Isolation, Identification, and Decomposition of Antibacterial Dialkylresorcinols from a Chinese Pseudomonas aurantiaca Strain

Article abstract of DOI:10.1021/acs.jnatprod.9b00315

A chemical investigation of a Chinese Pseudomonas aurantiaca strain has yielded a new benzoquinone (4) and furanone (5), in addition to the known dialkylresorcinols 1 and 2. Extensive decomposition studies on the major metabolite 1 produced an additional furanone derivative (6), a hydroxyquinone (7), and two unusual resorcinol and hydroxyquinone dimers (8 and 9). Structures were elucidated by nuclear magnetic resonance spectroscopy in combination with tandem mass spectrometry analysis. These studies illustrate the potential of artifacts as a source of additional chemical diversity. Compounds 1 and 2 showed moderate antibacterial activity against a panel of Gram-positive pathogens, while the antibacterial activities of the artifacts (4-9) were reduced.

Full text of DOI:10.1021/acs.jnatprod.9b00315

pubs.acs.org/jnp
Article
Isolation, Identification, and Decomposition of Antibacterial
Dialkylresorcinols from a Chinese Pseudomonas aurantiaca Strain
Yue Shi, Diana A. Zaleta-Pinet, and Benjamin R. Clark*
Cite This: https://dx.doi.org/10.1021/acs.jnatprod.9b00315
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: A chemical investigation of a Chinese Pseudomonas
aurantiaca strain has yielded a new benzoquinone (4) and
furanone (5), in addition to the known dialkylresorcinols 1 and
2. Extensive decomposition studies on the major metabolite 1
produced an additional furanone derivative (6), a hydroxyquinone
(7), and two unusual resorcinol and hydroxyquinone dimers (8
and 9). Structures were elucidated by nuclear magnetic resonance
spectroscopy in combination with tandem mass spectrometry analysis. These studies illustrate the potential of artifacts as a source of
additional chemical diversity. Compounds 1 and 2 showed moderate antibacterial activity against a panel of Gram-positive
pathogens, while the antibacterial activities of the artifacts (4−9) were reduced.
ith the widespread use of antibiotics, drug-resistant
bacteria have become increasingly prevalent and some
of detailed chemical analysis of the decomposition of the
dialkylresorcinol 1 under acidic and basic conditions, four new
products were elucidated, including two dimers. This report
differentiates the natural and artificial products produced
during the culture and isolation process, by demonstrating
transformation from dialkylresorcinols into mono- and dimeric
derivatives.
W
infectious pathogens have developed resistance to most
commonly available antibiotics, which increases the chance
of treatment failure.1 Microbial natural products are structur-
ally diverse and can serve as a source of novel molecular
scaffolds for drug development, especially in the field of
antibiotic discovery.2 In addition, metagenomic sequencing of
environmental samples has revealed that the majority of
microbial biodiversity has yet to be chemically explored, often
as a result of the difficulty of growing many species under
laboratory conditions.3 China, a country with a large territory
and highly varied climate, offers significant potential for
microbial natural product discovery,4−6 to discover novel
antibiotics that can meet the resistance challenge.7,8
Pseudomonas is a genus of Gram-negative bacteria commonly
acting as rhizosphere colonizers. They have been reported to
produce more than 100 antibacterial compounds,9 including
lipopeptides,10 lipodepsipeptides,11 diketopiperazines,12 rham-
nolipids,13 pseudotrienic acids,14 and dialkylresorcinols.15 One
of the characteristic metabolite classes produced by the
Pseudomonas genus are the dialkylresorcinols, which have
displayed antimicrobial activity against Gram-positive bacteria,
mycobacteria, yeasts, and filamentous fungi.16,17 A great
number of dialkylresorcinol derivatives have been described
from Pseudomonas and related genera.14−20 Biosynthetic
studies using isotopic labeling strategies have delineated the
biosynthetic pathway, an unusual head-to-head condensation
of two medium-chain-length fatty acid precursors.21
RESULTS AND DISCUSSION
■
The bacterium YM03-Y3 was isolated from a soil sample
collected in the Yellow Mountains, Anhui Province, China, and
was identified as P. aurantiaca by 16S rDNA sequencing.22 The
methanol extract of YM03-Y3 was analyzed by high-perform-
ance liquid chromatography (HPLC) (Figure S2 of the
Supporting Information), showing numerous high-intensity
peaks, indicating the presence of diverse secondary metabo-
lites. In addition, the extract also presented significant
biological activity against a panel of microbial pathogens
(Figure S1 of the Supporting Information), suggesting that it
was a suitable candidate for further investigation. Culture
conditions of YM03-Y3 were optimized by the selection of an
appropriate medium, pH, and temperature to maximize
metabolite production. Fractionation of a large-scale culture
led to the isolation of five compounds 1−5, of which the
structures (Figure 1) were elucidated by detailed spectroscopic
analysis.
