The Configuration of Distaminolyne A is S: Quantitative Evaluation of Exciton Coupling Circular Dichroism of N, O- Bis-arenoyl-1-amino-2-alkanols

Article abstract of DOI:10.1021/acs.jnatprod.8b00937

The 2S configuration of the marine natural product distaminolyne A was recently disputed based upon total synthesis, yet paradoxically supported by a second independent total synthesis from a different research group. We now verify the 2S configuration of

Full text of DOI:10.1021/acs.jnatprod.8b00937

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
The Configuration of Distaminolyne A is S: Quantitative Evaluation
of Exciton Coupling Circular Dichroism of N,O- Bis-arenoyl-1-amino-
2-alkanols
Tadeusz F. Molinski,*,†,‡ Mariam N. Salib,† A. Norrie Pearce,§ and Brent R. Copp§
†Department of Chemistry and Biochemistry and ‡Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California,
San Diego, La Jolla, California 92093-0358, United States
§School of Chemical Sciences, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand
S
* Supporting Information
ABSTRACT: The 2S configuration of the marine natural product
distaminolyne A was recently disputed based upon total synthesis, yet
paradoxically supported by a second independent total synthesis from a
different research group. We now verify the 2S configuration of
distaminolyne A by extensive chiroptical studies and support the veracity
of the EC ECD method originally used to prove it. The origin of the
apparent paradox appears to lie in the limits of precision of polarimetry in
the context of weakly rotatory molecules, which strikes a cautionary note
on the reliability of “reassignment” of natural product configurations based
solely on specific rotation.
n 2016, two of us reported a new diyne-containing amino
Ialcohol, distaminolyne A (1a), from a New Zealand
ascidian, Pseudodistoma opacum, along with an unrelated N-
hydroxy-β-carboline.1 Compound 1a is a member of a limited
family of linear, long-chain aminoalkanol (LCAA)2 natural
products that are associated with marine invertebrates, mostly
sponges within the genera Haliclona, Leucetta, Rhizochalina,
and Oceanapia and several genera of ascidians, including
Didemnum, Pseudodistoma,3 and Clavelina4 or, in rare
exceptions, algae and bivalves.5 LCAAs exhibit a range of
biological properties, perhaps due to their structural resem-
blance to sphingosine (2a) and the versatile mammalian
“second messenger”, D-sphingosine 1-O-phosphate (2b). For
example, most recently leucettamols A and B, originally
discovered by the Faulkner group,6 were identified by
Fattorusso and co-workers as nonelectrophilic activators of
TRPA1 and inhibitors of TRPM8 channels with micromolar
potency.7
The configurations of LCAA natural products are highly
heterogeneous. The simplest examples, 1-amino-2-alkanols,
have been found in both R and S configurations, while more
complex “two-headed” sphingolipids, such as rhizochalin8 and
specific rotations of aminoalkanols, creative solutions for their
oceanapiside,9 bearing four stereogenic centers, occur in
assignment have been developed and deployed. The
almost all of the 16 possible stereopermutations.10 Notori-
configuration of 1a was solved by a variant of the exciton
coupling (EC) “dibenzoate method” in electronic circular
dichroism (ECD) spectra pioneered by Nakanishi and
ously, 1-amino-2-alkanols are weakly rotatory:11 being either
levorotatory or dextrorotatory with no simple correlation to
the C-2 configuration (Table 1).12 Consequently, the problem
of assignment of configuration of LCAAs is not trivial and must
be approached on a case-by-case basis. Because of the small
Received: November 5, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. Specific Rotations of Natural and Synthetic
Products
a
cmpd
[α]D
c
solvent
ref
c
(S)-1a
−1
+1.9
+0.8
+1.2
−2.0
−0.9
−5.0
−1.4
+5.3
0.44
0.36
1.0
0.1
0.1
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
CHCl3
MeOH
CHCl3
MeOH
CHCl3
MeOH
CHCl3
MeOH
CHCl3
1
d
(R)-3a
15
17
17
17
17
b
(S)-1a
(S)-1a
c
b
(R)-1a
(R)-1a
c
0.1
0.1
de
,
(S)-1a
(S)-4a
(S)-4a
(S)-4a
(S)-8a
(S)-8a
18
h
b
1.76
1.78
2.46
1.02
0.45
1.76
1.72
2.30
2.02
2.82
2.82
0.55
0.43
dfg
, ,
h
h
h
h
h
h
h
h
h
h
h
h
c
+5.2
+28.3
+17.8
−26.1
−19.3
+11.5
−7.6
−12.9
+8.6
(R)-8b
(R)-8b
(S)-9a
Figure 1. Newman diagrams of the three major conformers of N,O-
acylated 1-amino-2-alkanols, (a) 2S enantiomer and (b) 2R
enantiomer, showing helicities (+, −, or 0) and predicted signs of
split Cotton effects in exciton-coupled ECD. Shaded box represents
Bz or 2-naphthoyl, and the dominant conformer is iii.
(S)-9a
(R)-9b
(R)-9b
(R)-10b
(R)-10b
−88.9
−104
a
CIP designation and values are taken, as reported, from original
references. Concentration, c g/100 mL. Solvents were spectroscopic
distaminolyne A should be revised from 2S to 2R. A second
asymmetric synthesis of (−)-1a ([α]D −5.0 (c 0.1, MeOH)),
published in the same year by Krishna and co-workers,18
contradicted the Guo result, “since our values matched with
reported one (sic)”, referring to the original Copp assignment.1
Introduction of asymmetry in both syntheses employed
different, but reliable catalytic methodology, and the
configurations of the synthetic samples of (−)- and (+)-1a
are not in doubt. Yet, the conclusions of all three studies are
inconsistent, and the true configuration of natural 1a is
brought into question.
