Identification and characterization of an anthrol reductase from: Talaromyces islandicus (Penicillium islandicum) WF-38-12

Article abstract of DOI:10.1039/c9gc03072g

An NADPH-dependent oxidoreductase from Talaromyces islandicus WF-38-12 has been identified through genome analysis. It has been shown to catalyze a regio- and stereoselective reduction of anthrols (formed in situ by the reduction of anthraquinones in the presence of Na₂S₂O₄) to (R)-dihydroanthracenones, with high enantiomeric excess (>99%). The implications of results on the biosynthesis of deoxygenated (bis)anthraquinones and modified (bis)anthraquinones are discussed.

Full text of DOI:10.1039/c9gc03072g

Green Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Identification and characterization of an anthrol
reductase from Talaromyces islandicus
(Penicillium islandicum) WF-38-12†
Cite this: Green Chem., 2019, 21,
6594
Received 31st August 2019,
Accepted 6th November 2019
Shailesh Kumar Singh, ‡ Amit Mondal, ‡ Nirmal Saha
Syed Masood Husain
and
DOI: 10.1039/c9gc03072g
*
rsc.li/greenchem
An NADPH-dependent oxidoreductase from Talaromyces islandi- yet another example, versicolorin A is being reduced, che-
cus WF-38-12 has been identified through genome analysis. It has moenzymatically using AflM and NADPH in the presence of
been shown to catalyze a regio- and stereoselective reduction of Na2S2O4 to reduced versicolorin A hydroquinone 9, which has
anthrols (formed in situ by the reduction of anthraquinones in the been proposed as a putative biosynthetic intermediate for afla-
presence of Na2S2O4) to (R)-dihydroanthracenones, with high enan- toxin biosynthesis.3
tiomeric excess (>99%). The implications of results on the bio-
Moreover, the isolation of related deoxyanthraquinones
synthesis of deoxygenated (bis)anthraquinones and modified (bis) such as islandicin (10), aloe-emodin (11), and rhein (12);
anthraquinones are discussed.
xanthone like ravenelin (13); and modified bisanthraquinones
such as (+)-rugulosin (14), (+)-2,2′-epi-cytoskyrin (15),
A
The asymmetric reduction of tricyclic aromatic polyketides by
fungal short-chain dehydrogenase/reductase (SDR) has gained
significant attention in recent years, due to their role in the
biosynthesis of monomeric and dimeric deoxyanthraquinones.1–3
For example, emodin hydroquinone 2b (formed in situ by the
reduction of emodin (1) using Na2S2O4) is reduced by MdpC
from Aspergillus nidulans using NADPH to give a putative bio-
synthetic intermediate, (R)-3,8,9,10-tetrahydroxy-6-methyl-3,4-
dihydroanthracene-1(2H)-one (3) (Fig. 1A).1 This transform-
ation is being recognized as a crucial step for the formation of
a commonly isolated deoxyanthraquinone, chrysophanol (4),
formed by dehydration and oxidation of 3, during the biosyn-
thesis of monodictyphenone (5),1 agnestin A (6)4 and cladoful-
vin (7)5 (Fig. 1). As a result, several SDRs have been identified
to possess anthrol reductase activity. They include 17β-HSD
from Cochliobolus lunatus,6 AflM from Aspergillus paraticus,3
ClaC from Cladosporium fulvum,5 and AgnL6 from Paecilomyces
variotii,4 which reduce emodin hydroquinone 2 to 3 in the
same manner as that of MdpC (Fig. 1A). Being a key biosyn-
thetic intermediate, 3 has been utilized in the biomimetic syn-
thesis of a dimeric bisanthraquinone, (−)-flavoskyrin (8),
which further supports the involvement of reduced anthrols
like 3 in the biosynthesis of modified bisanthraquinones.2 In
(+)-rugulosin C (16), (−)-rubroskyrin (17), (−)-rugulosin (18),
Centre of Biomedical Research, Sanjay Gandhi Postgraduate Institute of Medical
Sciences Campus, Raebareli Road, Lucknow 226014, India.
E-mail: smhusain@cbmr.res.in, smhusain.cbmr@gmail.com
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9gc03072g
Fig. 1 (A) Chemoenzymatic reduction of emodin (1) to reduced emodin
hydroquinone 3 by MdpC of Aspergillus nidulans.1 (B) Secondary metab-
olites (1, 4, 8, 10, 17–19)7 and (6, 7, 11–19)4,5,8–12 have been isolated
from P. islandicum and other fungi, respectively.