Received: April 5, 2019
The current report describes the chemical investigation of a
Chinese isolate of Pseudomonas aurantiaca. In addition to
describing the isolation and structure elucidation of secondary
metabolites, a study of the decomposition of its major
products, the dialkylresorcinols, was carried out. On the basis
© XXXX American Chemical Society and
American Society of Pharmacognosy
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX
A

Journal of Natural Products
pubs.acs.org/jnp
Article
181.1219 (C11H17O2, calculated for 181.1223) corresponding
to losses of C5H10 and C6H12 in compound 1, while m/z
209.1530 (C13H21O2, calculated for 209.1536) and m/z
195.1378 (C12H19O2, calculated for 195.1380) corresponded
to losses of C6H12 and C7H14 in compound 2. The intensity of
fragments resulting from the loss of the alkyl chains at C-2 was
significantly stronger than that resulting from the loss of the C-
5 alkyl chain. Compound 3 was structurally unrelated to the
other compounds in the extract and was identified as 2-
hydroxyphenazine by comparison of high-resolution electro-
spray ionization mass spectrometry (HRESIMS) and 1H NMR
spectra to reported data (Table S1 of the Supporting
Information).12
Compound 4 was isolated as a yellow pigment. The
ultraviolet (UV) spectrum suggested that it possessed a typical
benzoquinone chromophore with absorption maxima at 210,
270, and 390 nm. 13C and heteronuclear multiple-bond
correlation (HMBC) NMR revealed 6 downfield resonances,
including 2 carbonyl signals at 185.2 and 190.0, in addition to
13 resonances consistent with aliphatic carbons (Table 1). A
comparison of the proton and carbon NMR to literature data15
made it clear that compound 4 was an homologue of the
known compound, 2-hexyl-3-hydroxy-5-pentylbenzo-1,4-qui-
none15 (7), with the addition of two additional methylene
groups, which was supported by HRESIMS (measured
305.2126, [M − H]−, C17H30O3, calculated for 305.2122).
HMBC data showed correlations from the two downfield
methylene signals (δH 2.39 and 23.7) (Figure 2) to carbons at
δC 185.2 and 190.0, respectively, indicating the presence of two
carbonyl groups, consistent with a dialkyl-substituted hydrox-
ybenzoquinone. Correlation spectroscopy (COSY) experi-
ments showed a cross peak between 6.40 ppm (H5) and
2.39 ppm (H6a), resulting from a weak four-bond coupling.
The known dialkylhydroxyquinone 7 has been reported as an
oxidation product from compound 1; both possess n-hexyl and
n-pentyl substituents.15 On the basis of a comparison to
literature data, the alkyl groups were assigned as shown for
Figure 1. Natural products and artifacts isolated from P. aurantiaca
(YM03-Y3). All alkyl chains are linear.
Compounds 1 and 2 have been previously reported as
natural products from Pseudomonas sp. Ki19.15,16 Experimental
1H and 13C nuclear magnetic resonance (NMR) and high-
resolution mass spectrometry (HRMS) data corresponded well
with previously reported literature data (Table 2 and Table S1
of the Supporting Information). To further characterize these
molecules, high-resolution tandem mass spectrometry
(HRMS2) was conducted (Figures S3 and S4 of the
Supporting Information), revealing fragments at m/z
195.1378 (C12H19O2, calculated for 195.1380) and m/z
Table 1. NMR Spectroscopic Data for Compounds 4−6 (d4-MeOH, 400 MHz)
compound 4
compound 5
compound 6
δH, mult (J in Hz)
number
δC
δH, mult
number
δC
δH, mult (J in Hz)
δC
1
2
3
4
5
6
3a
3b
3c
3d
3e
3f
6a
6b
6c
6d
6e
6f
6g
185.0
155.1
122.1
189.8
134.5
147.0
23.2
30.1
32.9
30.4
23.7
14.4
29.4
29.0
32.8
30.4
29.2
23.7
14.4
a
a
2
3
4
5
105.9
104.9
145.6
139.3
173.6
204.6
37.5
24.3
29.8
32.4
23.4
14.3
26.1
28.1
32.7
23.6
14.3
a
a
145.7
139.4
173.1
204.4
37.5
24.3
30.1
32.7
23.6
14.4
26.2
28.4
30.2
29.8
32.7
23.7
14.4
7.00, s
6.99, s
6.41, s
2a
2b
2c
2d
2e
2f
2g
4a
4b
4c
4d
4e
4f
4g
2.37, m
1.44, m
1.35, m
1.31, m
1.31, m
0.89, m
2.39, m
1.52, m
1.35, m
1.31, m
1.31, m
1.31, m
0.89, m
2.62, t (7.0)
1.55 m
2.63, t (7.2)
1.56, m
1.36, m
1.30, m
1.30, m
0.91, m
2.30, t (7.6)
1.58, m
1.36, m
1.30, m
1.36, m
1.29, m
1.29, m
0.90, m
2.31, t (7.5)
1.57, m
1.36, m
1.29, m
1.29, m
1.29, m
0.90, m
0.91, m
a
Indicated carbon shifts were obtained from the HMBC spectrum.
B
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 2. NMR Spectroscopic Data for Compounds 1 and 7−9 (d4-MeOH, 400 MHz)
compound 1
compound 7
compound 8, fragment A
compound 8, fragment B
compound 9
a
number
δC
δH, mult
number
δC
δH, mult number
δC
δH, mult number
δC
δH, mult
number
δC
δH, mult
1
157.0
114.6
157.0
107.8
142.2
107.8
24.0
1
184.9
155.1
122.1
189.8
134.5
147.0
23.4
1
185.2
154.4
122.5
189.5
144.3
145.4
23.5
1′
2′
3′
4′
5′
6′
2′a
2′b
2′c
2′d
2′e
2′f
5′a
157.1
115.2
153.6
113.8
139.7
108.3
23.8
1/1′
2/2′
3/3′
4/4′
5/5′
6/6′
3a/3′a
3b/3′b
3c/3′c
3d/3′d
3e/3′e
3f/3′f
6a/6′a
183.2
151.0
121.8
185.7
139.7
142.4
23.0
2
2
2
3
3
3
4
6.13, s
6.13, s
4
4
5
5
6.41, s
5
6
6
6
6.30, s
2a
2b
2c
2d
2e
2f
5a
2.54, t (7.2)
1.49, m
3a
3b
3c
3d
3e
3f
6a
2.38, m
1.44, m
1.35, m
1.31, m
1.31, m
0.92, m
2.40, m
3a
3b
3c
3d
3e
3f
6a
2.44, m
1.47, m
1.37, m
1.14, m
1.14, m
0.78, m
2.23, m
2.12, m
1.37, m
1.22, m
2.59, m
1.47, m
1.31, m
1.31, m
1.31, m
0.87, m
2.13, m
2.42, m
1.44, m
1.26, m
1.26, m
1.26, m
0.82, m
2.31, m
2.07, m
1.26, m
1.26, m
1.21, m
1.21, m
0.86, m
30.3
29.4
30.3
29.4
28.3
33.1
1.31, m
32.8
29.8
33.2
29.5
32.7
1.31, m
30.4
33.0
32.9
32.1
23.6
1.31, m
23.4
23.2
24.1
22.4
14.4
0.90, m
14.4
14.5
14.4
14.4
36.7
2.39, t (7.2)
29.3
30.0
34.7
28.2
5b
5c
32.2
30.6
1.56, m
1.31, m
6b
6c
28.8
32.6
1.54, m
1.35, m
6b
6c
28.8
33.2
5′b
5′c
31.3
30.4
1.31, m
1.31, m
6b/6′b
6c/6′c
28.1
31.7
5d
5e
23.8
14.5
1.31, m
0.90, m
6d
6e
23.6
14.3
1.31, m
0.92, m
6d
6e
28.6
14.4
1.22, m
0.87, m
5′d
5′e
23.7
14.2
1.31, m
0.87, m
6d/6′d
6e/6′e
22.7
14.2
a
Coupling constants in hertz.
