Oxasqualenoids from Laurencia viridis: Combined spectroscopic-computational analysis and antifouling potential

Article abstract of DOI:10.1021/np5008922

The chemical study of the red alga Laurencia viridis has led to the isolation of four new polyether triterpenoids: 28-hydroxysaiyacenol B (2), saiyacenol C (3), 15,16-epoxythyrsiferol A (4), and 15,16-epoxythyrsiferol B (5). The structures of 2 and 3 were established mainly by NMR data analysis and comparison with the well-known metabolite dehydrothyrsiferol (1). However, due to the existence of a nonprotonated carbon within the epoxide functionality, stereochemical assignments in 4 and 5 required an in-depth structural study that included NOESY data, J-based configuration analysis, comparison with synthetic models, and DFT calculations. The biological activities of the new metabolites and other related oxasqualenoids were evaluated for the first time against a panel of relevant biofouling marine organisms, and structure-activity conclusions were obtained.

Full text of DOI:10.1021/np5008922

Article
pubs.acs.org/jnp
Oxasqualenoids from Laurencia viridis: Combined Spectroscopic−
Computational Analysis and Antifouling Potential
†,§
Francisco Cen-Pacheco,†,‡ Adrian J. Santiago-Benítez,†,§ Celina García,†,§ Sergio J. Alvarez-Mendez,
́
́
́
Alberto J. Martín-Rodríguez,†,⊥ Manuel Norte,†,§ Victor S. Martín,†,§ Jose A. Gavín,†,§
́
Jose J. Fernandez,*,†,§ and Antonio Hernandez Daranas*,†,∥
́
́
́
†Institute of Bio-Organic Chemistry “Antonio Gonzal
́
ez”, Center for Biomedical Research of the Canary Islands (CIBICAN),
§Department of Organic Chemistry, and Department of Chemical Engineering and Pharmaceutical Technology, Faculty of Health
Sciences, University of La Laguna, Avenida Astrofísico Francisco Sanchez 2, 38206, Tenerife, Spain
ico
∥
́
‡Faculty of Bioanalysis, Campus-Veracruz, Universidad Veracruzana, 91700, Veracruz, Mex
́
⊥Oceanic Platform of the Canary Islands (PLOCAN), Carretera de Taliarte s/n, 35214, Telde, Gran Canaria, Spain
S
* Supporting Information
ABSTRACT: The chemical study of the red alga Laurencia viridis has
led to the isolation of four new polyether triterpenoids: 28-
hydroxysaiyacenol B (2), saiyacenol C (3), 15,16-epoxythyrsiferol A
(4), and 15,16-epoxythyrsiferol B (5). The structures of 2 and 3 were
established mainly by NMR data analysis and comparison with the well-
known metabolite dehydrothyrsiferol (1). However, due to the existence
of a nonprotonated carbon within the epoxide functionality, stereo-
chemical assignments in 4 and 5 required an in-depth structural study
that included NOESY data, J-based configuration analysis, comparison
with synthetic models, and DFT calculations. The biological activities of
the new metabolites and other related oxasqualenoids were evaluated for
the first time against a panel of relevant biofouling marine organisms,
and structure−activity conclusions were obtained.
arine polyethers constitute an important class of
bioactive compounds among marine natural products.1,2
extracted with CHCl3/MeOH (1:1) at room temperature,
yielding 83.0 g of a dark green, viscous oil after solvent
evaporation. This extract was first separated on a Sephadex LH-
20 column using MeOH as the mobile phase, and the enriched
polyether fraction was subsequently processed using Lobar
LiChroprep-RP18 and Lobar LiChroprep Si-60 columns. The
resulting fractions were further purified by HPLC on a μ-Porasil
column using n-hexane/EtOAc/MeOH, 18:15:5, and n-
hexane/acetone, 7:3, sequentially to afford four new polyether
compounds, 28-hydroxysaiyacenol B (2), saiyacenol C (3),
15,16-epoxythyrsiferol A (4), and 15,16-epoxythyrsiferol B (5).
The molecular formula of 28-hydroxysaiyacenol B (2),
C30H51BrO7, was deduced by HRESIMS, accounting for five
unsaturations. This was consistent with its 13C NMR data,
where seven methyls, 11 methylenes, six methines, and six
nonprotonated carbons bearing oxygen were detected; the
absence of sp2 carbons and carbonyl groups indicates that 2
should be a pentacyclic molecule. Examination of the observed
13C chemical shifts indicated that 2 belongs to the marine
polyether triterpene family of compounds typically found in
Laurencia.7 Among these metabolites, saiyacenols A (6) and B
(7) showed similar NMR data, although with chemical shift
M
Many have been widely used as research tools to unravel
complex biochemical pathways, and some of them have entered
clinical trials or are close to that stage, as a result of their
attractive pharmacological properties.3,4 The red alga Laurencia
viridis produces an amazing variety of these metabolites, with
dehydrothyrsiferol (1) as a key example, some of which have
shown bioactivity as Ser-Thr protein phosphatase 2A inhibitors
or integrin antagonists or cytotoxicity.5,6 This report describes
the investigation of L. viridis collected from the coastal rocks of
the Canary Islands, leading to the isolation of four new
metabolites: 28-hydroxysaiyacenol B (2), saiyacenol C (3), and
15,16-epoxythyrsiferols A and B (4 and 5). Their structures
were determined on the basis of detailed spectroscopic studies,
including a J-based configuration approach, comparison with
synthetic models, and DFT theoretical calculations. These
compounds were evaluated for their antifouling activity toward
a panel of organisms composed by marine bacteria, marine-
derived fungi, benthic diatoms, and macroalgal zoospores.
RESULTS AND DISCUSSION
■
Isolation, Structure Determination, and Synthesis of
Simplified Models. Specimens of L. viridis collected in spring
2013 along the coast of Tenerife (Canary Islands, Spain) were
Received: November 10, 2014
Published: March 17, 2015
© 2015 American Chemical Society and
American Society of Pharmacognosy
712
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
Table 1. NMR Spectroscopic Data (1H 600 MHz, 13C 150
MHz, CDCl3) for Compounds 2 and 3
28-hydroxysaiyacenol B (2)
δC, type δH (J in Hz)
saiyacenol C (3)
δC, type δH (J in Hz)
C
1
31.0, CH3 1.26, s
75.0, C
31.0, CH3 1.26, s
75.1, C
2
3
59.0, CH
3.90, dd (4.2,
12.4)
59.0, CH
3.88, dd (4.5, 12.1)
4
5
28.2, CH2 α 2.10, m
β 2.24, m
28.2, CH2 α 2.08, m
β 2.24, m
37.1, CH2 α 1.54, m
β 1.81, m
37.4, CH2 α 1.53, m
β 1.83, m
6
7
74.4, C
74.5, C
86.5, CH
3.06, dd (2.1,
11.5)
86.3, CH
2.98, dd (1.5,11.0)
8
9
23.0, CH2 β 1.44, m
α 1.74, m
23.3, CH2 β 1.40, m
α 1.75, m
38.6, CH2 α 1.56, m
β 1.74, m
39.8, CH2 α 1.52, m
β 1.8, m
10
11
72.0, C
70.2, C
76.2, CH
3.62, dd (7.0,
11.5)
84.4, CH
3.09, dd (2.8, 9.9)
12
13
14
21.1, CH2 β 1.50, m
α 1.92, m
32.0, CH2 β 2.01, m
α 2.32, m
20.8, CH2 α 1.70, m
β 1.92, m
124.7, CH
5.64, ddd (6.1,
7.8,16.0)
73.7, CH
4.10, dd (2.0,
13.0)
137.3, CH
82.8, C
5.52, dd (16.0)
15
16
85.1, C
27.8, CH2 1.74, m
1.96, m
37.7, CH2 1.66, m
17
18
27.6, CH2 1.59, m
1.88, m
27.3, CH2 1.78, m
1.85, m
differences around the C-15→C-18 region (Table 1).8 In
addition, the molecular formula of 2 indicated the existence of
an additional oxygen atom compared to 6 and 7.
85.0, CH
3.85, dd (5.6,
9.5)
84.6, CH
3.89, dd (4.5/12.0)
Analysis of COSY and HSQC spectra disclosed the existence
of five 1H−1H spin systems with very similar chemical shifts to
those of 6 and 7: I [H-3→H2-5]; II [H-7→H2-9]; III [H-11→
H-14]; IV [H2-16→H-18]; and V [H2-20→H-22] (Figure 1).
Finally, both H2-28 showed an indistinguishable chemical shift
at δH 3.55 clearly suggesting the presence of an additional
19
20
84.0, C
84.7, C
34.5, CH2 1.64, m
2.12, m
35.4, CH2 1.72, m
1.95, m
21
22
26.2, CH2 1.81, m
1.86, m
26.6, CH2 1.83, m
87.2, CH
3.74, dd (5.8,
9.6)
87.0, CH
3.77, dd (6.3/8.9)
1
primary alcohol at C-28. Long-range H−13C connectivities
23
24
25
26
27
28
29
30
70.4, C
70.7, C
extracted from the HMBC experimental data allowed linking
1
the previous H−1H spin systems. Particularly important were
24.1, CH3 1.12, s
23.7, CH3 1.40, s
20.0, CH3 1.20, s
21.5, CH3 1.19, s
67.6, CH2 3.55, bs
24.4, CH3 1.19, s
27.7, CH3 1.21, s
24.1, CH3 1.12, s
23.5, CH3 1.40, s
20.2, CH3 1.19, s
20.4, CH3 1.15, s
27.2, CH3 1.28, s
23.1, CH3 1.16, s
27.7, CH3 1.21, s
the long-range correlations observed from the methyl groups,
CH3-1/CH3-25 to C-2 and C-3; CH3-26 to C-5, C-6, and C-7;
CH3-27 to C-9, C-10, and C-11; H3-29 to C-18, C-19, and C-
20; and finally CH3-24/CH3-30 to C-22 and C-23, that secured
the assignments. Moreover, HMBC correlations between CH2-
28 and C-14, C-15, and C-16 supported the previous
conclusions and allowed the determination of the planar
structure of 2 as shown in Figure 1.