Collectively, the [α]D values measured and reported for 1a
span the range of −5 to +2, with little correlation with
configuration; we concur with Krishna that, “inconsistencies
might be due to the low optical rotational values”.18 Most
importantly, neither of the two synthetic groups attempted to
address how their results could be consistent with the EC ECD
method used to assign (S)-1a1 and, by corollary, (R)-3a.15
Being sufficiently stimulated by this paradox, and in
confirming an unambiguous assignment of natural 1a, we
reexamined our EC ECD method by undertaking a chiroptical
study of enantiomeric pairs of synthetic, configurationally well-
defined LCAAs and their corresponding N,O-acylates and
matched these against the claims of configurational “reassign-
ment”. We can now state, without doubt, that the configuration
of 1a is, indeed, 2S as originally reported. In addition, we
conjecture that the origin of the paradox lies not in a need for
reassignment of 1a, but within instrumental limits of precision
of modern polarimetry and the low values of [α]D observed for
weakly rotatory molecules. A timely cautionary note is offered
for configurational assignment with sole reliance upon specific
rotations and, in particular, the attendant risks in facile claims
of “reassignment” based on the same.
b
c
grade MeOH or CHCl3 (stabilized with amylenes). Free base. TFA
d
e
salt. HCl salt. A slightly different value, [α]D −6.7 (c 0.1, MeOH), is
reported for synthetic (S)-1a·HCl in the Supporting Information.18
f
g
Replicate measurements, n = 10 (standard deviation, σ 1.4). For
[α] values of (S)-4a·HCl at different wavelengths, see Experimental
h
Section. This work.
Harada.13,14 Briefly, the helicity, or “twist”, of proximal
dibenzoate charge transfer (CT) electronic transition dipole
moments is correlated with a split-ECD spectrum due to
Davydov coupling. Positive dipole helicity manifests a large
maximum at a longer wavelength and a minimum at shorter
wavelength; both are centered approximately at the λmax of the
benzoate chromophore.
A similar analysis can be applied to benzoate−benzamide
chromophore pairs (Figure 1).13,14 The dominant gauche
rotameridentified by conformational analysis of the vicinal
benzoate−benzamide of (R)-3b, an N,O-dibenzoyl derivative
of the marine-derived (R)-3a, reported from an Australian
Didemnum sp. in 199315subtends negative helicity of the
benzoyl electronic transition dipole moments (cf. Figure 1(b),
iii) and manifests EC through a negative-split ECD spectrum.
For N,O-dibenzoyldistaminolyne A (1b), Copp and co-
workers measured an antipodal positive-split ECD spectrum
corresponding to a positive helicity (cf. Figure 1(a), iii),
thereby assigning the 2S configuration for the natural product
1a.1 Since then, EC ECD of the vicinal benzoate−benzamide
method has been applied successfully to other marine-derived
1-amino-2-alkanols.16
Recently, two independent reports of asymmetric synthesis
of 1a have appeared, one of which appears to contradict the
forgoing configurational assignment. In 2017, Guo and co-
workers reported the asymmetric syntheses of free bases of
both (−)-(R)-1a and (+)-(S)-1a ([α]D −2.0 and +0.9,
respectively; see Table 1) along with their trifluoroacetic acid
(TFA) salts.17 Guo, interpreting the specific rotation reported
for natural 1a by Copp and co-workers ([α]D −1 (c 0.44,
MeOH)), concluded that the absolute configuration (AC) of
RESULTS AND DISCUSSION
■
Distaminolyne A (1a) contains a conjugated 10,11-diyne
(UV−vis, λmax 215, 240, 253, 267, 283 nm). Based on prior
observations,19 this chromophore was deemed too remote
from the vicinal amino alcohol moiety to interfere with EC
B
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Scheme 1. Synthesis of (S)- and (R)-1-Amino-2-decanols (4a,b) and Selected Acylates (8−10)
ECD at the terminal benzoate−benzamide pair of 1b. To
simplify our analysis, we chose to synthesize S and R 1-amino-
2-decanol (4a and 4b, Scheme 1), by a well-defined
asymmetric methodology, to remove the component of
conformational analysis that was critical in proposing the AC
of 3a,b.15 Compounds 4a,b served as common platforms for
the preparation of N,O-acylates for critical examination of the
characteristics of both [α]D and EC ECD. 1-Decene was
epoxidized (m-CPBA, (CH2Cl)2, radical inhibitor,20 reflux,
96%), affording ( )-5, which was subjected to kinetic
hydrolytic resolution (H2O, AcOH) in the presence of
Jacobsen’s catalyst ((R,R)-Co(salen) complex 621 (0.48 mol
%)) to give unreacted, optically pure 1,2-epoxydecane
[(+)-(R)-5b, 41%, >97% ee, [α]D +14.8 (c 1.12, Et2O)].
The procedure was repeated with ( )-5 and (S,S)-6 to secure
optically pure (S)-5a ([α]D −12.3 (c 1.12, Et2O); lit.22 −12.9
(c 1.17, Et2O), lit.23 −13.9 (c 1.2, Et2O)) and the
corresponding diol (R)-7b. Each epoxide 5a and 5b was
subjected to ammoniolysis (∼20 w/v% NH4OH(aq), 80%
EtOH, 60 °C) to give (−)-(S)-4a and (+)-(R)-4b, respectively
(>90%). The latter amino alcohols were used without further
purification for the subsequent reactions to deliver the
following pairs of enantiomeric acylates after purification by
Figure 2. ECD spectra (MeOH, 23 °C) of (−)-(R)-8b (blue, solid
line) and (+)-(S)-8a (red, dashed line).
silica flash chromatography: (BzCl, dry pyridine, 60 °C)24
N,O-dibenzoyl derivatives, (+)-(S)-8a and (−)-(R)-8b, and
(Ac2O, pyridine, 23 °C) N,O-acetyl derivatives (S)-9a and (R)-
9b, respectively. Finally, (+)-(R)-4b was converted (2-
naphthoyl chloride, DMAP, pyridine, 60 °C, 9%) to the bis-
chromophoric derivative (−)-(R)-10b for comparison pur-
poses.