attached methylene protons H4a, in turn, showed HMBC
correlations to C3, C4, and C5, while H4b correlated to C4,
confirming the location of one of the aliphatic chains at C4.
For the other chain, the HMBC correlations observed from
two methylene protons H2b at 2.61 ppm and H2c at 1.55 ppm
to C2a (204.1 ppm), indicating that carbonyl carbon was
attached to the second alkyl chain. The only remaining
position to link the two major fragments was to connect C2a to
carbon 2; however, the high chemical shift of C2 and lack of
any directly attached protons suggested that it was further
substituted and that the remaining hydroxy group was located
at this position. As previously established, NMR signals
evidenced the presence of 13 aliphatic carbons (Table 1),
but as a result of the lack of HMBC correlation from H2c/C4b
to a methyl carbon, it was impossible to establish the length of
the individual aliphatic chains using NMR data alone.
Figure 2. Key HMBC (H → C) correlations and MS−MS fragments
for compounds 4−6.
To determinate the specific number of carbons in each
chain, ESIMS2 analysis was employed (Figure S5 of the
Supporting Information). The tandem mass spectrometry
(MS−MS) spectrum resulting from fragmentation of the
protonated molecule at m/z 311.2213 [M + H]+ revealed a
fragment at m/z 197.1171 (C11H17O3, calculated for
197.1172), consistent with cleavage between C2 and C2a
and consequent loss of C7H13O, while a strong signal at m/z
113.0963 (C7H13O, calculated for 113.0961) further supported
this hypothesis. They both indicted that the aliphatic chain at
position 2a was a n-hexyl chain; hence, the other alkyl chain
attached at C4 should be a n-pentyl chain. Thus, the structure
of compound 5 was assigned as shown in Figure 1. Circular
dichroism (CD) analysis of compound 5 did not reveal any
significant absorbances; this was attributed to the likely
presence of a racemic mixture, as detailed further below.
Previous reports described that a solution of compound 1
could turn from colorless to pale yellow when allowed to stand
in deuterated CDCl3 in an NMR tube for 2 days;15,16 a similar
observation was noted in this study. However, only one
oxidation product from compound 1 has been reported; the
dialkylhydroxyquinone 7 mentioned above.15 This change is
compound 4. It is plausible that compound 4 might be formed
by an analogous oxidation of compound 2.
Compound 5 was obtained as a white solid and showed
ultraviolet/visible (UV/vis) absorption maxima at 200, 220,
and 280 nm. Its molecular formula was determined to be
C18H30O4 by HRESIMS, indicating 4 degrees of unsaturation.
The 13C NMR spectrum revealed the presence of 13 aliphatic
carbon signals between 14.4 and 37.6 ppm, including two
1
methyl carbons. In addition to two downfield methylene H
NMR signals at 2.63 and 2.30 ppm, this suggested the presence
of two linear aliphatic chains in the structure. The carbon
signals in the HMBC spectrum (Figure 2) revealed the
presence of five distinct downfield carbons, one sp3 carbon at
105.9 ppm, two alkene carbons at 139.4 and 145.7 ppm, and
two carbonyl groups: one at 204.4 ppm, which suggested a
non-conjugated ketone, and another at 173.1 ppm, suggesting
the presence of an ester. The furanone core was established
with the HMBC spectrum, with the olefinic proton at 7.00
ppm showing four cross peaks to three sp2 carbon signals at
105.9 ppm (C2), 139.4 ppm (C4), 173.1 ppm (C5) and to one
aliphatic carbon at 26.1 ppm, which was assigned as C4a. The
C
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
caused by a two-step oxidation mechanism, previously
proposed by Singh et al.,23 coupled with a similar degradation
pathway reported by Philipp and Schink.24
provided by MS−MS data (Figure S6 of the Supporting
Information), which revealed MS2 fragments, m/z 169.0858
(C9H13O3, calculated for 169.0859) and m/z 113.0964
(C7H13O, calculated for 113.0961), analogous to those
observed for compound 5, which further supported the
assignment of the structure for compound 6 as shown.
Again, CD analysis did not reveal any significant absorbances,
which suggests the presence of a racemic mixture. Given the
artifactual nature of this compound and its likely formation
through an achiral chemical process, this is not surprising.
The dialkylresorcinol−quinone dimer (8) was isolated as a
red powder. The UV spectrum of compound 8 revealed
absorption maxima at 230, 275, 375, and up to 490 nm.
