cell width around 55 μm), as well
as the detection of malyngamide C acetate by LC-MS analysis (Figure
S4), suggested that S1501 was more closely related to O. hirsuta than
O. comitata.3 A voucher specimen (S1501) preserved in RNAlater was
deposited at Hokkaido University, Japan.
Serinolamide D (2): colorless, amorphous solid, [α]26 +14.3 (c
D
0.11, CHCl3); UV (MeOH) λmax 215 nm (log ε 4.01); IR (neat) λmax
1
2966, 2924, 2852, 1741, 1625, 1531, 1463, 1260 cm−1; H and 13C
NMR data, Table 1; ESITOFMS m/z 398.3255 [M + H]+ (calcd for
C23H44NO4, 398.3265).
Lyngbyabellin O (3): colorless, amorphous solid; [α]26 −9.5 (c
D
1
0.23, MeOH); UV (MeOH) λmax 243 nm (log ε 4.76); H and 13C
NMR data, Table 2; ESITOFMS m/z 627.1024 [M + H]+ (calcd for
C24H33Cl2N2O9S2, 627.0999).
Lyngbyabellin P (4): colorless, amorphous solid; 1H and 13C NMR
data, Table 2; ESITOFMS m/z 854.2544 [M + H]+ (calcd for
C36H54Cl2N3O12S2, 854.2520).
Absolute Configurations of 1 and 2. The absolute config-
urations of 1 and 2 were determined by the synthesis of (S)- and (R)-
O-methylserinol according to the method reported by Gao et al.19
A
small aliquot of 1 (0.1 mg) was hydrolyzed using 6 M HCl at 110 °C
for 16 h. Compound 2 was hydrolyzed in a similar manner using 12 M
HCl at 120 °C to convert its acetate group to an alcohol. The
hydrolysates were dried in vacuo and subjected to Marfey’s analysis,
along with the two synthesized O-methylserinol enantiomers. The
hydrolysates and derivatives were analyzed by RP-HPLC (Cosmosil
5C18-AR-II 4.6 × 250 mm, 5 μm, 1 mL/min, UV detection at 340 nm,
gradient 0−45 min, 10−50% MeCN containing 0.05% (v/v) TFA). A
comparison of the retention times (tR) of the standard (21 and 31 min
for (R)- and (S)-O-methylserinol, respectively) and the hydrolysates
(31 min for 1 and 2) led to the assignment of the stereogenic centers
in compounds 1 and 2 as S and R, respectively.
Absolute Configurations of 3 at C-14 and C-20. The absolute
configurations of 3 at C-14 and C-20 were determined by chiral-phase
chromatography techniques. A small portion of 3 (0.6 mg) was
dissolved in MeOH (2 mL) and ozonolyzed at −78 °C for 25 min.
The solvent was removed under a stream of Ar, and the sample was
hydrolyzed with 2 N NaOH at 90 °C for 7 h. To determine the
relative configuration at C-14, the hydrolysate was analyzed by chiral-
phase ESI-LC-MS (column Astec Chirobiotic TAG 2.1 × 250 mm; 0.2
mL/min; ESI negative low mode; 1:1 0.1% aqueous formic acid/1%
NH4OAc in MeOH; column oven: 20 °C), and it was found to contain
D-glyceric acid (3.8 min) by comparison of the retention times of the
commercially available L- and D-glyceric acids (3.5 and 3.8 min,
respectively). The hydrolysate was then methylated (2 M CH2N2 for
10 min at 0 °C) to produce α,β-dihydroxyisovaleric (Dhiv) methyl
esters. The standards R- and S-Dhiv methyl esters were synthesized by
Sharpless asymmetric dihydroxylation of methyl 3,3-dimethyl acrylate
using AD-mix-α and AD-mix-β, respectively.25 The product obtained
from 3 and the standards R- and S-Dhiv methyl esters were derivatized
using (R)-MTPA. The samples were dissolved in CH2Cl2, mixed with
(R)-MTPA (50 μL of a solution of 50 μL/mL of MeCN) and heated
for 1 h at 70 °C. The reaction was quenched by addition of EtOH
(100 μL) and heating for 15 min at 70 °C. After evaporation of the
solvent, the samples were dissolved in MeOH and subjected to HPLC
analysis (column Cosmosil 5C18-MS-II 4.6 × 250 mm, 5 μm, 1 mL/
min, UV detection 254 nm, gradient 0−30 min, 20−30% MeOH). The
retention time of the MTPA derivative from the hydrolysate matched
that of the S-Dhiv methyl ester standard (16.5 min, instead of 14.6 min
for the R-Dhiv methyl ester standard).