‡These authors contributed equally.
6594 | Green Chem., 2019, 21, 6594–6599
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Communication
and (−)-luteoskyrin (19) from different fungal species, implies 17β-HSD from C. lunatus, 77% with MdpC from A. nidulans,
for a broader substrate scope for anthrol reductases (Fig. 1B). and 66% with AflM from A. parasiticus. In addition, the align-
Therefore, in this work, we aim to identify a new fungal ment of protein sequences of the well-known naphthol
anthrol reductase, which can catalyze the reduction of natural reductases,
1,3,6,8-tetrahydroxynaphthalene
reductase
anthraquinones or their hydroquinones to give access to (T4HNR) and 1,6,8-trihydroxynaphthalene reductase (T3HNR)
unknown, not yet isolated putative biosynthetic intermediates. of Magnaporthe grisea involved in the dearomatization of
This would have implications on the understanding of the bio- naphthols during melanin biosynthesis16,17 with ARti shows
synthesis of modified anthraquinones and bisanthraquinones. 53% and 65% sequence identity, respectively. The conserved
Considering the isolation of secondary metabolites 1, 4, 8, domain analysis of the ARti with T4HNR, T3HNR, MdpC, AflM,
10, and 17–19 7 from Talaromyces islandicus (formerly and 17β-HSDcl shows a consensus of the NADP-binding site
Penicillium islandicum Sopp), which appears to involve a regio- for all the enzymes (Fig. S1†). The comparison of ARti with
and stereoselective reduction of anthraquinones or their 17β-HSDcl in Table 2, shows common amino acid residues for
hydroquinones, we propose the existence of anthrol reductase the substrate binding and active site.18 The high sequence
(s) in T. islandicus. Such an enzyme might be directly involved homology and putative domain architecture hint for a similar
in the biosynthesis of these natural products (Fig. 1B). To physiological function for ARti.
identify such an anthrol reductase(s) and its biosynthetic gene
Thus, based on sequence identity, conserved domains, and
cluster (BGC), we analyzed the genomic sequence of the position in the genome, we propose that the physiological
T. islandicus WF-38-12 genome, which has been published function of ARti is to catalyze the NAD(P)H-dependent regio-
recently.13 We searched for a BGC containing a putative oxido- and stereoselective reduction of naphthols and/or anthrols.6,19
reductase with high sequence homology to the known anthrol Therefore, we tested compounds that are known to be reduced
reductases MdpC, AflM, and 17β-HSDcl, all belonging to the by T4HNR of M. grisea, 17β-HSD of C. lunatus and also those
short-chain dehydrogenase/reductase (SDR) superfamily. The which might be reduced by related enzymes to identify the
fungiSMASH (fungal version of antiSMASH)14 analysis of the physiological function of ARti (Fig. 2). For enzyme preparation,
T. islandicus genome predicted nineteen type 1 polyketide the codon-optimized gene of ARti was synthesized and cloned
synthase gene clusters, out of which only one contains all the into the pET-19b vector between 5′-NdeI and 3′-XhoI encoding
essential genes coding for proteins involved in the biosyn- N-terminal his10-tag. The recombinant plasmid was then,
thesis and biotransformation of anthraquinones (Fig. S2†). In transformed into E. coli BL21 (DE3) and the protein was
Table 1, this cluster shows high similarity with other known expressed and purified using Ni-NTA affinity chromatography.
BGCs such as monodictyphenone (mdp)15 of A. nidulans, clado- The purity and size of the protein were confirmed by
fulvin (cla)5 of C. fulvum, and agnestin (agn)4 of P. divaricatus SDS-PAGE (Fig. S3†). To test the substrate scope, at first, we
(for more details see Table S1†).
synthesized 1,3,6,8-tetrahydroxynaphthalene (T4HN, 20)20 and
CRG86674.1 (polyketide synthase) and CRG86680.1 incubated it with NADPH (regenerated through the glucose/
(β-lactamase) show strong similarities to mdpG/claG and mdpF/ GDH system) and ARti_his in potassium phosphate buffer
claF, respectively which can synthesize atrochrysone.