HRESIMS data were consistent with a molecular formula of
C34H52O5. The proton NMR resonances of compound 8
consisted of one aromatic singlet (6.30 ppm), three deshielded
2H methylenes (2.59, 2.44, and 2.13 ppm), one methylene unit
with non-equivalent protons (2.23 and 2.12 ppm), four methyl
groups [0.87 (3×) and 0.78 ppm], and numerous methylene
groups around 1.30 ppm. The 13C NMR spectrum revealed the
presence of 12 carbons in the aromatic region: 189.5, 185.2,
157.1, 154.4, 153.5, 145.3, 144.3, 139.7, 122.5, 115.1, 113.2,
and 108.2 ppm. Taken together and compared to the NMR
data of compounds 1 and 7 suggested that compound 8 might
be a dimer of these two fragments. Carbon signals at 185.2,
154.4, 122.5, 189.5, 144.3, and 145.3 ppm were assigned as
part of hydroxyquinone (fragment A in Figure 4), while 157.1,
115.1, 153.5, 113.2, 139.7, and 108.2 ppm were attributed to a
dialkylresorcinol (fragment B in Figure 4).
Considering the structures of compounds 4 and 5 and the
reports from previous research, it seemed possible that these
compounds might also be artifact products from decom-
position of compound 2. As a preliminary experiment, a small
amount of compound 1 (3 mg) was dissolved in chloroform
(1.5 mL) and allowed to stand at room temperature; the
decomposition process was monitored by HPLC. After 7 days,
three additional major peaks could be observed in the HPLC
chromatogram of the decomposition mixture (Figure 3a).
Figure 3. HPLC isoplot spectrum from 190 to 600 nm for a 7 day
decomposition study of compound 1 (shown in black circles): (a)
artifacts produced after decomposition in CHCl3 and (b) artifacts
produced after decomposition in NaOH−MeOH.
Figure 4. Substructures of compound 8.
The HMBC data associated with substructure A indicated
that there were correlations from a deshielded methylene
proton at δH 2.44 (3a), correlating with carbon signals at 189.5
ppm (C4), 154.4 ppm (C2), and 122.5 ppm (C3), while a pair
of diasterotopic methylene protons at δH 2.23 and 2.12 (6a)
correlated with carbons at 185.2 ppm (C1) and 144.3 ppm
(C5). Crucially, the aromatic singlet of 7 was absent from the
spectrum. A suite of COSY correlations of mutually coupled
methylenes together with 13C NMR chemical shifts of
numerous signals between 14.4 and 32.9 ppm and a number
of deshielded methylenes confirmed the existence of two
aliphatic chains and established the partial structure as
substructure A.
Artifacts can often be a fortuitous route for the discovery of
new structures or stronger bioactivities.25−27 Hence, a larger
scale study (50 mg) with compound 1 was performed to
further explore the decomposition products.
Decomposition of the compound was conducted in the same
manner and yielded three major compounds: compounds 6
(2.4 mg), 7 (3.0 mg), and 8 (4.5 mg). Compound 7 possessed
very similar NMR data to compound 4 and was rapidly
identified as a known dialkylhydroxyquinone, with all
spectroscopic data consistent with literature values.15 However,
compounds 6 and 8 were previously unreported molecules.
A comparison of NMR data and molecular formulas
obtained from HRESIMS (305.1720, [M + Na]+, C16H26O4)
led to the identification of compound 6 as a homologue of
compound 5, with the loss of two methylene units. HMBCs
(Figure 2) as well as 13C NMR data were almost identical to
those of compound 5, with only minor changes to shifts of
alkyl carbons. However, one characteristic difference was
observed, with a change in the chemical shift of C4a to 32.7
ppm (cf. 30.2 ppm in compound 5). This indicated that the C4
alkyl chain had changed in length, while the C2 alkanone chain
was unchanged, providing further evidence for the assigned
structures of both compounds. Additional evidence was
HMBC signals associated with substructure B were similar
to the known dialkylresorcinols, except for the lack of one
aromatic proton at 6.13 ppm. There was a single aromatic
singlet (δH 6.30) correlated with 157.1 (C1′), 115.1 (C2′), and
34.7 (C5′a) and two deshielded methylenes (H2′a and 5′a):
2′a correlated with carbons at 157.1 ppm (C1′), 153.5 ppm
(C3′), and 115.1 ppm C2′), while H5′a correlated with
carbons at 139.7 ppm (C5′), 113.2 ppm (C4′), and 108.2 ppm
1
(C6′), respectively. Substructures A and B account for all H
and 13C NMR signals observed in the NMR spectrum, and
they are consistent with the assigned structure, in which the
D
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
a
Scheme 1. Inferred Decomposition Process of Compound 1, Including Proposed Minor Decomposition Products
a
Structures of compounds 10−15 are tentative and based on HRMS data. All alkyl chains are linear.
two fragments were directly linked from C5 to C4′. Under
these circumstances, a chiral axis may exist along this bond,
which could explain the presence of two non-equivalent
methylene protons (H6) at 2.23 and 2.12 ppm.
oxidatively dimerized with a second molecule of compound 1
to yield the heterodimer 8.
Acid- and base-catalyzed decompositions can often yield
dramatically different suites of compounds.27 Thus, the
decomposition procedure for compound 1 was repeated
under basic conditions, using NaOH−MeOH (pH 10) as
the solvent, which formed a pale purple solution over 7 days.
HPLC monitoring of a small-scale decomposition mixture
(Figure 3B) revealed products with retention times similar to
those formed during the acid decomposition, with one
exception, an intense peak at 21.2 min.
