Catenulobactins A and B, Heterocyclic Peptides from Culturing Catenuloplanes sp. with a Mycolic Acid-Containing Bacterium

Article abstract of DOI:10.1021/acs.jnatprod.8b00261

The production of two new heterocyclic peptide isomers, catenulobactins A (1) and B (2), in cultures of Catenuloplanes sp. RD067331 was significantly increased when it was cocultured with a mycolic acid-containing bacterium. The planar structures and absolute configurations of the catenulobactins were determined based on NMR/MS and chiral-phase GC-MS analyses. Catenulobactin B (2) displayed Fe(III)-chelating activity and moderate cytotoxicity against P388 murine leukemia cells.

Full text of DOI:10.1021/acs.jnatprod.8b00261

Note
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Catenulobactins A and B, Heterocyclic Peptides from Culturing
Catenuloplanes sp. with a Mycolic Acid-Containing Bacterium
Shotaro Hoshino,† Masahiro Ozeki,† Takayoshi Awakawa,†,‡ Hiroyuki Morita,§ Hiroyasu Onaka,‡,⊥
and Ikuro Abe*,†,‡
†Graduate School of Pharmaceutical Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan
‡Collaborative Research Institute for Innovative Microbiology, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657,
Japan
§Institute of Natural Medicine, University of Toyama, 2630-Sugitani, Toyama 930-0194, Japan
⊥Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113-8657, Japan
S
* Supporting Information
ABSTRACT: The production of two new heterocyclic
peptide isomers, catenulobactins A (1) and B (2), in cultures
of Catenuloplanes sp. RD067331 was significantly increased
when it was cocultured with a mycolic acid-containing
bacterium. The planar structures and absolute configurations
of the catenulobactins were determined based on NMR/MS and chiral-phase GC-MS analyses. Catenulobactin B (2) displayed
Fe(III)-chelating activity and moderate cytotoxicity against P388 murine leukemia cells.
RD067331 was significantly enhanced in the presence of the
ctinomycetes are Gram-positive bacteria and prolific
Asources of bioactive compounds. Phylogenetically, actino- mycolic acid-containing bacterium Tsukamurella pulmonis TP-
B0596 (Figure 1A). Herein, we report the isolation and
structural elucidation of two novel heterocyclic peptides,
named catenulobactins A and B (1 and 2, Figure 1B), as well
as the cytotoxicity and Fe(III)-chelating property of 2.
Initially, we cultured the Catenuloplanes sp. RD067331 in the
presence or absence of T. pulmonis TP-B0596 (a mycolic acid-
containing bacterium) and analyzed each metabolic profile by
HPLC. Intriguingly, the production of two metabolites was
significantly enhanced in the presence of the mycolic acid-
containing bacterium (1 and 2, Figure 1A). To determine their
chemical structures, the combined-culture was scaled up (2.0
L), and the extract from the culture was subjected to a series of
chromatographic purifications, yielding the catenulobactins A
(1, 3.1 mg) and B (2, 2.4 mg).
mycetes are often divided into Streptomyces and non-
Streptomyces (rare actinomycetes). In contrast to Streptomyces,
which is the largest genus of actinobacteria and whose
secondary metabolites have been extensively investigated,1,2
those from rare actinomycetes have been less exploited, except
as described in some reports.3−5 Recent genome sequencing
projects have revealed that rare actinomycetes are a rich source
of cryptic natural product biosynthetic gene clusters,6,7
whereas the number of identified secondary metabolites is
much less than that of natural product biosynthetic gene
clusters. Considering that most of the gene clusters are likely
not expressed in the traditional laboratory conditions,
activation of these cryptic natural product biosynthetic gene
clusters would lead to isolation of various novel compounds.
Coculture is a simple but potent approach to activate the
cryptic natural product biosynthetic gene clusters in certain
microorganisms.8−11 In nature, microorganisms often interact
via signaling molecules and/or physical contacts, which may
trigger the production of secondary metabolites. Coculture is
thought to mimic such an interaction factor and induce the
expression of silent natural product biosynthetic gene clusters.