The stereostructure of 2 was established by examination of
the ROESY experiment together with a comparison of its H
was complicated by the fact that these stereocenters are
connected by a single bond and C-15 is a nonprotonated
carbon, limiting the number of accessible geometrical restraints.
Therefore, in order to confirm the stereochemical proposal,
both C-15 epimers of compound 2 were obtained using a
semisynthetic approach starting from a known compound,
dehydrothyrisferol (1) (Scheme 1).9 Thus, a solution of 1 (1.5
mg/0.5 mL) in CH2Cl2 was stirred at room temperature while
adding m-chloroperbenzoic acid (MCPBA; 1.5 equiv) to afford
a mixture of two C-15 epimers (2 and 8), which were separated
by HPLC with a μ-Porasil column using a mixture of n-hexane/
acetone (7:3). The stereostructure of C-15 in 2 was secured as
1
and 13C chemical shift values with those available for other
related molecules. Thus, the relative configuration of the three
oxane rings and the C-19→C-22 oxolane ring was defined as
identical to those of 6 and 7. Key dipolar correlations were
observed between H3-1/H-3, H3-25/H3-26, H-7/H-11, H-11/
H-14, H-8β/H3-27, and H-12β/H3-27 (Figure 1). A cis-
configuration was predicted for the oxolane ring at C-15→C-
18 after the observation of a clear ROE correlation between H-
18 and H2-28. However, as it was the case for 6 and 7,
assignment of the relative configurations of C-14 and C-15 in 2
713
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
recognized by analysis of the COSY, TOCSY, and HSQC
1
spectra of 3. Four H−1H spin systems, I [C-3→C-5], II [C-
7→C-9], IV [C-16→C-18], and V [C-20→C-22], were
analogous to those observed in saiyacenols A and B (6 and
1
7). The remaining H−1H spin system, III [C-11→C-14], was
conveniently assigned starting from methine H-11 (δH 3.09),
which is coupled with methylene H2-12 (δH 2.01/2.32), and
these protons were coupled, in turn, with H-13 (δH 5.64),
which was further correlated to methine H-14 (δH 5.52). Long-
range 1H−13C correlations observed in the HMBC spectra were
used to link these 1H−1H spin systems. Thus, methyls CH3-1/
CH3-25 connected with C-2 and C-3; CH3-26 with C-5, C-6,
and C-7; CH3-27 with C-9, C-10, and C-11; CH3-29 with C-18,
C-19, and C-20; and CH3-24/CH3-30 with C-22 and C-23,
confirming that fragments C-1→C-11 and C-19→C-24 are
identical to those of saiyacenols A (6) and B (7). Finally,
HMBC correlations between H3-28 (δH 1.28) and C-14 (δC
137.3), C-15 (δC 82.8), and C-16 (δC 37.7) allowed us to
determine the planar structure of 3 as shown in Figure 2.
Figure 1. (Top) Selected NMR-derived correlations observed for 28-
1
hydroxysaiyacenol B (2). H,1H spin systems are numbered from I to
V. (Bottom) Key NOE correlations used to determine the relative
configuration of 2.
Scheme 1. Epoxidation of Dehydrothyrsiferol, Yielding
Compounds 2 and 8
Figure 2. (Top) Selected NMR-derived correlations observed for
saiyacenol C (3). 1H,1H spin systems are numbered from I to V.
(Bottom) Key NOE correlations used to determine the relative
configuration of 3.
Examination of the ROESY experiment allowed us to obtain a
number of key correlations (H3-1/H-3, H3-25/H3-26, H-7/H-
11, and H-8β/H3-27) that secured the relative configuration
within the two oxane rings present in 3. Relative configurations
within the C-15→C-18 oxolane ring were established by
observation of dipolar correlations between the pair of olefinic
protons H-13/H-14 and H-18, locating them in the same side
of the ring (Figure 2). Finally, the relationship between the
configurations of C-18 and C-19 was established comparing the
chemical shifts of 3 with those of two molecules sharing the
same C-15→C-24 moiety: saiyacenol A (6) and a synthetic
penta-THF (18S*,19R* and 18R*,19R*, respectively, Support-
ing Information).10 Thus, using the DP4 parameter,11 the
relative configuration 3R*, 6S*, 7R*, 10S*, 11R*, 15S*, 18S*,
19R*, and 22 R* was selected for 3 with a 92.3% probability.
This result is in perfect agreement with the fact that a 18S*,
19R* relationship has been found for all natural compounds
belonging to the thyrsiferol series.1
S* by observation of a ROE between H-18 and H3-28, leaving 8
as its epimer.
Finally, NMR data of the natural 28-hydroxysaiyacenol B
were compared with those of the synthetic compounds to find
identity with the 15S* epimer (2). Considering that in 1 the
stereochemical relationships of C-18 with all other stereo-
centers were known, this result sets the relative configuration of
2 as 3R*, 6S*, 7R*, 10S*, 11R*, 14R*, 15S*, 18S*, 19R*,
22R*.
The molecular formula of saiyacenol C (3), C30H51BrO6, was
deduced by HRESIMS. This observation was consistent with its
13C NMR spectrum, where eight methyls, nine methylenes,
seven methines, and six nonprotonated carbons bearing oxygen
carbons were detected. According to these data, five degrees of
unsaturation, corresponding to one double bond and four rings,
1
were present in 3 (Table 1). Comparison of the observed H
and 13C NMR chemical shifts of 3 with those previously
reported for saiyacenols A (6) and B (7) suggested the absence
of the characteristic C-7→C-14 dioxabicyclo[4.4.0]decane.
Now, two signals at δH 5.52 and 5.64 ppm, corresponding to
an E double bond (JH13H14 = 16.0 Hz), were clearly observed
15,16-Epoxythyrsiferol A (4) was isolated as an amorphous
solid, and its molecular formula was determined as C30H51BrO7
1
based on HRESIMS measurements. Analysis of its 13C and H
NMR data indicated the presence of eight methyls, nine
methylenes, seven methines, and six nonprotonated carbons
1
in the NMR spectrum. Five H−1H spin systems were easily
714
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
bearing oxygen (Table 2). The absence of olefinic carbons in
this molecule suggested the presence of five oxygenated rings.
Table 2. NMR Spectroscopic Data (1H 600 MHz, 13C 150
MHz, CDCl3) for Compounds 4 and 5
15,16-epoxythyrsiferol A (4)
δC, type δH (J in Hz)
31.0, CH3 1.27, s
15,16-epoxythyrsiferol B (5)
δC, type δH (J in Hz)
C
1
31.2, CH3 1.27, s
74.9, C
2
3
75.0, C
59.0, CH
3.89, dd (4.1, 12.4)
58.8, CH
3.89, dd (4.0,
12.4)
4
5
28.2, CH2 α 2.10, m
β 2.24, m
28.2, CH2 α 2.10, m
β 2.24, m
37.1, CH2 α 1.53, m
β 1.80, m
37.1, CH2 α 1.53, m
β 1.80, m
6
7
74.4, C
74.3, C
86.5, CH
3.04, dd (2.6, 11.5)
86.6, CH
3.04, dd (2.6,
11.5)
8
9
23.0, CH2 β 1.45, m
α 1.74, m
23.0, CH2 β 1.45, m
α 1.74, m
Figure 3. (Top) Selected NMR-derived correlations observed for
15,16-epoxythyrsiferol A (4). 1H,1H spin systems are numbered from I
to V. (Bottom) Key NOE correlations used to determine the relative
configuration of 4.
38.6, CH2 α 1.53, m
β 1.79, m
38.6, CH2 α 1.53, m
β 1.79, m
10
11
72.0, C
72.0, C
77.1, CH
3.44, dd (6.5, 11.1)
77.7, CH
3.38, dd (6.5,
11.1)
where H3-28 (δH 1.27) was correlated with C-14 (δC 74.2), C-
15 (δC 60.9), and C-16 (δC 61.6), as well as by the correlations
of methyl H3-29 (δH 1.15) with C-18 (δC 75.9), C-19 (δC 85.4),
and C-20 (δC 32.6). Other HMBC connectivities confirmed the
presence of the bromopyran dioxabicyclo[4.4.0]decane and the
terminal oxolane ring. Finally, due to the characteristic chemical
shifts of C-15 and C-16, the presence of an epoxide
functionality within fragment IV, accounting for the last cyclic
system within this molecule, was determined.