CRG86683.1 (decarboxylase or anthrone oxygenase) shows
strong similarity with mdpH/claH that catalyzes the decarboxyl-
Table 2 Comparison of amino acid residues at binding and active sites
ation of atrochrysone to produce emodin (1). CRG86682.1
(SDR) and CRG86684.1 (dehydratase) show strong similarities
with mdpC/claC and mdpB/claB, respectively. The former cata-
lyzes the reduction of emodin (1) while the latter might facili-
tate the dehydration of the reduced form of emodin that is 2b
leading to the formation of chrysophanol (3).5,15 The putative
anthrol reductase from T. islandicus (named herein “ARti”,
between 17β-HSDcl and ARti
Binding site
Active site
17β-HSDcl
ARti
17β-HSDcl
ARti
Ser153, Asn154,
Tyr167, Gly199,
Met204, Phe205,
Ser148, Asn149,
Tyr162, Gly194,
Met199, Phe200,
Ala204, Tyr207
Asn127,
Ser153,
Tyr167,
Lys171
Asn122,
Ser148,
Tyr162,
Lys166
accession CRG86682.1) shows 80% sequence identity with Ser209 and Tyr212
Table 1 Alignment of the BGC containing ARti of T. islandicus with other known BGCs involved in the biosynthesis of monodictyphenone (mdp),15
cladofulvin (cla),5 and agnestin (agn)4
Sequence identity
Proteins with putative function
mdp
cla
agn
CRG86674.1 (polyketide synthase)
CRG86680.1 (β-lactamase)
CRG86682.1 (SDR, “ARti”)
CRG86683.1 (decarboxylase)
CRG86684.1 (dehydratase)
mdpG (45%)
mdpF (48%)
mdpC (77%)
mdpH (57%)
mdpB (56%)
claG (43%)
claF (47%)
claC (77%)
claH (68%)
claB (60%)
agnpks1 (44%)
agnL7 (50%)
agnL6 (72%)
—
agnL8 (57%)
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6594–6599 | 6595

View Article Online
Communication
Green Chemistry
Fig. 2 Substrates 1, 20–25 tested for reduction with ARti_his using
NADPH, regenerated through the glucose/GDH system.
Scheme 1 ARti_his catalyzed reduction of emodin (1) hydroquinones
(2a/2b) formed in situ using Na2S2O4 to (R)-3,8,9,10-tetrahydroxy-6-
methyl-3,4-dihydroanthracene-1(2H)-one (3) in the presence of NADPH.
(50 mM, 1 mM EDTA, 1 mM DTT, pH 7.0) under anoxic con-
ditions. The transformation resulted in the reduction of 20 to
scytalone with >99% ee and 36% yield. (R)-Configuration was
assigned to scytalone based on the CD spectra (Fig. S14†).21 1H NMR spectra. (R)-Configuration was assigned to 3 in com-
However, when we tested other substrates such as lawsone (21) parison with the CD spectra of (R)-3 synthesized chemoenzy-
and 2-tetralone (22) which are known to be reduced by matically using 17β-HSDcl as reported earlier (Fig. S6†).6 We
T4HNR21,22 in the presence of NADPH, none of them were also incubated emodin (1) with ARti_his in potassium phos-
reduced by ARti (Fig. 2). This indicates that the two enzymes phate buffer (50 mM, 1 mM EDTA, 1 mM DTT, pH 7.0) using
have different physiological functions, which motivated us to DMSO as a co-solvent for 12 h in the absence of Na2S2O4,
test substrates such as 1-tetralone (23) and menadione (24), under anoxic conditions in the presence of NADPH, regener-
which are not reduced by T4HNR or even T3HNR of M. grisea ated through glucose/GDH. However, no reduced product was
(Fig. 2).22 However, none of these substrates resulted in any formed in such a case. To further characterize the catalytic
reduced product on incubation with ARti under the same function of ARti under different conditions, such as various
conditions.