A larger scale experiment (50 mg of compound 1) was
carried out in aqueous methanol adjusted to pH 10 with
NaOH, stirring at room temperature for 10 days. Isolation of
the decomposition products was carried out using semi-
preparative HPLC. Isolation resulted in compounds 6−8,
previously obtained in the acid decomposition, in addition to
one new dimer (compound 9, 5.1 mg). The different colors
observed in acid and base decompositions are likely due to
deprotonation of the hydroxyquinones; bathychromic shifts are
common for phenolic compounds under basic conditions.28
Similar to compound 5, the number of carbons in the C4
aliphatic chain was confirmed by HRESIMS2 (Figure S7 of the
Supporting Information). Some of these fragments were
consistent with the fragmentation of the dimer into its two
principal halves: m/z 265.2159 and 277.1795, which
corresponded to the dialkylresorcinol and hydroxybenzoqui-
none portions of the molecule, respectively. In addition,
fragments at m/z 457.2945 and 471.3107 corresponded to
losses of C5H10 and C6H12, similar to those observed for the
monomeric dialkylresorcinol 1. The remainder of the
fragmentation was more complex; however, identities for
some of the major fragments have been proposed (Scheme S1
and Table S12 of the Supporting Information). In combina-
tion, all spectral data were consistent with the structure shown
(8). Compound 8 might be formed by a two-step reaction: first
dialkylresorcinol 1 is oxidized into hydroxyquinone 7 via
atmospheric oxidation under acid catalysis, which is then
E
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 3. Minimum Inhibitory Concentration (MIC) (μg/mL) of Compounds 1−9 toward American Type Culture Collection
(ATCC) Bacterial Pathogens
Staphylococcus aureus
Staphylococcus epidermidis
Enterococcus hirae
ATCC 8043
Streptococcus mutans
ATCC 25175
Bacillus subtillis
ATCC 6633
compound
ATCC 12600
ATCC 14990
1
2
3
4
5
6
7
8
9
4.2
1.1
a
133
16.7
33.4
a
4.2
8.3
33.4
a
33.4
66.7
a
a
a
0.52
4.2
33.4
a
4.2
33.4
a
133
133
133
66.7
133
133
1.1
8.4
66.7
a
a
a
66.7
133
133
a
a
1.1
33.4
66.7
33.4
a
133
1.1
a
133
0.26
positive
control
b
negative
control
a
a
a
a
a
c
a
b
c
No activity detected at tested concentrations. See Table S7 of the Supporting Information for positive controls. Distilled water was used as a
negative control.
Compound 9 was isolated as a deep yellow solid. HRESIMS
established the molecular formula as C34H50O6. Carbon signals
at 185.8 and 183.0 ppm, in addition to a characteristic UV/vis
spectrum with maximum absorption at 230, 280, and 400 nm
suggested a quinone substructure. There were 17 resonances in
the 13C NMR spectrum, and comparing this to the predicted
molecular formula implied the presence of 2-fold symmetry in
the molecule. Considering that this compound was derived
from compound 1, it was proposed that this might be a
quinone dimer, with the same lengths of the alkyl chains as the
monomeric quinone 7. Ultimately, the structure of compound
9 is proposed as shown in Figure 1. Given the observation of
two pairs of non-equivalent methylene protons (H6a and H6c
and H6′a and H6′c), it suggested the presence of a chiral axis
in the structure. HMBCs between methylene 3a (2.42 ppm)
and C2−C4 (151.0, 121.8, and 185.8 ppm) and from H6a
(2.31 and 2.07 ppm) to C1, C5, and C6 (183.0, 139.6, and
142.3 ppm) were also consistent with the proposed structure.
Additional evidence obtained from HRESIMS2 (Figure S8 of
the Supporting Information) was the fragment at 277.1795
(calculated for C17H25O3, 277.1798) corresponding to the
monomeric fragment; major fragments are listed in Scheme S2
and Table S13 of the Supporting Information.
The structure of the major compounds identified in the
decomposition study, compounds 6−9, also suggested that
compounds 4 and 5, isolated during the initial isolation, could
be artifacts as a result of changes in pH and exposure to oxygen
during the culture and isolation process. It is assumed that
decomposition of compound 2 would yield a similar suite of
products to those derived from compound 1 but with n-heptyl
in place of n-pentyl substitution, although this was not tested in
this study.
In addition to the major products of the decomposition,
several minor products were detected by liquid chromatog-
raphy−mass spectrometry (LC−MS) under both acidic and
basic conditions. We have tentatively proposed structures
(compounds 10−15 in Scheme 1) based on the information
obtained by HRESIMS and the already known decomposition
outcomes (Table S14 of the Supporting Information). On the
basis of the isolated decomposition products, dialkylresorcinol
(1) could first be oxidized into the hydroxyquinones (7 and
10) and, subsequently, further oxidized and rearranged to give
the furanones 6, with the loss of one carbon unit.
Subsequently, hydroxyquinone is dimerized with either
dialkylresorcinol 8 or itself (9) through an aromatic coupling
process. Compound 12 might be an intermediate product in
the process of dimerization to form compound 9. In a similar
manner, hydroxyquinones, dialkylresorcinols, and furanones
might also be dimerized to form either homodimers (13 and
14) or heterodimers (11 and 15) (Scheme 1).
All isolated compounds were screened against a panel of
microorganisms (Table S7 of the Supporting Information).
The results (Figure S1 of the Supporting Information) showed
that compounds were active against Gram-positive bacteria.
Compound 1 showed moderate activity (4−8 μg/mL) against
all strains tested, while compound 2 displayed the strongest
activity against Staphylococcus aureus (1.1 μg/mL); the
decomposition products presented weaker biological activity
compared to the parent compounds (Table 3).