12
13
14
21.3, CH2 β 1.54, m
α 1.83, m
21.3, CH2 β 1.54, m
α 1.83, m
22.1, CH2 α 1.79, m
β 1.91, m
22.3, CH2 α 1.70, m
β 1.83, m
74.2, CH
3.56, dd (3.6, 10.7)
74.6, CH
3.66, dd (3.6,
10.7)
15
16
60.9, C
60.9, C
61.6, CH
3.08, dd (4.9, 7.3)
58.7, CH
3.12, dd (4.9,
7.3)
The stereostructure of 4 was initially accomplished using
dipolar correlations extracted from the NOESY experiment
(H3-1/H-3, H3-25/H3-26, H-7/H-11, and H-8β/H3-27) that
allowed us to ensure the relative configuration within the C-1→
C-14 moiety of 4 as identical to those identified in 1, 2, 6, 7,
and 8. Determination of the stereochemical relationship
between C-18 and C-19, connected by a single bond, was
complicated because C-19 is a nonprotonated carbon, thus
reducing the number of coupling constants available for
17
30.6, CH2 1.54, ddd (7.3, 10.5,
14.2)
31.1, CH2 1.56, m
1.88, ddd (2.1, 4.9,
14.2)
2.08, m
18
75.9, CH
85.4, C
3.75, dd (2.1, 10.5)
74.3, CH
85.8, C
3.76, dd (2.1,
10.5)
19
20
32.6, CH2 1.64, m
2.12, m
32.6, CH2 1.64, m
2.12, m
21
22
26.5, CH2 1.86, m
26.5, CH2 1.86, m
measurement.12 Nevertheless, the observed values of JC19−H18
2
87.5, CH
3.77, dd (5.9, 10.0)
87.9, CH
3.77, dd (5.9,
10.0)
3
3
= 2.5 Hz, JC20−H18 = 3.5 Hz, and JC29−H18 = 2.7 Hz, together
with the NOE between H-17α (δH 1.54) and H-20α (δH 2.12),
supported the 18S*, 19R* relative configuration (Figure 4),
equal to that observed in dehydrothyrsiferol (1). In addition,
1H and 13C chemical shifts and coupling constants along the C-
18→C-24 moiety of 4 exhibited very good correspondence
with those of 1 (Table S6, Supporting Information), reinforcing
the previous conclusion. The relative configurations of C-16
and C-18 were related using a J-based configurational approach,
23
24
25
26
27
28
29
30
70.4, C
70.4, C
24.0, CH3 1.13, s
23.6, CH3 1.40, s
20.1, CH3 1.19, s
20.7, CH3 1.23, s
12.4, CH3 1.27, s
23.0, CH3 1.15, s
27.7, CH3 1.22, s
24.0, CH3 1.13, s
23.6, CH3 1.40, s
20.0, CH3 1.20, s
20.7, CH3 1.23, s
13.6, CH3 1.30, s
23.8, CH3 1.16, s
27.7, CH3 1.22, s
2
3
measuring coupling constants (3JH,H, JC,H, and JC,H) that can
be used in cyclic or linear systems.13,14 The analysis was
conveniently started from the C-18→C-17 bond, where the
Analysis of COSY and HSQC spectra allowed us to build five
1H−1H spin systems: I [C-3→C-5], II [C-7→C-9], III [C-11→
C-14], IV [C-16→C-18], and V [C-20→C-22]. In this case,
spin system IV [C-16→C-18] was different from those of 2 and
3, as it contains only one methylene (Figure 3). This
substructure was conveniently started from methine H-16 (δH
3.08), which is coupled with methylene H2-17 (δH 1.54/1.88),
and this in turn with H-18 (δH 3.75). Next, IV was connected
with its neighbor fragments through the HMBC experiment,
3
3
observed values for JH18−H17b = 10.5 Hz and JH18−H17a = 2.1
Hz suggested an anti-conformation between H-18 (δH 3.75)
and H-17b (δH 1.54) and a gauche-relationship for H-18 and H-
17a (δH 1.88). Once the positions of both H-17 hydrogens
were identified, we proceeded with the C-16→C-17 bond;
however it has to be noted that H-16 is attached to one of the
epoxy-carbons, so the geometry along this bond is characteristic
3
of such functionality. Thus, the observed values of JH17a‑H16
=
715
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
4.7 Hz, 3JH17b‑H16 = 7.2 Hz, 3JC15−H17a = 2.3 Hz, and 3JC18−H16
=
2.6 Hz imply a threo-configuration between H-16 and H-17a
that can be explained by a conformational equilibrium, as
shown in Figure 4.10
Still, the relative configurations of C-16→C-24 had to be
associated with those of C-1→C-14 through the epoxide
functionality. Fortunately a very similar metabolite, 15,16-
epoxythyrsiferol B (5), that was also isolated showed the same
molecular formula observed for 4, C30H51BrO7, as deduced by
1
HRESIMS. In addition, comparison of their H and 13C NMR
chemical shifts together with an analysis of their 2D NMR data
allowed us to establish an identical planar structure for both
1
compounds. Nevertheless, H and 13C NMR chemical shifts of
the C-15→C-18 moiety, including methyl C-28, showed subtle
discrepancies. Therefore, it was concluded that these differ-
ences were due to different configurations at C-15 and/or C-16.
Next, a trans-configuration for the new epoxides (4 and 5) was
determined by evaluation of the NOESY experiment. Thus, the
observation of strong dipolar correlations between H-16 and H-
14 and H-18 as well as between H3-28 and H2-17 was very
informative. Further support for this conclusion was obtained
from long-range heteronuclear coupling constants. Thus,
Figure 4. Configurational analysis of fragment C-16→C-19 in 15,16-
epoxythyrsiferol A.
Scheme 2. Synthesis of Model Compounds 14−17
716
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
3JC14−H16 = 1.9 Hz was measured for 15,16-epoxythyrsiferol A
(4) using the JHMBC experiment.15 The measured dihedral
angle for a trans-configuration is ∼0°, while for a cis-
configuration it is ∼120°, and in consequence the expected
3JCH value is smaller in cis-isomers than in trans-isomers. Aliev
et al. reported values of <0.5 Hz for the cis-isomers and ∼2 Hz
for the trans-isomers in a similar structural motif, in agreement
with the previous observations.16 Nevertheless, differentiating
the two trans-isomers was clearly more challenging. Measure-
ment of a full set of nJCH was hindered by the fact that C-15 is a
nonprotonated carbon and because the carbon chemical shifts
for C-15 and C-16 are very similar in 4 (Δδ = 0.7 ppm).
Therefore, simplified models of the epoxides were synthesized
to address this problem. Thus, a stereocontrolled synthesis of
the four possible diastereomers of (3-methyl-3-((S)-tetrahydro-
2H-pyran-2-yl)oxiran-2-yl)methanol was carried out using 3,4-
dihydro-2H-pyran as starting material. All stereocenters were
introduced taking advantage of the asymmetric Katsuki−
Sharpless epoxidation (Scheme 2).17 The hemiacetal obtained
by treatment of 3,4-dihydro-2H-pyran with aqueous 0.2 M HCl
underwent a Wittig-type reaction in order to obtain the
hydroxy-α,β-unsaturated ester 9.18 This ester was reduced using
DIBAL-H, and the obtained diol 10 underwent a Katsuki−
Sharpless epoxidation reaction using (−)-DET as the chiral
ligand. After the oxidative cleavage of 11, the ketone obtained
reacted under Ando phosphonate19 to give a mixture of the two
possible α,β-unsaturated esters in a 1.6:1 ratio, where the E-
isomer 12a was the major product. Compounds 12a and 12b
were separated by chromatography and reduced to allylic
alcohols 13a and 13b, respectively, precursors required for the
subsequent Katsuki−Sharpless epoxidation. Using 13a as
starting material the epoxide 14 was obtained when (−)-DET
was used as the chiral ligand, and the epoxide 15 when
(+)-DET was the chiral ligand used. A mixture of the epoxides
16 and 17 in a 2:1 ratio, respectively, by treatment of both (−)-
and (+)-DET was obtained when 13b was used as starting
material. In order to confirm the absolute configurations of
these two epoxides, the p-bromobenzoate derivative 18 of the
epoxy alcohol 17 was prepared, and its structure confirmed by
X-ray crystallography.
the calculations were performed in vacuo, they predicted the
same equilibrium along the C-16→C-17 bond that was
proposed in CHCl3 on the basis of the coupling constant
information (providing a 58:42 ratio for rotamers A and B as
depicted in Figure 4).25 Using these data average chemical
shifts were calculated for each diastereoisomer. Finally, the
computed and the experimental values were compared using
the CP3 parameter, which can be applied to the case of
assigning a pair of diastereoisomers when one has both
experimental data sets.26 The result was the assignment of the
3R*, 6S*, 7R*, 10S*, 11R*, 14R*, 15S*, 16R*, 18S*, 19R*, and
22R* relative configuration for 15,16-epoxythyrsiferol A (4)
and 3R*, 6S*, 7R*, 10S*, 11R*, 14R*, 15R*, 16S*, 18S*, 19R*,
and 22R* for 15,16-epoxythyrsiferol B (5) with a probability of
1
99.9% when both H and 13C chemical shifts were taken into
account.