Indeed, we previously demonstrated that the cryptic natural
product biosynthetic gene clusters in several Streptomyces
species were efficiently activated by the “combined-culture”
with a mycolic acid-containing bacterium,8,9 which resulted in
the isolation of 17 novel secondary metabolites.8 Recently, we
also demonstrated that this method can be applied to rare
actinomycetes.12,13 During screening of metabolites from the
combined-culture with rare actinomycetes, we found that the
production of secondary metabolites in Catenuloplanes sp.
The molecular formula of catenulobactin A (1) was
established as C27H30N4O9 based on HR-ESIMS analysis,
which indicated 14 degrees of unsaturation. The 1D NMR
(1H, 13C) and HMQC data revealed the presence of 27
carbons (Table 1), consisting of five carbonyls/imidate (δC
173.5−165.9), two aromatic carbons adjacent to an oxygen
atom (δC 160.2 and 159.7), two sp2 nonprotonated carbons
(δC 115.7 and 110.6), eight aromatic methines (δC 134.7−
117.1), six aliphatic methines adjacent to oxygen/nitrogen
atoms (δC 79.5−52.5), three aliphatic methylenes (δC 45.0,
28.8, and 23.6), and two methyl groups (δC 21.1 and 17.2).
1
The H NMR spectrum also displayed three exchangeable
proton resonances, including typical amide NHs (δH 9.21 and
Received: March 27, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00261
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
Table 1. NMR Data for Catenulobactins A (1) and B (2)
Recorded in d6-DMSO (1H: 500 MHz, 13C: 125 MHz)
catenulobactin A (1)
catenulobactin B (2)
position
δC, type
159.7, C
δH (J in Hz)
δC, type
δH (J in Hz)
c
1
159.8, C
h
g
1-OH
(11.79, s)
12.02,
s
2
3
4
5
6
7
117.1, CH 6.96, d (7.5)
117.1, CH 6.97, m
a
b
d
134.7, CH 7.43, dd (7.5)
119.6, CH 6.90, m
128.6, CH 7.59, d (7.5)
110.6, C
134.6, CH 7.43, m
e
119.6, CH 6.91, m
f
128.6, CH 7.59, m
110.6, C
165.9, C
165.9, C
8
9
10
11
12(NH)
13
14
73.9, CH
169.9, C
79.5, CH
21.1, CH3
4.53, d (7.5)
73.9, CH
169.8, C
4.52, d (7.5)
4.84, dq (7.5, 6.5) 79.5, CH
4.81, m
1.43, d (6.5)
8.42, brd (7.5)
4.18, m
21.0, CH3
1.42, d (6.5)
8.39, brd (7.0)
4.18, m
52.5, CH
28.8, CH2
52.6, CH
29.0, CH2
1.79, m
1.75, m
1.69, m
1.60, m
15
16
23.6, CH2
45.0, CH2
1.61−1.57, m
3.56, m
23.2, CH2
47.8, CH2
1.63−1.58, m
3.65, m
3.45, m
3.45, m
17
18
19
173.5, C
167.6, C
57.7, CH
173.7, C
169.5, C
70.8, CH
4.70, dd (10.5,
7.5)
5.02, d (5.5)
20(NH)
21
9.21, d (7.5)
4.42, dq (10.5,
6.0)
78.6, CH
79.3, CH
4.91, m
22
23
24
25
26
27
28
29
17.2, CH3
169.4, C
115.7, C
1.35, d (6.0)
20.9, CH3
165.9, C
1.42, d (6.5)
110.6, C
f
128.8, CH 7.84, d (8.0)
119.3, CH 6.89, m
128.4, CH 7.59, m
Figure 1. (A) HPLC chromatograms of crude extracts from
Catenuloplanes sp. cocultured with a mycolic acid-containing
bacterium (top), the pure culture of Catenuloplanes sp. (bottom),
and the pure culture of the mycolic acid-containing bacterium
(bottom). (B) Chemical structures of 1 and 2 (Sal: salicylic acid,
MeOzn: methyloxazoline, OHOrn: N-hydroxyornithine).
