Discovery of McbB, an enzyme catalyzing the β-carboline skeleton construction in the marinacarboline biosynthetic pathway

Article abstract of DOI:10.1002/anie.201303449

Three genes, mcbABC, that drive the biosynthesis of marinacarbolines, have been elucidated through genome mining, gene inactivation, heterologous expression, feeding, and site-directed mutagenesis experiments. McbB is highlighted as a novel enzyme for the β-carboline core construction, which involves a Pictet-Spengler cyclization process and requiring E₉₇ for biochemical activity.

Full text of DOI:10.1002/anie.201303449

.
Angewandte
Communications
DOI: 10.1002/anie.201303449
Alkaloid Biosynthesis
Discovery of McbB, an Enzyme Catalyzing the b-Carboline Skeleton
Construction in the Marinacarboline Biosynthetic Pathway**
Qi Chen, Changtao Ji, Yongxiang Song, Hongbo Huang, Junying Ma, Xinpeng Tian, and
Jianhua Ju*
The b-carboline (bCs) alkaloids possess a tricyclic pyrido[3, 4-
b]indole ring system and are widely distributed in nature; the
heterocyclic skeleton endows the bCs with antiallergic,
antiviral, anti-inflammatory, antibacterial, and antitumor
activities, and enables interactions with various receptors
leading to neurotoxin and neuroprotectant activities.[1] How-
ever, only a few types of enzymes are known to catalyze the
Pictet–Spengler (PS) reaction, enabling the biosynthesis of
tetrahydro-bC scaffolds. Those characterized enzymes
include the strictosidine synthases from various higher
plants that are known to catalyze PS condensation of
tryptamine and the monoterpenoid aldehyde secologanin to
form 3a(S)-strictosidine, which serves as the precursor for
more than 2500 indole alkaloids.[2] The C-domain of the non-
ribosomal peptide synthetase SfmC from Streptomyces lav-
endulae also is known to mediate PS cyclization during the
course of saframycin biosynthesis.[3]
There exists a small group of C-ring unsaturated bCs in
which C1 is substituted by a two-carbon unit found in various
Scheme 1. Structures of marinacarbolines A–D (1–4) and structurally
related b-carboline alkaloids from higher plants, marine sponges,
fungi, and actinomycetes.
organisms, including medicinal plants,[4] marine tunicate,[5]
fungi,[6] and actinomycetes (Scheme 1).[7] Such compounds
have attracted significant interest from synthetic chemists.[8]
However, the enzymes driving biosynthesis of this group of
molecules have not yet been deciphered. We have isolated
marinacarbolines A–D (MCBs, 1–4) and 1-acetyl-b-carboline
(5) from the deep South China Sea-derived Marinactinospora
thermotolerans SCSIO 00652,[9] thus providing us an excellent
opportunity to explore the biosynthetic machinery of this
interesting scaffold at a genetic level. Notably, compounds 1–
4 are all associated with activities against Plasmodium
falciparum 3D7, a drug-sensitive line, and Dd2, a multi-
drug-resistant line.[9] Herein, we report the identification of
a gene cluster consisting of three genes mcbABC responsible
for MCB scaffold biosynthesis, and characterization of McbB,
a novel enzyme mediating b-carboline core construction
involving a PS cyclization/decarboxylation/oxidation process.
Considering that the bC metabolites 1–4 all contain an
aromatic substituted ethylamine group and that the C-ring is
unsaturated, we initially hypothesized that a decarboxylase
and a dehydrogenase/oxidase are required for MCB scaffold
construction. Previously, we shotgun sequenced the whole
genome of M. thermotolerans SCSIO 00652 and identified the
biosynthetic gene clusters governing the biosynthesis of
nucleoside antibiotic A201A and methylpendolmycin/pen-
dolmycin.[10] Bioinformatics analysis of the genome revealed
eight DNA segments distributed on scaffolds 1, 2, 11, 20, 21,
27, 47, and 93 coding for both types of enzymes, and therefore
having a potential role in MCB biosynthesis. Further analysis
revealed that the DNA segment on scaffold 27 harbors a fatty
acid CoA ligase (Supporting Information, Table S1), poten-
tially involved in amide bond formation on the way to 1. To
explore the possible involvement of these genes in MCB
biosynthesis, a pOJ436 vector-based genomic library was
constructed and cosmids containing the DNA segment
harboring the three types of enzymes were screened. Two
positive cosmids, 1610B and 273H, were transformed into
Streptomyces lividans TK64 through direct conjugation for
gene cluster expression. HPLC analyses of the fermentation
extracts revealed that all the exconjugants harboring either
[*] Q. Chen, C. Ji, Dr. Y. Song, Dr. H. Huang, Dr. J. Ma, Dr. X. Tian,
Prof. Dr. J. Ju
CAS Key Laboratory of Marine Bio-resources Sustainable Utilization,
Guangdong Key Laboratory of Marine Materia Medica, RNAM
Center for Marine Microbiology, South China Sea Institute of
Oceanology, Chinese Academy of Sciences
164 West Xingang Road, Guangzhou 510301 (China)
E-mail: jju@scsio.ac.cn
Q. Chen, C. Ji
University of Chinese Academy of Sciences
19 Yuquan Road, Beijing 110039 (China)
[**] This work was financially supported by grants from MOST
(2012AA092104, 2010CB833805), NSFC (31290233, 31000051,
41106138), and CAS (KSCX2-EW-G-12, SQ201015).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201303449.