co-solvents, pH and temperature, we used the reduction of
Considering, the high sequence identity (80%) of ARti with emodin (1) by ARti_his in the presence of Na2S2O4 and
that of 17β-HSDcl, which is known to reduce estrone (25) using NADPH to 3, as a model reaction (Scheme 1). At first, we tested
NADPH, we tested 25 as a probable substrate for ARti different co-solvents. For this, we incubated 1 dissolved in
(Fig. 2).23 We incubated 25 with ARti_his in the presence of different co-solvents (8% v/v) such as 2-propanol, ethanol,
NADPH (regenerated using the glucose/GDH system) in the t-butanol, acetone, acetonitrile, DMSO, DMF, or 1,4-dioxane,
same buffer for 12 h using DMF as a co-solvent. ARti_his cata- with ARti_his in the presence of Na2S2O4 (20 equiv.) and
lyzed the reduction of carbonyl of estrone (25) in the 17-posi- NADPH, regenerated through the glucose/GDH system under
tion to 17β-estradiol (>99% de) with 26% conversion (deter- anoxic conditions for 6 h. The conversions (%) under different
1
mined with H NMR spectroscopy). The product was purified co-solvents were determined from 1H NMR spectra of the
in 22% yield by silica gel chromatography and characterized by crude reaction mixture in acetonitrile-d3. The result shows
NMR spectroscopy and mass spectrometry (ESI). The above DMSO as the best co-solvent with 87% conversion of 1 into 3
transformation indicated that ARti might have a similar phys- (Fig. 3a). The use of 2-propanol as a co-solvent resulted in only
iological function as that of 17β-HSD from C. lunatus. Similar 34% conversion. This could be due to the low solubility of the
to MdpC, 17β-HSDcl has also been identified recently to cata- substrate in 2-propanol. All other solvents gave reasonable con-
lyze the regio- and stereoselective reduction of emodin version between 73–84% (Fig. 3a). Therefore, DMSO is chosen
anthrols (2a/2b) formed in situ by the reduction of 1 using as a co-solvent for further experiments. To determine the
Na2S2O4, to 3 in the presence of NADPH (Scheme 1).6
cofactor preference, we performed the transformation
This motivated us to test the reduction of emodin (1) with (Scheme 1) using NADPH or NADH, generated using the
ARti in the presence of Na2S2O4 which might also result in the glucose/GDH system keeping everything else the same. We
formation of 3. We incubated emodin (1) with ARti_his in the observed that only the NADPH containing reaction resulted in
same buffer using DMSO as a co-solvent in the presence of the formation of 3, which shows ARti as an NADPH-dependent
Na2S2O4 and NADPH (regenerated using glucose/GDH) for enzyme. To investigate the effect of pH on the conversion of
12 h under anoxic conditions (Scheme 1). This resulted in 93% emodin (1) to 3, we performed the same chemoenzymatic
conversion of emodin (1) to (R)-3,8,9,10-tetrahydroxy-6-methyl- transformation at pH 5.5–8.0 and determined the conversion
1
3,4-dihydroanthracene-1(2H)-one (3) with >99% ee. Due to the using the H NMR spectra of the crude reaction mixture. The
overlap of absorbance bands between the substrate and cofac- results show a low conversion (∼63%) at pH 5.5 and 8.0,
tor, UV based assay was not feasible; therefore to analyze the higher conversion (87–94%) at pH in between, and the highest
activity of ARti, we determined the conversion of 1 to 3 using (94%) at pH 7.0 (Fig. 3b).
6596 | Green Chem., 2019, 21, 6594–6599
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Communication
Fig. 3 (A) Effect of the co-solvent on the reduction of emodin hydro-
quinones (2a/2b) by ARti_his. (B) pH profile of ARti_his with 2a/2b. (C)
Temperature profile of ARti_his with 2a/2b. (D) Plot of substrate concen-
tration ([S]) versus time fitted exponentially.
Fig. 4 (A) Regio- and stereoselective, NADPH-dependent reduction of
anthrols formed in situ using Na2S2O4 by ARti_his. NADPH regenerated
using the glucose/GDH system. (B) Reduced compounds 29a–f obtained
by the reduction of natural anthraquinones 26a–f. (C) Reduced com-
pounds 29g obtained by the chemoenzymatic reduction of non-natural
anthraquinone 26g. (D) Substrates 26h–i were not reduced by ARti_his
chemoenzymatically.