237 (−10.2)]15 were the expected bisignate curvesopposite
Cotton effects at long and short wavelengthsthat were
matched in sign and magnitude, proving that both the original
conformational and chiroptical analysis of the EC ECD of (R)-
3b were correct.15 Moreover, comparison of the ECD spectra
of (−)-(R)-8b and 1b1 again showed the two molecules were
The ECD spectra (Figure 2, Table S1) of (−)-(R)-8b [λ 221
nm (Δε +3.7); 238 (−7.7)] and (R)-3b [λ 221 nm (Δε +5.4);
C
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
specific rotations (e.g., (S)-4a·HCl), these errors can be
mitigated.
Both reported syntheses of 1a utilized well-established
asymmetric methodologies, and configurational assignments of
the synthetic products are not in doubt.17,18 Indeed, Guo
prepared both enantiomers of 1a and verified their
configurations using Riguera’s MPA double derivatization
method26 and the modified Mosher’s method.27 Both groups,
however, recorded low-magnitude specific rotation values that
either refute (Guo)17 or confirm (Krishna)18 the proposed
configuration of the natural product 1a.1 Unfortunately neither
group took the opportunity to prepare the corresponding N,O-
dibenzoyl derivatives, which would have permitted comparison
with 1b by EC ECD as reported in the original isolation
paper.1
Regardless of the errors and level of uncertainty associated
with the use of specific rotation values to assign configuration
to 1-amino-2-alkanols, what is unmistakable is that the EC
ECD spectrum of 1b clearly correlates with 8a and the 2S
configuration.28 This current study also highlights that N,O-
dinaphthoyl derivative 10b demonstrates superior sensitivity
when applied to the EC ECD of 1-amino-2-alkanols. We
estimated the limits of detection (LOD) of the test cases 8b
and 10b for unambiguous assignment by EC ECD to be ∼5.7
and ∼0.38 nmol, respectively (Figure S25).29
Figure 3. ECD spectra (MeOH, 23 °C) of (−)-(R)-10b (blue, solid
line) and (−)-(R)-8b (red, dashed line).
It should be mentioned that other chiroptical methods for
assignment of AC have risen, of late, in popularity and
reliability, namely, vibrational circular dichroism (VCD)30 and
Raman optical activity (ROA),31 combined with quantum
chemical calculations reliant upon time-dependent density
functional theory (TDDFT). The advantages of both include
assignments based on “fingerprint” vibrational transitions,
independent of the presence of UV−vis chromophores, but at
the disadvantage of much lower sensitivity than EC ECD.
Finally, the nonempirical EC ECD method is of superior
reliability compared to single-point measurements, such as [α],
for several reasons, but mostly because the method is
exceptionally sensitive and absolute configurational assign-
ments are supported by the well-developed quantum chemical
theory of exciton coupling in the context of the dibenzoate
method.13
In conclusion, the configuration of distaminolyne A (1a) is
upheld as 2S by independent verification of the EC ECD
method with standard compounds of defined configuration.32
This suggests that “reassignment” of 1a to 2R is unwarranted
and in error, likely due to misinterpretation of the low rotatory
strengths of unmodified 1-amino-2-alcohols and measurements
of [α]D that lie within random error. Caution should be
applied to claims of reassignment of natural product
configurations when specific rotations lie within the latter
regime.
antipodal and upheld the original 2S assignment for 1a,b1 and
3a,b.15
Measurement of the ECD of compound (R)-10b revealed a
split-Cotton effect of the same sign as (R)-8b, but of much
larger magnitude [λ 229 nm (Δε −91.3); 242 (−91.3], as
expected from the stronger CT band of the 2′-naphthoyl
chromophore. The larger A value may also be a consequence of
tighter conformational restraints on the favored gauche
conformation imposed by the bulkier 2′-naphthoyl groups.
The optical rotations of 1-amino-2-decanols and their
acylates (Table1) are informative. When measured in
MeOH, the specific rotation of amino alcohol (S)-4a, like
that of 1a, is of low magnitude;25 however, those of 8−10
(MeOH) are larger and reliably measured. The R-configured
benzoyl or naphthoyl derivatives 8b and 10b are uniformly
levorotatory and, conversely, their S enantiomers are dextro-
rotatory. In MeOH, acetyl derivative (S)-9a is dextrorotatory,
but it is noteworthy that when measured in CHCl3, the sign of
[α]D of 9a inverts, but the signs of [α]D for the pairs 8a,b and
10a,b remain the same in both solvents. Generally, character-
ization of LCAAs by specific rotations of their acylated
derivatives is more reliable and reproducible than [α]Ds of the
free bases or salts.
A brief error analysis of measurements of [α]D on optically
pure samples 8−10 was conducted to identify the sources of
random error that propagate in calculations of low [α]D values
(see Figure S24). In low-rotatory samples, such as natural 1a
and synthetic 4a (free base, Table 1), the largest source of
error was α of the solvent blank, the value of which approaches
the sample α at close to the limit of detection of most modern
polarimeters, and should be considered within the range of
error. For example, back-calculation of α (Figure S24) from
the low values of [α]D of synthetic 1a reported (Table 1) by
Guo17 and Krishna18 leads to the same conclusion, only here,
with lower values for c, the errors must be even greater. When
samples of higher concentration are available, or with larger
EXPERIMENTAL SECTION
■
General Experimental Procedures. Spectrometric measure-
ments (UV, ECD) were measured in spectroscopic grade MeOH
(Fisher), CHCl3 (Acros, stabilized with amylene), or 2,2,2-
trifluoroethanol (TFE, Acros). Specific rotations were measured on
a Jasco P-2000 digital polarimeter at the D-line (Na0 emission) or an
Autopol IV at shorter wavelengths, with appropriate filters, in quartz
cells of path length 0.100 or 0.500 dm at 23 °C. UV−vis spectra were
measured on a Jasco V-630 spectrometer in a 1.00 or 0.20 cm quartz
cuvette. FTIR spectra were collected on thin film samples using a
Jasco FTIR-4100 spectrometer fitted with an ATR accessory (ZnSe
plate). ECD spectra were measured on a Jasco J810 spectropolarim-
D
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
eter on samples in 1.00 or 5.00 mm quartz cells. Inverse detected 2D
NMR spectra were measured on a Jeol ECA (500 MHz)
pentane): [α]23 −14.4 (c 3.06, Et2O); lit.22 −12.9 (c 1.17, Et2O),
D
lit.23 −13.9 (c 1.2, Et2O).