9980
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9980 –9984

Angewandte
Chemie
genotypes were subsequently confirmed by PCR analysis
(Supporting Information, Figures S1–S10). All mutant strains
were fermented alongside the wild-type control and the
resulting metabolites analyzed by HPLC. The upstream
cluster boundary was determined by gene inactivations of
orf((À7)-(À4)), orf(À3), orf(À2), and orf(À1); none of the
inactivated mutants accumulated MCBs or related congeners
at rates or efficiencies significantly different from those of the
wild-type strain. Similarly, the downstream cluster boundary
was determined by disruption of orf(+1), orf(+2) and
orf(+3); none of these mutants displayed changes in MCB
production relative to wild-type (Supporting Information,
Figure S11). Thus, the regulatory, resistance, and oxidoreduc-
tase coding genes (two of each category) listed in the
Supporting Information, Table S1 seem not to play a role in
MCB biosynthesis in the native M. thermotelerans SCSIO
00652 producer or their functions are supplemented by other
corresponding genes in the genome. In contrast, the remain-
ing three genes, mcbABC, appear to be solely responsible for
MCB scaffold construction.
McbA exhibits 42% identity and 55% similarity with
SRIM_40803, a putative fatty acid CoA ligase from Strepto-
myces rimosus subsp. rimosus ATCC 10970. McbB exhibits
48% identity and 62% similarity with BBA_06474, an
enzyme with unknown biosynthetic function from Beauveria
bassiana ARSEF 2860. McbC displays 32% identity and 49%
similarity with Gad2, an assigned glutamate decarboxylase
from Beggiatoa sp. PS. Generation of DmcbB mutant and
subsequent metabolite profiling revealed that mcbB inacti-
vation abolishes production of 1–5 and related analogues
demonstrating that McbB is absolutely essential for b-carbo-
line core construction (Figure 1B, trace vi). The DmcbA
mutant also failed to produce MCBs 1–4 but accumulated
compound 6 as its major product, along with a minor product
5 (Figure 1B, trace v). Metabolite profiling of the DmcbC
mutant afforded an HPLC profile similar to that observed
when evaluating fermentations of the DmcbA mutant
although the yields of 5 and 6 were relatively low in the
DmcbC case (Figure 1B, trace vii). Compound 6 has a molec-
ular formula of C14H10N2O3 as determined by HRMS and was
identified as 1-acetyl-3-carboxy-b-carboline, following MS, 1H
and 13C NMR spectroscopy, and HMBC data analyses (Sup-
porting Information). Importantly, 6 has been previously
reported to be isolated from higher plants[4b,c] and compound
5 was identified as 1-acetyl-b-carboline by comparison with an
authentic sample. Consequently, in vivo gene inactivation
results suggest that McbB plays a vital role in bC skeleton
assembly and that the effects of McbA and McbC knockout
are consistent with inactivation of CoA ligase and/or decar-
boxylase (of Phe/Tyr/Trp) functions. Ultimately, assignments
of McbA as a CoA ligase and McbC as a decarboxylase were
achieved on the basis of bioinformatics.