Next, we study the effect of heat on the above transform-
ation by performing the enzymatic reduction of emodin (1)
using ARti_his in the presence of Na2S2O4 and NADPH for 2 h
and 14 h at temperatures from 15 °C to 50 °C at the interval of
5 °C. ARti shows the highest activity at 40 °C and 35 °C as
observed by conversions of 1 to 3 when incubated for 2 h and
14 h, respectively (Fig. 3c). This shows that ARti remained
active for a longer period of time at a lower temperature of
35 °C. However, considering the formation of a deoxygenated such as citreorosein (26b, R1 = H, R2 = CH2OH), emodic acid
side product, chrysophanol (4) at a higher temperature, we per- (26d, R1 = H, R2 = CO2H) and its methyl ester (26g, R1 = H, R2 =
formed all further transformations at room temperature. In CO2CH3) were synthesized starting from emodin (1) in 3–4
addition, we determined the melting temperature (Tm) for steps (ESI†). Anthraquinones 26a–e and 26h–i have been
ARti_his using CD spectroscopy, to be 51 °C (Fig. S4, ESI, p. reportedly isolated from different fungi, so they could be
S7†). We also calculated the rate constant for the above trans- appropriate biosynthetic precursors for other modified anthra-
formation from the plot of substrate concentration versus time quinones and bisanthraquinones involving the reduction of
(Fig. 3d). For this, we carried out emodin (1) reduction in the anthrols.24 Therefore, anthraquinones 26a–e and 26h–i were
presence of Na2S2O4 with ARti_his using NADPH, regenerated incubated one by one with ARti_his in potassium phosphate
through the glucose/GDH system for different time intervals. buffer (50 mM, 1 mM EDTA, 1 mM DTT, pH 7.0) in the pres-
The substrate concentration was calculated from the conver- ence of Na2S2O4 (20 equiv.) and NADPH, regenerated through
sion of 1 into 3, determined through 1H NMR spectra in aceto- the glucose/GDH system, using DMSO as a co-solvent under
nitrile-d3. The rate constant was found to be 0.2298 min−1 anoxic conditions (ESI, pp. S22–S30†). After 14 h, the reaction
(Fig. 3d) (ESI, pp. S8 and S9†).
mixture was extracted in ethyl acetate and conversion to the
To support further the reduction of anthrols by the putative reduced product formed, if any, was determined using 1H
oxidoreductase of T. islandicus, ARti, as its physiological func- NMR spectra in acetone-d6. This resulted in the formation of
tion, we aimed to test the reduction of anthrols of several reduced hydroanthraquinones 29a–e (Fig. 4B), expectedly
natural and non-natural anthraquinones 26a–i as substrates formed by the reduction of tautomers of anthrols 27a–e/28a–e
for the newly characterized enzyme (Fig. 4A). For this purpose, catalyzed by ARti_his using NADPH (Fig. 4A). Except for
naturally occurring anthraquinones, lunatin (26a, R1 = H, R2 = lunatin (26a) which shows 82% conversion, we observed the
OCH3), 1-methylemodin (26c, R1 = CH3, R2 = CH3), 1,3,6,8-tet- quantitative conversion of anthraquinones 26b–e to their
rahydroxyanthraquinone (26e, R1 = H, R2 = OH), questin (26h), respective products 29b–e formed by the reduction of the
and physicon (26i) were synthesized using 2,6-dichlorobenzo- respective anthrols (27a–e/28a–e) catalyzed by ARti. No
quinone and various mixed vinyl ketene acetals via two con- reduced product was obtained in the case of questin (26h),
secutive Diels–Alder reactions (ESI†). Other anthraquinones and physicon (26i) (Fig. 4D).
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6594–6599 | 6597

View Article Online
Communication
Green Chemistry
The deoxygenation of 29e by heating under an oxygen atmo- anthraquinones 8, 14–18 (Fig. 1). In addition, the use of enan-
sphere gave access to another natural anthraquinone 26f (R1 = tiomerically pure, chemoenzymatically reduced anthrols 29a–f
R2 = H), which was also subjected to the chemoenzymatic presented here, will facilitate method development to syn-
reduction with ARti_his under the same condition to obtain thesize various natural products and their analogs and will
29f with quantitative conversion (Fig. 4B). A non-natural provide clues for the elucidation of the (bio)synthesis of such
anthraquinone 26g also resulted in the formation of 29g in natural products. We also believe that many more secondary
75% conversion (Fig. 4C), when incubated with ARti_his in the metabolites with dimeric structures related to bisanthraqui-
presence of Na2S2O4 and NADPH (regenerated using glucose/ none such as (−)-flavoskyrin (8) and 14–18 (Fig. 1) with a
GDH system). The in situ formation of anthrols 27a–g/28a–g by diverse array of biological activities will be isolated in future.