Antifouling Studies. Compounds 1−4, 7, and 8 were
evaluated for their antifouling activities toward a panel of
organisms composed of marine bacteria, marine-derived fungi,
benthic diatoms, and macroalgal zoospores.27 Whereas none of
them displayed activity toward the bacterial and fungal strains
used (data not shown), meaningful dose-dependent inhibitions
were observed in the diatom growth and zoospore germination
assays (Table 3). Thus, dehydrothyrsiferol (1) and its
Table 3. Activities (μM) Displayed by Tested Natural and
a
Synthetic Compounds
IC50
MIC
Phaeodactylum
tricornutum
Cylindrotheca
Navicula cf.
salinicola
Gayralia
oxysperma
compound
sp.
1
>100
18.3
17.8
11.6
23.7
13.7
100
25
2
3
>100
>100
67.7
17.1
46.7
>100
>100
4
>100
13.0
>100
17.2
b
7
n.t.
8
63.3
23.6
18.0
25
>100
>100
>100
14
15
16
17
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
A comparison of the NMR chemical shifts extracted from the
synthetic models with those of the natural products allowed us
to propose the relative configurations of 4 and 5. Thus, a
significant difference was observed in the natural products
between δC15 − δC16 = 0.7 ppm for 4 and δC15 − δC16 = 2.2 ppm
for 5. This pattern was also found for the synthetic
diastereoisomers that showed Δδ = 2.8 ppm for 14 and Δδ
= 0.1 ppm for 15. On the basis of these values, a 14R*, 15S*,
16R* relative configuration in 15,16-epoxythyrsiferol A (4) and
14R*, 15R*, 16S* for 15,16-epoxythyrsiferol B (5) were
proposed.
The synthetic approach was complemented using quantum
mechanical calculations of NMR chemical shifts, a tool that has
been shown to be successful even in highly complex marine
natural products.20,21 Thus, the two possible C-15→C-16
diastereoisomers were built and conformational searches were
performed for each using a hybrid MCMM and low-mode
sampling and the MMFF94 force field.22 Next, all conformers
within an energy threshold of 10 kJ/mol of the global minimum
found were used as input structures for density functional
theory (DFT) calculations using the B3LYP functional and the
LAVCP**+ basis set to calculate their relative energies and
isotropic chemical shieldings.23,24 It is noteworthy that although
a
b
IC50 values were calculated from three experimental replicates. n.t. =
not tested.
congeners saiyacenols B (7) and C (3) prevented Navicula cf.
salinicola and Cylindrotheca sp. growth at micromolar
concentrations, while both 28-hydroxysaiyacenols B and A (2
and 8) also inhibited the germination of Gayralia oxysperma
spores. On the basis of these results, it seems clear that the 28-
hydroxy functionality is essential for the activity of this
polyether skeleton. In fact, 2 and 8 exhibited the most potent
activities, showing IC50 values in the 10−60 μM range for the
three diatom strains tested and an MIC value of 25 μM for G.
oxysperma zoospore germination. However, it seems that the
influence of the C-15 configuration on the bioactivity can be
neglected, as both epimers showed similar activities. Another
observation that reinforces the previous conclusion about the
importance of the 28-hydroxy group is the fact that epoxidation
at C-15 and C-16 as in 15,16-epoxythyrsiferol A (4) leads to a
complete loss of bioactivity. This structural hint is further
supported by the lack of activity of the synthetic models (14−
17).
717
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
Biological Material. Specimens of Laurencia viridis were collected
in April 2013 in the intertidal zone at Paraiso Floral, Tenerife, Canary
Islands (28°07′12″ N, 16°46′45″ W). Dried material from the sterile
plants, sporophytes, and gametophytes was filed at TFC Phyc 7180
CONCLUSIONS
■
Four new oxasqualenoids (2−5) were isolated from the red alga
L. viridis. Their relative configurations were determined by a
combined experimental−computational analysis. The existence
of two diastereomeric epoxides (4 and 5) comprising a
nonprotonated carbon complicated the determination of their
configurations, making necessary the synthesis of simplified
models. The approach used combined the analysis of NOE
correlations, homo- and heteronuclear J couplings, and
chemical shift comparisons with synthetic and computational
models. The three different NMR approaches were in
agreement with the proposed structures. The methodological
approach and the conclusions obtained from the structural
study can be of general use for similar functionalities existing in
other molecules. Although different bioactivities have been
reported for oxasqualenoids, this is the first report on their
antifouling activity. Ten compounds (1−4, 7, 8, 14−17) were
tested against a panel of biofouling species including marine
bacteria and fungi, benthic diatoms, and macroalgal zoospores
to get a broad overview of their potential effects. Whereas none
of them displayed activity toward the bacterial and fungal
strains used, this was not the case in diatom growth and
zoospore germination assays (Table 3). On the basis of the
results obtained it was concluded that a 28-hydroxy
functionality improves the activity of this molecular backbone.
On the other hand, an epoxide functional group neighboring C-
28 as in 15,16-epoxytirsiferol A (4) leads to a complete loss of
bioactivity. Given the specific activity of these naturally
occurring oxasqualenoieds, particularly the 28-hydroxy deriva-
tives, toward both micro- and macroalgal colonizers at
micromolar concentrations, it seems reasonable to postulate a
possible ecological role of these secondary metabolites as
chemical defenses through a targeted action against algal
epibionts.28
́
(Herbario de la Universidad de La Laguna, Departamento de Biologia
Vegetal, Botanica, Tenerife, Spain).
́
Extraction and Isolation. The specimens of L. viridis were
extracted with CHCl3/MeOH (1:1) at room temperature (rt), and a
dark green, viscous oil was obtained (83.0 g) after concentration under
reduced pressure. The extract was first chromatographed using
Sephadex LH-20 (7 × 50 cm) using CH2Cl2/MeOH (1:1) as the
mobile phase. The enriched polyether fraction (53.5 g) collected
between 225 and 360 mL was subsequently processed on a silica gel
column (7 × 50 cm) using a linear gradient of n-hexane/EtOAc
(80:20−20:80), and fractions collected between 350 and 500 mL were
dried (17.1 g). Next, medium-pressure chromatography was done on
Lobar LiChroprep Si-60 using CH2Cl2/acetone (8:2) at 1 mL/min,
and fractions collected between 80 and 105 min were pooled together
(770 mg). Final purification was done on an HPLC with a μ-Porasil
column using n-hexane/EtOAc/MeOH, 18:15:5, sequentially to afford
four new polyethers: 15,16-epoxythyrsiferol A (4) (2.5 mg) from
fractions collected at 39 min, a 1:4 mixture of 4 and 15,16-
epoxythyrsiferol B (5) (0.9 mg) from fractions collected at 40 min, 28-
hydroxysaiyacenol B (2) (2.3 mg) from fractions collected between 54
and 57 min, and saiyacenol C (3) (9 mg) from fractions collected
between 82 and 88 min.
28-Hydroxysaiyacenol B (2): amorphous, white solid; [α]25 +4 (c
D
0.25, CH2Cl2); IR (CHCl3) νmax 3385, 2971, 2362, 2334, 1375, 1061
cm−1; 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 125
MHz), Table 1; HRESIMS m/z 627.2634/625.2709 [M + Na]+ (calcd
for C30H5179BrNaO7, 625.2716).
Saiyacenol C (3): amorphous, white solid; [α]25 +20 (c 0.15,
D
CH2Cl2); IR (CHCl3) νmax 3414, 2973, 2866, 2042, 1458, 1377, 1126
cm−1; 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 125
MHz), Table 1; HRESIMS m/z 611.2823/609.2724 [M + Na]+ (calcd
for C30H5179BrNaO6, 609.2767).
15,16-Epoxythyrsiferol A (4): amorphous, white solid; [α]25D +2 (c
0.14, CH2Cl2); IR (CHCl3) νmax 3550, 2980, 1470, 1455, 1125 cm−1;
1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 125 MHz),
Table 2; HRESIMS m/z 627.2694/625.2723 [M + Na]+ (calcd for
EXPERIMENTAL SECTION
■
C30H5179BrNaO7, 625.2716).
General Experimental Procedures. All reagents were commer-
cially available and used as received. All solvents were dried and
distilled under argon immediately prior to use or stored appropriately.
THF and Et2O were refluxed over sodium and benzophenone. CH2Cl2
was distilled from CaH2. Reactions were monitored by TLC. Flash
chromatography was performed with silica gel (230−400 mesh) as the
stationary phase and mixtures of n-hexane and EtOAc, in different
proportions given in each case, as the mobile phase. Melting points
Chemical Transformation of Dehydrothyrsiferol (1) into 28-
Hydroxysaiyacenol B (2) and 28-Hydroxysaiyacenol A (8). m-
Chloroperbenzoic acid (1.5 equiv) was added into a CH2Cl2 (0.5 mL)
solution containing 1.5 mg of dehydrothyrsiferol (1). The resulting
mixture was stirred for 3 h at rt. Afterward, the solution was filtered
and concentrated to give a solid residue, which was chromatographed
using HPLC (μ-Porasil column, n-hexane/acetone, 7:3, flow rate 1
mL/min), affording 0.8 mg of 28-hydroxysaiyacenol B (2) and 0.6 mg
of 28-hydroxysaiyacenol A (8).
were determined on a Buchi B-540 model. Optical rotations were
̈
determined on a PerkinElmer 343 polarimeter using a sodium lamp
operating at 589 nm. IR spectra were measured on a Bruker IFS 66
spectrometer using a methanolic solution over a NaCl disk or neat.