Figure 1. A) Organization of the marinacarboline gene cluster in M.
thermotolerans SCSIO 00652. B) Metabolite profiles upon HPLC analy-
sis using an Alltima C18 column: i)–iii) S. lividans TK64 harboring
pOJ436 vector, cosmids 1610B and 273H, respectively; iv)–vii) M.
thermotolerans SCSIO 00652 wild-type and its mutant strains; viii)–xii)
[10b]
~
*
authentic standards of 1–5. methylpendolmycin, pendolmycin.
cosmid successfully produce 1 and 5, and another analogue 6,
whereas the control S. lividans TK64 with an empty pOJ436
vector was unable to produce these compounds (Figure 1B,
traces i–iii). These findings suggest that cosmids 1610B and
273H harbor the complete gene set responsible for MCB
biosynthesis. Indeed, end sequencing, along with bioinfor-
matics analysis, indicated that a 11.2 kb DNA sequence
consisting of 13 open reading frames (orfs) on scaffold 27
most likely contain all the genes accounting for MCB
biosynthesis (Supporting Information, Table S1; orf(À4)
through orf(À7) not shown). The nucleotide sequences have
been deposited in the GenBank with accession number
KC541560.
To precisely investigate how many genes are involved in
MCB biosynthesis, we first determined the boundaries of the
MCB gene cluster. We screened the previously constructed
DNA library derived from supercos1 vector[10] and used the
positive cosmid 310A to inactivate the 13 genes by replacing
them individually with an apramycin resistance gene cassette
using l-RED-mediated PCR-targeting mutagenesis method-
ology.[10,11] The four-gene cassette orf(À7), orf(À6), orf(À5)
and orf(À4), encoding nitrate reductase a, b, g and d subunits
respectively, was proposed not to be involved in MCB
biosynthesis; hence these four genes were deleted in combi-
nation. The corresponding mutant strain is designated
orf((À7)-(À4)). The resulting 10 mutant strains were selected
on the basis of kanamycinS and apramycinR phenotypes;
To validate that mcbA, mcbB, and mcbC are sufficient for
MCB scaffold biosynthesis, we next carried out heterologous
expression (either in combinations or individually) of
mcbABC, mcbAB and mcbB in E. coli BL21. The genes
mcbABC, mcbAB and mcbB were cloned into NdeI and
HindIII sites of the pET28a(+) vector to generate
pET28a(+)/mcbABC, pET28a(+)/mcbAB and pET28a(+)/
Angew. Chem. Int. Ed. 2013, 52, 9980 –9984
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9981

.
Angewandte
Communications
mcbB; these were then transformed into E. coli BL21(DE3),
respectively. The three strains were cultured in LB medium to
an OD600 of 0.6, then induced with 0.05 mm IPTG and
cultured for another 12 h at 288C. HPLC analyses of the
culture extracts revealed that: 1) MCBs 1 and 4 were
produced as the major product in E. coli BL21/pET28a
(+)/mcbABC; 2) MCBs 1, 2, and 4 were produced in E. coli
BL21/pET28a (+)/mcbAB; and 3) the major product 6 and
two minor product 5 and 8 were produced in E. coli BL21/
pET28a (+)/mcbB (Figure 2, traces ii–iv). Scaled-up fermen-
glycolysis). To test this hypothesis, [U-13C6] glucose was fed to
the E. coli BL21/pET28a (+)/mcbB strain and the resultant
product 6 was purified. Measurement of the 13C NMR
spectrum of 6 revealed that all of the C3 unit (C1, C14, and
C15) was isotopically enriched. However, subsequent 2D
INADEQUATE experiment of 6 disclosed that direct corre-
lations were only observed between C1 and C14, indicating
that these two carbons are derived from a C2 unit; the C15
methyl group did not show an apprent correlation with C14.
Consequently, feeding experiments with 13C-labeled acetate
to E. coli BL21/pET28a (+)/mcbB were conducted and the
corresponding product 6 was purified. Inspection of the
13C NMR spectrum (Supporting Information) of correspond-
ing product 6 revealed that: 1) feeding with [1-13C] acetate led
to C1 enrichment in 6; 2) feeding with [2-13C] acetate led to
C14 and C15 enrichments in 6; and 3) feeding with [1, 2-13C]
acetate led to C1, C14, and C15 enrichments in 6; C1 and C14
appeared as coupled doublets (1JCC = 63.1 Hz) whereas C15
appeared as a singlet. These feeding experiments collectively
demonstrate that: 1) the bC core 6 originates from Trp and
two acetate units; 2) the C1 and C14 originate from an intact
acetate unit; and 3) C15 originates from C2 of the acetate.
