Characterization of diketopiperazine heterodimers as potential chemical markers for discrimination of two dominant black aspergilli, Aspergillus niger and Aspergillus tubingensis

Article abstract of DOI:10.1016/j.phytochem.2020.112399

Black aspergilli are distributed worldwide and represent one of the most prolific sources of metabolites with biomedical and agrochemical interests. However, due to their similar morphological characteristics and insufficient molecular identification, the taxonomic classification of black aspergilli remains ill-defined. The production of specialised metabolites is often unique for species among black aspergilli and could be used as diagnostic chemical markers for species identification. In this study, chemical investigation of Aspergillus tubingensis OUCMBIII 143291 led to the discovery of the diagnostic chemical marker asperazine, a complex diketopiperazine heterodimer, as well as two previously undescribed analogues, asperazine B and C. In addition, an undescribed 2-benzylpyridin-4(1H)-one-containing amide, pestalamide D, along with four known related metabolites were isolated. Their chemical structures, including their absolute configurations, were established on the basis of comprehensive spectral analysis and chiral HPLC analysis of the acidic hydrolysates. Asperazines B and C can serve as potential chemical markers for distinguishing A. tubingensis from A. niger, two representative species of black aspergilli that are usually incorrectly identified. Moreover, the isolated compounds were evaluated for their antifungal activity against eight phytopathogenic fungi including Alternaria alternata, A. brassicae, Botrytis cinerea, Colletotrichum lagenarium, Fusarium oxysporum, Gaeumannomyces graminis, Penicillium digitatum, and Valsa mali. Pestalamide D exhibited significant activities against B. cinerea, C. lagenarium, and V. mali, with MIC values of 4, 8, and 8 μg/mL, respectively, compared with the positive controls carbendazim (MICs = 8, 4, and 4 μg/mL) and prochloraz (MICs = 8, 8, and 4 μg/mL). The results of this study reveal two additional chemical markers and provide a powerful tool for the rapid identification of black aspergilli.

Full text of DOI:10.1016/j.phytochem.2020.112399

Phytochemistry 176 (2020) 112399
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Characterization of diketopiperazine heterodimers as potential chemical
markers for discrimination of two dominant black aspergilli, Aspergillus niger
and Aspergillus tubingensis
Ce Xua,b,1, Kuo Xu , Xiao-Long Yuan , Guang-Wei Ren , Xiao-Qiang Wang , Wei Li , Ning Deng ,
a,1
a
a
a
c
a
a,∗∗
a,∗
Xiu-Fang Wang , Peng Zhang
a
Tobacco Research Institute of Chinese Academy of Agricultural Sciences, Qingdao, 266101, People's Republic of China
Graduate School of Chinese Academy of Agricultural Sciences, Beijing, 100081, China
College of Marine Life Sciences, Ocean University of China, Qingdao, 266003, People's Republic of China
b
c
A R T I C L E I N F O
A B S T R A C T
Keywords:
Black aspergilli are distributed worldwide and represent one of the most prolific sources of metabolites with
biomedical and agrochemical interests. However, due to their similar morphological characteristics and in-
sufficient molecular identification, the taxonomic classification of black aspergilli remains ill-defined. The
production of specialised metabolites is often unique for species among black aspergilli and could be used as
diagnostic chemical markers for species identification. In this study, chemical investigation of Aspergillus tu-
bingensis OUCMBIII 143291 led to the discovery of the diagnostic chemical marker asperazine, a complex di-
ketopiperazine heterodimer, as well as two previously undescribed analogues, asperazine B and C. In addition,
an undescribed 2-benzylpyridin-4(1H)-one-containing amide, pestalamide D, along with four known related
metabolites were isolated. Their chemical structures, including their absolute configurations, were established
on the basis of comprehensive spectral analysis and chiral HPLC analysis of the acidic hydrolysates. Asperazines
B and C can serve as potential chemical markers for distinguishing A. tubingensis from A. niger, two representative
species of black aspergilli that are usually incorrectly identified. Moreover, the isolated compounds were
evaluated for their antifungal activity against eight phytopathogenic fungi including Alternaria alternata, A.
brassicae, Botrytis cinerea, Colletotrichum lagenarium, Fusarium oxysporum, Gaeumannomyces graminis, Penicillium
digitatum, and Valsa mali. Pestalamide D exhibited significant activities against B. cinerea, C. lagenarium, and V.
mali, with MIC values of 4, 8, and 8 μg/mL, respectively, compared with the positive controls carbendazim
Aspergillus tubingensis Mosseray (discellaceae)
Black aspergilla
Chemotaxonomy
Diketopiperazine heterodimers
Asperazines
Antifungal activity
(
MICs = 8, 4, and 4 μg/mL) and prochloraz (MICs = 8, 8, and 4 μg/mL). The results of this study reveal two
additional chemical markers and provide a powerful tool for the rapid identification of black aspergilli.
1. Introduction
industries for the production of extracellular or intracellular enzymes
(
such as amylases and lipases), organic acids (such as citric acid and
The Aspergillus (Aspergillaceae) genus has been well studied and
gluconic acid), and antibiotics in food fermentation processes (Samson
et al., 2007; Varga et al., 2011). Besides their economic importance,
many species can cause food spoilage and contamination by synthe-
sizing mycotoxins (such as ochratoxin A and fuminosins) (D'hooge
et al., 2019). Among the described species, A. niger (Moniliaceae) and
A. tubingensis (Discellaceae) are the most representative species of black
aspergilli (Mirhendi et al., 2016). A. niger is ubiquitous in soil and may
cause black mould on certain fruits and vegetables, such as grapes,
onions, and peanuts (Abdel-Aziz et al., 2019), while A. tubingensis is
proved to be the predominant producer of diverse specialised metabo-
lites with high economic, agronomic, and/or medical importance. The
section Nigri within this genus, also known as black aspergilli, com-
prises 26 closely related species (e.g., A. acidus, A. aculeatus, A. brasi-
liensis, A. carbonarius, A. heteromorphus, A. ibericus, A. lacticoffeatus, A.
japonicus, A. niger, A. tubingensis) that have mainly been isolated from
soil and plant sources (Varga et al., 2011; Xanthopoulou et al., 2019).