e
119.5, CH 6.91, m
a
b
d
134.6, CH 7.40, dd (7.5)
118.0, CH 6.91, m
160.2, C
134.5, CH 7.43, m
117.1, CH 6.97, m
c
158.7, C
h
g
29-OH
(11.79, s)
11.79,
s
a−g
h
Assignments are interchangeable. The signal at 11.79 ppm is
8.42) and a phenolic OH (δH 11.79). The presence of two sets
of o-substituted benzene rings was inferred from the COSY
correlations (Figure 2, H-2/H-3/H-4/H-5 and H-25/H-26/H-
27/H-28), and both were confirmed as salicylic acids (Sal1 and
Sal2 in Figure 1B) based on the HMBC correlations from
aromatic protons (Figure 2). Further NMR analyses unveiled
the presence of three amino acid residues (Thr1, Thr2, and
Orn, Figure 1B). The Thr1 residue was clearly indicated by the
COSY correlations of H-8/H-10/H3-11 and the HMBC
correlations of H-8/C-9 and H-10/C-9. Considering the
downfield-shifted 13C NMR resonance at C-8 (δC 73.9, Figure
S15)14,15 and the absence of an NH signal in the 1H NMR, the
C-8 position is thought to be adjacent to an imine nitrogen
atom (−CN−C) rather than to an amide nitrogen (C(
O)−NH−C) (Figure S15),14,15 and the Thr1 residue was
assumed to form a methyloxazoline (MeOzn) ring. Likewise,
the Thr2 residue was identified by the COSY correlations of
NH-20/H-19/H-21/H3-22 and the HMBC correlation of H-
19/C-18 (Figure 2). The Orn residue was inferred from the
COSY correlations of NH-12/H-13/H-14/H-15/H-16 and the
HMBC correlations of H-13/C-17 and H-14/C-17 (Figure 2).
The connection of each substructure was completely
determined based on further HMBC analyses. The HMBC
assigned to either 1-OH or 29-OH.
Figure 2. Key HMBC and COSY correlations in 1 and 2.
correlations from two NH protons (NH-12/C-9 and NH-20/
C-23) clearly indicated the connection of Thr1/Orn and
Thr2/Sal2 through amide bonds (Figure 2). In addition, the
HMBC correlations of H-8/C-7 and H-10/C-7 established the
connection of Sal1/Thr1 through methyloxazoline ring
B
DOI: 10.1021/acs.jnatprod.8b00261
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
formation. Finally, the HMBC correlation of H-16/C-18
revealed the amide bond formation between the Orn and Thr2
residues. The above observations could explain 13 out of the
14 degrees of unsaturation in 1, indicating that 1 has one more
ring system. Based on the molecular formula and the absence
of an NH signal at the side chain of the Orn residue, the final
ring system was proposed to be an isoxazolidinone ring
between the Orn and Thr2 residues, which contains an N−O
linkage (Figure 1B). The assignment of the isoxazolidinone
structure was further confirmed by the 13C NMR resonances of
the Thr2 residue, which includes characteristic upfield-shifted
α-carbon (C-19, δC 57.7) adjacent to an amide-type nitrogen
(C(O)−NH−C) (Figure S15).14,15 Finally, we analyzed the
fragment pattern in the ESIMS/MS spectrum of 1, which
supported the proposed structure of 1 (Figure S16).
The molecular formula of catenulobactin B (2) was also
established as C27H30N4O9, based on the HR-ESIMS analysis.