The labeling pattern of 6 strongly indicate that oxaloacetal-
dehyde derived from oxaloacetic acid which originates from
TCA cycle[12] might be used as the direct precursor, and the
single McbB-catalyzed process on the way to 6 involves
a Pictet–Spengler cyclization, a decarboxylation reaction, and
C-ring oxidation, as shown in Scheme 2.
Figure 2. HPLC profiles of culture extracts. i)–v) Recombinant E. coli
BL21 strains harboring various plasmids, vi) E. coli BL21/pET28a
(+)/mcbB fed with 5-F-DL-Trp.
Finally, we generated point mutations in McbB to probe
which amino acid residues are important for catalytic function
by expressing the mutated McbB proteins in E. coli BL21 and
then assaying for production of 6. BLAST analyses revealed
that McbB shows high identity and similarity to a group of
proteins primarily annotated from fungi and bacteria whole
genome sequences; no conserved domains could be found
from this class of uncharacterized proteins. We aligned McbB
with ten closely related proteins (Supporting Information,
Figure S12) and analyzed the conserved amino acid residues.
Site-directed mutagenesis of 42 highly or moderately con-
served amino acid residues in McbB was carried out using the
QuickChange site-directed mutagenesis kit or overlap exten-
sion-PCR methods.[13] Mutant plasmids were transformed
into E. coli BL21 (DE3) and the resulting cells fermented
with IPTG induction; wild-type McbB-bearing cells were
processed in an identical fashion thus serving as an enzymati-
cally intact control. Fermentation broths were extracted with
EtOAc and the resulting extracts then analyzed by HPLC.
These HPLC results revealed that the E97L mutation in
McbB completely abolished generation of 6 and minor
metabolite 5 (Figure 2, trace v). The yields of these two
compounds were sharply decreased to 1–2% in the N24L,
L27V, Q42I, H44F, S48A, S83A, Q86I, H87F, K91A, L115A,
E181L, and R212L mutants; compound production by the
I6V, L36A, D71L, G195A, Y53F, Y38F, Y92F, E110L, and
E170L mutants was diminished to 10–40% of that found with
the wild-type producer. All the remaining 20 mutants failed to
show obvious changes in metabolite yields relative to wild-
type (Supporting Information, Figure S14). Consequently, the
N-terminus region of McbB appears to contain more residues
tation of E. coli BL21/pET28a (+)/mcbB (4.5 L) and subse-
quent isolation afforded analytically pure quantities of 5, 6,
and 8. HRMS of 8 yielded a molecular formula of C13H10N2O2,
28 mass units smaller than that of 6. The 1H, 13C, and
HMBC NMR data analyses of 8 (Supporting Information) led
us to assign its structure as 1-acetyl-3-hydroxy-b-carboline
(Scheme 1). These results support the in vivo inactivation
results and further demonstrate that the minimal three-gene
cassette mcbABC is sufficient for MCB scaffold biosynthesis
(1, 2 and 4), and that McbB alone is able to construct the bC
skeleton intermediate 6, functioning as a Pictet–Spenglerase
that executes C-ring closure and desaturation but that also
leaves open the question of whether or not the C3 unit (C1,
C14, and C15) is a biosynthetic precursor to 6.
To explore the biosynthetic precursors that McbB may
utilize to biosynthesize bC core 6, we then performed feeding
experiments with 5-fluoro-Trp and 13C-labeled precursors.
E. coli BL21/pET28a(+)/mcbB was cultured, induced with
IPTG, and fed with 5-fluoro-DL-Trp. The culture extract,
upon HPLC analysis (Figure 2, trace vi), clearly yielded a new
product 7 that shows UVabsorption bands similar to those of
6 and that has a molecular weight of 272.1 corresponding to 6-
fluoro substituted 6. Scaled-up fermentation (1.5 L) and
subsequent HPLC-guided purification led to the isolation of
1
analytically pure 7 for structure elucidation. HRMS, H, 13C,
and HMBC NMR data analyses permitted unambiguous
assignment of 7 as 1-acetyl-3-carboxy-6-fluoro-b-carboline
(Supporting Information). We initially envisioned that the
remaining three-carbon unit (C1, C14, and C15) of 6 might
originate from a direct C3 precursor such as pyruvate (from
9982 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9980 –9984

Angewandte
Chemie
Scheme 2. Marinacarbolines 1–4 are biosynthesized by McbA, McbB, and McbC, and proposed McbB-catalyzed biosynthetic pathway for 6 using
Trp and acetate as precursors. Labeling pattern of 6 derived from the feeding of 13C-labeled sodium acetate is shown.
involved in McbB bioactivity; the conserved amino acid
residue E97 is absolutely required and may serve as the
catalytic center of McbB.
protein on the way to bC scaffold 6; the catalytic mechanisms
driven by McbB are the subject of future exploration.