They are an important group in the food mycology and biotechnology
∗
Corresponding author.
Corresponding author.
∗∗
E-mail addresses: wangxiufang02@caas.cn (X.-F. Wang), zhangpeng@caas.com (P. Zhang).
These authors contributed equally to this work.
1
https://doi.org/10.1016/j.phytochem.2020.112399
Received 29 January 2020; Received in revised form 27 April 2020; Accepted 27 April 2020
0031-9422/ © 2020 Elsevier Ltd. All rights reserved.

C. Xu, et al.
Phytochemistry 176 (2020) 112399
frequently responsible for the postharvest decay of some economic
fruits and vegetables (Apaliya et al., 2017; Perrone et al., 2007).
Although the section Nigri of Aspergillus has been studied for nearly
determination, chemotaxonomic significance, and antifungal activity of
these compounds.
9
0 years (the first species was discovered by Raoul Mosseray in 1934),
taxonomic classification in the Nigri section is a remaining challenge
D'hooge et al., 2019). Due to their extremely similar morphological
2
. Results and discussion
(
2.1. Structural elucidation of the isolated compounds
characteristics and insufficient molecular identification, distinguishing
these species is difficult and, in many cases, has resulted in mis-
classifications and conflicting phylogenetic trees (Nielsen et al., 2009).
The taxonomic classification of black aspergilli can be successfully
achieved based on a combination of phenotypic, chemotaxonomic,
molecular and chromatographic approaches (Xanthopoulou et al.,
Compound 2 was isolated as a white amorphous powder. Its mole-
cular formula was determined to be C40
ESI HRMS peak at m/z 681.2812 [M + H] (calcd. 681.2820), which
was in accordance with the H and C NMR data. Detailed analysis of
the H, C, and HSQC NMR spectroscopic data of 2 (Table 1) revealed
36 6 5
H N O from the prominent
+
1
13
1
13
2
019). New molecular DNA-based techniques for rapid and accurate
differentiation, such as amplified fragment length polymorphism
AFLP), restriction fragment length polymorphism (RFLP), random
amplified polymorphic DNA (RAPD), and multilocus sequence typing
3
the presence of four methylenes, five sp methines, one quaternary
carbon, 26 aromatic/olefinic carbons (17 of which were protonated),
(
and four amide carbonyls. Moreover, the 1H NMR spectrum showed
additionally six exchangeable protons [δ
NH), 9.64 (s, 30-NH), 7.88 (s, 35-NH), 7.96 (s, 38-NH), and 9.27 (s, 44-
OH)] in DMSO‑d . Comparison of the 1D NMR spectroscopic data of 2
with those of asperazine (1) revealed their structural similarities
Varoglu et al., 1997). The main differences were the 16 additional
mass units in the molecular weight of 2, as well as the presence of an
H
6.65 (s, 1-NH), 8.19 (s, 14-
(
(
MLST), have been investigated and applied in recent years
Xanthopoulou et al., 2019). Alternatively, unique chemical markers
6
provide an efficient chemotaxonomic methodology to identify these
closely-related species (Nielsen et al., 2009). Asperazine, an unusual
unsymmetrical diketopiperazine dimer isolated from A. tubingensis was
proposed as a distinctive chemical marker to distinguish A. tubingensis
from A. niger (Choque et al., 2018; Nielsen et al., 2009). Genomic
analyses revealed that A. tubingensis has the capacity to produce as-
perazine, while none of the A. niger strains possessed this capacity
(
2
oxygenated sp quaternary carbon (δ
2
stitution of a phenylalanine (Phe) unit for tyrosine (Tyr) in one of the
monomeric subunits (Ding et al., 2008). The key 1H– H COSY and
HMBC data of 2 further confirmed the presence of the Tyr residue
C
156.1, C-44) in 2, suggesting that
is also a diketopiperazine heterodimer equivalent to 1 with the sub-
1
(
Choque et al., 2018). Recently, we isolated A. tubingensis from inter-
tidal zone sediment. To satisfy our curiosity that whether this strain can
produce the chemical marker asperazine or its derivatives, we per-
formed chemical investigation of this fungal strain. As a result, asper-
azine (1) and two undescribed analogues, asperazine B (2) and C (3),
were subsequently isolated and identified (Fig. 1). Asperazines B (2)
and C (3) provided additional chemical markers for distinguishing A.
tubingensis and other related species of black aspergilli. In addition, an
undescribed 2-benzylpyridin-4(1H)-one-containing amide, pestalamide
D (4), and four known related metabolites (5–8) were also isolated
(
3
Fig. 2). Interpretation of the selected HMBC correlations from H-2 to C-
, C-4, C-12, and C-24; from H-12 to C-3, C-4, and C-24; and from H-25
to C-3 established a linkage envisioned between C-3 and C-24, which
was previously reported in asperazine (Varoglu et al., 1997) and pes-
talazine A (Ding et al., 2008) but was different from that in the C ,N
connected asperazine A (Li et al., 2015), cristatumin E (Li et al., 2013),
WIN 64821 and WIN 64745 (Barrow et al., 1993; Ovenden et al., 2004).
Accordingly, the collective 1D and 2D NMR data allowed the gross
structure of 2 to be assembled as shown in Fig. 1.
3
30
-
(
Fig. 1). In the present study, we reported the isolation, structure
The NOESY experiment was applied to assign the relative
Fig. 1. Structures of compounds 1–8.
2

C. Xu, et al.
Phytochemistry 176 (2020) 112399
Table 1
1
H (500 MHz) and13C NMR (125 MHz) data of compounds 2 and 3 in DMSO‑d
6
No.
