Structure assignment of lucentamycin e and revision of the olefin geometries of the marine-derived lucentamycins

Article abstract of DOI:10.1021/np3003854

A new lucentamycin analogue, lucentamycin E (5), was isolated from the culture broth of the marine-derived actinomycete Nocardiopsis lucentensis, strain CNR-712. The absolute stereostructure of 5 was assigned by comprehensive analyses of NMR data and by application of the advanced Marfey's method. The planar structure of 5 was analogous to lucentamycins A-D, whereas the olefin geometry of the 3-methyl-4-ethylideneproline moiety was found to be E, opposite of that previously reported. Consequently, a reinvestigation of the olefin geometries of the 3-methyl-4-ethylideneproline residues of lucentamycins A-D showed that the olefin geometries of the substituted proline functionalities must be revised to (2S,3R,E)-3-methyl-4-ethylideneproline.

Full text of DOI:10.1021/np3003854

Note
pubs.acs.org/jnp
Structure Assignment of Lucentamycin E and Revision of the Olefin
Geometries of the Marine-Derived Lucentamycins
Jin Wook Cha,† Jin-Soo Park,† Taebo Sim,‡ Sang-Jip Nam,§,⊥ Hak Cheol Kwon,*,† Juan R. Del Valle,∥
and William Fenical*,§
†Natural Medicine Center, Korea Institute of Science and Technology, Gangneung, Gangwon-do 210-340, Republic of Korea
‡Korea Institute of Science and Technology, Cheongryang, Seoul 130-650, Republic of Korea
§Center for Marine Biotechnology and Biomedicine, Scripps Institution of Oceanography, University of California, San Diego, La
Jolla, California 92093-0204, United States
∥Drug Discovery Department, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida 33612, United States
S
* Supporting Information
ABSTRACT: A new lucentamycin analogue, lucentamycin E
(5), was isolated from the culture broth of the marine-derived
actinomycete Nocardiopsis lucentensis, strain CNR-712. The
absolute stereostructure of 5 was assigned by comprehensive
analyses of NMR data and by application of the advanced
Marfey’s method. The planar structure of 5 was analogous to
lucentamycins A−D, whereas the olefin geometry of the 3-
methyl-4-ethylideneproline moiety was found to be E,
opposite of that previously reported. Consequently, a
reinvestigation of the olefin geometries of the 3-methyl-4-
ethylideneproline residues of lucentamycins A−D showed that the olefin geometries of the substituted proline functionalities
must be revised to (2S,3R,E)-3-methyl-4-ethylideneproline.
the ROESY NMR data derived from this compound, coupled
n an earlier publication the structures of lucentamycins A−D
I(1−4), peptide metabolites produced by the marine-derived with Marfey’s amino acid analysis of the acid hydrolysates of 5
actinomycete Nocardiopsis lucentensis, were reported.1 Sub-
and lucentamycin A (1) and syntheses of the four steroisomers
of 3-methyl-4-ethylideneproline, allowed the full stereostruc-
tures of 1−5 to be defined. When the original ROESY
spectroscopic data of lucentamycin A were re-examined, it was
clear that the incorrect Z-olefin assignment1 was based upon
observed 2D NMR contours that were nonreproducible.
Lucentamycin E (5), obtained as a colorless gum, analyzed
for the molecular formula C31H431N7O5 by interpretation of
HRFABMS data. Interpretation of H, 13C, and 2D NMR data
revealed that 5 was composed of three amino acid residues
[tryptophan (Trp), homoarginine (Har), 3-methyl-4-ethyl-
ideneproline (Mep)] and an isopentenoic acid moiety. The
structure of 5 was found to be analogous to lucentamycin B but
was composed of an isopentenoic acid residue instead of an
isopentanoic acid. The structure of the isopentenoic acid group
was assigned by interpretation of HMBC correlations from the
geminal methyl groups (δH 1.75 and 2.05) to the quaternary
carbon (C-29) and to the sp2 carbon resonance (C-28), as well
as from the olefinic proton signal (H-28) to the carbonyl
carbon (C-27).
sequently, several research groups have attempted a total
synthesis of lucentamycin A,2−5 based on the structural novelty
of the 3-methyl-4-ethylideneproline unit and their potent
cytotoxicities toward human colon carcinoma cells. Recently,
Del Valle and co-workers accomplished the total synthesis of
the reported structure of lucentamycin A.2 However, in that
report, a discrepancy in the NMR data between synthetic
lucentamycin A and natural lucentamycin A was observed, with
most of the differences attributed to resonances derived from
the ethylideneproline moiety. A total synthesis of 8-epi-
lucentamycin A was concurrently reported by Lindsley and
co-workers,3 and subsequent publications from the groups of
Del Valle4 and Sim5 described the synthesis of all four possible
stereoisomers of 3-methyl-4-ethylideneproline and their in-
corporation into target tripeptides. They also found that all of
their synthetic compounds did not match natural lucentamycin
A. In this paper, we correct the structures of the lucentamycins,
specifically revising the structure of the substituted proline
residue to (2S,3R,E)-3-methyl-4-ethylideneproline and also
adding another metabolite, lucentamycin E (5), to this group.
On the basis of the discrepancies defined by synthetic
studies, we decided to reisolate the lucentamycins from N.
lucentensis, strain CNR-712. During this re-examination, we
isolated a new metabolite, lucentamycin E (5). Re-evaluation of
The secondary amide proton signal (21-NH) at δH 7.94
exhibited a strong COSY correlation with a methine proton
Received: May 31, 2012
Published: September 6, 2012
© 2012 American Chemical Society and
American Society of Pharmacognosy
1648
dx.doi.org/10.1021/np3003854 | J. Nat. Prod. 2012, 75, 1648−1651