Gene Sequencing. The details of these procedures are reported in
the Supporting Information.
Extraction and Isolation. A sample of the cyanobacterium was
homogenized and extracted three times with MeOH. The dried
MeOH extract (3.5 g) was partitioned between EtOAc and H2O (1:1,
v/v). The dried EtOAc fraction (2.8 g) was purified by NP-column
chromatography over silica gel. A stepwise gradient composed of
hexane and EtOAc (100:0, 9:1, 8:2, 6:4, 4:6, 2:8 v/v), followed by
EtOAc and MeOH (100:0, 75:25, 0:100 v/v), was used to elute the
column. LC-MS analysis revealed that the 2:8 (v/v) hexane/EtOAc
fraction (118 mg) contained the main compound of the mother
fraction, with an m/z value of 356.31. Isolation was performed using
semipreparative RP-HPLC (gradient 0−30 min, 40−80% MeCN; 30−
40 min, 80−100% MeCN). The fraction eluted at 40 min was further
purified by RP-HPLC (gradient 0−40 min, 90−100% MeCN) to yield
serinolamide C (1) (tR 15.7 min, 12 mg). The fraction eluted with 6:4
(v/v) hexane/EtOAc (245 mg) was fractionated by RP-HPLC
[gradient 0−50 min, 75−90% MeCN containing 0.05% (v/v)
trifluoroacetic acid (TFA)]. The fraction eluted at 24.3 min was
further purified by RP-HPLC (isocratic 90% MeCN containing 0.05%
(v/v) TFA) to yield compound 1 (tR 14.6 min, 10 mg) and another
fraction (tR 18.4 min, 10 mg). This fraction was further purified by RP-
HPLC (Cosmosil MS-5C18-II 10 × 250 mm, 5 μm, 3 mL/min, UV
detection at 210 nm, isocratic 87% MeCN) to give 1 (tR 21.3 min, 1.7
mg) and serinolamide D (2) (tR 30.3 min, 1.6 mg). The fraction eluted
with 100% EtOAc (140 mg) was fractionated by RP-HPLC (gradient
0−30 min, 40−80% MeCN containing 0.05% (v/v) TFA) to yield
fraction A (tR 14.1 min, 3.4 mg), fraction B (tR 16.4 min, 6.7 mg),
fraction C (tR 20.4 min, 4.4 mg), fraction D (tR 21.8 min, 2.4 mg), and
fraction E (tR 22.1−25 min, 3.7 mg). Fraction A was further purified by
RP-HPLC (Cosmosil Cholester 4.6 × 250 mm, 5 μm, 1 mL/min, UV
detection at 210 nm, gradient 0−40 min, 40−70% MeCN containing
0.05% (v/v) TFA) to yield lyngbyabellin O (3) (tR 9.6 min, 1.8 mg).