Compound 2
(mult, J in Hz)
Compound 3
(mult, J in Hz)
δ
H
δ
C
, type
δ
H
C
δ , type
1
2
3
4
5
6
7
8
9
1
-NH
6.65 (1H, s)
5.83 (1H, s)
7.22 (1H, s)
5.71 (1H, s)
82.0, CH
57.0, C
132.2, C
79.1, CH
57.3, C
130.8, C
123.6, CH
119.0, CH
128.5, CH
110.0, CH
148.7, C
55.6, CH
6.77 (1H, d, 7.5)
6.53 (1H, t, 7.5)
7.01 (1H, overlap)
6.69 (1H, t, 7.5)
123.6, CH 7.04 (1H, overlap)
118.3, CH 6.59 (1H, t, 7.5)
128.3, CH 6.94 (1H, overlap)
110.2, CH 6.71 (1H, d, 7.5)
147.5, C
1
3.40 (1H, overlap)
56.0, CH
37.8, CH
168.3, C
3.90 (1H, dd, 12.0,
.5)
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
3
3
2a
2b
3
3.07 (1H, overlap)
2.24 (1H, t, 12.0)
2
3.03 (1H, overlap)
1.90 (1H, t, 12.0)
36.4, CH
167.4, C
2
4-NH 8.19 (1H, s)
5
6
7a
7b
8
8.14 (1H, s)
4.44 (1H, m)
4.02 (1H, m)
58.2, CH
166.9, C
38.6, CH
58.2, CH
165.0, C
38.8, CH
2
3.04 (1H, overlap)
2.88 (1H, overlap)
2
3.04 (1H, overlap)
2.98 (1H, 14.0, 3.5)
136.0, C
136.6, C
9, 23 7.03 (2H, overlap)
0, 22 7.04 (2H, overlap)
129.5, CH 6.98 (2H, overlap)
128.3, CH 6.99 (2H, overlap)
126.9, CH 7.14 (1H, overlap)
123.2, C
118.5, CH 6.93 (1H, overlap)
118.4, CH 6.96 (1H, overlap)
118.7, CH 7.45 (1H, d, 7.5)
129.1, C
129.7, CH
128.0, CH
126.4, CH
126.6, C
118.0, CH
118.4, CH
118.6, CH
128.7, C
1
4
5
6
7
8
9
7.12 (1H, t, 7.5)
6.92 (1H, overlap)
6.98 (1H, overlap)
7.53 (1H, d, 7.5)
Fig. 3. Key NOESY correlations of compounds 2 and 3.
2
indicating that H-11 and H -17 were on the same face of the diketo-
132.8, C
132.5, C
piperazine ring (Fig. 3). Additional NOESY correlations of H-11 with H-
0-NH 9.64 (1H, s)
1
2
3a
3b
10.24 (1H, s)
124.9, CH 7.03 (1H, overlap)
108.9, C
2
(δ
H
H
5.83) and H-25 (δ 6.92), and of H-2 with H-25, revealed that H-
6.97 (1H, overlap)
124.7, CH
109.0, C
28.2, CH
11, H-2, and the C-24 indole moiety were located on the same face of
the pyrrolidine ring. The absence of correlated signals for H-34 (δ
3.35) with H-37 (δ 3.42) led to the tentative assignment of these two
protons as having a trans configuration, which was supported by a
biosynthetic analogy to the stereochemistry of H-11 and H-15 for as-
perazine and asperazine A (Varoglu et al., 1997; Ding et al., 2008).
The absolute configuration of 2 was determined by acidic hydrolysis
and chiral HPLC analysis (Ding et al., 2008; Li et al., 2013; Zhang et al.,
H
3.11 (1H, overlap)
2.84 (1H, overlap)
27.9, CH
2
3.08 (1H, overlap)
2.85 (1H, dd, 15.0,
2
H
5
.0)
3
3
3
3
3
3
4
4
4
3.35 (1H, overlap)
54.3, CH
3.36 (1H, overlap)
7.97 (1H, s)
54.7, CH
5-NH 7.88 (1H, s)
6
7
8-NH 7.96 (1H, s)
9
0a
0b
167.6, C
55.1, CH
167.6, C
54.5, CH
3.42 (1H, overlap)
3.45 (1H, overlap)
7.93 (1H, s)
167.1, C
37.1, CH
167.0, C
37.6, CH
2
2
015). The acidic hydrolysates of 2 and standard compounds, D- and L-
Tyr, D- and L-Phe, were prepared and subjected to chiral HPLC analysis.
The retention times of the hydrolysates of 2 (t 14.1 and 25.9 min) were
the same as those of authentic D-Tyr (t 14.1) and D-Phe (t 25.9 min),
respectively, but were different from those of authentic L-Tyr (t
1.6 min) and L-Phe (t 38.0) (Fig. 4), suggesting the presence of D-Tyr
2.90 (1H, overlap)
2.60 (1H, dd, 13.5,
.5)
2
2.97 (1H, overlap)
2.71 (1H, dd, 13.5,
5.0)
4
R
4
4
4
4
4
1
125.8, C
136.1, C
R
R
2, 46 6.62 (2H, d, 8.0)
3, 45 6.90 (2H, d, 8.0)
4
131.1, CH 7.10 (2H, overlap)
114.8, CH 7.20 (2H, overlap)
130.1, CH
127.9, CH
126.6, CH
R
2
R
156.1, C
7.13 (1H, overlap)
and D-Phe residues in 2. Thus, the absolute configurations of 2 were
defined as (2R, 3R, 11S, 15R, 34S, 37R).
4-OH 9.27 (1H, s)
Compound 3 was also isolated as a white amorphous powder. On
+
configuration of the diketopiperazine rings. In the NOESY spectrum, H-
basis of ESI HRMS analysis (m/z 665.2866 [M + H] , calcd.