1
spectrometer, equipped with a 5 mm H{13C} probe, or a Bruker
Diol (+)-(R)-7b. Recrystallized from Et2O/pentane, mp 58−59 °C;
[α]23D +10.5 (c 2.92, MeOH); 1H NMR (500 MHz, CD3OD) δ 3.56
(1H, m, H-2), 3.46 (1H, dd, J = 11.0, 4.5 Hz, H-1a), 3.42 (1H, dd, J =
11.0, 6.5 Hz, H-1b), 3.31 (1H, m, H-2), 1.48 (2H, m, H2-3), 1.32
(12H, m), 0.90 (3H, t, J = 6.8 Hz, H3-10); 13C NMR (125 MHz,
CD3OD) δ 73.3 (CH, C-2), 67.4 (CH2, C-1), 34.5 (CH2), 33.1
(CH2), 30.9 (CH2), 30.8 (CH2), 30.5 (CH2), 26.7 (CH2), 23.8
(CH2), 14.5 (CH3, C-10); HR-ESIMS m/z 173.1548 [M − H]−
Avance III (600 MHz) NMR spectrometer with a 5 mm 1H{13C,15N}
1
cryoprobe or 1.7 mm H{13C} microcryoprobe. 13C NMR spectra
were collected on a Varian NMR spectrometer (125 MHz) equipped
with a 5 mm Xsens 13C{1H} cryoprobe. NMR spectra were referenced
to residual solvent signals, CDCl3 (δH 7.26, δC 77.16). High-
resolution ESITOF analyses were carried out on an Agilent 1200
HPLC coupled to an Agilent 6350 TOFMS. Low-resolution MS
measurements were made on a Thermoelectron Surveyor UHPLC
coupled to an MSD single-quadrupole detector. A semipreparative
and chiral-phase HPLC system comprising dual Jasco pumps (PU-
2086) and a UV−vis detector (UV-2075) or UV-CD detector
(CD2095) in tandem with an ELSD detector (Softa-A model 300).
Flash column chromatography was carried out with silica (Silicycle,
particle size 40−63 μm) using HPLC grade solvents. Pyridine was
distilled from CaH2 under N2. Commercial benzoyl chloride was
washed with ice-cold 5% NaHCO3, dried over anhydrous CaCl2, and
fractionally distilled under reduced pressure (32 mmHg). 2-
Naphthoyl chloride was prepared by refluxing 2-naphthoic acid in
SOCl2 (cat. DMF, 1 h). After removal of volatiles, the residue was
−
(calcd for C10H21O2 173.1547).
Epoxide (+)-(R)-5b. A sample of ( )-5 was resolved with
Jacobsen’s catalyst, (R,R)-Co(salen) (6), according to the published
procedure21 (3 °C, 21 h) to give optically enriched (R)-5b, which was
distilled under reduced pressure (90 °C, 120 mmHg). The distillate
was subjected to flash chromatography, as described above for ( )-5a,
to give pure (+)-(R)-5b (41%) as a clear oil ([α]D +14.8 (c 1.12,
1
Et2O). H and 13C NMR matched those of ( )-5.
Amino Alcohol (S)-4a. A solution of epoxide (−)-(S)-5a (228
mg, 1.45 mmol) in EtOH (7.0 mL) and concentrated NH4OH (7.0
mL) was heated at 60 °C for 17 h and worked up, as described below
for (R)-4b, to obtain (S)-4a (218 mg, 86%), which was used in the
subsequent acylation steps without further purification. A sample
(30.8 mg) of (S)-4a was repurified by flash chromatography on a
“pipet column” (5 × 80 mm, SiO2, 1:9 MeOH/CH2Cl2/0.7 M NH3)
to give the free base as a pale yellow solid (6.9 mg): [α]D −1.4 1.4
(c 1.76, MeOH). A sample of (S)-4a (6.9 mg) was converted to the
hydrochloride salt by dissolving in 1 M HCl in MeOH (1.8 mL) and
removing the volatiles under reduced pressure, followed by re-
evaporation from CHCl3 (×2) to give (S)-4a·HCl (8.8 mg): [α]D
distilled (Kugelrohr) under reduced pressure (32 mmHg) before use.