Fraction B was further purified by RP-HPLC (gradient 0−40 min, 40−
70% MeCN) to yield the known compound lynbyabellin G (6) (tR
20.1 min, 3 mg). Fraction C was further purified by RP-HPLC (UV
detection at 210 nm, gradient 0−40 min, 40−70% MeCN) to yield
lyngbyabellin P (4) (tR 27.5 min, 2 mg). Fraction D was further
purified by RP-HPLC (Cosmosil Cholester 10 × 250 mm, 5 μm, 3
mL/min, gradient 0−60 min, 50−70% MeCN) to give the known
compound lyngbyabellin F (5) (tR 20.7 min, 1.1 mg). Fraction E was
further purified by RP-HPLC (Cosmosil Cholester 4.6 × 250 mm, 5
μm, 1 mL/min, gradient 0−40 min, 60−80% MeCN containing 0.05%
(v/v) TFA) to yield a subfraction (tR 10.5 min, 7 mg), which was
further purified by RP-HPLC (Cosmosil Cholester 4.6 × 250 mm, 5
μm, 1 mL/min, UV detection at 254 nm, gradient 0−40 min, 50−70%
MeCN) to give the known compound dolastatin 16 (7) (tR 22.5 min,
1.5 mg).
Mild Methanolysis of 6. A portion of 6 (0.4 mg) was dissolved in
MeOH (100 μL), and 0.1 M HCl in MeOH was added (900 μL). The
mixture was heated at 85 °C for 45 min. The reaction was quenched
by evaporation of the solvent. The product was then dissolved in
MeOH for ESI-LC-MS analysis. The results showed that 6 was
converted into 3.
Deacetylation of 5. Compound 5 (1 μmol) was dissolved in
MeOH (100 μL), and 0.1 M NaOCH3 in MeOH was added (1 μL).
After 15 min at room temperature, the reaction was quenched by
evaporation of the solvent. The product was then dissolved in EtOH
for ESI-LC-MS analysis. The results showed that 5 was converted into
4.
Serinolamide C (1): white, amorphous solid, [α]26 +4.9 (c 0.51,
D
CHCl3); UV (MeOH) λmax 219 nm (log ε 3.42); IR (neat) λmax 3305
1
(br), 2957, 2917, 2848, 1610, 1552, 1469, 1260 cm−1; H and 13C
Antifouling Assay. An antifouling assay was performed on
Amphibalanus amphitrite cypris according to the method described
by Kitano et al.26 Adult barnacles were procured from oyster farms in
NMR data, Table 1; ESITOFMS m/z 356.3165 [M + H]+ (calcd for
C21H42NO3, 356.3159).
F
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Lake Hamana, Japan, as well as a pier at Shimizu Bay, Shizuoka, Japan.
All of these barnacles were kept in an aquarium at 20 °C, where they
were fed brine shrimp Artemia salina nauplii. The barnacles were dried
at room temperature overnight before being immersed in seawater,
which lead to the release of I−II nauplii from the broods. These
nauplii were cultured at 25 °C in 80% filtered (0.2 μm of pore filter)
natural seawater (diluted with deionized water) containing 30 μg/L
streptomycin and 20 mg/L penicillin G, where they were fed on the
diatom Chaetoceros gracilis until they reached the cyprid stage within 5
days. The day that the newly transformed cyprids were collected was
designated as day 0. The test samples and compounds were dissolved
in EtOH. The antifouling assays were conducted in flat-bottomed 24-
well polystyrene culture plates, and four wells containing six healthy
cyprids were used for each concentration. After an exposure time of 48
h in the dark at 25 °C, the numbers of larvae that had attached or
metamorphosed or did not settle were counted under a microscope.
After an exposure time of 120 h, the number of dead cyprids was
counted in the same way. Concerning the recovery test, floating larvae
after an exposure time of 120 h were collected and put in fresh marine
water without compound, and floating and settled larvae were counted
daily for 7 days. For the initial screening of the extracts, the samples
were tested at two different concentrations, namely, 10 and 100 μg/
mL. Pure compounds were tested at six different concentrations in
triplicate (n = 3), including 0.03, 0.1, 0.3, 1.0, 3.0, and 10 μg/mL. The
antifouling activity of each compound was expressed as an EC50 value,
which represents the concentration of compound required to reduce
larval settlement to 50% of the control. Probit analysis was used to
calculate the EC50 values.