In conclusion, genome scanning, bioinformatics analysis,
gene inactivation, and heterologous expression in Streptomy-
ces lividans TK64 and Escherichia coli BL21 have enabled us
to identify and elucidate three genes, mcbABC, that drive the
biosynthesis of MCBs. When expressed on its own in E. coli,
mcbB, produced 1-acetyl-3-carboxy-b-carboline (6) as the
major product and two other minor products 1-acetyl-b-
carboline (5) and 1-acetyl-3-hydroxy-b-carboline (8), thus
highlighting it as a novel enzyme for b-carboline scaffold
construction. Feeding experiments with 5-F-Trp and 13C-
labeled glucose and acetate reveal that 6 is derived from Trp
and two equivalents of acetate as precursors. Generation of 6
by McbB involves a Pictet-Spengler cyclization, a decarbox-
ylation and C-ring oxidation process. Site-directed muta-
genesis experiments with McbB reveal that E97 is absolutely
required for biochemical activity. All in all, these studies
enhance our understanding of the genetic and biochemical
aspects of MCB construction thereby expanding our reper-
toire of methods by which to search for and to develop new
bC-containing natural products.
In light of the above data, the chemistry carried out by
McbB appears to involve a domino reaction sequence
composed of initial Pictet–Spenger condensation followed
by decarboxylation, and a C-ring desaturation reaction on the
way to major compound 6 and minor metabolites 5 and 8. We
envision the chemistry of MCB scaffold construction to
proceed as depicted in Scheme 2. Oxaloacetaldehyde, derived
from oxaloacetic acid, might be utilized as the direct
precursor for initial Schiff base formation, as per standard
Pictet–Spengler chemistry. Concomitant decarboxylation
leads and C-ring installation affords key intermediate 6a. A
À
related means of effecting C C bond cleavage has been
recently implicated in the biosynthesis of aziridine-containing
natural products azicemicin A and B.[12] Importantly, 6a can
proceed to compounds 6, 8, and 5 through sequences that
include: 1) C-ring aromatization via two-fold oxidation
(affording 6); 2) C-ring oxidation followed by decarboxyla-
tion, hydration, and ultimate C-ring aromatization (affording
8); and 3) C-ring oxidation followed by decarboxylation and
C-ring aromatization (affording 5). It is important to note that
throughout the pathway to these products, oxidative trans-
formations play a central role, some of which may proceed
spontaneously, as evidenced by chemical synthesis of 5 and 8
using l-Trp, methylglyoxal, and p-toluenesulfonic acid
(1.0 equiv) as a catalyst in yields of 5% and 13%, respecti-
vely.[8b] The alkaloids harman and eleagnine are other
examples of natural products whose b-carboline scaffold is
attributed to oxidative transformations of indole intermedi-
ates.[14] Likewise, we envision that generation of the present
MCB scaffold relies, in large measure, on the oxidative
amenability of indole 6a and related compounds. No doubt
aided by this reactivity, it is remarkable to discover the
diversity of chemistries enabled by the multifaceted McbB
Received: April 23, 2013
Revised: June 9, 2013
Published online: August 1, 2013
Keywords: biosynthesis · b-carbolines · marinacarbolines ·
.
natural products · Pictet–Spengler reaction
[1] a) R. Cao, W. Peng, Z. Wang, A. Xu, Curr. Med. Chem. 2007, 14,
479 – 500; b) T. Herraiz in Isoquinolines and beta-carbolines as
neurotoxins and neuroprotectants (Eds.: L. Antkiewicz-Micha-
luk, H. Rommelspacher), Springer, New York, 2012, pp. 77 – 103.
[2] a) J. Stçckigt, A. P. Antonchick, F. Wu, H. Waldmann, Angew.
Chem. 2011, 123, 8692 – 8719; Angew. Chem. Int. Ed. 2011, 50,
Angew. Chem. Int. Ed. 2013, 52, 9980 –9984
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9983

.