1
3
1 (δ
H
3.40) exhibited unambiguous correlated signals with H
2
-17 (δ
4.02),
H
665.2871), the molecular formula of
C
40
H
36
N
6
O
4
was assigned.
Analysis of the 1H, C and DEPT NMR data for 3 (Table 1) and its 2D
13
.04 and 2.88) but showed no correlation with H-15 (δ
H
Fig. 2. Key HMBC and 1H– H COSY correlations of compounds 2, 3, and 4.
1
3

C. Xu, et al.
Phytochemistry 176 (2020) 112399
presence of 19 carbons, which were classified as two methyls (including
a methoxy group), two methylenes, one sp3 methine, and seven aro-
matic/olefinic methines, three aromatic/olefinic quaternary carbons,
and four carbonyl carbons. The 1D NMR data were highly similar to
those of pestalamide B, an amide containing a 2-benzylpyridin-4(1H)-
one substructure previously isolated from the plant pathogenic fungus
Pestalotiopsis theae (Ding et al., 2008). An additional methoxy group
(δH/C 3.58/51.5, C-21) was observed in 4, suggesting 4 as the methyl
ester of pestalamide B. The absolute configuration of C-18 in 4 was
assigned as S by comparison of its specific rotation ([α]25 D –10,
MeOH) with that of pestalamide B ([α]25 D –8, MeOH). It is worth
mentioning that specialised metabolites bearing 2-benzylpyridin-4(1H)-
one have been encountered in relatively few fungi outside of the black
aspergilli clade (Henrikson et al., 2011). Notably, to better understand
the distribution of 2-benzylpyridin-4(1H)-ones among black aspergilli,
Henrikson et al. analyzed 11 black aspergilli strains and found that
pestalamide B was the only metabolite common to all 11 strains
(Henrikson et al., 2011).
Fig. 4. The chiral HPLC analyses of the hydrolysates of compounds 2 and 3.
NMR spectra (Fig. 2) revealed that 3 has the same planar structure as
asperazine (1) (Varoglu et al., 1997). However, the chemical shifts of
the carbons surrounding the two chiral centers, C-2 and C-3, exhibited
clear changes when measured in the same solvent, DMSO‑d
chemical shifts of C-2 (δ 79.1), C-4 (δ 130.8), and C-12 (δ 36.4) of 3
82.0, 132.2, 37.6 for 1, respectively), while C-3
126.6) were shifted downfield (δ 57.0 and
23.6 for 1). These changes indicated that 3 was the C-2 and C-3 ste-
3.90) exhibited
3.04), but showed no
4.44). These results indicated that H-11 and
-17 were on the same face of the diketopiperazine-ring, while H-15
was on the opposite side as in 1 and 2 (Fig. 3). Additional unambiguous
NOESY correlations of H-2 (δ 5.71) with H-25 (δ 6.93) and H-30 (δ
3.90)
6
. The
2.2. Study of chemotaxonomic significance
C
C
C
were shifted upfield (δ
C
As mentioned above, black aspergilli species are phylogenetically
closely related, and their taxonomic classification is challenging and
has changed numerously in the last few decades (D'hooge et al., 2019;
Samson et al., 2007; Varga et al., 2011). In general, fungal species can
be distinguished by a combined polyphasic approach including eva-
luation of their morphological characteristics (colony morphology,
conidial size, and ornamentation), biochemical properties, and mole-
cular analysis (Bathoorn et al., 2013; D'hooge et al., 2019). Un-
fortunately, most of the species belonging to black aspergilli are mor-
phologically indistinguishable. Diagnostic molecular DNA-based
techniques, such as PCR-RFLP and RAPD-PCR, appear to be the most
useful and effective methods to successfully facilitate correct species
determination (Bathoorn et al., 2013).
In this study, we obtained two fungal strains, A. niger and A. tu-
bingensis, and performed species identification based on morphological
characteristics and molecular analysis. For the morphological char-
acteristics, both strains possessed very similar microscopic character-
istics, such as dark brown and spiny conidia, spherical vesicles, and
lightly pigmented hyphae near the apex (Fig. 5) (Mirhendi et al., 2016).
These two strains can be recognized as black aspergilli at only the genus
level and cannot be distinguished from one another using morpholo-
gical characteristics. Next, the strains were subjected to molecular re-
identification. The ITS region, 18S and 26S rRNA genes were amplified
and sequenced. Based on the sequences of the ITS rDNA and 18S and
(
δ
C
57.3) and C-24 (δ
C
C
1
reoisomer of 1. In the NOESY experiment, H-11 (δ
unambiguous correlated signals with H -17 (δ
correlation with H-15 (δ
CH
H
2
H
H
2
H
H
H
1
0.24), as well as the absence of the correlated signals of H-11 (δ
H
with H-2 and H-30 revealed that H-2 and the C24-indole moiety were
located both on the other face of the pyrrolidine-ring. The absense of
the correlated signals of H-34 (δ
H
H
3.36) with H-37 (δ 3.45) indicated
that these two protons were in trans-positions as in 1 and 2. To further
determine the configurations of the amino acid residues, acidic hy-
drolysates of 3 were analyzed by chiral HPLC (Fig. 4), which showed
the presence of D-Phe (t
Compound 4 was isolated as colorless oil. Its molecular formula of
was established based on pseudomolecular ion peaks at m/
R
25.9 min).
19 20 2 5
C H N O
z 357.1438 [M + H] (calcd. 357.1445) and 379.1255 [M + Na]+
+
(
calcd. 379.1264) in the HRESIMS spectrum. In accordance with the
1
3
proposed molecular formula, its C NMR data (Table 2) revealed the
Table 2
26S rRNA regions, phylogenetic analyses showed that the two fungal
1
H (500 MHz) and13C NMR (125 MHz) data of compound 4 in DMSO‑d
6
strains of A. niger and A. tubingensis were placed in different evolu-
tionary positions (Fig. 6). Therefore, the molecular identification of
these fungal strains was too ambiguous to determine their taxonomic
status.