NMR spectra were performed on Bruker Avance 400, 500, or 600
instruments at 300 K, and coupling constants are given in Hz. COSY,
1D/2D TOCSY, HSQC, HMBC, and ROESY experiments were
performed using standard pulse sequences. Phase-sensitive ROESY
28-Hydroxysaiyacenol A (8): amorphous, white solid; [α]25D +2 (c
0.05, CH2Cl2); IR (CHCl3) νmax 3385, 2971, 2362, 2334, 1375, and
1061 cm−1; 1H NMR (CDCl3, 600 MHz) and 13C NMR (CDCl3, 125
MHz), Table 2; HRESIMS m/z 627.2720/625.2711 [M + Na]+ (calcd
for C30H5179BrNaO7, 625.2716).
Preparation of Synthetic Models. (E)-Ethyl 7-Hydroxy-2-
methylhept-2-enoate (9). To 3,4-dihydro-2H-pyran (16.6 mL, 178.3
mmol) was added HCl(aq) (38 mL, 0.2 M) at 0 °C. The solution was
stirred for 15 min and then stirred at rt for 1 h. The resultant mixture
was extracted with CH2Cl2. The combined organic layer was washed
with saturated aqueous NaHCO3 and brine, dried over MgSO4,
filtered, and concentrated under vacuum to afford the crude hemiacetal
as a colorless oil.
The hemiacetal (6.6 g, 64.7 mmol) was added to a solution of
ethoxycarbonylethylidene triphenylphosphorane (24.2 g, 64.7 mmol)
in benzene (100 mL). The mixture was heated under reflux for 3 h,
then cooled to rt and concentrated under reduced pressure.
Purification by column chromatography afforded the α,β-unsaturated
ester 9 (7.88 g, 47% after two steps, E:Z = 94:6) as a colorless oil. The
3
spectra were measured using a mixing time of 500 ms. JH,H values
were measured from 1D H NMR and, when signal overlapping did
1
not permit it, from the TOCSY experiment. J-HMBC pulse sequence
was used to measure long-range heteronuclear coupling constants. A J-
scale factor of 52 was used, and the experiment was optimized for
long-range couplings of 2 Hz. Data were processed using Topspin or
MestRe software. Mass spectra were recorded on a LCT Premier XE
Micromass spectrometer using electrospray ionization. X-ray crystal-
lography was performed using an Oxford Diffraction Supernova
System. TLC was performed on AL Si gel. TLC plates were visualized
by UV light (254 nm) and by adding a phosphomolybdic acid solution
10 wt % in MeOH or a vanillin solution (6 g of vanillin, 450 mL of
EtOH, 40 mL of AcOH, and 30 mL of H2SO4).
718
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
observed 1H NMR spectrum was consistent with that previously
reported in the literature.29 Rf: 0.25 (n-hexane/EtOAct, 6:4, silica gel).
(E)-2-Methylhept-2-ene-1,7-diol (10). To a solution of α,β-
unsaturated ester 9 (7.88 g, 42.3 mmol) in diethyl ether (423 mL),
cooled to 0 °C and under argon atmosphere, was added dropwise
DIBAL-H 1 M in cyclohexane (85 mL, 85.0 mmol). The reaction was
allowed to warm for 1 h and then was recooled to 0 °C, diluted with
diethyl ether (100 mL), treated with distilled H2O (7.7 mL, 427.8
mmol), and stirred vigorously for 30 min. After that, MgSO4 was
added, and 15 min later the mixture was filtered over a Celite pad,
washed with diethyl ether, and concentrated. Diol 10 was obtained as a
colorless oil after purification by column chromatography (5.49 g,
31.0 (CH2, C-3′), 25.9 (CH2, C-4′), 23.9 (CH2, C-5′), 15.7 (CH3, C-
4); HRESIMS m/z 207.0989 [M + Na]+ (calcd for C10H16NaO3,
207.0997); Rf 0.53 (n-hexane/EtOAc, 9:1, silica gel).
12b: yellowish oil; 1H NMR (500 MHz, CDCl3) δ 5.60−5.62 (1H,
m, H-2), 5.10 (1H, d, J = 10.9 Hz, H-2′), 3.96−4.00 (1H, m, H-6′),
3.66 (3H, s, CO2Me), 3.50 (1H, td, J = 11.7, 2.7 Hz, H-6′), 1.91 (3H,
d, J = 1.2 Hz, H-4), 1.83−1.89 (1H, m, H-4′), 1.65−1.73 (1H, m, H-
4′), 1.61−1.64 (1H, m, H-3′), 1.53−1.60 (1H, m, H-5′), 1.48−1.52
(1H, m, H-5′), 1.39−1.47 (1H, m, H-3′); 13C NMR (100 MHz,
CDCl3) δ 166.3 (C, C-1), 161.6 (C, C-3), 114.9 (CH, C-2), 77.0 (CH,
C-2′), 68.3 (CH2, C-6′), 51.1 (CH3, CO2Me), 30.3 (CH2, C-3′), 26.0
(CH2, C-4′), 23.7 (CH2, C-5′), 19.7 (CH3, C-4); HRESIMS m/z
207.0989 [M + Na]+ (calcd for C10H16NaO3, 207.0997); Rf 0.65 (n-
hexane/EtOAc, 9:1, silica gel).
1
90%), and the observed H NMR spectrum was consistent with that
previously reported in the literature.30 Rf: 0.32 (n-hexane/EtOAc, 4:6,
silica gel).
(S,E)-3-(Tetrahydro-2H-pyran-2-yl)but-2-en-1-ol (13a). Com-
pound 13a was obtained using the same procedure as for diol 10,
using ester 12a (134 mg, 0.73 mmol), diethyl ether (7.3 mL), and
DIBAL-H 1 M in cyclohexane (1.8 mL, 1.8 mmol). The final product
was purified by column chromatography to give the alcohol 13a (95.8
(R)-2-((S)-Tetrahydro-2H-pyran-2-yl)propane-1,2-diol (11). A flask
with molecular sieves powder (4 Å) was flamed and then cooled to
−20 °C. A solution of diol 10 (1.34 g, 9.29 mmol) in CH2Cl2 (93 mL)
was added under argon atmosphere, followed by freshly distilled
titanium(IV) isopropoxide (3.3 mL, 11.15 mmol) and (−)-diethyl D-
tartrate (2.2 mL, 13.01 mmol). After 30 min, tert-butyl hydroperoxide
solution 5.06 M in isooctane (3.3 mL, 16.72 mmol) was added under
an argon atmosphere and then stirred for 18 h. Once the reaction was
finished, 15% tartaric acid(aq) (100 mL) was added, and the mixture
was vigorously stirred at rt for 15 min. Then the aqueous layer was
saturated with NaCl powder and extracted with CH2Cl2 (3 × 75 mL),
and the combined organic layer was concentrated under vacuum.
Purification by column chromatography afforded diol 11 (1.01 g,
68%): white solid; mp 35.8−38.3 °C; 1H NMR (400 MHz, CDCl3) δ
4.00 (1H, d, J = 11.0 Hz, H-6′), 3.74 (1H, d, J = 11.0 Hz, H-1), 3.27−
3.42 (3H, m, H-1, H-2′, H-6′), 2.86−2.99 (2H, m, 2 × OH), 1.84−
1.91 (1H, m, H-5′), 1.64 (1H, d, J = 12.0 Hz, H-4′), 1.42−1.55 (3H,
m, H-3′, H-4′, H-5′), 1.32 (1H, dtd, J = 12.0, 12.0, 3.0 Hz, H-3′), 1.06
(3H, s, H-3); 13C NMR (100 MHz, CDCl3) δ 84.6 (CH, C-2′), 73.3
(C, C-2), 69.3 (CH2, C-6′), 67.7 (CH2, C-1), 26.2 (CH2, C-3′ or C-
4′), 26.1 (CH2, C-3′ or C-4′), 23.5 (CH2, C-5′), 20.4 (CH3, C-3);
EIMS m/z 129 [M − CH2OH]+ (80), 111 (35), 85 [M −
C(Me)(OH)CH2OH]+ (100), 57 (59); HRESIMS m/z 183.0996
[M + Na]+ (calcd for C8H16NaO3, 183.0997); Rf 0.36 (n-hexane/
EtOAc, 4:6, silica gel).
mg, 84%): yellow oil; [α]25 −25 (c 0.7, CHCl3); ATR-FTIR (neat)
D
νmax 3375, 2935, 2851, 1440, 1264, 1204, 1085, 1038, 1009, 633 cm−1;
1H NMR (400 MHz, CDCl3) δ 5.65 (1H, t, J = 6.6 Hz, H-2), 4.19
(2H, d, J = 6.7 Hz, H-1), 4.03 (1H, d, J = 11.0 Hz, H-6′), 3.65 (1H, d,
J = 10.7 Hz, H-2′), 3.48 (1H, t, J = 11.1 Hz, H-6′), 1.83−1.91 (1H, m,
H-5′), 1.68 (3H, s, H-4), 1.42−1.66 (5H, m, 2 × H-3′, 2 × H-4′, 1 ×
H-5′); 13C NMR (100 MHz, CDCl3) δ 139.9 (C, C-3), 124.3 (CH, C-
2), 82.4 (CH, C-2′), 68.8 (CH2, C-6′), 59.3 (CH2, C-1), 30.6 (CH2,
C-3′), 26.1 (CH2, C-4′), 23.8 (CH2, C-5′), 13.1 (CH3, C-4); EIMS m/
z 156 [M]+ (1), 141 [M − Me]+ (1), 125 [M + 1 − CH2OH]+ (100),
69 (31); HRESIMS m/z 156.1148 (calcd for C9H16O2 [M + Na]+,
156.1150); Rf 0.12 (n-hexane/EtOAc, 7:3, silica gel).