̈
N,O-Dibenzoylation of Distaminolyne A. Method A. A
solution of 1a (17.5 mg, 46.6 μmol) and DMAP (1 crystal) was
dissolved in dry pyridine (2.0 mL), treated with benzoyl chloride (17
μL, 26.2 mg, 187 μmol, 4.0 equiv), and heated at 60 °C under
nitrogen for 21 h. The volatiles were removed under high vacuum,
and the residue was dissolved in 1:4 EtOAc/hexanes and filtered
through a plug of basic alumina. The fraction containing 1b (6.5 mg,
TLC) was concentrated and subjected to semipreparative HPLC
(silica, 10 × 250 mm, 5 μ Dynamax, 1:4 EtOAc/hexane, 3.0 mL·
min−1, UV detection) to obtain pure 1b (1.6 mg, 7%) with 1H NMR
and HRMS identical to the literature values.1
+5.3 1.7 (c 1.78, MeOH), [α]24365 +17.1 (c 0.387, MeOH), [α]24
405
1
+12.1, [α]24 +10.6, [α]24 + 5.7, [α]24 +3.9; see Table 1; H
436
546
633
NMR (400 MHz, CD3OD) δ 3.50 (1H, m, H-2), 2.64 (1H, dd, J =
16.0, 3.5 Hz, H-1a), 2.50 (1H, ddd, J = 16.0, 8.0 Hz, H-1b), 1.48−
1.25 (14H, m), 0.90 (3H, t, J = 6.8 Hz); 13C NMR (125 MHz,
CD3OD) δ 73.3 (CH), 48.3 (CH2), 36.0 (CH2), 33.1 (CH2), 30.8
(CH2), 30.7 (CH2), 30.5 (CH2), 26.8 (CH2), 23.8 (CH2), 14.5
(CH3); HR-ESIMS m/z 174.1851 [M + H]+ (calcd for C10H24NO
174.1852). Using a similar procedure, but with TFA in CH2Cl2, a
separate sample of free base (S)-4a (8.8 mg) was converted to (S)-4a·
Method B. A solution of benzoic acid (8.0 mg, 0.07 mmol), EDC·
HCl (16.0 mg, 0.08 mmol), and DMAP (15 mg, 0.125 mmol) was
stirred at 0 °C in CH2Cl2 (0.5 mL) for 20 min under N2, followed by
addition of 1a (3.7 mg, 0.014 mmol) in CH2Cl2 (0.5 mL). The
solution was allowed to come to room temperature and was stirred for
24 h, after which CH2Cl2 (20 mL) was added and the mixture was
washed successively with 10% HCl (20 mL), water (20 mL),
saturated NaHCO3 (20 mL), and finally water (20 mL) before
removal of the volatiles under reduced pressure. Purification of the
residue by reversed-phase column chromatography (C18, gradient
100:0 to 0:100 H2O (0.05% TFA)/MeOH) afforded 1b (2.6 mg,
TFA: [α]23 +5.2 1.4 (c 2.46, MeOH). See Table 1.
D
Amino Alcohol (R)-4b. A solution of epoxide (+)-(R)-5b (200
mg, 1.28 mmol) in EtOH (7.0 mL) and concentrated NH4OH (7.0
mL) was heated at 60 °C for 15 h.17 The mixture was cooled, and the
volatiles were removed under reduced pressure, giving 1-amino-2-
decanol, (R)-4b, as a colorless solid (159 mg, 72%), which was used
in the subsequent acylation steps without further purification. A
sample of crude amino alcohol (∼15 mg) was purified by flash
chromatography (silica, elution with CH2Cl2, followed by 14:86
MeOH (10% saturated with NH3)/CH2Cl2) to obtain pure free base
(8.8 mg) (R)-4b as a colorless solid: HR-ESIMS m/z 174.1854 [M +
H]+ (calcd for C10H24NO 174.1852).
1
1
40%) with H NMR and HRMS identical to the literature values.
Epoxide ( )-5. To a solution of 1-decene (2.0 g, 14 mmol) in 1,2-
dichloroethane (100 mL) were added m-CPBA (77% w/w, 3.81 g, 17
mmol) and the radical inhibitor TBM (300 mg).20 The mixture was
heated at reflux for 1.5 h, then cooled to room temperature and
washed, sequentially, with Na2SO3 (10% w/v), NaHCO3 (satd), and
brine. The organic layer was dried over MgSO4, and the volatiles were
removed under reduced pressure to give a pale yellow residue, which
was purified by flash chromatography (SiO2, 1:4 Et2O/pentane) to
give racemic ( )-5 as a clear oil (2.09 g, 96%): 1H NMR (500 MHz)
δ 2.90 (1H, m, H-2), 2.74 (1H, dd, J = 5.0, 4.0 Hz, H-1a), 2.45 (1H,
dd, J = 5.0, 3.0 Hz, H-1b), 1.52 (2H, m, H2-3), 1.43 (4H, m), 1.34−
1.26 (12H, m), 0.87 (3H, t, J = 6.8 Hz, H3-10); 13C NMR (125 MHz)
δ 52.6 (CH, C-2), 47.3 (C-1), 32.6 (CH2), 29.653 (CH2), 29.584
(CH2), 29.358 (CH2), 26.1 (CH2), 22.8 (CH2), 14.3 (CH3, C-
10).22,23
(+)-(S)-N,O-Dibenzoyl-1-amino-2-decanol, 8a. A solution of
amino alcohol (S)-4a (35 mg, 0.20 mmol) and DMAP (∼2 mg) in
dry pyridine (8 mL) was treated, dropwise, with freshly distilled
benzoyl chloride (93 μL, 0.808 mmol), and the mixture heated to 60
°C for 5 h. After cooling, pyridine was removed by azeotropic co-
distillation with n-heptane under reduced pressure, and the residue
purified by flash chromatography (SiO2, 1:9 EtOAc/hexanes) to
provide pure (+)-(S)-8a (11.2 mg, 15%): [α]D +28.3 (c 1.76,
MeOH), [α]D +17.8 (c 0.45, CHCl3); UV−vis (CF3CH2OH) λ 196
nm (log10 ε 4.65), 227 (4.21); FTIR (ZnSe) ν 3349 bs, 1710s,
Epoxide (−)-(S)-5a and Diol (+)-(R)-7b. Racemic ( )-5 (340
mg, 2.18 mmol) was kinetically resolved by hydrolysis, as described
above, using (S,S)-6 (2 °C, H2O, 0.55 equiv) to obtain a mixture that
1
1645m, 1271s, 745s, 1711s cm−1; H NMR (400 MHz, CDCl3) δ
8.11 (2H, d, J = 8.4 Hz, H-2′), 7.74 (2H, d, J = 7.6 Hz), 7.59 (2H, m),
7.5−7.39 (4H, m), 6.76 (1H, bt, NH), 5.30 (1H, m, H-2), 3.81 (1H,
ddd, J = 14.4, 4.0, 4.0 Hz, H-1a), 3.73 (1H, ddd, J = 14.4, 8.0, 6.0 Hz,
H-1b), 1.76−1.74 (2H, m, H2-3), 1.45 (2H, m, H2-4), 1.25 (10H, m),
0.86 (3H, t, J = 6.6 Hz); HR-ESIMS m/z 382.2376 [M + H]+ (calcd
for C24H31NO3 382.2377).