The production of specialised metabolites is often unique for species
within black aspergilli and could be used as diagnostic chemical mar-
kers for identification (Mirhendi et al., 2016). Since A. tubingensis was
often misidentified as A. niger, very recently, whole-genome sequencing
of A. tubingensis was carried out (Choque et al., 2018; Nielsen et al.,
No.
δ
H
(mult, J in Hz)
C
δ , type
1
2
3
4
5
6
7
8
13.11 (1H, s)
8.46 (1H, s)
143.9, CH
115.1, C
177.5, C
118.6, CH
152.4, C
6.34 (1H, s)
3.94 (2H, s)
37.9, CH
136.8, C
2
2009). The genomic analyses suggested that A. tubingensis has the ca-
9
1
1
1
1
1
1
1
1
1
2
2
, 13
0, 12
1
4
5
7.30–7.34 (2H, overlap)
7.32–7.36 (2H, overlap)
7.27 (1H, m)
128.7, CH
128.9, CH
127.0, CH
163.3, C
pacity to produce asperazine (Choque et al., 2018). Comprehensive
analysis of the specialised metabolites from 140 strains of A. niger re-
vealed that none of the A. niger strains possessed the capacity to pro-
duce asperazine, proposing asperazine as a distinctive chemical marker
to distinguish A. tubingensis from A. niger (Nielsen et al., 2009). This was
also concluded on the basis of the genomic features of the putative
secondary metabolite clusters classification for asperazine biosynthesis
12.75 (1H, s)
6
173.0, C
7a
7b
8
9
0
2.89 (1H, dd, 15.5, 8.5)
3.14 (1H, dd, 15.5, 8.0)
2.86 (1H, overlap)
34.4, CH
2
41.3, CH
175.5, C
16.9, CH
51.5, CH
(
Choque et al., 2018). Since the gene clusters might be silent, to eval-
1.14 (3H, d, 6.5)
3.58 (3H, s)
3
3
uate the authentic capacity to produce asperazine, crude extracts of A.
tubingensis and A. niger in PDB medium were analyzed by HPLC-DAD.
1
4

C. Xu, et al.
Phytochemistry 176 (2020) 112399
Fig. 5. Culture and morphological characteristics of Aspergillus niger (A) and A. tubingensis (D), colonies on PDA media; Conidia and conidiophores heads of A. niger
B–C) and A. tubingensis (E–F).
(
The results indicated that both species can produce dimeric naptho-γ-
pyrones, which represent the most abundant family of specialised me-
tabolites in black aspergilli (Choque et al., 2014; Nielsen et al., 2009).
However, the HPLC-DAD profile of A. tubingensis showed a prominent
peak attributed to asperazine (Fig. 7). This is the first time to report
asperazine isolated from A. tubingensis. Moreover, the isolation of two
further asperazine derivatives may serve as additional chemical mar-
kers, which provide a powerful tool for the rapid identification of black
aspergilli.
3. Conclusions
In conclusion, chemical investigation of the fungal strain A. tu-
bingensis OUCMBIII 143291 led to the isolation and identification of
eight specialised metabolites, including two undescribed diketopiper-
azine heterodimers, asperazine B (2) and C (3), and an undescribed 2-
benzylpyridin-4(1H)-one-containing amide pestalamide D (4). It has
been reported that A. niger and A. tubingensis are the two most re-
presentative species of black aspergilli, and due to their similar mor-
phological characteristics and insufficient molecular identification, A.
tubingensis has usually been misidentified as A. niger. This study firstly
reported the isolation of asperazine from A. tubingensis. Furthermore,
asperazines B (2) and C (3) were characterized as additional chemical
markers distinguishing A. tubingensis from other related strains, pro-
viding a powerful tool for the rapid identification of black aspergilli. In
addition, the undescribed compound 4 exhibited significant activity
against the necrotrophic pathogen B. cinerea, which may reveal the
chemical basis of the biocontrol efficacy of A. tubingensis.
2.3. Antifungal activity
Microbial specialised metabolites served as a rich resource for ex-
ploring lead compounds with medicinal and/or agricultural importance
Cao et al., 2019). A previous study indicated the good performance of
(
A. tubingensis as a potential biocontrol agent against grey mould on
tomato caused by Botrytis cinerea (Zhao et al., 2018). Thus, the spe-
cialised metabolites of this strain might possess considerable antifungal
activities and could be used as lead compounds with agrochemical
significance. Compounds 1–8 were evaluated for their antifungal ac-
tivity against eight widely distributed pathogenic fungi (A. alternata, A.
brassicae, B. cinerea, C. lagenarium, F. oxysporum, G. graminis, P. digi-
tatum, and V. mali). As shown in Table 3, compounds 2, 4, and 6 dis-
played potent and specific antifungal activity, with MIC values of 4 or
4. Experimental section
4.1. General experimental procedures
Optical rotations were measured with a Jasco P-1020 digital po-
larimeter. UV spectra were obtained with a Shimadzu UV-2700 spec-
trophotometer. ECD spectra were collected with a Jasco J-815-150S
circular dichroism spectrometer. Standard 1D and 2D NMR spectra
were recorded on an Agilent DD2 500 MHz NMR spectrometer (500 and
125 MHz for 1H and C, respectively) with TMS as internal standard.
HRESIMS spectra were measured on an LTQ Orbitrap XL spectrometer.
Commercially available silica gel (100–200 mesh and 200–300 mesh,
Qingdao Marine Chemical Co.), Lobar LiChroprep RP-18 (40–60 μm,
Merck), and Sephadex LH-20 (Merck) were used for open column
chromatography (CC). Precoated silica gel GF254 plates were used for
analytical TLC. All solvents used were either chemical or analytical
grade.