Innovation Plan−the King Abdulaziz City for Science and
Technology for the Kingdom of Saudi Arabia (11-ENV1540-
03)
REFERENCES
■
(1) Nunnery, J. K.; Mevers, E.; Gerwick, W. H. Curr. Opin. Biotechnol.
2010, 21, 787−793.
(2) Engene, N.; Paul, V. J.; Byrum, T.; Gerwick, W. H.; Thor, A.;
Ellisman, M. H. J. Phycol. 2013, 49, 1095−1106.
(3) Winnikoff, J. R.; Glukhov, E.; Dorrestein, P. C.; Gerwick, W. H. J.
Antibiot. 2014, 67, 105−112.
(4) Shao, C.-L.; Linington, R. G.; Balunas, M. J.; Centeno, A.;
Boudreau, P.; Zhang, C.; Engene, N.; Spadafora, C.; Mutka, T. S.;
Kyle, D. E.; Gerwick, L.; Wang, C.; Gerwick, W. H. J. Org. Chem. 2015,
80, 7849−7855.
(5) Okamoto, S.; Iwasaki, A.; Ohno, O.; Suenaga, K. J. Nat. Prod.
2015, 78, 2719−2725.
(6) Navarro, G.; Cummings, S.; Lee, J.; Moss, N.; Glukhov, E.;
Valeriote, F. A.; Gerwick, L.; Gerwick, W. H. Environ. Sci. Technol. Lett.
2015, 2, 166−170.
(7) Ogawa, H.; Iwasaki, A.; Sumimoto, S.; Kanamori, Y.; Ohno, O.;
Iwatsuki, M.; Ishiyama, A.; Hokari, R.; Otoguro, K.; Omura, S.;
Suenaga, K. J. Nat. Prod. 2016, 79, 1862−1866.
(8) Sueyoshi, K.; Kaneda, M.; Suminoto, S.; Oishi, S.; Fujii, N.;
Suenaga, K.; Teruya, T. Tetrahedron 2016, 72, 5472−5478.
(9) Kanamori, Y.; Iwasaki, A.; Sumimoto, S.; Suenaga, K. Tetrahedron
Lett. 2016, 57, 4213−4216.
Cytotoxicity Assay. The cytotoxicities of the compounds were
determined according to the procedure reported by Lopez et al.
(2016).27
(10) Al-Lihaibi, S. S.; Abdel-Lateff, A.; Alarif, W. M.; Nogata, Y.;
Ayyad, S-E. N.; Okino, T. Asian J. Chem. 2015, 27, 2252−2256.
(11) Engene, N.; Choi, H.; Esquenazi, E.; Rottacker, E. C.; Ellisman,
M. H.; Dorrestein, P. C.; Gerwick, W. H. Environ. Microbiol. 2011, 13,
1601−1610.
ASSOCIATED CONTENT
* Supporting Information
■
S
(12) Nogle, L. M.; Gerwick, W. H. J. Nat. Prod. 2002, 65, 21−24.
(13) Han, B.; McPhail, K. L.; Gross, H.; Goeger, D. E.; Mooberry, S.
L.; Gerwick, W. H. Tetrahedron 2005, 61, 11723−11729.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.7b00449.
(14) Engene, N.; Rottacker, E. C.; Kast
Ellisman, M. H.; Komar
Microbiol. 2012, 62, 1171−1178.
(15) Gutierrez, M.; Pereira, A. R.; Debonsi, H. M.; Ligresti, A.; Di
̌
́
ovsky, J.; Byrum, T.; Choi, H.;
ek, J.; Gerwick, W. H. Int. J. Syst. Evol.
́
Figures S1−S29; procedure of 16S rRNA gene
sequencing and phylogenetic analysis experiments; 1D
and 2D NMR spectra for serinolamides C (1) and D (2)
and lyngbyabellins O (3) and (4); 1D NMR for
lyngbyabellins F (5) and G (6) (PDF)
́
Marzo, V.; Gerwick, W. H. J. Nat. Prod. 2011, 74, 2313−2317.