The NMR spectra and the fragment pattern of the ESIMS/MS
spectrum of 2 were similar to those of 1, and the detailed
spectroscopic analyses indicated the presence of two salicylic
acids, two Thr, and Orn substructures with the same sequences
as in 1 (Figures 1B, 2, and S17). In contrast, the 13C NMR
resonances of the Thr2 residue were quite different from those
1
of 1 (Figure S15), and the H NMR signal corresponding to
NH-20 was not observed. Therefore, the Thr2 and Sal2
residues were expected to be connected through the
methyloxazoline ring in 2, and this was further supported by
the HMBC correlation of H-21/C-23 (Figure 2). Moreover, to
satisfy the molecular formula of 2, a hydroxy group should be
attached to the nitrogen atom in the side chain of the Orn
residue (OHOrn, Figure 1B).
Figure 3. (A) Proposed mechanism of preacinetobactin isomerization
to acinetobactin. The hydroxamate oxygen attacks the adjacent
oxazoline ring in an SN2 manner. (B) Proposed mechanism of the
isomerization of 2 to 1 in 0.1 M phosphate buffer (pH = 7.0). (C)
HPLC chromatograms of 2 incubated in 0.1 M phosphate buffer for 5,
40, 80, and 150 min. (D) Key NOESY correlations and vicinal
coupling constants in the isoxazolidinone moiety of 1.
The relationship between 1 and 2 resembles that of
acinetobactin and preacinetobactin, which are bacterial side-
rophores produced by Acinetobacter baumannii (Figure
3A).16−18 Walsh and co-workers reported that acinetobactin
was initially biosynthesized and released from a nonribosomal
peptide synthase (NRPS) assembly line, as the preacineto-
bactin containing an N-OH oxazoline structure.17 However,
the N-OH oxazoline structure was unstable under physiological
conditions (aqueous, pH > 7) and rapidly isomerized to
acinetobactin with an isoxazolidinone structure by the attack of
a hydroxamate oxygen to the oxazoline ring in an SN2 manner
(Figure 3A).17,18 Therefore, the isoxazolidinone ring of 1 was
likely derived from the N-OH oxazoline structure of 2 by
nonenzymatic rearrangement (Figure 3B). To test our
hypothesis, we incubated the purified compound 2 in
phosphate buffer (pH = 7.0). As expected, 2 was indeed
rapidly isomerized to 1 (Figure 3C), suggesting that 1 is likely
an artifact created from 2.
To determine their absolute configurations, 1 and 2 were
subjected to total acid hydrolysis and subsequent chiral-phase
GC-MS analysis. Under the hydrolysis conditions, the MeOzn/
OHOrn moieties were converted to Thr/Orn with retention of
the stereochemistry (Figure S18). The amino acids in the
hydrolysates were then derivatized to N-trifluoroacetyl methyl
esters (Figure S18). Each hydrolysate was subsequently
subjected to chiral-phase GC-MS analysis, and the retention
time was compared with those of amino acid standards
derivatized in the same manner. As a result, L-Thr and D-Orn
were detected in both hydrolysates (Figure S19), indicating
that all of the MeOzn units are derived from L-Thr, and the
Orn residues have the D-configuration in 1 and 2. Considering
the mechanism of the isoxazolidinone ring formation in 1
(Figure 3B), the absolute configuration of C-19 should be
retained between 1 and 2, while that of C-21 was expected to
be inverted. Consistent with this expectation, the large
coupling constant between H-19 and H-21 and the NOESY
correlation of H-19/H3-22 indicated the 1,2-trans relationship
between H-19 and H-21 (Figure 3D). Therefore, the absolute
configuration of the isoxazolidinone ring in 1 was determined
to be (19S, 21S).
Since the chemical structures of catenulobactins are closely
related to those of known bacterial siderophores,16,19,20 we
expected that the catenulobactins would also act as Fe(III)
chelators. To test the Fe(III)-binding activities, 1 and 2 were
treated with an excess amount of tris(2,4-pentanedionato)-
iron(III) (Fe(acac)3) and analyzed by LC-MS. As expected, 2
formed a 1:1 Fe(III) complex, which displayed the character-
istic 56Fe- and 54Fe-containing positive ion species (Figures 4
and S20). In contrast, 1 did not show a noticeable m/z change
upon treatment with Fe(acac)3 (Figure S21), indicating that
the N-OH ornithine and oxazoline moieties of 2 are essential
for the Fe(III)-binding activity, as described in the previous
structure−function studies of acinetobactin analogues.14 We
also evaluated the cytotoxicities of 1 and 2 against P388
murine leukemia cells by the methylthiazole tetrazolium
(MTT) assay.21 2 showed moderate cytotoxicity against
P388 murine leukemia cells, with an IC50 of 22.4 μM, while
1 did not exhibit any cytotoxicity up to a 100 μM
concentration.