8
μg/mL. Compound 2 demonstrated strong activity against V. mali
with an MIC value of 4 μg/mL, which was equal to that of the positive
controls carbendazim and prochloraz (MICs = 4 μg/mL). Compound 4
exhibited significant activity against B. cinerea, C. lagenarium, and V.
mali, with MIC values of 4, 8, and 8 μg/mL, respectively. Interestingly,
compound 4 showed higher growth-inhibitory activity (MIC = 4 μg/
mL) than carbendazim and prochloraz towards B. cinerea (MICs = 8 μg/
mL) (Table 3). B. cinerea attacks more than 200 crops of economic
importance and represents one of the most extensively studied necro-
trophic pathogens. A. tubingensis was evaluated as a potential biocontrol
agent against B. cinerea (Zhao et al., 2018), and the isolation of com-
pound 4 described herein may provide the chemical basis for its bio-
control efficacy.
13
5

C. Xu, et al.
Phytochemistry 176 (2020) 112399
Fig. 6. Neighbor-joining (NJ) trees based on the 18S (A), 26S (B) and ITS rDNA regions (C). HQM289 and TB3291 represent A. niger and A. tubingensis, respectively.
NJ bootstrap values were estimated and marked above the branches.
6

C. Xu, et al.
Phytochemistry 176 (2020) 112399
neighbor-joining (NJ) model. Branch confidences were estimated based
on 1000 bootstrap replications for each gene.
4.4. Fermentation, extraction, and isolation
The fungal strain was cultivated in 100 × 1 L Erlenmeyer flasks,
each containing 300 mL of modified PDB liquid medium (2% mannitol,
% glucose, 0.3% peptone, 0.5% yeast extract, and 300 mL of naturally
sourced and filtered seawater). Static fermentation was carried out for
0 days at room temperature. Afterwards, the fresh mycelia and culture
1
3
broth were homogenized and then exhaustively extracted three times
with EtOAc. The filtrates were evaporated under reduced pressure at
45 °C to afford 12 g of soluble crude extract. Initial chromatographic
separation of the crude extract was performed by vacuum liquid chro-
matography (VLC) on silica gel, eluting with mixed petroleum ether
(PE)−EtOAc and dichloromethane (DCM)−MeOH to give five fractions
(Frs. 1–5). Since the HPLC-DAD spectrum of Fr. 3 (eluted with
PE–EtOAc 1:1, v/v) contained absorptions characteristic of asperazine,
this fraction was further purified by reversed-phase CC on Lobar
Fig. 7. HPLC-DAD spectra of the crude extracts of A. niger (A) and A. tubingensis
(
B) cultivated in PDB medium. (Chromatographic conditions: Column: SunFire®
C18, 250 mm × 4.60 mm, 5 μm; Mobile phase: MeCN–H O, 0–30 min,
0%–100% MeCN; Flow rate: 1 mL/min; UV detection: 235 nm).
2
3
4.2. Fungal material and genomic DNA extraction
LiChroprep RP-18 with a stepwise MeOH–H
0:0) to give five subfractions (Frs. 3.1–3.5). Fr. 3.1 (1.0 g) was further
purified by silica gel eluting with a DCM−MeOH gradient (from 30:1 to
0:1) and then by Sephadex LH-20 (MeOH) to obtain compounds 7
13.6 mg) and 8 (10.5 mg). Fr. 3.2 (38 mg) was subjected to Sephadex
2
O gradient (from 1:9 to
1
The fungal strains Aspergillus niger Tiegh. (Moniliaceae) and
Aspergillus tubingensis Mosseray (Discellaceae) were previously isolated
from the intertidal sediment sample, collected in Dongying, China
1
(
(
longitude 119°08′E; latitude 37°45′N), in June 2014. Morphologically,
LH-20 (MeOH) and silica gel (DCM−acetone 2:1, v/v) CC to yield
compounds 4 (3.9 mg) and 6 (5.5 mg). Fr. 3.3 (95 mg) was separated on
silica gel (PE–EtOAc 1:1, v/v) to yield compound 5 (7.2 mg). Further
purification of Fr. 3.4 (120 mg) by preparative TLC (pTLC; plate size,
their black colonies enabled them to be easily classified as species of
black aspergilli. Next, molecular reidentification was carried out. Both
strains were cultured for 3 days at 28 °C under shaking conditions of
120 rpm in PDB medium. The mycelia were harvested, ground into a
2
2
0 × 20 cm; precoating, GF254; developing solvents, DCM–MeOH,
0:1) and then by Sephadex LH-20 (MeOH) afforded compounds 1
fine powder with liquid nitrogen and stored at −80 °C. The genomic
DNA of these fungi was extracted from the powder with a protocol
adapted from the trimethyl ammonium bromide (CTAB) method (Gao
et al., 2019). The genomic DNA concentration and quality were esti-
mated using a Nanodrop 2000 (Thermo Fisher Scientific, Wilmington,
MA). A. niger and A. tubingensis were preserved at the College of Marine
Life Sciences, Ocean University of China, under voucher numbers
HQM289 and OUCMBIII 143291, respectively.
(
7.8 mg) and 2 (20.0 mg). Finally, compound 3 (4.0 mg) was isolated
from Fr. 3.5 (60 mg) by silica gel (DCM−acetone 5:1, v/v) and
Sephadex LH-20 (MeOH).
4.4.1. Asperazine B (2)
White amorphous powder; [α]25 D +80 (c 0.5, MeOH); UV (MeOH)
1
13
λ
max (log ε) 224 (4.20), 286 (3.91), 303 (3.92); H and C NMR data,
+
see Table 1; ESI HRMS at m/z 681.2812 [M + H] (calcd for
4.3. PCR amplification, DNA sequencing and phylogenetic analysis
40 37 6 5
C H N O , 681.2820).