was distilled (Kugelrohr, 90°/120 mmHg). The distillate was taken up
̈
in 1:9 Et2O/hexane, and, upon standing, the solution deposited
colorless plates of (+)-(R)-7b, which were filtered, washed with
pentane, and dried (65 mg). Optically pure (−)-(S)-5a (45%) was
obtained by flash chromatography of the mother liquors (1:9 Et2O/
E
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(−)-(R)-N,O-Dibenzoyl-1-amino-2-decanol, (R)-8b. Amino
alcohol (R)-4b (25 mg, 0.14 mmol) was benzoylated, using the
above procedure (method A) to deliver (−)-(R)-8b (13.5 mg, 25%):
[α]D −26.1 (c 1.76, MeOH); [α]D −19.3 (c 1.72, CHCl3); 1H NMR
(400 MHz, CDCl3) and 13C NMR (125 MHz, CDCl3) identical with
8a; HR-ESIMS m/z 382.2373 [M + H]+ (calcd for C24H31NO3
382.2377).
AUTHOR INFORMATION
■
Corresponding Author
*Tel (T. F. Molinski): +1 858-534-7115. Fax: +1 (858) 822-
0386. E-mail: tmolinski@uscd.edu.
ORCID
Tadeusz F. Molinski: 0000-0003-1935-2535
Brent R. Copp: 0000-0001-5492-5269
Notes
(S)-N,O-Diacetyl-1-amino-2-decanol, 9a. To a solution of
amino alcohol (S)-4a (24 mg, 0.139 mmol) in dry pyridine (1.0
mL) was added acetic anhydride, and the mixture stirred at 23 °C
under N2 for 17 h. The volatiles were removed by bulb-to-bulb
distillation under high vacuum, and the straw yellow residue was
purified by flash chromatography (silica, 2:98 MeOH/CH2Cl2) to
give pure N,O-diacetyl derivative (S)-9a as a colorless oil (9.2 mg,
26%): [α]D +11.5 (c 2.3, MeOH); [α]D −7.6 (c 2.02, CHCl3); FTIR
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This fruitful collaboration was initiated after convivial
conversations at the 10th European Conference on Marine
Natural Products, Kolymbari, Crete, September 2017. We
thank Y. Su for MS data measurements, A. Mrse for assistance
with NMR measurements, and B. Lucero (all UCSD) for
preparation of some compounds. We are grateful to R. B.
Taylor (University of Auckland) for the image of P. opacum
used in the Graphical Abstract. The 500 MHz NMR
spectrometer and the HPLC TOFMS were purchased with
funding from the NSF (Chemical Research Instrument Fund,
CHE0741968) and the NIH Shared Instrument Grant
(S10RR025636) programs, respectively. This work was partly
funded by New Zealand Foundation for Research Science and
Technology, contract CO1X0205 (to B.R.C.), and grants from
the NIH (R01 AI100776-01 and R21 AT009783-01 to
T.F.M.). To the memory of Koji Nakanishi (1925−2019) a
giant in the field of natural products chemistry who inspired a
generation.
1
(ZnSe, neat) ν 3300bs, 1740s, 1665s, 1238s cm−1; H NMR (600
MHz) and 13C NMR (125 MHz) identical with (R)-9b; HR-ESIMS
m/z 280.1881 [M + Na]+ (calcd for C14H27NNaO3, 280.1883).
(R)-N,O-Diacetyl-1-amino-2-decanol, 9b. Compound (R)-4b
(32.3 mg, 0.186 mmol) was acetylated, as described for (S)-9a, above,
to give pure N,O-diacetyl derivative (R)-9b as a colorless oil (22.3 mg,
1
46%): [α]D −6.8 (c 1.29, MeOH); [α]D +7.1 (c 1.29, CHCl3); H
NMR (600 MHz, CDCl3) δ 5.74 (1H, m, NH), 4.90 (1H, m, H-2),
3.46 (1H, ddd, J = 14.4, 5.4, 3.0 Hz, H-1a), 3.38 (1H, ddd, J = 14.4,
7.2, 6.0 Hz, H-1b), 2.08 (3H, s, OAc), 1.98 (3H, s, OAc), 1.51 (2H,
m), 1.25 (12H), 0.87 (3H, t, J = 6.9 Hz, H3-10); 13C NMR (125
MHz, CDCl3) δ 171.5 (C, (O(CO)CH3), 170.4 (O(CO)CH3), 73.7
(CH, C-2), 43.3 (CH2, C-1), 32.0 (CH2), 29.5 (CH2), 29.3 (CH2),
25.4 (CH2), 22.4 (CH3, O(CO)CH3), 22.8 (CH2), 21.3 (CH3,
O(CO)CH3), 14.3 (CH3, C-10); HR-ESIMS m/z 280.1885 [M +
Na]+ (calcd for C14H27NNaO3, 280.1883).