In conclusion, we identified two novel heterocyclic peptides,
catenulobactins A (1) and B (2), by combined-culture. To the
best of our knowledge, the catenulobactins are the first
secondary metabolites identified from the genus Catenulo-
C
DOI: 10.1021/acs.jnatprod.8b00261
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
1.0% HP-20 resin, pH = 7.2) and shaken under the same conditions
for 5 days. After the fermentation, a 10 mL portion of the culture
broth was lyophilized, extracted with 10 mL of an equal mixture of
methanol−chloroform, and concentrated in vacuo. The residual
material was dissolved in 0.5 mL of extractant, and an aliquot (10
μL) was analyzed by HPLC with a Cosmosil 5C18-MS-II column (4.6
× 250 mm, Nacalai Tesque, Japan) in a CH3CN (solvent A)−0.05%
formic acid (solvent B) gradient system (solvent A: 20% (5 min), 70%
(12 min), 100% (15−25 min), 1.0 mL min−1). In parallel, we also
prepared the pure cultures of Catenuloplanes/Tsukamurella, and their
metabolic profiles were analyzed in the same manner. The production
levels of 1 and 2 were evaluated based on the peak areas of HPLC
chromatograms detected at 300 nm. Under the combined-culture
conditions, the production of 1 increased 12.9-fold and the
production of 2 increased 19.4-fold, as compared with the pure-
culture conditions. Examination of the temporal production of 1 and
2 during a 5-day cultivation revealed that the formation of 1 and 2 was
not observed at 48, 72, and 96 h.
Isolation of Catenulobactins. Catenuloplanes sp. RD067331 was
combined-cultured with a mycolic acid-containing bacterium on a
large scale (100 mL × 20), as described above. The mixture of the cell
pellet and HP-20 resin was collected by centrifugation (6000 rpm),
lyophilized, and extracted with a 1:1 mixture of methanol−chloroform
(1.0 L). Evaporation of the organic solvent gave a crude extract (3.6
g), which was subjected to flash silica gel column chromatography (19
× 320 mm, methanol−chloroform, 0/100, 10/90, 20/80, 50/50, and
100/0, 50 mL each). Both 1 and 2 were found in the 50/50
methanol−chloroform fraction, and they were further purified by
semipreparative reverse-phase HPLC with a Cosmosil C18-ARII
column (10 × 250 mm, Nacalai Tesque, Japan) in a CH3CN (solvent
A)−0.05% formic acid (solvent B) gradient system (solvent A: 55−
65% (12 min), 100% (13−16 min), 3.0 mL min−1). The final yields of
1 (tR = 8.8 min) and 2 (tR = 10.2 min) were 3.1 and 2.4 mg,
respectively.
Figure 4. LC-MS charts of (i) total ion chromatogram (TIC) of 2
treated with Fe(acac)3, (ii) extracted ion chromatogram (EIC) of 2
treated with Fe(acac)3 at m/z 555.2 ([M + H]+), (iii) EIC of 2
treated with Fe(acac)3 at m/z 608.1 ([M + 56Fe − 2H]+), (iv) TIC of
2 (untreated), (v) EIC of 2 (untreated) at m/z 555.2, and (vi) EIC of
2 (untreated) at m/z 608.1.