(S,Z)-3-(Tetrahydro-2H-pyran-2-yl)but-2-en-1-ol (13b). Com-
pound 13b was obtained using the same procedure as for diol 10,
using ester 12b (254.4 mg, 1.38 mmol), diethyl ether (14 mL), and
DIBAL-H 1 M in cyclohexane (3 mL, 3 mmol, 1 h). The final product
was purified by column chromatography to give the alcohol 13b
(202.9 mg, 94%): yellow oil; [α]25D −12.0 (c 1.0, CHCl3); ATR-FTIR
(neat) νmax 3371, 2935, 2852, 1440, 1269, 1205, 1085, 1020, 997, 631
1
cm−1; H NMR (500 MHz, CDCl3) δ 5.51 (1H, t, J = 7.0 Hz, H-2),
4.08−4.17 (2H, m, H-1), 4.05−4.07 (1H, d, J = 10.4 Hz, H-2′), 3.99−
4.04 (1H, m, H-6′), 3.47 (1H, t, J = 11.5 Hz, H-6′), 1.99 (1H, br s,
OH), 1.85−1.90 (1H, m, H-5′), 1.74 (3H, s, H-4), 1.49−1.60 (5H, m,
2 × H-3′, 2 × H-4′, 1 × H-5′); 13C NMR (100 MHz, CDCl3) δ 140.1
(C, C-3), 126.1 (CH, C-2), 77.9 (CH, C2′), 68.8 (CH2, C-6′), 58.5
(CH2, C-1), 30.4 (CH2, C-3′), 25.9 (CH2, C-4′), 23.8 (CH2, C-5′),
20.1 (CH3, C-4); EIMS m/z 156 [M]+ (3), 141 [M − Me]+ (3), 125
[M + 1 − CH2OH]+ (100), 69 (68); HRESIMS m/z 179.1044 [M +
Na]+ (calcd for C9H16NaO2, 179.1048); Rf 0.26 (n-hexane/EtOAc,
9:1, silica gel).
(S,E)-Methyl 3-(Tetrahydro-2H-pyran-2-yl)but-2-enoate (12a)
and (S,Z)-Methyl 3-(Tetrahydro-2H-pyran-2-yl)but-2-enoate (12b).
To a solution of diol 11 (1 g, 6.26 mmol) in a mixture of THF/H2O,
1:1 (42 mL), was added NaIO4 (2 g, 9.35 mmol), and the mixture was
stirred for 30 min. Once the reaction was finished, diethyl ether (20
mL) and H2O (20 mL) were added, the layers were separated, and the
aqueous one was saturated with NaCl powder, extracted with diethyl
ether (3 × 20 mL), dried over MgSO4, filtered, and concentrated.
NMR analysis of the crude revealed high purity of the yellowish oil
(605.8 mg, 76%), so it was immediately employed in the following
reaction without further purification.
((2R,3R)-3-Methyl-3-((S)-tetrahydro-2H-pyran-2-yl)oxiran-2-yl)-
methanol (14). Compound 14 was obtained using the same
procedure as in the case of diol 11, using alcohol 13a (41.7 mg,
0.27 mmol), DCM (2.7 mL), titanium(IV) isopropoxide (0.09 mL,
0.32 mmol), (−)-diethyl D-tartrate (0.06 mL, 0.37 mmol), and tert-
butyl hydroperoxide solution 5.06 M in isooctane (0.1 mL, 0.48
mmol). The final product was purified by column chromatography to
To a suspension of NaH (109.2 mg, 2.73 mmol) in THF (10 mL)
was slowly added, at 0 °C and under an argon atmosphere, a solution
of methyl 2-(bis(o-tolyloxy)phosphoryl)acetate17 (977.8 mg, 2.93
mmol) in THF (10 mL). After 10 min, a solution of the crude ketone
(250 mg, 1.95 mmol) in THF (10 mL) was added dropwise. The
mixture was allowed to warm for 2 h and then was quenched with
brine (40 mL). The layers were separated, and the aqueous layer was
extracted with diethyl ether (3 × 40 mL). The combined organic layer
was dried over MgSO4, filtered, and concentrated in vacuum, and then
the crude was purified by column chromatography. A 1.6:1 mixture of
the E-isomer/Z-isomer of the final esters (12a: 183.5 mg, 51%; 12b:
114.7 mg, 32%) was obtained, both as yellowish oils.
give the epoxy alcohol 14 (18.8 mg, 41%): yellow oil; [α]25 −7.3 (c
D
0.8, CHCl3); ATR-FTIR (neat) νmax 3424, 2931, 2852, 1441, 1379,
1
1263, 1205, 1087, 1031, 630 cm−1; H NMR (400 MHz, CDCl3) δ
3.97−4.04 (1H, m, H-6′), 3.82 (1H, dd, J = 12.1, 4.2 Hz, H-1), 3.70
(1H, dd, J = 12.6, 6.5 Hz, H-1), 3.37−3.45 (1H, m, H-6′), 3.13 (1H, t,
J = 5.4 Hz, H-2), 3.08 (1H, d, J = 10.9 Hz, H-2′), 2.13 (1H, br s, OH),
1.83−1.91 (1H, m, H-5′), 1.35−1.61 (5H, m, 2 × H-3′, 2 × H-4′, 1 ×
H-5′), 1.30 (3H, s, Me); 13C NMR (100 MHz, CDCl3) δ 81.6 (CH,
C-2′), 68.8 (CH2, C-6′), 62.6 (C, C-3), 61.1 (CH2, C-1), 59.8 (CH, C-
2), 27.2 (CH2, C-3′), 25.9 (CH2, C-4′), 23.3 (CH2, C-5′), 13.3 (CH3,
Me); EIMS m/z 172 [M]+ (1), 141 [M − CH2OH]+ (9), 112 (19), 85
(100); HRESIMS m/z 172.1102 [M]+ (calcd for C9H16O3, 172.1099);
Rf 0.16 (n-hexane/EtOAc, 6:4, silica gel).
12a: yellowish oil; 1H NMR (500 MHz, CDCl3) δ 5.94 (1H, t, J =
1.2 Hz, H-2), 4.03−4.07 (1H, m, H-6′), 3.70 (1H, d, J = 11.2 Hz, H-
2′), 3.68 (3H, s, CO2Me), 3.47 (1H, td, J = 11.5, 2.5 Hz, H-6′), 2.12
(3H, d, J = 1.2 Hz, H-4), 1.86−1.92 (1H, m, H-5′), 1.74−1.80 (1H, m,
H-3′), 1.49−1.61 (3H, m, 2 × H-4′, H-5′), 1.26−1.35 (1H, m, H-3′);
13C NMR (100 MHz, CDCl3) δ 167.7 (C, C-1), 159.1 (C, C-3), 114.1
(CH, C-2), 81.7 (CH, C-2′), 68.6 (CH2, C-6′), 51.1 (CH3, CO2Me),
719
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
((2S,3S)-3-Methyl-3-((S)-tetrahydro-2H-pyran-2-yl)oxiran-2-yl)-
methanol (15). Compound 15 was obtained using the same
procedure as in the case of diol 11, using alcohol 13a (41.7 mg,
0.27 mmol), CH2Cl2 (2.7 mL), titanium(IV) isopropoxide (0.09 mL,
0.32 mmol), (+)-diethyl D-tartrate (0.06 mL, 0.37 mmol), and tert-
butyl hydroperoxide solution 5.06 M in isooctane (0.1 mL, 0.48
mmol). The final product was purified by column chromatography to
yield the epoxy alcohol 15 (20.5 mg, 45%): yellow oil; [α]25D −21.9 (c
0.9, CHCl3); ATR-FTIR (neat) νmax 3421, 2937, 2854, 1442, 1380,
((2S,3R)-3-Methyl-3-((S)-tetrahydro-2H-pyran-2-yl)oxiran-2-yl)-
methyl 4-Bromobenzoate (18). To a solution of epoxy alcohol 17
(22.9 mg, 0.13 mmol) in dry CH2Cl2 (1.3 mL) was added sequentially,
at 0 °C and under an argon atmosphere, triethylamine (0.11 mL, 0.78
mmol), 4-bromobenzoyl chloride (116.2 mg, 0.52 mmol), and a
catalytic amount of DMAP. The reaction was allowed to warm for 1 h
and then was diluted with CH2Cl2 (1 mL) and quenched with brine (2
mL). Layers were separated, and the aqueous layer was extracted with
CH2Cl2 (3 × 2 mL). The combined organic layers was dried over
MgSO4, filtered, and concentrated, and the crude was purified by
chromatographic column, to yield the p-bromobenzoate 18 (43.3 mg,
92%) as a white solid. Pure monoclinic crystals were obtained from
1
1263, 1205, 1089, 1042, 630 cm−1; H NMR (400 MHz, CDCl3) δ
3.94−4.01 (1H, m, H-6′), 3.79 (1H, dd, J = 12.0, 3.8 Hz, H-1), 3.67
(1H, dd, J = 12.0, 6.3 Hz, H-1), 3.35−3.43 (1H, m, H-6′), 3.11 (1H, t,
J = 5.3 Hz, H-2), 3.01 (1H, d, J = 10.1 Hz, H-2′), 2.67 (1H, br s, OH),
1.82−1.89 (1H, m, H-5′), 1.37−1.67 (5H, m, 2 × H-3′, 2 × H-4′, 1 ×
H-5′), 1.25 (3H, s, Me); 13C NMR (100 MHz, CDCl3) δ 81.3 (CH,
C-2′), 68.8 (CH2, C-6′), 61.8 (CH, C-2), 61.7 (C, C-3), 61.0 (CH2,
C-1), 26.8 (CH2, C-3′), 26.0 (CH2, C-4′), 23.1 (CH2, C-5′), 13.0
(CH3, Me); EIMS m/z 155 [M − OH]+ (10), 141 [M − CH2OH]+
(1), 112 (30), 85 (100); HRESIMS m/z 155.1066 [M − OH]+ (calcd
for C9H15O2, 155.1072); Rf 0.16 (n-hexane/EtOAc, 6:4, silica gel).