the reduction of respective anthraquinones 26a–g through
1
Na2S2O4, was confirmed by the H NMR spectra, supported by
their respective HRMS data. The reduced compounds 29a–g Conflicts of interest
were purified using silica gel column chromatography in
66–84% yield and characterized by NMR spectroscopy and
There are no conflicts to declare.
mass spectrometry. The configuration (R) has been assigned to
all the reduced compounds 29a–g by comparison of their CD
spectra with that of (R)-3,8,9,10-tetrahydroxy-6-methyl-3,4-dihy-
Acknowledgements
droanthracene-1(2H)-one (3) synthesized by the chemoenzy-
We are thankful to Prof. Michael Müller for his useful com-
ments on the manuscript, Council of Scientific and Industrial
Research, New Delhi (Project No. 02(0258)/16/EMR-II) and
matic reduction of emodin (1) using 17β-HSDcl as reported
previously (Fig. S4–S12†).6 We could also determine the enan-
tiomeric excess for all the reduced compounds 29a–g using
SERB
(CRG/2018/002682)
funding,
Department
of
HPLC by comparison with the racemic 29a–g made by the
reduction of anthraquinones 26a–g using NaBH4 in the pres-
ence of Na2S2O4 in water. For all the reduced anthraquinones
29a–g obtained through chemoenzymatic reduction, ee was
found to be >99%. Overall, the result shows tolerance of
various substituents on the ring A of the anthraquinones as
observed for the reduction of 26a–g while any changes to the
ring C of anthraquinones, as in the case of 26h–i, were not tol-
erated. Therefore, no reduced products were obtained in the
case of questin (26h) and physicon (26i). The asymmetric
reduction of hydroanthraquinones 27a–g/28a–g shown above
by ARti, supports anthrol reduction as a physiological function
for the putative oxidoreductase (accession no. CRG86682.1)
from T. islandicus. As shown here, the use of ARti provides easy
access to many of the putative biosynthetic intermediates 29a–
f in optically pure form for the first time and had also broad-
ened the scope of anthrol reductases which remained limited
to the reduction of emodin anthrols (2a/2b). In comparison
with the only non-enzymatic procedure that has been reported
for the synthesis of (S)-29a and involves 14 steps with an
overall yield of 5% (Scheme S7†),25 our biocatalytic procedure
is simple, atom economical, green and results in high yields.
In conclusion, an SDR from Talaromyces islandicus WF-38-
12 has been identified through genome analysis and proposed
to possess anthrol reductase activity. The enzyme has been
shown to catalyze an asymmetric reduction of anthrols 27a–g/
28a–g, formed in situ by the reduction of variously substituted
natural anthraquinones 26a–f and a non-natural analog 26g in
Biotechnology, India for fellowship for SKS and Director,
Centre of Biomedical Research, Lucknow for research
facilities.
Notes and references
1 M. A. Schätzle, S. M. Husain, S. Ferlaino and M. Müller,
J. Am. Chem. Soc., 2012, 134, 14742–14745.
2 N. Saha, A. Mondal, K. Witte, S. K. Singh, M. Müller and
S. M. Husain, Chem. – Eur. J., 2018, 24, 1283–1286.
3 D. Conradt, M. A. Schätzle, J. Haas, C. A. Townsend and
M. Müller, J. Am. Chem. Soc., 2015, 137, 10867–10869.
4 A. J. Szwalbe, K. Williams, Z. Song, K. De Mattos-Shipley,
J. L. Vincent, A. M. Bailey, C. L. Willis, R. J. Cox and
T. J. Simpson, Chem. Sci., 2019, 10, 233–238.
5 S. Griffiths, C. H. Mesarich, B. Saccomanno, A. Vaisberg,
P. J. De Wit, R. Cox and J. Collemare, Proc. Natl. Acad.
Sci. U. S. A., 2016, 113, 6851–6856.
6 L. Fürtges, D. Conradt, M. A. Schätzle, S. K. Singh,
N. Kraševec, T. L. Rižner, M. Müller and S. M. Husain,
ChemBioChem, 2017, 18, 77–80.
7 U. Sankawa, in The Biosynthesis of Mycotoxins, ed.
P. S. Steyn, Academic Press, New York, 1980, pp. 357–394.
8 P.-L. Wang, D.-Y. Li, L.-R. Xie, X. Wu, H.-M. Hua and
Z.-L. Li, Nat. Prod. Res., 2014, 28, 290–293.