(16) Montaser, R.; Paul, V. J.; Luesch, H. ChemBioChem 2012, 13,
2676−2681.
(17) Wannikoff, J. R.; Glukhov, E.; Watrous, J.; Dorrestein, P. C.;
Gerwick, W. H. J. Antibiot. 2014, 67, 105−112.
AUTHOR INFORMATION
Corresponding Author
*Tel: +81 11 706 4519. Fax: +81 11 706 4867. E-mail: okino@
ees.hokudai.ac.jp.
■
(18) Dorman, D. E.; Jautelat, M.; Roberts, J. D. J. Org. Chem. 1971,
36, 2757−2766.
(19) Gao, Y.-R.; Guo, S.-H.; Zhang, Z.-X.; Mao, S.; Zhang, Y.-L.;
Wang, Y.-Q. Tetrahedron Lett. 2013, 54, 6511−6513.
(20) Luesch, H.; Yoshida, W. Y.; Moore, R. E.; Paul, W. J.
Tetrahedron 2002, 58, 7959−7966.
ORCID
Julie G. Petitbois: 0000-0002-5602-0591
Julius Adam V. Lopez: 0000-0002-1812-8104
Taiki Umezawa: 0000-0003-4280-6574
(21) Tan, L. T.; Goh, B. P. L.; Tripathi, A.; Teo, S. L. M. Biofouling
2010, 26, 685−695.
(22) Pettit, G. R.; Xu, J.-P.; Hogan, F.; Williams, M. D.; Doubek, D.
L.; Schmidt, J. M.; Cerny, R. L.; Boyd, M. R. J. Nat. Prod. 1997, 60,
752−754.
Notes
The authors declare no competing financial interest.
(23) Pettit, G. R.; Smith, T. H.; Arce, P. M.; Flahive, E. J.; Anderson,
C. R.; Chapuis, J.-C.; Xu, J.-P.; Groy, T. L.; Belcher, P. E.; Macdonald,
C. B. J. Nat. Prod. 2015, 78, 476−485.
ACKNOWLEDGMENTS
■
We are grateful to Dr. M. Morikawa and Dr. M. Kurasaki of the
Faculty of Environmental Earth Science, Hokkaido University,
Japan, for allowing us to use their genetics and cell culture
facilities, respectively. We would like to thank Dr. Y. Kumaki of
the High-Resolution NMR laboratory and Dr. Y. Fukushi of the
GC-MS and NMR laboratory, Hokkaido University, Japan, for
conducting the NMR analyses described in this study. This
work was supported by JSPS KAKENHI Grant Number
16H04975 and the National Science, Technology and
(24) Casalme, L. O.; Yamauchi, A.; Sato, A.; Petitbois, J. G.; Nogata,
Y.; Yoshimura, E.; Okino, T.; Umezawa, T.; Matsuda, F. Org. Biomol.
Chem. 2017, 15, 1140−1150.
(25) Sharpless, K. B.; Amberg, W.; Bennami, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.;
Xu, D.; Zhang, X.-L. J. Org. Chem. 1992, 57, 2768−2771.
(26) Kitano, Y.; Ito, T.; Suzuki, T.; Nogata, Y.; Shinshima, K.;
Yoshimura, E.; Chiba, K.; Tada, M.; Sakaguchi, I. J. Chem. Soc., Perkin
Trans.1 2002, 2251−2255.
G
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(27) Lopez, J. A. V.; Al-Lihaibi, S. S.; Alarif, W. M.; Abdel-Lateff, A.;
Nogata, Y.; Washio, K.; Morikawa, M.; Okino, T. J. Nat. Prod. 2016,
79, 1213−1218.
H
DOI: 10.1021/acs.jnatprod.7b00449
J. Nat. Prod. XXXX, XXX, XXX−XXX