To infer the evolutionary locations of these two isolated strains, we
4.4.2. 0Asperazine C (3)
performed phylogenetic analysis using the sequences of the internal
transcribed spacer rDNA (ITS region) and the 18S and 26S ribosomal
RNA genes of these two species, together with those from the other
Aspergillus species (Table S1), respectively. Detailed operating proce-
dures including the information of the primers were presented in the
Supplementary data. After PCR amplification, these three amplicons
were separated and then sequenced by the Sanger method. Subse-
quently, the ITS rDNA and 18S and 26S rRNA sequences were aligned
with those of other known species using Mega 6. Then, the alignments
were used to construct the phylogenetic tree with Mega 6 using the
White amorphous powder; [α]25 D +57 (c 0.6, MeOH); UV (MeOH)
1
13
λ
max (log ε) 225 (4.36), 286 (4.01), 301 (4.00) nm; H and C NMR
+
data, see Table 1; ESI HRMS at m/z 665.2866 [M + H] (calcd for
, 665.2871).
40 37 6 4
C H N O
4.4.3. Pestalamide D (4)
Colorless oil; [α]25 D –10 (c 0.4, MeOH); UV (MeOH) λmax (log ε)
1
13
246 (4.04), 311 (4.01) nm; H and C NMR data, see Table 2; ESI
+
HRMS at m/z 357.1438 [M + H] (calcd for C19
21 2 5
H N O , 357.1445)
and m/z 379.1255 [M + Na]+ (calcd for C19
H
20
N
O
2
5
Na, 379.1264).
Table 3
Antifungal activities of compounds 1–8 (MIC, μg/mL).
no.
A. alternata
A. brassicae
B. cinerea
C. lagenarium
F. oxysporum
G. graminis
P. digitatum
V. mali
1
2
3
4
5
6
7
8
32
16
> 64
> 64
> 64
32
32
> 64
8
32
32
16
16
> 64
16
16
> 64
8
16
> 64
16
> 64
4
> 64
16
32
> 64
8
8
> 64
32
> 64
8
> 64
> 64
> 64
> 64
4
32
32
32
> 64
> 64
16
> 64
> 64
4
> 64
32
> 64
> 64
16
> 64
> 64
> 64
4
> 64
> 64
> 64
32
16
16
> 64
> 64
4
16
4
32
8
16
8
16
> 64
4
Ca*
Pr*
16
8
8
4
16
4
*
Positive controls, carbendazim (Ca) and prochloraz (Pr).
7

C. Xu, et al.
Phytochemistry 176 (2020) 112399
4.5. Acidic hydrolysis of compounds 2 and 3 (Zhang et al., 2015)
onion bulbs. Postharvest Biol. Technol. 156, 110959.
Apaliya, M.T., Zhang, H., Yang, Q., Zheng, X., Zhao, L., Kwaw, E., Mahunu, G.K., 2017.
Hanseniaspora uvarum enhanced with trehalose induced defense-related enzyme ac-
tivities and relative genes expression levels against Aspergillus tubingensis in table
grapes. Postharvest Biol. Technol. 132, 162–170.
Compounds 2 (2.2 mg) and 3 (3.0 mg) were separately hydrolyzed
in HCl solution (6 mol/L, 10 mL) at 110 °C for 24 h. The hydrolysate
was further evaporated under vacuum at 55 °C. The obtained residues
were subjected to chiral HPLC analysis, which was performed with the
aid of Daicel Chiral Technologies (China) Co., Ltd., through the fol-
lowing procedure: The samples were separately dissolved in mobile
Barrow, C.J., Cai, P., Snyder, J.K., Sedlock, D.M., Sun, H.H., Cooper, R., 1993. WIN
6
4821, a new competitive antagonist to substance P, isolated from an Aspergillus
species: structure determination and solution conformation. J. Org. Chem. 58,
6016–6021.
Bathoorn, E., Salazar, N.E., Sepehrkhouy, S., Meijer, M., de Cock, H., Haas, P.J., 2013.
Involvement of the opportunistic pathogen Aspergillus tubingensis in osteomyelitis of
the maxillary bone: a case report. BMC Infect. Dis. 13, 59.
4
phase (pH 1.0 HClO a. q.; final concentration of 1.0 mg/mL) and then
analyzed on a Shimadzu LC-20A system equipped with a CROWNPAK®
CR (+) column (CRP0CB-QB037) (0.40 cm I.D. × 15 cm L × 5 μm)
and SPD-20A detector (205 nm). The injection volume was 8 μL, the
flow rate was 0.4 mL/min, and the temperature was 25 °C. The presence
of D-Tyr and D-Phe was confirmed by comparing the obtained retention
times with those of standards (D-Tyr, 14.1 min; L-Tyr, 21.6 min; L-Phe,
Cao, F., Meng, Z.H., Mu, X., Yue, Y.F., Zhu, H.J., 2019. Absolute configuration of
bioactive azaphilones from the marine-derived fungus Pleosporales sp. CF09-1. J. Nat.
Prod. 82, 386–392.
Choque, E., El Rayess, Y., Raynal, J., Mathieu, F., 2014. Fungal naphtho-γ-pyr-
ones−secondary metabolites of industrial interest. Appl. Microbiol. Biotechnol. 99,
1081–1096.
Choque, E., Klopp, C., Valiere, S., Raynal, J., Mathieu, F., 2018. Whole-genome sequen-
cing of Aspergillus tubingensis G131 and overview of its secondary metabolism po-
tential. BMC Genom. 19, 200–215.
25.9 min; D-Phe 38.0 min (Supporting Information)). Standard amino
acids were purchased from Shanghai Yuanye Bio-Technology Co., Ltd.
D’hooge, E., Becker, P., Stubbe, D., Normand, A.C., Piarroux, R., Hendrickx, M., 2019.
Black aspergilli: a remaining challenge in fungal taxonomy? Med. Mycol. 57,
7
73–780.