((2R,3S)-3-Methyl-3-((S)-tetrahydro-2H-pyran-2-yl)oxiran-2-yl)-
methanol (16) and ((2S,3R)-3-Methyl-3-((S)-tetrahydro-2H-pyran-2-
yl)oxiran-2-yl)methanol (17). A flask with molecular sieves powder (4
Å) was flamed and then cooled to −20 °C. A solution of allylic alcohol
13b (60 mg, 0.39 mmol) in CH2Cl2 (3.9 mL) was added under an
argon atmosphere, followed by freshly distilled titanium(IV)
isopropoxide (0.15 mL, 0.45 mmol) and (−)-diethyl D-tartrate (0.09
mL, 0.54 mmol). After 30 min, a tert-butyl hydroperoxide solution 5.06
M in isooctane (0.15 mL, 0.69 mmol) was added under an argon
atmosphere and then stirred for 18 h. Once the reaction was finished,
15% tartaric acid(aq) (9 mL) was added and the mixture was
vigorously stirred at rt for 15 min. Then the aqueous layer was
extracted with CH2Cl2 (3 × 9 mL), and the combined organic layer
was concentrated under vacuum. The crude was redissolved in diethyl
ether (10 mL), cooled to 0 °C, and treated with previously cooled 15%
sodium hydroxide(aq) (10 mL) for 1 min. After that, layers were
separated, and the aqueous layer was extracted with diethyl ether (3 ×
10 mL). The combined organic layer was dried over MgSO4, filtered,
and concentrated under vacuum. After purification by column
chromatography, a 2:1 mixture of epoxyalcohol 16 (40.8 mg, 62%)
and epoxy alcohol 17 (20.4 mg, 31%) was obtained as an oil. The same
yield and proportion were obtained when (+)-diethyl L-tartrate was
used instead of (−)-diethyl D-tartrate.
crystallization in CH2Cl2/n-hexane, 9:1: white solid; [α]25 −29.9 (c
D
1.6, CHCl3); mp 78.3−80.5 °C; ATR-FTIR (neat) νmax 2941, 2855,
1
1725, 1590, 1264, 1173, 1087, 1007, 865, 753 cm−1; H NMR (400
MHz, CDCl3) δ 7.91 (2H, d, J = 8.1 Hz, Ar-H), 7.58 (2H, d, J = 8.1
Hz, Ar-H), 4.59 (1H, dd, J = 12.2, 3.4 Hz, H-1), 4.33 (1H, dd, J = 12.2,
7.2 Hz, H-1), 4.04 (1H, d, J = 11.6 Hz, H-6′), 3.44 (1H, t, J = 11.2 Hz,
H-6′), 3.21 (1H, t, J = 6.3 Hz, H-2′), 3.14 (1H, dd, J = 7.0, 3.7 Hz, H-
2), 1.84−1.92 (1H, m, H-5′), 1.46−1.63 (5H, m, 1 × H-3′, 2 × H-4′, 2
× H-5′), 1.37 (3H, s, Me); 13C NMR (100 MHz, CDCl3) δ 165.8 (C,
C-1″), 131.9 (C, 2 × Ar), 131.4 (CH, 2 × Ar), 128.7 (C, Ar), 128.6
(C, Ar), 79.4 (CH, C-2′), 68.7 (CH2, C-6′), 63.5 (CH, C-2), 62.7 (C,
C-3), 60.5 (CH2, C-1), 28.4 (CH2, C-3′ or C-4′), 25.8 (CH2, C-3′ or
C-4′), 23.3 (CH2, C-5′), 17.6 (CH3, Me); EIMS m/z 356 [M]+ (1),
155 [M − OC(O)C6H4Br]+ (22), 141 [M − CH2OC(O)C6H4Br]+
(9), 85 (100); HRESIMS m/z 356.0444 [M]+ (calcd for C16H19BrO4,
356.0446); Rf 0.28 (n-hexane/EtOAc, 9:1, silica gel).
Computational Methods. Molecular mechanics conformational
searches were undertaken using the Macromodel software (version 8.5,
Schrodinger Inc.) and the MMFF94 force field.31 Solvation effects of
̈
CHCl3 were simulated using the generalized Born/surface area
(GBSA) solvation model. Extended nonbonded cutoff distances (van
der Waals cutoff of 8.0 Å and an electrostatic cutoff of 20.0 Å) were
used. All local minima within 10 kJ of the global minimum were saved,
and the analysis of the results was undertaken using Maestro software.
Quantum mechanical calculations were carried out using the Jaguar
package (Jaguar; Schrodinger LLC). Single-point energy calculations
̈
were performed at the DFT theoretical level in the gas phase. The
B3LYP hybrid functional with the LACVP**+ basis set was used.
Chemical shifts were calculated from their shielding constants that
were first averaged according to their relative Boltzmann populations
using a Schrodinger Inc. Python script. Chemical shifts were calculated
̈
using the gauge-including atomic orbital (GIAO) method. Proton
chemical shifts for each methyl group were averaged due to their
conformational freedom.
16: yellowish oil; [α]25 −2.3 (c 0.1, CHCl3); [α]25 −5.2 (c 0.7,
D
D
(CH3)2CO); ATR-FTIR (neat) νmax 3417, 2938, 2854, 1441, 1380,
1
1261, 1204, 1086, 1039, 656 cm−1; H NMR (400 MHz, CDCl3) δ
Diatom Growth Inhibition. Three strains of benthic diatoms,
Phaeodactylum tricornutum, Cylindrotheca sp., and Navicula cf. salinicola
BEA0055, were used to test the effect of the compounds on microalgal
growth. Diatoms were cultured at 19 1 °C in Erlenmeyer flaks (250
mL) containing 150 mL of Guillard’s F/2 medium and subjected to a
photoperiod of 18:6 (L:D). Tests were run in 48-well plates. Inocula
were prepared by adjusting the diatom concentration to (1−2) × 106
cells/mL using a Neubauer chamber. Test products were dissolved in
F/2 medium (500 μL) to which diatom inocula (500 μL) were added.
Plates were incubated under the above-mentioned conditions for 5
days, and then chlorophyll-a (Chla) was extracted. In order to extract
Chla, the plates were centrifuged at 5500 rpm for 15 min. The
supernatants were discarded, and 200 μL of DMSO was added to the
wells. Then, the content of each well was transferred to a 96-well plate,
3.99−4.02 (1H, m, H-6′), 3.89 (1H, dd, J = 11.4, 4.9 Hz, H-1), 3.66
(1H, dd, J = 11.6, 7.3 Hz, H-1), 3.42−3.47 (1H, m, H-6′), 3.20 (1H, d,
J = 10.7 Hz, H-2′), 3.06 (1H, t, J = 6.2 Hz, H-2), 2.48 (1H, br s, OH),
1.88−1.96 (1H, m, H-5′), 1.47−1.74 (5H, m, 2 × H-3′, 2 × H-4′, 1 ×
H-5′), 1.32 (3H, s, Me); 13C NMR (100 MHz, CDCl3) δ 79.4 (CH,
C-2′), 69.1 (CH2, C-6′), 63.6 (CH, C-2), 61.26 (C, C-3), 61.23 (CH2,
C-1), 27.3 (CH2, C-3′), 26.0 (CH2, C-4′), 23.1 (CH2, C-5′), 18.0
(CH3, Me); EIMS m/z 172 [M]+ (1), 141 [M − CH2OH]+ (82), 112
(24), 85 (100); HRESIMS m/z 172.1106 [M − OH]+ (calcd for
C9H16O3, 172.1099); Rf 0.39 (n-hexane/EtOAc, 6:4 (eluted twice),
silica gel).