Journal of Natural Products
Note
Chart 1
signal at δH 4.65 (H-21), which is the α-proton of the
homoarginine (Har) residue of 5. A serial COSY correlation
was observed from H-21 to the methylene proton signals at δH
3.01 and 3.15 (H2-25), which were connected to 25-NH [δH
10.47] of the guanidine unit. An HMBC correlation of the α-
proton signal at δH 4.65 (H-21) to the carbonyl carbon at δC
165.3 (C-27) in the isopentenoic acid moiety indicated the
connectivity of the Har and isopentenoic acid residues.
(unpublished data). The Mep moiety of 5 can be in two
stereoisomeric forms (2R,3S and 2S,3R configuration) based on
the analysis of ROESY correlations and the value of the
coupling constant between H-13 and H-14. The HPLC
retention time of the L-FDLA derivative of (2S,3R)-Mep was
34.97 min (run at a different time), and the L-FDLA derivative
of (2R3S)-Mep was 37.17 min, whereas the L-FDLA-derivatized
Mep in the acid hydrolysate of 5 (33.40 min, run earlier)
matched best with that of (2S,3R)-Mep. Thus, the absolute
configuration of the Mep moiety in 5 was defined as 2S, 3R,
which is the same absolute configuration defined earlier for
lucentamycin A (1). By extensive NMR studies and application
of the advanced Marfey’s method, the absolute configurations
of 5 and 1 can be assigned as 2S, 13S, 14R, 15E, 21S and 2S, 8S,
9R, 10E, 16S, respectively.
Comprehensive analysis of 1D and 2 D NMR data (Table 1)
showed that the C-1 to C-21 core of lucentamycin E was
identical to that of lucentamycin B (2), thus establishing their
relationship. However, the geometrical configuration of the
exocyclic double bond in the Mep moiety of 5 was determined
as E on the basis of clear ROESY correlations between H3-18
and H-14 and between H3-18 and H-19, coupled with a
ROESY correlation between H-17 and H2-16 (Figure 1). The
relative configuration at C-13 and C-14 in the Mep moiety of 5
was determined to be syn based on the analysis of ROESY
correlations and the value of the coupling constant (3J = 8.5
Hz) between H-13 and H-14 in comparison with the previously
reported NMR data for the four synthetic stereoisomers of
lucentamycin A (with Z-olefin).5 Because the E double-bond
geometry of 5 was opposite of that previously assigned to 1−4,
the alkene geometry of lucentamycin A (1) was reinvestigated
by careful analysis of ROESY NMR data under several
conditions. A strong ROESY correlation between H3-13 and
H3-14 revealed that the double-bond geometry of Mep of
lucentamycin A was also in the E configuration.
The absolute configurations of the amino acid residues in 5
were defined by applying the advanced Marfey’s method.6 The
acid hydrolysate of 5 was derivatized, in separate experiments,
with 1-fluoro-2,4-dinitrophenyl-5-L-leucinamide (L-FDLA) and
D-FDLA and analyzed by C8 RP HPLC LC-MS. The retention
times, molecular ion peaks, and UV spectra of Trp and Har
derivatives were identical with those of authentic L-Trp and L-
Har derivatives. In addition, the HPLC elution time of L-FDLA-
derivatized Mep (33.40 min) was earlier than D-FDLA-
derivatized Mep (36.20 min). For the determination of the
absolute configuration of Mep, Marfey’s analysis was performed
with four stereoisomers of Mep, which were obtained from
Taebo Sim at the Korea Institute of Science and Technology
EXPERIMENTAL SECTION
■
General Experimental Procedures. The optical rotation was
measured on a PerkinElmer model 343 polarimeter. UV and IR spectra
were recorded using a Perkin-Elmer Lambda 35 UV/vis spectrometer
and a Thermo Scientific Nicolet iS10 spectrometer, respectively. High-
resolution FABMS data were obtained using a JEOL/JMS-AX505WA
1
instrument. H, 13C, and 2D NMR spectral data were obtained in
DMSO-d6 on a Varian UNITY 500 MHz NMR spectrometer. Low-
resolution ESIMS data were acquired by an Agilent 1200 series HPLC
system/6120 quadrupole MSD. A Waters 1525 HPLC-PDA system
with a Phenomenex Luna C18 (2) 5 μm column (150 × 4.6 mm) was
used for the analysis of the extract and chromatography fractions.
Diaion HP-20 (Supelco) resin was used for flash column
chromatography. HPLC separation was performed with a Gilson
321 HPLC system with a Phenomenex Luna 10 μm C18(2) (250 × 10
mm) column for semipreparative separation and a YMC-Pack ODS-A
5 μm (250 × 10 mm) column for final purification.
Collection, Identification, and Cultivation of Strain CNR-712.
Actinomycete strain CNR-712 was isolated from a sediment sample
collected in a shallow saline pond on the island of Little San Salvador
in the Bahamas. The strain shares 99.5% sequence identity with
Nocardiopsis lucentensis. The 16S rRNA gene sequence has been
deposited in GenBank (accession number EF392847). Actinomycete
strain CNR-712 was cultured in 32 replicates of 500 mL volume using
an A1 liquid culture medium (10 g of starch, 4 g of peptone, 2 g of
yeast extract in 1 L of natural seawater, SIO Pier) for 7 days at 27 °C
while shaking at 200 rpm. After 7 days, the culture media was extracted
1649
dx.doi.org/10.1021/np3003854 | J. Nat. Prod. 2012, 75, 1648−1651