Catenulobactin A (1): white powder; [α]25 −19.9 (c 0.033,
planes, illustrating the broad applicability of our combined-
culture strategy. Intriguingly, 2 formed the Fe(III) complex,
suggesting its functional role as a bacterial siderophore. From
the perspective of competition for environmental Fe(III)
species, it is reasonable that Catenuloplanes sp. increased the
production of 2 in response to a mycolic acid-containing
bacterium. Although the detailed mechanism of the formation
of the siderophore still remains to be elucidated, we propose
that physical interactions with a mycolic acid-containing
bacterium is crucial, and this triggers the catenulobactin
biosynthesis as in the case of previously reported cryptic
natural product biosynthetic gene clusters.8,9
D
MeOH); UV (MeOH) λmax (log ε) 206, 238, 303 nm; IR (KBr) νmax
3394, 2980, 2362, 2339, 1683, 1640, 1451, 1309, 1256, 1229, 1074,
1028, 828, 758 cm−1; 1H and 13C NMR data, Table 1; HR-ESIMS m/
z 555.2075 [M + H]+ (calcd. for C27H31N4O9, 555.2092).
Catenulobactin B (2): white powder; [α]25 +55.7 (c 0.023,
D
MeOH); UV (MeOH) λmax (log ε) 206, 244, 306 nm; IR (KBr) νmax
3393, 2925, 2361, 2334, 1721, 1641, 1493, 1454, 1258, 1231, 1074,
1
1036, 825, 758 cm−1; H and 13C NMR data, Table 1; HRESIMS
555.2071 [M + H]+ (calcd for C27H31N4O9, 555.2092).
Isomerization of 2 to 1 in Phosphate Buffer. Compound 2
(0.1 mg) was placed into a microtube containing 0.1 M phosphate
buffer (0.1 mL, pH = 7.0) and incubated at 30 °C. After 5, 40, 80, and
150 min of incubation, 10 μL portions of the solution were subjected
to an HPLC analysis with a Cosmosil 5C18-MS-II column (4.6 × 250
mm, Nacalai Tesque, Japan) in a CH3CN (solvent A)−0.05% formic
acid (solvent B) gradient system (solvent A: 20% (5 min), 70% (12
min), 100% (15−25 min), 1.0 mL min−1), to monitor the conversion
from 2 to 1.
Chiral-Phase GC-MS Analysis. In a sealed tube, each
catenulobactin (0.1 mg) was completely hydrolyzed by heating in 6
M HCl (0.4 mL) at 110 °C for 12 h, and the resulting hydrolysate was
lyophilized to dryness. The hydrolysate was then derivatized with a
methyl ester group by heating in 5−10% HCl−MeOH reagent (0.4
mL, Tokyo Chemical Industry Co., Ltd.) at 110 °C for 30 min. After
the solvent was removed by evaporation, the hydrolysate was further
derivatized with an N-trifluoroacetyl (N-TFA) group, by boiling in an
equal mixture of trifluoroacetic acid anhydride and CH2Cl2 (total 0.4
mL) at 110 °C for 5 min. The reaction mixture was dried under an
argon atmosphere, and the residue was dissolved in acetone (0.3 mL)
and used as the sample for the chiral-phase GC-MS analysis. All of the
amino acid standards (DL-Thr, DL-allo-Thr, and DL-Orn) were
derivatized with a methyl ester−N-TFA in the same manner.
Chiral-phase GC-MS analyses were performed with a CP-Chirasil-
DEX column (Alltech, 0.25 mm × 25 m; He as the carrier gas;
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
obtained with a JASCO DIP-1000 digital polarimeter. The UV
spectra were measured by a Shimadzu UV-1280. Infrared spectra were
recorded with KBr pellets on a Jasco FT/IR-460 Plus spectrometer.
All NMR analyses were performed on a JEOL ECX-500 spectrometer
(1H NMR, 500 MHz; 13C NMR, 125 MHz). The HR-ESIMS and
ESIMS/MS spectra were obtained with a Bruker Compact QqTOF
mass spectrometer.
Bacterial Strains. Catenuloplanes sp. RD067331 was purchased
from the National Institute of Technology and Evaluation (NITE).