(−)-(R)-N,O-Bis(2-naphthoyl)-1-amino-2-decanol, 10b. To a
solution of amino alcohol (R)-4b (24 mg, 0.14 mmol) and DMAP
(∼2 mg) in dry pyridine (5.5 mL) was added freshly distilled 2-
naphthoyl chloride (105 mg, 0.55 mmol), and the mixture heated at
60 °C for 17 h. After cooling, the residue was subjected to several
rounds of coazeotropic distillation with n-heptane to provide a crude
mixture of (R)-10b and 2-naphthoic acid, which were separated by
“pencil column” chromatography (5 × 87 mm, basic Al2O3, elution
with 1:9 i-PrOH/CH2Cl2) to deliver pure (−)-(R)-10b (6.6 mg, 9%):
[α]D −88.9 (c 0.55, MeOH); [α]D −104 (c 0.43, CHCl3); UV−vis
(CF3CH2OH) λ 231 nm (log10 ε 4.93), 271 sh, 280 (4.05), 290 sh;
1H NMR (600 MHz, CDCl3) δ 8.64 (1H, s, H-1′), 8.27 (1H, s, H-
1″), 8.08 (1H, dd, J = 8.4, 1.8 Hz, H-3′), 7.97 (1H, d, J = 8.4 Hz, H-
3″), 7.91−7.84 (5H, m), 7.8 (1H, dd, J = 8.4, 1.8 Hz), 7.61 (1H, td, J
= 7.5, 1.2 Hz), 7.56−7.26 (3H, m), 6.95 (1H, bt, NH), 5.42 (1H, m,
H-2), 3.90 (1H, ddd, J = 14.4, 4.8, 3.0 Hz, H-1a), 3.83 (1H, ddd, J =
14.4, 9.0, 5.4 Hz, H-1b), 1.95 (1H, m, H-3a), 1.83 (1H, m, H-3b),
1.51(2H, m, H2-4), 1.40−1.26 (13H, m), 0.86 (3H, t, J = 6.9 Hz, H3-
10); 13C NMR (150 MHz, APT, CDCl3) δ 167.8 (C, C = O), 167.7
(C, C = O), 141.6 (CH), 135.8 (C), 134.8 (C), 132.7 (C), 132.6 (C),
131.6 (C), 131.5 (CH), 129.5 (CH), 129.1 (CH), 128.6 (CH), 128.5
(CH), 127.9 CH), 127.8 (CH), 127.74 (CH), 127.67 (CH), 127.2
(C), 126.9 (CH), 126.8 (CH), 125.3 (CH), 123.6 (CH), 74.7 (CH),
44.6 (CH2), 32.6 (CH2), 32.0 (CH2), 29.6 (2×CH2), 29.4 (CH2),
25.6 (CH2), 22.8 (CH2), 14.3 (CH3); HR-ESIMS m/z 504.2508 [M
+ Na]+ (calcd for C32H35NNaO3 504.2509).
REFERENCES
■
(1) Wang, J.; Pearce, A. N.; Chan, S. T. S.; Taylor, R. B.; Page, M. J.;
Valentin, A.; Bourguet-Kondracki, M.-L.; Dalton, J. P.; Wiles, S.;
Copp, B. R. J. Nat. Prod. 2016, 79, 607−610.
́
́
(2) Martinkova, M.; Gonda, J.; Jackova, D. Tetrahedron: Asymmetry
2016, 27, 1187−1212.
́
(3) Garrido, L.; Zubía, E.; Ortega, M. J.; Naranjo, S.; Salva, J.
Tetrahedron 2001, 57, 4579−4588.
(4) (a) Aiello, A.; Fattorusso, E.; Giordano, A.; Menna, M.;
Navarrete, C.; Munoz, E. Bioorg. Med. Chem. 2007, 15, 2920−2926.
̃
(b) Kossuga, M. H.; MacMillan, J. B.; Rogers, E. W.; Molinski, T. F.;
Nascimento, G. G. F.; Rocha, R. M.; Berlinck, R. G. S. J. Nat. Prod.
2004, 67, 1879−1881.
(5) Examples from outlying phyla are calyxoside from the green alga
Calyx: (a) Zhou, B. N.; Mattern, M. P.; Johnson, R. K.; Kingston, D.
G. I. Tetrahedron 2001, 57, 9549−9554. epi-Xestoaminol C from the
brown alga Xiphophora chondrophylla: (b) Dasyam, N.; Munkacsi, A.
B.; Fadzilah, N. H.; Senanayake, D. S.; O’Toole, R. F.; Keyzers, R. A.
J. Nat. Prod. 2014, 77, 1519−1523. Spisulosin from the clam, Spisula
polynyma: (c) Cuadros, R.; Montejo de Garcini, E.; Wandosell, T.;
́
Faircloth, G.; Fernandez-Sousa, J. M.; Avil, J. Cancer Lett. 2000, 152,
23−29.
(6) (a) Kong, F. H.; Faulkner, D. J. J. Org. Chem. 1993, 58, 970−
971. (b) Dalisay, D. S.; Tsukamoto, S.; Molinski, T. F. J. Nat. Prod.
2009, 72, 353−359.
(7) Chianese, G.; Fattorusso, E.; Putra, M.; Calcinai, B.; Bavestrello,
G.; Moriello, A.; Petrocellis, L.; Marzo, V.; Taglialatela-Scafati, O.
Mar. Drugs 2012, 10, 2435−2447.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00937.
(8) (a) Makarieva, T. N.; Denisenko, V. A.; Stonik, V. A.
Tetrahedron Lett. 1989, 30, 6581−6584. (b) Molinski, T. F.;
Makarieva, T. N.; Stonik, V. A. Angew. Chem., Int. Ed. 2000, 39,
4076−4079.
(9) (a) Nicholas, G. M.; Hong, T. W.; Molinski, T. F.; Lerch, M. L.;
Cancilla, M. T.; Lebrilla, C. B. J. Nat. Prod. 1999, 62 (12), 1678−
1H NMR, 13C NMR, ESI HRMS of 4a and 5−10, APT
spectra of 9b and 10b, ECD data for 8a,b and 10a,b,
error analysis of [α]D, and ECD of (−)-10b (sensitivity,
limits of detection) (PDF)
F
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1681. (b) Nicholas, G. M.; Molinski, T. F. J. Am. Chem. Soc. 2000,
122, 4011−4019.
(10) It should be stressed that each two-headed sphingolipid
characterized, so far, appears to be enantiopure. In other words,
absolute and relative configuration is highly conserved in biosynthesis
within each specimen of sponge.
(29) The boundary conditions set for LOD (Figure S25) were the
ability to discriminate a distinct bisignate Cotton effect with a signal
of 3× the noise (S/N ≈ 3) at wavelengths where no ECD signal
occurs (λ 360−400 nm), corresponding to the minimum detected
mass in a 1 mm path length ECD cell of volume 330 μL and
accumulated (averaged) number of scans n = 10 (20 min
measurement). Obviously, larger values of n will improve LOD.