Reasons for this loss of activity are uncertain; however,
oxidation to quinones reduces the polarity of the substituents
and may negatively affect their bioavailability. Likewise, the
formation of dimers increases the molecular mass and may lead
to a reduction in solubility. Little is known about the mode of
action of the dialkylresorcinols; they are known to interact with
the PauR receptor in Photorhabdus asymbiotica in quorum-
sensing processes,29 but the precise details of the interaction
have not yet been delineated.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Samples were dried in a
Savant SC210A SpeedVac concentrator with VLP120 vacuum pump
(Ameritech). HPLC and UV spectra were obtained on an Agilent
spectrophotometer (Agilent 1260 Infinity). Semi-preparative HPLC
purifications were performed on LC-20AR Series semi-preparative
HPLC (Shimadzu, Japan) using a 5 μm Agilent ZORBAX SB-C18, 9.4
× 250 mm column with a flow rate of 2.5 mL/min. 1H and 13C NMR
spectra were recorded on a 400 or 600 MHz spectrometer (Bruker
Avance III) at 25 °C. HRMS was obtained on a Q-TOF micro
spectrometer (Agilent micrOTOF-Q II) and an ESI-Orbitrap
spectrometer (Thermo Fisher Q Exactive HF). Electronic circular
dichroism (ECD) spectra were measured on a Bio-Logic MOS-500
spectrometer using a 1 cm path length quartz cuvette at 2 nm slit size.
Microbial Material. The strain YM03-Y3 was isolated from soil
collected in the Yellow Mountains (Huangshan), Anhui Province,
China, on July 2, 2016. Soil was taken from the top of the mountain,
approximately 1 m away from a river, during the raining season, at
GPS coordinates 30° 08′ 25″ N, 118° 10′ 03″ E. The isolate was
F
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
identified as P. aurantiaca by sequence analysis of 16S rDNA; the
sequence has been submitted to GenBank under accession number
SUB5407677 Seq1 MK744062. It was cultured in Czapek−Dox
medium and stored at −80 °C with glycerol. All strains used for
antimicrobial activity assays in this study are ATCC pathogens (Table
S7 of the Supporting Information).
Strain Identification. Purified samples of bacteria were used for
DNA extractions. DNA isolation, amplification, and sequence analysis
were carried out using an adaption of existing protocols (page S42 of
the Supporting Information).22
Standard Analytical HPLC Method. Column, ZORBAX SB-C18,
3.5 μm, 4.6 × 150 mm; flow rate, 0.8 mL/min; and gradient, 10%
acetonitrile (ACN) in the first minutes, up to 100% ACN at 15 min,
and retained at 100% ACN for 3 min before returning to original
conditions for the next run.
2-Hexyl,5-heptyl-1,3-resorcinols (2): White, amorphous powder;
UV (MeOH) λmax, 210, 270 nm; 1H and 13C NMR data, see Table S1
of the Supporting Information; positive HRESIMS ion, m/z 293.2468
[M + H]+ (calculated for C19H33O2, 293.2475).
2-Hydroxyphenazine (3): Orange, amorphous powder; UV
1
(MeOH) λmax, 210, 260, 350, 400 nm; H and 13C NMR data, see
Table S1 of the Supporting Information; positive HRESIMS ion, m/z
197.0713 [M + H]+ (calculated for C12H9N2O, 197.0709).
Hydroxyquinone (4): Yellow, amorphous powder; UV (MeOH)
1
λmax, 210, 270, 390 nm; H and 13C NMR data, see Table S3 of the
Supporting Information; negative HRESIMS ion, m/z 305.2126 [M −
H]− (calculated for C19H29O3, 305.2122).
Furanone (5): White, amorphous powder; UV (MeOH) λmax, 200,
220, 280 nm; 1H and 13C NMR data, see Table 1 and Table S2 of the
Supporting Information; negative HRESIMS ion, m/z 309.2073 [M −
H]− (calculated for C18H29O4, 309.2071).
Isolation and Purification. After 10 days of fermentation under
optimized conditions (shaking under 200 rpm, at 30 °C, in a 2 L
Erlenmeyer flask containing 1 L of Czapek−Dox broth, total 15 L
broth), cell and supernatant fractions of the cultures were separated
by centrifugation at 4000 rpm for 20 min. Supernatant was extracted
by liquid−liquid extraction with ethyl acetate (22.5 L) and then n-
butanol (22.5 L) to yield 902 mg and 1.35 g of crude extract,
respectively. Biomass was separately extracted with methanol, yielding
600 mg of crude extract. All fractions possessed similar HPLC profiles
and were combined and concentrated together in a SpeedVac prior to
further separation. The combined extracts (2.85 g) were separated
through on a HP20 resin column using a stepwise elution of
methanol−water (200 mL of each gradient) and finally ethyl acetate
to give four fractions (100% H2O, 50% H2O−MeOH, 100% MeOH,
and 100% EtOAc). The 100% H2O fraction contained largely large
polarity metabolites and water-soluble medium components and other
high-polarity metabolites and was not investigated further. The three
remaining fractions were (1.97 g) recombined and fractionated using
silica gel open column chromatography with a hexane−ethyl acetate
gradient (step gradient: hexane−EtOAc of 20:1, 10:1, 5:1, 3:1, 2:1,
1:1, 1:3, 1:5, 1:10, and 1:20, 100% EtOAc, 100% EtOH, 150 mL of
each gradient) to afford 10 fractions (F1−F10). The major fraction, F3
(650 mg), was subjected to semi-preparative HPLC (linear gradient
elution of 55:45−100:0 ACN−H2O over 60 min and then kept at
100% ACN for 20 min) to yield dialkylresorcinol (1, tR = 15.5 min
with standard analytical HPLC gradient, 125 mg) and dialkylresorci-
nol (2, tR = 16.5 min, 110 mg). The remaining material of fraction F3
was further purified by another round of preparative HPLC (linear
gradient elution of 65:35−85:15 ACN−H2O over a period of 40 min,
then gradient elution up to 100% ACN over 10 min, and kept at 100%
ACN for 15) to give a new furanone derivative (5, tR = 17.1 min, 2.4
mg). Fraction F10 (155 mg) was separated by semi-preparative HPLC
using a linear gradient elution of 25:75−70:30 ACN−H2O over 50
min to yield 2-hydroxyphenazine (3, tR = 9.1 min, 3.5 mg). The deep
red-colored fraction F1 (420 mg) was fractionated using HPLC with
an isocratic elution of 90% ACN−H2O and yielded quinone (4, tR =
19.3 min, 4.5 mg).