17: yellowish oil; [α]25 −34.1 (c 0.9, CHCl3); ATR-FTIR (neat)
D
νmax 3425, 2937, 2853, 1441, 1379, 1259, 1205, 1086, 1039, 630 cm−1;
1H NMR (400 MHz, CDCl3) δ 3.99−4.06 (1H, m, H-6′), 3.78 (2H, d,
J = 5.6 Hz, H-1), 3.39−3.47 (1H, m, H-6′), 3.25 (1H, t, J = 6.2 Hz, H-
2′), 2.99 (1H, t, J = 5.6 Hz, H-2), 2.13 (1H, br s, OH), 1.83−1.91 (1H,
m, H-5′), 1.44−1.62 (5H, m, 1 × H-3′, 2 × H-4′, 2 × H-5′), 1.34 (3H,
s, Me); 13C NMR (100 MHz, CDCl3) δ 79.7 (CH, C-2′), 68.8 (CH2,
C-6′), 63.6 (CH, C-2), 63.3 (C, C-3), 60.8 (CH2, C-1), 28.3 (CH2, C-
3′), 25.9 (CH2, C-4′), 23.4 (CH2, C-5′), 18.2 (CH3, Me); EIMS m/z
172 [M]+ (1), 141 [M − CH2OH]+ (100), 112 (16), 85 (84);
HRESIMS m/z 172.1104 [M]+ (calcd for C9H16O3, 172.1099); Rf 0.29
(n-hexane/EtOAc, 6:4 (eluted twice), silica gel).
and the amount of Chla was determined spectrophotometrically.32
A
pathlength correction for the DMSO extracts was applied.33
Inhibition of Macroalgal Spore Germination. Gayralia oxy-
sperma specimens were collected from the intertidal zone at El
́
Medano beach, Tenerife, Spain. Spores were released in Von Stosch
Solution (VSS) by the osmotic method.34,35 Bioassays were conducted
in flat-bottom 96-well plates as described by Chambers et al. with
slight modifications.36 Each well was filled with 50 μL of the
appropriate dilution of the products in VSS to which 50 μL of spore
inoculum (1.6 × 105 spores/mL) was added. Plates were incubated at
720
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721

Journal of Natural Products
Article
19
1 °C for 6 days under a L:D 18:6 photoperiod. After the
(14) Napolitano, J.; Gavín, J.; García, C.; Norte, M.; Fernan
́
dez, J. J.;
incubation time, the bottom of each well was inspected for the
presence of germinated spores with an inverted microscope. A spore
was considered to be germinating if the germ tube was visible. The
MIC was recorded as the lowest concentration inhibiting spore
germination. Three replicates were prepared for each compound and
test concentration. Eight serial 2-fold dilutions, from 100 to 0.7 μM,
were assayed for each compound.
Daranas, A. H. Chem.Eur. J. 2011, 17, 6338−6347.
(15) Meissner, A.; Sorensen, O. W. Magn. Reson. Chem. 2001, 39,
49−52.
(16) Aliev, A. E.; Zakirin, A. M.; Busson, M. J. M.; Fitzmaurice, R. J.;
Caddick, S. J. Org. Chem. 2012, 77, 6290−6295.
(17) Katsuki, T.; Martín, V. S. In Organic Reactions; Paquette, L. A.,
et al., Eds.; Wiley & Sons: New York, 1996; Vol. 48, pp 1−299.
(18) Hayashi, Y.; Kanayama, J.; Yamaguchi, J.; Shoji, M. J. Org. Chem.
2002, 67, 9443−9448.
ASSOCIATED CONTENT
■
(19) Ando, K. J. Org. Chem. 1997, 62, 1934−1939.
S
* Supporting Information
́
(20) Cen-Pacheco, F.; Rodríguez, J.; Norte, M.; Fernandez, J. J.;
General experimental procedures, NMR and X-ray crystallo-
graphic data, as well as details on computational calculation
results. This material is available free of charge via the Internet
at http://pubs.acs.org.
Daranas, A. H. Chem.Eur. J. 2013, 19, 8525−8532.
(21) Domínguez, H. J.; Crespín, G. D.; Santiago-Benítez, A. J.; Gavín,
J. A.; Norte, M.; Fernan
́
dez, J. J.; Daranas, A. H. Mar. Drugs 2014, 12,
176−192.
(22) Kolossvar
1684.
́
y, I.; Guida, W. C. J. Comput. Chem. 1999, 20, 1671−
AUTHOR INFORMATION
■
(23) Lodewyk, M. W.; Siebert, M. R.; Tantillo, D. R. Chem. Rev.
2012, 112, 1839−1862.
Corresponding Authors
*E-mail (J. J. Fernan
́
dez): jjfercas@ull.es.
(24) Rassolov, V. A.; Ratner, M. A.; Pople, J. A.; Redfern, P. C.;
Curtiss, L. A. J. Comput. Chem. 2001, 22, 976−984.
(25) Saielli, G.; Nicolaou, K. C.; Ortiz, A.; Zhang, H.; Bagno, A. J.
Am. Chem. Soc. 2011, 133, 6072−6077.
*E-mail (A. H. Daranas): adaranas@ull.es.
Notes
The authors declare no competing financial interest.
(26) Smith, S. G.; Goodman, J. M. J. Org. Chem. 2009, 74, 4597−
4607.
ACKNOWLEDGMENTS
■
(27) Briand, J.-F. Biofouling 2009, 25, 297−311.
(28) Lane, A. L., Kubanek, J. Secondary Metabolite Defenses against
Pathogens and Biofoulers. In Algal Chemical Ecology; Amsler, C. D.,
Ed.; Springer-Verlag: Berlin, 2008; pp 229−243.
(29) Hayashi, Y.; Kanayama, J.; Yamaguchi, J.; Shoji, M. J. Org. Chem.
2002, 67, 9443−9448.
(30) Orsini, F.; Pelizzoni, F. J. Org. Chem. 1980, 45, 4726−4727.
(31) Macromodel, version 10.0, and Jaguar, version 8.0; Schrodinger,
LLC: New York, NY, 2013.
(32) Wellburn, A. R. J. Plant Physiol. 1994, 144, 307−313.
(33) Warren, C. R. J. Plant Nutr. 2008, 31, 1321−1332.
(34) Fletcher, R. L. Int. Biodeterior. 1989, 25, 407−422.
(35) Callow, M. E.; Callow, J. A.; Pickett-Heaps, J. D.; Wetherbee, R.
J. Phycol. 1997, 33, 938−947.
(36) Chambers, L. D.; Hellio, C.; Strokes, K. R.; Dennington, S. P.;
Goodes, L. R.; Wood, R. J. K.; Walsh, F. C. Int. Biodeterior. Biodegrad.
2011, 65, 939−946.
This research was funded by SAF2011-28883-C03-01 and
CTQ2011-28417-C02-01/BQU from MINECO and CEI10/
00018 from MECyD, Spain, as well as by EU Grants FP7-
KBBE-3-245137-MAREX and FP7-REGPOT-2012-CT2012-
31637-IMBRAIN. A.J.M.R. holds a 2+2 contract from
PLOCAN. S.J.A.-M. thanks the Spanish MECyD for an FPU
grant. A.J.S.B. thanks the CajaCanarias Foundation for a grant.
̈
́
The authors are grateful to Dr. M. Sanson, Department of
Vegetal Biology, ULL, for her valuable help in the collection
and identification of G. oxysperma specimens, as well as for
insightful comments about algal biology and experimental
procedures with zoospores.
REFERENCES
■
(1) Fernan
́
dez, J. J.; Souto, M. L.; Norte, M. Nat. Prod. Rep. 2000, 17,
235−246.
(2) Blunt, J. W.; Copp, B. R.; Keyzers, R. A.; Munro, M. H. G.;
Prinsep, M. R. Nat. Prod. Rep. 2014, 31, 160−258.
(3) Cragg, G. M.; Grothaus, P. G.; Newman, D. J. Chem. Rev. 2009,
109, 3012−3043.
(4) Cragg, G. M.; Grothaus, P. G.; Newman, D. J. J. Nat. Prod. 2014,
77, 703−723.
(5) Cen-Pacheco, F.; Villa-Pulgarin, J. A.; Mollinedo, F.; Norte, M.;
Daranas, A. H.; Fernan
3308.
(6) Cen-Pacheco, F.; Villa-Pulgarin, J. A.; Mollinedo, F.; Norte, M.;
Fernandez, J. J.; Daranas, A. H. Mar. Drugs 2011, 9, 2220−2235.
(7) Cen-Pacheco, F.; Nordstrom, L.; Souto, M. L.; Norte, M.;
́
dez, J. J. Eur. J. Med. Chem. 2011, 46, 3302−
́
̈
Fernan
(8) Cen-Pacheco, F.; Mollinedo, F.; Villa-Pulgarin, J. A.; Norte, M.;
Fernandez, J. J.; Daranas, A. H. Tetrahedron 2012, 68, 7275−7279.
(9) Gonzalez, A. G.; Arteaga, J. M.; Fernandez, J. J.; Martín, J. D.;
́
dez, J. J.; Daranas, A. H. Mar. Drugs 2010, 8, 1178−1188.
́
́
́
Norte, M.; Ruano, J. Z. Tetrahedron 1984, 40, 2751−2755.
(10) Caserta, T.; Piccialli, V.; Gomez-Paloma, L.; Bifulco, G.
Tetrahedron 2005, 61, 927−939.
(11) Smith, S. G.; Goodman, J. M. J. Am. Chem. Soc. 2010, 132 (37),
12946−12959.
(12) Bifulco, G.; Dambruoso, P.; Gomez-Paloma, L.; Riccio, R. Chem.
Rev. 2007, 107, 3744−3779.
(13) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.;
Tachibana, K. J. Org. Chem. 1999, 64, 866−876.
721
DOI: 10.1021/np5008922
J. Nat. Prod. 2015, 78, 712−721