(30) (a) Polarvarapu, P. L. In Comprehensive Chiroptical Spectros-
copy: Applications in Stereochemical Analysis of Synthetic Compounds,
Natural Products, and Biomolecules; Berova, N.; Polavarapu, P. L.;
Nakanishi, K.; Woody, R. W., Eds.; Wiley, 2012; Vol. 2, Chapter 11.
(b) Stephens, P. J.; McCann, D. M.; Devlin, F. J.; Smith, A. B. J. Nat.
(11) Similar notoriety is associated with other amino alkanols; for
example, see clavaminols A−F from Clavelina phlegraea (a) ref 7 and
(2S,3R)-2-aminododecan-3-ol from C. oblonga from Brazil (b) ref 4b.
(12) Other sponge-derived aliphatic amines share the property of
weak rotation. For example the dimeric oxaquinolizidine (−)-xesto-
spongin C from Xestospongia exigua ([α]D −2.7 (c 0.54, CHCl3):
Nakagawa, M.; Endo, M.; Tanaka, N.; Gen-Pei, L. Tetrahedron Lett.
1984, 25, 3227−3230. [α]D −2.4 (c 0.34, CHCl3): Moon, S. S.;
MacMillan, J. B.; Olmstead, M. M.; Anh Ta, T.; Pessah, I. N.;
Molinski, T. F. J. Nat. Prod. 2002, 65, 249−254. This exists in Nature
in both enantiomeric forms and has been the subject of total
synthesis: Baldwin, J. E.; Melman, A.; Lee, V.; Firkin, C. R.;
Whitehead, R. C. J. Am. Chem. Soc. 1998, 120, 8559−8560. And
misassignment: Hoye, T. R.; North, J. T.; Yao, L. J. J. Am. Chem. Soc.
1994, 116, 2617−2618. Hoye, T. R.; Ye, Z.; Yao, L. J.; North, J. T. J.
Am. Chem. Soc. 1996, 118, 12074−12081.
Prod. 2006, 69, 1055−1064. (c) Burgueno-Tapia, E.; Joseph-Nathan,
̃
P. Nat. Prod. Commun. 2015, 10, 1785−1795.
(31) (a) Parchansky, V.; Kapitan, J.; Bour, P. RSC Adv. 2014, 4,
́
57125−57136. (b) Nieto-Ortega, B.; Casado, J.; Blanch, E. W.; Lopez
Navarrete, J. T.; Quesada, A. R.; Ramírez, F. R. J. Phys. Chem. A 2011,
115, 2752−2755.
(32) For more recent investigations of EC ECD of additional
samples of 1b and synthetic analogues that support this conclusion,
see: Pearce, A. N.; Copp, B. R.; Molinski, T. F. Molecules 2019, 24,
90−104.
(13) Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy:
Exciton Coupling in Organic Stereochemistry; University Science Books:
Mill Valley, CA, USA, 1983.
(14) Kawai, M.; Nagai, U.; Katsumi, M. Tetrahedron Lett. 1975, 16,
3165−3166.
(15) Searle, P. A.; Molinski, T. F. J. Org. Chem. 1993, 58, 7578−
7580.
(16) Won, T.; You, M.; Lee, S.-H.; Rho, B.; Oh, D.-C.; Oh, K.-B.;
Shin, J. Mar. Drugs 2014, 12, 3754.
(17) Sun, D.-Y.; Han, G.-Y.; Gong, J.-X.; Nay, B.; Li, X.-W.; Guo, Y.-
W. Org. Lett. 2017, 19, 714−717.
(18) Dumpala, M.; Srinivas, T.; Krishna, P. R. Tetrahedron Lett.
2017, 58, 1273−1275.
(19) Nicholas, G. M.; Molinski, T. F. J. Am. Chem. Soc. 2000, 122,
4011−4019.
(20) TBM: 4,4′-thiobis(tert-butyl-3-methylphenol): Kishi, Y.;
Aratani, M.; Tanino, H.; Fukuyama, T.; Goto, T.; Inoue, S.;
Sugiura, S.; Kakoi, H. J. Chem. Soc., Chem. Commun. 1972, 64−65.
(21) (a) Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N.
Science 1997, 277, 936−938. (b) Schaus, S. E.; Brandes, B. D.;
Larrow, J. F.; Tokunaga, M.; Hansen, K. B.; Gould, A. E.; Furrow, M.
E.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 1307−1315.
(22) Masaoka, Y.; Sakakibara, M.; Mori, K. Agric. Biol. Chem. 1982,
46, 2319−2324.
(23) Oppolzer, W.; Dudfield, P. Tetrahedron Lett. 1985, 26, 5037−
5040.
(24) For reasons that are unclear, earlier yields of 1b (9%, ref 1) and
3b (14%, ref 15) were dismal and not much improved in this study of
1b, (S)-8a, and (R)-8b (7%, 15%, and 24%, respectively) under
standard conditions (BzCl, pyridine, 60 °C). An alternative method
using EDC-mediated coupling between 1a and benzoic acid gives 1b
with a more satisfactory yield of 40%. See Experimental Section and
ref 32 for details.
(25) It is noteworthy that the signs of rotation of HCl and TFA salts
of (S)-4a are opposite that of the free base. Also, as expected, [α]
values measured at shorter wavelengths are larger (see Experimental
Section).
́
̃
oa, E.; Riguera, R. Chem. Rev. 2012, 112,
(26) (a) Seco, J. M.; Quin
4603−4641. (b) Leiro, V.; Seco, J. M.; Quin
Lett. 2008, 10, 2729−2732.
́
oa, E.; Riguera, R. Org.
̃
(27) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092−4096.
(28) See also Table S1. A slight difference is evident in the ratio of
Cotton effects (negative vs positive peaks R = |Δε(λ 237)/Δε(λ 221)|
for (R)-8b (R = 0.48) compared to (S)-1b (R = 1.1),1 which may
reflect a weak interaction with the conjugated 1,3-diyne of the latter.
G
DOI: 10.1021/acs.jnatprod.8b00937
J. Nat. Prod. XXXX, XXX, XXX−XXX