Furanone (6): White, amorphous powder; UV (MeOH) λmax, 200,
220, 280 nm; 1H and 13C NMR data, see Table 1 and Table S4 of the
Supporting Information; positive HRESIMS ion, m/z 305.1720 [M +
Na]+ (calculated for C16H26O4Na, 305.1723).
Hydroxyquinone (7): Yellow, amorphous powder; UV (MeOH)
λmax, 210, 270, 390 nm; 1H and 13C NMR data, see Table 2; negative
HRESIMS ion, m/z 277.1814 [M − H]− (calculated for C17H25O3,
277.1809).
Dialkylresorcinol−Hydroxyquinone Dimer (8): Red, amorphous
1
powder; UV (MeOH) λmax, 230, 275, 375, and 490 nm; H and 13C
NMR data, see Table 2 and Table S5 of the Supporting Information;
positive HRESIMS ion, m/z 541.3887 [M − H]− (calculated for
C34H53O5, 541.3888).
Hydroxyquinone Dimer (9): Yellow, amorphous powder; UV
(MeOH) λmax, 230, 280, 400 nm; 1H and 13C NMR data, see Table 2
and Table S6 of the Supporting Information; positive HRESIMS ion,
m/z 553.3536 [M − H]− (calculated for C34H49O6, 553.3535).
Antimicrobial Testing. For agar well diffusion assays, dried
samples were dissolved in sterile water and dimethyl sulfoxide
(DMSO, <5%) to make a sample solution with a concentration at 10
mg/mL. Appropriate antibiotics (2 mg/mL; see Table S7 of the
Supporting Information) were used as the positive control, while
sterile water with DMSO (<5%) was used as the negative control.
ATCC pathogens were suspended in sterile water until optical
density at 585 nm reached 1.00 ( 0.02) [cell density is about 108
colony-forming units (CFU)/mL]. A 200 μL pathogen suspension
was spread on the appropriate solid media and let to stand for 10 min.
Wells (5 mm) were then punched out of the media plates using a
sterile pipet tip. Samples (30 μL) were added to the well. Plates were
incubated for 12 h at the optimal growth temperature (see Table S7
of the Supporting Information), and inhibition zones were observed
and measured. For MIC determination, a modified form of a literature
protocol was used with appropriate antibiotics as the positive control
and sterile water/DMSO (<5%) as the negative control (page S43 of
the Supporting Information).30
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.9b00315.
■
Large-Scale Decomposition Study. Acid Decomposition.
Compound 1 (50 mg) was dissolved with 10 mL of chloroform
and stirred at room temperature for 10 days. Purification of the
resulting mixture was carried out using reversed-phase preparative
HPLC (gradient elution from 55 to 90% ACN over 40 min), to yield
compounds 6 (tR = 15.6 min, 2.4 mg), 7 (tR = 17.8 min, 3.0 mg), and
8 (tR = 19.1 min, 4.5 mg).
Base Decomposition. Compound 1 (50 mg) was dissolved in
methanol (20 mL), with the pH adjusted to 10 with 0.1 mol/L
NaOH, and then stirred at room temperature for 10 days. Isolation of
the decomposition products was performed using reversed-phase
semi-preparative HPLC as described above to yield compounds 6−8
again and compound 9 (tR = 21.2 min, 5.1 mg).
sı
Tabulated NMR data for compounds 2−9, including
two-dimensional (2D) data for compounds 4−9, details
on antimicrobial test strains, HRMS2 data for com-
pounds 1, 2, 5, 6, 8, and 9, NMR and HRESIMS spectra
for compounds 1−9, and detailed protocols for microbe
identification and antimicrobial tests (PDF)
AUTHOR INFORMATION
Corresponding Author
■
2-Hexyl,5-pentyl-1,3-resorcinol (1): White, amorphous powder;
UV (MeOH) λmax, 210, 270 nm; 1H and 13C NMR data, see Table 2;
positive HRESIMS ion, m/z 265.2155 [M + H]+ (calculated for
C17H29O2, 265.2162).
Benjamin R. Clark − School of Pharmaceutical Science and
Technology, Tianjin University, Tianjin 300072, People’s
G
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
(19) Beresovsky, D.; Hadas, A.; Livne, O.; Sukenik, A.; Kaplan, A.;
Carmeli, S. Isr. J. Chem. 2006, 46, 79−87.
Republic of China; orcid.org/0000-0001-8987-7092;
Phone: +86-22-87401830; Email: bclark@tju.edu.cn
(20) Imai, S.; Fujioka, K.; Furihata, K.; Furihata, K.; Seto, H. J.
Antibiot. 1993, 46, 1323−1325.
Authors
(21) Nowak-Thompson, B.; Hammer, P. E.; Hill, D. S.; Stafford, J.;
Yue Shi − School of Pharmaceutical Science and Technology,
Tianjin University, Tianjin 300072, People’s Republic of China
Diana A. Zaleta-Pinet − School of Pharmaceutical Science and
Technology, Tianjin University, Tianjin 300072, People’s
Republic of China
́
Torkewitz, N.; Gaffney, T. D.; Lam, S. T.; Molnar, I.; Ligon, J. M. J.