9 X. You, S. Feng, S. Luo, D. Cong, Z. Yu, Z. Yang and
J. Zhang, Fitoterapia, 2013, 85, 161–168.
the presence of Na2S2O4, using NADPH to 29a–g (Fig. 4). This 10 A. J. Birch, J. Baldas, J. R. Hlubucek, T. J. Simpson and
chemoenzymatic transformation supports anthrol reduction as
the physiological function of ARti. Based on the earlier work of
P. W. Westerman, J. Chem. Soc., Perkin Trans. 1, 1976, 898–
904.
Shibata7,26 and our recent biomimetic approach towards the 11 H. Yamazaki, N. Koyama, S. Ōmura and H. Tomoda, Org.
synthesis of bisanthraquinones,2,27 we propose that 29a–f may
act as precursors for the biosynthesis of deoxy (bis)anthraqui- 12 A. Agusta, K. Ohashi and H. Shibuya, Chem. Pharm. Bull.,
Lett., 2010, 12, 1572–1575.
nones like 7, 10–12, xanthones like 6, 13 and modified (bis)
2006, 54, 579–582.
6598 | Green Chem., 2019, 21, 6594–6599
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Communication
13 T. Schafhauser, D. Wibberg, C. Rückert, A. Winkler, L. Flor, 19 S. M. Husain, M. A. Schätzle, S. Lüdeke and M. Müller,
K.-H. van Pée, D. P. Fewer, K. Sivonen, L. Jahn, J. Ludwig- Angew. Chem., Int. Ed., 2014, 53, 9806–9811.
Müller, T. Caradec, P. Jacques, M. M. E. Huijbers, 20 K. Ichinose, Y. Ebizuka and U. Sankawa, Chem. Pharm.
W. J. H. van Berkel, T. Weber, W. Wohlleben and
J. Kalinowski, J. Biotechnol., 2015, 211, 101–102.
14 K. Blin, T. Wolf, M. G. Chevrette, X. Lu, C. J. Schwalen,
S. A. Kautsar, H. G. Suarez Duran, E. L. C. de los Santos,
Bull., 2001, 49, 192–196.
21 M. A. Schätzle, S. Flemming, S. M. Husain, M. Richter,
S. Günther and M. Müller, Angew. Chem., Int. Ed., 2012, 51,
2643–2646.
H. U. Kim, M. Nave, J. S. Dickschat, D. A. Mitchell, 22 N. Saha, M. Müller and S. M. Husain, Org. Lett., 2019, 21,
E. Shelest, R. Breitling, E. Takano, S. Y. Lee, T. Weber and
M. H. Medema, Nucleic Acids Res., 2017, 45, W36–W41.
15 Y.-M. Chiang, E. Szewczyk, A. D. Davidson, R. Entwistle,
2204–2208.
23 T. L. Rižner, M. Žakelj-Mavrič, A. Plemenitaš and M. Zorko,
J. Steroid Biochem. Mol. Biol., 1996, 59, 205–214.
N. P. Keller, C. C. C. Wang and B. R. Oakley, Appl. Environ. 24 M. Fouillaud, M. Venkatachalam, E. Girard-Valenciennes,
Microbiol., 2010, 76, 2067–2074.
Y. Caro and L. Dufossé, Mar. Drugs, 2016, 14, 1–64.
25 K. C. Nicolaou, Y. H. Lim, J. L. Piper and C. D. Papageorgiou,
J. Am. Chem. Soc., 2007, 129, 4001–4013.
26 S. Seo, U. Sankawa, Y. Ogihara, Y. Iitaka and S. Shibata,
Tetrahedron, 1973, 29, 3721–3726.
16 S. M. Husain and M. Müller, Synlett, 2017, 2360–2372.
17 D. Conradt, M. A. Schätzle, S. M. Husain and M. Müller,
ChemCatChem, 2015, 7, 3116–3120.
18 A. Cassetta, J. Stojan, I. Krastanova, K. Kristan,
M. Brunskole Švegelj, D. Lamba and T. Lanišnik Rižner, 27 A. Mondal, N. Saha, A. Rajput, S. K. Singh, B. Roy and
J. Steroid Biochem. Mol. Biol., 2017, 171, 80–93.
S. M. Husain, Org. Biomol. Chem., 2019, 17, 8711–8715.
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6594–6599 | 6599