The characterization of Tsukamurella pulmonis TP-B0596 was
previously reported.11
Monitoring the Production of Catenulobactins. Initially,
Catenuloplanes sp. RD067331 and T. pulmonis TP-B0596 were
separately precultured in 500 mL flasks containing 100 mL of V-22
medium11 up to full growth (Catenuloplanes: 3 days, Tsukamurella: 2
days) on a rotary shaker at 160 rpm and 30 °C. Then, aliquots of each
preculture (Catenuloplanes: 3 mL, Tsukamurella: 0.3 mL) were
combined in a 500 mL baffled flask containing production medium
(2.5% starch, 1.5% soy flour, 0.2% yeast extract, 0.4% CaCO3, and
D
DOI: 10.1021/acs.jnatprod.8b00261
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Note
program rate: 50−200 °C at 4 °C/min). As a result, L-Thr and D-Orn
were confirmed to be present in both hydrolysates of 1 and 2.
Monitoring the Fe(III)-Binding Activity of Catenulobactins.
Methanol solutions of compounds 1 and 2 (50 μg dissolved in 50 μL
of methanol) were treated with 2.0 equiv of tris(2,4-pentanedionato)-
iron(III) (Fe(acac)3, 64 μg dissolved in 50 μL of methanol). Each
mixture was incubated at room temperature for 1 h and analyzed by
LC-MS with a Cosmosil 2.5C18-MS-II column (2.0 × 75 mm, Nacalai
Tesque, Japan) in a CH3CN (solvent A)−20 mM formic acid (solvent
B) gradient system (solvent A: 20% (3 min), 100% (15−20 min), 0.2
mL min−1). As references, methanol solutions of 1 and 2 (untreated
with Fe(acac)3) were also analyzed by LC-MS in the same manner.
Cytotoxicity Assay. The cytotoxicities of 1 and 2 against P388
murine leukemia cells were evaluated using the MTT assay.21 P388
murine leukemia cells were cultured in RPMI 1640 (Wako
Chemicals) medium, containing 10 μg/mL of penicillin−streptomy-
cin and 10% fetal bovine serum (MP Biomedicals), at 37 °C under a
5% CO2 atmosphere. The methanol solutions of 1 and 2 (100 μL
each; 100, 50, 25, 12.5, 6.3, 3.1, 1.6, and 0.78 μM) were added to the
wells of a 96-well microplate containing 100 μL of 1 × 105 cells/mL
tumor cell suspension. The plates were incubated for 4 days at 37 °C
under a 5% CO2 atmosphere, and then 50 μL of MTT solution (1
mg/mL in DMSO) was added. The plates were further incubated for
4 h under the same conditions. The plates were centrifuged, and the
precipitates were dissolved in 50 μL of DMSO. Finally, the
absorbance at 570 nm was measured with a microplate reader. As a
positive control, we employed doxorubicin, and its IC50 value was
determined to be 5.2 μM.
REFERENCES
■
(1) Watve, M. G.; Tickoo, R.; Jog, M. M.; Bhole, B. D. Arch.
Microbiol. 2001, 176, 386−390.
́
(2) Berdy, J. J. Antibiot. 2005, 58, 1−26.
(3) Nakashima, T.; Takahashi, Y.; Omura, S. Biochem. Pharmacol.
2017, 134, 42−55.
(4) Jensen, P. R.; Moore, B. S.; Fenical, W. Nat. Prod. Rep. 2015, 32,
738−751.
(5) Subramani, R.; Aalbersberg, W. Appl. Microbiol. Biotechnol. 2013,
97, 9291−9321.
(6) Nett, M.; Ikeda, H.; Moore, B. S. Nat. Prod. Rep. 2009, 26,
1362−1384.
(7) Doroghazi, J. R.; Albright, J. C.; Goering, A. W.; Ju, K. S.;
Haines, R. R.; Tchalukov, K. A.; Labeda, D. P.; Kelleher, N. L.;
Metcalf, W. W. Nat. Chem. Biol. 2014, 10, 963−968.