Bacteriol. 2003, 185, 860−869.
(22) Kolbert, C. P.; Persing, D. H. Curr. Opin. Microbiol. 1999, 2,
299−305.
(23) Singh, U. S.; Scannell, R. T.; An, H.; Carter, B. J.; Hecht, S. M.
J. Am. Chem. Soc. 1995, 117, 12691−12699.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.9b00315
(24) Philipp, B.; Schink, B. J. Bacteriol. 1998, 180, 3644−3649.
(25) Akee, R. K.; Ransom, T.; Ratnayake, R.; McMahon, J. B.;
Beutler, J. A. J. Nat. Prod. 2012, 75, 459−463.
Notes
(26) Clark, B. R.; Capon, R. J.; Lacey, E.; Tennant, S.; Gill, J. H. Org.
Biomol. Chem. 2006, 4, 1520−1528.
The authors declare no competing financial interest.
(27) Clark, B. R.; Capon, R. J.; Lacey, E.; Tennant, S.; Gill, J. H. Org.
Biomol. Chem. 2006, 4, 1512−1519.
ACKNOWLEDGMENTS
■
(28) Scott, A. I. Interpretation of the Ultraviolet Spectra of Natural
Products; Pergamon Press: Oxford, U.K., 1964; pp 95−96.
(29) Brameyer, S.; Kresovic, D.; Bode, H. B.; Heermann, R. Proc.
Natl. Acad. Sci. U. S. A. 2015, 112, 572−577.
(30) Wayne, P. A. Methods for Dilution Antimicrobial Susceptibility
Tests for Bacteria That Grow Aerobically: Approved Standard; Clinical
and Laboratory Standards Institute (CLSI): Wayne, PA, 2015.
This work was funded in part by the award from the Research
Fund for International Young Scientists (81850410550) from
the National Science Foundation of China and a grant from
the National Basic Research Program of China
(2015CB856500). The authors acknowledge Y. Gao for
assistance in acquisition of MS−MS data and Prof. J. De
Voss for his helpful comments on the decomposition
mechanism.
REFERENCES
■
́
́
́
(1) Bolivar, P.; Cruz-Paredes, C.; Hernandez, L. R.; Juarez, Z. N.;
Sanchez-Arreola, E.; Av-Gay, Y.; Bach, H. J. Ethnopharmacol. 2011,
137, 141−147.
(2) Koehn, F. E.; Carter, G. T. Nat. Rev. Drug Discovery 2005, 4,
206−220.
(3) Bachmann, B. O.; Van Lanen, S. G.; Baltz, R. H. J. Ind. Microbiol.
Biotechnol. 2014, 41, 175−184.
(4) Patwardhan, B.; Warude, D.; Pushpangadan, P.; Bhatt, N. Evid-
Based. Compl. Alt. 2005, 2, 465−473.
(5) Wang, Q.; Hu, Z. Q.; Li, X. X.; Wang, A. L.; Wu, H.; Liu, J.; Cao,
S. G.; Liu, Q. S. J. Nat. Prod. 2018, 81, 2531−2538.
(6) Sun, N.; Zhu, Y.; Zhou, H. F.; Zhou, J. F.; Zhang, H. Q.; Zhang,
M. K.; Zeng, H.; Yao, G. M. J. Nat. Prod. 2018, 81, 2673−2681.
(7) Wang, J. J.; Chen, F. M.; Liu, Y. H.; Liu, Y. X.; Li, K. L.; Yang, X.
L.; Liu, S. W.; Zhou, X. F.; Yang, J. J. Nat. Prod. 2018, 81, 2722−2730.
(8) Yang, L.; Li, H. X.; Wu, P.; Mahal, A.; Xue, J. H.; Xu, L. X.; Wei,
X. Y. J. Nat. Prod. 2018, 81, 1928−1936.
(9) Feklistova, I. N.; Maksimova, N. P. Microbiology 2008, 77, 176−
180.
(10) Raaijmakers, J. M.; De Bruijn, I.; Nybroe, O.; Ongena, M.
FEMS Microbiol. Rev. 2010, 34, 1037−1062.
(11) Bassarello, C.; Lazzaroni, S.; Bifulco, G.; Lo Cantore, P.;
Iacobellis, N. S.; Riccio, R.; Gomez-Paloma, L.; Evidente, A. J. Nat.
Prod. 2004, 67, 811−816.
(12) Mehnaz, S.; Saleem, R. S. Z.; Yameen, B.; Pianet, I.;
Schnakenburg, G.; Pietraszkiewicz, H.; Valeriote, F.; Josten, M.;
Sahl, H. G.; Franzblau, S. G.; Gross, H. J. Nat. Prod. 2013, 76, 135−
141.
(13) Sharma, A.; Jansen, R.; Nimtz, M.; Johri, B. N.; Wray, V. J. Nat.
Prod. 2007, 70, 941−947.
(14) Pohanka, A.; Broberg, A.; Johansson, M.; Kenne, L.; Levenfors,
J. J. Nat. Prod. 2005, 68, 1380−1385.
(15) Pohanka, A.; Levenfors, J.; Broberg, A. J. Nat. Prod. 2006, 69,
654−657.
(16) Kanda, N.; Ishizaki, N.; Inoue, N.; Oshima, M.; Handa, A.;
Kitahara, T. J. Antibiot. 1975, 28, 935−942.
(17) Kato, S.; Shindo, K.; Kawai, H.; Matsuoka, M.; Mochizuki, J. J.
Antibiot. 1993, 46, 1024−1026.
(18) Stodola, F. H.; Weisleder, D.; Vesonder, R. F. Phytochemistry
1973, 12, 1797−1798.
H
https://dx.doi.org/10.1021/acs.jnatprod.9b00315
J. Nat. Prod. XXXX, XXX, XXX−XXX