4.6. Antifungal assay
Ding, G., Jiang, L., Guo, L., Chen, X., Zhang, H., Che, Y., 2008. Pestalazines and pesta-
lamides, bioactive metabolites from the plant pathogenic fungus Pestalotiopsis theae.
J. Nat. Prod. 71, 1861–1865.
Gao, Y., Li, L., Zhang, X., Wang, X., Ji, W., Zhao, J., Ozaki, Y., 2019. CTAB-triggered Ag
aggregates for reproducible SERS analysis of urinary polycyclic aromatic hydro-
carbon metabolites. Chem. Commun. 55, 2146–2149.
The antifungal activity of the isolated compounds 1–8 against eight
phytopathogenic fungi (Alternaria alternata, Alternaria brassicae, Botrytis
cinerea, Colletotrichum lagenarium, Fusarium oxysporum,
Gaeumannomyces graminis, Penicillium digitatum, and Valsa mali) were
determined in 96-well microtitration plates with a modified method
previously described in the literature (Xiao et al., 2014). All the plant
pathogens were purchased from Qingdao Agricultural University
Qingdao, People's Republic of China). Carbendazim and prochloraz,
ubiquitously applied as broad-spectrum fungicides in agriculture, were
chosen as positive controls.
Henrikson, J.C., Ellis, T.K., King, J.B., Cichewicz, R.H., 2011. Reappraising the structures
and distribution of metabolites from black aspergilli containing uncommon 2-benzyl-
4H-pyran-4-one and 2-benzylpyridin-4(1H)-one systems. J. Nat. Prod. 74,
1
959–1964.
Li, X.B., Li, Y.L., Zhou, J.C., Yuan, H.Q., Wang, X.N., Lou, H.X., 2015. A new diketopi-
perazine heterodimer from an endophytic fungus Aspergillus niger. J. Asian Nat. Prod.
Res. 17, 182–187.
Li, Y., Sun, K.L., Wang, Y., Fu, P., Liu, P.P., Wang, C., Zhu, W.M., 2013. A cytotoxic
pyrrolidinoindoline diketopiperazine dimer from the algal fungus Eurotium herbar-
iorum HT-2. Chin. Chem. Lett. 24, 1049–1052.
(
Mirhendi, H., Zarei, F., Motamedi, M., Nouripour-Sisakht, S., 2016. Aspergillus tubingensis
and Aspergillus niger as the dominant black Aspergillus, use of simple PCR-RFLP for
preliminary differentiation. J. Mycol. Med. 26, 9–16.
Author contributions
Nielsen, K.F., Mogensen, J.M., Johansen, M., Larsen, T.O., Frisvad, J.C., 2009. Review of
secondary metabolites and mycotoxins from the Aspergillus niger group. Anal. Bioanal.
Chem. 395, 1225–1242.
Ovenden, S.P.B., Sberna, G., Tait, R.M., Wildman, H.G., Patel, R., Li, B., Steffy, K.,
Nguyen, N., Meurer-Grimes, B.M., 2004. A diketopiperazine dimer from a marine-
derived isolate of Aspergillus niger. J. Nat. Prod. 67, 2093–2095.
Perrone, G., Susca, A., Cozzi, G., Ehrlich, K., Varga, J., Frisvad, J.C., Meijer, M., Noonim,
P., Mahakarnchanakul, W., Samson, R.A., 2007. Biodiversity of Aspergillus species in
some important agricultural products. Stud. Mycol. 59, 53–66.
Samson, R.A., Noonim, P., Meijer, M., Houbraken, J., Frisvad, J., Varga, J., 2007.
Diagnostic tools to identify black aspergilli. Stud. Mycol. 59, 129–145.
Varga, J., Frisvad, J.C., Kocsubé, S., Brankovics, B., Tóth, B., Szigeti, G., Samson, R.A.,
C. Xu and K. Xu contributed equally to this work.
Declaration of competing interest
The authors declare no competing financial interest.
Acknowledgments
This work was supported by the Agricultural Science and
Technology Innovation Program (Grant No. ASTIP-TRIC04 and ASTIP-
TRIC05) and Major Projects [No.110201601023 (LS-03) and
2
011. New and revisited species in Aspergillus section Nigri. Stud. Mycol. 69, 1–17.
Varoglu, M., Corbett, T.H., Valeriote, F.A., Crews, P., 1997. Asperazine, a selective cy-
totoxic alkaloid from a sponge-derived culture of Aspergillus niger. J. Org. Chem. 62,
7078–7079.
Xanthopoulou, A., Ganopoulos, I., Tryfinopoulou, P., Panagou, E.Z., Osanthanunkul, M.,
Madesis, P., Kizis, D., 2019. Rapid and accurate identification of black aspergilli from
grapes using high-resolution melting (HRM) analysis. J. Sci. Food Agric. 99, 309–314.
Xiao, J., Zhang, Q., Gao, Y.Q., Tang, J.J., Zhang, A.L., Gao, J.M., 2014. Secondary me-
tabolites from the endophytic Botryosphaeria dothidea of Melia azedarach and their
antifungal, antibacterial, antioxidant, and cytotoxic activities. J. Agric. Food Chem.
110201901040 (LS-03)].
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.phytochem.2020.112399.
6
2, 3584–3590.
Zhang, P., Li, X.M., Wang, J.N., Li, X., Wang, B.G., 2015. Prenylated indole alkaloids from
the marine-derived fungus Paecilomyces variotii. Chin. Chem. Lett. 26, 313–316.
Zhao, J., Liu, W., Liu, D., Lu, C., Zhang, D., Wu, H., Dong, D., Meng, L., 2018.
Identification and evaluation of Aspergillus tubingensis as a potential biocontrol agent
against grey mould on tomato. J. Gen. Plant Pathol. 84, 148–159.
References
Abdel-Aziz, M.M., Emam, T.M., Elsherbiny, E.A., 2019. Effects of Mandarin (Citrus re-
ticulata) peel essential oil as a natural antibiofilm agent against Aspergillus niger in
8