Journal of Natural Products
Note
Isolation and Purification of Lucentamycin E (5). The extract
(1.06 g) was absorbed on Celite (1.5 g) and subjected to flash column
chromatography using Diaion HP-20 (Supelco) resin as the adsorbent
phase. The column was eluted with aqueous MeOH using a step
gradient approach (20%, 50%, 70%, 100% MeOH in H2O) to yield
four fractions. The 70% aqueous MeOH fraction was dried under
vacuum and refractionated by reversed-phase HPLC (Gilson 321;
Phenomenex Luna 10 μm C18(2) (250 × 10 mm) column; at a flow
rate of 4 mL/min) using isocratic elution with 35% acetonitrile in
H2O. The subfraction containing 5 was then finally purified by
reversed-phase HPLC by isocratic elution with 35% MeCN in H2O
[(YMC-Pack ODS-A 5 μ (250 × 10 mm) column; at a flow of 2 mL/
min)] to yield 5 (0.8 mg).
Table 1. NMR Spectroscopic Data (DMSO-d6) for
1
Lucentamycin E (5) (500 MHz for H, 125 MHz for 13C)
a
position
δC type
δH (J in Hz)
1
175.5, C
54.2, CH
28.6, CH2
2
4.19, m
b
3
3.03, m
c
4
111.0, C
5
127.9, C
6
118.7, CH
117.9, CH
120.6, CH
7.57, d (7.5)
6.91, t (7.5)
7.02, t (7.5)
7.27, d (7.5)
7
8
Lucentamycin E (5). The new compound, a colorless gum: [α]20
c
9
111.0, C
D
−19 (c 0.1, MeOH); UV (MeOH) λmax (log ε) 206 (4.34), 220 (4.33),
280 (3.62) nm; IR (KBr) νmax 3273, 1635, 1180 cm−1; see Table 1 and
Supporting Information for 1H NMR data and 13C NMR data;
HRFABMS [M + H]+ m/z 594.3407 (calcd C31H44N7O5, 594.3404).
Acid Hydrolysis and Advanced Marfey’s Analysis. Lucentamycin E
(5, 0.7 mg) was hydrolyzed with 6 N HCl at 110 °C for 2 h, and the
hydrolysate was divided into two equal portions. Each portion was
treated with 100 μL of 1 N NaHCO3 and 50 μL of 1% w/v 1-fluoro-
2,4-dinitrophenyl-L-5-leucinamide (L-FDLA or D-FDLA) solution in
acetone, followed by heating at 80 °C for 3 min. The reaction mixtures
were cooled at room temperature, neutralized by adding 30 μL of 2 N
HCl, and diluted with 300 μL of 50% aqueous MeCN solution. The
resulting solution was analyzed by LC-MS [Agilent 6120 quadrupole
ESI-MS (positive mode)] using a Phenomenex Luna C18 (2) 5 μm
column (150 × 4.6 mm) and an aqueous MeCN solution applied in a
linear gradient (10−100% for 60 min at a flow 0.7 mL/min). The
retention times of the L- and D-FDLA-derivatized amino acids in the
acid hydrolysate of 5 were 19.65 and 19.55 min for L-Har, 33.00 and
34.76 min for L-Trp, and 33.40 and 36.20 min for L-Mep, respectively.
The retention times of the L- and D-FDLA derivatives of authentic L-
Har were 19.74 and 19.50 min, respectively, whereas those of an
authentic L-Trp were 33.24 and 34.73 min. The HPLC retention times
for the L-FDLA derivatives of four synthetic isomers of 3-methyl-4-
ethylideneproline, run at a different time, were 37.82 min for 2R,3R-
Mep, 37.17 min for 2R,3S-Mep, 34.97 min for 2S,3R-Mep, and 36.11
min for 2S,3S-Mep.
10
135.9, C
123.1, CH
167.6, C
65.6, CH
35.0, CH
139.4, C
51.0, CH2
11
7.05, br s
12
13
4.21, d (8.5)
3.17, m
14
15
16a
16b
17
4.40, d (14.0)
4.26, d (14.0)
5.34, q (6.5)
1.57, d (6.5)
0.83, d (7.0)
116.3, CH
13.2, CH3
14.9, CH3
171.4, C
18
19
20
21
49.9, CH
31.9, CH2
4.65, br m
1.80, m
1.47, m
1.47, m
1.36, m
1.52, m
1.34, m
3.12, m
2.99, m
22a
22b
23a
23b
24a
24b
25a
25b
26
23.0, CH2
28.7, CH2
41.3, CH2
157.2, C
165.3, C
118.6, CH
149.5, C
27.0, CH3
19.4, CH3
27
28
5.74, s
29
ASSOCIATED CONTENT
30
1.75, s
2.05, s
■
31
S
* Supporting Information
b
2-NH
10-NH
21-NH
25-NH
26-NH
26-NH2
7.01, br s
The spectroscopic data and further details of the structure
assignment are available free of charge via the Internet at
http://pubs.acs.org.
10.6, s
7.94, d (8.5)
b
10.47, s
not observed
AUTHOR INFORMATION
■
6.89, s
a
b
Corresponding Author
Number of attached protons assigned by DEPT NMR. Multiplicity
patterns were unclear due to signal overlap. Signals were overlapped.
c
*(W.F.) Tel: +1-858-534-2133. Fax: +1-858-534-1318. E-mail:
wfenical@ucsd.edu. (H.C.K.) Tel: + 82 33 650 3504. E-mail:
hkwon@kist.re.kr.
Present Address
⊥College of Pharmacy and Research Institute of Life and
Pharmaceutical Sciences, Sunchon National University, Sun-
cheon 540-950, Republic of Korea.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Figure 1. Key ROESY correlations for the 3-methyl-4-ethylidinepro-
line moieties in lucentamycins A (1) and E (5).
The present study was supported by the National Cancer
Institute, NIH, under grant R37CA044848, and by the Korea
Institute of Science and Technology institutional program,
under grant number 2Z03550.
with EtOAc (16 L overall). Concentration afforded 1.06 g of EtOAc
extract.
1650
dx.doi.org/10.1021/np3003854 | J. Nat. Prod. 2012, 75, 1648−1651