(8) Onaka, H. J. Antibiot. 2017, 70, 865−870.
(9) Onaka, H.; Mori, Y.; Igarashi, Y.; Furumai, T. Appl. Environ.
Microbiol. 2011, 77, 400−406.
(10) Bertrand, S.; Bohni, N.; Schnee, S.; Schumpp, O.; Gindro, K.;
Wolfender, J. L. Biotechnol. Adv. 2014, 32, 1180−1224.
(11) Adnani, N.; Chevrette, M. G.; Adibhatla, S. N.; Zhang, F.; Yu,
Q.; Braun, D. R.; Nelson, J.; Simpkins, S. W.; McDonald, B. R.;
Myers, C. L.; Piotrowski, J. S.; Thompson, C. J.; Currie, C. R.; Li, L.;
Rajski, S. R.; Bugni, T. S. ACS Chem. Biol. 2017, 12, 3093−3102.
(12) Hoshino, S.; Okada, M.; Awakawa, T.; Asamizu, S.; Onaka, H.;
Abe, I. Org. Lett. 2017, 19, 4992−4995.
(13) Hoshino, S.; Wong, C. P.; Ozeki, M.; Zhang, H.; Hayashi, F.;
Awakawa, T.; Asamizu, S.; Onaka, H.; Abe, I. J. Antibiot. 2018, 71,
653−657.
ASSOCIATED CONTENT
■
(14) Shapiro, J. A.; Wencewicz, T. A. Metallomics. 2017, 9, 463−470.
(15) Seyedsayamdost, M. R.; Cleto, S.; Carr, G.; Vlamakis, H.; Joao
̃
S
* Supporting Information
Vieira, M.; Kolter, R.; Clardy, J. J. Am. Chem. Soc. 2012, 134, 13550−
13553.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00261.
(16) Yamamoto, S.; Okujo, N.; Sakakibara, Y. Arch. Microbiol. 1994,
162, 249−254.
(17) Wuest, W. M.; Sattely, E. S.; Walsh, C. T. J. Am. Chem. Soc.
2009, 131, 5056−5057.
Copies of NMR and ESIMS/MS spectra of catenulo-
bactins A and B (1 and 2), 13C NMR resonances of
isoxazolidinone/oxazoline structures in 1 and 2
compared with those of reference compounds, graphical
outline of acid hydrolysis and chiral-phase GC-MS
analysis, GC-MS chromatograms, HR-ESIMS spectrum
of 2-Fe(III) complex, and LC-MS chromatograms of 2
treated with Fe(acac)3 (PDF)
(18) Shapiro, J. A.; Wencewicz, T. A. ACS Infect. Dis. 2016, 2, 157−
168.
(19) Anthoni, U.; Christophersen, C.; Nielsen, P. H.; Gram, L.;
Petersen, B. O. J. Nat. Prod. 1995, 58, 1786−1789.
(20) Actis, L. A.; Fish, W.; Crosa, J. H.; Kellerman, K.; Ellenberger,
S. R.; Hauser, F. M.; Sanders-Loehr, J. J. Bacteriol. 1986, 167, 57−65.
(21) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.
AUTHOR INFORMATION
■
Corresponding Author
*Tel: +81-3-5841-4740. Fax: +81-3-5841-4744. E-mail: abei@
mol.f.u-tokyo.ac.jp.
ORCID
Ikuro Abe: 0000-0002-3640-888X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Grants-in-Aid from the Ministry
of Education, Culture, Sports, Science and Technology
(MEXT), Japan (JSPS KAKENHI Grant Nos. JP16H06443
and JP18H02120), JST/NSFC Strategic International Collab-
orative Research Program, and JSPS Research Fellowships for
Young Scientists (to S.H.). We thank Dr. S. Asamizu (The
University of Tokyo) for helpful advice and the maintenance of
Tsukamurella pulmonis TP-B0596.
E
DOI: 10.1021/acs.jnatprod.8b00261
J. Nat. Prod. XXXX, XXX, XXX−XXX