Angewandte
Communications
8538 – 8564; b) L. F. Szabꢀ, Molecules 2008, 13, 1875 – 1896;
c) J. J. Maresh, L. A. Giddings, A. Friedrich, E. A. Loris, S.
Panjikar, B. L. Trout, J. Stçkigt, B. Peters, S. E. OꢁConnor, J. Am.
Chem. Soc. 2008, 130, 710 – 723; d) P. Bernhardt, E. McCoy, S. E.
OꢁConnor, Chem. Biol. 2007, 14, 888 – 897; e) X. Y. Ma, S.
Panjikar, J. Koepke, E. Loris, J. Stçckigt, Plant Cell 2006, 18,
907 – 920.
Llona-Minguez, S. Lang, S. P. Mackay, C. J. Suckling, O. B.
Sutcliffe, Beilstein J. Org. Chem. 2011, 7, 1407 – 1411.
[9] H. Huang, Y. Yao, Z. He, T. Yang, J. Ma, X. Tian, Y. Li, C.
Huang, X. Chen, W. Li, S. Zhang, C. Zhang, J. Ju, J. Nat. Prod.
2011, 74, 2122 – 2127.
[10] a) Q. Zhu, J. Li, J. Ma, M. Luo, B. Wang, H. Huang, X. Tian, W.
Li, S. Zhang, C. Zhang, J. Ju, Antimicrob. Agents Chemother.
2012, 56, 110 – 114; b) J. Ma, D. Zuo, Y. Song, B. Wang, H.
Huang, Y. Yao, W. Li, S. Zhang, C. Zhang, J. Ju, ChemBioChem
2012, 13, 547 – 552.
[11] a) Y. Zhang, F. Buchholz, J. P. Muyrers, A. F. Stewart, Nat.
Genet. 1998, 20, 123 – 128; b) B. Gust, G. L. Challis, K. Fowler, T.
Kieser, K. F. Chater, Proc. Natl. Acad. Sci. USA 2003, 100, 1541 –
1546.
[12] Y. Ogasawara, H.-w. Liu, J. Am. Chem. Soc. 2009, 131, 18066 –
18068.
[3] K. Koketsu, K. Watanabe, H. Suda, H. Oguri, H. Oikawa, Nat.
Chem. Biol. 2010, 6, 408 – 410.
[4] a) B. Sun, T. Morikawa, H. Matsuda, S. Tewtrakul, L.-J. Wu, S.
Harima, M. Yoshikawa, J. Nat. Prod. 2004, 67, 1464 – 1469; b) L.-
H. Cao, W. Zhang, J.-G. Luo, L.-Y. Kong, Helv. Chim. Acta 2012,
95, 1018 – 1025; c) F. Faini, R. Torres, F. D. Monache, G. B.
Marini-Bettolo, M. Castillo, Planta Med. 1980, 38, 128 – 132.
[5] T. Suzuki, T. Kubota, J. Kobayashi, Bioorg. Med. Chem. Lett.
2011, 21, 4220 – 4223.
[6] M.-L. Yang, P.-C. Kuo, T.-L. Hwang, T.-S. Wu, J. Nat. Prod. 2011,
74, 1996 – 2000.
[7] H. J. Shin, H.-S. Lee, D.-S. Lee, J. Microbiol. Biotechnol. 2010,
20, 501 – 505.
[8] a) A. L. Pumphrey, H. Dong, T. G. Driver, Angew. Chem. 2012,
124, 6022 – 6025; Angew. Chem. Int. Ed. 2012, 51, 5920 – 5923;
b) M.-L. Yang, P.-C. Kuo, A. G. Damu, R.-J. Chang, W.-F. Chiou,
T.-S. Wu, Tetrahedron 2006, 62, 10900 – 10906; c) J. Baiget, S.
[13] a) S. Simionatto, S. B. Marchioro, V. Galli, T. D. Luerce, D. D.
Hartwig, A. N. Moreira, O. A. Dellagostin, J. Microbiol. Meth-
ods 2009, 79, 101 – 105; b) X. Mo, J. Ma, H. Huang, B. Wang, Y.
Song, S. Zhang, C. Zhang, J. Ju, J. Am. Chem. Soc. 2012, 134,
2844 – 2847.
[14] R. B. Herbert, J. Mann, J. Chem. Soc. Perkin Trans. 1 1982, 1523 –
1525.
9984 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9980 –9984