Journal of Natural Products
Note
REFERENCES
■
(1) Cho, J. Y.; Williams, P. G.; Kwon, H. C.; Jensen, P. R.; Fenical,
W. J. Nat. Prod. 2007, 70, 1321−1328.
(2) Pal, U.; Ranatunga, S.; Ariyarathna, Y.; Del Valle, J. R. Org. Lett.
2009, 11, 5298−5301.
(3) Daniels, R. N.; Melancon, B. J.; Wang, E. A.; Crews, B. C.;
Marnett, L. J.; Sulikowski, G. A.; Lindsley, C. W. J. Org. Chem. 2009,
74, 8852−8855.
(4) Ranatunga, S.; Kim, J. S.; Pal, U.; Del Valle, J. R. J. Org. Chem.
2011, 76, 8962−8976.
(5) Ham, Y. J.; Yu, H.; Kim, N. D.; Hah, J.-M.; Selim, K. B.; Choi, H.
G.; Sim, T. Tetrahedron 2012, 68, 1918−1925.
(6) Fujii, K.; Ikai, Y.; Oka, H.; Suzuki, M.; Harada, K. Anal. Chem.
1997, 69, 5146−5151.
1651
dx.doi.org/10.1021/np3003854 | J. Nat. Prod. 2012, 75, 1648−1651